Surgical evaluation

Surgical staging of the mediastinum is considered standard if accurate evaluation of the nodal status is needed to determine therapy.

Accurate staging of the mediastinal lymph nodes provides important prognostic information.

Evidence (nodal status):

1. The association between survival and the number of examined lymph nodes during surgery for patients with stage I NSCLC treated with definitive surgical resection was assessed from the population-based Surveillance, Epidemiology, and End Results (SEER) Program database for the period from 1990 to 2000.[2] A total of 16,800 patients were included in the study.

   1. The overall survival analysis for patients without radiation therapy was done by comparing the reference group (one to four lymph nodes) with the following groups:

      • Patients with five to eight lymph nodes examined during surgery had a modest but statistically significant increase in survival, with a proportionate hazard ratio (HR) of 0.90 (95% confidence interval [CI], 0.84–0.97).
      • For patients with 9 to 12 examined lymph nodes, the HR was 0.86 (95% CI, 0.79–0.95).
      • For patients with 13 to 16 examined lymph nodes, the HR was 0.78 (95% CI, 0.68–0.90).
      • There appeared to be no incremental improvement after evaluating more than 16 lymph nodes.
      The corresponding results for lung cancer–specific mortality and for patients who received radiation therapy were not substantially different.
   2. These results indicate that patient survival following resection for NSCLC is associated with the number of lymph nodes evaluated during surgery. Because this is most likely the result of a reduction-of-staging error, namely, a decreased likelihood of missing positive lymph nodes with an increasing number of lymph nodes sampled, it suggests that an evaluation of nodal status should include 11 to 16 lymph nodes.

CT imaging

CT scanning is primarily used for determining the size of the tumor. The CT scan should extend inferiorly to include the liver and adrenal glands. MRI scans of the thorax and upper abdomen do not appear to yield advantages over CT scans.[3]

Evidence (CT scan):

1. A systematic review of the medical literature relating to the accuracy of CT scanning for noninvasive staging of the mediastinum in patients with lung cancer was conducted. In the 35 studies published between 1991 and June 2006, 5,111 evaluable patients were identified. Almost all studies specified that CT scanning was performed following the administration of intravenous contrast material and that a positive test result was defined as the presence of one or more lymph nodes that measured larger than 1 cm on the short-axis diameter.[4]

   • The median prevalence of mediastinal metastasis was 28% (range, 18%–56%).
   • The pooled estimates of sensitivity and specificity of CT scanning for identifying mediastinal lymph node metastasis were 51% (95% CI, 47%–54%) for sensitivity and 86% (95% CI, 84%–88%) for specificity. Corresponding positive (3.4%) and negative (0.6%) likelihood ratios were provided.
2. The results from the systematic review are similar to those of a large meta-analysis that reported the median sensitivity and specificity of CT scanning for identifying malignant mediastinal nodes as 61% for sensitivity and 79% for specificity.[5]
3. An earlier meta-analysis reported an average sensitivity rate of 64% and specificity rate of 74%.[6]

18F-FDG PET scanning

The wider availability and use of 18F-FDG PET scanning for staging has modified the approach to staging mediastinal lymph nodes and distant metastases.

Randomized trials evaluating the utility of 18F-FDG PET scanning in potentially resectable NSCLC patients reported conflicting results in terms of the relative reduction in the number of noncurative thoracotomies.

Although the current evidence is conflicting, 18F-FDG PET scanning may improve results of early-stage lung cancer by identifying patients who have evidence of metastatic disease that is beyond the scope of surgical resection and that is not evident by standard preoperative staging procedures.

Evidence (18F-FDG PET scan):

1. A systematic review, an expansion of a health technology assessment conducted in 2001 by the Institute for Clinical and Evaluative Sciences, evaluated the accuracy and utility of 18F-FDG PET scanning in the diagnosis and staging of lung cancer.[7] Through a systematic search of the literature, 12 evidence summary reports and 15 prospective studies of the diagnostic accuracy of 18F-FDG PET scanning were identified.

   • 18F-FDG PET scanning appears to be superior to CT imaging for mediastinal staging in NSCLC.
   • 18F-FDG PET scanning also appears to have high sensitivity and reasonable specificity for differentiating benign from malignant lesions as small as 1 cm.
2. A systematic review of the medical literature relating to the accuracy of 18F-FDG PET scanning for noninvasive staging of the mediastinum in patients with lung cancer identified 44 studies published between 1994 and 2006 with 2,865 evaluable patients.[4]

   • The median prevalence of mediastinal metastases was 29% (range, 5%–64%).
   • Pooled estimates of sensitivity and specificity for identifying mediastinal metastasis were 74% (95% CI, 69%–79%) for sensitivity and 85% (95% CI, 82%–88%) for specificity.
   • Corresponding positive (4.9%) and negative (0.3%) likelihood ratios were provided for mediastinal staging with 18F-FDG PET scanning.
   • These findings demonstrated that 18F-FDG PET scanning is more accurate than CT scanning for staging of the mediastinum in patients with lung cancer.

Decision analyses demonstrate that 18F-FDG PET scanning may reduce the overall costs of medical care by identifying patients with falsely negative CT scans in the mediastinum or otherwise undetected sites of metastases.[8-10] Studies concluded that the money saved by forgoing mediastinoscopy in 18F-FDG PET-positive mediastinal lesions was not justified because of the unacceptably high number of false-positive results.[8-10] A randomized study found that the addition of 18F-FDG PET scanning to conventional staging was associated with significantly fewer thoracotomies.[11] A second randomized trial evaluating the impact of 18F-FDG PET scanning on clinical management found that 18F-FDG PET scanning provided additional information regarding appropriate stage but did not lead to significantly fewer thoracotomies.[12]

Combination of CT imaging and 18F-FDG PET scanning

The combination of CT imaging and 18F-FDG PET scanning has greater sensitivity and specificity than CT imaging alone.[13]

Evidence (CT/18F-FDG PET scan):

1. If there is no evidence of distant metastatic disease on CT scan, 18F-FDG PET scanning complements CT scan staging of the mediastinum. Numerous nonrandomized studies of 18F-FDG PET scanning have evaluated mediastinal lymph nodes using surgery (i.e., mediastinoscopy and/or thoracotomy with mediastinal lymph node dissection) as the gold standard of comparison.
2. A meta-analysis evaluated the conditional test performance of 18F-FDG PET scanning and CT scanning.

   • The median sensitivity and specificity of 18F-FDG PET scans were reported as 100% for sensitivity and 78% for specificity in patients with enlarged lymph nodes.[5]
   • 18F-FDG PET scanning is considered very accurate in identifying malignant nodal involvement when lymph nodes are enlarged. However, 18F-FDG PET scanning will falsely identify a malignancy in approximately one-fourth of patients with lymph nodes that are enlarged for other reasons, usually as a result of inflammation or infection.[14,15]
3. The median sensitivity and specificity of 18F-FDG PET scanning in patients with normal-sized mediastinal lymph nodes were 82% for sensitivity and 93% for specificity.[5] These data indicate that nearly 20% of patients with normal-sized lymph nodes but with malignant involvement had falsely negative 18F-FDG PET scan findings.

For patients with clinically operable NSCLC, the evidence supports performing a biopsy of mediastinal lymph nodes that are found to be larger than 1 cm in shortest transverse axis on chest CT scan or are found to be positive on 18F-FDG PET scan. Negative 18F-FDG PET scanning does not preclude biopsy of radiographically enlarged mediastinal lymph nodes. Mediastinoscopy is necessary for the detection of cancer in mediastinal lymph nodes when the results of the CT scan and 18F-FDG PET scan do not corroborate each other.

Adjuvant chemotherapy

Based on a meta-analysis, postoperative chemotherapy is not recommended outside of a clinical trial for patients with completely resected stage I NSCLC.[10] However, there may be some benefit of adjuvant chemotherapy in patients with stage IB tumors that are larger than 4 cm.

Evidence (adjuvant chemotherapy for patients with stage IB NSCLC):

1. The Cancer and Leukemia Group B study (CALGB-9633 [NCT00002852]) addressed the results of adjuvant carboplatin and paclitaxel versus observation for OS in 344 patients with resected stage IB (i.e., pathological T2, N0) NSCLC. Within 4 to 8 weeks of resection, patients were randomly assigned to postoperative chemotherapy or observation.[11]

   • Survival was not significantly different (HR, 0.83; 90% CI, 0.64–1.08; P = .12) at a median follow-up of 74 months.
   • Grades 3 to 4 neutropenia were the predominant toxicity; there were no treatment-related deaths.
   • A post-hoc exploratory analysis demonstrated a significant survival difference in favor of postoperative chemotherapy for patients who had tumors 4 cm or larger in diameter (HR, 0.69; 90% CI, 0.48–0.99; P = .043).

Given the magnitude of observed survival differences, CALGB-9633 may have been underpowered to detect small but clinically meaningful improvements in survival. In addition, the use of a carboplatin versus a cisplatin combination might have affected the results. At present, there is no reliable evidence that postoperative chemotherapy improves survival of patients with stage IB NSCLC.[11][Level of evidence A1]

Adjuvant targeted therapy (for patients with stage IB NSCLC with EGFR mutations)

Adjuvant targeted therapy with osimertinib for patients with EGFR-mutated NSCLC and resected stage IB to IIIA NSCLC was studied in a phase III clinical trial and showed improved OS.

Evidence (adjuvant targeted therapy with osimertinib for patients with stage IB EGFR-mutated NSCLC):

1. The ADAURA (NCT02511106) phase III, double-blind, placebo-controlled trial randomly assigned 682 patients with surgically resected stage IB to stage IIIA NSCLC with EGFR-sensitizing mutations (centrally determined, deletion in exon 19 or L858R mutation in exon 21) to receive either osimertinib 80 mg by mouth daily or a placebo for 3 years. Standard postoperative adjuvant chemotherapy was allowed but not mandatory; decisions regarding adjuvant chemotherapy were made by the physician and patient before trial enrollment. There were 399 patients who received osimertinib and 342 patients who received placebo.[12][Level of evidence A1]

   • In the overall population, the 5-year OS rate was 88% in the osimertinib group and 78% in the placebo group (overall HR_death, 0.49; 95.03% CI, 0.34–0.70; P < .001).
   • Among patients with stage II to IIIA disease, the 5-year OS rate was 85% in the osimertinib group and 73% in the placebo group (overall HR_death, 0.49; 95.03% CI, 0.33–0.73; P < .001).
   • The adverse event profile is consistent with other studies that used osimertinib except for pneumonia related to COVID-19, which was reported later.

The U.S. Food and Drug Administration (FDA) approved osimertinib as adjuvant therapy for patients with stage IB to IIIA NSCLC with EGFR exon 19 deletions or exon 21 L858R mutations.

Adjuvant immunotherapy

Evidence (adjuvant immunotherapy with pembrolizumab for patients with stage IB tumors >4 cm):

1. The phase III, multicenter, open-label PEARLS/KEYNOTE-091 trial (NCT02504372) randomly assigned 1,177 patients with completely resected stage IB (tumor >4 cm) to stage IIIA NSCLC to receive pembrolizumab (200 mg every 3 weeks) or placebo for up to 18 cycles, or until disease progression or unacceptable toxicity. Patients started study treatment after resection or, if indicated, after adjuvant chemotherapy (maximum of four cycles). The dual primary end points were DFS in the overall study population and DFS in patients with a programmed death-ligand 1 (PD-L1) tumor proportion score (TPS) of 50% or greater, as determined using the PD-L1 IHC 22C3 pharmDx assay. These end points were reported in a prespecified interim analysis after a median follow-up of 35.6 months (interquartile range, 27.1–45.5).[13][Level of evidence B1]

   • In the overall study population, the median DFS was 53.6 months (95% CI, 39.2–not reached [NR]) in the pembrolizumab group and 42.0 months (95% CI, 31.3–NR) in the placebo group (HR, 0.76; 95% CI, 0.63–0.91; P = .0014).
   • In the PD-L1 TPS ≥50% population, the median DFS was not reached with either pembrolizumab (95% CI, 44.3–NR) or placebo (95% CI, 35.8–NR) (HR, 0.82; 95% CI, 0.57–1.18; P = .14).
   • OS data were immature at the time of the prespecified interim analysis.
   • No new safety signals were identified in this study.

The FDA approved pembrolizumab as a single agent for adjuvant treatment following resection and platinum-based chemotherapy for patients with stage IB (T2a ≥4 cm), II, or IIIA NSCLC. Of note, the FDA label specifies that pembrolizumab can be used as adjuvant therapy after platinum-based chemotherapy. However, chemotherapy was not required in the overall study patient population evaluated in KEYNOTE-091.

Adjuvant external radiation therapy

The value of postoperative (adjuvant) radiation therapy (PORT) has been evaluated and has not been found to improve the outcome of patients with completely resected stage I NSCLC.[14]

Adjuvant brachytherapy

The value of intraoperative (adjuvant) brachytherapy applied to the suture line has been evaluated in patients undergoing sublobar resections for stage I NSCLC to improve local control; it has not been found to improve outcomes.

Evidence (adjuvant brachytherapy):

1. A phase III trial that randomly assigned 222 patients to undergo sublobar resection with or without suture line brachytherapy reported the following:[15]

   • No difference in the primary end point of local recurrence (5-year estimate, 14.0% vs. 16.7%; P = .59).
   • No difference in OS rates (5-year estimate, 61.4% vs. 55.6%; P = .38).[15][Level of evidence B1] vs. [Level of evidence A1]

Conventional radiation therapy

Historically, conventional primary radiation therapy consisted of approximately 60 Gy to 70 Gy delivered with megavoltage equipment to the midplane of the known tumor volume using conventional fractionation (1.8–2.0 Gy per day).

Improvements in radiation techniques include planning techniques to account for tumor motion, more conformal planning techniques (e.g., 3-D conformal radiation therapy and intensity-modulated radiation therapy), and image guidance during treatment. Modern approaches to delivery of external-beam radiation therapy (EBRT) include hypofractionated radiation therapy and stereotactic body radiation therapy (SBRT). However, there are limited reliable data from comparative trials to determine which approaches yield superior outcomes.[17,18]

Evidence (conventional radiation therapy):

1. In the largest retrospective conventional radiation therapy series, patients with inoperable disease were treated with definitive radiation therapy.[20-22]

   • Patients achieved 5-year survival rates of 10% to 30%.[20-22]
   • Several series demonstrated that patients with T1, N0 tumors had better outcomes, and 5-year survival rates of 30% to 60% were found in this subgroup.[20,21,23]
   • However, local-only failure occurs in as many as 50% of patients treated with conventional radiation therapy to doses in the range of 60 Gy to 65 Gy.[24,25]
2. A single report of patients older than 70 years who had resectable lesions smaller than 4 cm but who had medically inoperable disease or who refused surgery reported the following:[23]

   • Survival at 5 years after radiation therapy with curative intent was comparable with a historical control group of patients of similar age who were resected with curative intent.
3. A small case series using matched controls reported the following:[4]

   • The addition of endobronchial brachytherapy improved local disease control compared with EBRT.[4][Level of evidence C2]

Hypofractionated radiation therapy

Hypofractionated radiation therapy involves the delivery of a slightly higher dose of radiation therapy per day (e.g., 2.4–4.0 Gy) over a shorter period of time compared with conventionally fractionated radiation therapy. Multiple prospective phase I/II trials have demonstrated that hypofractionated radiation therapy to a dose of 60 Gy to 70 Gy delivered over 3 to 4 weeks with 2.4 Gy to 4.0 Gy per day resulted in a low incidence of moderate to severe toxicity, 2-year OS rates of 50% to 60%, and 2-year tumor local control of 80% to 90%.[26-28][Level of evidence C1]

Stereotactic body radiation therapy (SBRT)

SBRT involves the delivery of highly conformal, high-dose radiation therapy over an extremely hypofractionated course (e.g., one to five treatments) delivered over 1 to 2 weeks. Commonly used regimens include 18 Gy × 3, 12 Gy to 12.5 Gy × 4, and 10 Gy to 12 Gy × 5, and deliver a substantially higher biologically effective dose compared with historic conventional radiation therapy regimens.

Multiple prospective phase I/II trials and institutional series have demonstrated that SBRT results in a low incidence of pulmonary toxicity (<10% risk of symptomatic radiation pneumonitis), 2-year OS rates of 50% to 60%, and 2-year tumor control of 90% to 95%.[29-35][Level of evidence C1]

Evidence (SBRT):

1. Early phase I/II trials from Indiana University identified the maximum tolerated dose of three-fraction SBRT at 18 Gy × 3 for T1 tumors.

   • This regimen resulted in a 2-year OS rate of 55% and 2-year local tumor control of 95%.
   • An unacceptably high incidence (8.6%) of grade 5 toxicity was observed in patients with central tumors (defined as within 2 cm of the tracheobronchial tree from the trachea to the level of the lobar bronchi).[30]
2. A subsequent multicenter trial (RTOG-0236 [NCT00087438]) studied the 18 Gy × 3 regimen in 55 patients with peripheral T1 to T2 tumors only.

   • This trial demonstrated a 3-year OS rate of 56% and 3-year primary tumor control of 98%.
   • The incidence of moderate to severe toxicity was low, with grade 3 toxicity in 24% of patients, grade 4 toxicity in 4% of patients, and no grade 5 toxicity, with a 4% incidence of grade 3 radiation pneumonitis.[34]
3. In the largest reported series from VU University Medical Center Amsterdam, 676 patients with T1 to T2 tumors were treated with three-, five-, and eight-fraction SBRT using a risk-adapted approach (a tailored fractionation regimen based on tumor proximity to critical organs).

   • With a median follow-up of 32.9 months, the median OS was 40.7 months, and 2-year local tumor control was 95%.[35]
4. While central location is a contraindication to three-fraction SBRT based on data from the Indiana phase II study, a subsequent systematic review of published reports of 315 patients with 563 central tumors demonstrated a much lower incidence of severe toxicity, including a 1% to 5% risk of grade 5 events with more protracted SBRT regimens (e.g., four to ten fractions).[36]
5. A multicenter phase I/II trial (RTOG-0813 [NCT00750269]) is ongoing to identify the maximum tolerated dose for a five-fraction SBRT regimen for central tumors.

A randomized trial of hypofractionated radiation therapy versus SBRT (LUSTRE [NCT01968941]) is ongoing to determine the optimal radiation therapy regimen, but SBRT has been widely adopted for patients with medically inoperable stage I NSCLC.

Surgery alone

Surgery is the treatment of choice for patients with stage II NSCLC. A lobectomy, pneumonectomy, segmental resection, wedge resection, or sleeve resection may be performed as appropriate. Careful preoperative assessment of the patient’s overall medical condition, especially the patient’s pulmonary reserve, is critical in considering the benefits of surgery. In addition to the immediate and age-related postoperative mortality rate, a 5% to 8% mortality rate with pneumonectomy or a 3% to 5% mortality rate with lobectomy can be expected.

Evidence (surgery):

1. The Cochrane Collaboration reviewed 11 randomized trials with a total of 1,910 patients who underwent surgical interventions for early-stage (I–IIIA) lung cancer.[1] A pooled analysis of three trials reported the following:

   • Four-year survival was superior in patients with resectable stage I, II, or IIIA NSCLC who underwent resection and complete ipsilateral mediastinal lymph node dissection (CMLND), compared with those who underwent resection and lymph node sampling; the hazard ratio (HR) was estimated to be 0.78 (95% confidence interval [CI], 0.65–0.93; P = .005).[1][Level of evidence A1]
   • There was a significant reduction in any cancer recurrence (local or distant) in the CMLND group (relative risk [RR], 0.79; 95% CI, 0.66–0.95; P = .01) that appeared mainly as the result of a reduction in the number of distant recurrences (RR, 0.78; 95% CI, 0.61–1.00; P = .05).
   • There was no difference in operative mortality.
   • Air leak lasting more than 5 days was significantly more common in patients assigned to CMLND (RR, 2.94; 95% CI, 1.01–8.54; P = .05).
2. CMLND versus lymph node sampling was evaluated in a large randomized phase III trial (ACOSOG-Z0030 [NCT00003831]).[2]

   • Preliminary analyses of operative morbidity and mortality showed comparable rates from the procedures.[2]
   • There was no difference in overall survival (OS), disease-free survival (DFS), local recurrence, and regional recurrence.[3][Level of evidence A1]

Evidence suggests that lung cancer resection combined with CMLND is not associated with improvement in survival compared with lung cancer resection combined with systematic sampling of mediastinal lymph nodes in patients with stage I, II, or IIIA NSCLC.[3][Level of evidence A1]

Conclusions about the efficacy of surgery for patients with local and locoregional NSCLC are limited by the small number of participants studied and potential methodological weaknesses of the trials.

Adjuvant chemotherapy

The preponderance of evidence indicates that postoperative cisplatin combination chemotherapy provides a significant survival advantage to patients with resected stage II NSCLC. Preoperative chemotherapy may also provide survival benefit. The optimal sequence of surgery and chemotherapy and the benefits and risks of postoperative radiation therapy in patients with resectable NSCLC remain to be determined.

After surgery, many patients develop regional or distant metastases.[4] Several randomized controlled trials and meta-analyses have evaluated the use of postoperative chemotherapy in patients with stage I, II, and IIIA NSCLC.[5-11]

Evidence (adjuvant chemotherapy):

1. Data on individual patient outcomes were collected and pooled into a meta-analysis from the five largest trials (4,584 patients) of cisplatin-based chemotherapy in patients with completely resected NSCLC that were conducted after 1995.[7]

   • With a median follow-up time of 5.2 years, the overall HR_death was 0.89 (95% CI, 0.82–0.96; P = .005), corresponding to a 5-year absolute benefit of 5.4% from chemotherapy.
   • The benefit varied with stage (test for trend, P = .04; HR for stage IA, 1.40; 95% CI, 0.95–2.06; HR for stage IB, 0.93; 95% CI, 0.78–1.10; HR for stage II, 0.83; 95% CI, 0.73–0.95; and HR for stage III, 0.83; 95% CI, 0.72–0.94).
   • The effect of chemotherapy did not vary significantly (test for interaction, P = .11) with the associated drugs, including vinorelbine (HR, 0.80; 95% CI, 0.70–0.91), etoposide or vinca alkaloid (HR, 0.92; 95% CI, 0.80–1.07), or other drugs (HR, 0.97; 95% CI, 0.84–1.13).
   • The greater effect on survival observed with the doublet of cisplatin plus vinorelbine compared with other regimens should be interpreted cautiously as the total dose of cisplatin received was significantly higher in patients treated with vinorelbine.
2. The meta-analysis [7] and the individual studies [5,12] support the administration of postoperative cisplatin-based chemotherapy in combination with vinorelbine.

