Ozanimod for treating moderately to severely active ulcerative colitis

3.1 Ulcerative colitis is a chronic, unpredictable condition characterised by relapsing and remitting periods of inflammation of the rectal and colonic lining. The patient experts explained that the symptoms of the condition and the side effects of treatments can have a profound and devastating impact on all aspects of a person's life. Symptoms can affect the ability to work, study, socialise, take part in leisure activities and have intimate relationships. They explained that fears of needing surgery or developing cancer or even dying can affect a person's mental and emotional wellbeing, leading to feelings of anxiety, stress, depression and hopelessness. They explained that the impact of living with ulcerative colitis can be exacerbated because of the stigma of having the condition and frustration from a lack of understanding of the condition by others. Unlike symptoms of gastroenteritis that may last for 1 or 2 days, symptoms of active ulcerative colitis can last for months. People with active disease may experience debilitating, multiple daily episodes of explosive, painful diarrhoea with blood loss and pus. The committee concluded that symptoms of ulcerative colitis and the side effects of treatments for moderately to severely active ulcerative colitis can have a profound impact on quality of life.

3.2 The clinical experts explained that management of ulcerative colitis is highly individualised. Factors including previous therapies, the need for concomitant medicines such as corticosteroids and thiopurines, tolerability, and personal preferences and circumstances such as planned pregnancy can affect treatment options. They explained that current treatments are not very effective at achieving and maintaining clinical remission, so treatment switching often occurs. Some treatments also have serious side effects, for example thiopurines are associated with an increased risk of cancer, and there are also some serious side effects associated with biological treatments. At initial diagnosis, people typically have conventional treatments such as corticosteroids, mesalazine and thiopurines (azathioprine and mercaptopurine). If the condition does not respond well enough or stops responding to conventional treatment, a biological treatment, usually a tumour necrosis factor (TNF)‑alpha inhibitor, commonly infliximab, is most often offered. The clinical experts explained that TNF‑alpha inhibitors should be used with thiopurines to be most effective. In about 30% of people, the condition does not respond to a TNF‑alpha inhibitor (primary non-response), and about 40% of people with the condition will lose response over 12 months (secondary non-response). For a minority of people, another TNF‑alpha inhibitor such as adalimumab or golimumab may be offered, but other options include vedolizumab, ustekinumab, tofacitinib, and filgotinib. These treatments have different mechanisms of action to TNF‑alpha inhibitors which people may benefit from. However, the clinical experts explained that at this point in the treatment pathway, response to further treatment, regardless of its mechanism of action, is likely to be lower. They emphasised the importance of using the most effective option for the person in the first instance. They explained that medicines that result in clinical remission are not stopped if they continue to be effective. They highlighted that an oral treatment, such as tofacitinib, is attractive because it is easy to take and any response can be judged after 1 or 2 weeks. However, response is usually assessed after a few months for TNF‑alpha inhibitors and ustekinumab. Importantly, tofacitinib can be used without corticosteroids or thiopurines. They highlighted that tofacitinib use is variable across England and suggested that uptake may have been affected by safety concerns about deep vein thrombosis (DVT). They explained that under-treated ulcerative colitis also increases the risk of DVTs and that tofacitinib has been effectively used in carefully selected people. The patient experts emphasised the importance of having access to a range of effective treatment options to avoid or delay the need for surgery. The committee acknowledged the need for new treatments that can be taken without concomitant medicines and that are fast-acting, well tolerated and easy to administer. The committee concluded that treatment options are highly individualised and that there is an unmet need for new treatments with different mechanisms of action.

