Tolvaptan for treating autosomal dominant polycystic kidney disease

Clinical effectiveness

4.4 The Committee was aware that the marketing authorisation for tolvaptan is for people with chronic kidney disease (CKD) stages 1 to 3 with rapidly progressing disease. It heard from clinical experts that there is no definition of rapidly progressing disease, but that in clinical practice they were able to identify people at greater risk of rapid decline in renal function using a variety of observations, such as baseline GFR, and a person's symptoms and quality of life over time. In addition, clinical experts stated that a person's risk of progression could be assessed according to the experience of progression by family members with ADPKD. The Committee concluded that clinicians would use a combination of clinical variables to identify patients in clinical practice who may be more likely to benefit from tolvaptan when used within its licensed indication.

4.5 The Committee noted that the company's additional evidence submission focused only on a proposed optimised subgroup of people with CKD stages 2 and 3. The Committee heard from clinical experts that this subgroup could be readily identified in clinical practice because CKD stage is routinely recorded in clinical practice. The Committee considered whether it was appropriate to only define the subgroup according to the stage of CKD. It noted that to be eligible for entry into TEMPO 3:4, patients had to have a total kidney volume (TKV) of 750 ml or more and the TKV in the optimised subgroup had increased considerably compared with the intention‑to‑treat population. It heard from the company and from clinical experts that TKV is not routinely recorded in clinical practice, and that the most accurate method to determine TKV is MRI, which may be associated with potential access issues. It was also mindful of its previous conclusion that several clinical variables are often used in clinical practice to identify people at risk of rapid progression (see section 4.4). The Committee heard from the patient and clinical experts that they considered it acceptable to narrow the eligible population by excluding people with CKD stage 1. The Committee concluded that CKD stage is routinely recorded in clinical practice, and that it was appropriate to consider tolvaptan for the subgroup of people with CKD stages 2 and 3 as well as for people within the broader licensed indication for tolvaptan.

4.6 The Committee considered whether the evidence for the clinical effectiveness of tolvaptan could be generalised to patients in clinical practice. It was aware that the main evidence in the company's submission came from the pivotal TEMPO 3:4 randomised controlled trial (n=1445) that compared tolvaptan with placebo. The Committee noted that a high number of patients who had been considered eligible for treatment had been excluded from the trial because they did not meet the inclusion criteria (TKV 750 ml or more, and an estimated glomerular filtration rate [eGFR] of 60 ml/min or more). It was also aware that the trial included a small percentage of CKD stage 3 (17%, n=247) patients. The Committee noted that the average baseline TKV was 1692 ml and the average eGFR was 82 ml/min/1.73 m² (measured with the Chronic Kidney Disease Epidemiology Collaboration [CKD‑EPI] equation). The Committee understood that these factors reflected a population with a greater probability of rapidly progressing disease, as defined by the inclusion criteria for the clinical trial. The Committee noted that for the subgroup presented by the company, the average baseline TKV was 2300 ml and the average eGFR was 60 ml/min/1.73 m² (midpoint of CKD stages 2 to 3). It considered that the subgroup reflected a population with an even greater probability of rapidly progressing disease than that of the intention‑to‑treat population. The Committee recalled that in the pivotal trial 5% of the patients were from the UK, although it also noted comments from the company that there were similarities in ethnicity between the trial population and people in England. The Committee noted that the age range of patients included in the trial was between 18 and 50. However, the marketing authorisation for tolvaptan has no upper age limit. The Committee noted comments from clinical experts stating that TEMPO 3:4 reflected UK clinical practice. The Committee concluded that TEMPO 3:4 was relevant to UK clinical practice and that the results could be used for decision‑making.

4.7 The Committee considered the most appropriate outcomes in the TEMPO 3:4 trial for measuring the progression of ADPKD and the relative treatment effect of tolvaptan. It was aware that the primary outcome was TKV, but that the NICE scope for the appraisal listed the rate of decline in renal function as the main outcome. The Committee understood from the company that rate of decline in renal function, as assessed by eGFR, was included as a secondary outcome, and it noted the post‑hoc power calculations presented by the company in its additional evidence submission that suggested it was very likely that the trial was adequately powered to assess this outcome. The Committee considered the approach to assessing eGFR in the trial, and noted that this had been undertaken using the reciprocal of serum creatinine in the primary analysis and was also estimated using the CKD‑EPI equation formula. The Committee heard from the clinical experts that CKD‑EPI is the accepted method in the clinical community for measuring eGFR and understood that this was the preferred measure for diagnosing, staging and estimating treatment effect in NICE's guideline on chronic kidney disease. The Committee concluded that the preferred measure of both the progression of ADPKD and the relative treatment effect of tolvaptan is eGFR as estimated using the CKD‑EPI formula.

