Inflammatory bowel disease: Novel therpaies NZ November 2018

Transcription

1 Inflammatory bowel disease: Novel therpaies NZ November 2018 Dr Charlie Lees Consultant Gastroenterologist From the Edinburgh IBD Unit at the Western General Hospital

2 History of treatment for IBD 1950s First use of steroids 1989 First MTX study in CD 2015 Start of the biosimilar era 2017 Ustekinumab launched for CD 1960 First use of thiopurines 1990s First approval of anti-tnfs 2014 Vedolizumab launched; first gut-selective agent for IBD Tofacitinib approved for UC /JAK inhibitors 2

3 Current therapies for IBD (September 2018) Induction agents Corticosteroids 5-ASA (UC only) Anti-TNF agents Vedolizumab Ustekinumab (Crohn s only) Exclusive enteral feeding (ileal Crohn s) Limited surgical resection Maintenance agents 5-ASA (UC only) Azathioprine / MP Methotrexate (Crohn s) Anti-TNF agents Vedolizumab Ustekinumab (Crohn s only) Smoking cessation

4 Benefits of biologic agents Clinical trials in Crohn s disease have shown: Adalimumab Infliximab* Vedolizumab Ustekinumab Induction of remission CLASSIC-I 1 Targan et al. 7 GEMINI II 9 UNITI-1 12 GEMINI lll 10 UNITI-2 12 Fast onset of action Long-term remission CLASSIC-I 1 GAIN 2 Targan et al. 7 GEMINI ll 9 UNITI-1 12 UNITI-2 12 ADHERE (3yr) 3 ACCENT-1 (1yr) 8 GEMINI ll (1 yr) 9 GEMINI LTS (2 yr) 11 IM-UNITI (44 wks) 12 Mucosal healing EXTEND 4 ACCENT-1 8 VERSIFY GETAID 13 Reduced hospital admissions and operations CHARM / ADHERE 5 ACCENT-1 8 EXTEND 6 *Remicade (MSD) 1. Hanauer SB, et al. Gastroenterol 2006;130: Sandborn WJ, et al. Ann Intern Med 2007;146: Panaccione R, et al. Aliment Pharmacol Ther 2013;38: Rutgeerts P, et al. Gastroenterology 2012;142: Panaccione R, et al. Aliment Pharmacol Ther 2010;31: Colombel J-F, et al. Clin Gastro Hepatol 2014;12: Targan SR, et al. N Engl J Med. 1997;337: Rutgeerts P, et al. Gastroenterology 2004;126: Sandborn WJ, et al. N Engl J Med 2013;369: Sands BE, et al. Gastroenterology 2014;147: GEMINI LTS. NCT Available at (accessed January 2018) 12. Feagan BG, et al. N Eng J Med 2016;375: Wils P, et al. Clin Gastroenterol Hepatol 2016;14:

5 Do Conventional and Anti-TNFα Therapies Meet the Goals of Crohn s Disease Treatment? Goal 5-ASA Steroids AZA MTX Anti-TNFα Short-term endpoints Clinical remission Steroid-free clinical remission Clinical and mucosal remission 5? 3,4 Long-term disease modification Reduction of surgical risk?? Conflicting data 6? 7 Reduction of disability????? Reduction of damage????? No prospective head-to-head trials exist. No comparative conclusions should be made. 5-ASA=5-aminosalicylic acid; AZA=azathioprine; MTX=methotrexate; TNFα=tumor necrosis factor alpha. 1. Candy S et al. Gut 1995;37: Feagan B et al. N Engl J Med 1995;332: Rutgeerts P et al. Gastroenterology 2012;142: Colombel JF et al. N Engl J Med 2010;362: D Haens G et al. Gastroenterology 1997;112: Lakatos PL et al. Am J Gastroenterol 2012;107: Peyrin-Biroulet L et al. J Crohns Colitis 2011;5:

6 Crohn s: Predictors of Serious Infection Hazard Ratio (95% CI) TREAT Registry: Predictors of Serious Infection Moderate-to-severe disease activity Narcotic analgesic treatment Prednisone therapy Infliximab treatment 4.0 P<0.001 P<0.001 P=0.002 P= Crohn s disease: TREAT registry, >5 years of follow-up (N=5394) AZA, azathioprine; IFX, infliximab; MTX, methotrexate. Lichtenstein GR et al. Am J Gastroenterol. 2012;107:

7 Sustained clinical response, % Anti-TNFα Therapies Are Associated with Gradual Loss of Response Over Time ±30% primary nonresponse 1 60 ±20% loss of response in first year 2 40 ±13% annual loss of response Time, months TNFα=tumor necrosis factor alpha. Graph based on data from 1. Yanai H et al. Am J Gastroenterol 2011;106: Ben-Horin S et al. Aliment Pharmacol Ther 2011;33: Gisbert JP et al. Am J Gastroenterol 2009;104:

8 Evolution of treatment goals Clinical remission

9 Evolution of treatment goals Clinical remission Steroid-free remission Biochemical and endoscopic remission Increasing use of calprotectin

10 Evolution of treatment goals Clinical remission Steroid free remission Deep remission Biochemical and endoscopic remission Increasing use of calprotectin Treat-to-target

11 Treating a patient with Crohn s disease AND Treating intestinal inflammation Resolution of abdominal pain and normalisation of bowel habit should be the target Absence of ulceration is the target Endoscopy or cross-sectional imaging should be performed within 6 9 months after treatment starts Assess outcomes at least every 3 months until resolution After symptom resolution, outcomes should be assessed at least every 6 12 months Histological remission and biomarkers (CRP/FC) are not targets Failure of CRP or FC normalisation should prompt further endoscopic evaluation, irrespective of symptoms

12 Induction therapy & tight monitoring Mild disease Moderate disease Severe disease risk factors + risk factors Budesonide/ corticosteroids EEN or CS + AZA Anti-TNF + AZA Anti-TNF + AZA Tight monitoring Tight monitoring AZA: Azathioprine; CS: corticosteroid therapy; EEN: exclusive enteral nutrition Dr Charlie Lees, personal algorithm UK/EYV/1803/0027k April 2018

13 Induction therapy & tight monitoring Mild disease Moderate disease Severe disease risk factors + risk factors Budesonide/ corticosteroids Anti-TNF monotherapy Anti-TNF + AZA Anti-TNF + AZA Tight monitoring Tight monitoring AZA: Azathioprine; CS: corticosteroid therapy; EEN: exclusive enteral nutrition Dr Charlie Lees, personal algorithm Poor predictive ability means we are very reliant on tight monitoring

14 Young age (Beaugerie L. et al, Gastroenterology. 2006;130: Loly C, et al. Scand J Gastroenterol. 2008;43:948-54) Smoking (Franchimont D, et al. Eur J Gastroenterol Hepatol. 1998;10:821-5) + Risk factors Extensive small bowel disease (Munkholm P, et al. Gastroenterology. 1993;105: ) Peri-anal disease (Beaugerie L. et al, Gastroenterol. 2006;130: Loly C, et al.scand J Gastroenterol. 2008;43:948-54) Steroids at diagnosis (Beaugerie L, et al. Gastroenterology. 2006;130:650-6) Weight loss (Loly C, et al. Scand J Gastroenterol. 2008;43:948-54) Deep ulcerations at endoscopy (Allez M, et al. Am J Gastroenterol. 2002;89:454-9)

15 Young age (Beaugerie L. et al, Gastroenterology. 2006;130: Loly C, et al. Scand J Gastroenterol. 2008;43:948-54) Smoking (Franchimont D, et al. Eur J Gastroenterol Hepatol. 1998;10:821-5) Layer in big data from multiple omics sources - clinical and molecular phenotyping - machine learning and AI - hyper-personalized care + Risk factors Extensive small bowel disease (Munkholm P, et al. Gastroenterology. 1993;105: ) Peri-anal disease (Beaugerie L. et al, Gastroenterol. 2006;130: Loly C, et al.scand J Gastroenterol. 2008;43:948-54) Steroids at diagnosis (Beaugerie L, et al. Gastroenterology. 2006;130:650-6) Weight loss (Loly C, et al. Scand J Gastroenterol. 2008;43:948-54) Deep ulcerations at endoscopy (Allez M, et al. Am J Gastroenterol. 2002;89:454-9)