   1. Superior OS for the trial population and patients with stage II disease was reported for the Lung Adjuvant Cisplatin Evaluation (LACE) pooled analysis (pooled HR, 0.83; 95% CI, 0.73–0.95); the Adjuvant Navelbine International Trialist Association (ANITA) trial (HR, 0.71; 95% CI, 0.49–1.03); and the National Cancer Institute of Canada Clinical Trials Group JBR.10 trial (HR, 0.59; 95% CI, 0.42–0.85).
   2. Chemotherapy effect was higher in patients with better performance status.
   3. There was no interaction between chemotherapy effect and any of the following:

      • Sex.
      • Age.
      • Histology.
      • Type of surgery.
      • Planned radiation therapy.
      • Planned total dose of cisplatin.
3. In a retrospective analysis of a phase III trial of postoperative cisplatin and vinorelbine, patients older than 65 years were found to benefit from treatment.[13]

   • Chemotherapy significantly prolonged OS for patients older than 65 years (HR, 0.61; 95% CI, 0.38–0.98; P = .04).
   • There were no significant differences in toxic effects, hospitalization, or treatment-related death by age group, although patients older than 65 years received less treatment.[13]
4. Several other randomized controlled trials and meta-analyses have evaluated the use of postoperative chemotherapy in patients with stages I, II, and IIIA NSCLC.[5-11]

Based on these data, patients with completely resected stage II lung cancer may benefit from postoperative cisplatin-based chemotherapy.[13][Level of evidence A1]

Adjuvant targeted therapy (for patients with EGFR mutations)

Adjuvant targeted therapy with osimertinib for patients with EGFR-mutated NSCLC and resected stage IB to IIIA NSCLC was studied in a phase III clinical trial and showed improved OS.

Evidence (adjuvant targeted therapy with osimertinib for patients with stages IIA and IIB EGFR-mutated NSCLC):

1. The ADAURA (NCT02511106) phase III, double-blind, placebo-controlled trial randomly assigned 682 patients with surgically resected stage IB to stage IIIA NSCLC with EGFR-sensitizing mutations (centrally determined, deletion in exon 19 or L858R mutation in exon 21) to receive osimertinib 80 mg by mouth daily or a placebo for 3 years. Standard postoperative adjuvant chemotherapy was allowed but not mandatory; decisions regarding adjuvant chemotherapy were made by the physician and patient before trial enrollment. There were 399 patients who received osimertinib and 342 patients who received placebo.[14][Level of evidence A1]

   • In the overall population, the 5-year OS rate was 88% in the osimertinib group and 78% in the placebo group (overall HR_death, 0.49; 95.03% CI, 0.34–0.70; P < .001).
   • Among patients with stage II to IIIA disease, the 5-year OS rate was 85% in the osimertinib group and 73% in the placebo group (overall HR_death, 0.49; 95.03% CI, 0.33–0.73; P < .001).
   • The adverse event profile is consistent with other studies that used osimertinib except for pneumonia related to COVID-19, which was reported later.

The U.S. Food and Drug Administration (FDA) approved osimertinib as adjuvant therapy for patients with stage IB to IIIA NSCLC with EGFR exon 19 deletions or exon 21 L858R mutations.

Adjuvant immunotherapy

Adjuvant immunotherapy for patients with resected stage IB to IIIA NSCLC has been found to significantly increase DFS.[15,16]

Evidence (adjuvant immunotherapy with pembrolizumab for patients with stage IIA and IIB tumors >4 cm):

1. The phase III, multicenter, open-label PEARLS/KEYNOTE-091 trial (NCT02504372) randomly assigned 1,177 patients with completely resected stage IB (tumor >4 cm) to stage IIIA NSCLC to receive pembrolizumab (200 mg every 3 weeks) or placebo for up to 18 cycles, or until disease progression or unacceptable toxicity. Patients started study treatment after resection or, if indicated, after adjuvant chemotherapy (maximum of four cycles). The dual primary end points were DFS in the overall study population and DFS in patients with a programmed death-ligand 1 (PD-L1) tumor proportion score (TPS) of 50% or greater, as determined using the PD-L1 IHC 22C3 pharmDx assay. These end points were reported in a prespecified interim analysis after a median follow-up of 35.6 months (interquartile range [IQR], 27.1–45.5).[15][Level of evidence B1]

   • In the overall study population, the median DFS was 53.6 months (95% CI, 39.2–not reached [NR]) in the pembrolizumab group and 42.0 months (95% CI, 31.3–NR) in the placebo group (HR, 0.76; 95% CI, 0.63–0.91; P = .0014).
   • In the PD-L1 TPS ≥50% population, the median DFS was not reached with either pembrolizumab (95% CI, 44.3–NR) or placebo (95% CI, 35.8–NR) (HR, 0.82; 95% CI, 0.57–1.18; P = .14).
   • OS data were immature at the time of the prespecified interim analysis.
   • No new safety signals were identified in this study.

The FDA approved pembrolizumab as a single agent for adjuvant treatment following resection and platinum-based chemotherapy for patients with stage IB (T2a ≥4 cm), II, or IIIA NSCLC. Of note, the FDA label specifies that pembrolizumab can be used as adjuvant therapy after platinum-based chemotherapy. However, chemotherapy was not required in the overall study patient population evaluated in KEYNOTE-091.

Evidence (adjuvant immunotherapy with atezolizumab for patients with stages IIA and IIB NSCLC):

1. IMpower010 (NCT02486718) was a phase III, multicenter, open-label trial that randomly assigned 1,005 patients with surgically resected stage IB (tumor >4 cm) to stage IIIA NSCLC. Patients received atezolizumab (1,200 mg every 21 days intravenously) or best supportive care for 16 cycles or 1 year after standard adjuvant platinum-based chemotherapy. Patients were enrolled after resection if they were eligible for cisplatin-based chemotherapy and were randomized after completion of chemotherapy if they remained eligible and did not experience disease progression. The primary end point was investigator-assessed DFS.[16]

   • The primary end point was tested hierarchically, first in the stage II to IIIA population subgroup whose tumors expressed PD-L1 on at least 1% of tumor cells (using the SP263 antibody), then in all patients in the stage II to IIIA population, and finally in the intention-to-treat (ITT) population (stage IB to IIIA). Of the 882 patients who were randomly assigned and had stage II to IIIA disease, 476 had tumors expressing PD-L1 on at least 1% of tumor cells per SP263.[16][Level of evidence B1]
   • After a median follow-up of 32.2 months, atezolizumab treatment improved DFS compared with best supportive care in patients in the stage II to IIIA population whose tumors expressed PD-L1 on at least 1% of tumor cells (HR, 0.66; 95% CI, 0.50–0.88; P = .0039). At 24 months, the DFS rate was 74.6% for the atezolizumab group and 61.0% for the best supportive care group.
   • Atezolizumab also improved DFS in all patients in the stage II to IIIA population (HR, 0.79; 95% CI, 0.64–0.96; P = .020). At 24 months, the DFS rate was 70.2% for the atezolizumab group and 61.6% for the best supportive care group.
   • In the ITT population, which included patients with stage IB to IIIA disease, HR_DFS was 0.81 (95% CI, 0.67–0.99; P = .040). However, the boundary for statistical significance for DFS was not crossed.
   • OS data are immature.
   • No new safety signals were noted.

The FDA approved atezolizumab for adjuvant treatment of patients with stage II to IIIA NSCLC whose tumors express PD-L1 on at least 1% of tumor cells.

Adjuvant radiation therapy

The value of postoperative (adjuvant) radiation therapy (PORT) has been evaluated.[17]

Evidence (adjuvant radiation therapy):

1. A meta-analysis, based on the results of ten randomized controlled trials and 2,232 individuals, reported the following:[17]

   • An 18% relative increase in the risk of death for patients who received PORT compared with surgery alone (HR, 1.18; P = .002). This is equivalent to an absolute detriment of 6% at 2 years (95% CI, 2%–9%), reducing OS from 58% to 52%. Exploratory subgroup analyses suggested that this detrimental effect was most pronounced for patients with stage I/II, N0 to N1 disease, whereas for patients with stage III, N2 disease there was no clear evidence of an adverse effect.
   • Results for local (HR, 1.13; P = .02), distant (HR, 1.14; P = .02), and overall (HR, 1.10; P = .06) recurrence-free survival similarly showed a detriment of PORT.[17][Level of evidence A1]

Further analysis is needed to determine whether these outcomes can potentially be modified with technical improvements, better definitions of target volumes, and limitation of cardiac volume in the radiation portals.

Neoadjuvant chemotherapy

The role of chemotherapy before surgery was tested in clinical trials. The proposed benefits of preoperative chemotherapy include:

• A reduction in tumor size that may facilitate surgical resection.
• Early eradication of micrometastases.
• Better tolerability.

Preoperative chemotherapy may, however, delay potentially curative surgery.

Evidence (neoadjuvant chemotherapy):

1. The Cochrane Collaboration reported a systematic review and meta-analysis of seven randomized controlled trials that included 988 patients and evaluated the addition of preoperative chemotherapy to surgery versus surgery alone. These trials evaluated patients with stages I, II, and IIIA NSCLC.[18]

   • Preoperative chemotherapy provided an absolute benefit in survival of 6% across all stages of disease, from 14% to 20% at 5 years (HR, 0.82; 95% CI, 0.69–0.97; P = .022).[18][Level of evidence A1]
   • This analysis was unable to address questions such as whether particular types of patients may benefit more or less from preoperative chemotherapy.
2. In the largest trial reported to date, 519 patients were randomly assigned to receive either surgery alone or three cycles of platinum-based chemotherapy followed by surgery. Most patients (61%) had clinical stage I disease; 31% had stage II disease; and 7% had stage III disease.[19]

   • No survival advantage was seen.[19]
   • Postoperative complications were similar between groups, and no impairment of quality of life was observed.
   • There was no evidence of a benefit in terms of OS (HR, 1.02; 95% CI, 0.80–1.31; P = .86).
   • Updating the systematic review by addition of the present result suggests a 12% relative survival benefit with the addition of neoadjuvant (preoperative) chemotherapy (1,507 patients; HR, 0.88; 95% CI, 0.76–1.01; P = .07), equivalent to an absolute improvement in survival of 5% at 5 years.

Neoadjuvant immunotherapy with chemotherapy

Perioperative (neoadjuvant and adjuvant) immunotherapy with chemotherapy

Several immune checkpoint inhibitors have been approved by the FDA for select patient populations with potentially resectable NSCLC, either in the neoadjuvant setting (nivolumab) or adjuvant setting (atezolizumab, durvalumab, or pembrolizumab). Ongoing phase III trials are evaluating the role of perioperative immune checkpoint inhibitors. These regimens for patients with potentially resectable stages II to III NSCLC include neoadjuvant immune checkpoint inhibitors with chemotherapy followed by surgery and adjuvant immune checkpoint inhibitors. Compared with neoadjuvant chemotherapy alone, early results from studies of perioperative immune checkpoint inhibitor regimens have shown improvements in several key outcomes including EFS, major pathological response, pathological complete response, and OS.

Neoadjuvant chemotherapy

The role of chemotherapy before surgery in patients with stage IIIA NSCLC has been extensively tested in clinical trials. The proposed benefits of preoperative (neoadjuvant) chemotherapy include:

• A reduction in tumor size that may facilitate surgical resection.
• Early eradication of micrometastases.
• Better tolerability.

Evidence (neoadjuvant chemotherapy):

1. The Cochrane Collaboration provided a systematic review and meta-analysis of seven randomized controlled trials that included 988 patients and evaluated the addition of preoperative chemotherapy to surgery versus surgery alone.[5] These trials evaluated patients with stages I, II, and IIIA NSCLC.

   • Preoperative chemotherapy provided an absolute benefit in survival of 6% across all stages of disease, from 14% to 20% at 5 years (HR, 0.82; 95% CI, 0.69–0.97; P = .022).[5][Level of evidence A1]
   • This analysis was unable to address questions such as whether particular types of patients may benefit more or less from preoperative chemotherapy.[6]
2. In the largest trial reported to date, 519 patients were randomly assigned to receive either surgery alone or three cycles of platinum-based chemotherapy followed by surgery.[7] Most patients (61%) had clinical stage I disease, 31% had stage II disease, and 7% had stage III disease.

   • Postoperative complications were similar between groups, and no impairment of quality of life was observed.
   • There was no evidence of a benefit in terms of OS (HR, 1.02; 95% CI, 0.80–1.31; P = .86).
   • Updating the systematic review by addition of the present result suggests a 12% relative survival benefit with the addition of preoperative chemotherapy (1,507 patients; HR, 0.88; 95% CI, 0.76–1.01; P = .07), equivalent to an absolute improvement in survival of 5% at 5 years.[7]

Neoadjuvant chemoradiation therapy

Administering concurrent neoadjuvant chemotherapy and radiation therapy before surgery may intensify treatment and increase the likelihood of downstaging the tumor burden. Commonly used regimens that have been tested in the phase II setting include cisplatin/etoposide (EP5050) and weekly carboplatin/paclitaxel.[8,9] In a randomized trial of neoadjuvant chemoradiation therapy and surgery versus concurrent chemoradiation therapy alone, there was no difference in OS, but surgery improved PFS and local control.[2][Level of evidence B1]

Evidence (neoadjuvant chemoradiation therapy):

1. The Intergroup-0139 trial (NCT00002550) compared chemoradiation therapy alone with neoadjuvant chemoradiation followed by surgery in 396 patients with stage IIIA (N2) NSCLC.[2]

   • Surgery did not improve OS (5-year OS rate, 27% vs. 20%; HR, 0.87; 0.70–1.10; P = .24).
   • Surgery improved PFS (5-year PFS rate, 22% vs. 11%; HR, 0.77; 0.62–0.96; P = .017) and decreased the risk of local recurrence (10% vs. 22%; P = .002).
   • There was increased treatment mortality with neoadjuvant chemoradiation with surgery (8% vs. 2%), particularly in the subset of patients who underwent pneumonectomy.

A direct comparison of neoadjuvant chemotherapy versus neoadjuvant chemoradiation therapy using modern treatment regimens has not been performed to date; the optimal neoadjuvant approach remains unclear.

Neoadjuvant immunotherapy with chemotherapy

Perioperative (neoadjuvant and adjuvant) immunotherapy with chemotherapy

Several immune checkpoint inhibitors have been approved by the FDA for select patient populations with potentially resectable NSCLC, either in the neoadjuvant setting (nivolumab) or adjuvant setting (atezolizumab, durvalumab, or pembrolizumab). Ongoing phase III trials are evaluating the role of perioperative immune checkpoint inhibitors. These regimens for patients with potentially resectable stages II to III NSCLC include neoadjuvant immune checkpoint inhibitors with chemotherapy followed by surgery and adjuvant immune checkpoint inhibitors. Compared with neoadjuvant chemotherapy alone, early results from studies of perioperative immune checkpoint inhibitor regimens have shown improvements in several key outcomes including EFS, major pathological response, pathological complete response, and OS.

Adjuvant chemotherapy

Patients with completely resected stage IIIA NSCLC may benefit from postoperative cisplatin-based chemotherapy.[16][Level of evidence A1]

Evidence (adjuvant chemotherapy):

Evidence from randomized controlled clinical trials indicates that when stage IIIA NSCLC is encountered unexpectedly at surgery, chemotherapy given after complete resection improves survival.

Several randomized, controlled trials and meta-analyses have evaluated the use of postoperative chemotherapy in patients with stages I, II, and IIIA NSCLC.[16-22]

1. Data on individual patient outcomes from the five largest trials (4,584 patients) that were conducted after 1995 of cisplatin-based chemotherapy in patients with completely resected NSCLC were collected and pooled into a meta-analysis.[16]

   • With a median follow-up of 5.2 years, the overall HR_death was 0.89 (95% CI, 0.82–0.96; P = .005), corresponding to a 5-year absolute benefit of 5.4% from chemotherapy.
   • The effect of chemotherapy did not vary significantly (test for interaction, P = .11) with the associated drugs, including vinorelbine (HR, 0.80; 95% CI, 0.70–0.91), etoposide or vinca alkaloid (HR, 0.92; 95% CI, 0.80–1.07), or other drugs (HR, 0.97; 95% CI, 0.84–1.13).
   • The benefit varied with stage (HR for stage IIIA, 0.83; 95% CI, 0.72–0.94).
   • The greater effect on survival observed with the doublet of cisplatin plus vinorelbine compared with other regimens should be interpreted with caution as the total dose of cisplatin received was significantly higher in patients treated with vinorelbine.
2. Two trials (FRE-IALT and the Adjuvant Navelbine International Trialist Association [ANITA] trial) reported significant OS benefits associated with postoperative chemotherapy in stage IIIA disease.[6,18]

   1. For the subgroup of stage IIIA patients in the ANITA trial (n = 325), the HR was 0.69 (95% CI, 0.53–0.90), and the result for the FRE-IALT trial (n = 728) was HR, 0.79 (95% CI, 0.66–0.95).
   2. The chemotherapy effect was higher in patients with a better performance status.
   3. There was no interaction between the chemotherapy effect and any of the following:

      • Sex.
      • Age.
      • Histology.
      • Type of surgery.
      • Planned radiation therapy.
      • Planned total dose of cisplatin.
3. In a retrospective analysis of a phase III trial of postoperative cisplatin and vinorelbine, patients older than 65 years were found to benefit from treatment.[23]

   • Chemotherapy significantly prolonged OS for patients older than 65 years (HR, 0.61; 95% CI, 0.38–0.98; P = .04).
   • There were no significant differences in toxic effects, hospitalization, or treatment-related death by age group, although patients older than 65 years received less treatment.

Adjuvant targeted therapy (for patients with EGFR mutations)

Adjuvant targeted therapy with osimertinib for patients with EGFR-mutated NSCLC and resected stage IB to IIIA NSCLC was studied in a phase III clinical trial and showed improved OS.

Evidence (adjuvant targeted therapy with osimertinib for patients with stage IIIA EGFR-mutated NSCLC):

1. The ADAURA (NCT02511106) phase III, double-blind, placebo-controlled trial randomly assigned 682 patients with surgically resected stage IB to stage IIIA NSCLC with EGFR-sensitizing mutations (centrally determined, deletion in exon 19 or L858R mutation in exon 21) to osimertinib 80 mg by mouth daily or a placebo for 3 years. Standard postoperative adjuvant chemotherapy was allowed but not mandatory; decisions regarding adjuvant chemotherapy were made by the physician and patient before trial enrollment. There were 399 patients who received osimertinib and 342 patients who received placebo.[24][Level of evidence A1]

   • In the overall population, the 5-year OS rate was 88% in the osimertinib group and 78% in the placebo group (overall HR_death, 0.49; 95.03% CI, 0.34–0.70; P < .001).
   • Among patients with stage II to IIIA disease, the 5-year OS rate was 85% in the osimertinib group and 73% in the placebo group (overall HR_death, 0.49; 95.03% CI, 0.33–0.73; P < .001).
   • The adverse event profile is consistent with other studies that used osimertinib except for pneumonia related to COVID-19, which was reported later.

The FDA approved osimertinib as adjuvant therapy for patients with stage IB to IIIA NSCLC with EGFR exon 19 deletions or exon 21 L858R mutations.

Adjuvant immunotherapy

Adjuvant immunotherapy for patients with resected stage IB to IIIA NSCLC has been found to significantly increase DFS.[25,26]

Evidence (adjuvant immunotherapy with pembrolizumab for patients with stage IIIA NSCLC):

1. The phase III, multicenter, open-label PEARLS/KEYNOTE-091 trial (NCT02504372) randomly assigned 1,177 patients with completely resected stage IB (tumor >4 cm) to stage IIIA NSCLC to receive pembrolizumab (200 mg every 3 weeks) or placebo for up to 18 cycles, or until disease progression, or unacceptable toxicity. Patients started study treatment after resection or, if indicated, after adjuvant chemotherapy (maximum of four cycles). The dual primary end points were DFS in the overall study population and DFS in patients with a PD-L1 tumor proportion score (TPS) of 50% or greater, as determined using the PD-L1 IHC 22C3 pharmDx assay. These end points were reported in a prespecified interim analysis after a median follow-up of 35.6 months (interquartile range, 27.1–45.5).[25][Level of evidence B1]

   • In the overall study population, the median DFS was 53.6 months (95% CI, 39.2 to NR) in the pembrolizumab group and 42.0 months (95% CI, 31.3–NR) in the placebo group (HR, 0.76; 95% CI, 0.63–0.91, P = .0014).
   • In the PD-L1 TPS ≥50% population, the median DFS was not reached with either pembrolizumab (95% CI, 44.3–NR) or placebo (95% CI, 35.8–NR) (HR, 0.82; 95% CI, 0.57–1.18; P = .14).
   • OS data were immature at the time of prespecified interim analysis.
   • No new safety signals were identified in this study.

The FDA approved pembrolizumab as a single agent for adjuvant treatment following resection and platinum-based chemotherapy for patients with stage IB (T2a ≥4 cm), II, or IIIA NSCLC. Of note, the FDA label specifies that pembrolizumab can be used as adjuvant therapy after platinum-based chemotherapy. However, chemotherapy was not required in the overall study patient population evaluated in KEYNOTE-091.

Evidence (adjuvant immunotherapy with atezolizumab for patients with resected stage IIIA NSCLC):

1. IMpower010 (NCT02486718) was a phase III, multicenter, open-label trial that randomly assigned 1,005 patients with surgically resected stage IB (tumor >4 cm) to stage IIIA NSCLC. Patients received atezolizumab (1,200 mg every 21 days intravenously) or best supportive care for 16 cycles or 1 year after standard adjuvant platinum-based chemotherapy. Patients were enrolled after resection if they were eligible for cisplatin-based chemotherapy and were randomized after completion of chemotherapy if they remained eligible and did not experience disease progression. The primary end point was investigator-assessed DFS.[26]

   • The primary end point was tested hierarchically, first in the stage II to IIIA population subgroup whose tumors expressed PD-L1 on at least 1% of tumor cells (using the SP263 antibody), then in all patients in the stage II to IIIA population, and finally in the ITT population (stage IB to IIIA). Of the 882 patients who were randomly assigned and had stage II to IIIA disease, 476 had tumors expressing PD-L1 on at least 1% of tumor cells per SP263.[26][Level of evidence B1]
   • After a median follow-up of 32.2 months, atezolizumab treatment improved DFS compared with best supportive care in patients in the stage II to IIIA population whose tumors expressed PD-L1 on at least 1% of tumor cells (HR, 0.66; 95% CI, 0.50–0.88; P = .0039). At 24 months, the DFS rate was 74.6% for the atezolizumab group and 61.0% for the best supportive care group.
   • Atezolizumab also improved DFS in all patients in the stage II to IIIA population (HR, 0.79; 95% CI, 0.64–0.96; P = .020). At 24 months, the DFS rate was 70.2% for the atezolizumab group and 61.6% for the best supportive care group.
   • In the ITT population, which included patients with stage IB to IIIA disease, HR_DFS was 0.81 (95% CI, 0.67–0.99; P = .040). However, the boundary for statistical significance for DFS was not crossed.
   • OS data are immature.
   • No new safety signals were noted.

The FDA approved atezolizumab for adjuvant treatment of patients with stage II to IIIA NSCLC whose tumors express PD-L1 on at least 1% of tumor cells.

Adjuvant chemoradiation therapy

Combination chemotherapy and radiation therapy administered before or following surgery should be viewed as investigational and requiring evaluation in future clinical trials.

