Study design

Patients with PAH who completed the PATENT-1 study were invited to participate in the open-label, single-group PATENT-2 study to assess the long-term safety and efficacy of riociguat. Patients from PATENT-1 were excluded from entering PATENT-2 if they were experiencing riociguat-related serious adverse events (SAEs) or they withdrew from PATENT-1 due to clinical worsening of their pulmonary hypertension.

PATENT-2 was conducted in 97 of the 124 centres involved in the PATENT-1 study, which was an eight-week, double-blind, dose-adjustment phase followed by the open-label portion whereby patients received riociguat doses of up to 2.5 mg three times daily. Permission was allotted to investigators to adjust doses in accordance with the occurrence of adverse events (AEs), systolic blood pressure findings, and progression of patient’s PAH.

Assessment

The assessment of the safety and tolerability of riociguat were the primary objectives of the PATENT-2 study and included observations of both AEs and certain laboratory parameters. Efficacy assessments included the following variables:

• Six-minute walk distance (6MWD) test
• N-terminal pro-brain natriuretic peptide (NT-proBNP)
• World Health Organization (WHO) functional class change
• time to clinical worsening
• Borg dyspnea score
• EuroQol 5-Dimensions (EQ-5D) questionnaire
• Living with Pulmonary Hypertension (LPH) questionnaire.

Outcome assessments were performed at PATENT-2 entry and at weeks two, four, six, eight, 12, and every three months thereafter. A follow-up assessment 30 days after any patient discontinued riociguat was also performed. All assessments were analyzed descriptively and were non-comparative. The start of the PATENT-1 study was considered baseline.

Results

Three hundred and ninety-six patients (98%) entered the PATENT-2 study out of the 405 patients who completed the PATENT-1 study. Of these patients, 50% were using additional PAH medications, 43% were using endothelin receptor antagonists (ERAs), 6% were using non-intravenous prostanoids, and 1% were using both ERAs and prostanoids. Seventy-seven per cent, 91%, and 79% of patients from the former placebo, riociguat 1.5 mg, and riociguat 2.5 mg groups, respectively, were receiving a dose of 2.5 mg riociguat three times daily at the end of the eight-week dose-adjustment period. This dosage was sustained at year one of PATENT-2 study in 86% of patients. In addition, 46% were also using ERAs, 4% were receiving prostanoids, and 4% were using both ERAs and prostanoids at year 1.

Of the 396 patients who started PATENT-2, 81%, 86%, and 81% of patients from the former placebo, riociguat 1.5 mg, and riociguat 2.5 mg groups, respectively, were still receiving treatment at the March 2013 cut-off point of this ongoing study. The median treatment duration was 91 weeks (mean 95 weeks) with a cumulative exposure of 718 person-years. The most common reasons for discontinuation included AEs, death, and lack of efficacy. Detailed patient disposition is provided in Table 23.

Safety

The majority of patients experienced AEs, with the most common being nasopharyngitis, dizziness, peripheral edema, cough, and gastrointestinal events. Syncope was reported in 7% of all patients, while hypotension was reported in 9%. Hemoptysis/pulmonary hemorrhage was reported as an AE in 6% of patients and as an SAE in 3% of patients throughout the PATENT-2 study. The PATENT-2 AE rate of hemoptysis/pulmonary hemorrhage per 100 patient-years was 6.3 versus 9.9 or 11.1 in the PATENT-1 riociguat or placebo groups, respectively. SAEs were reported in 52% of patients, leading to discontinuations in 6%, 7%, and 8% of the patients from the former placebo, riociguat 1.5 mg, and riociguat 2.5 mg groups, respectively. SAEs were graded as moderate in five patients and severe in another five patients. In addition, the PATENT-2 exposure-adjusted SAE rate of hemoptysis/pulmonary hemorrhage per 100 patient-years was 1.8 versus 4.3 in the PATENT-1 riociguat groups. A total of 27 deaths were reported in the study (7%), three of which were considered drug-related and included hypoxic-ischemic encephalopathy, hemoptysis/pulmonary hemorrhage and pneumonia, and cardio-respiratory arrest. Detailed harms data are presented in Table 24.