3.3 The marketing authorisation for ozanimod is for people with moderately to severely active ulcerative colitis whose condition had an inadequate response, lost response, or were intolerant to either conventional therapy or a biological treatment. The company submission presented ozanimod at 2 positions in the treatment pathway:

3.4 The company proposed the following comparators:

3.5 The main clinical evidence for ozanimod came from a phase 3, double-blind, randomised, placebo-controlled, multicentre trial that included a 10‑week induction phase and a 42‑week maintenance phase. It included 1,012 adults (18 to 75 years) with moderately to severely active ulcerative colitis. This was defined by a 4‑component Mayo score of between 6 and 12, and component subscores of at least 1 for stool frequency and rectal bleeding, and at least 2 for endoscopic findings. It had a re‑randomised design and included 2 cohorts. Cohort 1 included 645 adults who were randomised to either ozanimod or placebo. Cohort 2 included an open-label group of 367 adults who had ozanimod. People taking corticosteroids maintained their dose during induction, and then tapered their dose on entering the maintenance phase. The key primary end point, the proportion of people in clinical remission at induction (week 10) and maintenance (week 52) used data from Cohort 1. In addition to results for the overall population, the company presented the results based on the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups. The clinical experts confirmed that the most important clinical end point is clinical remission. For both the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups, a greater proportion of people who had ozanimod experienced clinical remission than those in the placebo group at the end of induction and maintenance. These findings were all statistically significant except for the TNF‑alpha inhibitor-experienced subgroup at the end of induction. These results are considered academic in confidence by the company and cannot be presented here. The committee concluded that ozanimod improved clinical remission compared with placebo for both the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups.

3.6 The company confirmed that some people in the TNF‑alpha inhibitor-experienced subgroup had received more than 1 biological treatment. The committee noted that the company had not further stratified the TNF‑alpha inhibitor-experienced subgroup based on the number of previous biological treatments, that is, for second-line or third-line biological treatment. The ERG highlighted that an older group of adults may be missing in the trial. The clinical experts confirmed that people seen in the NHS are unlikely to be much older than people in the trial. They considered that the people in TRUENORTH are representative of people likely to have ozanimod in clinical practice. They explained that people in the TNF‑alpha inhibitor-experienced subgroup are likely to have more severe ulcerative colitis and that any treatment is likely to be less effective at this stage. The committee concluded that the people in TRUENORTH are representative of people that would be treated in NHS clinical practice.

3.7 The committee noted the high placebo response in TRUENORTH for both the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups. The clinical experts explained that high placebo responses are typically seen in these trials and that concomitant corticosteroid use may have a large impact on placebo response. The ERG noted that there was little detail on the use of concomitant medicines in the company submission. The committee noted that placebo response contributed to the heterogeneity across trials in the network meta-analysis.

3.8 Because there were no head-to-head trials, the company did network meta-analyses to compare the effectiveness of ozanimod with TNF‑alpha inhibitors (adalimumab, infliximab, golimumab), vedolizumab, ustekinumab, and tofacitinib. The network meta-analyses included 25 trials stratified by TNF‑alpha inhibitor experience (TNF‑alpha inhibitor-naive compared with TNF‑alpha inhibitor-experienced) and treatment phase (induction 6 to 14 weeks compared with maintenance 52 to 60 weeks). This was done using random effects ordinal models with a probit link to assess clinical remission and clinical response. The company assessed baseline risk in the placebo arm of its revised network meta-analyses using data from single trials in the network meta-analyses that it considered representative of UK clinical practice. This was in line with the ERG's approach. The ERG explained that overall, the company's revised network meta-analyses are in line with recommended practice and have accounted for some potential effect modifiers (mainly previous TNF‑alpha inhibitor treatment and differences in trial design). But it considered that there is uncertainty as to whether the most suitable source of evidence had been used to estimate baseline placebo risk. The committee recognised that the company tried to address the limitation of identifying suitable data sources to estimate baseline placebo risk during technical engagement but was unsuccessful. It concluded that although the approach to estimate baseline placebo risk was suboptimal, overall the company's revised network meta-analyses were appropriate for decision making.