Evidence (adjuvant chemoradiation therapy):

1. Five randomized trials have assessed the value of postoperative combination chemoradiation therapy versus radiation therapy following surgical resection.[5,7,27-29][Level of evidence A1]

   • Only one trial reported improved DFS, and no trial reported improved OS.
2. Three trials have evaluated platinum-based combination chemotherapy followed by surgery versus platinum-based chemotherapy followed by radiation therapy (60–69.6 Gy) alone to determine whether surgery or radiation therapy was most efficacious.[29-31] Although the studies were small, enrolling 73 (Radiation Therapy Oncology Group [RTOG]) (RTOG 89-01), 107 (The University of Texas M.D. Anderson Cancer Center), and 333 (European Organisation for Research and Treatment of Cancer [EORTC-08941; NCT00002623]) patients with stage IIIA (N2) disease, no trial reported a difference in local control or survival.[29-31][Level of evidence A1]

   1. In the largest series (EORTC-08941), 579 patients with histologically- or cytologically-proven stage IIIA (N2) NSCLC were given three cycles of platinum-based induction chemotherapy.[31] The 333 responding patients were subsequently randomly assigned to surgical resection or radiation therapy. Of the 154 patients (92%) who underwent surgery, 50% had a radical resection, 42% had a pathological downstaging, and 5% had a pathological complete response; 4% died after surgery. Postoperative (adjuvant) radiation therapy (PORT) was administered to 62 patients (40%) in the surgery arm. Among the 154 patients (93%) who received radiation therapy, overall compliance to the radiation therapy prescription was 55%, and grade 3 to 4 acute and late esophageal and pulmonary toxic effects occurred in 4% and 7% of patients; one patient died of radiation pneumonitis.

      • Median OS was 16.4 months for patients assigned to resection versus 17.5 months for patients assigned to radiation therapy; the 5-year OS rate was 15.7% for patients assigned to resection versus 14% for patients assigned to radiation therapy (HR, 1.06; 95% CI, 0.84–1.35).[31]
      • Rates of PFS were also similar in both groups. In view of its low morbidity and mortality, it was concluded that radiation therapy should be considered the preferred locoregional treatment for these patients.[31]

Adjuvant radiation therapy

The value of PORT has been assessed.[27] Although some studies suggest that PORT can improve local control for node-positive patients whose tumors were resected, it remains controversial whether it can improve survival. The optimal dose of thoracic PORT is not known at this time. Most studies cited used doses ranging from 30 Gy to 60 Gy, typically provided in 2 Gy to 2.5 Gy fractions.[27]

As referred to in the National Cancer Institute of Canada (NCIC) Clinical Trials Group JBR.10 study (NCT00002583), PORT may be considered in selected patients to reduce the risk of local recurrence, if any of the following are present:[23]

• Involvement of multiple nodal stations.
• Extracapsular tumor spread.
• Close or microscopically positive resection margins.

Evidence (adjuvant radiation therapy):

Evidence from one large meta-analysis, subset analyses of randomized trials, and one large population study suggest that PORT may reduce local recurrence. Results from these studies on the effect of PORT on OS are conflicting.

1. A meta-analysis of ten randomized trials that evaluated PORT versus surgery alone showed the following:

   • No difference in OS for the entire PORT group or for the subset of N2 patients.[18][Level of evidence A1]
2. Results from a nonrandomized subanalysis of the ANITA trial, comparing 5-year OS in N2 patients who did or did not receive PORT, found the following:[6]

   • Higher survival rates in patients who received radiation therapy in the observation arm (21% in patients who received PORT vs. 17% in patients who did not receive PORT) and in the chemotherapy arm (47% with PORT vs. 34% without PORT); however, statistical tests of comparison were not conducted.[6]
3. Results from the Surveillance, Epidemiology, and End Results (SEER) Program [28] suggest the following:

   • The large SEER retrospective study (N = 7,465) found superior survival rates associated with radiation therapy in N2 disease (HR, 0.855; 95% CI, 0.762–0.959).

There is benefit of PORT in stage IIIA (N2) disease, and the role of PORT in early stages of NSCLC should be clarified in ongoing phase III trials. Further analysis is needed to determine whether these outcomes can be modified with technical improvements, better definitions of target volumes, and limitation of cardiac volume in the radiation portals.[18]

Concurrent versus sequential chemoradiation therapy

The results from two randomized trials (including RTOG-9410 [NCT01134861]) and a meta-analysis indicate that concurrent chemotherapy and radiation therapy may provide greater survival benefit, albeit with more toxic effects, than sequential chemotherapy and radiation therapy.[35-37][Level of evidence A1]

Evidence (concurrent vs. sequential chemoradiation therapy):

1. In the first trial, the combination of mitomycin C, vindesine, and cisplatin were given concurrently with split-course daily radiation therapy to 56 Gy compared with chemotherapy followed by continuous daily radiation therapy to 56 Gy.[35]

   • Five-year OS rates favored concurrent therapy (27% vs. 9%).
   • Myelosuppression was greater among patients in the concurrent arm, but treatment-related mortality was less than 1% in both arms.[35]
2. In the second trial, 610 patients were randomly assigned to sequential chemotherapy with cisplatin and vinblastine followed by 63 Gy of radiation therapy, concurrent chemoradiation therapy using the same regimen, or concurrent chemotherapy with cisplatin and etoposide with twice-daily radiation therapy.[37]

   • Median and 5-year survival were superior in the concurrent chemotherapy with daily radiation therapy arm (17 months vs. 14.6 months and 16% vs. 10% for sequential regimen [P = .046]).[37]
3. Two smaller studies also reported OS results that favored concurrent over sequential chemotherapy and radiation, although the results did not reach statistical significance.[36,38][Level of evidence A1]
4. A meta-analysis of three trials evaluated concurrent versus sequential treatment (711 patients).[34]

   • The analysis indicated a significant benefit of concurrent over sequential treatment (RR, 0.86; 95% CI, 0.78–0.95; P = .003). All studies used cisplatin-based regimens and once-daily radiation therapy.[34]
   • More deaths (3% OS rate) were reported in the concurrent arm, but this did not reach statistical significance (RR, 1.60; CI, 0.75–3.44; P = .2).
   • There was more acute esophagitis (grade 3 or worse) with concurrent treatment (range, 17%–26%) compared with sequential treatment (range, 0%–4%; RR, 6.77; P = .001). Overall, the incidence of neutropenia (grade 3 or worse) was similar in both arms.

Radiation therapy dose escalation for concurrent chemoradiation

With improvement in radiation therapy–delivery technology in the 1990s, including tumor-motion management and image guidance, phase I/II trials demonstrated the feasibility of dose-escalation radiation therapy to 74 Gy with concurrent chemotherapy.[39-41] However, a phase III trial of a conventional dose of 60 Gy versus dose escalation to 74 Gy with concurrent weekly carboplatin/paclitaxel did not demonstrate improved local control or PFS, and OS was worse with dose escalation (HR, 1.38; 95% CI, 1.09–1.76; P = .004). There was a nonsignificant increase in grade 5 events with dose escalation (10% vs. 2%) and higher incidence of grade 3 esophagitis (21% vs. 7%; P = .0003). Thus, there is no clear benefit in radiation dose escalation beyond 60 Gy for stage III NSCLC.[42][Level of evidence A1]

Consolidation therapy following concurrent chemoradiation

Evidence (consolidation therapy following concurrent chemoradiation):

1. The randomized phase III PROCLAIM study [NCT00686959] enrolled 598 patients with newly diagnosed, stage IIIA/B, unresectable, nonsquamous NSCLC.[43] Patients were randomly assigned on a 1:1 ratio to either of two arms:

   • Arm A: Pemetrexed (500 mg/m²) and cisplatin (75 mg/m²) intravenously every 3 weeks for three cycles plus concurrent thoracic radiation therapy (60 to 66 Gy) followed by pemetrexed consolidation every 3 weeks for four cycles.
   • Arm B: Standard therapy with etoposide (50 mg/m²) and cisplatin (50 mg/m²) intravenously every 4 weeks for two cycles plus concurrent thoracic radiation therapy (60 to 66 Gy) followed by two cycles of consolidation platinum-based doublet chemotherapy.
   The primary objective was OS. The study was designed as a superiority trial with 80% power to detect an OS HR of 0.74 with a type 1 error of .05. This study randomly assigned 598 patients (arm A, 301; arm B, 297) and treated 555 patients (arm A, 283; arm B, 272).

   • Enrollment was stopped early because of futility.
   • OS in arm A was not superior to arm B (HR, 0.98; 95% CI, 0.79–1.20; median, 26.8 vs. 25.0 months; P = .831).
   • Arm A had a significantly lower incidence of any drug-related grade 3 to 4 adverse events (64.0% vs. 76.8%; P = .001), including neutropenia (24.4% vs. 44.5%; P < .001), during the overall treatment period.

Locally advanced unresectable tumors

Radiation therapy alone may provide benefit to patients with locally advanced unresectable stage IIIA NSCLC.

Radiation therapy with traditional dose and fractionation schedules (1.8–2.0 Gy per fraction per day to 60–70 Gy in 6–7 weeks) results in reproducible long-term survival benefit in 5% to 10% of patients and significant palliation of symptoms.[51]

Evidence (radiation therapy for locally advanced unresectable tumor):

1. One prospective randomized clinical study showed the following:[52]

   • Radiation therapy given continuously (including weekends) as three daily fractions (continuous hyperfractionated accelerated radiation therapy) improved OS compared with radiation therapy given as one daily fraction.[52][Level of evidence A1]
   • Patterns of failure for patients treated with radiation therapy alone included both locoregional and distant failures.

Although patients with unresectable stage IIIA disease may benefit from radiation therapy, long-term outcomes have generally been poor because of local and systemic relapse.

Palliative treatment

Radiation therapy may be effective in palliating symptomatic local involvement with NSCLC, such as:

• Tracheal, esophageal, or bronchial compression.
• Pain.
• Vocal cord paralysis.
• Hemoptysis.
• Superior vena cava syndrome.

In some cases, endobronchial laser therapy and/or brachytherapy has been used to alleviate proximal obstructing lesions.[53]

Evidence (radiation therapy for palliative treatment):

1. A systematic review identified six randomized trials of high-dose rate endobronchial brachytherapy (HDREB) alone or with external-beam radiation therapy (EBRT) or laser therapy.[54]

   • Better overall symptom palliation and fewer re-treatments were required in previously untreated patients using EBRT alone.[54][Level of evidence A3]
   • Although EBRT is frequently prescribed for symptom palliation, there is no consensus about when the fractionation scheme should be used.
   • For EBRT, different multifraction regimens appear to provide similar symptom relief;[55-60] however, single-fraction radiation therapy may be insufficient for symptom relief compared with hypofractionated or standard regimens, as seen in the NCIC Clinical Trials Group trial (NCT00003685).[57][Level of evidence A3]
   • Evidence of a modest increase in survival in patients with better performance status given high-dose EBRT is available.[55,56][Level of evidence A1]
   • HDREB provided palliation of symptomatic patients with recurrent endobronchial obstruction previously treated by EBRT, when it was technically feasible.

Radiation therapy dose escalation for concurrent chemoradiation

With improvement in radiation therapy–delivery technology in the 1990s, including tumor-motion management and image guidance, phase I/II trials demonstrated the feasibility of dose-escalation radiation therapy to 74 Gy with concurrent chemotherapy.[39-41] However, a phase III trial of a conventional dose of 60 Gy versus dose escalation to 74 Gy with concurrent weekly carboplatin/paclitaxel did not demonstrate improved local control or PFS, and OS was worse with dose escalation (HR, 1.38 [1.09–1.76]; P = .004). There was a nonsignificant increase in grade 5 events with dose escalation (10% vs. 2%) and higher incidence of grade 3 esophagitis (21% vs. 7%; P = .0003). Thus, there is no clear benefit in radiation dose escalation beyond 60 Gy for stage III NSCLC.[42][Level of evidence A1]

Radiation therapy dose escalation for concurrent chemoradiation

With improvement in radiation therapy–delivery technology in the 1990s, including tumor-motion management and image guidance, phase I/II trials demonstrated the feasibility of dose-escalation radiation therapy to 74 Gy with concurrent chemotherapy.[12-14] However, a phase III trial of a conventional dose of 60 Gy versus dose escalation to 74 Gy with concurrent weekly carboplatin/paclitaxel did not demonstrate improved local control or progression-free survival (PFS), and OS was worse with dose escalation (HR, 1.38 [1.09–1.76]; P = .004). There was a nonsignificant increase in grade 5 events with dose escalation (10% vs. 2%) and higher incidence of grade 3 esophagitis (21% vs. 7%; P = .0003).[15][Level of evidence A1]

Systemic consolidation therapy before or after concurrent chemoradiation therapy

For treatment of locally advanced unresectable tumor in patients who are not candidates for chemotherapy

Radiation therapy alone may provide benefit to patients with locally advanced unresectable stage III NSCLC.

Radiation therapy with traditional dose and fractionation schedules (1.8–2.0 Gy per fraction per day to 60–70 Gy in 6–7 weeks) results in reproducible long-term survival benefit in 5% to 10% of patients and significant palliation of symptoms.[22]

Evidence (radiation therapy for locally advanced unresectable tumor):

1. One prospective randomized clinical study showed the following:

   • Radiation therapy given as three daily fractions improved OS compared with radiation therapy given as one daily fraction.[23][Level of evidence A1]
   • Patterns of failure for patients treated with radiation therapy alone included both locoregional and distant failures.

For patients requiring palliative treatment

Radiation therapy may be effective in palliating symptomatic local involvement with NSCLC, such as:

• Tracheal, esophageal, or bronchial compression.
• Pain.
• Vocal cord paralysis.
• Hemoptysis.
• Superior vena cava syndrome.

In some cases, endobronchial laser therapy and/or brachytherapy has been used to alleviate proximal obstructing lesions.[24]

Evidence (radiation therapy for palliative treatment):

1. A systematic review identified six randomized trials of high-dose rate endobronchial brachytherapy (HDREB) alone or with external-beam radiation therapy (EBRT) or laser therapy.[25]

   • Better overall symptom palliation and fewer re-treatments were required in previously untreated patients using EBRT alone.[25][Level of evidence A3]
   • HDREB provided palliation of symptomatic patients with recurrent endobronchial obstruction previously treated by EBRT, when it was technically feasible.
   • Although EBRT is frequently prescribed for symptom palliation, there is no consensus about when the fractionation scheme should be used.
   • Although different multifraction regimens appear to provide similar symptom relief,[26-31] single-fraction radiation may be insufficient for symptom relief compared with hypofractionated or standard regimens, as shown in the National Cancer Institute of Canada Clinical Trials Group trial (NCT00003685).[28][Level of evidence A3]
   • Evidence of a modest increase in survival in patients with better performance status given high-dose radiation therapy is available.[26,27][Level of evidence A1]

Patients with stages IIIB or IIIC disease with poor performance status are candidates for chest radiation therapy to palliate pulmonary symptoms (e.g., cough, shortness of breath, hemoptysis, or pain).[22][Level of evidence C1] For more information, see Cardiopulmonary Syndromes and Cancer Pain.

Combination chemotherapy

The type and number of chemotherapy drugs to be used for the treatment of patients with advanced NSCLC has been extensively evaluated in randomized controlled trials and meta-analyses.

Several randomized trials have evaluated various drugs combined with either cisplatin or carboplatin in previously untreated patients with advanced NSCLC. On the basis of meta-analyses of the trials, the following conclusions can be drawn:

• Certain three-drug combinations that add so-called targeted agents may result in superior survival.
• EGFR inhibitors may benefit selected patients with EGFR mutations.
• Maintenance chemotherapy after four cycles of platinum combination chemotherapy may improve PFS and OS.
• Platinum combinations with vinorelbine, paclitaxel, docetaxel, gemcitabine, irinotecan, protein-bound paclitaxel, and pemetrexed yield similar improvements in survival. Types and frequencies of toxic effects differ, and these may determine the preferred regimen for an individual patient. Patients with adenocarcinoma may benefit from pemetrexed.
• Cisplatin and carboplatin yield similar improvements in outcome with different toxic effects. Some, but not all, trials and meta-analyses of trials suggest that outcomes with cisplatin may be superior, although with a higher risk of certain toxicities such as nausea and vomiting.
• Nonplatinum combinations offer no advantage to platinum-based chemotherapy, and some studies demonstrate inferiority.
• Three-drug combinations of the commonly used chemotherapy drugs do not result in superior survival and are more toxic than two-drug combinations.

Evidence (combination chemotherapy):

1. The Cochrane Collaboration reviewed data from all randomized controlled trials published between January 1980 and June 2006, comparing a doublet regimen with a single-agent regimen or comparing a triplet regimen with a doublet regimen in patients with advanced NSCLC.[24] Sixty-five trials (13,601 patients) were identified.

   • In the trials that compared a doublet regimen with a single-agent regimen, a significant increase was observed in tumor response (odds ratio [OR], 0.42; 95% CI, 0.37–0.47; P < .001) and 1-year survival (OR, 0.80; 95% CI, 0.70–0.91; P < .001) in favor of the doublet regimen. The absolute benefit in 1-year survival was 5%, which corresponds to an increase in 1-year survival from 30% with a single-agent regimen to 35% with a doublet regimen. The rates of grades 3 and 4 toxic effects caused by doublet regimens were statistically increased compared with rates after single-agent therapy, with ORs ranging from 1.2 to 6.2. Infection rates did not increase in doublet regimens.
   • There was no increase in 1-year survival (OR, 1.01; 95% CI, 0.85–1.21; P = .88) for triplet regimens versus doublet regimens. The median survival ratio was 1.00 (95% CI, 0.94–1.06; P = .97).
2. Several meta-analyses have evaluated whether cisplatin or carboplatin regimens are superior, with variable results.[25-27] One meta-analysis reported individual patient data for 2,968 patients entered in nine randomized trials.[25]

   • The objective response rate was higher for patients treated with cisplatin (30%) than for patients treated with carboplatin (24%); (OR, 1.37; 95% CI, 1.16–1.61; P < .001).
   • Carboplatin treatment was associated with a nonstatistically significant increase in the hazard of mortality relative to treatment with cisplatin (HR, 1.07; 95% CI, 0.99–1.15; P = .100).
   • In patients with nonsquamous cell tumors and in patients treated with third-generation chemotherapy, carboplatin-based chemotherapy was associated with a statistically significant increase in mortality (HR, 1.12; 95% CI, 1.01–1.23 in patients with nonsquamous cell tumors and HR, 1.11; 95% CI, 1.01–1.21 in patients treated with third-generation chemotherapy).
   • Treatment-related toxic effects were also assessed in the meta-analysis. More thrombocytopenia was seen with carboplatin than with cisplatin (12% vs. 6%; OR, 2.27; 95% CI, 1.71–3.01; P < .001), but cisplatin caused more nausea and vomiting (8% vs. 18%; OR, 0.42; 95% CI, 0.33–0.53; P < .001) and renal toxic effects (0.5% vs. 1.5%; OR, 0.37; 95% CI, 0.15–0.88; P = .018).
   • The authors concluded that treatment with cisplatin was not associated with a substantial increase in the overall risk of severe toxic effects. This comprehensive individual-patient meta-analysis is consistent with the conclusions of other meta-analyses that were based on essentially the same clinical trials, but which used only published data.
3. Three literature-based meta-analyses have trials that compared platinum with nonplatinum combinations.[28-30]

   1. The first meta-analysis identified 37 assessable trials that included 7,633 patients.[28]

      • A 62% increase in the OR for response was attributable to platinum-based therapy (OR, 1.62; 95% CI, 1.46–1.8; P < .001). The 1-year survival rate was increased by 5% with platinum-based regimens (34% vs. 29%; OR, 1.21; 95% CI, 1.09–1.35; P = .003).
      • No statistically significant increase in 1-year survival was found when platinum therapies were compared with third-generation-based combination regimens (OR, 1.11; 95% CI, 0.96–1.28; P = .17).
      • The toxic effects of platinum-based regimens was significantly higher for hematologic toxic effects, nephrotoxic effects, and nausea and vomiting but not for neurological toxic effects, febrile neutropenia rate, or toxic death rate. These results are consistent with the second literature-based meta-analysis.
   2. The second meta-analysis identified 17 trials that included 4,920 patients.[29]

      • The use of platinum-based doublet regimens was associated with a slightly higher survival at 1 year (relative risk [RR], 1.08; 95% CI, 1.01%–1.16%; P = .03) and a better partial response (RR, 1.11; 95% CI, 1.02–1.21; P = .02), with a higher risk of anemia, nausea, and neurological toxic effects.
      • In subanalyses, cisplatin-based doublet regimens improved survival at 1 year (RR, 1.16%; 95% CI, 1.06–1.27; P = .001), complete response (RR, 2.29; 95% CI, 1.08–4.88; P = .03), and partial response (RR, 1.19; 95% CI, 1.07–1.32; P = .002), with an increased risk of anemia, neutropenia, neurological toxic effects, and nausea.
      • Conversely, carboplatin-based doublet regimens did not increase survival at 1 year (RR, 0.95; 95% CI, 0.85–1.07; P = .43).
   3. The third meta-analysis of phase III trials randomizing platinum-based versus nonplatinum combinations as first-line chemotherapy identified 14 trials.[30] Experimental arms were gemcitabine and vinorelbine (n = 4), gemcitabine and taxane (n = 7), gemcitabine and epirubicin (n = 1), paclitaxel and vinorelbine (n = 1), and gemcitabine and ifosfamide (n = 1). This meta-analysis was limited to the set of 11 phase III studies that used a platinum-based doublet (2,298 patients in the platinum-based arm and 2,304 patients in the nonplatinum arm).

      • Patients treated with a platinum-based regimen benefited from a statistically significant reduction in the risk of death at 1 year (OR, 0.88; 95% CI, 0.78–0.99; P = .044) and a lower risk of being refractory to chemotherapy (OR, 0.87; CI, 0.73–0.99; P = .049).
      • Forty-four (1.9%) toxic-related deaths were reported for platinum-based regimens and 29 (1.3%) toxic-related deaths were reported for nonplatinum regimens (OR, 1.53; CI, 0.96–2.49; P = .08). An increased risk of grade 3 to 4 gastrointestinal and hematologic toxic effects for patients treated with platinum-based chemotherapy was statistically demonstrated. There was no statistically significant increase in the risk of febrile neutropenia (OR, 1.23; CI, 0.94–1.60; P = .063).

Drug and dose schedule

Among the active combinations, definitive recommendations regarding drug dose and schedule cannot be made, except for carboplatin, pemetrexed, and pembrolizumab for patients with nonsquamous tumor histology.

Evidence (drug and dose schedule):

1. One meta-analysis of seven trials that included 2,867 patients assessed the benefit of docetaxel versus vinorelbine.[31] Docetaxel was administered with a platinum agent in three trials, with gemcitabine in two trials, or as monotherapy in two trials. Vinca alkaloid (vinorelbine in six trials and vindesine in one trial) was administered with cisplatin in six trials or alone in one trial.

