Upadacitinib for treating moderately to severely active ulcerative colitis

3.1 Upadacitinib is a type of treatment called a Janus kinase (JAK) inhibitor. It has a marketing authorisation for treating moderately to severely active ulcerative colitis when conventional or biological treatment cannot be tolerated, or if the disease has not responded well enough or stopped responding to treatment. The committee noted that choice of treatment for ulcerative colitis is highly individualised based on a range of factors. The company's submission considered 2 subgroups:

3.2 The clinical-effectiveness evidence for upadacitinib comes from placebo-controlled, double-blind randomised controlled trials:

3.3 At the end of induction treatment, the rate of remission was statistically significantly higher in the upadacitinib 45 mg group than the placebo group (adjusted treatment difference compared with placebo of 22% and 29% in the 2 studies) for the overall population. Remission rates were consistent in the biologic-naive and biologic-exposed subgroups. At week 52 of the maintenance phase, a statistically significantly greater proportion of people who had upadacitinib were in remission compared with those who had placebo (adjusted treatment difference compared with placebo of 31% [15 mg upadacitinib] and 39% [30 mg upadacitinib]). Remission rates were consistent in the biologic-naive and biologic-exposed subgroups. Clinical experts noted that the trials show upadacitinib will provide clinically meaningful benefits. The committee concluded that upadacitinib is more effective than placebo at inducing and maintaining remission.

3.4 In the induction trials, the most common adverse events with upadacitinib were creatine phosphokinase increase, acne and nasopharyngitis. In the maintenance trial these were nasopharyngitis, worsening of ulcerative colitis and creatine phosphokinase increase. Discontinuation from the trial because of adverse events was more common with placebo than with upadacitinib. There were no deaths in any trial. The committee was aware of the European Medicines Agency safety committee review of JAK inhibitors for inflammatory disorders (including upadacitinib for ulcerative colitis). This is because a clinical trial of tofacitinib in rheumatoid arthritis showed people with risk of heart disease were more likely to experience a major cardiovascular problem and had a higher risk of developing cancer with tofacitinib than TNF-alpha inhibitors. The company submitted a summary of the most recent data from a long-term extension study of upadacitinib (U‑ACTIVATE). No new safety risks were identified in people who completed maintenance treatment in U‑ACHIEVE and continued upadacitinib in the extension study. The committee concluded that based on the evidence presented, upadacitinib has an acceptable safety profile.

3.5 Because there were no head-to-head studies, the company did network meta-analyses of 2 randomised placebo-controlled trials of upadacitinib (U‑ACHIEVE induction and maintenance, and U‑ACCOMPLISH induction) and 18 trials of comparators. The network meta-analyses for efficacy endpoints assessed clinical remission and clinical response in 2 subgroups:

3.6 The company's network meta-analyses suggested that upadacitinib is as effective and sometimes more effective than comparators in the biologic-naive and biologic-exposed subgroups. The company considers the results confidential so they cannot be reported here. The company's safety network analysis suggested that upadacitinib has a low risk of serious infection. The committee concluded that upadacitinib was at least as effective as its comparators, and has a similar low risk of serious infection.

3.7 The company estimated the cost effectiveness of upadacitinib using a model with a hybrid structure (the induction phase was modelled using a decision tree and the maintenance phase was modelled using a Markov structure). The company provided cost-effectiveness estimates for the biologic-naive and biologic-exposed subgroups. The company's model structure was similar to those used in earlier ulcerative colitis technology appraisals. It included health states defined by remission and response without remission (in which people in the model had maintenance treatment with upadacitinib and comparators), active ulcerative colitis (with no biological drug treatment), and surgery and post-surgery. The company base case did not include a subsequent biological therapy. The EAG stated that it was not plausible for the company to assume that after having only 1 treatment people would move to an 'active ulcerative colitis' health state for the remaining duration of the model. Instead, in clinical practice, people would either be offered surgery within 12 months or other drug treatments. Clinical advisers to the EAG suggested these other medicines may include the treatment which previously gave the best symptom relief, even if the person's disease had not responded to it. The committee noted that the company's approach of not modelling further treatment after first-line treatment exaggerates the differences between treatments over a lifetime. The committee agreed that the company's model did not reflect NHS clinical practice. The committee added that it expects company submissions in future appraisals to reflect more up-to-date practice, recognising there are many treatments available for ulcerative colitis. This should include a structured decision-analytic model for ulcerative colitis that realistically represents what happens in NHS clinical practice. The committee concluded that the company's modelled treatment pathway did not reflect NHS clinical practice.

