The Risks and Benefits of JAK Inhibitors in Acute Severe UC Peter D.R. Higgins, MD, PhD, MSc 

Acute Severe Ulcerative Colitis (ASUC) • Roughly 30% admitted for ASUC will have a colectomy in 1 year – 25% at 90 days, most during initial hospital stay • High risk of VTE due to inflammation, steroids • Rapid leak of proteins, including biologics • Extended delay of surgery = worse outcomes, death • Best predictors of steroid response - # BMs, CRP at 72 h – Lichtiger, Travis, Ho indices include these • Increasing numbers of ASUC patients have failed at least one biologic, often an aTNF 2 MICHIGAN IBD ● JAKs for ASUC 

Standard of Care for ASUC • IV fluids – 5-10L in first 12 h • IV corticosteroids ~ 60 mg solumedrol qday/divided • SQ Enoxaparin 40 mg daily or equivalent • Assess indices at 72h, make rescue decision – Doing well, can taper to po steroids – Doing poorly, rescue with • Colectomy • Cyclosporine, as a bridge to vedolizumab/ustekinumab/thiopurines • Accelerated IFX • UM Colectomy rate - 25.8% at 90 days - unchanged in years 3 MICHIGAN IBD ● JAKs for ASUC http://www.med.umich.edu/ibd/docs/severeucprotocol.pdf 

New Options for ASUC • JAK inhibitors are rapid (change in rectal bleeding by day 3)1 – Stop the inflammatory momentum of severe UC • High risk ASUC is identifiable – Already on steroids – Already failed biologic – High CRP (30-300 mg/L) and low Albumin (<3.3) after hydration – OSH Transfer already failing IV steroids • Should we wait 72h to fail IV steroids in high-risk ASUC? 4 MICHIGAN IBD ● JAKs for ASUC 1Hanauer, S., et al. Clin Gastroenterol Hepatol. 2019; 17: 139-147. 

Rationale for Accelerated JAKi for ASUC • Approved for moderate to severe UC • Rapid onset of action • Rapidly cleared (t1/2 = 3.2h)2 – Rapid washout in 18h if need for surgery • Efficacy of 15 mg tid in phase 2 trial • Less susceptible to colon leak & drug loss • No risk of anti-drug antibodies to JAKi • Option for prior aTNF failure • Short trial of JAKi is cheaper than 10 mg/kg IFX – 10 mg tid x 3 days $1,444 vs 700 mg single dose $3,220 • High dose use is short-term, limited risk 5 MICHIGAN IBD ● JAKs for ASUC 2Lambal, Clin Pharm. 2016 Sandborn. N Engl J Med. 2012 

New Option for ASUC • Standard of Care IV steroids & tofacitinib 10 mg tid – Based on 15 mg bid in phase 2, adapted for short half life • Start ASAP upon admission (infectious studies pending) – Taper to 10 bid before discharge with po steroids • Potentially rapid and beneficial • But what are the Risks and Benefits? 6 MICHIGAN IBD ● JAKs for ASUC 

Risks of Accelerated Tofacitinib/Steroids • Venous thromboembolism • Acne • Infection, including shingles • Adverse cardiac events • Perforation • Worse surgical outcomes • Every adverse event with corticosteroids • Unable to get tofacitinib outpatient prior authorization 7 MICHIGAN IBD ● JAKs for ASUC 

Risks of Accelerated Tofacitinib/Steroids • Venous thromboembolism – Increased by ASUC, immobility, steroids – Increased by high dose tofacitinib – Mostly in elderly with RA on chronic steroids or malignancy, quite rare in UC trials population without other risk factors (e. g. cancer)3 – Must use daily anticoagulation. Patient must understand the rationale, must not refuse dosing. 8 MICHIGAN IBD ● JAKs for ASUC 3Sandborn, WJ. Aliment Pharmacol Ther. 2019; 50: 1068-1076. 

Risks of Accelerated Tofacitinib/Steroids • Adverse Cardiac Events – RA patients 50 years of age and older with at least one CV risk factor treated with tofacitinib 10 mg BID had a higher rate of all-cause mortality, including sudden CV death. – Occurred in elderly with RA on methotrexate and chronic steroids, but quite rare in UC trials population3. – Must assess CV risk. http://tools.acc.org • Not a viable strategy with elevated CV risk in > 50 yo – Must stop other small molecules (MTX, Aza, 6-MP) 9 MICHIGAN IBD ● JAKs for ASUC 3Sandborn, WJ. Aliment Pharmacol Ther. 2019; 50: 1068-1076. 

