42 - One-Year Comparative Effectiveness of Ustekinumab vs Tofacitinib for Ulcerative Colitis After Anti-Tumor Necrosis Factor Failure

Tuesday, October 25, 2022

3:05 PM – 3:15 PM ET

Location: Hall C2

Presenting Author(s)

Rahul Dalal, MD, MPH

Brigham and Women's Hospital
Boston, MA

PDF of Slides View Slides

Rahul Dalal, MD, MPH¹, Puza Sharma, MD, PhD², Kanwal Bains, MBBS¹, Jordan Pruce, BS¹, Jessica R. Allegretti, MD, MPH³
¹Brigham and Women's Hospital, Boston, MA; ²Pfizer Inc., New York, NY; ³Brigham and Women’s Hospital Crohn’s and Colitis Center, Boston, MA

Introduction: Tofacitinib (tofa) is an oral small molecule JAK inhibitor for treatment of ulcerative colitis (UC). Real-world data comparing effectiveness of ustekinumab (uste) vs tofa are limited. We compared 52 wk outcomes and drug survival of uste vs tofa for UC.

Methods: In this retrospective cohort study, adults initiated uste or tofa after failure of >1 anti-TNF agent 5/1/18-4/1/21 at a large US academic center. Electronic records were reviewed. The primary outcome was steroid-free clinical remission (SFCR; i.e. simple clinical colitis activity index <2 or provider assessment and no use of oral/IV steroids for >30 days) at 12 and 52 (+/-4) weeks. Other outcomes: drug survival, endoscopic response/remission, biochemical response/remission, improvement in arthralgia, hospitalization, and colectomy within 52 wks. Reasons for discontinuation are reported descriptively. Inverse probability of treatment weighted (IPTW) logistic and Cox regression were used to calculate adjusted odds ratios (aORs) and hazard ratios (aHRs). Kaplan-Meier drug survival curves were compared using log-rank tests.

Results: 97 pts initiated uste and 69 initiated tofa with median follow-up of 62.0 wks (IQR 35.6-97.0 wks) and 88.0 wks (IQR 40.9-153.1 wks), respectively. Baseline characteristics were similar except for immunomodulator use, Mayo endoscopic subscore, and CRP (Table 1). At 12 wks, 31/96 (32.3%) uste pts and 36/68 (52.9%) tofa pts were in SFCR (p< 0.01). At 52 wks, 44/90 (48.9%) uste pts and 37/66 (56.1%) tofa pts were in SFCR (p=0.38). 38/97 uste pts and 30/69 tofa pts discontinued tofa during follow-up: non-response (includes colectomy; 89.5% uste, 83.3% tofa), dysplasia requiring colectomy (7.9% uste, 6.7% tofa), insurance or adherence (0% uste, 6.6% tofa), and adverse events (2.6% uste, 3.3% tofa). Patients who discontinued for adverse events had nausea with arthralgia (n=1, uste) and elevated liver enzymes (n=1, tofa). Unadjusted outcomes are presented in Fig 1A. After IPTW, covariate balance was confirmed with < |10%| standardized differences (Fig 1B). There was no association between tofa vs uste for SFCR 12 wks (aOR 1.94, 95% CI 0.96-3.92) or 52 wks (aOR 1.16, 95% CI 0.58-2.31). Drug survival was similar between groups (aHR 1.26, 95% CI 0.74-2.15) (Fig 1C).

Discussion: In a real-world, anti-TNF exposed cohort of UC pts, uste and tofa showed similar effectiveness/safety at 52 wks with approximately 50% of pts in SFCR. Due to limited sample size, larger real-world studies are needed to confirm these findings.

Figure: Figure 1. A. Unadjusted outcomes. Definitions of other outcomes: endoscopic response (improvement in Mayo endoscopic subscore by >1 point) and remission (Mayo endoscopic subscore=0) within 52 weeks, biochemical response (improvement in elevated C-reactive protein [CRP] or fecal calprotectin [FC] by >25% from baseline) and remission (normalized CRP or FC) within 52 weeks. Denominators for outcomes vary due to differences in available data pre- and post-drug initiation: SFCR 12 wks (n = 96 uste, 68 tofa), SFCR 52 wks (n = 90 uste, 66 tofa), endoscopy (n = 31 uste, 28 tofa), arthralgia (n = 19 uste, 22 tofa), biochemical (n = 35 uste, 28 tofa), hospitalization (n = 66 uste, 45 tofa), colectomy (n = 67 uste, 50 tofa), discontinuation (n = 93 uste, 66 tofa). Hospitalization and colectomy only consider patients who remained on treatment for full duration of 52 weeks unless outcome was met earlier. *P-values comparing uste vs tofa proportions for each outcome were calculated using Fisher’s exact test; only SFCR 12 wks was significant at p < 0.05. B. Covariate balance before and after IPTW. These covariates, which were used to calculate propensity scores, were chosen a priori based on clinical significance and data availability. C. Kaplan-Meier analysis stratified by treatment group. Patients were censored at loss to follow-up or when they discontinued treatment for reasons unrelated to efficacy (e.g. adverse event, dysplasia, adherence). Abbreviations: SFCR = steroid-free clinical remission, IPTW = inverse probability of treatment weighting

