CADTH Reimbursement Review

Upadacitinib (Rinvoq)

Sponsor: AbbVie Corporation

Therapeutic area: Crohn disease

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Table 1: Background Information of Application Submitted for Review

NOC = Notice of Compliance.

Introduction

Crohn disease (CD) is a chronic and progressive form of inflammatory bowel disease (IBD) that leads to significant disability and negatively affects a patient’s health-related quality of life (HRQoL).^1-3 It is characterized by recurrent, uncontrolled inflammation that can affect any part of the gastrointestinal (GI) tract from mouth to anus, and primarily affects the ileum, colon, and rectum.^4,5 Common symptoms include diarrhea, abdominal pain, fatigue, fever, rectal bleeding, loss of appetite, weight loss, and malnutrition.^6,7 Complications associated with CD include bowel obstructions, fistulas, anal fissures, intra-abdominal and other abscesses, and ulcers.^8,9

Canada has the highest incidence and prevalence of IBD in the world.^10,11 According to the 2018 Impact of Inflammatory Bowel Disease in Canada report, approximately 270,000 Canadians were living with IBD, of whom 135,000 had CD. For every 100,000 Canadians, 16.3 new cases of CD are diagnosed each year.¹²

Diagnosis of CD requires a combination of a medical history and physical examinations.⁶ Key biomarkers used in laboratory tests include C-reactive protein (CRP), erythrocyte sedimentation rate, and fecal calprotectin.^6,13-15 In active CD, CRP levels can range from 5 mg/L to 20 mg/L, depending on disease severity, compared to typical baseline levels of 1 mg/L.¹⁵ Diagnostic imaging, including endoscopy, radiology, and ultrasound, are also used to examine findings of typical CD cases, including bowel obstruction, stenosis, fistulae, abscesses, and atrophy.^14,16-19

Upadacitinib is a selective and reversible Janus kinase (JAK) inhibitor engineered to have greater inhibitory potency against JAK1 proteins versus JAK2, JAK3, and TYK2 in human cellular assays. Upadacitinib preferentially inhibits signalling by JAK1 or JAK1/JAK3 with functional selectivity over cytokine receptors that signal via pairs of JAK2 proteins.^20,21

The indication for upadacitinib is for the treatment of adult patients with moderately to severely active CD who have demonstrated prior treatment failure (i.e., an inadequate response, loss of response, or intolerance to at least 1 conventional and/or biologic therapy). The sponsor’s reimbursement request is consistent with the indication.

Upadacitinib is administered orally. The recommended dosing schedule involves 2 phases; a 12-week induction phase consisting of 45 mg upadacitinib once daily, followed by an ongoing maintenance phase. The recommended dosage of upadacitinib for maintenance treatment of patients aged between 18 and 64 years is 15 mg or 30 mg once daily, based on patient presentation; a dosage of 30 mg once daily may be appropriate for patients with high disease burden (such as refractory or severe disease) or those who do not show adequate therapeutic benefit with 15 mg once daily. The lowest effective dose for maintenance should be used. For patients who are aged at least 65 years, the recommended maintenance dosage is 15 mg once daily. In patients responding to induction or maintenance treatment with upadacitinib, corticosteroids may be reduced and/or discontinued in accordance with standard of care.^22,23

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient and clinician groups who responded to CADTH’s call for input and from 1 clinical expert consulted by CADTH for the purpose of this review.

Patient Input

Two patient groups, Crohn and Colitis Canada (CCC) and the GI Society, provided input for this review. CCC gathered information from a report published in 2018 (Impact of Inflammatory Bowel Disease in Canada), a survey, and interviews with patients who participated in the upadacitinib clinical trials. The patient input provided by the GI Society was based on surveys, interviews, a patient roundtable, and media interactions.

The 2 patient groups emphasized that CD has a tremendous impact on every aspect of a person’s life. The most frequent symptoms associated with CD reported by the patients are diarrhea, rectal bleeding, abdominal pain, and weight loss. Other symptoms included inflammation of the eyes or joints, ulcers of the mouth or skin, tender and inflamed nodules on the shins, anemia, anxiety, and stress. Both CCC and the GI Society stated that being unable to predict when the next urgent need for a bowel movement would occur, and the inability to control the flare, had a significant negative impact on the personal and social lives of patients with CD.

The GI Society described the treatment of CD as multifaceted as it involves managing symptoms and consequences of the disease and reducing inflammation. Patients also rely on medications to reduce the need for surgery. First-line treatments include 5-aminosalicylate (5-ASA) and corticosteroids to reduce inflammation in moderate to severe cases of CD. When 1 medication fails, patients must try another. According to the patient input, these treatments are inconvenient therapies that make it difficult for patients to keep a normal routine. As described by the GI Society, JAK inhibitors (such as the drug under review) are a newer class of medication. Unlike other biologics delivered by infusion therapy, JAK inhibitors are easier and more convenient to take as they are in pill form. Patients have difficulty achieving remission or adequate symptom relief despite available treatment options. Even after surgery to repair fistulas and fissures or removal of diseased bowel tissue, CD symptoms tend to recur in most patients.

Improved outcomes described as important by the patient groups included symptom mitigation and a reduction in preventable patient suffering. Managing unpredictable and frequent bowel movements, pain, and fatigue were also noted as important by CCC respondents. Unmet patient needs were noted by CCC to vary among individuals depending on their specific symptoms and life circumstances. Both patient groups emphasized the importance of a treatment option that is easy to administer and can provide symptom relief, achieve remission, and improve subsequent HRQoL.

Three patients from CCC and 2 patients from the GI Society group who had experience with upadacitinib reported near-immediate improvements in their health, alleviation of disease symptoms, and either no side effects or few mild side effects, such as weight gain. Patients noted the convenience of pill-based administration and the lack of need to refrigerate the medication or attend a clinic for infusions.

Clinician Input

Input From the Clinical Expert Consulted by CADTH

The clinical expert consulted by CADTH for this review indicated that there are profound treatment gaps in the management of IBD, including CD. Over time, transmural damage results in complications (e.g., stenosis and penetrating complications that often require surgery). Early treatment initiation is key to limiting disease activity and stopping progression. Although several effective drugs are available for the treatment of moderate to severe CD, there are significant limitations in efficacy, in addition to the frequency of loss of response over time and reduced efficacy with the introduction of each subsequent biologic after failure. This was described as the greatest treatment challenge in the management of CD.

The clinical expert noted that both primary nonresponse and secondary loss of response are common in the treatment of CD with advanced therapies. Therapies that remain efficacious in patients who have experience with biologics are therefore needed.

Clinical trial design has historically focused on clinical symptoms of response and remission, which may not always correlate with objective measures such as endoscopic remission and mucosal healing. The clinical expert pointed out that long-term longitudinal studies to evaluate the modification of bowel damage are lacking as most clinical trials are up to 2 years in duration.

According to the clinical expert, some treatments are particularly inconvenient and can affect a patient’s lifestyle due to the need to visit an infusion clinic for a few hours every 4 to 8 weeks. Therapies given by subcutaneous injection are more convenient, but patients may find them painful. No orally administered advanced therapies for CD are currently available.

The clinical expert noted that the current treatment paradigm for adults with moderately to severely active CD is complex and dictated by disease phenotype. Conventional therapies (e.g., steroids and immune suppressants) are not typically used in isolation over the long-term, and most patients with moderately to severely active disease go on to receive advanced therapy, such as a biologic. The clinical expert noted that the first therapy to be prescribed has the best chance of resulting in improvement and healing due to the aforementioned pattern of a lower likelihood of a robust response with subsequent advanced therapies. Selecting the optimal therapy from the start is a challenge and is based on disease phenotype, disease severity, and the risks and expected onset of action of each available therapy. Particularly severe disease would warrant selection of a therapy with rapid onset, high efficacy, and steroid-sparing effects (e.g., anti–tumour necrosis factors [TNFs] or anti–interleukin [IL]-23 and IL-12/IL-23).

Nearly half of patients have extra-intestinal manifestations (EIMs) of CD, which can be disabling, and only a select few of the currently available medications address them, with a preference for the anti-TNF category. Approximately a quarter of patients with CD have fistulizing perianal disease, which is a marker of severe disease, and again anti-TNFs are the preferred therapeutic option for this subpopulation. Other options for patients with these disease phenotypes are needed.

The clinical expert indicated that upadacitinib would be used as a first-line drug for patients receiving advanced therapies for CD, and that no mechanism-, efficacy-, or sequencing-based argument requires the failure of other advanced drugs before initiation of upadacitinib.

The clinical expert acknowledged that there is increasing off-label use of combination therapies with complementary mechanisms of effect in particularly severe, high-risk patients with prior drug failures, surgeries, or other markers of disabilities. The expert stated that combinations would typically include a low-risk, safe drug, such as an anti-integrin, with other more systemically active drugs. This may become relevant in the case of upadacitinib. However, the expert emphasized that this use is limited to dire situations in which there is a risk of extensive surgery or disability.

According to the clinical expert, patients with confirmed pathologic or histologic diagnosis of moderate to severe CD are typically diagnosed by a gastroenterologist. Misdiagnosis is rare, but diagnosis may be delayed, as previously noted. Patients with EIMs (e.g., inflammatory arthropathy, peripheral or axial) are a priority for treatment. Although there are no clear “stages” of CD, objective measures such as endoscopic activity and the requirement for, or dependence on, corticosteroids are important, while the presence of clinical symptoms is not as critical. There are no established predictors of disease response.

The clinical expert noted that assessment of response in clinical practice differs from clinical trials due to logistics and patient preference. The most easily accessed marker of response is improvement in clinical symptoms (abdominal pain [AP] and frequency of soft or liquid stools in particular), but this is weakly correlated with objective markers of disease activity and may be heterogeneous according to disease phenotype. For example, patients with bowel stricture may experience constipation instead of diarrhea, and patients with prior surgeries may have differing symptoms caused by anatomic alteration rather than inflammation. Objective measures of disease activity are important, particularly endoscopy (i.e., ileocolonoscopy). The clinical expert noted that, although the clinical trials assessed endoscopic outcomes at 12 weeks, endoscopy is rarely performed at 12 weeks in clinical practice, and instead is typically performed at 6 to 9 months, and can be challenging to repeat. Other objective measures may include biomarkers (CRP and fecal calprotectin) as well as noninvasive intestinal ultrasound scans.

The clinical expert indicated treatment discontinuation should be considered in a manner similar to that of other advanced therapies for adults with CD, and based on a combination of clinical symptoms and objective data that support primary nonresponse or loss of response, including:

• persistence or worsening of clinical symptoms, most importantly diarrhea and AP

• persistence or worsening of endoscopic activity

• worsening or persistent elevation of biomarkers, including CRP and fecal calprotectin

• worsening or development of complications (including strictures and penetrating disease) on cross-sectional imaging

• dependence on or need for recurrent courses of corticosteroids (e.g., 2 or more full courses of oral prednisone within 1 year, but details may be debated)

• development of adverse events (AEs), weighed on a case-by-case basis depending on treatability and severity of the AE; all patients should be vaccinated appropriately (e.g., for varicella) to avoid any preventable AEs potentially associated with treatment

• circumstances when patients with severe disease may require a single course of corticosteroids, which may not preclude ongoing maintenance and therefore the need for discontinuation would be judged by the treating physician.

Prescription of upadacitinib should not be limited to IBD specialists, as general gastroenterologists would have the expertise required to initiate therapy. General internists with a special interest in IBD and/or GI may have sufficient experience and training to prescribe upadacitinib, and this may be important for accessibility in rural regions of Canada.

The clinical expert noted that initiation criteria should be similar to those for other biologics. However, the expert emphasized that the current requirements for previously failed therapies are not up to date with clinical practice. In particular, the current requirements for prior drug failures in prescribing advanced therapies includes 5-ASA, which was considered by the clinical expert to be out of date due to its known lack of efficacy in this population. The clinical expert noted that, in clinical practice, this results in short-term prescriptions of 5-ASA to meet the requirements when it is not expected to have patient benefit, and the expert recommended that this not be a requirement for prior treatment failures when prescribing upadacitinib. The expert noted that, if treatment is interrupted for at least 2 weeks, the patient may need to undergo another course of induction therapy.

Clinician Group Input

One clinician group, the Canadian IBD Specialist Group, responded to CADTH’s call for input. The input was based on a discussion held by the group in March 2023.

The clinician group emphasized that CD has a tremendous impact on the physical, emotional, and social aspects of those living with the disease, affecting HRQoL and causing a significant economic burden. The current treatment paradigm for CD includes 5-ASA, corticosteroids, immune modifiers, and biologics that include anti-TNFs, anti-integrin, and anti–IL-12/IL-23 and anti–IL-23 agents.

The input from the clinician groups identified the same unmet medical needs for patients with CD and potential place in therapy for the drug under review as the clinical experts consulted by CADTH.

The clinician group identified significant unmet therapeutic needs for patients living with moderate to severe CD. There is a lack of safe and effective treatments that could rapidly improve endoscopic appearance, maintain long-term improvement and remission, and reduce the risk of complications and need for surgery.

In addition to relieving clinical symptoms, the clinician group emphasized that the goal of treatment should focus on changing the course of disease for patients with CD, preventing further intestinal damage, avoiding disability, and reducing the overall cost of care.

The clinician group noted that upadacitinib has a novel mechanism of action, and it is the first oral therapy for CD to be evaluated to meet the treatment goals. According to the clinician group, any patient with an inadequate response or intolerance to corticosteroids or multiple advanced therapies and those with 1 or more extraintestinal IBD manifestations could benefit from upadacitinib.