   • The pooled estimate for OS showed an 11% improvement in favor of docetaxel (HR, 0.89; 95% CI, 0.82–0.96; P = .004). Sensitivity analyses that considered only vinorelbine as a comparator or only the doublet regimens showed similar improvements.
   • Grade 3 to 4 neutropenia and grade 3 to 4 serious adverse events were less frequent with docetaxel-based regimens (OR, 0.59; 95% CI, 0.38–0.89; P = .013) versus vinca alkaloid-based regimens (OR, 0.68; 95% CI, 0.55–0.84; P < .001).
2. Two randomized trials compared weekly versus every-3-week dosing of paclitaxel and carboplatin, which reported no significant difference in efficacy and better tolerability for weekly administration.[32,33] Although meta-analyses of randomized controlled trials suggest that cisplatin combinations may be superior to carboplatin or nonplatinum combinations, the clinical relevance of the differences in efficacy must be balanced against the anticipated tolerability, logistics of administration, and familiarity of the medical staff in making treatment decisions for individual patients.
3. A large, noninferiority, phase III randomized study compared the OS in 1,725 chemotherapy-naïve patients with stage IIIB/IV NSCLC and a performance status of 0 to 1.[4] Patients received cisplatin 75 mg/m² on day 1 and gemcitabine 1,250 mg/m² on days 1 and 8 (n = 863) or cisplatin 75 mg/m² and pemetrexed 500 mg/m² on day 1 (n = 862) every 3 weeks for up to six cycles.

   • OS for cisplatin and pemetrexed (median survival, 10.3 months) was noninferior to cisplatin and gemcitabine (median survival, 10.3 months; HR, 0.94; 95% CI, 0.84%–1.05%).
   • In patients with adenocarcinoma (n = 847), OS was statistically superior for cisplatin and pemetrexed (12.6 months) versus cisplatin and gemcitabine (10.9 months); in patients with large cell carcinoma (n = 153), OS was statistically superior for cisplatin and pemetrexed (10.4 months) versus cisplatin and gemcitabine (6.7 months).
   • In contrast, in patients with squamous cell histology (n = 473), there was a significant improvement in survival with cisplatin and gemcitabine (10.8 months) versus cisplatin and pemetrexed (9.4 months). For cisplatin and pemetrexed, rates of grade 3 or 4 neutropenia, anemia, and thrombocytopenia (P ≤ .001); febrile neutropenia (P = .002); and alopecia (P < .001) were significantly lower, whereas grade 3 or 4 nausea (P = .004) was more common.
   • The results of this study suggested that the cisplatin and pemetrexed doublet is another alternative doublet for first-line chemotherapy for advanced NSCLC. The results also suggested that there may be differences in outcome depending on histology.

Bevacizumab

Evidence (bevacizumab):

1. Two randomized trials have evaluated the addition of bevacizumab, an antibody targeting vascular endothelial growth factor, to standard first-line combination chemotherapy.

   1. In a randomized study of 878 patients with recurrent or advanced stage IIIB/IV NSCLC, 444 patients received paclitaxel and carboplatin alone, and 434 patients received paclitaxel and carboplatin plus bevacizumab.[34] Chemotherapy was administered every 3 weeks for six cycles, and bevacizumab was administered every 3 weeks until disease progression was evident or toxic effects were intolerable. Patients with squamous cell tumors, brain metastases, clinically significant hemoptysis, or inadequate organ function or performance status (ECOG performance status >1) were excluded.

      • Median survival was 12.3 months in the group assigned to chemotherapy plus bevacizumab, as compared with 10.3 months in the chemotherapy-alone group (HR_death, 0.79; P = .003).
      • Median PFS was 6.2 months in the group assigned to chemotherapy plus bevacizumab (HR_{disease progression}, 0.66; P < .001), with a 35% response rate (P < .001), and 4.5 months in the chemotherapy-alone group (HR_{disease progression}, 0.66; P < .001), with a 15% response rate (P < .001).
      • Rates of clinically significant bleeding were 4.4% in the group assigned to chemotherapy plus bevacizumab and 0.7% in the chemotherapy-alone group (P < .001). There were 15 treatment-related deaths in the chemotherapy-plus-bevacizumab group, including five from pulmonary hemorrhage.
      • For this subgroup of patients with NSCLC, the addition of bevacizumab to paclitaxel and carboplatin may provide survival benefit.[34][Level of evidence A1]
   2. Another randomized, phase III trial investigated the efficacy and safety of cisplatin-gemcitabine plus bevacizumab.[35] Patients were randomly assigned to receive cisplatin (80 mg/m²) and gemcitabine (1,250 mg/m²) for up to six cycles, plus low-dose bevacizumab (7.5 mg/kg), high-dose bevacizumab (15 mg/kg), or placebo every 3 weeks until disease progression. The primary end point was amended from OS to PFS during the course of the study. A total of 1,043 patients were accrued (placebo group, n = 347; low-dose group, n = 345; high-dose group, n = 351).

      • PFS was significantly prolonged with the addition of bevacizumab; the HRs for PFS were 0.75 in the low-dose group (median PFS, 6.7 months vs. 6.1 months for the placebo group; P = .03) and 0.82 in the high-dose group compared with the placebo group (median PFS, 6.5 months vs. 6.1 months for the placebo group; P = .03).[35][Level of evidence A1]
      • Objective response rates were also improved with the addition of bevacizumab, and they were 20.1% for placebo, 34.1% for low-dose bevacizumab, and 30.4% for high-dose bevacizumab plus cisplatin/gemcitabine.
      • Incidence of grade 3 or greater adverse events was similar across arms.
      • Grade 3 or greater pulmonary hemorrhage rates were 1.5% or less for all arms, despite 9% of patients receiving therapeutic anticoagulation.
      • These results support the addition of bevacizumab to platinum-containing chemotherapy, but the results are far less impressive than when the carboplatin-paclitaxel combination was used.
      • Furthermore, no significant difference in survival was shown in this study, as reported in abstract form.
      • Altogether, these findings may suggest that the backbone of chemotherapy may be important when bevacizumab is added.

Cetuximab

Evidence (cetuximab):

1. Two trials have evaluated the addition of cetuximab to first-line combination chemotherapy.[36,37]

   1. In the first trial, 676 chemotherapy-naïve patients with stage IIIB (pleural effusion) or stage IV NSCLC, without restrictions by histology or EGFR expression, received cetuximab with taxane (paclitaxel or docetaxel with carboplatin) or combination chemotherapy.[36]

      • The addition of cetuximab did not result in a statistically significant improvement in PFS, the primary study end point, or OS.
      • Median PFS was 4.40 months for patients in the cetuximab-chemotherapy arm versus 4.24 months for patients in the taxane-carboplatin arm (HR, 0.902; 95% CI, 0.761–1.069; P = .236).
      • Median OS was 9.69 months for patients in the cetuximab-chemotherapy arm versus 8.38 months for patients in the chemotherapy-alone arm (HR, 0.890; 95% CI, 0.754–1.051; P = .169).
      • No significant associations were found between EGFR expression, EGFR mutation, EGFR copy number, or KRAS mutations and PFS, OS, and response in the treatment-specific analyses.[38]
   2. The second trial was composed of 1,125 chemotherapy-naïve patients with advanced EGFR-expressing stage IIIB/IV NSCLC treated with cisplatin-vinorelbine chemotherapy plus cetuximab or chemotherapy alone.[37]

      • The primary study end point, OS, was longer for patients treated with cetuximab and chemotherapy (median 11.3 months vs. 10.1 months; HR_death, 0.871; 95% CI, 0.762–0.996; P = .044).
      • A survival benefit was seen in all histological subgroups; however, survival benefit was not seen in non-White or Asian patients. Only the interaction between the treatment and the ethnic origin was significant (P = .011).
      • The main cetuximab-related adverse event was acne-like rash (grade 3, 10%).
   3. It is not clear whether the differences in outcome in these two studies are the result of differences in the study populations, tumor characterization for EGFR expression, or chemotherapy regimens.

Necitumumab

Evidence (necitumumab):

1. Two phase III trials have evaluated the addition of the second-generation, recombinant, human immunoglobulin G1 EGFR antibody, necitumumab, to platinum-doublet chemotherapy in the first-line treatment of patients with advanced nonsquamous cell and squamous cell NSCLC.[39,40]

   1. The SQUIRE trial (NCT00981058) randomly assigned 1,093 patients with advanced squamous NSCLC to receive either first-line chemotherapy with cisplatin and gemcitabine or the same regimen with the addition of necitumumab (800 mg on day 1 and day 8 of each cycle).[40]

      • Median OS was prolonged with the addition of necitumumab (11.5 months vs. 9.9 months; P = .01).
      • PFS was also prolonged with the addition of necitumumab (5.7 months vs. 5.5 months); however, the overall response rate was similar in both groups (31% vs. 28%).
      • Grades 3 and 4 adverse events were higher in the necitumumab-containing arm (72% vs. 62%).
      • Necitumumab is associated with higher toxicity and relatively modest benefit.
   2. The INSPIRE trial (NCT00982111) randomly assigned 633 patients with advanced nonsquamous NSCLC to receive either first-line chemotherapy with cisplatin and pemetrexed or to cisplatin and pemetrexed with the addition of necitumumab (800 mg on day 1 and day 8 of each cycle).[39]

      • This study showed no benefit from the addition of necitumumab to standard first-line chemotherapy for advanced nonsquamous NSCLC.
      • OS was 11.3 months (95% CI, 9.5–13.4) for patients in the necitumumab-containing arm versus 11.5 months (95% CI, 10.1–13.1) for patients in the chemotherapy alone arm; P = .96. Similarly, there was no difference between the arms in terms of objective response or PFS.
      • Serious adverse events and rates of grades 3 and 4 adverse events, including thromboembolic events, were higher in patients in the necitumumab-containing arm; the incidence of treatment-related deaths was also higher (5% vs. 3%).
      • On the basis of these results, necitumumab is not recommended as combination therapy with standard first-line chemotherapy for patients with advanced nonsquamous NSCLC.

Osimertinib alone

Evidence (osimertinib alone):

1. A phase III, multicenter, randomized, double-blind, controlled trial (FLAURA [NCT02296125]) compared osimertinib, an oral, third-generation, irreversible EGFR TKI that inhibits both EGFR-TKI–sensitizing mutations and the EGFR T790M resistance mutation, with standard of care EGFR TKIs (gefitinib or erlotinib) as first-line treatment of patients with previously untreated, EGFR mutation-positive (exon 19 deletion or L858R), advanced NSCLC, as detected by a U.S. Food and Drug Administration (FDA)-approved test.[49] The 556 patients were randomly assigned in a 1:1 ratio.

   • The primary end point of PFS was significantly longer with osimertinib (18.9 months vs. 10.2 months; HR, 0.46; 95% CI, 0.37–0.57, P < .001).[49][Level of evidence B1]
   • The objective response rate was similar for both groups (80% for the osimertinib group vs. 76% for the standard EGFR TKI group).
   • Central nervous system (CNS) progression was observed less often in the osimertinib group compared with the standard EGFR TKI group (6% vs. 15%).
   • The median duration of response (DOR) was 17.2 months (95% CI, 13.8–22.0) with osimertinib versus 8.5 months (95% CI, 7.3–9.8) with standard EGFR TKIs.
   • OS was a key secondary end point. With a follow-up of at least 39 months in each group, the median OS was 38.6 months (95% CI, 34.5–41.8) in the osimertinib group and 31.8 months (95% CI, 26.6–36.0) in the standard EGFR TKI group (HR_death 0.80; 95.05% CI, 0.64–1.00; P = .046).[50][Level of evidence A1]
   • The crossover rate from the standard EGFR TKI group to the osimertinib group was 31% (85 of 277) among patients assigned to the standard EGFR TKI group and 47% (85 of 180) among patients discontinuing the EGFR TKI. The authors noted that this crossover probably contributed to the duration of OS in the EGFR TKI (31.8 months).
   • No new safety signals were observed. Rates of adverse events of grade 3 or higher and adverse events leading treatment discontinuations were similar between groups. Adverse events leading to dose interruptions, reductions, or permanent discontinuations were 43%, 5%, and 15%, respectively, in the osimertinib group and 41%, 4%, and 18%, respectively, in the EGFR TKI group.

The FDA approved osimertinib for first-line treatment of EGFR-mutant NSCLC (exon 19 deletion or L858R).

Longer PFS and OS, activity against the EGFR T790M mutation in addition to EGFR-TKI−sensitizing mutations, decreased frequency of CNS progression, and good tolerability make osimertinib the preferred choice for treatment of patients with advanced EGFR-mutated NSCLC compared with first- and second-generation EGFR TKIs.

Osimertinib plus chemotherapy

Evidence (osimertinib plus chemotherapy):

1. A multicenter, randomized, open-label, phase III trial (FLAURA2 [NCT04035486]) compared osimertinib plus chemotherapy with osimertinib alone in patients with EGFR-mutated (exon 19 deletion or L858R mutation) advanced NSCLC who had not previously received treatment for advanced disease. Patients received osimertinib (80 mg once daily) with chemotherapy (pemetrexed [500 mg/m² of body-surface area] plus either cisplatin [75 mg/m²] or carboplatin [pharmacologically guided dose, area under the curve (AUC) = 5]) or osimertinib monotherapy (80 mg once daily). Chemotherapy in the combination arm was given for four 21-day cycles and was followed by osimertinib and pemetrexed (500 mg/m²) maintenance every 3 weeks. A total of 557 eligible patients were randomly assigned in a 1:1 ratio. The primary end point was investigator-assessed PFS.[51]

   • PFS was 25.5 months (24.7–not calculable) for patients in the osimertinib-plus-chemotherapy group and 16.7 months (14.1–21.3) for patients in the osimertinib-monotherapy group (HR, 0.62; 95% CI, 0.49–0.79; P < .001).[51][Level of evidence B1] PFS was assessed according to blinded independent central review and was consistent with the primary analysis (HR, 0.62; 95% CI, 0.48–0.80).
   • At 24 months, 57% (95% CI, 50%–63%) of the patients in the osimertinib-plus-chemotherapy group and 41% (95% CI, 35%–47%) of those in the osimertinib-alone group were alive and progression-free.
   • An objective response (complete or partial) occurred in 83% of patients who received osimertinib plus chemotherapy and 76% of patients who received osimertinib alone.
   • The median DOR was 24.0 months (95% CI, 20.9–27.8) in the osimertinib-plus-chemotherapy group and 15.3 months (95% CI, 12.7–19.4), in the osimertinib-alone group.
   • Grade 3 or higher adverse events from any cause were more common with the combination (64%) than with monotherapy (27%); this is consistent with known chemotherapy-related adverse events. Osimertinib plus pemetrexed with a platinum-based agent had a safety profile that was consistent with the established profiles of these agents.
   Analysis of OS, a secondary end point, requires further follow-up (data maturity, 27%).

Dacomitinib

Evidence (dacomitinib):

1. A multicenter, randomized, open-label, phase III trial (ARCHER 1050 [NCT01774721]) compared dacomitinib, a second-generation, irreversible EGFR TKI, administered orally at a dose of 45 mg per day with gefitinib administered orally at a dose of 250 mg per day, as first-line therapy in patients with newly diagnosed advanced NSCLC harboring the following EGFR mutations: exon 19 deletion or exon 21 L858R substitution mutations, as detected by an FDA-approved test.[52] Four hundred and fifty-two eligible patients were randomly assigned in a 1:1 ratio. The primary end point was PFS assessed by masked independent review in the intention-to-treat (ITT) population.

   • Median PFS was 14.7 months in the dacomitinib group and 9.2 months in the gefitinib group (HR, 0.59; 95% CI, 0.47−0.74; P < .0001).[52][Level of evidence B1]
   • The objective response rate was similar between the two groups (75% for the dacomitinib group vs. 72% for the gefitinib group; P = 0.42).
   • The median DOR was longer in the dacomitinib group (14.8 months vs. 8.3 months; HR, 0.4; 95% CI, 0.31−0.53; P < .0001).
   • The median OS was 34.1 months with dacomitinib vs. 26.8 months with gefitinib (HR, 0.76; 95% CI, 0.58−0.99; P = .44).[52]
   • Grade 3 or higher adverse events of any cause occurred in 63% of patients who received dacomitinib and 41% of patients who received gefitinib. The most common grade 3 or 4 adverse events were dermatitis acneiform (14% in the dacomitinib group vs. none in the gefitinib group), diarrhea (8% vs. 1%), and raised alanine aminotransferase (ALT) levels (1% vs. 8%). Serious treatment-related adverse events were more frequent in the dacomitinib group (9% vs. 4%). Permanent discontinuation of the study drug because of treatment-related adverse events occurred more often in the dacomitinib group (10% vs. 7%). Dose reductions were also more frequent in the dacomitinib group (66% vs. 8%).

The FDA approved dacomitinib for first-line treatment of patients with metastatic NSCLC with EGFR exon 19 deletion or exon 21 L858R substitution mutations as detected by an FDA-approved test.

Gefitinib

Evidence (gefitinib):

1. A phase III, multicenter, randomized trial compared gefitinib with carboplatin plus paclitaxel as first-line treatment in clinically selected patients in East Asia who had advanced adenocarcinoma of the lung and had never smoked or were former light smokers.[53]

   • The study met its primary objective of demonstrating the superiority of gefitinib compared with the carboplatin-paclitaxel combination for PFS (HR_{progression or death}, 0.74; 95% CI, 0.65–0.85; P < .001).
   • The median PFS was 5.7 months in the gefitinib group and 5.8 months in the carboplatin-paclitaxel group.[53][Level of evidence B1]
   • Following the time that chemotherapy was discontinued and while gefitinib was continued, the PFS curves clearly separated and favored gefitinib.
   • The 12-month PFS rates were 24.9% with the gefitinib group and 6.7% with the carboplatin-paclitaxel group.
   • More than 90% of the patients in the trial with mutations had either del19 or exon 21 L858R mutations, which have been shown to be sensitive to EGFR inhibitors. In the subgroup of patients with a mutation, PFS was significantly longer among those who received gefitinib (HR, 0.48; 95% CI, 0.36–0.64; P < .001); however, in the subgroup of patients who were negative for a mutation, PFS was significantly longer in those who received the carboplatin-paclitaxel combination (HR with gefitinib, 2.85; 95% CI, 2.05–3.98; P < .001). There was a significant interaction between treatment and EGFR mutation with respect to PFS (P < .001).[53]
   • OS was similar for patients who received gefitinib and carboplatin-paclitaxel, with no significant difference between treatments overall (HR, 0.90; 95% CI, 0.79–1.02; P = .109) or in EGFR mutation–positive (HR, 1.00; 95% CI, 0.76–1.33; P = .990) or EGFR mutation–negative (HR, 1.18; 95% CI, 0.86–1.63; P = .309; treatment by EGFR mutation interaction P = .480) subgroups. A high proportion (64.3%) of EGFR mutation–positive patients randomly assigned to the carboplatin-paclitaxel regimen received subsequent EGFR TKIs. PFS was significantly longer with gefitinib for patients whose tumors had both high EGFR gene copy number and EGFR mutation (HR, 0.48; 95% CI, 0.34–0.67) but significantly shorter when high EGFR gene copy number was not accompanied by EGFR mutation (HR, 3.85; 95% CI, 2.09–7.09).
2. A phase III trial from Japan prospectively confirmed that patients with NSCLC and EGFR mutations have improved PFS but not OS when treated with gefitinib. The trial included 230 chemotherapy-naïve patients with metastatic NSCLC and EGFR mutations who were randomly assigned to receive gefitinib or carboplatin-paclitaxel.[54]

   • In the planned interim analysis of data for the first 200 patients, PFS was significantly longer in the gefitinib group than in the standard-chemotherapy group (HR_death or disease progression with gefitinib, 0.36; P < .001), resulting in early termination of the study.
   • The gefitinib group had a significantly longer median PFS (10.8 months vs. 5.4 months in the chemotherapy group; HR, 0.30; 95% CI, 0.22–0.41; P < .001).[54][Level of evidence B1] The median OS was 30.5 months in the gefitinib group and 23.6 months in the standard chemotherapy group (P = .31).
3. Another phase III trial from Japan also prospectively confirmed that patients with NSCLC and EGFR mutations have improved PFS but not OS when treated with gefitinib. In the second trial, the West Japanese Oncology Group conducted a phase III study (WJTOG3405) in 177 chemotherapy-naïve patients aged 75 years or younger and diagnosed with stage IIIB/IV NSCLC or postoperative recurrence harboring EGFR mutations (either the exon 19 deletion or L858R-point mutation).[55]

   • Patients were randomly assigned to receive either gefitinib or cisplatin plus docetaxel (administered every 21 days for three to six cycles). The primary end point was PFS.
   • The gefitinib group had significantly longer PFS than the cisplatin-plus-docetaxel group, with a median PFS of 9.2 months (95% CI, 8.0–13.9) versus 6.3 months (range, 5.8–7.8 months; HR, 0.489; 95% CI, 0.336–0.710, log-rank P < .0001).[55][Level of evidence B1]

Erlotinib

Evidence (erlotinib):

1. In an open-label, randomized, phase III trial (NCT00874419) from China, 165 patients older than 18 years with histologically confirmed stage IIIB/IV NSCLC and a confirmed activating mutation of EGFR (i.e., exon 19 deletion or exon 21 L858R-point mutation) received either oral erlotinib (150 mg/day) until they experienced disease progression or unacceptable toxic effects, or up to four cycles of gemcitabine plus carboplatin.[56]

   • Median PFS was significantly longer in erlotinib-treated patients than in patients treated with chemotherapy (13.1 months [95% CI, 10.58–16.53] vs. 4.6 months [range, 4.21–5.42 months]; HR, 0.16; 95% CI, 0.10–0.26; P < .0001).[56][Level of evidence B1]
2. In a European study (EURTAC [NCT00446225]), 1,227 patients with advanced NSCLC were screened for EGFR mutations. Of these, 174 patients with EGFR mutations were randomly assigned to receive erlotinib or platinum-based chemotherapy.[57] The primary end point was PFS.

   • In an interim analysis of the first 153 patients, PFS in the chemotherapy arm was 5.2 months (95% CI, 4.5–5.8) compared with 9.7 months (95% CI, 8.4–12.3) in the erlotinib arm (HR, 0.37; P < .0001). Median survival was 19.3 months in patients in the chemotherapy arm and 19.5 months in patients in the erlotinib arm (HR, 0.80; P = .42).[58][Level of evidence B1]

Afatinib

Evidence (afatinib):

1. In an open-label, randomized, phase III study (LUX-Lung 3 [NCT00949650]), 345 Asian (72%) and White (26%) patients with stage IIIB/IV NSCLC and confirmed EGFR mutations (i.e., exon 19 deletion, L858R, or other [38 of 345 patients had other less-common mutations]) were screened, and 340 patients received at least one dose of study medication, which was either 40 mg of oral afatinib, an irreversible EGFR/human epidermal receptor TKI, daily or up to six cycles of cisplatin and pemetrexed for first-line treatment.[59]

   1. The primary end point was PFS. In this study, the afatinib group had significantly longer PFS than the cisplatin-plus-pemetrexed group, with a median PFS of 11.1 months for afatinib and 6.9 months for chemotherapy (HR, 0.58; 95% CI, 0.43–0.78; P = .001).[59][Level of evidence B1]
   2. Assessment of OS was a secondary end point and was reported separately.[60] Similar to the PFS analysis, OS was stratified on the basis of EGFR-mutation type and ethnic origin.

