5.2.1. Efficacy
The key reason that led to the negative recommendation by CEDAC following the original review of Kuvan was that patient details were insufficient to identify a subpopulation for which SAP may provide a significant clinical benefit that is cost-effective. Thus, the interpretation of the new clinical evidence provided in the resubmission requires careful consideration of whether or not the new information reduces uncertainty regarding which patients will ultimately benefit from SAP treatment and whether or not the outstanding questions arising from the original review have been addressed. Although sufficient evidence that SAP lowers blood Phe levels in certain patients with PKU was demonstrated in the original review of Kuvan, the clinical significance of the magnitude of the reduction and uncertainty regarding whether the reduction in Phe levels are associated with clinically meaningful outcomes are important considerations in interpreting the new clinical evidence.
Both the PKU-016 and SPARK trials included in the current review were associated with various limitations, and some common to both trials are the large placebo effect, short duration and small number of included patients, and the lack of validation of measured outcomes and MCIDs in patients with PKU. While the large placebo effect (or the effect of Phe-restricted diet alone) may, in part, be attributed to participation in a clinical trial and increased attention to dietary therapy, it underscores the fact that dietary therapy is an effective intervention for patients with PKU. In light of the variability in patient response to SAP and the uncertainty of the clinical significance of the magnitude of Phe blood level reductions provided by SAP, a Phe-restricted diet remains a key component of ongoing therapy for PKU. This is reiterated by the Health Canada—approved indication for Kuvan, which states that SAP should be used only in conjunction with a Phe-restricted diet.5 The relatively short duration of the included trials does not address the outstanding question regarding long-term safety and cognitive outcomes in the original CDR review. It is noted, however, that additional new evidence provided in the resubmission did include two registry studies15,18 and an ongoing, open-label, single-arm observational study14 of longterm SAP treatment, which have been summarized in Appendix 6. Despite the limitations of these studies (discussed in Appendix 6), it appears that SAP is associated with favourable long-term safety and tolerability and that with uninterrupted use, it is effective in sustaining reductions in Phe blood levels and increasing dietary Phe tolerance in Phe responders. The small number of patients included in the efficacy analyses populations of the included trials in the current review affect the generalizability of the results to a broader patient population and preclude identification of a subpopulation that may benefit from SAP treatment, which was the main reason for the negative recommendation following the original CDR review. Although the primary objective of PKU-016 was to assess the efficacy and safety of SAP in patients with PKU and ADHD symptoms, only 38 patients (i.e., approximately 32% of the total Phe responders in the trial, who in turn represented only about 57% of the entire trial population) met this definition. According to the clinical expert consulted on this review, ADHD is not highly prevalent among patients with PKU, so it is unclear why the investigators would have considered this to be a potential target population. Therefore, the new clinical information does not clearly identify a subpopulation that may benefit from SAP treatment. Overall, the interpretation of the clinical significance of the reported outcomes in both included trials is difficult, as the MCIDs for the reported outcomes are unknown and the majority of instruments used in the trials have not been validated in patients with PKU. Furthermore, neither included trial reported on outcomes that were identified as being most important to patients, were included in the review protocol, and which were also missing in the trials included in the original CDR review, namely quality of life and diet liberalization.