Drug Program Input

Input was obtained from the drug programs that participate in the CADTH reimbursement review process. The following were identified as key factors that may affect the implementation of a CADTH recommendation for upadacitinib:

• relevant comparators

• consideration for continuation or renewal of therapy

• considerations for discontinuation of therapy

• consideration for prescribing of therapy

• care provision issues

• system and economic issues.

The clinical expert consulted by CADTH provided advice on the potential implementation issues raised by the drug program. Table 5 provides more details.

Clinical Evidence

Systematic Review

Description of Studies

Three phase III, double-blind, placebo-controlled, multicentre, international randomized controlled trials (RCTs) are included in this review. Two of the RCTs were induction studies in adult patients with moderately to severely active CD and a history of either biologic failure (U-EXCEED) or biologic and/or conventional therapy failure (U-EXCEL). Patients in both induction studies were randomized in a 2:1 ratio to receive upadacitinib 45 mg or placebo once daily. The primary results for randomized cohorts were evaluated at 12 weeks (“part 1”), although patients who did not achieve an adequate response could carry on to part 2 or 3 for extended induction. An adequate response was defined as a decrease of 30% or more in average daily very soft or liquid stool frequency (SF) and/or a decrease of 30% or more in the average daily AP score (neither worse than baseline). The third RCT (U-ENDURE) was a maintenance study of upadacitinib 15 mg or 30 mg once daily versus placebo in patients who had achieved an adequate response in either the U-EXCEED or U-EXCEL trial, and the primary results for re-randomized patients from part 1 of the induction studies were evaluated at 52 weeks. Patients who had carried on to extended induction therapy in either of the induction studies and thereafter achieved a response could also enrol in cohort 2 or 3 of the U-ENDURE trial, which were not randomized. The coprimary outcomes in all trials included clinical remission (based on patient-reported outcomes [PROs] or the Crohn Disease Activity Index [CDAI]), and endoscopic response (based on the Simple Endoscopic Score for Crohn Disease [SES-CD]). Other important outcomes included endoscopic remission, the proportion of patients who discontinued corticosteroid use for CD and achieved clinical remission (among patients taking corticosteroids at induction baseline), the proportion of patients who achieved both clinical remission and endoscopic remission, change in HRQoL (using the Irritable Bowel Disease Questionnaire [IBDQ]), a decrease of at least 100 points from baseline in the CDAI (CR-100), resolution of EIMs in patients who had EIMs at induction baseline, the proportion who experienced CD-related hospitalizations or surgeries, and the proportion who experienced harms, including serious adverse events (SAEs) or adverse events of special interest (AESIs).

The U-EXCEED trial enrolled 624 patients across 229 sites in 39 countries; the U-EXCEL trial enrolled 526 patients across 209 sites in 42 countries; and the U-ENDURE trial enrolled 901 patients across 277 sites in 43 countries. In the double-blind cohorts of the U-EXCEED, U-EXCEL, and U-ENDURE trials, there were slightly more male than female patients (53.5%, 53.8%, and 55.4%, respectively), and the mean ages were 38.1, 39.6, and 37.0 years, respectively. The majority of enrolled patients were white (approximately 70% in each trial, followed by Asian, Black, multiple races (unspecified in the study), and American Indian or Alaska Native). The mean durations of CD were 9.4 years in the U-EXCEED trial, 6.1 years in the U-EXCEL trial, and 7.2 years in the U-ENDURE trial; the differences were expected given that the U-EXCEL trial included patients who may not have experienced failure with biologics, indicating an earlier point in treatment history on average. Overall, most enrolled patients in the randomized cohorts had a history of biologic failure (100% in the U-EXCEED trial by design, 45.4% in the U-EXCEL trial, and 75.6% in the U-ENDURE trial).

Efficacy Results: Induction

Induction Outcomes (12 Weeks)

• Clinical Remission as Measured by PROs at 12 Weeks: In the U-EXCEED and U-EXCEL trials, a higher percentage of patients achieved clinical remission as measured by PROs among those treated with upadacitinib 45 mg compared with those who received placebo. In the U-EXCEED trial, the response-rate difference compared to placebo was 25.9% (95% confidence interval [CI], 18.7% to 33.1%), and in the U-EXCEL trial it was 28.7% (95% CI, 20.9% to 36.4%).

  Results were consistent across subgroups based on the number of prior biologics failed and the analysis for clinical remission as measured by PROs at 12 weeks.

• Clinical Remission as Measured by CDAI at 12 Weeks: In the U-EXCEED and U-EXCEL trials, a higher percentage of patients achieved clinical remission as measured by CDAI among those treated with upadacitinib 45 mg compared to those who received placebo. In the U-EXCEED trial, the response-rate difference compared to placebo was 17.9% (95% CI, 10.0% to 25.8%) and in the U-EXCEL trial it was 20.8% (95% CI, 12.7% to 28.8%).

  Results were consistent across subgroups based on the number of prior biologics failed and the analysis for clinical remission as measured by CDAI at 12 weeks.

• Endoscopic Response at 12 Weeks: In the U-EXCEED and U-EXCEL trials, a higher percentage of patients achieved endoscopic response among patients treated with upadacitinib 45 mg compared to those who received placebo. In the U-EXCEED trial, the response-rate difference compared to placebo was 31.2% (95% CI, 25.5% to 37.0%) and in the U-EXCEL trial it was 33.0% (95% CI, 26.2% to 39.9%).

  Results were consistent across subgroups based on the number of prior biologics failed and the analysis for endoscopic response at 12 weeks.

• Endoscopic Remission at 12 Weeks: In the U-EXCEED and U-EXCEL trials, a higher percentage of patients achieved endoscopic remission among those treated with upadacitinib 45 mg compared with those who received placebo. In the U-EXCEED trial, the difference compared to placebo was 16.8% (95% CI, 12.0% to 21.6%) and in the U-EXCEL trial it was 21.8% (95% CI, 15.8% to 27.8%).

• Discontinuation of Corticosteroid Use and Clinical Remission as Measured by CDAI at 12 Weeks Among Patients Taking Corticosteroids at Baseline: In the U-EXCEED and U-EXCEL trials, a higher proportion of patients treated with upadacitinib 45 mg discontinued corticosteroid use and had CDAI clinical remission at week 12 compared to the placebo group. In the U-EXCEED trial, the difference compared to placebo was 22.5% (95% CI, 11.1% to 34.0%) and in the U-EXCEL it was 27.7% (95% CI, 15.7% to 39.8%).

  Results were similar for discontinuation of corticosteroid use and clinical remission as measured by PROs at 12 weeks.

• Clinical Remission as Measured by CDAI and Endoscopic Remission at 12 Weeks: | |||||||| | |||||||| |||||| |||||||||| || ||||||||||||||||||| |||||||||||| || || || |||||||| ||||||||| || |||| || |||||||||| ||||||||| || || |||||| || |||||||| |||||||||| |||||||| || ||||||| || ||||| |||| || |||| || |||||| || ||||| |||| ||||| || ||||||| |||||||||| |||||||| |||||||||||||

• Change From Baseline in IBDQ Total Score at 12 Weeks: In the U-EXCEED and U-EXCEL trials, there was a larger within-group change from baseline in IBDQ total score in patients treated with upadacitinib 45 mg than in patients treated with placebo. The between-group difference compared to placebo (least squares mean) was 24.3 (95% CI, 17.2 to 31.5) in the U-EXCEED trial and 21.8 (95% CI, 15.6 to 28.1) in the U-EXCEL trial.

CR-100 at 12 Weeks

In the U-EXCEED and U-EXCEL trials, a higher percentage of patients achieved CR-100 among patients treated with upadacitinib 45 mg than in those who received placebo. In the U-EXCEED trial, the difference compared to placebo was 22.8% (95% CI, 14.4% to 31.2%), and in the U-EXCEL it was 19.8% (95% CI, 11.3% to 28.4%).

• Resolution of EIMs at 12 Weeks in Patients With EIMs at Baseline: In the U-EXCEED trial, resolution of EIMs at week 12 in patients with any EIMs at baseline was 32.8% for upadacitinib 45 mg versus 21.7% for PBO (between-group difference = 11.5%; 95% CI, −1.5% to 24.4%). In the U-EXCEL trial, resolution of EIMs at week 12 in patients with any EIMs at baseline was 28.5% for upadacitinib 45 mg versus 20.9% for placebo (between-group difference = 9.0%; 95% CI, −1.9% to 19.9%). In both cases, the 95% CI crossed the threshold between potential benefit and potential harm (i.e., null).

• Proportion With CD-Related Hospitalizations Through 12 Weeks: No substantial differences were observed in the proportion of patients with CD-related hospitalizations between patients treated with upadacitinib 45 mg in the U-EXCEED trial (20 of 324) and the U-EXCEL trial (13 of 350) compared to patients treated with placebo (15 of 171 and 9 of 176, respectively). The differences compared to placebo were −2.6% (95% CI, −7.6% to 2.4%) in the U-EXCEED trial and −1.4% (95% CI, −5.2% to 2.4%) in the U-EXCEL trial. In both cases, the 95% CI crossed the threshold between potential benefit and potential harm (i.e., null).

• Proportion With CD-Related Surgeries Through 12 Weeks: |||||| ||||||||| ||||||| | ||||||||||| ||||||||||| |||||||||| ||||||||| |||| |||||||| || |||||||||| || ||||||||||||| || || ||||| |||||||| || | ||||||| ||||| || |||| |||| || ||||| || ||||| || |||||||| || |||| |||||| || ||||| || |||||||| || |||| |||||| || || || ||||||| || ||||||||| ||||||| ||||||||| |||||||| ||||||||| |||| |||||| ||||||

Maintenance Outcomes (52 Weeks)

• Clinical Remission as Measured by PROs at 52 Weeks: In the U-ENDURE trial, the upadacitinib 15 mg and 30 mg groups had higher percentages of patients who achieved response rates in clinical remission as measured by PROs compared to the placebo group. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups, were 21.9% (95% CI, 13.7% to 30.0%) and 31.8% (95% CI, 23.2% to 40.3%), respectively.

  The results were similar for the subgroups of patients for whom treatment had failed using 1 or more prior biologics, 1 prior biologic, and more than 1 prior biologic, as well as for those in the 30 mg group for whom treatment with 0 prior biologics had failed. However, for the subgroup of 0 prior biologics failed in the 15 mg group, the difference (versus placebo) was smaller, and the 95% CI crossed the null value (11.7%; −9.1% to 32.5%).

• Clinical Remission as Measured by CDAI at 52 Week: The upadacitinib 15 mg and 30 mg groups had higher percentages of patients who achieved clinical remission as measured by CDAI compared to the placebo group. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups, were 23.7% (95% CI, 15.2% to 32.1%) and 32.8% (95% CI, 23.9% to 41.6%), respectively.

  Results in the subgroups based on the number of prior biologics failed were consistent with the analysis for both dosage groups.

• Endoscopic Response at 52 Weeks: The upadacitinib 15 mg and 30 mg groups had higher percentages of patients who achieved an endoscopic response compared to the placebo group. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups were 21.0% (95% CI, 13.6% to 28.4%) and 33.7% (95% CI, 26.0% to 41.3%), respectively.

  Results in the subgroups based on number of prior biologics failed were consistent with the analysis for both dosage groups.

• Endoscopic Remission at 52 Weeks: The upadacitinib 15 mg and 30 mg groups had higher percentages of patients who achieved endoscopic remission compared to the placebo group. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups, were 14.4% (95% CI, 7.7% to 21.0%) and 23.6% (16.1% to 31.0%), respectively.

  Discontinuation of Corticosteroid Use at Least 90 Days Prior to Week 52 and Clinical Remission as Measured by CDAI at 52 Weeks Among Patients Taking Corticosteroids for CD at Induction Baseline.

  Among patients taking corticosteroids for CD at induction baseline, a higher proportion of patients in the upadacitinib 15 mg and 30 mg groups discontinued corticosteroid use and had CDAI clinical remission at week 52 compared to the placebo group. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups, were 35.4% (95% CI, 23.3% to 47.5%) and 32.3% (95% CI, 20.1% to 44.5%), respectively. Among all patients (i.e., not limited to those taking corticosteroids at induction baseline), the differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups, were 23.8% (95% CI, 15.5% to 32.1%) and 32.2% (95% CI, 23.4% to 40.9%), respectively.

• Clinical Remission as Measured by CDAI and Endoscopic Remission at 52 Weeks: A higher proportion of patients in the upadacitinib 15 mg and 30 mg groups had both CDAI clinical remission and endoscopic remission at week 52 compared to the placebo group. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups were 12.2% (95% CI 6.3% to 18.1%) and 19.8% (95% CI 13.0% to 26.6%), respectively.

• Change From Baseline in IBDQ Total Score at 52 Weeks: The within-group change from baseline in IBDQ total score was larger in the patients treated with upadacitinib (15 mg or 30 mg) than in those treated with placebo. The between-group difference (least squares mean) was 12.9 (95% CI, 4.3 to 21.4) when the upadacitinib 15 mg group was compared to the placebo group and 18.1 (95% CI, 9.8, 26.4) when the upadacitinib 30 mg group was compared to the placebo group. Only the between-group difference in the latter comparison (i.e., 30 mg upadacitinib versus placebo) was greater than the reported minimally important difference (MID) of 16 points in the IBDQ total score for patients with CD; the 95% CIs of both comparisons include values both greater than and less than this MID.

CR-100 at 52 Weeks

There was a higher percentage of patients who achieved CR-100 among patients treated with upadacitinib 30 mg or 15 mg compared with placebo. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups, were 27.1% (95% CI, 18.3% to 35.8%) and 36.4% (95% CI, 27.5% to 45.2%), respectively.