      • With a median follow-up of 41 months, median OS was 28.2 months in patients in both arms (HR, 0.88; 95% CI, 0.66–1.17; P = .39).
      • In patients harboring common EGFR mutations (i.e., exon 19 deletion and L858R), survival did not differ significantly between treatment arms (HR, 0.78; 95% CI, 0.58–1.06; P = .11). However, prespecified subgroup analyses demonstrated a survival advantage with afatinib compared with chemotherapy in patients with tumors harboring the EGFR del19 mutation (median OS, 33.3 months vs. 21.1 months; HR, 0.54; 95% CI, 0.36–0.79; P = .0015) but no significant difference between treatment arms in patients with tumors harboring the L858R mutation (median OS, 27.6 months vs. 40.3 months; HR, 1.30; 95% CI, 0.80–2.11; P = .29).
      • First-line afatinib was associated with a significant survival advantage compared with chemotherapy in patients with NSCLC-harboring EGFR del19 mutations but not in patients with EGFR L858R mutations or in the overall EGFR–mutation-positive patient population.[60][Level of evidence A1]
2. In an open-label, randomized, phase III study (LUX-Lung 6 [NCT01121393]), 364 East Asian patients with stage IIIB/IV NSCLC and confirmed EGFR mutations (i.e., exon 19 deletion, L858R, or other) were randomly assigned (in a 2:1 ratio) to 40 mg of afatinib daily or gemcitabine and cisplatin for up to six cycles for first-line treatment.[61]

   1. The primary end point was PFS. Median PFS was significantly longer in the afatinib group (11.0 months; 95% CI, 9.7–13.7) than in the gemcitabine and cisplatin group (5.6 months, [range, 5.1–6.7 months]; HR, 0.28; 95% CI, 0.20–0.39; P < .0001).[61][Level of evidence B1]
   2. Assessment of OS was a prespecified secondary end point and was reported separately.[60] Similar to the PFS analysis, OS was stratified on the basis of EGFR-mutation type and ethnic origin.

      • With a median follow-up of 33 months, median OS was 23.1 months in patients in the afatinib arm and 23.5 months in patients in the chemotherapy arm (HR, 0.93; 95% CI, 0.72–1.22; P = .61).
      • In patients harboring common EGFR mutations (i.e., exon 19 deletion and L858R), survival did not differ significantly between treatment arms (HR, 0.83; 95% CI, 0.62–1.09; P = .18). However, prespecified subgroup analyses demonstrated a survival advantage with afatinib compared with chemotherapy in patients with tumors harboring the EGFR del19 mutation (median OS, 31.4 months vs. 18.4 months; HR, 0.64; 95% CI, 0.44–0.94; P = .023), but no significant difference between treatment arms was seen in patients with tumors harboring the L858R mutation (median OS, 19.6 months vs. 24.3 months; HR, 1.22; 95% CI, 0.81–1.83; P = .34).
      • First-line afatinib was associated with a significant survival advantage compared with chemotherapy in patients with NSCLC-harboring EGFR del19 mutations but not in patients with EGFR L858R mutations or in the overall EGFR-mutation-positive patient population.[60][Level of evidence A1]

Amivantamab

Amivantamab has been previously approved for patients with locally advanced or metastatic NSCLC harboring EGFR exon 20 insertion mutations whose disease has progressed on or after platinum-based chemotherapy.

Evidence (amivantamab plus chemotherapy):

1. PAPILLON (NCT04538664) was a phase III randomized trial that compared amivantamab plus chemotherapy with chemotherapy alone as first-line treatment for patients with advanced NSCLC and EGFR exon 20 insertions. Patients in the chemotherapy-alone group who had disease progression were allowed to cross over to receive amivantamab monotherapy. A total of 308 patients were randomly assigned 1:1 to receive either amivantamab plus carboplatin and pemetrexed or carboplatin plus pemetrexed. The primary end point was PFS according to blinded independent central review. The median follow-up was 14.9 months.[62][Level of evidence B1]

   • At 18 months, the PFS rate was 31% for patients who received amivantamab plus chemotherapy, and 3% for patients who received chemotherapy alone.
   • The objective response rate was higher in the amivantamab-plus-chemotherapy group (73%) than the chemotherapy-alone group (47%).
   • In an interim OS analysis, there was no statistically significant difference.
   • The most common adverse events with amivantamab plus chemotherapy were hematologic toxicities, rash, and paronychia. Infusion reactions occurred in 42% of patients.
   The study supports amivantamab plus chemotherapy as an effective first-line treatment option for patients with NSCLC and EGFR exon 20 insertions based on superior PFS when compared with chemotherapy alone.[62]

Alectinib

Evidence (alectinib):

1. In an open-label, randomized, phase III study (the ALEX trial [NCT02075840]), 303 patients with previously untreated, advanced ALK-rearranged NSCLC received either alectinib (600 mg twice a day) or crizotinib (250 mg twice a day).[63] The primary end point was investigator-assessed PFS.

   • The rate of PFS was significantly higher with alectinib than crizotinib; the 12-month event-free survival rate was 68.4% for the alectinib group (95% CI, 61.0%–75.9%) compared with 48.7% for the crizotinib group (95% CI, 40.4%–56.9%) (HR, 0.47; 95% CI, 0.34–0.65; P < .001). The median PFS was not reached with alectinib. The results of independent review committee-assessed PFS were consistent.[63][Level of evidence B1]
   • CNS progression events were less frequent with alectinib (12%) than with crizotinib (45%) (HR, 0.16; 95% CI, 0.10–0.28; P <.001).
   • The response rate was similar for both groups, 82.9% for the alectinib group compared with 75.5% for the crizotinib group (P = .09).
   • Grade 3 to 5 adverse events were less frequent with alectinib (41%) than with crizotinib (50%).
2. A second, open-label, randomized, phase III trial (J-ALEX) recruited 207 ALK-inhibitor–naïve Japanese patients with ALK-positive NSCLC who were chemotherapy-naïve or had received one previous chemotherapy regimen. Patients were randomly assigned in a 1:1 ratio to receive alectinib (300 mg twice daily, which is the dose approved in Japan and is lower than the 600 mg twice daily dose approved elsewhere) versus crizotinib (250 mg twice daily).[64] The primary end point was PFS-assessed by an independent review committee.

   • At data cutoff for the second primary interim analysis, the independent data monitoring committee determined that the primary end point was met (HR, 0.34; 99.7% CI, 0.17–0.71; P <.0001) and recommended immediate release of the data. Median PFS had not been reached with alectinib but was reached at 10.2 months with crizotinib.
   • Grade 3 or 4 adverse events occurred less frequently with alectinib (26% occurrence rate) than with crizotinib (52% occurrence rate).

Lorlatinib

Evidence (lorlatinib):

1. The phase III CROWN trial (NCT03052608) included patients with advanced ALK-rearranged NSCLC who had received no prior systemic therapy for metastatic disease. The trial randomly assigned 296 patients to receive either lorlatinib (100 mg daily) or crizotinib (250 mg twice daily). The primary end point was PFS as determined by blinded independent central review.[65][Level of evidence B1]

   • A planned interim analysis was conducted after approximately 75% of expected events of disease progression or death had occurred. The percentage of patients alive without progression at 12 months was 78% (95% CI, 70%–84%) in the lorlatinib group and 39% (95% CI, 30%–48%) in the crizotinib group. Objective responses occurred in 76% of patients who received lorlatinib and 58% of patients who received crizotinib.
   • Post hoc analysis showed improved PFS for the lorlatinib group, compared with the crizotinib group in patients with and without brain metastases at baseline (12-month PFS rates: 78% vs. 22% and 78% vs. 45%, respectively).[66]
   • There was a lower 12-month cumulative incidence of CNS progression with lorlatinib, compared with crizotinib in patients with (7% vs. 72%) and without (1% vs. 18%) brain metastases at baseline.
   • Lorlatinib was associated with more grade 3 to 4 adverse events than crizotinib (72% vs. 56%), the most common being altered lipid levels. Treatment discontinuation occurred in 7% of patients who received lorlatinib and 9% of patients who received crizotinib.[65]
   • Updated safety data in the post hoc analysis showed that 35% of patients had CNS adverse events with lorlatinib, most of grade 1 severity. These included memory impairment and mood effects, including anxiety, depression, and lability. The occurrence of CNS adverse events did not result in a clinically meaningful difference in patient-reported quality of life. At analysis, 56% of CNS adverse effects had resolved (33% without intervention; 17% with lorlatinib dose modification), and 38% were unresolved. Lorlatinib dose modification did not notably influence PFS.

The FDA approved lorlatinib for patients with metastatic NSCLC whose tumors are ALK-positive, as detected by an FDA-approved test.

Crizotinib

Evidence (crizotinib):

1. In an open-label, randomized, phase III study, 343 patients with stage IIIB/IV NSCLC harboring translocations in ALK received either 250 mg of crizotinib orally twice a day or the combination of pemetrexed and cisplatin or carboplatin for up to six cycles.[67] At the time of disease progression, patients on the chemotherapy arm were allowed to cross over to crizotinib; 60% of patients in the chemotherapy arm subsequently received crizotinib. The primary end point of this study was PFS.

   • The study met its primary end point and demonstrated that crizotinib is superior to chemotherapy in prolonging PFS (median, 10.9 months vs. 7.0 months; HR, 0.454; 95% CI, 0.346–0.596; P < .0001).[68][Level of evidence B1]

Ceritinib

Evidence (ceritinib):

1. In an open-label, randomized, phase III study, 376 patients with stage IIIB/IV ALK-rearranged nonsquamous NSCLC received either oral ceritinib 750 mg daily or platinum-based chemotherapy (cisplatin or carboplatin and pemetrexed) every 3 weeks for four cycles, followed by maintenance pemetrexed.[69] The primary end point was PFS and crossover from chemotherapy to ceritinib was allowed upon documented progression.

   • Median PFS, assessed by blinded independent review, was 16.6 months in the ceritinib group and 8.1 months in the chemotherapy group (HR, 0.55; 95% CI, 0.42–0.73; P < .00001).
   • The median OS was not reached with ceritinib, and it was 26.2 months with chemotherapy (HR, 0.73; 95% CI, 0.50–1.08; P = .056).[69][Level of evidence B1]

Brigatinib

Evidence (brigatinib):

1. A phase II, open-label trial (NCT02094573) enrolled 222 patients with ALK-translocated locally advanced or metastatic NSCLC who had disease progression after crizotinib treatment. Patients were randomly assigned to receive 90 mg every day (n = 112; 109 treated) or 180 mg every day with a 7-day lead-in at 90 mg every day (n = 110).[70]

   • The primary end point assessed by the investigators was objective response rate. The objective response rate was 45% (97.5% CI, 34%–56%) for patients who received the 90 mg dose and 54% (97.5% CI, 43%–65%) for patients who received the 180 mg dose.
   • Median PFS was 9.2 months (95% CI, 7.4–15.6) for patients who received the 90 mg dose and 12.9 months (95% CI, 11.1–not reached) for patients who received the 180 mg dose.
   • At data cutoff, the median DOR was 13.8 months (95% CI, 5.6–13.8) for patients who received the 90 mg dose and 11.1 months (95% CI, 9.2–13.8) for patients who received the 180 mg dose.[70][Level of evidence B3]
   • The CNS objective response rate in patients with measurable CNS lesions was 42% in patients who received 90 mg every day (n = 26) and 67% in patients who received 180 mg every day (n = 18).
   • Common adverse events, which were mainly grade 1 or 2 and occurred in 27% to 38% of patients at the higher dose, were nausea, diarrhea, headache, and cough. A subset of pulmonary adverse events with early onset (median onset, day 2) occurred in 14 of 219 treated patients (all grades, 6%; grade ≥3, 3%); none occurred after escalation to 180 mg. These events included dyspnea, hypoxia, cough, pneumonia, or pneumonitis. They were managed with dose interruption. Seven of the 14 patients were successfully retreated with brigatinib.
   • The FDA-approved dose of brigatinib is 90 mg every day for 7 days; if tolerated, the dose is increased to 180 mg every day.

Dabrafenib and trametinib

Evidence (dabrafenib and trametinib):

1. In a phase II, multicenter, nonrandomized, open-label study (NCT01336634), 36 patients with previously untreated metastatic NSCLC who tested positive for BRAF V600E mutations were treated with dabrafenib (a BRAF inhibitor) 150 mg twice a day and trametinib (a MEK inhibitor) 2 mg every day.[71] BRAF V600E mutations were identified by the Oncomine Dx Target Test (ThermoFisher Scientific). The primary end point was investigator-assessed overall response.

   • The overall response rate was 64% (95% CI, 46%–79%). Six percent of patients had a complete response, and 58% of patients had a partial response.
   • The median investigator-assessed PFS was 10.9 months (95% CI, 7.0–16.6). The estimated median DOR was 10.4 months (95% CI, 8.3–17.9). At data cutoff, 47% of patients had died, and the median OS was 24.6 months (95% CI, 12.3–not estimable).
   • Sixty-nine percent of patients had at least one grade 3 or 4 adverse event, of which the most common were pyrexia, ALT increase, hypertension, or vomiting. Adverse events led to permanent discontinuation in 22% of patients, dose interruption or delay in 75% of patients, and dose reduction in 39% of patients.[71][Level of evidence C3]

The FDA approved the combination of dabrafenib and trametinib in the treatment of patients with NSCLC whose tumors harbor BRAF V600E mutations as detected by an FDA-approved test.

Entrectinib

The FDA approved entrectinib for treatment of patients with metastatic NSCLC whose tumors are ROS1-positive, regardless of the number of previous systemic therapies.

Evidence (entrectinib):

1. The safety and clinical activity of entrectinib in ROS1 fusion-positive metastatic NSCLC was determined by integrated analysis of three multicenter, single-arm, open-label clinical trials (ALKA-372-001/EudraCT, 2012-000148-88, STARTRK-1 [NCT02097810], and STARTRK-2 [NCT02568267]).[73] Entrectinib was administered orally at a dose of at least 600 mg once daily. Primary end points were the objective response rate and the DOR determined by blinded independent central review. Of note, time-to-event end points are difficult to interpret in the absence of a control arm. Evaluation of tumor samples for the ROS1 gene fusion was conducted prospectively in local laboratories using either a FISH or next-generation sequencing (NGS) laboratory-developed test.
   Seventeen (32%) patients had received no previous systemic therapy, 23 (43%) had received one previous therapy, and 13 (25%) had received two or more lines of treatment. CNS disease was present in 23 (43%) patients at baseline. Thirty-one (59%) patients were never-smokers and 52 (98%) patients had adenocarcinoma histology.

   • The objective response rate in 53 efficacy-evaluable patients was 77% (95% CI, 64%−88%). Six percent of patients had a complete response and 72% had a partial response. Among patients with CNS disease at baseline, the objective response rate was 74% (95% CI, 52%−90%) and all patients had a partial response. Among patients without CNS disease at baseline, the overall response rate was 80% (95% CI, 61%−92%) (10% had a complete response and 70% had a partial response).[73][Level of evidence C3]
   • The median DOR was 24.6 months (95% CI, 11.4−34.8) in efficacy-evaluable patients; 12.6 months (95% CI, 6.5−not estimable) in patients with baseline CNS disease, and 24.6 months (95% CI, 11.4−34.8) in those without CNS disease at baseline.
   • Treatment-related adverse events were assessed in 134 patients in the safety-evaluable population. Grade 1 or 2 treatment-related adverse events were observed in 79 patients (59%). Grade 3 or 4 treatment-related adverse events were observed in 46 patients (34%). Fifteen patients (11%) had serious treatment-related adverse events. There were no treatment-related deaths.
   • The median PFS was 19 months (95% CI, 12.2−36.6) in efficacy-evaluable patients; 13.6 months (95% CI, 4.5−not estimable) in patients with baseline CNS disease, and 26.3 months (95% CI, 15.7−36.6) in patients with no baseline CNS disease.

Crizotinib

Crizotinib was approved for patients with metastatic NSCLC whose tumors are ROS1-positive, regardless of the number of previous systemic therapies.

Evidence (crizotinib):

1. In an expansion cohort of a phase I study of crizotinib, 50 patients with advanced NSCLC who tested positive for ROS1 rearrangement were treated with oral crizotinib 250 mg twice daily.[74] ROS1 rearrangements were identified using break-apart FISH or reverse transcriptase−polymerase chain reaction assay. Seven patients (14%) had not had any previous treatment for advanced disease, 21 patients (42%) had one prior treatment, and 22 patients (44%) had more than one prior treatment. The primary end point was response rate.

   • The overall response rate was 72% (95% CI, 58%–84%). Six percent of patients had a complete response, 66% had a partial response, and 18% had stable disease as their best response.
   • Median PFS was 19.2 months (95% CI, 14.4–not reached). The estimated DOR was 17.6 months (95% CI, 14.5–not reached).[74][Level of evidence C3]
2. In a phase II, open-label, single-arm trial, 127 East Asian patients with ROS1-positive NSCLC were treated with crizotinib 250 mg twice daily.[75] Twenty-four patients (18.9%) had not had any previous treatment for advanced disease, 53 patients (41.7%) had one previous treatment, and 50 patients (39%) had two or three previous treatments. The primary end point was objective response rate by independent review.

   • The objective response rate was 71.7% (95% CI, 63.0%–79.3%). Response rates were similar, irrespective of the number of previous therapies. Complete responses occurred in 13.4% of patients, while 58.3% of patients had partial responses, and 16.5% of patients had stable disease as their best response.[75][Level of evidence C3]
   • Median PFS was 15.9 months (95% CI, 12.9–24). The DOR was 19.7 months (95% CI, 14.1–not reached).
   • OS was 32.5 months (95% CI, 32.5–not reached).

Larotrectinib

Evidence (larotrectinib):

1. Larotrectinib was studied in three protocols: a phase I study involving adults, a phase I/II study involving children, and a phase II study involving adolescents and adults.[78] Fusions were confirmed in the tumors using either FISH or NGS methods. The primary end point for the combined analysis was objective response rate by independent review and was conducted with input from regulators with the goal of excluding a lower bound of less than 30% for response rate. In total, 55 patients with a median age of 45 years (range, 4 months‒76 years) were enrolled across 17 different NTRK fusion-positive tumor types. All patients had either metastatic disease (82%) or locally advanced unresectable disease (18%). Enrolled patients had received a median of two previous systemic therapies.

   • The objective response rate was 75% (95% CI, 61%‒75%) and 73% of these responses lasted at least 6 months.[78][Level of evidence C3]
   • Treatment was well tolerated with 93% of adverse events being grade 1 to 2; the most common grade 3 to 4 adverse events were anemia (11% of patients), transaminitis (7%), and neutropenia (7%).

The FDA approved larotrectinib for the treatment of patients who have locally advanced or metastatic tumors that harbor an NTRK gene fusion without a known acquired resistance mutation, and who have no satisfactory alternative treatments or whose cancer has progressed following treatment.

Entrectinib

The FDA granted accelerated approval to entrectinib for the treatment of solid tumors that have an NTRK gene fusion without a known acquired resistance mutation, are metastatic, have progressed after treatment, have no satisfactory alternative therapy, or for cases in which surgical resection is likely to result in severe morbidity.

Evidence (entrectinib):

1. The safety and clinical activity of entrectinib in NTRK inhibitor-naïve patients with metastatic or locally-advanced solid tumors (including NSCLC) harboring NTRK1, NTRK2, or NTRK3 gene fusions was determined by integrated analysis of three early-phase, multicenter, single-arm, open-label clinical trials (ALKA-372-001/EudraCT, 2012-000148-88, STARTRK-1 [NCT02097810], and STARTRK-2 [NCT02568267]).[79] Treatment consisted of entrectinib administered orally at a dose of at least 600 mg once per day. The primary end points were objective response rate and median DOR, which were assessed by blinded independent central review. Of note, time-to-event end points are difficult to interpret in the absence of a control arm. Identification of positive NTRK gene fusion status was conducted prospectively in local laboratories or a central laboratory using various nucleic acid–based tests.
   Of 54 patients in the NTRK gene fusion-positive efficacy-evaluable population, 20 (37%) had received no previous systemic therapy, 11 (20%) had received one previous systemic therapy, and 23 (43%) had received two or more systemic therapies. Twelve (22%) patients had CNS disease at baseline. Ten (19%) patients had NSCLC. Fifty-two (96%) patients had an NTRK gene fusion detected by NGS and 2 (4%) had an NTRK gene fusion detected by other nucleic acid–based tests.

   • The objective response rate in 54 patients was 57% (95% CI, 43.2%−70.8%). Seven percent of patients had a complete response and 50% had a partial response. In patients with baseline CNS disease, 50% achieved a response (all partial responses), whereas in patients without baseline CNS disease, 60% achieved a response (10% complete response; 50% partial response).[79][Level of evidence C3]
   • The median DOR in efficacy-evaluable patients was 10.4 months (95% CI, 7.1−not estimable). In patients with baseline CNS disease DOR was not estimable, and in patients with no baseline CNS disease it was 12.9 months (95% CI, 7.1−not estimable).
   • Among 10 patients with NSCLC, the response rate was 70% (95% CI, 35%−93%) and DOR ranged between 1.9 months and 20.1 months. For more information, see the prescribing information.
   • The safety-evaluable population consisted of 68 patients with NTRK fusion-positive tumors. Most treatment-related adverse events were grade 1 or 2 and reversible. The most frequent grade 3 or 4 treatment-related adverse events were increased weight gain (10%) and anemia (12%). Serious treatment-related adverse events were reported in 7 (10%) patients. Three (4%) patients had dose interruptions and 27 (40%) patients had dose reductions due to treatment-related adverse events. There were no treatment-related deaths.
   • Median PFS was 11.2 months (95% CI, 8.0−14.9). In patients with baseline CNS disease, median PFS was 7.7 months (95% CI, 4.7−not estimable), and it was 12 months (95% CI, 8.7−15.7) in patients with no baseline CNS disease.