Risks of Accelerated Tofacitinib/Steroids • Infections, including shingles • Most common infections in UC trials: – Pneumonia – ideally Prevnar and Pneumovax – Cellulitis – inspect skin daily while on steroids plus JAKi – Herpes zoster – Shingrix if available – Urinary tract infection – No catheters – Diverticulitis – Appendicitis 10 MICHIGAN IBD ● JAKs for ASUC 

Risks of Accelerated Tofacitinib/Steroids 11 MICHIGAN IBD ● JAKs for ASUC Risk Mitigation Venous Thromboembolism Enoxaparin 40 mg SQ daily, discuss rationale with patient (no refusals) Acne Topical antibiotic cream prn Infection, including shingles Shingrix if available Adverse cardiac events Assess risk, avoid in ASCVD risk > 20% http://tools.acc.org Toxic Megacolon/Perforation Avoid opioids, control inflammation Corticosteroids AEs Monitor glucose, BP Litigation Discuss risks and alternatives with patient, including colectomy. Early surgical consultation (d2). Unable to get tofa prior auth Start outpatient approval on day 1, apply for patient assistance program as backup plan ASAP. 

Accelerated Tofacitinib/Steroids for Whom? 12 MICHIGAN IBD ● JAKs for ASUC Will a colectomy prevent me from having kids? 10-year CV risk 0.9% Could this clog up my heart stents? 10-year CV risk 67% 

But Do Accelerated JAKi/Steroids Help? • Case series of 4 patients, 9 doses of tofacitinib over 3 days3 • Several struggle with outpatient authorization • One failed after 9 doses – CRP rises rapidly – to colectomy. 13 MICHIGAN IBD ● JAKs for ASUC 3Berinstein, JA, et al. CG&H 2019;17:988–990. 

But Do Accelerated JAKi/Steroids Help? • Case-control study of accelerated inpatient ASUC therapy • Compare: – ASUC patients who received tofacitinib (10 bid or 10 tid) – Contemporary controls (not as sick as tofa & steroids group) – Historical controls (includes full range of severe to toxic) • Control for: – Prior biologic failure – CRP peak – Albumin nadir – Endoscopic Mayo score – Colonic dilation >5.5 cm at any point during hospitalization 14 MICHIGAN IBD ● JAKs for ASUC 

Patient Selection for Case-Control • ASUC defined by Truelove & Witts’ criteria PLUS started IV steroids • Period 2010-2020 – Matched each tofacitinib case (2016-2020) to 3 controls • Two control groups: – Contemporary control group (match by sex and month of admission) – Historical control group (match by sex and month, and date of admission was prior to 2016) 15 MICHIGAN IBD ● JAKs for ASUC 

Patient Selection 16 MICHIGAN IBD ● JAKs for ASUC Tofacitinib Group: Inclusion Criteria: - Dx of UC - Age >18 - IV CS + T & W - Initiation of tofa inpatient Exclusion Criteria: - Biologic naïve - Dual IFX and Tofa - Post-colectomy Control Group: Inclusion Criteria: - Dx of UC - Age >18 - IV CS + T & W Exclusion Criteria: - Post-colectomy - Part of Tofa group Cases matched to 3 randomly selected controls according sex date of admission 

Outcomes and Analysis • Outcomes: – Colectomy at 30 days and 90 days – Complications at 90 days • Analysis: – Kaplan Meier plots and Cox proportional hazards regression – Adjusted hazard ratios (HR) and 95% confidence intervals (CI) • HR >1 àincreased colectomy risk • HR <1 àreduced colectomy risk • Sub-Group Analysis: – Risk of colectomy according to Tofacitinib dose 17 MICHIGAN IBD ● JAKs for ASUC 