The first efficacy outcome in the current review protocol was quality of life; however, no quality of life outcomes were reported in either the PKU-016 trial or the SPARK study. According to the patient input received, patient expectations are that SAP will lower Phe levels and increase Phe tolerance so that patients are able to eat a more varied diet and be less isolated from their peers, both of which are expected to substantially affect patients’ quality of life. As per Appendix 7 and as corroborated by the clinical expert, a Phe-restricted diet is expensive, unpalatable, and extremely difficult to maintain. There is evidence that adherence to the diet wanes over time and with increasing age.19-21 Therefore, a clinically important treatment benefit of SAP could be its ability to facilitate diet liberalization by increasing dietary Phe tolerance. It follows that if SAP is able to maintain Phe blood levels within the desired range even when diet is not carefully restricted, this could have an important impact on patients’ quality of life. Unfortunately, neither included trial has provided evidence that the reduction in Phe blood levels observed in Phe responders translates into improved quality of life. The only trial to specifically measure change in dietary Phe tolerance was the SPARK trial, which included children four years of age and younger. These children were already within the target Phe blood level range of 120 to 360 µmol/L with dietary therapy alone. Therefore, this patient population may not be the most appropriate in this context, as it is expected that their adherence to diet is very good due to vigilant control by parents and caregivers at this age. As noted previously, although there was a statistically significant increase in dietary Phe tolerance with SAP + diet compared with diet alone in the SPARK study, this was not linked with improved quality of life, or any measure of diet liberalization. Of note, the long-term studies summarized in Appendix 6 also do not provide evidence of an association of improved Phe tolerance with diet liberalization.
In both the PKU-016 (Phe responders) and SPARK trials, it was shown that mean Phe blood levels were reduced over the trial durations in both the SAP + diet arms and the placebo + diet or diet alone arms, respectively. In PKU-016, no statistical comparisons were made between treatment groups, but at week 13 (randomized double-blind treatment period), mean Phe blood levels decreased by approximately 30% in the SAP + diet arm, while remaining relatively unchanged in the placebo + diet arm. Following crossover to open-label SAP in the second 13-week open-label treatment period, mean Phe levels in patients who received placebo + diet in the first part of the trial also fell by approximately 30% after crossover with levels at week 26 being similar between both treatment groups. Of note, the mean Phe blood levels at week 26 were ▬ (SAP + diet) and ▬ (placebo + diet) which both exceed the upper limit of the blood Phe target range (360 µmol/L) as per the ACMG guidelines.1 In the SPARK study, mean Phe levels were also reduced from baseline in both the SAP + diet and diet alone groups at week 26 (i.e., mean Phe levels were ▬, respectively). The difference between groups was not statistically significant; however, both groups remained in the Phe blood target range of 120 to 360 µmol/L, as they were at baseline. In both included trials, diet was to remain unchanged throughout the duration of the trials which was verified by review of dietary report cards at clinic visits. Thus, there is no evidence from the included trials to show that increased diet liberalization occurred due to the addition of SAP to dietary therapy. Based on the results of the diet alone arm in the SPARK trial, it appears that diet alone is able to maintain patients within the target Phe blood level range.1
The results pertaining to the magnitude of reduction in Phe blood levels associated with SAP are difficult to interpret as the MCID for Phe blood level reduction (i.e., Phe response) is unknown, as shown in Appendix 5. In addition, the per cent reduction in Phe blood levels used to identify Phe responders (i.e., 20% or 30%) is arbitrary, as confirmed by the clinical expert. The ACMG guidelines state that clinical judgment is required to determine what constitutes a significant or beneficial decline in Phe blood levels from baseline in an individual and that 30% is often cited as evidence of effective Phe reduction, although some centres do consider 20% reduction as a measure of response.1 In addition, although the ACMG guidelines recommend a target range of 120 to 360 µmol/L for mean Phe blood levels,1 there is controversy over the threshold for treatment, as it has been suggested that the target range should be less than 360 µmol/L up to age 12 years and then less than 600 µmol/L thereafter.8 There appears to be agreement that Phe blood levels of 360 µmol/L and less are not toxic to the brain and the clinical expert advised that in North America, the treatment aim is to maintain levels at below 360 µmol/L, although as shown in Appendix 8, it is suggested that in those aged 16 years and older, the target for Phe blood levels should be less than 600 µmol/L. Furthermore, as discussed in Appendix 5, there is also speculation that fluctuations in Phe levels are of potential clinical significance and that there may be up to 400% variation in day-to-day blood Phe levels in adults with well-controlled PKU, which may be influenced by age, genotype, rates of growth, dietary adherence, diet, and illness.22 There is also evidence that Phe concentrations in the blood are higher than in the brain and that the disposition of Phe in the brain does not appear to parallel that in the blood.22 Despite there being conflicting evidence regarding the effects of Phe fluctuations in patients with PKU,22,23 it is still maintained that continual blood Phe concentrations should be monitored and controlled in childhood and that this should continue throughout the life of the patient with PKU.22-24 It does not appear that there is yet widespread consensus regarding the threshold for initiating or maintaining treatment with SAP based on Phe blood levels; however, this is discussed in more detail below with regard to proposed reimbursement criteria.