Selpercatinib

Evidence (selpercatinib):

1. A phase I/II study (LIBRETTO-001 [NCT03157128]) enrolled patients with RET fusion−positive solid tumors. RET fusion status was determined by local molecular testing (NGS, FISH, or polymerase chain reaction assay) without central confirmation. The primary end point was objective response.[81][Level of evidence C3]

   • Updated analysis was conducted in 316 patients with RET fusion–positive NSCLC.[81]
   • Among the 69 treatment-naïve patients, the objective response rate was 84% (95% CI, 73%–92%), and 6% achieved complete responses. The median DOR was 20.2 months (95% CI, 13.0–could not be evaluated); 40% of responses were ongoing at the data cutoff (median follow-up, 20.3 months). The median PFS was 22.0 months; 35% of patients were alive and progression-free at the data cutoff (median follow-up, 21.9 months).
   • Among the 247 patients who had received prior platinum-based chemotherapy, the objective response rate was 61% (95% CI, 55%–67%), and 7% achieved complete responses. The median DOR was 28.6 months (95% CI, 20.4–could not be evaluated); 49% of responses were ongoing (median follow-up, 21.2 months). The median PFS was 24.9 months; 38% of patients were alive and progression-free at the data cutoff (median follow-up, 24.7 months).
   • Among the 26 patients with measurable baseline CNS metastasis by the independent review committee, the intracranial objective response rate was 85% (95% CI, 65%–96%), and 27% had complete responses.
   • In the full safety population (n = 796), the median treatment duration was 36.1 months.
   • There was no significant change in the safety profile. Most adverse events were grade 1 to 2. The most common adverse events were edema, diarrhea, fatigue, dry mouth, hypertension, increased ALT and aspartate aminotransferase (AST), and rash.

The FDA approved selpercatinib to treat adults with locally advanced or metastatic NSCLC with RET gene fusion, as detected by an FDA-approved test.

Pralsetinib

Evidence (pralsetinib):

1. A phase I/II study (ARROW [NCT03037385]) enrolled patients with RET fusion−positive solid tumors. Two hundred thirty-three patients had RET fusion−positive NSCLC. RET fusion status was determined by local molecular testing of tumor or circulating tumor nucleic acid (ctDNA) in blood, without central confirmation. The primary end point was objective response.[82][Level of evidence C3]

   • Ninety-two patients who had received platinum-based chemotherapy and 29 patients who were treatment-naive (and not candidates for standard platinum-based treatment) received pralsetinib before the efficacy enrollment cutoff (July 11, 2019). Eighty-seven previously treated patients and 27 treatment-naive patients had centrally adjudicated baseline measurable disease, and thus formed the efficacy cohort.
   • The overall response rate was 61% (95% CI, 50%–71%) in the 87 patients who had received platinum-based chemotherapy, including complete responses in 6%. The median DOR was not reached (15.2 months–not estimable).
   • The overall response rate was 70% (95% CI, 50%–86%) in the 27 treatment-naive patients, including complete responses in 11%. The median DOR was 9.0 months (6.3–not estimable).
   • In the 233-patient safety cohort, 93% had treatment-related adverse events, including 48% with grade 3 or worse events. The most common grade 3 or worse treatment-related adverse events were neutropenia (18%), hypertension (11%), and anemia (10%). Dose reductions occurred in 38% of patients, and 6% discontinued treatment because of adverse events.

Tepotinib

Evidence (tepotinib):

1. An open-label phase II study (VISION [NCT02864992]) enrolled patients with MET exon 14 skipping mutations. The trial included 152 patients who received tepotinib (500 mg orally once daily). MET status was determined centrally, either via liquid biopsy (from circulating free DNA obtained from plasma; n = 66) or via tissue biopsy (n = 60). Twenty-seven patients had positive results from both methods. The primary end point was objective response.[84][Level of evidence C3]

   • Among the 99 patients who had been followed for at least 9 months (i.e., the efficacy population), the objective response rate as assessed by independent review was 46% (95% CI, 36%–57%), with a median DOR of 11.1 months (95% CI, 7.2–not estimable). Response rates were similar in the liquid biopsy and tissue biopsy groups.
   • Responses were similar regardless of prior therapy.
   • Grade 3 or higher adverse events occurred in 28% of patients, including peripheral edema in 7% of patients. Adverse events led to therapy discontinuation in 11% of patients.

Capmatinib

Evidence (capmatinib):

1. A phase II study (GEOMETRY [NCT02414139]) evaluated capmatinib (400 mg orally twice daily) in patients with MET exon 14 skipping mutations or MET amplification. MET status was performed centrally. A total of 364 patients were enrolled. The primary end point was overall response.[85][Level of evidence C3]

   • Of the 69 patients with MET exon 14 skipping mutations who had received one or two prior lines of therapy, the overall response rate was 41% (95% CI, 29%–53%). The median DOR was 9.7 months (95% CI, 5.6–13.0).
   • Of the 28 patients with MET exon 14 skipping mutations who had not received any prior treatment, the overall response rate was 68% (95% CI, 48%–84%). The median DOR was 12.6 months (95% CI, 5.6–not estimable).
   • Response rates in patients with MET amplification without the exon 14 skipping mutation did not meet the prespecified threshold for clinically relevant activity.
   • Grade 3 to 4 adverse events of occurred in 67% of patients. The most common events, regardless of causality, were peripheral edema, nausea, vomiting, and increased creatinine. Adverse events led to therapy discontinuation 11% of patients.

Pembrolizumab plus chemotherapy

Evidence (pembrolizumab plus chemotherapy):

1. A phase III double-blind trial (KEYNOTE-189 [NCT02578680]) randomly assigned, in a 2:1 ratio, 616 patients with metastatic nonsquamous NSCLC without sensitizing EGFR mutations or ALK rearrangements who had received no previous treatment for metastatic disease. Patients received pemetrexed and a platinum-based drug plus either 200 mg of pembrolizumab or placebo every 3 weeks for 4 cycles, followed by pembrolizumab or placebo for up to a total of 35 cycles plus pemetrexed maintenance.[1] Crossover to pembrolizumab monotherapy was permitted after verified progression among patients in the placebo-containing combination group. The primary end points were OS and PFS as assessed by blinded independent central committee radiological review.

   • In the 5-year updated analysis, the median time from random assignment to data cutoff was 64.6 months (range, 60.1–72.4).[87]
   • After 5 years, pembrolizumab plus pemetrexed-platinum was associated with improved OS and PFS, compared with placebo plus pemetrexed-platinum in patients with metastatic nonsquamous NSCLC, regardless of PD-L1 expression. In the ITT population, 5-year OS rates were 19.4% in the pembrolizumab plus pemetrexed-platinum group, compared with 11.3% in the placebo plus pemetrexed-platinum group.
   • Survival was higher in patients with a higher PD-L1 tumor proportion score (TPS), especially in the TPS >50% subgroup (29.6% vs. 21.4%).
   • There were 57 patients who completed 35 cycles of pembrolizumab. For these patients, the objective response rate was 86.0% and the 3-year OS rate after completing 35 cycles (approximately 5 years after random assignment) was 71.9%.[87]
   • Immune-mediated adverse events and infusion reactions occurred in 113 (27.9%) and 27 (13.4%) patients.
   • Adverse events of grade 3 or higher occurred with similar frequency in both treatment groups (71.9% in the pembrolizumab combination group vs. 66.8% in the placebo combination group).
2. A phase III, randomized, double-blind study (KEYNOTE-407 [NCT02775435]) included previously untreated patients with metastatic squamous cell NSCLC. Patients were randomly assigned 1:1 to receive pembrolizumab 200 mg or placebo plus carboplatin and paclitaxel/nab-paclitaxel once every 3 weeks for four cycles, followed by pembrolizumab or placebo for up to 35 cycles (pembrolizumab-plus-chemotherapy, n = 5,278; placebo-plus-chemotherapy, n = 5,281). Primary end points were OS and PFS per RECIST version 1.1 by blinded independent central review.[88]

   • The median time from random assignment to data cutoff was 56.9 months (range, 49.9–66.2). OS and PFS were improved with pembrolizumab-plus-chemotherapy versus placebo-plus-chemotherapy (HR, 0.71 [0.59–0.85] and 0.62 [0.52–0.74]), respectively; 95% CI). The 5-year OS rates were 18.4% and 9.7%, respectively.[88][Level of evidence A1]
   • A total of 55 patients completed 35 cycles of pembrolizumab. The objective response rate was 90.9% and the 3-year OS rate after completion of 35 cycles (approximately 5 years after random assignment) was 69.5%.

Pembrolizumab alone

Evidence (pembrolizumab alone):

1. A phase III open-label study (KEYNOTE-024) randomly assigned 305 patients with previously untreated, advanced NSCLC with PD-L1 expression on 50% or more tumor cells and no sensitizing EGFR mutations or ALK translocations to either intravenous pembrolizumab (200 mg every 3 weeks for up to 35 cycles) or platinum-based chemotherapy (four to six cycles, investigator’s choice; pemetrexed maintenance was allowed for nonsquamous tumors).[86] The primary end point was PFS.

   1. PD-L1 expression was centrally assessed using the PD-L1 immunohistochemistry 22C3 pharmDx assay. PD-L1 tumor expression of 50% or more was found in 30.2% of 1,653 patient samples that were examined.
   2. Pembrolizumab demonstrated significant improvement in median PFS (10.3 months vs. 6.0 months; HR, 0.50; 95% CI, 0.37–0.68; P < .001). The overall response rate (44.8% vs. 27.8%), the median DOR (not reached, [range, 1.9–14.5 months] vs. 6.3 months [range, 2.1–12.6 months]), and the estimated rate of OS at 6 months (80.2% vs. 72.4%; HR, 0.60; 95% CI, 0.41–0.89; P = .005) were all higher with pembrolizumab than with chemotherapy.
   3. Further follow-up of this study confirmed an OS advantage in favor of pembrolizumab; the median OS for patients who received pembrolizumab was 30 months (95% CI, 18.3 months–not reached) versus 14.2 months for patients who received chemotherapy, with a 75% crossover to immunotherapy afterwards, suggesting the crossover did not impact survival.[89]
   4. Adverse events (any grade) were less frequent with pembrolizumab than with chemotherapy (73.4% vs. 90.0%).

      • Grade 3 to 5 adverse events occurred in 26.6% of patients treated with pembrolizumab and 53.3% of patients treated with chemotherapy.
      • Grade 3 or 4 immune-related events occurred in 9.7% of patients treated with pembrolizumab and 0.7% of patients treated with chemotherapy.
      • The most common grade 3 or 4 immune-related events associated with pembrolizumab were severe skin reactions (3.9%), pneumonitis (2.6%), and colitis (1.3%).
      • There were no grade 5 immune-related events.
   5. Pembrolizumab treatment demonstrated significant improvement in PFS, OS, and DOR with less frequent adverse events compared with chemotherapy treatment.[86][Level of evidence B1]
2. A phase III open-label study (KEYNOTE-042 [NCT02220894]) included patients with locally advanced or metastatic NSCLC without EGFR or ALK alterations and with a PD-L1 TPS score greater than 1%. Patients were randomly assigned to receive either pembrolizumab 200 mg once every 3 weeks for 35 cycles or chemotherapy (carboplatin plus paclitaxel or pemetrexed) for four to six cycles with optional maintenance pemetrexed (pembrolizumab, n = 637; chemotherapy, n = 637). The primary end points were OS in the populations with a PD-L1 TPS greater than 50%, greater than 20%, and greater than 1%.[90]

   1. The median follow-up was 61.1 months (range, 50.0–76.3).
   2. OS outcomes favored pembrolizumab versus chemotherapy, regardless of the PD-L1 TPS.

      • The HR was 0.68 (95% CI, 0.57–0.81) for the TPS >50% group, 0.75 (95% CI, 0.64–0.87) for the TPS >20% group, and 0.79 (95% CI, 0.70–0.89) for the TPS >1% group.
      • The OS rates for patients who received pembrolizumab were 21.9% (TPS >50%), 19.4% (TPS >1%), and 16.6% (TPS >1%).
   3. The most common adverse reactions reported in at least 10% of patients who received pembrolizumab as a single agent in KEYNOTE-042 included fatigue, decreased appetite, dyspnea, cough, rash, constipation, diarrhea, nausea, hypothyroidism, pneumonia, pyrexia, and weight loss.

The FDA approved pembrolizumab in combination with pemetrexed and carboplatin as first-line treatment of patients with metastatic nonsquamous NSCLC, regardless of PD-L1 expression. The FDA also approved pembrolizumab as a first-line monotherapy for patients with NSCLC whose tumors express PD-L1 (>1%) (staining as determined by an FDA-approved test). Patients with EGFR or ALK genomic tumor aberrations should have disease progression on FDA-approved therapies before receiving pembrolizumab (see the FDA label for pembrolizumab).

Cemiplimab-rwlc plus chemotherapy

Evidence (cemiplimab-rwlc plus chemotherapy):

1. A phase III, double-blind, placebo-controlled trial (EMPOWER-Lung 3 [NCT03409614]) examined cemiplimab-rwlc plus platinum-doublet chemotherapy in 466 patients with stage III or IV advanced NSCLC. Patients had not received previous chemotherapy and had no EGFR, ALK, or ROS1 genomic tumor aberrations. Patients were randomly assigned (2:1) to receive cemiplimab-rwlc 350 mg (n = 312) or placebo (n = 154) every 3 weeks for up to 108 weeks along with four cycles of platinum-doublet chemotherapy. Patients also received pemetrexed maintenance as indicated. The primary end point was OS. The trial met preset OS efficacy criteria and was stopped early on the recommendation of the independent data monitoring committee.[91][Level of evidence A1]

   • The median OS was 21.9 months (95% CI, 15.5–not evaluable [NE]) in the cemiplimab-rwlc-plus-chemotherapy group and 13 months (95% CI, 11.9–16.1) in the placebo-plus-chemotherapy group (HR, 0.71; 95% CI, 0.53–0.93; P = .014).
   • The secondary end point of median PFS was 8.2 months (95% CI, 6.4–9.3) in the cemiplimab-rwlc-plus-chemotherapy group and 5.0 months (95% CI, 4.3–6.2) in the placebo-plus-chemotherapy group (HR, 0.56; 95% CI, 0.44–0.70; P < .0001).
   • Another secondary end point, the estimated proportion of patients alive at 12 months, was 65.7% (95% CI, 59.9%–70.9%) in the cemiplimab-rwlc-plus-chemotherapy group and 56.1% (95% CI, 47.5%–63.8%) in the placebo-plus-chemotherapy group.
   • Grade 3 or greater adverse events occurred in 43.6% (136 of 312) of patients who received cemiplimab-rwlc plus chemotherapy and 31.4% (48 of 153) of patients who received placebo plus chemotherapy.
   • The most common (≥15%) adverse reactions were alopecia, musculoskeletal pain, nausea, fatigue, peripheral neuropathy, and decreased appetite.

The FDA approved cemiplimab-rwlc in combination with platinum-based chemotherapy for adult patients with advanced NSCLC and no EGFR, ALK, or ROS1 aberrations.

Cemiplimab-rwlc alone

Evidence (cemiplimab-rwlc alone):

1. A phase III open-label study (EMPOWER-Lung 1 [NCT03088540]) enrolled 710 patients with advanced NSCLC and PD-L1 tumor expression of at least 50%. Patients were randomly assigned (1:1) to receive cemiplimab-rwlc 350mg every 3 weeks for up to 108 weeks or platinum-doublet chemotherapy. Patients could cross over from chemotherapy to cemiplimab-rwlc in the event of disease progression. There was also the option to cross over to continue cemiplimab-rwlc plus four cycles of chemotherapy in the event of progression with cemiplimab alone. Primary end points were OS and PFS per blinded independent central review. The median follow-up was 37 months for the ITT population.[92,93][Level of evidence A1]

   • At 35 months of follow-up, in patients with PD-L1 expression ≥50%, the median OS was 26.1 months with cemiplimab-rwlc and 13.3 months with chemotherapy (HR, 0.57; P < .0001).
   • The median PFS was 8.1 months for patients who received cemiplimab-rwlc and 5.3 months for patients who received chemotherapy (HR, 0.51; P < .0001).
   • The objective response rate was 46% for patients who received cemiplimab-rwlc and 21% for patients who received chemotherapy (OR, 3.264; P < .0001).
   • Benefits were greater in patients with PD-L1 expression ≥90% versus 50% to 89%.
   • Among 64 patients who received cemiplimab-rwlc plus chemotherapy after initial progression on cemiplimab-rwlc alone, the median PFS was 6.6 months, the objective response rate was 31%, and the OS was 15.1 months.
   • The most common adverse reactions (>10%) with cemiplimab-rwlc were musculoskeletal pain, rash, anemia, fatigue, decreased appetite, pneumonia, and cough. The safety profile of cemiplimab-rwlc was consistent over longer follow-up.

The FDA approved cemiplimab-rwlc for patients with advanced NSCLC (locally advanced who are not candidates for surgical resection or definitive chemoradiation or metastatic) and PD-L1 tumor expression of at least 50% with no EGFR, anaplastic ALK, or ROS1 genomic aberrations.

Tremelimumab

Tremelimumab is a fully human monoclonal antibody against cytotoxic T-lymphocyte associated antigen 4 (CTLA-4). It is an immune checkpoint blocker.

Atezolizumab alone

Evidence (atezolizumab alone):

1. A phase III open-label study (IMpower110 [NCT02409342]) included 572 patients with previously untreated metastatic nonsquamous or squamous NSCLC. Patients had PD-L1 expression on at least 1% of tumor cells or on at least 1% of tumor-infiltrating immune cells. Patients were randomly assigned to receive either atezolizumab (1,200 mg intravenously) or platinum-based chemotherapy (4 or 6 cycles) once every 3 weeks. The primary end point was OS in the PD-L1–selected population that excluded sensitizing EGFR mutations or ALK translocations.[95][Level of evidence A1]

   • PD-L1 expression was assessed by the SP142 immunohistochemical assay. High expression was defined as more than 50% of tumor cells or more than 10% of tumor-infiltrating immune cells expressing PD-L1.
   • In the 205 patients with high PD-L1 expression, the median OS was 20.2 months for patients who received atezolizumab and 13.1 months for patients who received chemotherapy (HR_death, 0.59; P = .01).
   • Grade 3 to 4 adverse events occurred in 30.1% of patients who received atezolizumab and 52.5% of patients who received chemotherapy.

Atezolizumab monotherapy is approved for first-line treatment of patients with high PD-L1 expression (PD-L1 staining ≥50% of tumor cells or PD-L1 stained tumor-infiltrating immune cells covering ≥10% of the tumor area), as determined by an FDA-approved test, in the absence of EGFR or ALK genomic aberrations.

Atezolizumab plus chemotherapy

Evidence (atezolizumab in combination with carboplatin and nab-paclitaxel chemotherapy):

1. A phase III open-label study (IMpower130 [NCT02367781]) included 724 patients with previously untreated, stage IV, nonsquamous NSCLC. Patients were randomly assigned 2:1 to receive atezolizumab (1,200 mg intravenously every 3 weeks) plus chemotherapy (carboplatin, AUC 6 mg/mL per minute every 3 weeks with nab-paclitaxel 100 mg/m² intravenously every week), or chemotherapy alone given once every 3 weeks for four or six cycles. All patients received maintenance therapy as follows: (1) patients in the atezolizumab-plus-chemotherapy group received atezolizumab 1,200 mg intravenously every 3 weeks until investigator-assessed loss of clinical benefit or toxicity, and (2) patients in the chemotherapy-alone group received best supportive care or pemetrexed switch maintenance therapy until disease progression or toxicity. Coprimary end points were investigator-assessed PFS and OS in the ITT population with EGFR wild-type and ALK wild-type tumors.[96][Level of evidence A1]

   • In the ITT wild-type population, the median OS was 18.6 months (95% CI, 16.0–21.2) in the atezolizumab-plus-chemotherapy group and 13.9 months (95% CI, 12.0–18.7) in the chemotherapy group (stratified HR, 0.79; 95% CI, 0.64–0.98; P = .033).
   • The median PFS was 7 months (95% CI, 6.2–7.3) in the atezolizumab-plus-chemotherapy group and 5.5 months (95% CI, 4.4–5.9) in the chemotherapy group (stratified HR, 0.64; 95% CI, 0.54–0.77; P < .0001).
   • Subgroup analyses showed OS and PFS benefit with atezolizumab across several clinical subgroups, with the exception of patients with liver metastases where the additional of atezolizumab did not improve OS versus chemotherapy alone, and for patients with EGFR and ALK genomic alterations.
   • OS and PFS benefit with atezolizumab was also observed in the ITT wild-type population independent of PD-L1 expression.
   • Grade 3 or 4 adverse events occurred in 81% of patients who received atezolizumab plus chemotherapy versus 71% of patients who received chemotherapy alone. Immune-related adverse events occurred in 45% of patients treated with atezolizumab plus chemotherapy and most were grade 1 or 2 in severity. The most common immune-related adverse events were rash (24%), hypothyroidism (15%), and hepatitis (10%).

Atezolizumab in combination with nab-paclitaxel and carboplatin is approved for the first-line treatment of patients with metastatic nonsquamous NSCLC with no EGFR or ALK genomic aberrations.

Atezolizumab plus bevacizumab plus chemotherapy

Evidence (atezolizumab in combination with carboplatin, paclitaxel, and bevacizumab):

1. In a phase III open-label study (IMpower150 [NCT02366143]),1,202 patients with stage IV or recurrent metastatic nonsquamous NSCLC were randomly assigned in a 1:1:1 ratio to receive either atezolizumab plus carboplatin plus paclitaxel (ACP group), atezolizumab plus bevacizumab plus carboplatin plus paclitaxel (ABCP group), or bevacizumab plus carboplatin plus paclitaxel (BCP group). Treatment consisted of four or six 21-day cycles. Atezolizumab was given intravenously at a dose of 1,200 mg, bevacizumab at a dose of 15 mg per kilogram of body weight, carboplatin at an area under the concentration-time curve of 6 mg/mL per minute and paclitaxel at a dose of 200 mg/m² (175 mg/m² for Asian patients). Patients continued to receive atezolizumab, bevacizumab, or both until disease progression or development of intolerable toxicity. Coprimary end points were PFS, both in the ITT population with EGFR wild-type and ALK wild-type tumors and among patients with wild-type tumors who had high expression of an effector T-cell (Teff) gene signature in the tumor, and OS in the wild-type population.[97][Level of evidence A1]

   • Median PFS was longer in the ABCP group (8.3 months) than the BCP group (6.8 months) (HR, 0.62; 95% CI, 0.52–0.74; P < .001). In the Teff-high wild-type population, PFS was 11.3 months versus 6.8 months (HR, 0.51; 95% CI, 0.38–0.68; P .001). PFS was also longer in the ABCP group versus the BCP group in the ITT population with EGFR and ALK genomic alterations, among patients with low or negative PD-L1 expression, low Teff gene-signature expression, and in patients with liver metastases.
   • Median OS among patients with wild-type tumors was longer in the ABCP group (19.2 months), compared with the BCP group (14.7 months) (HR, 0.78; 95% CI, 0.64–0.96; P = .02).
   • Grade 3 or 4 treatment-related adverse events occurred in 56% of patients in the ABCP group versus 48% of patients in the BCP group. Most immune-related adverse events in the ABCP group were grade 1 or 2, and rash, hypothyroidism, hyperthyroidism, hepatitis, pneumonitis, and colitis were most common. Treatment-related deaths occurred in 11 patients (2.8%) in the ABCP group and 9 patients (2.3%) in the BCP group. Five deaths in the ABCP group were caused by pulmonary hemorrhage or hemoptysis, and four of five occurred in patients with high-risk features, including tumors infiltrating great vessels or tumor cavitation.

Atezolizumab in combination with bevacizumab, paclitaxel, and carboplatin is approved for the first-line treatment of patients with metastatic nonsquamous NSCLC with no EGFR or ALK genomic aberrations.