Study Inclusion 18 MICHIGAN IBD ● JAKs for ASUC 

Participant Characteristics 19 MICHIGAN IBD ● JAKs for ASUC Tofacitinib (N=40) Contemporary Control (N=113) Historic Control (N=115) Age, years 34.41 38.42 40.61 Female 24 (60.0%) 58 (51.3%) 69 (60.0%) Race White 36 (90.0%) 100 (88.5%) 103 (89.6%) Black 3 (7.5%) 5 (4.4%) 8 (7.0%) Other 1 (2.5%) 8 (7.1%) 4 (3.5%) Duration IBD, years 10.35 8.41 5.64 Prior Biologic Use 40 (100.0%) 44 (38.9%) 30 (26.1%) OSH Transfer 1 (2.6%) 7 (6.2%) 16 (14.0%) Duration Steroids, days 50.50 9.12 15.43 Truelove Witt’s Score 1.82 1.99 1.77 CRP (mg/dL) 6.29 7.71 8.18 Albumin(g/dL) 3.40 3.28 3.18 Admission Hgb 11.71 12.07 12.14 Colonic Dilation ` 6 (5.7%) 9 (8.4%) Endoscopic Mayo Score Mayo 1 1 (2.6%) 5 (5.0%) 1 (1.1%) Mayo 2 10 (25.6%) 40 (39.6%) 43 (46.7%) Mayo 3 28 (71.8%) 56 (55.4%) 48 (52.2%) C. diff Infection 6 (15.4%) 10 (8.8%) 15 (13.2%) CMV Colitis 3 (8.1%) 1 (1.0%) 4 (4.3%) Rescue Cyclosporine N/A 2 (1.8%) 3 (2.6%) Rescue Infliximab N/A 43 (38.1%) 22 (19.5%) Accelerated Rescue Infliximab N/A 13 (11.5%) 6 (5.2%) Tofacitinib dose 10mg BID 16 (40.0%) N/A N/A 10mg TID 24 (60.0%) N/A N/A Follow-up, years 0.83 0.99 5.54 

Participant Characteristics 20 MICHIGAN IBD ● JAKs for ASUC Tofacitinib (N=40) Contemporary Control (N=113) Historic Control (N=115) Age, years 34.41 38.42 40.61 Female 24 (60.0%) 58 (51.3%) 69 (60.0%) Race White 36 (90.0%) 100 (88.5%) 103 (89.6%) Black 3 (7.5%) 5 (4.4%) 8 (7.0%) Other 1 (2.5%) 8 (7.1%) 4 (3.5%) Duration IBD, years 10.35 8.41 5.64 Prior Biologic Use 40 (100.0%) 44 (38.9%) 30 (26.1%) OSH Transfer 1 (2.6%) 7 (6.2%) 16 (14.0%) Duration Steroids, days 50.50 9.12 15.43 Truelove Witt’s Score 1.82 1.99 1.77 CRP (mg/dL) 6.29 7.71 8.18 Albumin(g/dL) 3.40 3.28 3.18 Admission Hgb 11.71 12.07 12.14 Colonic Dilation ` 6 (5.7%) 9 (8.4%) Endoscopic Mayo Score Mayo 1 1 (2.6%) 5 (5.0%) 1 (1.1%) Mayo 2 10 (25.6%) 40 (39.6%) 43 (46.7%) Mayo 3 28 (71.8%) 56 (55.4%) 48 (52.2%) C. diff Infection 6 (15.4%) 10 (8.8%) 15 (13.2%) CMV Colitis 3 (8.1%) 1 (1.0%) 4 (4.3%) Rescue Cyclosporine N/A 2 (1.8%) 3 (2.6%) Rescue Infliximab N/A 43 (38.1%) 22 (19.5%) Accelerated Rescue Infliximab N/A 13 (11.5%) 6 (5.2%) Tofacitinib dose 10mg BID 16 (40.0%) N/A N/A 10mg TID 24 (60.0%) N/A N/A Follow-up, years 0.83 0.99 5.54 

Participant Characteristics 21 MICHIGAN IBD ● JAKs for ASUC Tofacitinib (N=40) Contemporary Control (N=113) Historic Control (N=115) Age, years 34.41 38.42 40.61 Female 24 (60.0%) 58 (51.3%) 69 (60.0%) Race White 36 (90.0%) 100 (88.5%) 103 (89.6%) Black 3 (7.5%) 5 (4.4%) 8 (7.0%) Other 1 (2.5%) 8 (7.1%) 4 (3.5%) Duration IBD, years 10.35 8.41 5.64 Prior Biologic Use 40 (100.0%) 44 (38.9%) 30 (26.1%) OSH Transfer 1 (2.6%) 7 (6.2%) 16 (14.0%) Duration Steroids, days 50.50 9.12 15.43 Truelove Witt’s Score 1.82 1.99 1.77 CRP (mg/dL) 6.29 7.71 8.18 Albumin(g/dL) 3.40 3.28 3.18 Admission Hgb 11.71 12.07 12.14 Colonic Dilation ` 6 (5.7%) 9 (8.4%) Endoscopic Mayo Score Mayo 1 1 (2.6%) 5 (5.0%) 1 (1.1%) Mayo 2 10 (25.6%) 40 (39.6%) 43 (46.7%) Mayo 3 28 (71.8%) 56 (55.4%) 48 (52.2%) C. diff Infection 6 (15.4%) 10 (8.8%) 15 (13.2%) CMV Colitis 3 (8.1%) 1 (1.0%) 4 (4.3%) Rescue Cyclosporine N/A 2 (1.8%) 3 (2.6%) Rescue Infliximab N/A 43 (38.1%) 22 (19.5%) Accelerated Rescue Infliximab N/A 13 (11.5%) 6 (5.2%) Tofacitinib dose 10mg BID 16 (40.0%) N/A N/A 10mg TID 24 (60.0%) N/A N/A Follow-up, years 0.83 0.99 5.54 