• Resolution of EIMs at 52 Weeks in Patients With EIMs at Induction Baseline: The proportion of patients who achieved resolution of EIMs at week 52 in those with any EIMs at induction baseline was 24.6% (upadacitinib 15 mg), 35.6% (upadacitinib 30 mg), and 15.2% (placebo). The difference versus placebo was 9.6% (95% CI, −3.4% to 22.6%) for upadacitinib 15 mg and 22.0% (95% CI, 9.3% to 34.8%) for upadacitinib 30 mg. For the 15 mg dose but not the 30 mg dose, the 95% CI crossed the threshold between potential benefit and potential harm (i.e., null).

• Proportion With CD-Related Hospitalizations Through 52 Weeks: No substantial differences were observed in the percentages of patients who experienced CD-related hospitalizations across the upadacitinib 30 mg, upadacitinib 15 mg, and placebo groups. The differences between the upadacitinib 15 mg and placebo groups, and between the upadacitinib 30 mg and placebo groups were −0.78% (95% CI, −10.4% to 8.8%) and −4.17 (95% CI, −13.1% to 4.7%), respectively. In both cases, the 95% CI was wide and crossed the threshold between potential benefit and potential harm (i.e., null).

• Incidence of CD-Related Surgeries Through 52 Weeks: |||||| || ||||||||| ||||||| ||||| || |||| ||||||||| |||| || |||||||||| ||||||||| || || ||||||||||||| |||| |||||||| || |||| |||||||||| |||||| |||||||||||||| || | |||||||||||| || || || || |||||| |||||||| || || ||||||| ||||| |||| |||| |||| || ||||| || ||||| || |||| |||||| || |||||| ||||||||||||| || |||| |||||| | | ||| |||| | ||||||| || ||||||||| ||||||| ||||||||| ||||||| || ||||||||| |||| |||||| ||||||

Harms Results

Across the trials, AEs were common and were experienced by approximately 58% to 76% of patients. In the placebo-controlled parts of the trials, the rates of AEs and withdrawals due to AEs were generally similar between treatment groups. SAEs were reported for approximately 7% to 15% of patients across the different treatment groups and cohorts of the included trials, and they were approximately similar between patients treated with upadacitinib and those who received placebo in the comparative cohorts. Some of the most frequently reported SAEs among all trials were GI disorders, infections, and infestations.

An evaluation of SAEs using Grading of Recommendations Assessment, Development and Evaluation (GRADE) found that upadacitinib induction or maintenance may result in little to no difference in the incidence of SAEs compared to placebo in a 12-week or 52-week time period.

Selection of AESIs was based on safety concerns reported for other JAK inhibitors, upadacitinib data obtained from preclinical studies, and the upadacitinib development program, as well as customary regulatory concerns for novel small-molecule drugs. Across the trials, AESIs of serious infection, opportunistic infection, herpes zoster, adjudicated GI perforation, anemia, neutropenia, lymphopenia, elevated creatine phosphokinase (CPK), hepatic disorder, renal dysfunction, and adjudicated venous thromboembolic events were observed. The most commonly reported AESIs (≥ 4% in any part or cohort of any included trial) included anemia, lymphopenia, serious infections, infections and infestations, herpes zoster, hepatic disorder, and elevated CPK. One adjudicated cardiovascular event was observed in a patient who was treated with placebo in part 1 of the U-EXCEL trial. Malignancies (all types), malignancies (excluding nonmelanoma skin cancer, and nonmelanoma skin cancer were rare in the U-ENDURE trial and not observed in the induction trials (U-EXCEED and U-EXCEL). No cases of lymphoma or active tuberculosis were observed in any included trial.

Critical Appraisal

All 3 trials were phase III, double-blind, placebo-controlled, multicentre assessments of several important clinical, endoscopic, and HRQoL-related outcomes. A review of internal validity raised no concerns related to study design (e.g., method of randomization, concealment of allocation, maintenance of blinding, or balance of patient characteristics between treatment groups). The U-ENDURE maintenance study included an enriched population as only patients with a response to and adequate tolerance of the study drug during induction could enrol, but this is representative of the reality of clinical practice. All trials included nonrandomized cohorts to accommodate patients who needed more than 12 weeks of induction to reach an adequate response; although not represented in the primary analysis, these patients reflect a minority of patients in real-world practice. Only the randomized data are discussed in detail here. In the U-ENDURE trial, patients who enrolled after achieving a response at 12 weeks of induction were re-randomized, which preserved the strength of randomized study design. Additionally, the use of separate induction and maintenance studies was consistent with European Medicines Agency (EMA) guidance for the development of drugs for the treatment of CD. Discontinuation rates were potentially imbalanced, with higher placebo-group withdrawals due to lack of efficacy in the U-EXCEL trial, and they were generally high during the U-ENDURE trial (20% to 28% across cohorts and treatment groups).

The clinical expert consulted by CADTH indicated that the study populations were wholly representative of the target population of adults with moderate to severe CD and a history of treatment failure. The dosage of the intervention, upadacitinib, was 45 mg once daily during the induction studies and either 15 mg or 30 mg once daily during the maintenance study. The clinical expert noted that treatment of moderate to severe CD in clinical practice would lean more often toward a maintenance dosage of 30 mg once daily due to evidence of greater efficacy and a reluctance to potentially undertreat, given the irreversible nature of the bowel damage that can occur. However, the clinical expert and the product monograph also noted that patients should be treated with the lowest effective dose in the interest of safety, and that the approach to dosing may vary by treating physician and disease severity. All 3 RCTs were placebo-controlled, and there is a lack of direct, head-to-head comparisons of active therapies. The 3 RCTs were unique among CD trials in that there was a mandatory tapering of corticosteroids, which was considered to be reasonably similar to clinical practice. Overall, the selected primary and key secondary outcomes were relevant to decision-making and/or clinical practice, and adequately reflected measures of both efficacies and harms. The duration of follow-up was appropriate for the induction and maintenance phase of treatment. However, when measuring the proportion of patients who experienced events such as hospitalizations or surgeries related to CD, both a 12-week and 52-week time frame were considered to be inadequate to experience a difference between arms, which contributed to uncertainty in interpreting these outcomes. Additionally, the clinical expert noted that, in clinical practice, endoscopy is not typically conducted at 12 weeks but rather after 6 to 9 months of initiating treatment due to practical limitations and the invasiveness of the procedure. This logistical limitation was also considered by the expert to be a factor in decision-making around dosing, as patients without symptoms may be experiencing endoscopic activity that would not be seen until the procedure could be completed.

GRADE Summary of Findings and Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to informing CADTH’s expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^24,25 Following the GRADE approach, evidence from RCTs started as high-certainty evidence and could be rated down in response to concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias.

The selection of outcomes for GRADE assessment was based on the sponsor’s Summary of Clinical Evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members: clinical remission as measured by PROs, clinical remission as measured by CDAI, endoscopic response, endoscopic remission, discontinuation of corticosteroid use for CD and CDAI clinical remission in patients taking corticosteroids at induction baseline, endoscopic remission and CDAI clinical remission, change in IBDQ, CR-100, resolution of EIMs among patients who had EIMs at induction baseline, CD-related hospitalization, CD-related surgery, and SAEs.

The induction studies (U-EXCEED and U-EXCEL) were assessed together due to their similarities in population and study design and are reported in Table 2. The maintenance study (U-ENDURE) is reported separately in Table 3, and GRADE assessments were conducted independently for the 2 dosages of upadacitinib maintenance therapy (15 mg or 30 mg once daily).

Indirect Comparisons

One indirect treatment comparison (ITC) was submitted by the sponsor to estimate the relative efficacy and safety of upadacitinib versus advanced therapies for the treatment of adult patients with moderately to severely active CD.^27,28

Description of Studies

Studies included in the ITC enrolled adult and adolescent patients with moderately to severely active CD receiving advanced treatments for CD, namely upadacitinib, vedolizumab, ustekinumab, risankizumab, adalimumab, and infliximab. Efficacy outcomes included clinical outcomes (remission and response), endoscopic outcomes (remission and response), and safety outcomes (any AE, SAE, serious infection, or AE leading to discontinuation), which generally aligned with the outcomes that were important to patients and clinicians.

Efficacy Results

The results of the ITC suggest that upadacitinib was ||| |||||||||||| ||||||||| |||| other advanced therapies for clinical remission and response induction. The endoscopic outcomes networks were sparse with only | treatments, making it challenging to |||| |||||| ||||||||||| from the data. Moreover, the credible intervals (CrIs) for risk-difference point estimates, including those |||| |||||||| || ||||| |||| |||||||||| |||| indicating imprecise treatment-effect estimates. ||||| |||| |||||||||| ||||||||||||| |||||| |||||||||| || |||, particularly for the maintenance phase, |||||||||| meaningful conclusions whether upadacitinib is |||| || |||| ||||||||||| than the comparators |||| ||||| ||||.

Table 2: Summary of Findings for Upadacitinib Induction Versus Placebo for Patients With Moderately to Severely Active CD and a History of Treatment Failure

CD = Crohn disease; CDAI = Crohn Disease Activity Index; CI = confidence interval; CR-100 = a decrease of at least 100 points from baseline in the CDAI; EIM = extra-intestinal manifestation; HRQoL = health-related quality of life; IBDQ = Inflammatory Bowel Disease Questionnaire; PRO = patient-reported outcome; RCT = randomized controlled trial; SAE = serious adverse event; UPA = upadacitinib.

Note: Study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes.

^aA difference of 15% between groups was identified by the clinical expert consulted by CADTH as a threshold of clinical importance for this outcome.

^bRated down 1 level for serious imprecision as the 95% CI for the between-group difference for each trial crossed the MID of 15% identified by the clinical expert consulted by CADTH for this outcome.

^cA difference of 5% between groups was identified by the clinical expert consulted by CADTH as a threshold of clinical importance for this outcome.

^dRated down 1 level for serious imprecision as the 95% CI for the between-group difference in the U-EXCEED trial crossed the threshold of 15% that was identified by the clinical expert consulted by CADTH for this outcome.

^eA MID of ≥ 16 points on the IBDQ was identified from the literature as clinically important. Although the lower boundary of the 95% CI in the U-EXCEL trial was 15.6, this was not considered a serious imprecision due to its proximity to 16.

^fA difference of 15% between groups was identified by the clinical expert consulted by CADTH as a threshold of clinical importance for this outcome. Although the lower boundaries of the 95% CIs were below this threshold, given the sample size and the proximity of the lower bound of the confidence intervals to the estimated threshold across both trials, the imprecision was not considered serious.

^gRated down 1 level for serious concern regarding imprecision because the point estimates are below the difference of 15% between groups identified by the clinical expert as clinically important, and the upper bound of the 95% CIs include the possibility of an important benefit.

^hRated down 2 levels for serious concerns regarding indirectness and imprecision. Longer-term outcome assessment would be required to compare the effect of treatment more meaningfully on these outcomes. The point estimates are close to null and the 95% CIs cross the null threshold. The clinical expert consulted by CADTH could not provide a threshold of important difference; however, the CADTH review team judged that the effect estimate and CI were unlikely to include any important effect.

ⁱRated down 1 level for serious concerns regarding imprecision. As no 95% CI of the difference was available, the optimal information size approach was used to judge imprecision. The clinical expert consulted by CADTH could not provide a threshold of important difference; however, the CADTH review team judged that the effect estimate and CI were unlikely to include any important effect.

Source: Clinical Study Reports of U-EXCEED²⁶ and U-EXCEL.²² Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 3: Summary of Findings for Upadacitinib Maintenance Versus Placebo for Patients With Moderately to Severely Active CD and History of Treatment Failure

CD = Crohn disease; CDAI = Crohn Disease Activity Index; CI = confidence interval; CR-100 = decrease of at least 100 points from baseline in the CDAI; EIM = extra-intestinal manifestation; HRQoL = health-related quality of life; IBDQ = Inflammatory Bowel Disease Questionnaire; MID = minimally important difference; PRO = patient-reported outcome; RCT = randomized controlled trial; SAE = serious adverse event; UPA = upadacitinib.

Note: Study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes.

^aRated down 1 level for serious imprecision as the 95% CI of each trial crossed the difference of 15% between groups identified by the clinical expert consulted by CADTH as clinically important for this outcome.

^bA difference of 15% between groups was identified by the clinical expert consulted by CADTH as a threshold of clinical importance for this outcome.

^cA difference of 5% between groups was identified by the clinical expert consulted by CADTH as a threshold of clinical importance for this outcome.

^dRated down 1 level for serious concerns regarding imprecision. As no 95% CI of the difference was available, the optimal information size approach was used to judge imprecision. There is no established MID.

^eRated down 1 level for serious concern regarding imprecision because the point estimate was lower than the literature-reported MID of ≥ 16 points on the IBDQ, and the 95% CI crossed the MID.

^fRated down 1 level for serious concern regarding imprecision because the 95% CI crossed the literature-reported MID of 16 points or higher on the IBDQ.

^gRated down 2 levels for very serious concerns regarding imprecision because the point estimate was lower than the difference of 15% between groups identified by the clinical expert consulted by CADTH as clinically important for this outcome, and the 95% CI crossed the clinical importance threshold.

^hRated down 1 level for serious concerns regarding imprecision because and the 95% CI crossed the difference of 15% between groups identified by the clinical expert consulted by CADTH as clinically important for this outcome.

ⁱRated down 2 levels for serious concerns regarding indirectness and imprecision. Longer-term outcome assessment would be required to compare the effect of treatment more meaningfully on these outcomes. The point estimates are close to null and the 95% CIs crossed the null threshold. These outcomes may not have been tested for multiplicity in the trial and should be considered as supportive evidence. The clinical experts consulted by CADTH could not provide a threshold of important difference; however, the CADTH review team judged that the effect estimate and CI were unlikely to include any important effect.