One concern expressed by CEDAC in the original CDR review of Kuvan was that no neuropsychological outcomes were measured in the trials included in the review. In PKU-016, the effects of SAP treatment on neuropsychiatric and neurocognitive effects were investigated using various instruments (e.g., ADHDRS/ASRS, HAM-D, HAM-A, BRIEF). As shown in Appendix 5, only the Inattention subscale of the ADHDRS/ASRS appears to have been validated in children with PKU; however, no MCIDs for either the total score or either subscale score have been established.25 The change from baseline to week 13 in the ADHD-RS/ASRS total score in Phe responders with ADHD symptoms was a co-primary end point in PKU-016 and results showed that while the change from baseline in the individual treatment arms were both statistically significant (i.e., indicating less severity of ADHD symptoms in both groups), the difference between the SAP + diet treated patients and placebo + diet patients was not statistically significant. At week 13, for the subscale score of Inattention, the difference between arms was statistically significant, whereas for the subscale score of Hyperactivity-Impulsivity, the difference between arms was not statistically significant. At week 26, the difference between arms was not statistically significantly different for the ADHD-RS/ASRS total score or either subscale score. These results are also difficult to interpret, as it appears that, in general, ADHD symptoms improved from baseline to week 13 in all treatment arms, but that there were no differences between arms with the exception of the Inattention subscale score. These results are also complicated by the fact that the MCIDs have not been established for this instrument in patients with PKU and that a correction factor was applied to the ASRS score, which has not been used previously or validated in PKU or other disease areas.9 Other measures in PKU-016, such as the HAM-D and HAM-A, also showed a similar pattern as the ADHD-RS/ASRS total score, in that in individual treatment arms, the improvement from baseline to week 13 was statistically significant, but the difference between arms was not statistically significant.
Furthermore, in PKU-016 the results of the BRIEF assessment were not consistent. For the BRIEF-Parent results (i.e., patients younger than 18 years, for whom parents or caregivers completed the instrument), differences for the GEC and MI index scales were statistically significant between arms at week 13, but not for the BRI index scale. For the BRIEF-A (i.e., patients aged 18 years and older who self-completed the instrument), no statistically significant differences between arms were observed for any of the index subscales. It could be inferred from these findings that SAP does not appear to offer a significant additional benefit on improving neuropsychiatric outcomes compared with a well-controlled Phe-restricted diet. In the SPARK study, the effects of SAP treatment on four areas of neuromotor developmental milestones (i.e., fine motor, gross motor, language, and personal-social) were assessed. In all four areas, the majority of children were classified as being normal and there were no statistically significant differences between SAP + diet compared with diet alone reported in any area. This may be due to the children being within the Phe blood level target range at study entry or the short duration of the trial (26 weeks) precluding any significant change being observed in these assessments.