16 Induction therapy & tight monitoring Mild disease Moderate disease Severe disease risk factors + risk factors >?>?> Anti-TNF monotherapy vedolizumab Anti-TNF + AZA JAK inhibitors ustekinumab

17 Treating to target is established in other chronic inflammatory diseases Rheumatoid arthritis 1,2 Remission / low disease activity Psoriasis 3 5 BSA, PASI, DLQI PGA score What is treating to target? Why is treating to target important? Psoriatic arthritis 6,7 Remission / low disease activity Proposed treat-to-target approach in IBD BSA, body surface area; PASI, psoriasis area severity index; PGA, physician global assessment; DLQI, dermatology life quality index 1. Smolen JS, et al. Ann Rheum Dis 2010;69:631 7; 2. Smolen JS, et al. Ann Rheum Dis 2017;76: ; 3. Armstrong AW, et al. J Am Acad Dermatol 2017;76:291 98; 4. Mrowietz U, et al. Arch Dermatol Res 2011;303:1 10; 5. Gulliver W, et al. J Cutaneous Med Surg 2015;19:22 27; 6. Smolen JS, et al. Ann Rheum Dis 2014;73:6 16; 7. Gossec L, et al. Ann Rheum Dis 2016;75: Summary 19

18 TICORA demonstrated the value of tight control in rheumatoid arthritis Rheumatoid arthritis 1,2 Remission / low disease activity Patients with active RA randomised to intensive management or routine care for 18 months 3 Intensive management (n=55) Visit every month: Disease activity score calculated; If score >2.4, therapy escalated as per protocol What is treating to target? Routine care (n=55) Visit every 3 months: No disease activity score calculated; If symptomatic, therapy escalated at investigator s discretion Why is treating to target important? Compared with routine care, intensive management: Reduced disease activity Reduced radiographic disease progression Improved physical function and quality of life Incurred no additional costs Proposed treat-to-target approach in IBD Summary 1. Smolen JS, et al. Ann Rheum Dis 2014;73:6 16; 2. Gossec L, et al. Ann Rheum Dis 2016;75: ; 3. Grigor C, et al. Lancet 2004:364;

19 5-year surgery rate Use of biologics and reduced surgical rates in Crohn s disease Surgical trends in CD population-based studies PRE-BIOLOGIC ERA Year BIOLOGIC ERA IBSEN Can treating to target further decrease the number of surgeries? Stockholm County Cardiff Olmstead County Copenhagen County Manitoba Danish National Patient Register Size of circle represents number of patients in cohort What is treating to target? Why is treating to target important? Proposed treat-to-target approach in IBD Summary Olivera P, et al. Curr Opin Gastroenterol 2017;33(4): This meeting has been organised and funded by AbbVie 21

20 Treating to target in IBD: proposed stepwise algorithm for clinical practice Active IBD Target Sustained target What is treating to target? TREAT-TO-TARGET STRATEGY Why is treating to target important? 1 Assess risk factors 2 Set appropriate target 3 Treat in a timely manner 4 Monitor regularly 5 Optimise therapy as necessary TARGET ACHIEVED Continue monitoring to sustain the target Proposed treat-to-target approach in IBD TARGET NOT ACHIEVED Summary Adapted from: Bouguen G, et al. Clin Gastroenterol Hepatol 2015;13: ; Smolen JS, et al. Ann Rheum Dis 2015;0:1 13 This meeting has been organised and funded by AbbVie 22