^jRated down 1 level for serious concern regarding imprecision. No 95% CI of the difference was available. As there is no established MID, the optimal information size approach was used. The clinical experts consulted by CADTH could not provide a threshold of important difference; however, the CADTH review team judged that the effect estimate and CI were unlikely to include any important effect.

Source: Clinical Study Reports of U-ENDURE.²³ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Harms Results

Induction-phase safety results had |||| |||| || || ||||||| || |||| for comparisons of upadacitinib versus other advanced treatments. Likewise, ||||||||||||| || |||||||||| |||| |||| ||||||||| comparative safety conclusions |||| ||||| |||| |||| ||||||||||.

Critical Appraisal

The networks included relatively few direct comparisons between active treatments or were small (e.g., endoscopic outcomes). In general, based on the baseline patient characteristics, the clinical expert was of the opinion that the patients in the studies were similar and generalizable to those who could receive upadacitinib in Canadian clinical practice. However, patient characteristics (and potential treatment-effect modifiers) varied across the studies for disease duration, CRP and fecal calprotectin values, and concomitant medication use, and it is likely that differences in patients’ experience with previous treatments (number and type) introduced further bias to the analysis. There was heterogeneity in the duration of induction and maintenance phases and in the definition of induction response (which dictated entry to the maintenance phase); re-randomization into the maintenance phase invalidated the assumption of a common placebo comparator; and studies used different definitions for safety outcomes, creating issues when analyzing harms results. ||| ||||||||||||| ||||| ||||||||| ||||||||| ||||||||||||| |||||||||||||| || |||| |||| |||| || ||||||||||| || |||| |||||||||| |||||||| | |||||| ||||||||||| ||||||| |||||||||||| | |||||| || ||||| |||| ||||| |||||||| || |||||||||| || |||| ||||||||||.

Conclusions

Three phase III, multicentre, double-blind RCTs evaluated the efficacy and safety of upadacitinib compared to placebo in adult patients with moderately to severely active CD and a history of treatment failure. Two of the RCTs evaluated induction therapy of upadacitinib 45 mg once daily while a third RCT evaluated maintenance therapy of upadacitinib 15 mg once daily and 30 mg once daily.

Compared to placebo at 12 weeks, upadacitinib 45 mg induction results in an increase in the proportion of patients who have clinical remission as measured by PROs; endoscopic response, endoscopic remission, both endoscopic remission and clinical remission as measured by CDAI, and CR-100; and an improvement in HRQoL based on IBDQ. It also likely results in an increase in the proportion who discontinued corticosteroid use for CD with concurrent clinical remission as measured by CDAI (in patients who were taking corticosteroids at baseline), and in the proportion with clinical remission as measured by CDAI. It likely results in an increase in the proportion of patients who have resolution of EIMs among those who had EIMs at baseline, and it may result in little to no difference in CD-related hospitalization, or |||||||||| ||||||||| at 12 weeks. Compared to placebo at 52 weeks, outcomes with upadacitinib 15 mg and 30 mg maintenance were generally consistent with the induction-related outcomes, albeit with slightly more imprecision in some outcomes for the 15 mg dose, and for the outcome of IBDQ for either dose. The time frame was considered insufficient to detect differences in the outcomes of CD-related hospitalization and |||||||||| |||||||||. The evidence shows that upadacitinib likely results in little to no difference in SAEs compared to placebo, and no new safety signals were identified. Longer-term data are currently being collected in substudy 2 of the U-ENDURE trial, but results are not yet available.

There is a data gap in head-to-head, direct evidence between upadacitinib and other advanced therapies for CD. Indirect evidence provided by the sponsor demonstrated | |||||||||| |||||||||| || |||||||| || |||||| |||||||| between upadacitinib and other advanced therapies during induction or maintenance, but the ITCs |||||||| |||| ||||||||||| ||||||||||| || |||||||||| |||||||||||||| |||||||||| |||||||||| ||||||||||| for both efficacy and safety outcomes.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of an oral induction dose of upadacitinib 45 mg once daily and maintenance doses of 15 mg or 30 mg once daily in the treatment of CD in adults with moderately to severely active disease who have demonstrated prior treatment failure, i.e., an inadequate response, loss of response, or intolerance to at least 1 conventional and/or biologic therapy.

Disease Background

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following summary was validated by the CADTH review team.

Crohn disease is a chronic progressive form of IBD that leads to significant disability and negatively affects HRQoL.^1-3 It is characterized by recurrent, uncontrolled inflammation that can affect any part of the GI tract from mouth to anus,⁸ primarily affecting the ileum (i.e., small intestine), colon (i.e., beginning of the large intestine), and rectum.^4,5 As CD is a fluctuating disease, patients experience periods of flare-ups.²⁹

It is thought that CD is triggered by a disturbance of the intestinal epithelial barrier due to environmental factors and/or genetic susceptibility.^30,31 The disruption of the epithelial barrier and the normal homeostasis of gut microbiota and mucosal immune system stimulates aberrant activation of the gut immune response.²¹ Common symptoms of CD include diarrhea, AP, fatigue, fever, and rectal bleeding. Other symptoms include loss of appetite, weight loss, and malnutrition.^6,7 Complications associated with CD can include bowel obstructions, fistulas, anal fissures, intra-abdominal and other abscesses, and ulcers that often require hospitalization or surgery.^8,9

Canada has the highest incidence and prevalence of IBD in the world.^10,11 The incidence of CD varies across the Canadian provinces, with the highest rate reported in Nova Scotia at 22.6 per 100,000 persons, whereas in Alberta, British Columbia, Manitoba, Ontario, Quebec, and Saskatchewan rates range from 8.8 to 16.6 per 100,000 persons.^10,32 According to the 2018 Impact of Inflammatory Bowel Disease in Canada report, approximately 270,000 Canadians were living with IBD, of whom 135,000 had CD. For every 100,000 Canadians, 16.3 new cases of CD are diagnosed each year.¹² This translates to more than 6,000 new cases annually. The number of people living with IBD is expected to exceed 400,000, or approximately 1% of the population, in 2030.¹²

Diagnosis of CD requires a combination of a medical history and physical examinations.⁶ Laboratory testing measures key biomarkers, including CRP, erythrocyte sedimentation rate, and fecal calprotectin.^6,13-15 In active CD, depending on disease severity, CRP levels can range from 5 to 20 mg/L compared to typical baseline levels of 1 mg/L.¹⁵ Diagnostic imaging, including endoscopy, radiology, and ultrasound, are also used to examine findings of typical CD cases, including bowel obstruction, stenosis, fistulae, abscesses, and, atrophy.^14,16-19

For many patients with CD, symptoms are chronic and intermittent, and disease activity and severity can vary widely. Disease severity is measured using the CDAI and Harvey-Bradshaw Index (HBI), which are designed to evaluate bowel-related symptoms, including SF, AP, arthritis and/or arthralgia, uveitis, skin and mouth lesions, and perianal disease. While the HBI is commonly used in routine gastroenterology practice, the CDAI remains the most common comparable end point across biologics in CD.³³ Less precision is expected with the HBI as it is a subset of the CDAI (e.g., the HBI uses single-day readings, only 5 of the 8 CDAI variables, and sums variables instead of applying weighted coefficients).³³ The correlation coefficient between HBI and CDAI is reportedly between 0.80 and 0.93.^33,34 The IBDQ incorporates social and emotional symptoms to give a meaningful indication of the impact of the disease on quality of life.^35,36 In addition, the Crohn Disease Endoscopic Index of Severity and SES-CD are scoring systems based on endoscopic observations of the size and location of ulceration and lesions.³⁷

Standards of Therapy

Contents within this section were informed by materials submitted by the sponsor and clinical expert input. The following summary was validated by the CADTH review team.

Selection of therapy is based on the location, extent, phenotype, and severity of disease.³⁸ As there is no cure for CD, the therapeutic goal is to induce and maintain clinical and endoscopic remission and reduce the need for long-term corticosteroid use.³⁸ Treatment of CD with pharmacotherapies is typically divided into induction, in which the goal is to achieve control of inflammation within a short period of time (e.g., 3 months or less), followed by maintenance, in which that control is sustained beyond 3 months.^38,39 The clinician group input, patient group input, and the clinical expert consulted by CADTH agreed that treatment goals for patients include short-term improvement of symptoms, and intermediate- to long-term maintenance of overall wellness and ability to return to work, school, and regular functions of daily life, with reductions in pain, fatigue, bowel movements, and improvement in HRQoL. While symptomatic control and clinical remission are important treatment goals, the clinical expert consulted by CADTH and published treatment guidelines for CD^38,39 emphasize that the modern treatment paradigm increasingly values the objective outcomes of endoscopic remission and mucosal healing, which may not always be correlated with clinical measures or symptom burden but are important in the prevention of downstream complications of CD.

Two major categories of pharmacotherapies are used to treat CD: conventional and biologic.³⁸ The conventional therapies include corticosteroids (e.g., prednisone), 5-ASA, and immunomodulators (e.g., azathioprine, cyclosporine, methotrexate, and 6-mercaptopurine).³⁸ Conventional therapies, particularly 5-ASA, are typically used in first-line treatment of patients with mild to moderate disease despite poor efficacy due to the risks associated with more effective therapies (i.e., biologics).³⁸ Corticosteroids are primarily used as rescue medication to treat CD flares, and systemic formulations are effective in short-term induction of remission in patients with moderate to severe CD, but are not appropriate for long-term use due to the associated side effects and poor efficacy in achieving mucosal healing.³⁸ Reducing dependence on corticosteroid use is a common goal in treatment of CD.³⁸

For moderate to severe CD, more effective therapies are required.³⁸ Biologic therapies for CD include TNF-alpha antagonists (e.g., infliximab and adalimumab), integrin inhibitors (e.g., vedolizumab), and IL-12/IL-23 inhibitors (e.g., ustekinumab). Selection of biologic therapy depends on patient phenotype and treatment experience, risks and benefits of each therapy, and patient preference. If patients develop an inadequate response, loss of response, or intolerance to the biologic agents, treatment will be escalated to higher doses or to newer biologic classes. Patients with serious complications or who have medically refractory disease may be candidates for surgery (e.g., total colectomy or ileostomy), which is associated with important morbidities.

Drug Under Review

Upadacitinib is a selective and reversible JAK inhibitor engineered to have greater inhibitory potency for JAK1 versus JAK2, JAK3, and TYK2 in human cellular assays.⁴⁰ Upadacitinib preferentially inhibits signalling by JAK1 or JAK1/JAK3 with functional selectivity over cytokine receptors that signal via pairs of JAK2 proteins. JAK inhibitors such as upadacitinib can inhibit intracellular downstream signalling elicited by multiple cytokines known to have a role in the inflammatory environment in CD.⁴¹ For example, pro-inflammatory cytokines such as the IL-23 pathway have a substantial role in the pathogenesis of CD.^42,43 IL-23 interacts with the IL-23 receptor to activate JAK2 and TYK2, stimulating the activation of key immune cells through the P38 MAPK, PI3K-Akt and nuclear factor kappa–beta signalling pathways.^20,21

The approved indication for upadacitinib is for the treatment of adult patients with moderately to severely active CD who have demonstrated prior treatment failure, i.e., an inadequate response, loss of response, or intolerance to at least 1 conventional and/or biologic therapy. The sponsor’s reimbursement request is consistent with the approved indication. Upadacitinib is administered orally. The recommended dosing schedule is in 2 phases: a 12-week induction phase consisting of 45 mg upadacitinib once daily, followed by an ongoing maintenance phase. The recommended dosage of upadacitinib for maintenance treatment for patients who are aged 18 to 64 years is 15 mg or 30 mg upadacitinib once daily, based on patient presentation; 30 mg once daily may be appropriate for patients with high disease burden (such as in cases of refractory or severe disease) or those who do not show adequate therapeutic benefit with 15 mg once daily, and the lowest effective dose for maintenance should be used. For patients who are aged 65 years or older, the recommended maintenance dose is 15 mg once daily. In patients who are responding to induction or maintenance treatment with upadacitinib, corticosteroids may be reduced and/or discontinued in accordance with standard of care.^22,23 Key characteristics of upadacitinib and other treatments available for CD are summarized in Table 4.

Table 4: Key Characteristics of Upadacitinib, Adalimumab, Infliximab, Ustekinumab, Vedolizumab, and Risankizumab

CD = Crohn disease; CS = corticosteroid; IgG1 = immunoglobin G1; IL = interleukin; JAK = Janus kinase; q.d. = once daily; q.2.w. = every 2 weeks; q.8.w. = every 8 weeks; q.12.w. = every 12 weeks; SC = subcutaneous; TNF = tumour necrosis factor.

^aHealth Canada indication.

Source: Product monographs of upadacitinib (Rinvoq)^4,4 risankizumab (Skyrizi),⁴⁵ vedolizumab (Entyvio),⁴⁶ infliximab (Remicade),⁴⁷ adalimumab (Humira),⁴⁸ and ustekinumab (Stelara).⁴⁹

Stakeholder Perspectives

Patient Group Input

This section was prepared by the CADTH review team based on the input provided by patient groups. The full original patient inputs received by CADTH are included in the Stakeholder section of this report.

Two patient groups, CCC and the GI Society, provided input for this review. The CCC is a national, volunteer-based charity dedicated to finding cures for patients with CD and ulcerative colitis and improving the lives of affected children and adults. The GI Society is committed to improving the lives of people with GI and liver conditions, supporting research, advocating for appropriate patient access to health care, and promoting GI and liver health.