Exploratory efficacy analysis

Following the changeover to riociguat, considerable improvements in the 6MWD test were reported in the PATENT-1 placebo group along with sustained improvements reported at both week 12 and year one in the PATENT-1 riociguat groups. Year one mean changes (± standard deviation [SD]) in the 6MWD test were 53 m (± 70 m), 56 m (± 88 m), and 46 m (± 76 m) in the former riociguat 2.5 mg, riociguat 1.5 mg, and placebo groups, respectively. The absolute mean (± SD) 6MWD at year one in the overall population was 419 m (± 97 m) (n = 327) compared with the baseline value of 367 m (± 67 m) (n = 396). Lower improvements in 6MWD test results were noted upon sensitivity analyses for missing data using various imputation methods; however, the overall interpretation was not affected. At year one of PATENT-2, 36%, 57%, and 7%, of patients were reported to have improved, stabilized, or worsened, respectively, in their WHO functional status. At year one of PATENT-2, 8%, 61%, 28%, and 2% of the total population (n = 339) were observed as WHO functional class I, II, III, and IV, respectively. This was compared with WHO functional class I, II, III, and IV of 3%, 43%, 54%, and 1%, respectively, of the overall population (n = 395) at baseline. Similar to the 6MWD test results, WHO functional class missing data sensitivity analyses reported reduced improvements; albeit, this did not affect the interpretation of the study results.

Decreases in the PATENT-1 riociguat NT-proBNP were sustained for up to year one of PATENT-2. PATENT-1 placebo patients who had transitioned to riociguat in PATENT-2 also showed similar decreases in NT-proBNP. Improvements in both EQ-5D and Borg dyspnea scores in those taking riociguat in PATENT-1 were sustained up to year one of PATENT-2. In addition, LPH scores also improved though to week eight of the PATENT-2 study in patients having received either riociguat 2.5 mg or 1.5 mg maximum in the PATENT-1 study, while minimal changes were observed in those patients in the PATENT-1 placebo group. Detailed efficacy results are presented in Table 25.

At least one clinical worsening event was experienced by 21% of the patients during PATENT-2, with an estimated rate of clinical worsening–free survival of 88% (95% confidence interval [CI], 84% to 91%) and an estimated survival rate of 97% (95% CI, 94% to 98%) at one year (with the results assuming that patients who drop out have the same event rates as patients continuing, thus leading to potentially positive bias, especially if the dropout rate was, in fact, higher.) Worst-case sensitivity analyses of the rate of clinical worsening–free survival and rate of survival at one year were 84% (95% CI, 80% to 87%) and 90% (95% CI, 87% to 93%), respectively. These sensitivity analyses assumed that every patient who dropped out had died or experienced a clinical worsening event. Estimated one-year survival rates were also similar between treatment-naive and pre-treated (with ERAs and prostanoids) patients at 97% (95% CI, 93% to 99%). Detailed results regarding clinical worsening events are presented in Table 24.

Limitations

The main limitation of PATENT-2 remains the fact that it was an open-label, long-term extension study, with no comparator group. In addition, with approximately 50% of patients on concomitant PAH medications, it is difficult to ascertain the absolute safety and efficacy of only riociguat. While informative regarding the safety information, any analysis of efficacy remains exploratory. The potential for an overestimation of the one-year efficacy results is possible due to the fact that the patients who discontinued may not have been doing well on riociguat, leaving only those able to tolerate and, subsequently, do well long-term on riociguat remaining. In addition, AEs may have been underestimated and the safety data could also be biased for the same reasons. Recall bias could also have been a factor due to the decreased frequency of visits in PATENT-2 (every three months) compared with PATENT-1 (every two weeks).

Discussion

It appeared that the safety and tolerability results were similar across the PATENT-1 and PATENT-2 studies, as most patients were continuing riociguat 2.5 mg three times daily at the March 2013 year one cut-off. No new safety signals were observed and the most common AEs were nasopharyngitis, dizziness, peripheral edema, cough, and gastrointestinal events. Notable AEs of interest were syncope, hypotension, and hemoptysis/pulmonary hemorrhage, which occurred in 7%, 9%, and 6% of patients, respectively. Three per cent of patients in PATENT-2 compared with 1% of patients in PATENT-1 experienced hemoptysis/pulmonary hemorrhage as an SAE. The estimated one-year survival rate in PATENT-2 was 97%, with 21% and 10% of patients experiencing incidences of clinical worsening or hospitalization, respectively. There were 27 deaths in total, with three attributed to treatment.

Efficacy results were exploratory in nature. Improvements in both 6MWD test results and WHO functional class were sustained up until year one of the PATENT-2 study; however, the standard deviations were large in most cases. Caution is required when observing the efficacy results of the aforementioned efficacy parameters.

Summary

The safety and tolerability of riociguat 2.5 mg three times daily was similar in both the PATENT-1 study and the PATENT-2 extension study. Most patients experienced AEs (the most common being nasopharyngitis, dizziness, peripheral edema, cough, and gastrointestinal events), with fewer than 10% of all patients experiencing syncope, hypotension, or hemoptysis/pulmonary hemorrhage. The estimated survival rate at one year in PATENT-2 was 97%, with three deaths being attributed to riociguat treatment. The potential for selection bias was a possibility, potentially leading toward bias for patients who responded better to riociguat treatment. While improvements in the 6MWD test and WHO functional class were reported, caution is required for the interpretation of all efficacy variables due to the exploratory nature of the study and the non-comparative analysis.