21 Target recommendations for Crohn s disease: treat beyond symptoms * Clinical/PRO remission Endoscopic remission What is treating to target? Resolution of abdominal pain and normalisation of bowel habit Assess at least every 3 months during active disease Resolution of ulceration Assess 6 9 months during the active phase Why is treating to target important? Biomarkers: CRP and FCP are adjunctive measures of inflammation for monitoring CD (not targets) Failure of CRP or FCP normalisation should prompt further endoscopic evaluation, irrespective of symptoms Proposed treat-to-target approach in IBD Summary *Resolution of symptoms alone is not a sufficient target; objective evidence of inflammation of the bowel is necessary when making clinical decisions STRIDE, Selecting Therapeutic Targets in IBD; CRP, C-reactive protein; FCP: faecal calprotectin; PRO, patient-reported outcome Peyrin-Biroulet L, et al. Am J Gastroenterol 2015;110: This meeting has been organised and funded by AbbVie 23

22 Quality of life Timely referral and diagnosis, followed by implementing a treat-to-target approach, could potentially improve outcomes for patients with IBD Achieve target 1 Life before illness Timely referral 2 Timely diagnosis Assess risk factors Set appropriate target Treat in a timely manner Onset of symptoms 3 Diagnosis and management plan Monitor regularly 4 Monitor regularly Steroid use Optimise therapy 5 Biologic initiation Monitor regularly 6 Adherence challenges This meeting has been organised and funded by AbbVie 25

23 Quality of life Timely referral and diagnosis, followed by implementing a treat-to-target approach, could potentially improve outcomes for patients with IBD Achieve target 1 Life before illness Timely referral 2 Timely diagnosis Assess risk factors Set appropriate target Treat in a timely manner Window of opportunity Onset of symptoms 3 Diagnosis and management plan Monitor regularly 4 Monitor regularly Steroid use Optimise therapy 5 Biologic initiation Monitor regularly 6 Adherence challenges This meeting has been organised and funded by AbbVie 26

24 CD patients in remission at various time points (%) Starting anti-tnf therapy: the importance of timing 100 PRECiSE 2* <1 year SONIC 2 2 years SUTD 1 SONIC 2 >2 years CHARM <2 years 6 DIAMOND 9 CHARM 2 5 years 6 GETAID 3 PRECiSE 2* PRECiSE 2 7 CHARM 5 5 years 8 ACCENT 1 4 CHARM >5 years 6 Tailor your treat-to-target approach Assess risk of disease progression Refer and diagnose patients quickly Engage patients 20 Treat in a timely manner Disease duration (years) The products used in the clinical trials listed here may differ with respect to indications, safety, and efficacy, and are not bioequivalent. Patient baseline characteristics and concomitant therapies in trials can differ. No conclusions regarding comparative safety or efficacy can be drawn from this information *Remission evaluated after 26 weeks of treatment 1. D Haens G, et al. Lancet 2008;371:660 67; 2. Peyrin-Biroulet L, et al. Gut 2014;63:88 95; 3. Lémann M, et al. Gastroenterology 2006;130: ; 4. Hanauer S, et al. Lancet 2002;359: ; 5. Colombel JF, et al. Gastroenterology 2007;132:52 65; 6. Schreiber S, et al. J Crohns Colitis 2013;7:213 21; 7. Schreiber S, et al. New Eng J Med 2007;357:239 50; 8. Schreiber S, et al. Am J Gastroenterol 2010;105: ; 9. Matsumoto T, et al. J Crohns Colitis 2016 Aug 26 [Epub ahead of print] Monitor disease regularly Summary This meeting has been organised and funded by AbbVie 27

25 Treating to target is a disease-management strategy intended to improve patient outcomes Treating to target goes beyond symptom resolution to also reduce inflammation and improve patient quality of life Control of inflammation may help halt disease progression and prevent bowel damage and disability Treating to target involves: 1. Assessing risk factors 2. Setting an appropriate target 3. Treating in a timely manner 4. Monitoring regularly 5. Optimising therapy as necessary MODULE 1 What is treating to target? Why is treating to target important? Proposed treat-to-target approach in IBD Summary This meeting has been organised and funded by AbbVie 28