Results: Colectomy Risk 22 MICHIGAN IBD ● JAKs for ASUC Risk of Colectomy at 30 Days Risk of Colectomy at 90 Days vs Contemporary Controls vs Historic Controls HR 95% CI p-value HR 95% CI p-value Risk of Colectomy at 30 Days Tofacitinib 0.29 0.09, 0.97 0.044 0.29 0.09, 0.95 0.041 Albumin (g/dL) 0.19 0.07, 0.49 <0.001 0.19 0.07, 0.48 <0.001 Prior Biologic 4.24 1.33, 13.5 0.014 4.38 1.37, 14.0 0.013 Colonic Dilation 5.07 1.52, 16.9 0.008 5.05 1.51, 16.9 0.009 Endo Mayo Score 7.12 1.45, 35.0 0.016 7.19 1.48, 34.9 0.014 Risk of Colectomy at 90 Days Tofacitinib 0.28 0.11, 0.76 0.012 0.28 0.10, 0.75 0.011 Albumin (g/dL) 0.28 0.13, 0.61 0.002 0.27 0.12, 0.60 0.001 Prior Biologic 4.27 1.62, 11.3 0.003 4.40 1.66, 11.7 0.003 Colonic Dilation 6.53 2.02, 21.1 0.002 6.53 2.02, 21.1 0.002 Endo Mayo Score 6.71 1.90, 23.7 0.003 6.60 1.90, 22.9 0.003 vs Contemporary Controls vs Historic Controls HR 95% CI p-value HR 95% CI p-value Risk of Colectomy at 90 Days Tofacitinib 0.28 0.11, 0.76 0.012 0.28 0.10, 0.75 0.011 Albumin (g/dL) 0.28 0.13, 0.61 0.002 0.27 0.12, 0.60 0.001 Prior Biologic 4.27 1.62, 11.3 0.003 4.40 1.66, 11.7 0.003 Colonic Dilation 6.53 2.02, 21.1 0.002 6.53 2.02, 21.1 0.002 Endo Mayo Score 6.71 1.90, 23.7 0.003 6.60 1.90, 22.9 0.003 

Results: 90-Day Colectomy Risk 23 MICHIGAN IBD ● JAKs for ASUC 90 Day Survival Analysis HR 0.28 (95% CI 0.11−0.76, p=0.012) 0.75 0.80 0.85 0.90 0.95 1.00 0 20 40 60 80 Time (Days) Colectomy Avoidance Rate Tofacitinib vs Contemporary Controls HR 0.28 (95% CI 0.10−0.75, p=0.011) 0.75 0.80 0.85 0.90 0.95 1.00 0 20 40 60 80 Time (Days) Colectomy Avoidance Rate Tofacitinib vs Historic Controls Controls Tofacitinib 

Results: Colectomy Risk by Tofacitinib Dose 24 MICHIGAN IBD ● JAKs for ASUC 90 Day Survival Analysis HR 0.13 (95% CI 0.03 to 0.60, p=0.008) HR 0.63 (95% CI 0.20 to 1.96, p=0.4) 0.75 0.80 0.85 0.90 0.95 1.00 0 20 40 60 80 Time (Days) Colectomy Avoidance Rate Tofacitinib vs Contemporary Controls HR 0.13 (95% CI 0.03 to 0.58, p=0.008) HR 0.62 (95% CI 0.20 to 1.93, p=0.4) 0.75 0.80 0.85 0.90 0.95 1.00 0 20 40 60 80 Time (Days) Colectomy Avoidance Rate Tofacitinib vs Historic Controls Controls Tofacitinib 10mg BID Tofacitinib 10mg TID 10 mg tid 10 mg tid 10 mg bid 10 mg bid CS alone CS alone 