Nivolumab plus ipilimumab

Nivolumab, a fully human anti–PD-1 antibody, and ipilimumab, a fully human anti–CTLA-4 antibody, are immune checkpoint inhibitors with distinct but complementary mechanisms of action.[98]

Evidence (nivolumab plus ipilimumab):

1. A phase III open-label study (CheckMate 227 [NCT02477826]) evaluated nivolumab in combination with ipilimumab versus chemotherapy as first-line treatment for stage IV or recurrent NSCLC without sensitizing EGFR mutations or ALK translocations. Patients (n = 1,739) were grouped by PD-L1 tumor status (either ≥1% or <1%). Patients with PD-L1 expression of at least 1% were randomly assigned to receive either nivolumab (3 mg/kg every 2 weeks) plus ipilimumab (1 mg/kg every 6 weeks), nivolumab (240 mg every 2 weeks) alone, or platinum-doublet chemotherapy every 3 weeks for up to four cycles. Patients with PD-L1 expression less than 1% were randomly assigned to receive nivolumab with ipilimumab, nivolumab with platinum-doublet chemotherapy, or platinum-doublet chemotherapy (every 3 weeks). Patients were treated until disease progression or unacceptable toxicity or up to 2 years for immunotherapy. Coprimary end points were OS with nivolumab-plus-ipilimumab compared with chemotherapy in patients with tumor PD-L1 expression of at least 1%, and PFS with nivolumab-plus-ipilimumab compared with chemotherapy in patients with high tumor mutational burden (TMB) (≥10 mutations per megabase).[2,99][Level of evidence A1]

   • Among patients with tumor PD-L1 ≥1% (n = 1,189), the median OS was 17.1 months (95% CI, 15.0–20.2) with nivolumab-plus-ipilimumab and 14.9 months (95% CI, 12.7–16.7) with chemotherapy (HR, 0.77; 95% CI, 0.66–0.91; P = .007). Five-year outcomes with nivolumab-plus-ipilimumab versus chemotherapy showed durable clinical benefit, with an OS rate of 24% with nivolumab-plus-ipilimumab and 14% for chemotherapy alone.[98]
   • In patients with TMB-high NSCLC, the median PFS was 7.2 months (95% CI, 5.5–13.2) with nivolumab-plus-ipilimumab, versus 5.5 months (95% CI, 4.4–5.8) with chemotherapy alone (HR, 0.58; 97.5% CI, 0.41–0.81; P < .001).
   • Among patients with tumor PD-L1 <1% (n = 550), the median OS was 17.4 months (95% CI, 13.2–22.0) with nivolumab-plus-ipilimumab, and 12.2 months (95% CI, 9.2–14.3) with chemotherapy alone (HR, 0.65; 95% CI, 0.52–0.81). Five-year outcomes with nivolumab-plus-ipilimumab versus chemotherapy alone showed durable clinical benefit, with an OS rate of 19% for nivolumab-plus-ipilimumab and 7% for chemotherapy alone.[98]
   • The frequency of grade 3 to 4 treatment-related adverse events was similar in both groups (32.8% with nivolumab-plus-ipilimumab vs. 36.0% with chemotherapy alone). Treatment-related adverse events leading to therapy discontinuation were more common with nivolumab-plus-ipilimumab than with chemotherapy alone (24.5% vs. 13.9%).
   • Treatment-related deaths occurred in eight patients who received nivolumab-plus-ipilimumab (pneumonitis in four patients; shock, myocarditis, acute tubular necrosis, and cardiac tamponade in one patient each) and in six patients who received chemotherapy (sepsis in two patients; febrile neutropenia, multifocal brain infarctions, interstitial lung disease, and thrombocytopenia in one patient each).

The FDA approved nivolumab-plus-ipilimumab as first-line therapy for patients with advanced NSCLC with PD-L1 expression of at least 1% and no EGFR or ALK genomic aberrations. While this regimen is not FDA-approved for patients with PD-L1 expression less than 1%, these patients were noted to have durable clinical benefit in CheckMate 227.

Everolimus

Everolimus is used for patients with unresectable, locally advanced or metastatic, progressive, well-differentiated, nonfunctional, neuroendocrine tumors.

Everolimus, an oral mTOR inhibitor, is clinically active against advanced pancreatic and nonpancreatic neuroendocrine tumors.[100] Based on the results of the RADIANT-4 clinical trial,[100] the FDA approved everolimus for the treatment of adult patients with unresectable, locally advanced or metastatic, progressive, well-differentiated (low or intermediate grade), nonfunctional neuroendocrine tumors of lung or gastrointestinal origin.

Evidence (everolimus):

1. A randomized, double-blind, placebo-controlled, phase III trial (RADIANT-4 [NCT01524783]) evaluated everolimus in patients older than 18 years with advanced, progressive, well-differentiated, nonfunctional neuroendocrine tumors of lung or gastrointestinal origin.[100] Eligible patients were randomly assigned in a 2:1 ratio to received everolimus 10 mg daily orally or placebo, both with best supportive care. A total of 302 patients were enrolled (205 in the everolimus arm and 97 in the placebo arm), including 90 patients with neuroendocrine tumors of lung origin (63 in the everolimus arm and 27 in the placebo arm). The primary end point was PFS assessed by central radiology review in the ITT population.

   • Median PFS was 11.0 months in the everolimus arm and 3.9 months in the placebo group (HR, 0.48; 95% CI, 0.35−0.67; P < .00001).[100][Level of evidence B1]
   • In a post hoc analysis of the lung subgroup, median PFS by central review was 9.2 months in the everolimus arm and 3.6 months in the placebo arm (HR, 0.50; 95% CI, 0.28−0.88).[100]
   • The objective response rate was 2% in patients who received everolimus and 1% in patients who received placebo. Disease stabilization was observed in 81% of patients in the everolimus arm and 64% of patients in the placebo arm.
   • The median duration of treatment was longer in the patients who received everolimus compared with those who received placebo (40.4 weeks vs. 19.6 weeks).
   • A planned interim analysis of OS showed a 36% reduction in the estimated risk of death with everolimus relative to placebo (HR, 0.64; 95% CI, 0.40−1.05). These results were not statistically significant.
   • The most common treatment-related adverse events were stomatitis, diarrhea, fatigue, infections, rash, and peripheral edema. The most common drug-related grade 3 or 4 adverse events were stomatitis, diarrhea, infections, anemia, and fatigue. Grade 3 or 4 adverse events resulted in treatment discontinuation in 12% of patients in the everolimus group and 3% of patients in the placebo group.

Endobronchial laser therapy and/or brachytherapy (for obstruction lesions)

Radiation therapy may be effective in palliating symptomatic patients with local involvement of NSCLC with any of the following:

• Tracheal, esophageal, or bronchial compression.
• Pain.
• Vocal cord paralysis.
• Hemoptysis.
• Superior vena cava syndrome.

In some cases, endobronchial laser therapy and/or brachytherapy have been used to alleviate proximal obstructing lesions.[20]

EBRT (primarily for palliation of local symptomatic tumor growth)

Although EBRT is frequently prescribed for symptom palliation, there is no consensus on which fractionation scheme should be used. Although different multifraction regimens appear to provide similar symptom relief,[101-106] single-fraction radiation may be insufficient for symptom relief compared with hypofractionated or standard regimens, as evidenced in the NCT00003685 trial.[21][Level of evidence A3] Evidence of a modest increase in survival in patients with a better performance status given high-dose radiation therapy is available.[23,107][Level of evidence A1] In closely observed asymptomatic patients, treatment may often be appropriately deferred until symptoms or signs of a progressive tumor develop.

Evidence (radiation therapy):

1. A systematic review identified six randomized trials of high-dose rate endobronchial brachytherapy (HDREB) alone or with EBRT or laser therapy.[108]

   • Better overall symptom palliation and fewer re-treatments were required in previously untreated patients using EBRT alone.[108][Level of evidence A3]
   • HDREB provided palliation of symptomatic patients with recurrent endobronchial obstruction previously treated by EBRT, when it was technically feasible.

Treatment of second primary tumor

A solitary pulmonary metastasis from an initially resected bronchogenic carcinoma is unusual. The lung is frequently the site of second primary malignancies in patients with primary lung cancers. Whether the new lesion is a new primary cancer or a metastasis may be difficult to determine. Studies have indicated that in most patients the new lesion is a second primary tumor, and after its resection, some patients may achieve long-term survival. Thus, if the first primary tumor has been controlled, the second primary tumor should be resected, if possible.[109,110]

Treatment of brain metastases

Patients who present with a solitary cerebral metastasis after resection of a primary NSCLC lesion and who have no evidence of extracranial tumor can achieve prolonged disease-free survival with surgical excision of the brain metastasis and postoperative whole-brain radiation therapy.[111,112] Unresectable brain metastases in this setting may be treated with stereotactic radiosurgery.[113]

Approximately 50% of patients treated with resection and postoperative radiation therapy will develop recurrence in the brain; some of these patients will be suitable for additional treatment.[114] In those selected patients with good performance status and without progressive metastases outside of the brain, treatment options include reoperation or stereotactic radiation surgery.[113,114] For most patients, additional radiation therapy can be considered; however, the palliative benefit of this treatment is limited.[115][Level of evidence C2]

Docetaxel

Evidence (docetaxel):

1. Two prospective randomized studies have shown an improvement in survival with the use of docetaxel compared with vinorelbine, ifosfamide, or best supportive care;[4,5] however, criteria for the selection of appropriate patients for second-line treatment are not well defined.[6]
2. A meta-analysis of five trials of 865 patients assessing the efficacy and safety of docetaxel administered weekly or every 3 weeks has been reported.[7] In that analysis, the following was shown:

   • Median survival was 27.4 weeks for patients treated every 3 weeks and 26.1 weeks for patients treated weekly (log-rank P = .24).
   • Significantly less severe neutropenia and febrile neutropenia were reported with weekly docetaxel (P < .001 for both); however, no significant differences were observed for anemia, thrombocytopenia, and nonhematologic toxic effects.

Docetaxel plus ramucirumab

Evidence (docetaxel plus ramucirumab):

1. In a double-blind, placebo-controlled, phase III study, 1,253 patients with an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 1 who had progressive disease after first-line chemotherapy were randomly assigned to receive docetaxel and placebo or docetaxel and ramucirumab.[8][Level of evidence A1] Ramucirumab is a human immunoglobulin G1 monoclonal antibody that targets the extracellular domain of vascular endothelial growth factor receptor 2. The primary end point of the study was overall survival (OS), with secondary end points of progression-free survival (PFS) and objective response rate. The study enrolled patients with either nonsquamous or squamous NSCLC; however, patients with poorly controlled hypertension, gastrointestinal perforation or fistulae, arterial thromboembolic event within 6 months (before random assignment), gross hemoptysis within 2 months, or grade 3 to 4 gastrointestinal bleeding within 3 months were excluded. In addition, the trial did not include patients with tumors that had major blood vessel involvement or intratumor cavitation.

   • The addition of ramucirumab to docetaxel compared with placebo plus docetaxel led to an increase in median OS (10.5 months vs. 9.1 months; hazard ratio [HR], 0.86; 95% confidence interval [CI], 0.75–0.98), objective response rate (23% vs. 14%), and PFS (4.5 months vs. 3 months). The improvement in OS from the addition of ramucirumab appeared consistent across subgroups including squamous and nonsquamous histologies.
   • Grade 3 to 4 treatment-related adverse events occurred in 79% of patients who received docetaxel and ramucirumab as compared with 71% of patients who received docetaxel and placebo. Febrile neutropenia, fatigue, and hypertension were among the toxicities that were more common with the addition of ramucirumab to docetaxel. There was no significant difference in the incidence of grades 3 to 4 hemorrhage between the groups.
   • On the basis of this study, the addition of ramucirumab to docetaxel chemotherapy can be considered for patients with good performance status with advanced NSCLC who have progressive disease after first-line chemotherapy.

Pemetrexed

Evidence (pemetrexed):

1. A randomized phase III trial of 571 patients was designed to demonstrate the noninferiority of pemetrexed compared with docetaxel.

   • The trial showed no difference in response rates, PFS, or OS between treatments.[9][Level of evidence A1]
   • Of note, patients with squamous histology benefited from docetaxel, and those with nonsquamous histologies appeared to benefit more from pemetrexed.[10]

EGFR-directed therapy after first-line chemotherapy (for patients with EGFR-sensitizing mutations)

EGFR-directed therapy (for patients with acquired EGFR T790M mutations after previous EGFR-directed therapy)

EGFR-directed therapy after first-line chemotherapy (for patients with EGFR exon 20 insertion mutations)

ALK-directed TKIs after first-line chemotherapy

ALK-directed TKIs after prior ALK TKI therapy

Dabrafenib and trametinib

Evidence (dabrafenib and trametinib):

1. In a phase II, multicenter, nonrandomized, open-label study (NCT01336634), 57 patients with progression after at least one to three previous platinum-containing regimens for treatment of metastatic NSCLC, who tested positive for BRAF V600E mutations, were treated with dabrafenib (a BRAF inhibitor) 150 mg twice a day and trametinib (a MEK inhibitor) 2 mg every day.[36] BRAF V600E mutations were ascertained by local testing. The primary end point was investigator-assessed overall response.

   • The overall response rate was 63.2% (95% CI, 49.3%–75.6%), as determined independently by investigator and independent review committee assessments. There were 2 out of 36 complete responses by investigator assessment; all responses were deemed partial by the independent review committee.
   • The median investigator-assessed PFS was 9.7 months (95% CI, 6.9–19.6 months). The estimated median DOR was 9.0 months (95% CI, 6.9–18.3). The OS data are immature.
   • Forty-nine percent of patients had at least one grade 3 or 4 adverse event, the most common of which were neutropenia, hyponatremia, and anemia.[36][Level of evidence C3]

The FDA approved the combination of dabrafenib and trametinib for patients with NSCLC whose tumors harbor BRAF V600E mutations as detected by an FDA-approved test.

Entrectinib

The FDA approved entrectinib for treatment of patients with metastatic NSCLC whose tumors are ROS1-positive, regardless of the number of previous systemic therapies.

Evidence (entrectinib):

1. The safety and clinical activity of entrectinib in ROS1 fusion-positive metastatic NSCLC was determined by integrated analysis of three multicenter, single-arm, open-label clinical trials (ALKA-372-001/EudraCT, 2012-000148-88, STARTRK-1 [NCT02097810], and STARTRK-2 [NCT02568267]).[38] Entrectinib was administered orally at a dose of at least 600 mg once daily. Primary end points were the overall response rate and the DOR determined by blinded independent central review. Of note, time-to-event end points are difficult to interpret in the absence of a control arm. Evaluation of tumor samples for the ROS1 gene fusion was conducted prospectively in local laboratories using either a FISH or next-generation sequencing (NGS) laboratory-developed test.
   Seventeen (32%) patients had received no previous systemic therapy, 23 (43%) had received one previous therapy, and 13 (25%) had received two or more lines of treatment. CNS disease was present in 23 (43%) patients at baseline. Thirty-one (59%) patients were never smokers and 52 (98%) patients had adenocarcinoma histology.

   • The overall response rate in 53 efficacy-evaluable patients was 77% (95% CI, 64%−88%). Six percent of patients had a complete response and 72% had a partial response. Among patients with CNS disease at baseline, the overall response rate was 74% and all patients had a partial response. Among patients without CNS disease at baseline, the overall response rate was 80% (10% complete response rate; 70% partial response rate). [38][Level of evidence C3]
   • The median DOR was 24.6 months (95% CI, 11.4−34.8); 12.6 months (95% CI, 6.5−not estimable) in patients with baseline CNS disease, and 24.6 months (95% CI, 11.4−34.8) in those without CNS disease at baseline.
   • Treatment-related adverse events were assessed in 134 patients in the safety-evaluable population. Grade 1 or 2 treatment-related adverse events were observed in 79 (59%) patients. Grade 3 or 4 treatment-related adverse events were observed in 46 (34%) patients. Fifteen (11%) patients had serious treatment-related adverse events. There were no treatment-related deaths.
   • The median PFS was 19 months (95% CI, 12.2−36.6); 13.6 months (95% CI, 4.5−not estimable) in patients with baseline CNS disease, and 26.3 months (95% CI, 15.7−36.6) in patients with no baseline CNS disease.

Crizotinib

Crizotinib was approved for patients with metastatic NSCLC whose tumors are ROS1-positive, regardless of the number of previous systemic therapies.

Evidence (crizotinib):

1. In an expansion cohort of a phase I study of crizotinib, 50 patients with advanced NSCLC who tested positive for ROS1 rearrangement were treated with oral crizotinib 250 mg twice daily.[39] ROS1 rearrangements were identified using break-apart FISH or reverse transcriptase−polymerase chain reaction assay. Seven patients (14%) had not had any previous treatment for advanced disease, 21 patients (42%) had one previous treatment, and 22 patients (44%) had more than one previous treatment. The primary end point was response rate.

   • The overall response rate was 72% (95% CI, 58%–84%). Six percent of patients had a complete response, 66% had a partial response, and 18% had stable disease as their best response.
   • Median PFS was 19.2 months (95% CI, 14.4–not reached). The estimated DOR was 17.6 months (95% CI, 14.5–not reached).[39][Level of evidence C3]
2. In a phase II, open-label, single-arm trial, 127 East Asian patients with ROS1-positive NSCLC were treated with crizotinib 250 mg twice daily.[40] Twenty-four patients (18.9%) had not had any previous treatment for advanced disease, 53 patients (41.7%) had one previous treatment, and 50 patients (39%) had two or three previous treatments. The primary end point was objective response rate by independent review.

   • The objective response rate was 71.7% (95% CI, 63.0%–79.3%). Response rates were similar, irrespective of the number of previous therapies. Complete responses occurred in 13.4% of patients, while 58.3% of patients had partial responses and 16.5% of patients had stable disease as their best response.[40][Level of evidence C3]
   • Median PFS was 15.9 months (95% CI, 12.9–24). The DOR was 19.7 months (95% CI, 14.1–not reached).
   • OS was 32.5 months (95% CI, 32.5–not reached).

Larotrectinib

Evidence (larotrectinib):

1. Larotrectinib was studied in three protocols: a phase I study involving adults, a phase I/II study involving children, and a phase II study involving adolescents and adults.[43] Fusions were confirmed in the tumors using either FISH or NGS methods. The primary end point for the combined analysis was objective response rate by independent review and was conducted with input from regulators with the goal of excluding a lower bound of less than 30% for response rate. In total, 55 patients with a median age of 45 years (range, 4 months‒76 years) were enrolled across 17 different NTRK fusion-positive tumor types. All patients had either metastatic disease (82%) or locally advanced unresectable disease (18%). Enrolled patients had received a median of two previous systemic therapies.

   • The objective response rate was 75% (95% CI, 61%‒75%) and 73% of these responses lasted at least 6 months.[43][Level of evidence C3]
   • Treatment was well tolerated with 93% of adverse events being grade 1 to 2; the most common grade 3 to 4 adverse events were anemia (11% of patients), transaminitis (7%), and neutropenia (7%).

The FDA approved larotrectinib for the treatment of patients who have locally advanced or metastatic tumors that harbor an NTRK gene fusion without a known acquired resistance mutation, and who have no satisfactory alternative treatments or whose cancer has progressed following treatment.

Entrectinib

The FDA granted accelerated approval to entrectinib for the treatment of solid tumors that have an NTRK gene fusion without a known acquired resistance mutation, are metastatic, have progressed after treatment, have no satisfactory alternative therapy, or for cases in which surgical resection is likely to result in severe morbidity.

Evidence (entrectinib):

1. The safety and clinical activity of entrectinib in NTRK inhibitor–naïve patients with metastatic or locally-advanced solid tumors (including NSCLC) harboring NTRK1, NTRK2, or NTRK3 gene fusions was determined by integrated analysis of three early-phase, multicenter, single-arm, open-label clinical trials (ALKA-372-001/EudraCT, 2012-000148-88, STARTRK-1 [NCT02097810], and STARTRK-2 [NCT02568267]).[44] Treatment consisted of entrectinib administered orally at a dose of at least 600 mg once per day. The primary end points were objective response and median DOR, which were assessed by blinded independent central review. Of note, time-to-event end points are difficult to interpret in the absence of a control arm. Identification of positive NTRK gene fusion status was conducted prospectively in local laboratories or a central laboratory using various nucleic acid–based tests.
   Of 54 patients in the NTRK gene fusion-positive efficacy-evaluable population, 20 (37%) had received no previous systemic therapy, 11 (20%) had received one previous systemic therapy, and 23 (43%) had received two or more systemic therapies. Twelve (22%) patients had CNS disease at baseline. Ten (19%) patients had NSCLC. Fifty-two (96%) patients had an NTRK gene fusion detected by NGS and two (4%) had an NTRK gene fusion detected by other nucleic acid–based tests.

   • The objective response rate in 54 patients was 57% (95% CI, 43.2%−70.8%). Seven percent of patients had a complete response and 50% had a partial response. In patients with baseline CNS disease the overall response rate was 50% (all partial responses), whereas in patients without baseline CNS disease, the overall response rate was 60% (10% complete response; 50% partial response).[44][Level of evidence C3]
   • The median DOR was 10.4 months (95% CI, 7.1−not estimable). In patients with baseline CNS disease, the DOR was not estimable, whereas it was 12.9 months (95% CI, 7.1−not estimable) in patients with no baseline CNS disease.
   • Among 10 patients with NSCLC, the overall response rate was 70% (95% CI, 35%−93%) and the DOR ranged between 1.9 months and 20.1 months. For more information, see the prescribing information.
   • The safety-evaluable population consisted of 68 patients with NTRK fusion-positive tumors. Most treatment-related adverse events were grade 1 or 2 and reversible. The most frequent grade 3 or 4 treatment-related adverse events were increased weight gain (10%) and anemia (12%). Serious treatment-related adverse events were reported in 7 (10%) patients. Three (4%) patients had dose interruptions and 27 (40%) patients had dose reductions due to treatment-related adverse events. There were no treatment-related deaths.
   • Median PFS was 11.2 months (95% CI, 8.0−14.9). In patients with baseline CNS disease, median PFS was 7.7 months (95% CI, 4.7−not estimable), and it was 12 months (95% CI, 8.7−15.7) in patients with no baseline CNS disease.