3.9 The results of the company's induction and maintenance network meta-analyses showed that ozanimod is likely to be as effective as some comparators in the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups. The results are considered academic in confidence by the company and cannot be presented here. The clinical experts highlighted that different network meta-analyses may provide different rankings, and that it is difficult to choose a specific treatment based only on small differences in efficacy. They explained that this is because people respond differently, and no single treatment is very effective in terms of maintaining clinical remission. They emphasised that treatment choice is highly dependent on individual needs and preferences. The committee concluded that in the network meta-analyses, ozanimod did not stand out as being substantially more or less effective than the comparators.

3.10 The company used a Markov model to estimate the cost effectiveness of ozanimod compared with adalimumab, golimumab, infliximab, tofacitinib, ustekinumab and vedolizumab for the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups. The company's model structure was similar to those used in previous ulcerative colitis technology appraisals. It included health states defined by the type of treatment and degree of disease control, to replicate the relapsing and remitting nature of ulcerative colitis. The types of treatment were active treatment (including both induction and maintenance phases), post-active treatment (including surgery and post-surgery), and best supportive care, comprising components of conventional treatment. Degrees of disease control were remission, response without remission, and active ulcerative colitis. The Markov model had a lifetime time horizon and a cycle length of 2 weeks and included clinical response, clinical remission and serious infections. The ERG agreed that the company captured all relevant health states and that its approach was appropriate. The committee concluded that the company's model was appropriate for decision making.

3.11 The company modelled efficacy estimates for best supportive care in the post-active treatment phase. It used data from the placebo arms of the trials included in the network meta-analysis from the TNF‑alpha inhibitor-experienced subgroup to inform the transition probabilities for the TNF‑alpha inhibitor-naive subgroup. The company considered that the data from the TNF‑alpha inhibitor-experienced subgroup was more appropriate for the TNF‑alpha inhibitor-naive subgroup because people in the post-active treatment phase had already had at least 1 TNF‑alpha inhibitor. The ERG considered that subgroup-specific data should inform transition probabilities in the best supportive care arm. It considered that the loss of response and loss of response (no remission) transition probabilities should be based on both the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced estimates. The committee noted that the 2 approaches resulted in only a modest difference in the incremental cost-effectiveness ratio (ICER), but considered that subgroup-specific data is preferred.

3.12 The committee considered the cost effectiveness of ozanimod compared with adalimumab, golimumab, infliximab, tofacitinib, ustekinumab and vedolizumab for the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups. The company's and ERG's base case results included confidential commercial discounts, so they cannot be reported here. The committee considered that the ICERs are potentially unstable because of very small and uncertain differences in efficacy between the treatments that were used in the economic model. However, considering the TNF‑alpha inhibitor-naive and TNF‑alpha inhibitor-experienced subgroups separately, the committee concluded that overall, ozanimod could be considered a cost-effective option when conventional or biological treatments are not tolerated or working well enough. The committee noted the exception that infliximab was more cost effective than ozanimod in the TNF‑alpha inhibitor-naive subgroup. It concluded that ozanimod should only be offered to people in the TNF‑alpha inhibitor-naive subgroup if infliximab cannot be tolerated or is contraindicated.

3.13 The patient experts explained that for some religious groups the impact of active disease and the effects of surgery may interfere with religious practices and cause distress. The committee did not consider this an equality issue that could be resolved by this appraisal. No other equality or social value judgement issues were identified.

3.14 The company considered ozanimod to be innovative because it has a novel mechanism of action that uses sphingosine 1‑phosphate (S1P) receptors to decrease circulating lymphocytes to reduce inflammation. Ozanimod addresses an unmet need by providing people with a new therapeutic oral option to treat symptoms and induce remission. The committee acknowledged the benefits offered by ozanimod and that people value an oral treatment. But it noted that it had not been presented with evidence of any additional benefits that were not captured in the quality-adjusted life year (QALY) measurements. The committee concluded that the benefits of ozanimod were adequately captured in the model.