Results: 90-Day Complication Rate 25 MICHIGAN IBD ● JAKs for ASUC Tofacitinib (N=34*) Contemporary Control (N=90*) p value VTE 1 (2.9%) 1 (1.1%) 0.470 Any Infection 8 (23.5%) 11 (12.2%) 0.119 COVID-19* 0 (0.0%) 0 (0.0%) Bacteremia 1 (2.9%) 0 (0.0%) 0.102 Urinary Tract Infection 1 (2.9%) 1 (1.1%) 0.470 Pneumonia 1 (2.9%) 2 (2.2%) 0.816 Cellulitis 0 (0.0%) 0 (0.0%) Intra-abdominal Infection 0 (0.0%) 0 (0.0%) Other Bacterial Infection 0 (0.0%) 1 (1.1%) 0.537 C. Diff Infection 4 (11.8%) 8 (8.9%) 0.629 Non-C. Diff Enteric Infection 1 (2.9%) 1 (1.1%) 0.470 Opportunistic Infection 1 (2.9%) 2 (2.2%) 0.816 Herpes Zoster 0 (0.0%) 0 (0.0%) Cardiovascular Events 0 (0.0%) 0 (0.0%) Post-Surgical Infection*** 0/6 (0.0%) 9/23 (39.1%) 0.065 Unplanned Readmissions 9 (26.5%) 16 (17.8%) 0.282 Steroids Dependence 17 (50.0%) 28 (31.1%) 0.051 

Results: 90-Day Complication Rate 26 MICHIGAN IBD ● JAKs for ASUC Tofacitinib (N=34*) Contemporary Control (N=90*) p value VTE 1 (2.9%) 1 (1.1%) 0.470 Any Infection 8 (23.5%) 11 (12.2%) 0.119 COVID-19* 0 (0.0%) 0 (0.0%) Bacteremia 1 (2.9%) 0 (0.0%) 0.102 Urinary Tract Infection 1 (2.9%) 1 (1.1%) 0.470 Pneumonia 1 (2.9%) 2 (2.2%) 0.816 Cellulitis 0 (0.0%) 0 (0.0%) Intra-abdominal Infection 0 (0.0%) 0 (0.0%) Other Bacterial Infection 0 (0.0%) 1 (1.1%) 0.537 C. Diff Infection 4 (11.8%) 8 (8.9%) 0.629 Non-C. Diff Enteric Infection 1 (2.9%) 1 (1.1%) 0.470 Opportunistic Infection 1 (2.9%) 2 (2.2%) 0.816 Herpes Zoster 0 (0.0%) 0 (0.0%) Cardiovascular Events 0 (0.0%) 0 (0.0%) Post-Surgical Infection*** 0/6 (0.0%) 9/23 (39.1%) 0.065 Unplanned Readmissions 9 (26.5%) 16 (17.8%) 0.282 Steroids Dependence 17 (50.0%) 28 (31.1%) 0.051 ***of the 6 and 23 who went to colectomy, after 18-hour washout for tofacitinib Rx 

Strengths and Limitations • First study to compare inpatient tofacitinib induction to matched controls • Largest study to date • Relatively small size of our tofacitinib group N=40 • Non-randomized retrospective study • Possible signal of longer steroid dependence/taper • May be related to prior authorization issues 27 MICHIGAN IBD ● JAKs for ASUC 

Case-Control Conclusions • Tofa 10 mg TID with IV steroids is associated with a lower risk of colectomy vs. IV steroid SoC in biologic-experienced patients • Unadjusted 2/24 (tid) vs. 4/16 (bid) vs. 23/90 (controls) • Tofacitinib 10 mg BID was not associated with reduced colectomy • Tofacitinib 10 mg TID & SoC IV Steroids may be an effective inpatient induction strategy for biologic-exposed ASUC patients • Inpatient Tofacitinib does not appear to increase complication risk after a small N of 40 • Prospective trials are needed to define the safety profile, optimal dose, frequency, and duration of JAK/CS induction for ASUC 28 MICHIGAN IBD ● JAKs for ASUC 

The Risks and Benefits of JAKs for ASUC • A potential benefit to accelerated, early dosing with IVCS • Risks are poorly defined, and will require larger N • Short-term induction use and mitigation strategies will likely limit the risks • Patient selection is critical (no cancer, CV risk < 20%) • 2nd generation JAK1-selective agents appear more potent, could be safer • 3rd generation TEC, Tyk2 kinase inhibitors could be even more effective • RCTs of rapid small molecule induction strategies are needed to reduce colectomy rates in ASUC 29 MICHIGAN IBD ● JAKs for ASUC 

Thanks to • Jeffrey Berinstein • Jessica Sheehan • Elliot Bernstein • Calen Steiner • Michael Dias • Randolph Regal • Kelly C. Cushing • Ryan W. Stidham • Shrinivas Bishu • Jami A.R. Kinnucan • Shirley A. Cohen-Mekelburg 30 MICHIGAN IBD ● JAKs for ASUC Jeffrey Berinstein, MD