4.8 The Committee considered the relative benefit of tolvaptan compared with placebo as reported in the TEMPO 3:4 trial. The Committee noted the relative reduction in the annual rate of renal decline, measured by CKD‑EPI, for tolvaptan compared with placebo of 26.4% for the intention‑to‑treat population. It also noted that the equivalent rate for the subgroup with CKD stages 2 and 3a was 29.7%. The Committee was aware that, out of 2122 patients who were assessed for eligibility in the trial, 677 patients (32%) had been excluded (530 of whom were excluded due to the inclusion criteria). The Committee expressed concern that data were not available for a large number of people for whom tolvaptan could have been considered suitable in clinical practice. The Committee accepted, however, from the 'tipping point' and 'jump to placebo' analyses (see section 3.54) presented by the company that the impact of missing data had not negatively affected the overall conclusions about the efficacy of tolvaptan based on the TEMPO 3:4 study. The Committee concluded that tolvaptan offers clinical benefit compared with standard care in both the intention‑to‑treat population and in the subgroup with CKD stages 2 and 3a.

4.9 The Committee noted that in the company's additional evidence submission, it presented clinical effectiveness evidence for only those people with CKD stages 2 or 3a, and that people with CKD stage 3b had been excluded. It understood from the company that this was an oversight and that the subgroup should have contained people with CKD stages 2, 3a and 3b. The Committee understood that for people with CKD stage 3, eGFR ranged from 30–59 ml/min, and that this group could be subdivided into CKD stages 3a (eGFR of 45–59 ml/min) or 3b (eGFR of 30–44 ml/min). The Committee accepted that the company had submitted evidence showing that the difference in treatment effect when people with CKD stage 3b were added to the population was marginal, and that it improved the cost effectiveness of tolvaptan. The Committee was mindful that the ERG had not had an opportunity to review this evidence, nevertheless it concluded that when considering the company's revised base case it was appropriate to consider all patients with CKD stages 2 and 3 (including stages 3a and 3b).

4.10 The Committee was aware of the requirement for liver function testing for patients taking tolvaptan, as detailed in the summary of product characteristics. It noted that significant abnormal liver‑function test results (determined by Hy's law; see section 3.7) were recorded for 3 people across the TEMPO 3:4 and TEMPO 4:4 trials. The Committee considered the potential for serious liver injury to be a concern with tolvaptan treatment, but noted that the effect reversed after discontinuing the drug. The Committee also noted comments from the company in its additional evidence submission that there had been no cases of severe drug‑induced liver injury (Hy's law) since the frequency of monitoring had been increased after the initial Hy's law cases, although it understood that the possibility of future Hy's law cases could not be ruled out. The Committee was aware from the patient experts that the main adverse reaction is thirst, and that in the patients' own personal experience it is necessary to drink at least 6 litres of water each day to overcome this thirst. The patient experts stressed that, over time, people can adjust to drinking this quantity of water. The Committee concluded that tolvaptan is associated with adverse reactions and effects, the more serious of which can be avoided through increased monitoring, and that some people do not need additional clinical assistance.

Cost effectiveness

4.11 The Committee discussed the economic model developed by the company for this appraisal. It considered that the model was acceptable for assessing the cost effectiveness of tolvaptan. The Committee noted that in the additional evidence submission the company had provided analyses for the optimised subgroup, but had not provided analyses for the intention‑to‑treat population. It noted that in the ERG's critique of the additional evidence submission, it had carried out additional exploratory analyses to understand the cost effectiveness of tolvaptan in the intention‑to‑treat population. In these exploratory analyses, the ERG reported the incremental cost‑effectiveness ratio (ICER) for tolvaptan compared with standard treatment of £43,500 per quality‑adjusted life year (QALY) gained (with the patient access scheme; see section 3.56).