Selpercatinib

Evidence (selpercatinib):

1. A phase I/II study (LIBRETTO-001 [NCT03157128]) enrolled patients with RET fusion−positive solid tumors RET fusion status was determined by local molecular testing (NGS, FISH, or polymerase chain reaction assay) without central confirmation. The primary end point was objective response.[46][Level of evidence C3]

   • Updated analysis was conducted in 316 patients with RET fusion–positive NSCLC.[46]
   • Among the 69 treatment-naïve patients, the objective response rate was 84% (95% CI, 73%–92%), and 6% achieved complete responses. The median DOR was 20.2 months (95% CI, 13.0–could not be evaluated); 40% of responses were ongoing at the data cutoff (median follow-up, 20.3 months). The median PFS was 22.0 months; 35% of patients were alive and progression-free at the data cutoff (median follow-up, 21.9 months).
   • Among the 247 patients who had received prior platinum-based chemotherapy, the objective response rate was 61% (95% CI, 55%–67%), and 7% achieved complete responses. The median DOR was 28.6 months (95% CI, 20.4–could not be evaluated); 49% of responses were ongoing (median follow-up, 21.2 months). The median PFS was 24.9 months; 38% of patients were alive and progression-free at the data cutoff (median follow-up, 24.7 months).
   • Among the 26 patients with measurable baseline CNS metastasis by the independent review committee, the intracranial objective response rate was 85% (95% CI, 65%–96%), and 27% had complete responses.
   • In the full safety population (n = 796), the median treatment duration was 36.1 months.
   • There was no significant change in the safety profile. Most adverse events were grade 1 to 2. The most common adverse events were edema, diarrhea, fatigue, dry mouth, hypertension, increased alanine aminotransferase (ALT) and aspartate aminotransferase (AST), and rash.

The FDA approved selpercatinib to treat adults with locally advanced or metastatic NSCLC with RET gene fusion, as detected by an FDA-approved test.

Pralsetinib

Evidence (pralsetinib):

1. A phase I/II study (ARROW [NCT03037385]) enrolled patients with RET fusion−positive solid tumors. Two hundred thirty-three patients had RET fusion−positive NSCLC. RET fusion status was determined by local molecular testing of tumor or circulating tumor nucleic acid (ctDNA) in blood, without central confirmation. The primary end point was objective response.[47][Level of evidence C3]

   • Ninety-two patients who had received platinum-based chemotherapy and 29 patients who were treatment-naive (and not candidates for standard platinum-based treatment) received pralsetinib before the efficacy enrollment cutoff (July 11, 2019). Eighty-seven previously treated patients and 27 treatment-naive patients had centrally adjudicated baseline measurable disease, and thus formed the efficacy cohort.
   • The overall response rate was 61% (95% CI, 50%–71%) in the 87 patients who had received platinum-based chemotherapy, including complete responses in 6%. The median DOR was not reached (15.2 months–not estimable).
   • The overall response rate was 70% (95% CI, 50%–86%) in the 27 treatment-naive patients, including complete responses in 11%. The median DOR was 9.0 months (6.3–not estimable).
   • In the 233-patient safety cohort, 93% had treatment-related adverse events, including 48% with grade 3 or worse events. The most common grade 3 or worse treatment-related adverse events were neutropenia (18%), hypertension (11%), and anemia (10%). Dose reductions occurred in 38% of patients, and 6% discontinued treatment because of adverse events.

Tepotinib

Evidence (tepotinib):

1. An open-label phase II study (VISION [NCT02864992]) enrolled patients with MET exon 14 skipping mutations. The trial included 152 patients who received tepotinib (500 mg orally once daily). MET status was determined centrally either via liquid biopsy (from circulating free DNA obtained from plasma; n = 66) or via tissue biopsy (n = 60). Twenty-seven patients had positive results from both methods. The primary end point was objective response.[49][Level of evidence C3]

   • Among the 99 patients who had been followed for at least 9 months (i.e., the efficacy population), the objective response rate as assessed by independent review was 46% (95% CI, 36%–57%), with a median DOR of 11.1 months (95% CI, 7.2 –not estimable). Response rates were similar in the liquid biopsy and tissue biopsy groups.
   • Responses were similar regardless of prior therapy.
   • Grade 3 or higher adverse events occurred in 28% of patients, including peripheral edema in 7% of patients. Adverse events led to therapy discontinuation in 11% of patients.

Capmatinib

Evidence (capmatinib):

1. A phase II study (GEOMETRY [NCT02414139]) evaluated capmatinib (400 mg orally twice daily) in patients with MET exon 14 skipping mutations or MET amplification. MET status was determined centrally. A total of 364 patients were enrolled. The primary end point was overall response.[50][Level of evidence C3]

   • Of the 69 patients with MET exon 14 skipping mutations who had received one or two prior lines of therapy, the overall response rate was 41% (95% CI, 29%–53%). The median DOR was 9.7 months (95% CI, 5.6–13.0).
   • Of the 28 patients with MET exon 14 skipping mutations who had not received any prior treatment, the overall response rate was 68% (95% CI, 48%–84%). The median DOR was 12.6 months (95% CI, 5.6–not estimable).
   • Response rates in patients with MET amplification without the exon 14 skipping mutation did not meet the prespecified threshold for clinically relevant activity.
   • Grade 3 to 4 adverse events occurred in 67% of patients. The most common events, regardless of causality, were peripheral edema, nausea, vomiting, and increased creatinine. Adverse events led to therapy discontinuation in 11% of patients.

Adagrasib

Evidence (adagrasib):

1. KRYSTAL-1 (NCT03785249) was a phase I/II multiple expansion cohort clinical trial that investigated adagrasib (600 mg orally twice daily) in patients with advanced solid tumors. A phase II cohort of the trial included patients with NSCLC and KRAS G12C mutations who were previously treated with platinum-based chemotherapy and anti–programmed death 1 (PD-1) or anti–programmed death-ligand 1 (PD-L1) therapy. The cohort enrolled 116 patients; 98.3% had previously received both chemotherapy and immunotherapy. The primary end point was objective response assessed by blinded independent central review.[52][Level of evidence C3]

   • Among 112 evaluable patients, the confirmed objective response rate was 42.9% (95% CI, 33.5%–52.6%), including one complete response (0.9%). The median DOR was 8.5 months (95% CI, 6.2–13.8). Disease control occurred in 79.5% of patients (95% CI, 70.8%–86.5%).
   • The median PFS was 6.5 months (95% CI, 4.7–8.4), and the median OS was 12.6 months (95% CI, 9.2–19.2).
   • Among 33 patients with previously treated stable CNS metastases, the intracranial confirmed objective response rate was 33.3% (95% CI, 18.0%–51.8%).
   • Grade 3 or higher adverse events occurred in 44.8% of patients and included two deaths. The most common adverse events were diarrhea, nausea, fatigue, vomiting, dyspnea, increased creatinine, increased ALT, increased AST, and decreased appetite. Adverse events led to therapy discontinuation in 6.9% of patients.

The FDA approved adagrasib for the treatment of adult patients with KRAS G12C-mutated locally advanced or metastatic NSCLC, as determined by an FDA-approved test, who received at least one prior systemic therapy.[53]

Sotorasib

Evidence (sotorasib):

1. A phase II clinical trial (CodeBreaK 100 [NCT03600883]) evaluated sotorasib (960 mg orally once daily) in patients with KRAS G12C mutations who were previously treated with standard therapies. A total of 126 patients were enrolled. Of 124 evaluable patients, most (81%) received prior platinum-based therapy and inhibitors of PD-1 or PD-L1. The primary end point was overall response according to independent central review.[51][Level of evidence C3]

   • For the 124 evaluable patients, the overall response rate was 37.1% (95% CI, 28.6%–46.2%), including complete responses in 3.2%. The median DOR was 11.1 months (95% CI, 6.9–could not be evaluated). Disease control occurred in 80.6% of patients (95% CI, 72.6%–87.2%).
   • Median PFS was 6.8 months (95% CI, 5.1–8.2) and median OS was 12.5 months (95% CI, 10.0–could not be evaluated).
   • Grade 3 to 4 adverse events occurred in 20.6% of patients. The most common adverse events that were considered treatment related were diarrhea, nausea, increases in ALT or AST, and fatigue. Adverse events led to therapy discontinuation in 7.1% of patients.

The FDA approved sotorasib for the treatment of adult patients with KRAS G12C-mutated locally advanced or metastatic NSCLC, as determined by an FDA-approved test, who have received at least one prior systemic therapy.

Trastuzumab deruxtecan

Trastuzumab deruxtecan is an antibody-drug conjugate consisting of a humanized anti-HER2 monoclonal antibody linked to a topoisomerase I inhibitor.

Evidence (trastuzumab deruxtecan):

1. DESTINY-Lung01 (NCT03505710) was a multicenter, open-label, two-cohort, phase II clinical trial that investigated trastuzumab deruxtecan at a dose of 6.4 mg/kg of body weight. Enrolled patients had HER2-overexpressing or HER2-mutant unresectable or metastatic nonsquamous NSCLC with disease relapse during standard treatment or with disease refractory to standard treatment. Patients who had previously received a HER2 antibody or an antibody-drug conjugate were ineligible. However, those who had received a HER2 TKI were eligible. The primary end point was objective response as assessed by independent central review. Ninety-one patients with HER2-mutant NSCLC were enrolled; 95% had received previous platinum-based therapy, and 66% had received anti-PD-1 or anti-PD-L1 immunotherapy.[55][Level of evidence C3]

   • All 91 patients were evaluable for response. The confirmed objective response rate was 55% (95% CI, 44%–65%), including one complete response (1%). The median DOR was 9.3 months (95% CI, 5.7–14.7). Disease control was observed in 92% of patients (95% CI, 85%–97%).
   • The median PFS was 8.2 months (95% CI, 6.0–11.9), and the median OS was 17.8 months (95% CI, 13.8–22.1).
   • Among 33 patients with CNS metastases at baseline, 14 had previously received radiation therapy to the brain, and 19 had not. Of these patients, 8 (57%) and 10 (53%), respectively, had a partial response. The median PFS for patients with CNS metastases at baseline was 7.1 months (95% CI, 5.5–9.8), and the median OS was 13.8 months (95% CI, 9.8–20.9).
   • Grade 3 or higher drug-related adverse events occurred in 46% of patients. The most common adverse events were nausea, fatigue, alopecia, vomiting, neutropenia, anemia, diarrhea, decreased appetite, leukopenia, and constipation. The most common grade 3 or higher drug-related adverse events were neutropenia (19%) and anemia (10%).
   • Drug-related adverse events led to treatment discontinuation in 23 patients (25%) and included pneumonitis in 12 patients (13%) and interstitial lung disease in 5 patients (5%).
   • There were two drug-related deaths caused by interstitial lung disease.

The FDA granted accelerated approval to trastuzumab deruxtecan for patients with unresectable or metastatic NSCLC whose tumors have activating HER2 mutations, as detected by an FDA-approved test, and who have received a prior systemic therapy. This approval was based on objective response rate and DOR.

Nivolumab

Evidence (nivolumab):

In two phase III clinical trials, one conducted in patients with advanced platinum-pretreated squamous NSCLC and the other trial conducted in patients with nonsquamous NSCLC, nivolumab demonstrated a significant improvement in OS compared with the previous standard treatment of docetaxel chemotherapy.[56,57][Level of evidence A1] In addition, the rates of grade 3 and 4 treatment-related toxicity in both trials were significantly lower with nivolumab than with docetaxel. Of note, all patients enrolled in phase III studies of nivolumab had an ECOG performance status of 0 or 1; patients with autoimmune disease, symptomatic interstitial lung disease, or those receiving systemic immunosuppression were excluded from enrollment.

1. A randomized, open-label, phase III trial randomly assigned 272 patients with advanced squamous cell NSCLC who had received one regimen of platinum-containing chemotherapy to receive either nivolumab (3 mg/kg every 2 weeks) or docetaxel (75 mg/m² every 3 weeks), administered until disease progression.[56] The primary end point of this study was OS.

   • Nivolumab demonstrated a significant improvement in median OS compared with docetaxel (9.2 months vs. 6 months; P < .001). In addition, the response rate (20% vs. 9%; P = .008) and median PFS (3.5 months vs. 2.8 months; P < .001) favored nivolumab.
   • Rates of treatment-related toxicity were significantly lower with nivolumab than with docetaxel (all grades, 58% for nivolumab vs. 86% for docetaxel; grades 3–4, 7% for nivolumab vs. 55% for docetaxel).
2. A randomized, open-label, phase III trial included 582 patients with advanced nonsquamous NSCLC who had received one regimen of platinum-containing chemotherapy. Patients were randomly assigned to receive either nivolumab (3 mg/kg every 2 weeks) or docetaxel (75 mg/m² every 3 weeks), administered until disease progression.[57] Previous maintenance chemotherapy after first-line platinum-doublet was allowed; patients with EGFR mutations or ALK translocations were allowed to have received an additional regimen of therapy with a TKI. The primary end point of this study was OS.

   • Patients who received nivolumab had a significant improvement in median OS compared with patients who received docetaxel (12.2 months vs. 9.4 months; HR, 0.73; 96% CI, 0.59–0.89; P = .002). In this study, the response rate (19% vs. 12%; P = .02) but not median PFS (2.3 months for nivolumab vs. 4.2 months for docetaxel) favored nivolumab. The median DOR in patients was 17.2 months for nivolumab and 5.6 months for docetaxel.
   • Rates of treatment-related toxicity were significantly lower with nivolumab than with docetaxel (all grades, 69% for nivolumab vs. 88% for docetaxel; grades 3–4, 10% for nivolumab vs. 54% for docetaxel).

Both of these trials demonstrated long-term clinical benefit at the 2-year outcomes. The OS rates for nivolumab at 2 years compared with docetaxel in squamous NSCLC were 23% (95% CI, 16%–30%) versus 8% (95% CI, 4%–13%), and OS rates in nonsquamous NSCLC were 29% (95% CI, 24%–34%) versus 16% (95% CI, 12%–20%).[59] Ongoing responses at 2 years were observed in 10 (37%) confirmed responders with squamous NSCLC and 19 (34%) of 56 responders with nonsquamous NSCLC. No patient treated with docetaxel in either study had an ongoing response.

Nivolumab is now considered a standard second-line therapy for patients with metastatic NSCLC with progression on or after first-line platinum-based chemotherapy and is associated with improved survival and lower rates of toxicity than docetaxel. However, clinical trials of nivolumab to date have not enrolled patients with a history of autoimmune disease, interstitial lung disease, or an ECOG performance status higher than 1. Patients with active autoimmune conditions cannot be treated with nivolumab. Closely monitoring all patients for autoimmune toxicities from treatment is required. Specific algorithms for the management of autoimmune toxicity are included in the FDA label for nivolumab.

Pembrolizumab

Evidence (pembrolizumab):

1. In a phase I study with multiple expansion cohorts, pembrolizumab demonstrated significant activity with respect to response rate and DOR.[58][Level of evidence C3]

   • In the study, 495 patients received either pembrolizumab 2 mg/kg every 3 weeks, 10 mg/kg every 3 weeks, or 10 mg/kg every 2 weeks. No significant differences were seen among the different treatment schedules. Key exclusion criteria were autoimmune disease, history of pneumonitis, requirement for systemic immunosuppressive therapy, and a performance status higher than 1. The objective response rate was 19.4% (95% CI, 16.0%–23.2%), which included a response rate of 18.0% (95% CI, 14.4%–22.2%) in 394 previously treated patients and 24.8% (95% CI, 16.7%–34.3%) in 101 previously untreated patients. Median PFS was 3.7 months (95% CI, 2.9–4.1) for all patients, 3.0 months (95% CI, 2.2–4.0) for previously treated patients, and 6.0 months (95% CI, 4.1–8.6) for previously untreated patients. The median DOR was 12.5 months (range, 1.0–23.3 months) in all patients.
   • The study evaluated the efficacy of pembrolizumab in patients with high levels of PD-L1, as assessed by the anti-PD-L1 antibody clone 22C3. Using the cutoff of membranous staining in at least 50% of tumor cells in a validation group of 73 patients, the response rate was 45.2% (95% CI, 33.5%–57.3%), and the median PFS in this group was 6.3 months (95% CI, 2.9–12.5). Median OS was not reached at the time of publication.
   • The estimated prevalence of PD-L1 tumor staining from 1,143 screened patients, of whom 824 had evaluable samples, is as follows: 23.2% had 50% or more tumor cells with staining; 37.6% had between 1% and 49% tumor cells with staining; and 39.2% had less than 1% of tumor cells with staining.
   • The most common adverse events were fatigue, pruritus, and decreased appetite. Grade 3 or higher adverse events were reported in 9.5% of patients. Inflammatory and immune-mediated adverse events that occurred in more than 2% of patients were infusion-related reactions (3.0%), hypothyroidism (6.9%), and pneumonitis (3.6%).
2. In a phase II/III randomized clinical trial, patients with previously treated NSCLC with PD-L1 expression on at least 1% of tumor cells were randomly assigned (1:1:1) to receive pembrolizumab (2 mg/kg), pembrolizumab (10 mg/kg), or docetaxel (75 mg/m²) every 3 weeks.[60][Level of evidence A1] The primary end points were OS and PFS in the total population and in patients with PD-L1 expression on at least 50% of tumor cells. This study enrolled 1,034 patients; 345 of them were allocated to pembrolizumab (2 mg/kg); 346 were allocated to pembrolizumab (10 mg/kg); and 343 were allocated to docetaxel.

   • In the total population, median OS was 10.4 months with pembrolizumab (2 mg/kg), 12.7 months with pembrolizumab (10 mg/kg), and 8.5 months with docetaxel. OS was significantly longer for pembrolizumab (2 mg/kg) versus docetaxel (HR 0.71; 95% CI, 0.58–0.88; P = .0008) and for pembrolizumab (10 mg/kg) versus docetaxel (HR, 0.61; CI, 0.49–0.75; P < .0001).
   • In the total population, PFS was not prolonged in the pembrolizumab groups compared with the docetaxel group.
   • Among patients with at least 50% of tumor cells expressing PD-L1, OS was significantly longer with pembrolizumab (2 mg/kg) than with docetaxel (median, 14.9 months vs. 8.2 months; HR, 0.54; 95% CI, 0.38–0.77; P = .0002) and with pembrolizumab (10 mg/kg) than with docetaxel (median, 17.3 months vs. 8.2 months; HR, 0.50; CI, 0.36–0.70; P < .0001).
   • In the group of patients with at least 50% of tumor cells expressing PD-L1, PFS was significantly longer with pembrolizumab (2 mg/kg) than with docetaxel (median, 5.0 months vs. 4.1 months; HR, 0.59; 95% CI, 0.44–0.78; P = .0001) and with pembrolizumab (10 mg/kg) than with docetaxel (median, 5.2 months vs. 4.1 months; HR, 0.59; CI, 0.45–0.78; P < .0001).
   • Grade 3 to 5 treatment-related adverse events were less common with pembrolizumab than with docetaxel (43 [13%] of 339 patients given pembrolizumab (2 mg/kg), 55 [16%] of 343 patients given pembrolizumab (10 mg/kg), and 109 [35%] of 309 patients given docetaxel).

The FDA granted accelerated approval to pembrolizumab as a second-line therapy for patients with NSCLC whose tumors express PD-L1 (>50% staining as determined by an FDA-approved test) with progression on or after first-line chemotherapy. Patients with EGFR or ALK genomic tumor aberrations should have disease progression on FDA-approved therapies before receiving pembrolizumab (see the FDA label for pembrolizumab).

Atezolizumab

Evidence (atezolizumab):

1. Two international, randomized, open-label clinical trials (OAK [NCT02008227] and POPLAR [NCT01903993]) demonstrated efficacy and safety in a total of 1,137 patients with NSCLC who previously received platinum chemotherapy.[61,62][Level of evidence A1] Compared with docetaxel, treatment with atezolizumab in the intended patient population resulted in improved OS rates of 4.2 months in the OAK study and 2.9 months in the POPLAR study.

   • In the OAK trial, the median OS was 13.8 months in the atezolizumab arm (95% CI, 11.8–15.7) compared with 9.6 months in the docetaxel arm (95% CI, 8.6–11.2) (HR, 0.74; 95% CI, 0.63–0.87; P = .0004).
   • The median OS in the POPLAR trial was 12.6 months in the atezolizumab arm (95% CI, 9.7–16.0) and 9.7 months in the docetaxel arm (95% CI, 8.6–12.0) (HR, 0.69; 95% CI, 0.52–0.92).
   • Although the magnitude of improvement correlated with PD-L1 immunohistochemistry expression on tumor cells and tumor-infiltrating immune cells, survival benefit with atezolizumab was seen in patients with tumors with and without PD-L1 expression.
   • In the POPLAR trial, the most common (≥20%) adverse reactions were in patients treated with atezolizumab and included fatigue, decreased appetite, dyspnea, cough, nausea, musculoskeletal pain, and constipation.
   • The most common (≥2%) grade 3 to 4 adverse events in patients treated with atezolizumab were dyspnea, pneumonia, hypoxia, hyponatremia, fatigue, anemia, musculoskeletal pain, ALT and AST increase, dysphagia, and arthralgia.
   • Clinically significant immune-related adverse events for patients who received atezolizumab included pneumonitis, hepatitis, colitis, and thyroid disease.

Everolimus

Everolimus is used for patients with unresectable, locally advanced or metastatic, progressive, well-differentiated, nonfunctional, neuroendocrine tumors.

Everolimus, an oral mTOR inhibitor, is clinically active against advanced pancreatic and nonpancreatic neuroendocrine tumors.[63] Based on the results of the RADIANT-4 clinical trial,[63] the FDA approved everolimus for the treatment of adult patients with unresectable, locally advanced or metastatic, progressive, well-differentiated (low or intermediate grade), nonfunctional neuroendocrine tumors of lung or gastrointestinal origin.

Evidence (everolimus):

1. A randomized, double-blind, placebo-controlled, phase III trial (RADIANT-4 [NCT01524783]) evaluated everolimus in patients older than 18 years with advanced, progressive, well-differentiated, nonfunctional neuroendocrine tumors of lung or gastrointestinal origin.[63] Eligible patients were randomly assigned in a 2:1 ratio to receive everolimus 10 mg daily orally or placebo, both with best supportive care. A total of 302 patients were enrolled (205 in the everolimus arm and 97 in the placebo arm), including 90 patients with neuroendocrine tumors of lung origin (63 in the everolimus arm and 27 in the placebo arm). The primary end point was PFS assessed by central radiology review in the intention-to-treat population.

   • Median PFS was 11.0 months in the everolimus arm and 3.9 months in the placebo group (HR, 0.48; 95% CI, 0.35−0.67; P < .00001).[63][Level of evidence B1]
   • In a post hoc analysis of the lung subgroup, median PFS by central review was 9.2 months in the everolimus arm and 3.6 months in the placebo arm (HR, 0.50; 95% CI, 0.28−0.88).[63]
   • The objective response rate was 2% in patients who received everolimus and 1% in patients who received placebo. Disease stabilization was observed in 81% of patients in the everolimus arm and 64% of patients in the placebo arm.
   • The median duration of treatment was longer in the patients who received everolimus compared with those who received placebo (40.4 weeks vs. 19.6 weeks).
   • A planned interim analysis of OS showed a 36% reduction in the estimated risk of death with everolimus relative to placebo (HR, 0.64; 95% CI, 0.40−1.05). These results were not statistically significant.
   • The most common treatment-related adverse events were stomatitis, diarrhea, fatigue, infections, rash, and peripheral edema. The most common drug-related grade 3 or 4 adverse events were stomatitis, diarrhea, infections, anemia, and fatigue. Grade 3 or 4 adverse events resulted in treatment discontinuation in 12% of patients in the everolimus group and 3% of patients in the placebo group.