Information was gathered by CCC from a report published in 2018 (Impact of Inflammatory Bowel Disease in Canada), a 2022 survey involving 1,700 respondents (687 with moderate to severe CD), and interviews with 3 patients who participated in the upadacitinib clinical trials. The patient input provided by the GI Society was based on surveys and interviews, including a 2015 survey on biologics and biosimilars involving 423 Canadian patients with IBD, a 2018 survey on the unmet needs involving 432 Canadians with IBD, a 2020 survey on the unmet needs of IBD involving 579 respondents, a 2020 survey on biosimilars completed by 145 respondents (most of whom had IBD), a 2022 survey on the IBD patient journey completed by 54 Canadians with IBD, a 2022 focus group with patients living with CD, interviews with 2 patients with CD who participated in the upadacitinib trial, one-on-one conversations with patients with IBD, a patient roundtable, and media interactions.

The 2 patient groups emphasized that CD has a tremendous impact on every aspect of a person’s life. Common symptoms experienced by patient respondents included diarrhea, nausea, unpredictable urgency to use the washroom, bloating, vomiting, rectal bleeding, AP, weight loss, stress and mental health disorders, anemia resulting from blood loss, skin conditions, adhesions or scar tissue, bowel obstruction, eye inflammation, fistulas and abscesses (perianal, anal, internal, or intra-abdominal), stricture, arthritis of the spine, liver conditions, and cancer. EIMs can include fever, inflammation of the eyes or joints (arthritis), ulcers of the mouth or skin, tender and inflamed nodules on the shins, and numerous other conditions. Anxiety, stress, and mental health are major factors, and the symptoms were described by respondents as relentless, unpredictable, embarrassing, and scary. Respondents to both patient groups stated that the inability to predict when the next urgency of bowel movements or flare would occur had a significant negative impact on the personal and social lives of patients with CD.

The GI Society described the treatment of CD as multifaceted as it involves managing symptoms and consequences of the disease and reducing inflammation. Patients also rely on medications to reduce the need for surgery. When 1 medication fails, patients must try another. First-line treatments include 5-ASA and corticosteroids to reduce inflammation in moderate to severe cases of CD. According to the patient input, these treatments are inconvenient therapies that make it difficult for patients to keep a normal routine. Also, if a patient has significant diarrhea, then the rectal medications may be difficult to hold in place for sufficient time to be effective. Immunosuppressive drugs reduce dependence on steroids and help patients who have a steroid-resistant disease; however, these treatments take a long time (e.g., as much as 6 months or more) for patients to experience results. A variety of biologic therapies are used to treat CD when older medications fail to relieve symptoms. As described by the GI input society, JAK inhibitors (such as the drug under review) are a newer class of medication. Unlike other biologics delivered by infusion therapy, JAK inhibitors are easier and more convenient to take as they are in pill form. Despite available treatment options, patients have difficulty achieving remission or adequate symptom relief. Even after surgery to repair fistulas and fissures or removal of diseased bowel tissue, CD symptoms tend to recur in most patients.

Improved outcomes noted as important by the patient groups included symptom mitigation such that patients can work productively; attend school and social events; and perform basic activities of daily living such as errands, maintaining a home, and raising children. Other improved outcomes of interest identified by respondents included reducing preventable patient suffering, such as debilitating symptoms, secondary illnesses (e.g., depression and anxiety), loss of family and social interactions, and reliance on health care resources such as hospital stays, surgeries, and other medications. Managing unpredictable and frequent bowel movements, pain, and fatigue were also considered important by CCC respondents. Unmet patient needs were noted by the CCC to vary between individuals depending on their specific symptoms and life circumstances. Both patient groups emphasized the importance of a treatment option that is easy to administer and can provide symptom relief, achieve remission, and improve subsequent HRQoL.

Three patients from the CCC and 2 patients from the GI Society group with experience with upadacitinib reported near-immediate improvements in their health, and alleviation of symptoms of their CD, with no side effects or few mild side effects such as weight gain. Patients noted the convenience of pill-based administration and the lack of a need to refrigerate the medication or attend a clinic for infusions.

Clinician Input

Input From the Clinical Expert Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 1 clinical specialist with expertise in the diagnosis and management of CD.

Unmet Needs

The clinical expert indicated there are profound treatment gaps in the management of IBD, including CD. A significant portion of patients do not exhibit symptoms associated with disease activity. Subclinical disease progression is common and patients often present at diagnosis with significant established, longstanding disease characterized by deep ulcers in the bowel. Over time, transmural damage leads to complications (e.g., stenosis and penetrating complications that often require surgery). Early treatment initiation is key to limiting disease activity and stopping progression. Although a number of effective drugs are available for the treatment of moderate to severe CD, there are significant limitations in efficacy. For example, the expert noted that the average likelihood of endoscopic remission across drugs ranges from approximately 20% to 40%, and emphasized the frequency of loss of response over time and the reduced efficacy with the introduction of each subsequent biologic after failure. This was described as the greatest treatment challenge in the management of CD.

The clinical expert added that primary nonresponse and secondary loss of response are both common in the treatment of CD with advanced therapies, creating a need for therapies that remain efficacious in patients with experienced with biologics.

Clinical trial design has historically focused on clinical symptoms of response and remission, which may not always correlate with objective measures, such as endoscopic remission and mucosal healing. The clinical expert noted that long-term longitudinal studies to evaluate the modification of bowel damage are lacking as most clinical trials are up to 2 years in duration.

The clinical expert acknowledged that some treatments are particularly inconvenient and affect a patient’s lifestyle due to the need to visit an infusion clinic for a few hours every 4 to 8 weeks. Therapies given by subcutaneous injection are more convenient, but patients may find them painful. No available orally administered advanced therapies are currently available to treat CD.

Place in Therapy

The clinical expert noted that the current treatment paradigm for adults with moderately to severely active CD is complex and dictated by disease phenotype. Conventional therapies (e.g., steroids and immune suppressants) are not typically used in isolation over the long-term, and most patients with moderately to severely active disease would go on to receive advanced therapy, such as a biologic drug. The clinical expert noted that the first therapy to be prescribed has the best chance for improvement and healing due to the aforementioned pattern of a lower likelihood of a robust response with subsequent advanced therapies. Selecting the optimal therapy from the start is a challenge and is based on disease phenotype, disease severity, and the risks and expected onset of action of each available therapy. For example, particularly severe disease would warrant selection of a therapy with rapid onset, high efficacy, and steroid-sparing effects (e.g., anti-TNFs or anti–IL-23 and anti–IL-12/IL-23).

Nearly half of patients have EIMs of CD, which can be disabling, and only a select few medications that are currently available address them, with a preference for the anti-TNF category. Approximately a quarter of patients with CD have fistulizing perianal disease, which is a marker of severe disease, and again anti-TNFs are the preferred therapeutic option for this subpopulation. Other options are needed for patients with these disease phenotypes.

The clinical expert indicated that upadacitinib would be used as a first-line option for patients receiving advanced therapies for CD, and that there is no mechanistic, efficacy, or sequencing-based argument to require the failure of other advanced drugs before initiation of upadacitinib.

The clinical expert also acknowledged increased off-label use of combination therapies with complementary mechanisms of effect in particularly severe, high-risk patients with prior drug failures, surgeries, or other markers of disabilities. According to the expert, combinations would typically include a low-risk, safe drug, such as an anti-integrin, with other more systemically active drugs. This may become relevant in the case of upadacitinib. However, the expert emphasized that this use is limited to dire situations in which there is a risk of extensive surgery or disability.

Patient Population

The clinical expert stated that patients with confirmed pathologic or histologic diagnosis of moderate to severe CD are typically diagnosed by a gastroenterologist. Misdiagnosis is rare, but a diagnosis may be delayed as previously described. Patients with EIMs (e.g., inflammatory arthropathy, peripheral or axial) are a priority for treatment. Although there are no clear “stages” of CD, objective measures such as endoscopic activity and the requirement or dependence on corticosteroids are important, while the presence of clinical symptoms is not as critical. There are no established predictors of disease response.

Assessing the Response Treatment

The clinical expert noted that assessment of response in clinical practice differs from clinical trials due to logistics and patient preference. The most easily accessed marker of response is improvement in clinical symptoms (particularly AP and frequency of soft or liquid stools), but this is only weakly correlated with objective markers of disease activity and may be heterogeneous according to disease phenotype. For example, patients with bowel stricture may experience constipation instead of diarrhea, and patients with prior surgeries may have differing symptoms caused by anatomic alteration rather than inflammation. Objective measures of disease activity, particularly endoscopy (i.e., ileocolonoscopy), are important. The clinical expert added that, although the clinical trials assessed endoscopic outcomes at 12 weeks, in clinical practice endoscopy is rarely performed at 12 weeks, and instead is typically performed at 6 to 9 months and can be challenging to repeat. Other objective measures may include biomarkers (CRP and fecal calprotectin) as well as noninvasive cross-sectional imaging, including intestinal ultrasound.

Discontinuing Treatment

The clinical expert indicated that treatment discontinuation should be considered in a manner similar to that of other advanced therapies for adults with CD, and include a combination of clinical symptoms and objective data to support primary nonresponse or loss of response:

• persistence or worsening of clinical symptoms, most importantly diarrhea and AP

• persistence or worsening of endoscopic activity

• worsening or persistent elevation of biomarkers including CRP and fecal calprotectin

• worsening or development of complications (including strictures and penetrating disease) on cross-sectional imaging

• dependence on or need for recurrent courses of corticosteroids (e.g., 2 or more full courses of oral prednisone within 1 year, but details may be debated)

• development of AEs should be weighed on case-by-case basis depending on treatability and severity of the AE; all patients should be vaccinated appropriately (e.g., varicella) to avoid any preventable AEs potentially associated with treatment

• patients with severe disease may require a single course of corticosteroids, which again may not preclude ongoing maintenance, and the need for discontinuation would therefore be judged by the treating physician.

Prescribing Considerations

Prescription of upadacitinib should not be limited to IBD specialists, as general gastroenterologists would have the expertise required to initiate therapy. General internists with a special interest in IBD and/or GI may have sufficient experience and training to prescribe upadacitinib, and this may be important for accessibility in rural regions of Canada.

The clinical expert noted that initiation criteria should be similar to other biologics. However, the expert emphasized that the current requirements for previously failed therapies are not up to date with clinical practice. In particular, the current requirements for prior drug failures in prescribing advanced therapies includes 5-ASA, which the clinical expert considered to be out of date due to its known lack of efficacy in this population. The clinical expert noted that, in clinical practice, this results in short-term prescriptions of 5-ASA to meet the requirements when it is not expected to have patient benefit, and the expert recommended that this not be included as a requirement for prior treatment failures when prescribing upadacitinib.

The expert noted that, if treatment is interrupted for at least 2 weeks, the patient may need to undergo induction therapy again.

Clinician Group Input

This section was prepared by the CADTH review team based on the input provided by 1 clinician group. The full original clinician group input received by CADTH is included in the Stakeholder section of this report.

One clinician group, the Canadian IBD Specialist Group, responded to CADTH’s call for input. This group consists of specialists in gastroenterology from across Canada caring for patients with CD. The clinician group input was based on a discussion held by the Canadian IBD Specialist Group in March 2023. The group reviewed the safety and efficacy data from the upadacitinib phase III trials, described the CD burden in Canada, and discussed the unmet treatment needs of CD, treatment goals, and how access to upadacitinib could benefit patients and society in the short and long-term.

The clinician group emphasized that CD has a tremendous impact on the physical, emotional, and social aspects, including HRQoL, of those living with the disease, and causes a significant economic burden. The current treatment paradigm for CD includes 5-ASA, corticosteroids, immune modifiers, and biologics that include anti-TNF, anti-integrin, and anti–IL-12/IL-23 and anti–IL-23 agents.

The input from the clinician groups identified the same unmet medical needs for patients with CD and potential place in therapy for the drug under review as the clinical experts consulted by CADTH.

The clinician group noted that there are significant unmet therapeutic needs for patients living with moderate to severe CD. There is a lack of safe and effective treatments that could rapidly improve the endoscopic appearance, maintain long-term improvement and remission, and reduce the risk of complications and the need for surgery. The group claimed that there is a high rate of surgery and postoperative recurrence in patients with CD despite currently available treatment options.

In addition to relieving clinical symptoms, the clinician group emphasized that the goal of treatment should focus on changing the course of disease for patients with CD, preventing further intestinal damage, avoiding disability, and reducing the overall cost of care.

The clinician groups noted that upadacitinib, with its selective blockade of the JAK1 signalling pathway, has a novel mechanism of action, and it is the first oral therapy for CD that has ever been evaluated to meet the treatment goals, treat symptoms of CD, improve the underlying inflammation, and provide sustained clinical and endoscopic outcomes.

According to the clinician group, any patient with inadequate response or intolerance to corticosteroids or multiple advanced therapies and those with 1 or more extraintestinal IBD manifestations could benefit from upadacitinib. It could also be used by patients with moderately to severely active CD who prefer an orally administered drug or those in whom there may be advantages to having both rapid onset and offset of action without concerns about immunogenicity or the altered pharmacokinetics that may occur with the use of biologics. However, for patients with ocular complications of CD, or where perianal disease is the dominant disease manifestation, the use of anti-TNF drugs would be preferred.

The clinician group indicated that the outcomes of interest include a meaningful improvement in symptoms as measured by SF and AP, symptomatic remission and off corticosteroids for 3 to 6 months, a decrease in biomarkers (CRP and fecal calprotectin) of inflammatory activity in the first 3 months, and a robust endoscopic response by 6 to 12 months. The group added that HRQoL should be taken into consideration.

The clinician group noted that discontinuation of therapy should be considered if there is a worsening of symptoms, inadequate clinical response to therapy by the end of the 12-week induction period, or severe AEs. The clinician group stated that treatment with upadacitinib should be prescribed by a CD specialist. Assessment of short-term clinical response should be made within 8 to 12 weeks of initiating therapy, biomarker response should be assessed by 3 to 6 months, and endoscopic response by 6 to 12 months.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s Reimbursement Review processes by identifying issues that may affect their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 5.