4.12 The Committee considered the company's additional evidence containing the revised base case (including the revised patient access scheme) for a subgroup with CKD stages 2 and 3, acknowledging that the company's analyses resulted in a base‑case ICER for tolvaptan compared with standard treatment of approximately £23,500 per QALY gained (see section 3.50). The Committee noted that the ICER presented in the ERG's additional analyses for the subgroup was approximately £30,000 per QALY gained (see section 3.56). The Committee understood that there were several differences in the assumptions adopted by the company in its additional evidence submission compared with those preferred by the ERG in its critique of the company's additional evidence, including assumptions relating to: treatment‑related utility decrement for tolvaptan; and probability of kidney pain for tolvaptan and placebo. The Committee discussed each of these in turn.

4.13 The Committee noted that the company used a utility decrement for tolvaptan treatment of 0.0025 whereas the ERG's additional analyses used a value of 0.0123. The Committee noted the ERG's view that the value of 0.0123 was conservatively incorporated because no adverse events other than kidney pain were incorporated in the model. The Committee understood that the company had estimated the utility decrement of 0.0025 by assuming the QALYs gained because of pain reduction were equal to the QALYs lost because of the negative effect of being on tolvaptan. The Committee was not convinced that this was a valid assumption; however, it was also aware that there was little evidence to support the utility decrement of 0.0123 applied by the ERG. The Committee heard from the patient experts that although they experienced some treatment‑related adverse reactions, their bodies had adapted to the drug and therefore it had a small effect on their quality of life. The Committee also heard from patient experts that people who have a large reduction in their quality of life as a result of tolvaptan would be likely to stop treatment and therefore the quality of life decrement would be relatively small for long‑term patients. The Committee concluded that the true utility value decrement as a result of tolvaptan treatment was unknown, but that it was likely to be less than 0.0123 and may diminish over time.

4.14 Regarding the probability of kidney pain, the Committee noted that in the company's additional evidence for its revised base case it had not applied an equal probability of kidney pain for tolvaptan and placebo (see section 3.48) because it had stated that this contradicted the findings of TEMPO 3:4. The Committee also noted the ERG's comments that the company's assumption inferred that the difference in kidney pain as observed in TEMPO 3:4 was independent from the effect of tolvaptan on disease progression. The Committee noted the ERG's view that pain was a known symptom of chronic kidney disease, increasing with disease progression, and that the separate modelling of pain may have led to double counting (that is, the higher utility for lower CKD stages may have already been captured by the effect of kidney pain). The Committee noted comments from clinical experts that kidney pain is not necessarily reflective of CKD stage and that reduction in pain could be seen as an effect of the drug because of the reduction in kidney size. The Committee was aware that the effect of this assumption was very small (see ERG exploratory analysis in section 3.27). The Committee concluded that the conservative approach incorporating an equal kidney pain probability for both arms was appropriate for the base case.

4.15 The Committee discussed the health‑state utility values used by the company in the model. It noted that health‑related quality of life was not assessed in the TEMPO 3:4 trial and the company used health‑state utility values from a study published by Gorodetskaya et al. (2005), identified from a literature search (see section 3.13). The Committee noted that the utility values published by Gorodetskaya et al. were not ADPKD specific, and consequently, it agreed that the results using these utility values were associated with a considerable degree of uncertainty. However, in a late submission that the ERG did not have the opportunity to critique, the company presented an analysis using EQ‑5D data from patients with ADPKD from the OVERTURE trial (details of the OVERTURE study are presented in section 3.6; however, details of the utility values are not presented because these were designated confidential by the company). In these analyses, health‑state decrements were modified in the model for CKD stages 1 to 4. The results of these analyses found there was only a small effect on the ICER, which the company stated was in line with the low degree of sensitivity demonstrated in scenario analyses on the health‑state utility decrements for CKD stages 3 and 4 provided in the company's original submission. The Committee concluded that although the analyses presented had not yet been critiqued by the ERG, they were persuaded that the ADPKD‑specific EQ‑5D data from OVERTURE were unlikely to significantly alter the outcome of the revised base‑case analysis.