3.8 The EAG preferred to adapt the company's model using an alternative health state, 'on subsequent treatment', after treatment failure in its base case. In this, people had a 'basket' of biologic treatments (the comparators). The committee noted that this alternative approach led to different costs and quality-adjusted life years (QALYs) being accrued. The EAG acknowledged that its own modelling approach had some limitations. It did not include surgery and it allowed for a more expensive second-line treatment option than might be used in clinical practice. But it did more closely reflect NHS practice than the company's approach. The committee highlighted that a preferable model would include likely sequences of treatment that are used in NHS practice, but acknowledged that it is unlikely that data exists to populate such a model. The committee agreed that neither treatment pathway reflected NHS practice but concluded that the EAG's simplified adjustment of the model was preferred for decision making.

3.9 The company's utility values for remission, response without remission and active ulcerative colitis came from Woehl et al. (2008). The company noted that this was consistent with previous appraisals in ulcerative colitis. It suggested that utility values collected in real-world clinical practice (as done in Woehl et al. [2008]) can provide more representative values of the population having treatment than those collected in clinical trials. The EAG preferred to use health state utility values from EQ‑5D data collected in the upadacitinib trials in its base case, in line with the NICE reference case. It noted that these utility estimates were mostly higher than in the source used by the company. The committee noted that the Woehl et al. (2008) data used by the company had been considered in previous ulcerative colitis appraisals but that the reliability of the utility estimates had also been a source of discussion. The committee noted that utility values for remission, response without remission, and active ulcerative colitis health states were collected in upadacitinib trials and therefore could have been used. The committee concluded that using the upadacitinib trial-based utility estimates is preferred.

3.10 The committee recalled that upadacitinib maintenance treatment is 15 mg ('standard') or 30 mg ('high') daily dosing. The company and EAG agreed that, as was done for comparator drugs, it would be assumed that upadacitinib is prescribed in a 70:30 ratio of 'standard' to 'high' doses in the maintenance phase. The committee was concerned that this assumption was not based on evidence and that the true proportions of people who would have standard or high-dose levels of these treatments was uncertain. However, it noted that this assumption was consistent with the approach taken in NICE's technology appraisal guidance on ustekinumab. The committee reviewed alternative explorations of the ratio and noted these did not have a substantial impact on costs or effects. The committee concluded that some people have a high dose of maintenance treatment, but the proportion is uncertain.

3.11 Fully incremental analyses were done on the company's and EAG's preferred base cases by prior biological treatment status. For the biologic-naive subgroup, the analyses compared upadacitinib with adalimumab, adalimumab biosimilar, golimumab, infliximab, infliximab biosimilar, tofacitinib, ustekinumab and vedolizumab. For the biologic-exposed subgroup, upadacitinib was compared with adalimumab, adalimumab biosimilar, tofacitinib, ustekinumab and vedolizumab. The committee recalled that there was some uncertainty associated with the results of the network meta-analyses (see section 3.5). It also recalled that neither the company's nor the EAG's model truly reflected NHS clinical practice but that the EAG's simplified approach was preferred (see section 3.8). The proportion of people who have high-dose maintenance treatment in clinical practice is also uncertain (see section 3.10). Looking at the cost-effectiveness results from the EAG's model, upadacitinib had the greatest net health benefit suggesting that it is a cost-effective use of NHS resources compared with existing NICE-recommended treatments. However, the committee noted that the differences between costs and QALYs were very small, and noted the uncertainties described above. The committee agreed it was likely that upadacitinib is a cost-effective use of NHS resources when conventional or biological treatments are not tolerated or are not working well enough. The committee concluded that upadacitinib is considered cost effective for treating moderately to severely active ulcerative colitis.

3.12 The committee recalled that treatment of ulcerative colitis is highly individualised based on a range of factors. It also recalled that upadacitinib is likely to be a cost-effective use of NHS resources. But it noted that because the cost-effectiveness estimates were similar it could not consider it in preference to other options. This is because the value that each provides is uncertain and is likely to vary from case to case. It recalled that a range of treatments are already available for people with moderately to severely active ulcerative colitis, including some that have not yet been widely used in clinical practice (see section 3.5). The committee concluded that healthcare professionals should choose the most appropriate treatment after discussing the advantages and disadvantages of the treatments available with the person having treatment. If patients and clinicians consider upadacitinib to be one of a range of suitable options, they should choose the least expensive treatment (taking into account drug administration costs, dose needed and frequency, and product price per dose). This may vary from person to person because of differences in how the drugs are taken and treatment schedules.