Table 5: Summary of Drug Plan Input and Clinical Expert Response

5-ASA = 5-aminosalicylates; CD = Crohn disease; CDEC = CADTH Canadian Drug Expert Committee; IBD = inflammatory bowel disease; JAK = Janus kinase.

Clinical Evidence

The objective of CADTH’s Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of upadacitinib (Rinvoq) 15 mg, 30 mg, and 45 mg once daily extended-release oral tablets for the treatment of adult patients with moderately to severely active CD who have demonstrated prior treatment failure, i.e., an inadequate response, loss of response, or intolerance to at least 1 conventional and/or biologic therapy. The focus is on comparing upadacitinib to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of upadacitinib is presented in 3 sections, with CADTH’s critical appraisal of the evidence included at the end of each section. The first section, the systematic review, includes pivotal studies and RCTs that were selected according to the sponsor’s systematic review protocol. CADTH’s assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section includes sponsor-submitted long-term extension studies. The third section includes indirect evidence from the sponsor.

Included Studies

Clinical evidence from 3 RCTs and a single ITC are included in the CADTH review and appraised in this document.

Systematic Review

Contents within this section were informed by materials submitted by the sponsor. The following summary was validated by the CADTH review team.

Description of Included Studies

Based on a systematic literature review, characteristics of the included studies are summarized in Table 6.

Table 6: Details of Studies Included in the Systematic Review

AE = adverse event; AP = abdominal pain; CD = Crohn disease; CDAI = Crohn Disease Activity Index; CR-100 = decrease of at least 100 points from baseline in the Crohn Disease Activity Index; EIM = extra-intestinal manifestation; EMA = European Medicines Agency; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; IBDQ = inflammatory Bowel Disease Questionnaire; NR = not reported; PRO = patient-reported outcome: SES-CD = Simple Endoscopic Score for Crohn Disease; SF = stool frequency; UPA = upadacitinib.

Source: Sponsor’s Summary of Clinical Evidence.

The objective of the randomized study of the U-EXCEED trial was to evaluate the efficacy and safety of upadacitinib 45 mg as an induction treatment compared to placebo in patients with moderately to severely active CD. This study was a multicentre, randomized, double-blind, placebo-controlled trial. Eligible patients were adults with a confirmed diagnosis of moderately to severely active CD for at least 3 months who had responded inadequately or were intolerant to biologic therapy. A total of 624 patients were enrolled and randomized between November 2017 and August 2021. Of the 229 study sites across 39 countries, 15 were in Canada. No screening phase was reported. More than 30% of enrolled patients had demonstrated an inadequate response or intolerance to 3 or more biologics. Treatment was split into 3 parts. In part 1, 495 patients were randomized double-blind 2:1 to upadacitinib or matching placebo for a 12-week induction period. Patients were stratified by baseline corticosteroid use (yes or no), endoscopic disease severity (SES-CD < 15 or ≥ 15), and number of prior biologics (> 1 or ≤ 1). In part 2, a total of 129 patients from part 1 were enrolled for a 12-week open-label induction period of an oral daily dose of upadacitinib 45 mg. Patients who achieved a clinical response in part 1 and 2 were eligible to enter a double-blind maintenance portion of the U-ENDURE trial and patients who did not achieve a clinical response in part 1 or 2 were enrolled in a 12-week part 3 extended treatment period. The extended treatment period was divided into 3 cohorts. Patients in cohort 1 who received placebo in part 1 received double-blind induction treatment with upadacitinib 45 mg daily for 12 weeks until week 24. Patients in cohort 2 who received upadacitinib 45 mg in part 1 were given double-blind induction treatment with upadacitinib 30 mg daily for 12 weeks until week 24. Patients in cohort 3 who received open-label upadacitinib 45 mg in part 2 were given open-label upadacitinib 30 mg daily for 12 weeks until week 24.

Figure 1: Study Design Schematic for U-EXCEED

QD = once daily; Rand = randomized.

Source: U-EXCEED Clinical Study Report.²⁶

The U-EXCEL study shared the same objective and study design as the U-EXCEED study but had different patient eligibility criteria. Eligible patients were adults with a confirmed diagnosis of moderately to severely active CD for at least 3 months who were biologic therapy–intolerant or inadequate responders to conventional therapy but had not failed biologic therapy (non-bio-IR population) and patients who were biologic therapy–intolerant or inadequate responders to 1 or more biologic agents for CD (bio-IR population). A total of 526 patients were enrolled and randomized between December 2017 and January 2022. There were 209 study sites across 42 countries, including 17 sites in Canada. No screening phase was reported. Treatment was split into 2 parts. In part 1, a total of 526 patients were randomized double-blind 2:1 to upadacitinib 45 mg or matching placebo for a 12-week induction period. Patients were stratified based on baseline steroid use (yes or no), endoscopic disease severity (SES-CD < 15 or ≥ 15) and number of prior biologics (0, 1, or > 1). Patients who achieved a clinical response in part 1 were eligible to enter a double-blind maintenance portion of the U-ENDURE trial, and patients who did not achieve a clinical response in part 1 were enrolled in a 12-week part 2 extended treatment period, which was categorized into 2 cohorts. Patients in cohort 1 who received placebo in part 1 were given double-blind induction treatment with upadacitinib 45 mg daily for 12 weeks until week 24. Patients in cohort 2 who received upadacitinib 45 mg in part 1 were given double-blind induction treatment with upadacitinib 30 mg daily for 12 weeks until week 24.

Figure 2: Study Design Schematic for U-EXCEL

QD = once daily; UPA = upadacitinib.

Source: U-EXCEL Clinical Study Report.²²

The U-ENDURE trial was designed to evaluate the efficacy and safety of upadacitinib 15 mg and 30 mg daily as maintenance treatments versus placebo in patients with moderately to severely active CD who responded to upadacitinib induction treatment in the U-EXCEED and U-EXCEL trials. This study was a multicentre, randomized, double-blind, placebo-controlled, long-term extension trial. A total of 901 patients were enrolled between March 2018 and March 2022. The 277 study sites across 43 countries included 17 sites in Canada. No screening phase was reported. Treatment was split into substudy 1 (maintenance) and substudy 2 (long-term extension). In cohort 1 of substudy 1, patients from the U-EXCEED or U-ENDURE trial who received 45 mg of upadacitinib for the 12-week induction (including patients who did not initially achieve a clinical response with placebo and received a second 12-week induction of upadacitinib 45 mg) and achieved a clinical response were re-randomized to either upadacitinib 30 mg daily, upadacitinib 15 mg daily, or matching placebo in a 1:1:1 ratio. Patients were stratified by prior population (blinded bio-IR, open-label bio-IR, and non–bio-IR status) in the induction studies, as well as by clinical remission (as measured by PROs) and endoscopic response. In cohort 2, patients who received placebo and achieved a clinical response during the induction studies continued to receive placebo. In cohort 3, patients who achieved a clinical response during the extended treatment period with 30 mg upadacitinib after not achieving a clinical response in the induction period continued to receive 30 mg upadacitinib daily in the U-ENDURE trial. Reported results in this section are presented for substudy 1, which included a 52-week maintenance treatment for patients previously enrolled in the U-EXCEL or U-EXCEED studies. Substudy 2 of the U-ENDURE trial includes a 240-week long-term extension study, which is ongoing.

Figure 3: Study Design Schematic for U-ENDURE

QD = once daily.

Source: U-ENDURE Clinical Study Report.²³

Populations

Inclusion and Exclusion Criteria

The U-EXCEED trial enrolled patients who had an inadequate response or were intolerant to 1 or more biologics for CD (bio-IR population).

The U-EXCEL trial enrolled patients who had an inadequate response or intolerance to conventional therapies but had not failed biologic therapy (non-bio-IR population), and patients who had an inadequate response or were intolerant to 1 or more biologics for CD (bio-IR population).

Patients in the U-EXCEED and U-EXCEL trials had a confirmed diagnosis of CD (an average daily soft or liquid SF of at least 4 and/or average daily AP score of at least 2). Patients were not included if they had a current diagnosis of ulcerative or indeterminate colitis, fulminant colitis, ongoing abscess, symptomatic bowel strictures, or ileoanal pouch, or were planning bowel surgery.

Patients in the U-ENDURE trial had a confirmed diagnosis of CD, achieved a clinical response, and completed the induction studies. Patients were not included if they had hypersensitivity or a history of an AE with recurring infections.

Interventions

All 3 trials compared upadacitinib against placebo. Upadacitinib was administered for 12 weeks at 45 mg once daily for parts 1 and 2, and 30 or 45 mg once daily for part 3 of the U-EXCEED trial. For the U-EXCEL trial, upadacitinib was administered for 12 weeks at 45 mg once daily for part 1, and at 30 mg or 45 mg once daily for part 2. Patients enrolled in the U-ENDURE trial received upadacitinib for 52 weeks at 15 or 30 mg once daily for cohort 1, and 30 mg once daily for cohort 3. In all 3 trials, the matching placebo was identical in appearance to the upadacitinib tablets and packaging. The investigators, study-site personnel, data-monitoring committee, and enrolled patients were to remain blinded to treatment assignment throughout the study, except in the event of a medical situation that required unblinding.

Prior and Concomitant Medications

For the induction trials (U-EXCEL and U-EXCEED), patients must have discontinued biological therapy before the first dose of the study drug as specified in the washout procedures. Biologics and medications containing strong CYP3A inhibitors and inducers were prohibited during the study periods. Patients were allowed to enrol with concomitant use of a systemic oral corticosteroid, methotrexate, aminosalicylate, or CD-related antibiotics on stable doses.

The mandatory tapering schedule for corticosteroids began at week 4. All patients continuing in the study were to have corticosteroids discontinued no later than week 11. Patients who did not achieve a clinical response at week 12 and enter part 2 of either study without having completed the steroid taper were to resume tapering at week 16. For patients who entered the study on a corticosteroid dose that was between 2 specific levels in the tapering schedule, the dose beginning at baseline was to be rounded up to the closest higher dose. Patients were to reduce their corticosteroid dose weekly by a set amount dictated by their baseline dose of prednisone, prednisolone, budesonide, methylprednisolone, or hydrocortisone. Patients were not allowed to be on both budesonide (for CD) and prednisone (or equivalent) simultaneously.

Initiation of locally acting (rectal or suppository) or systemic corticosteroids for any reason was prohibited during the study and considered a protocol deviation. The use of inhaled or topical (except rectal or suppository) corticosteroids was not restricted.

Patients receiving a stable dose of CD-related antibiotics, aminosalicylates, or methotrexate were to maintain their concomitant treatments without dose changes through the end of the study; initiating and/or changing doses of these medications was prohibited during the study, except for decreases in the event of moderate to severe treatment-related toxicities. Use of setons as concomitant therapy was authorized for patients with perianal fistulas.

In the maintenance trial (U-ENDURE substudy 1), patients on concomitant CD-related antibiotics were allowed to discontinue them at study entry. Patients on concomitant methotrexate or aminosalicylate were to continue them on stable doses through the end of the study. Doses of CD-related antibiotics, aminosalicylates, or methotrexate could be decreased in the event of treatment-related toxicities or demonstrated inadequate response and need for rescue therapy. Patients who were on a corticosteroid were encouraged to continue the mandatory gradual taper, in which doses were reduced on a weekly basis according to baseline dose.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 7, followed by descriptions of the outcome measures and a summary of the interpretation and validity of PROs in Table 8. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence as well as any outcomes identified as important to this review by the clinical expert consulted by CADTH and stakeholder input from patient and clinician groups and public drug plans. Using the same considerations, the CADTH review team selected end points that were considered to be most relevant to inform CADTH’s expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important for informing CADTH’s expert committee deliberations were also assessed using GRADE.

Table 7: Outcomes Summarized From the Studies Included in the Systematic Review

AE = adverse event; CD = Crohn disease; CDAI = Crohn Disease Activity Index; CR-100 = decrease of at least 100 points from baseline in the Crohn Disease Activity Index; EIM = extra-intestinal manifestation; EMA = European Medicines Agency; IBDQ = Inflammatory Bowel Disease Questionnaire; PRO = patient-reported outcome.

^aStatistical testing for these end points was adjusted for multiple comparisons (e.g., hierarchal testing).

Source: Clinical Study Reports of U-EXCEED,²² U-EXCEL,²³ and U-ENDURE.²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Clinical Remission via PROs or via CDAI

The primary outcomes were the same across all 3 trials, assessed at 12 weeks in the U-EXCEL and U-EXCEED trials, and at 52 weeks in the U-ENDURE trial. The primary outcomes were clinical remission as measured by PROs for EMA regulatory purposes, and clinical remission as measured by CDAI for FDA regulatory purposes.

Both the PRO-based and CDAI-based metrics of clinical remission were included in CADTH’s assessment because, depending on jurisdiction of regulatory body (i.e., FDA versus EMA), both were considered end points coprimary with endoscopic response. Among the clinical expert consulted by CADTH, the clinician group input submitted to the review, the design of the included trials, and previous publications of trials in CD, there is a diversity of opinion on preference between PRO- and CDAI-based clinical remission for decision-making. As a result, CADTH concluded it was prudent to assess both.

Clinical remission as measured by PROs was defined as an average daily very soft or liquid SF of less than equal to 2.8 and average daily AP score of less than equal to 1.0 and both not greater than baseline, based on patient diary entries in the prior 7 days. Very soft and liquid stools were assessed using the Bristol Stool Chart, which patients were trained to recognize and calculate with during the screening period.