4.16 The Committee considered whether it was appropriate to model the Hy's law cases, noting comments from the ERG that the possibility of future Hy's law cases cannot be eliminated. However, the Committee was mindful of its previous conclusion that the possibility of such adverse effects could be reduced by increased monitoring, The Committee also understood that liver biochemistry monitoring was relatively infrequent in the TEMPO studies, and that more frequent monitoring would be expected in clinical practice, which would further lower the risk of liver failure. The Committee concluded that the inclusion of Hy's law cases in the ERG's exploratory analyses reflected a 'worst‑case' scenario and with the additional monitoring measures in place it was reasonable not to include Hy's law cases in the base case.

4.17 The Committee considered the most plausible ICER for tolvaptan compared with standard care for adults with CKD stages 1 to 3. It noted that the company had not presented an estimate of the ICER for this population in its revised base case, and that the ERG had estimated an ICER of £43,500 per QALY gained (see section 3.56). The Committee was aware that this estimate had not included all of its preferred assumptions, but even accounting for this, the Committee considered that the most plausible ICER was still not in the range normally considered to be a cost‑effective use of NHS resources and concluded that it could not recommend tolvaptan for people with CKD stages 1 to 3. The Committee considered the most plausible ICER for tolvaptan compared with standard care for adults with CKD stages 2 to 3. The Committee noted that in the ERG's additional exploratory analyses, it had presented an ICER for tolvaptan compared with standard treatment of approximately £30,000 per QALY gained. The Committee estimated that this ICER was likely to have overestimated the most plausible ICER for 2 reasons:

4.18 The Committee noted comments from the company in its submission and from the clinical experts about tolvaptan being an innovative treatment. The company stated that tolvaptan represents a 'step‑change' in managing ADPKD, because this is the first drug available that slows cyst growth and reduces the decline in renal function. The company emphasised that this is an area of high unmet medical need and the burden of the disease can be extremely high for people. The company further stated that tolvaptan has a significant and substantial impact on health‑related benefits, and it can delay time to end‑stage renal disease and reduce the strain on renal replacement therapy resources. The Committee heard from clinical experts that tolvaptan represents a step‑change in treatment and from the patient experts that it may also have a positive psychological benefit for people with ADPKD. The Committee understood the importance of such benefits, which may be difficult to capture in measures of health‑related quality of life in addition to those already included in the QALY calculations. The Committee concluded that tolvaptan is an innovative treatment and it is the first treatment that has been shown to specifically impact on the progression of ADPKD. The Committee considered that the most plausible ICER for adults who have CKD stages 2 to 3 was approximately £23,500 per QALY gained. Taking all of these factors into account, the Committee concluded that tolvaptan represented a cost‑effective use of NHS resources in adults who have CKD stages 2 to 3. The Committee therefore recommended tolvaptan as an option for treating ADPKD to slow the progression of cyst development and renal insufficiency only in adults who have CKD stages 2 to 3 at the start of treatment and evidence of rapidly progressing disease.

4.19 The Committee considered whether there were any equality issues associated with recommending tolvaptan for people with CKD stage 2 and 3, considering that people with CKD stage 1 would not get access to treatment, and whether this could be considered unfair. It heard from clinical and patient experts that this was not an equality issue, and that people with CKD stage 1 would eventually progress to CKD stages 2 and 3 and therefore would become eligible for treatment. The Committee also heard from patient experts who considered it fair to exclude people with CKD stage 1 so that people with CKD stages 2 and 3 could gain access to the treatment. The Committee considered that people with CKD stage 1 did not differ from people with CKD stage 2 and 3 as far as any protected characteristics are concerned. It concluded that there was no unfairness or unlawful discrimination, and as a result there were no equality issues associated with recommending tolvaptan for use in patients with CKD stages 2 and 3 with high risk of progression.

4.20 The Committee considered whether it should take into account the consequences of PPRS 2014, and in particular the PPRS payment mechanism, when appraising tolvaptan. The Committee noted NICE's position statement in this regard, and accepted the conclusion 'that the 2014 PPRS payment mechanism should not, as a matter of course, be regarded as a relevant consideration in its assessment of the cost effectiveness of branded medicines'. The Committee heard nothing to suggest that there is any basis for taking a different view on the relevance of the PPRS to this appraisal. It therefore concluded that the PPRS payment mechanism was not applicable when considering the cost effectiveness of tolvaptan.

Summary of Appraisal Committee's key conclusions