Baseline Characteristics	Ustekinumab (n=97)	Tofacitinib (n=69)	P-value*
Female	49 (51%)	42 (61%)	0.19
Age, y, median (IQR)	35.5 (29.4, 50.4)	41.2 (28.1, 54.0)	0.25
UC duration, y, median (IQR)	9.0 (4.1, 13.5)	9.5 (4.4, 15.5)	0.39
Race
Caucasian	85 (88%)	63 (91%)	0.40
Black	4 (4%)	0 (0%)
Asian	5 (5%)	4 (6%)
Other/Unknown	3 (3%)	2 (3%)
Ethnicity
Non-Hispanic	89 (92%)	69 (100%)	0.05
Hispanic	4 (4%)	0 (0%)
Unknown	4 (4%)	0 (0%)
Prior malignancy	5 (5%)	4 (6%)	0.86
Number of prior biologics, median (IQR)	2 (2,3)	2 (2,3)	0.62
Number of prior anti-TNFs, median (IQR)	1 (1,2)	2 (1,2)	0.18
Prior vedolizumab	64 (66%)	51 (74%)	0.27
Prior tofacitinib	25 (26%)	0 (0%)	n/a
Prior ustekinumab	0 (0%)	8 (12%)	n/a
Prior 5-ASA	94 (97%)	67 (97%)	0.94
Current 5-ASA	19 (20%)	10 (14%)	0.39
Prior immunomodulator	70 (72%)	54 (78%)	0.37
Current immunomodulator	24 (25%)	6 (9%)	0.008
Current Oral/IV corticosteroids			0.41
Prednisone/Methylprednisolone	51 (53%)	30 (43%)
Budesonide	11 (11%)	7 (10%)
BMI, kg/m2, median (IQR)	25.1 (21.7, 29.0)	25.79 (21.8, 28.9)	0.97
Arthralgia at time of drug initiation	26 (27%)	26 (38%)	0.14
Last Montreal disease extent >E1 (i.e. >proctitis)	75 (77%)	59 (86%)	0.19
Last Mayo endoscopic subscore (severity)			0.049
0 (None)	10 (10%)	6 (9%)
1 (Mild)	20 (21%)	7 (10%)
2 (Moderate)	32 (33%)	37 (54%)
3 (Severe)	35 (36%)	19 (28%)
Smoking			0.31
Never	70 (72%)	56 (81%)
Current	2 (2%)	2 (3%)
Former	25 (26%)	11 (16%)
Current cannabis use	22 (23%)	9 (13%)	0.12
Current opioid use	3 (3%)	6 (9%)	0.12
UC hospitalization within 12 months	21 (22%)	18 (26%)	0.51
Serum albumin, g/dL, median (IQR)	4.1 (3.8, 4.4)	4.1 (3.8, 4.3)	0.47
C-reactive protein, mg/L, median (IQR)	2.8 (1, 7)	5.1 (1.8, 22.8)	0.01
Fecal calprotectin > 120 ug/g	49 (88%)	25 (89%)	0.81
SCCAI, median (IQR)	5 (3, 7)	5 (4, 8)	0.46
Daily bowel movement frequency, median (IQR)	6 (4, 9)	6 (4, 10)	0.57

Table: Table 1. Baseline Characteristics
*Calculated using Pearson’s chi squared and Wilcoxon rank-sum tests
Albumin, C-reactive protein, fecal calprotectin, SCCAI, and bowel movements were the most recent values available within 3 month prior to drug initiation. The most recent endoscopic data preceding drug initiation was used. Median time from endoscopic evaluation to drug initiation was 22.4 weeks (IQR 3.9 -44.1 weeks) for ustekinumab and 19.1 weeks (IQR 6.1-46.1 weeks) for tofacitinib. Abbreviations: IQR = interquartile range, TNF= tumor necrosis factor, ASA = aminosalicylic acid, SCCAI = simple clinical colitis activity index

Disclosures:

Rahul Dalal: Centaur – Consultant. Janssen – Grant/Research Support. Pfizer – Grant/Research Support.

Puza Sharma: Pfizer – Employee.

Kanwal Bains indicated no relevant financial relationships.

Jordan Pruce indicated no relevant financial relationships.

Jessica Allegretti: Artugen – Consultant. Baccain – Consultant. Bristol Myers Squibb – Consultant, Speaker. Ferring Pharmaceuticals – Consultant. Finch Therapeutics – Consultant, Grant/Research Support. Iterative Scopes – Consultant. Janssen – Consultant, Grant/Research Support. Merck – Grant/Research Support. Morphic – Consultant. Pfizer – Consultant, Grant/Research Support. Seres Therapeutics – Consultant. Servatus – Consultant. Takeda – Consultant.

Rahul Dalal, MD, MPH¹, Puza Sharma, MD, PhD², Kanwal Bains, MBBS¹, Jordan Pruce, BS¹, Jessica R. Allegretti, MD, MPH³, 42, One-Year Comparative Effectiveness of Ustekinumab vs Tofacitinib for Ulcerative Colitis After Anti-Tumor Necrosis Factor Failure, ACG 2022 Annual Scientific Meeting Abstracts. Charlotte, NC: American College of Gastroenterology.