The National Cooperative Crohn Disease Study Group developed the CDAI using prospective data from 187 visits of 112 patients with CD.⁵⁰ The CDAI is a disease-specific index that is considered the standard for assessing CD activity in clinical trials, but is also considered impractical for real-world practice. The overall score is based on the sum of the weighted value of 8 domains and ranges between 0 to 600; a score of 150 is the threshold between remission and active disease.^51,52 Scores ranging between 150 to 219 indicate mild to moderate CD, scores between 250 and 450 indicate moderate to severe CD, and scores above 450 indicate very severe CD.^50,53 Scores for each item are derived using entries from patient diaries from the 7 days preceding each visit. The items are assessed on the daily sum per week and include the following: number of liquid or very soft stools; AP score (from 0 to 3), general well-being (from 0 to 4); sum of events per week of arthritis and/or arthralgia, mucocutaneous lesions, iris and/or uveitis, anal disease, external fistula, or fever above 37.8°C; antidiarrheal use; abdominal mass (none = 0, equivocal = 2, present = 5); a hematocrit level of 47 (males) or 42 (females); and 100 × (1 – [body weight + standard weight]).⁵⁰

Endoscopic Response (SES-CD)

Endoscopic response was a coprimary outcome in all 3 trials for both the EMA and FDA. It was assessed at 12 weeks in the U-EXCEL and U-EXCEED trials, and at 52 weeks in the U-ENDURE trial. Endoscopic response was defined as a decrease of greater than 50% from baseline in SES-CD (or for patients with an SES-CD of 4 at baseline of the induction study, at least a 2-point reduction from baseline), as scored by the central reviewer.

The SES-CD was designed to assess 4 endoscopic items: size of ulcers, ulcerated surface, affected surface, and presence of narrowing.⁵⁴ Each item is scored from 0 to 3 with a total score ranging from 0 to 56. Higher scores indicate more severe disease.

Endoscopic Remission (SES-CD)

Endoscopic remission was defined as an SES-CD of 4 or lower and at least 2-point reduction from baseline and no subscore greater than 1 in any individual variable, as scored by the central reviewer. It was assessed at 12 weeks in the U-EXCEED and U-EXCEL trials, and at 52 weeks in the U-ENDURE trial.

Corticosteroid Discontinuation (for CD) and Clinical Remission as Measured by CDAI

The 12-week outcome evaluated in the U-EXCEED and U-EXCEL trials was the proportion of patients who discontinued corticosteroids and had clinical remission as measured by CDAI among those who had been receiving corticosteroids at baseline.

The 52-week outcome evaluated in the U-ENDURE trial was the proportion of patients who had not used a corticosteroid to treat CD for at least 90 days before week 52 and who had clinical remission as measured by CDAI among those who had used a corticosteroid at induction baseline.

The trials also reported corticosteroid discontinuation and clinical remission as measured by PROs. Although not formally assessed in this review because clinical remission as measured by PROs does not have validated thresholds for disease severity or MIDs, brief comparisons are made in the results section between the CDAI-based and PRO-based results. Results were also generated for the subset of patients who were taking corticosteroids at baseline; again, although not formally assessed, the results are discussed in comparison with the outcome of interest.

Clinical Remission as Measured by CDAI and Endoscopic Remission

The proportion of patients who achieved both clinical remission as measured by CDAI and endoscopic remission as measured by SES-CD as previously described. It was assessed at 12 weeks in the U-EXCEED and U-EXCEL trials, and at 52 weeks in the U-ENDURE trial.

The trials also reported clinical remission as measured by PROs and endoscopic remission. Although not formally assessed in this review, brief comparisons between the CDAI-based end point and the PRO-based end point are made in the results section.

IBDQ, Change in Total Score From Baseline

The IBDQ is a disease-specific instrument composed of 32 Likert-scaled items in the form of a physician-administered questionnaire.^55,56 The questionnaire is divided into 4 dimensions: bowel symptoms (10 items), systemic symptoms (5 items), emotional function (12 items), and social function (5 items). Patients are asked to recall symptoms and quality of life from the last 2 weeks, and each item is graded on a 7-point Likert scale, where 1 represents the worst situation and 7 represents the best situation. The total score ranges from 32 to 224, where higher scores represent better quality of life. Scores of patients in remission typically range from 170 to 190. It was assessed at 12 weeks in the U-EXCEED and U-EXCEL trials, and at 52 weeks in the U-ENDURE trial.

CR-100

A CR-100 was defined as a decrease of at least 100 points in CDAI from baseline, which has been suggested by the FDA and EMA as a meaningful response threshold.⁵⁷ It was assessed as a proportion of patients achieving CR-100 at 12 weeks in the U-EXCEED and U-EXCEL trials, and 52 weeks in the U-ENDURE.

Resolution of EIMs in Patients With EIMs at Induction Baseline

Extra-intestinal manifestations were defined as manifestations of CD in areas of the body other than the digestive tract, including the eyes, skin, joints, mouth, and liver. It is not clear how the resolution of EIMs was defined.

Proportion With SAE

All AEs reported from the time of study drug administration until 30 days following discontinuation of study drug administration were collected, whether solicited or spontaneously reported by the patient. Patients who discontinued study drug treatment but continued to participate in the study had SAEs and nonserious AEs collected for the remainder of their study participation. In addition, SAEs and protocol-related nonserious AEs were collected from the time the patient signed the study-specific informed consent.

An AE was defined as any untoward medical occurrence in a patient given a pharmaceutical product, which does not necessarily have a causal relationship with this treatment. If an AE met any of the following criteria, it was considered to be serious:

• death of patient

• life-threatening

• hospitalization or prolonging of hospitalization

• congenital anomaly

• persistent or significant disability or incapacity

• important medical event requiring medical or surgical intervention to prevent serious outcome.

Proportion of Patients With Hospitalizations Due to Crohn Disease

The proportion of patients who had any hospitalizations due to CD at 12 weeks or at 24 weeks. A hospitalization was defined as any event that resulted in an admission to the hospital for any length of time and does not include visits to an emergency department visits or admission to an outpatient facility.

Other Outcomes

Additional AESIs, which were reported but not assessed using GRADE, include:

• serious infections

• opportunistic infections

• herpes zoster

• active tuberculosis

• malignancy (all types)

• adjudicated gastrointestinal perforations

• adjudicated cardiovascular events (e.g., major adverse cardiac event)

• anemia

• neutropenia

• lymphopenia

• renal dysfunction

• hepatic disorder

• elevated CPK

• adjudicated embolic and thrombotic events (noncardiac, non–central nervous system).

Additional outcomes reported only to inform the economic evaluation are summarized in Appendix 1:

• 5-Level EQ-5D at 12 and 52 weeks.

Table 8: Summary of Outcome Measures and Their Measurement Properties

AP = abdominal pain; CD = Crohn disease; CDAI = Crohn Disease Activity Index; CDEIS = Crohn Disease Endoscopic Index of Severity; CI = confidence interval; EMA = European Medicines Agency; EQ-5D-5-Level = 5-Level EQ-5D questionnaire; HRQoL = health-related quality of life; IBD = inflammatory bowel disease; IBDQ = Inflammatory Bowel Disease Questionnaire; ICC = intraclass correlation coefficient; MID = minimal important difference; PRO = patient-reported outcome; SES-CD = Simple Endoscopic Score for Crohn Disease; SF = stool frequency; VAS = visual analogue scale.

Statistical Analysis

A summary of statistical analyses by end point is presented in Table 9.

Primary End Points

Endoscopic response and clinical remission as measured by PROs or CDAI were the coprimary end points in all trials. All primary and secondary end points were analyzed based on the intention-to-treat population for part 1 in each study (ITT1). The comparison between the treatment groups for the coprimary end points was performed using the Cochran-Mantel-Haenszel (CMH) test and was adjusted for stratification factors including baseline corticosteroid use (yes or no), endoscopic disease severity (SES-CD < 15 or ≥ 15), and number of prior biologics used (> 1 or ≤ 1) in the U-EXCEED trial. In the U-EXCEL trial, stratification was based on baseline corticosteroid use (yes or no), endoscopic disease severity (SES-CD < 15 or ≥ 15) and number of prior biologics with prior inadequate response or intolerance (0, 1, or > 1). A CMH-based 2-sided 95% CI for the difference between the treatment groups compared to placebo was constructed for all end points. Clinical remission according to both PROs and CDAI used 3 sensitivity analyses in both the U-EXCEED and U-EXCEL trials: a “9-2 approach” (using a 7-day window before a visit, excluding 2 days around the endoscopic procedure; i.e., a total of 9 days before the visit excluding the day before and the day of endoscopy), a nonresponder imputation with no special data-handling to account for COVID-19 (NRI-NC), and “as observed” (AO). Endoscopic remission was analyzed using the NRI-NC and AO approaches in all trials.

For all primary and secondary end points, the primary approach for handling missing data was nonresponder imputation while incorporating multiple imputation to handle data missing due to COVID-19. Moreover, if the average daily very soft or liquid SF or AP score or CDAI at week 12 were missing, the nonresponder-imputation approach was applied for the clinical remission as measured by PROs at week 12 and clinical remission as measured by CDAI at the week 12 end points, respectively. Patients who discontinued the study drug before week 12 were considered “not achieved” for clinical remission or endoscopic response end points in both the U-EXCEED and U-EXCEL trials. The method used to treat missing patients was also applied to the U-ENDURE trial at week 52.

In the U-ENDURE trial, the coprimary end points between the upadacitinib and placebo group were analyzed using the CMH test and stratified by randomization factors (prior induction population), clinical remission (as measured by PROs), status (yes or no), and endoscopic response status (yes or no). A multiple-testing procedure was used to control of the type I error rate at a 0.05 significance level comparing the upadacitinib dose group to the placebo group followed by a test using a 0.025 significance level.

Secondary End Points

Categorical secondary efficacy variables were analyzed using a CMH test for stratification variables. A CMH-based 2-sided 95% CI for the difference between treatment groups versus placebo was constructed for all end points. The same stratification factors were used for each trial. Covariates including continuous secondary efficacy variables with repeated measurements, such as the change in baseline for IBDQ total scores for all trials, were analyzed using a mixed model for repeated measures (MMRM). Hospitalizations due to CD during the induction period at week 12 (U-EXCEED and U-EXCEL trials) were analyzed based on the observed data using the AO approach only. For the U-EXCEED and U-EXCEL trials, all other secondary efficacy end points were presented using both the NRI-NC and AO approaches. For the ENDURE trial, the continuous secondary efficacy variables with repeated measurements also used multiple imputation incorporating a return to baseline. All other efficacy end points except for the exposure-adjusted rate for CD-related hospitalizations (which used only the AO approach) were presented using primary-analysis NRI-NC and AO approaches. For all primary and secondary end points with a small sample size or small number of responders (e.g., occurrence of CD-related hospitalizations during part 1), upadacitinib and placebo groups were compared using a chi-square test (or Fisher's exact test if more than 20% of the cells had expected counts of less than 5).

Table 9: Statistical Analysis of Efficacy End Points in Each Trial

ANCOVA = analysis of covariance; bio-IR = inadequate response or intolerant to 1 or more biologics; CD = Crohn disease; CDAI = Crohn Disease Activity Index; CMH = Cochran-Mantel-Haenszel; CR-100 = decrease of at least 100 points from baseline in the Crohn Disease Activity Index; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; IBDQ = Inflammatory Bowel Disease Questionnaire; ITT1 = intention-to-treat (part 1 or cohort 1 of study); MI = multiple imputation; MMRM = mixed model for repeated measures; NA = not applicable; NRI-NC = nonresponder imputation with no special data-handling to account for COVID-19; PRO = patient-reported outcome; RTB-MI = return-to-baseline multiple imputation.

^aNRI-NC categorized any patients who did not have an evaluation during a prespecified window (either due to missing assessment or due to early withdrawal from the study) as a nonresponder for the visit. The only exception is that patients with missing data due to COVID-19 infection or logistical restriction were handled by MI and the patients will be characterized as responders or nonresponders based on MI imputed values. In addition, at and after the CD-related corticosteroids intercurrent event and on after the date of initiation of CD-related confounding medications after premature discontinuation of study drug, patients were considered nonresponders.

^bUsing a 7-day window before a visit, excluding 2 days around the endoscopic procedure (i.e., a total of 9 days before the visit excluding the day before and the day of endoscopy).

^cAll available measurements are used for analysis regardless of intercurrent events and do not include imputed values for missing data.

^dThe repeat measurement analysis will be conducted using a mixed model including observed measurements at all postbaseline visits up to the end of the analysis period. The mixed model includes the categorical fixed effects of treatment, visit and treatment-by-visit interaction, stratification factors at randomization, and the continuous fixed covariates of induction baseline and week 0 measurement.

^eFor continuous efficacy variables that are collected at only 1 postbaseline visit (such as the Simplified Endoscopic Score of Crohn Disease), ANCOVA will be used. The model includes the categorical fixed effects of treatment, stratification factors at randomization, and the continuous fixed covariates of induction baseline and week 0 measurement.

^fRTB-MI is multiple imputation incorporating return-to-baseline to handle visits after occurrence of intercurrent events. The least squares mean, 95% CI and standard error are synthetic result based on ANCOVA with baseline FACIT-F, week 0 FACIT-F, stratification factors (prior induction), population (non–bio-IR, double-blind bio-IR, and open-label bio-IR), clinical remission as measured by PRO status (yes or no), and endoscopic response status (yes or no).

Source: Sponsor’s Summary of Clinical Evidence.

Sample Size and Power Calculation

The U-EXCEED trial enrolled 624 patients, with 495 assigned to the randomized, double-blind, part 1 phase. Assuming a rate of 12% for clinical remission as measured by PROs in the placebo group and 29% in the upadacitinib group at week 12, a total sample size of 495 patients randomized in a 2:1 ratio (330 patients in the upadacitinib group and 165 patients in the placebo group) would be sufficient to detect at least a 17% difference between the treatment groups in clinical remission rates at week 12 using Fisher’s exact test with at least 95% power at a 2-sided significance level of 0.05. Assuming a rate of 20% for clinical remission as measured by CDAI in the placebo group and 40% in the upadacitinib group at week 12, a same sample size of 495 patients would be sufficient to detect at least a 20% difference between the treatment groups in clinical remission rates at week 12 using Fisher’s exact test with at least 95% power at a 2-sided significance level of 0.05.