Another concern expressed by CEDAC in the original CDR review of Kuvan was that no information on growth parameters was provided in the included trials. The new clinical information did provide limited information on various growth parameters from the SPARK study where treatment differences were investigated for height, weight, BMI, and head circumference from baseline to week 26. ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬ ▬
The manufacturer has requested reimbursement criteria for Kuvan (as per section 1.3) that differ from the Exceptional Access Program (EAP) criteria (detailed in Appendix 8). In the manufacturer’s suggested criteria, ongoing funding of SAP should be considered in patients who demonstrate increased Phe tolerance or improvement in neurobehavioural or neurocognitive function or improved quality of life, whereas the EAP criteria require that patients demonstrate increased Phe tolerance and improvement in neurobehavioural or neurocognitive function and improved quality of life. The full EAP criteria, which also encompass initial funding of Kuvan for six months (which is funded by the manufacturer), the requirements for a 72-hour Kuvan challenge to determine eligibility, and the management of pregnant patients are provided in Appendix 8. Also included in Appendix 8 are alternate EAP reimbursement criteria for Kuvan proposed by Ontario PKU experts, as in their view, the current EAP criteria are difficult, if not impossible, to meet due to contradictory requirements and lack of resources to perform serial neurobehavioural/neurocognitive testing, as required by the criteria. The rationale and background for the changes proposed by the Ontario specialists are also provided in Appendix 8; however, the key differences between the respective criteria are the Phe blood level at which treatment should be initiated, the length of time required to determine a response to Kuvan, and the therapeutic Phe blood level to be maintained in order to continue funding of Kuvan. To reiterate, the initial six months of funding for Kuvan is provided by the manufacturer and the key criteria for this initial coverage are compliance with a Phe-restrictive diet and a baseline Phe blood level greater than 360 µmol/L (despite dietary compliance) based on two levels over three to six months. Eligibility for initial funding is assessed by completion of a “72-hour Kuvan challenge” with SAP 20 mg/kg/day. Patients must demonstrate a reduction in Phe blood level of at least 30% and also have a baseline assessment of neurobehavioural or neurocognitive impairment and quality of life. In contrast, the Ontario specialists recommend initial funding of Kuvan in patients based on at least one Phe plasma level of 600 µmol/L or greater in non-pregnant patients and a four-week Kuvan challenge to determine response to SAP 20 mg/kg/day. Response to Kuvan is similarly defined as a 30% decrease in blood Phe levels from baseline. The rationale for the higher threshold by the Ontario specialists is that there is a high degree of consensus that sustained Phe blood levels less than 360 µmol/L do not cause damage to the brain, but that levels 600 µmol/L and higher can cause significant damage.1,26,27 The longer four-week challenge is proposed as it is widely accepted and can detect more patients who are responsive to SAP. For continued ongoing funding of Kuvan, the current EAP criteria require dietary compliance, Phe blood levels between 120 and 360 µmol/L, or sustained reductions of blood Phe of at least 30% if baseline levels are less than 1,200 µmol/L or at least 50% if baseline levels are greater than 1200 µmol/L (all based on two levels measured at least one month apart), demonstrated increase in dietary protein (Phe) tolerance, and clinically meaningful improvement in neurobehavioural/neurocognitive function or impairment as determined by valid quality of life instruments. In comparison, the Ontario specialists recommend ongoing funding with Kuvan if Phe blood levels are consistently in the control range, which is defined as less than 360 µmol/L in more than 50% of monitoring samples and less than 600 µmol/L in more than 80% of monitoring samples in patients younger than 16 years of age, and as less than 600 µmol/L in more than 80% of monitoring samples in patients aged 16 years or older (i.e., monitoring samples should be collected at least monthly). The Ontario specialists also advocate removal of the requirements for dietary compliance, demonstration of increased dietary Phe tolerance, and the need to demonstrate neurobehavioural/neurocognitive or quality of life improvements. For recommendations pertaining to pregnant patients, patients planning pregnancy, or neonatal patients, please see Appendix 8. Based on the new clinical evidence provided in this review, the included trials both employed four-week time frames to determine Phe responders and the Phe blood level target range was identified as 120 to 360 µmol/L for the patient populations in both trials. Although PKU-016 did show that treatment arms had improvements from baseline in some neurobehavioural/neurocognitive outcomes at week 13, for the most part, the differences between arms were not statistically significant or were inconsistent.