The U-EXCEL trial had a sample size of 526, with all 526 assigned to the randomized, double-blind, part 1 phase. Assuming a clinical remission rate of 15% in the placebo group and 33% in the upadacitinib group at week 12, a total sample size of 501 patients randomized in a 2:1 ratio (334 patients in the upadacitinib group and 167 patients in the placebo group) would be sufficient to detect at least an 18% difference between the treatment groups in clinical remission rates at week 12 using Fisher’s exact test with at least 95% power at a 2-sided significance level of 0.05. Assuming a clinical remission rate of 21.5% as measured by CDAI in the placebo group and 45% in the upadacitinib group at week 12, this sample size would be sufficient to detect at least a 23.5% difference between the treatment groups in clinical remission rates at week 12 using Fisher’s exact test with at least 95% power at a 2-sided significance level of 0.05. Assuming an endoscopic response rate of 11.5% in the placebo group and 28.5% in the upadacitinib group at week 12, this sample size would be sufficient to detect at least a 17% difference between the treatment groups in endoscopic response rates at week 12 using Fisher’s exact test with at least 95% power at a 2-sided significance level of 0.05.

The U-ENDURE trial enrolled 901 patients, with 674 assigned to cohort 1. Assuming a week 52 clinical remission (as measured by PROs) rate of 42% for 1 of the upadacitinib dose groups and 17% for the placebo group, a total sample size of 501 patients randomized in a 1:1:1 ratio (167 patients each in the upadacitinib 30 mg daily, upadacitinib 15 mg daily, and placebo groups) would have approximately 99% power to detect at least a 25% difference between the treatment and placebo groups in clinical remission rates at week 52 using Fisher’s exact test at a 2-sided significance level of 0.025. Assuming a week-52 clinical remission (as measured by CDAI) rate of 50% for 1 of the upadacitinib dose groups and 22% for the placebo group, this would have approximately 99% power to detect at least a 28% difference between the treatment and placebo groups in clinical remission rates at week 52 using Fisher’s exact test at a 2-sided significance level of 0.025. Assuming an endoscopic response rate of 35% for 1 of the upadacitinib dose groups and 17% for the placebo group, this sample size would have approximately 94% power to detect at least a difference between the treatment and placebo groups of 18% in endoscopic response rates at week 52 using Fisher’s exact test at a 2-sided significance level of 0.025.

Statistical Testing

In both the U-EXCEED and U-EXCEL trial, the overall type I error rates of the coprimary and the secondary end points were controlled using a fixed-sequence, multiple-testing procedure as well as the Holm procedure. Multiplicity adjustment was applied to the additional efficacy end points. The analyses for additional efficacy end points were performed at the nominal alpha level of 0.05 (2-sided). Covariates such as categorical variables used a CMH test with a 95% CI. Continuous end points with repeated measurements collected longitudinally were analyzed using an MMRM and those collected at only 1 postbaseline visit were analyzed using analysis of covariance with a 95% CI.

In the U-ENDURE trial, the overall type I error rate of the coprimary and the secondary end points were controlled using a graphical multiple-testing procedure. Coprimary end points were tested using a 2-sided alpha level of 0.025 for each upadacitinib dose compared to placebo based on ITT1 for cohort 1. No statistical comparisons were performed for cohort 2 and cohort 3. The analysis of additional efficacy end points was performed at a nominal alpha level of 0.05 (2-sided) for each dose versus placebo. No multiplicity adjustment was applied to the additional efficacy end points. Similar to the other trials, covariates such as categorical end points used a CMH test with a 95% CI. Continuous secondary end points with repeated measurements were analyzed using an MMRM, and those collected at only 1 postbaseline visit were analyzed using analysis of covariance with a 95% CI.

Subgroup Analyses

Each trial conducted predefined subgroup analyses for several baseline demographic and disease-related variables, including sex, age, race, biomarkers, and clinical metrics of disease severity, treatment history, disease duration, and others. In this report, subgroup results for a number of prior biologics failed are presented for the coprimary outcomes. The clinical expert consulted by CADTH also noted that potentially important subgroups include patients with EIMs at baseline and patients with perianal disease; however, these subgroups were not prespecified in the clinical trials.

For the subgroup analysis, 95% CIs and point estimates for each treatment group as well as for treatment differences between the upadacitinib and placebo groups were presented. No P value was provided. If a subgroup population was less than 10% of the planned study size, subgroup analyses were not presented for that category. Comparability between the upadacitinib and placebo groups was presented for each dose, and no multiplicity adjustment was applied to the end points.

Analysis Populations

The analysis sets for each study are summarized in Table 10.

Table 10: Analysis Populations of U-EXCEED, U-EXCEL, and U-ENDURE

ITT = intention-to-treat population; ITT1 = intention-to-treat population in part 1, ITT2 = intention-to-treat population in part 2, ITT3 = intention-to-treat population in part 3, SA = safety population, UPA = upadacitinib.

Source: Sponsor’s Summary of Clinical Evidence.

Results

Patient Disposition

In the part 1 double-blind period of the U-EXCEED and U-EXCEL trials, which informs the 12-week outcomes, 11.1% and 8.0% of patients discontinued from the studies, respectively. Discontinuation was typically due to AEs, withdrawal by patients, or lack of efficacy. Comparing the placebo and upadacitinib groups, the rate of discontinuation was similar but slightly higher in the placebo group.

In substudy 1 of the U-ENDURE trial, which informs the 52-week outcomes, discontinuation occurred in almost a quarter of randomized patients; ||||||| ||||||||||||||| |||||||| | || |||||||||| |||||||| ||||||| || || ||||||||| ||||||| || |||||||||||||||| |||||||||||| || || || || |||||||||| || ||||||||| || |||||||||||||||| || || ||||| || ||||| || || ||||||||| |||||| || |||| || ||||||| |||||||| |||||||| || |||||||||||| || || || || |||||||| ||||||||| ||||||| ||||||||||||||| |||||||| || ||||| || ||||| || ||||||||| |||||||||||||

Table 11: Summary of Patient Disposition in U-EXCEED

ITT = intention-to-treat; NA = not applicable; NR = not reported; OL = open label; PBO = placebo; PP = per protocol; UPA = upadacitinib.

Sources: U-EXCEED Clinical Study Report.²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 12: Summary of Patient Disposition in U-EXCEL

ITT = intention-to-treat; NR = not reported; PBO = placebo; PP = per protocol; UPA = upadacitinib.

Sources: U-EXCEL Clinical Study Report.²² Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 13: Summary of Patient Disposition in U-ENDURE

ITT = intention-to-treat; OL = open-label; PBO = placebo, PP = per protocol, q.d. = once daily; UPA = upadacitinib.

Sources: U-ENDURE Clinical Study Report.²³ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Baseline Characteristics

In all 3 trials, baseline characteristics appeared to be relatively balanced between treatment arms. In each trial, there were slightly more male patients than female patients. The mean age was approximately 37 to 40 years in each trial. The majority of patients (approximately 70% to 75%) were white across all trials.

The median CD duration was 9.4 years in the U-EXCEED trial, 5.7 years in the U-EXCEL trial, and 7.6 years in the U-ENDURE trial. The longer median CD duration in the U-EXCEED trial may be reflective of the inclusion criterion requirement of inadequate response or intolerance to biologic therapy only, in contrast to the U-EXCEL trial, in which patients may have had an inadequate response or intolerance to conventional and/or biologic therapy; conventional therapy tends to occur earlier in the treatment pathway.

In the U-EXCEED trial, patients in part 1 had failed no more than 1 biologic (39.2%), 2 biologics (29.7%), or 3 or more biologics (31.1%). More than 95% of patients had prior failure to anti-TNF agents, and vedolizumab or natalizumab, whereas approximately 36% had prior failure to ustekinumab. In the U-EXCEL trial, patients were divided into subgroups based on biologic exposure: bio-IR patients had prior failure or intolerance to biologics, while non–bio-IR patients did not, but may not have necessarily been biologic-naive. Among the non–bio-IR patients in part 1 of the U-EXCEL trial, 8.7% had prior exposure to biologic therapy. Among bio-IR patients in part 1, the distribution of the number of biologics was roughly similar to that in the U-EXCEED trial (36.0%, 31.8%, and 32.2%, respectively). As the U-ENDURE cohorts comprised patients from the U-EXCEED and U-EXCEL trial, both bio-IR patients (75.6% in cohort 1) and non–bio-IR patients (24.4% in cohort 1) were included in the study, with similar proportions of patients having experienced 1, 2, or at least 3 prior biologic failures as in the parent studies.

The median baseline CDAI was 299.80 in part 1 of the U-EXCEED trial, 285.00 in part 1 of the U-EXCEL trial, and 300.50 in cohort 1 of the U-ENDURE trial, all of which fall into the “moderate to severe” category of CD activity. The median baseline SES-CD in part 1 of the U-EXCEED trial was 13.0, in part 1 of the U-EXCEL trial it was 12.0, and in cohort 1 of U-ENDURE trial it was 14.0, suggesting that patients in the U-ENDURE had more severe disease in terms of endoscopic outcomes, and patients in the U-EXCEL had less severe disease than the other trials. As the total score ranged from 0 to 56 and there was no reported MID for SES-CD, it is unknown whether these differences are substantial.

The studies were similar or different in the CD location as measured by SES-CD. The majority of patients in part 1 or cohort 1 of each trial (approximately 49% to 50%) had ileal-colonic CD, followed by colonic-only CD (approximately 35% to 41%), and finally ileal-only CD (approximately 12% to 17%). The U-EXCEL trial had the highest proportion of ileal-only CD out of the 3 included trials. The average daily frequency of very soft or liquid stools ranged from 5.5 to 5.9, and the average daily AP score was approximately 1.9 in all 3 trials. Baseline corticosteroid use ranged from approximately 36% to 38%. Baseline immunosuppressant use was less common (3.7% to 7.5%).

Exposure to Study Treatments

| | |||||||| |||||| | |||| |||||||| || ||||||||| |||||||| |||||| |||| |||| |||| || |||| |||| ||||||||| || | |||| || ||||||||||||| |||||| |||| || ||||| || |||||||| |||||||| || || || |||||||||||| | || |||||| |||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| |||||||||| ||||||||

Table 14: Summary of Baseline Characteristics of Patients in U-EXCEED

CD = Crohn disease; CDAI = Crohn Disease Activity Index; OL = open label; PBO = placebo; SD = standard deviation; SES-CD = Simplified Endoscopic Score of Crohn Disease; TNF = tumour necrosis factor; UPA = upadacitinib.

^an = 322.

^bn = 493.

^cn = 323.

^dn = 494.

Sources: Sponsor’s Summary of Clinical Evidence.

Table 15: Summary of Baseline Characteristics of Patients in U-EXCEL

bio-IR = biologic therapy–intolerant or inadequate responder; CD = Crohn disease; CDAI = Crohn Disease Activity Index; OL = open label; SD = standard deviation; SES-CD = Simplified Endoscopic Score of Crohn Disease; TNF = tumour necrosis factor; UPA = upadacitinib.

^an = 349.

^bn = 525.

Sources: Sponsor’s Summary of Clinical Evidence.

Concomitant Medications and Cointerventions

The list of concomitant medications shown in the tables in the following section were deemed appropriate by clinical data repository experts. Treatments may be recommended practice, rather than actual practice. ||||| || || |||||||| ||||||||||| ||||||||||| || || |||||||||||||| || |||| |||||| |||||| |||| |||| |||| |||||| || |||||||||| |||||||| || |||||||| | |||| |||||| ||||||||||| |||||||||| |||||||||| || ||||||||||| | |||| |||||| |||||||||| |||||||||| || |||| | |||||||||||| ||||||||||| |||||||||| || || |||||||| ||||| |||||| || |||||| |||| ||||| || |||||| |||

In the U-EXCEED and U-EXCEL trials, all patients were to undergo mandatory tapering beginning at week 4, with the goal of discontinuing corticosteroid treatment by week 11 (i.e., before the end of the induction period). In the U-ENDURE trial, at week 0, all patients who were taking corticosteroids at baseline in the U-EXCEED or U-EXCEL trials and had not completed the taper were to continue the mandatory tapering beginning at week 4, with the goal of discontinuing corticosteroid treatment by week 11.

Table 16: Summary of Baseline Characteristics of Patients in U-ENDURE

bio-IR = biologic therapy–intolerant or inadequate responder; CD = Crohn disease; CDAI = Crohn Disease Activity Index; SD = standard deviation; SES-CD = Simplified Endoscopic Score of Crohn Disease; TNF = tumour necrosis factor, UPA = upadacitinib.

Sources: Sponsor’s Summary of Clinical Evidence.

Table 17: Summary of Patient Exposure in U-EXCEED [Redacted]

|||||||||||||| || |||||||||||||| || |||||||||||| || ||| |||||| || ||||||||| || |||||| || |||||||| |||||||||| || |||||||||||||||||||||| ||||||| |||||||| || || ||||| | |||| ||

Note: This table has been redacted at the request of the sponsor.

Table 18: Summary of Patient Exposure in U-EXCEL [Redacted]

|||||||||||||| || |||||||||||||| || |||||||||||| || ||| |||||| || ||||||||| || |||||| || |||||||| |||||||||| || |||||||||||||||||||||| ||||||| |||||||| || || ||||| | |||| ||

Note: This table has been redacted at the request of the sponsor.

Table 19: Summary of Patient Exposure in U-ENDURE [Redacted]