26 CALM: Evidence for a treat-to-target approach in IBD Open-label, multicentre study in patients with early * moderate-to-severe CD Patients (n=244) randomised to: Tight control (treat-to-target approach) Treatment optimisation based on biomarkers (CRP, FCP), steroid use and clinical symptoms (CDAI) Clinical management Treatment optimisation based on steroid use and clinical symptoms (CDAI) Monitored every 12 weeks Tight control Primary endpoint was mucosal healing (CDEIS <4) with absence of deep ulcers at week 48 CALM is the first study to show that timely optimisation of therapy based on clinical symptoms combined with biomarkers in patients with early * CD results in improved clinical and endoscopic outcomes than optimisation based on symptoms alone *Early CD defined in CALM as CD diagnosis confirmed by endoscopy not >6 years before baseline CDEIS, Crohn's Disease Endoscopic Index of Severity; CRP, C-reactive protein; FCP, faecal calprotectin Colombel JF, et al. Lancet 2018;390: Clinical management Clinical evidence from CALM Quality-of-life data Costeffectiveness data Evidence from other studies Summary This meeting has been organised and funded by AbbVie 29

27 Patients (%) CALM primary outcomes: Superior endoscopic and deep remission outcomes with treat-to-target Primary endpoint 45.9 p= CDEIS <4 and absence of deep ulcer 36.9 p= Deep remission CDAI <150; no steroids for 8 weeks; no draining fistula; CDEIS <4; no deep ulcers Treat-to-target group (n=122) Clinical management (n=122) 29.5 p= Biologic remission Calprotectin <250 µg/g; CRP <5 mg/l; CDEIS <4 Colombel JF, et al. Presented at Digestive Disease Week, 6 9 May 2017, Chicago, IL: Abstract 718. CDAI, Crohn s Disease Activity Index; CDEIS, Crohn s Disease Endoscopic Index of Severity.

28 New mechanisms of action and future therapies for IBD? This non-promotional meeting has been organised and funded by AbbVie 31

29 Patients (%) CALM revisited Superior endoscopic and deep remission outcomes with treat-to-target Primary endpoint 45.9 p= CDEIS <4 and absence of deep ulcer 36.9 p= Deep remission CDAI <150; no steroids for 8 weeks; no draining fistula; CDEIS <4; no deep ulcers Treat-to-target group (n=122) Clinical management (n=122) 29.5 p= Biologic remission Calprotectin <250 µg/g; CRP <5 mg/l; CDEIS <4 Colombel JF, et al. Presented at Digestive Disease Week, 6 9 May 2017, Chicago, IL: Abstract 718. CDAI, Crohn s Disease Activity Index; CDEIS, Crohn s Disease Endoscopic Index of Severity.

30 Blood vessel Pathogen Intestinal lumen Lamina propria Dendritic cell AMG 139 Risankizumab LY IEL Etrolizumab Laquinimod PF Tofacitinib Filgotinib Upadacitinib Peficitinib Ustekinumab IL-6R JAKs STATs IL-12/23R SMADs SMAD7 TGF-βRI/II Mongersen IL-10 IL-6 TNF-α IL-17A/F IL-12 (p40/p35) TGF-β1 IL-23 (p40/p19) Ozanimod T E Estrasimod Amiselimod S1P 1 S1P gradient Lymph node TNFR NF- B Transcription of target genes Macrophages Infliximab Adalimumab Golimumab Certolizumab pegol PF α4β1 MAdCAM-1 Natalizumab AJM 300 α4β7 Vedolizumab AMG 181 Etrolizumab αeβ7 VCAM-1 S1P T-cells Endothelial cell 33 Figure adapted from Coskun M, et al. Trends Pharmacol Sci. 2017;38:

31

32 Anti-a4 antibody Anti-a4b7 antibody Anti-b7 antibody Anti-Madcam1 Natalizumab (Tysabri), Biogen Vedolizumab (Entyvio), Takeda Etrolizumab, Genentech PF , Pfizer Vedolizumab is the only gut-selective integrin inhibitor with marketing authorization for moderate to severe UC or CD NVMDL symposium 2016

33 Scottish Vedolizumab Cohort Plevris N and Lees CW (2018; submitted)

34 Scottish Vedolizumab Cohort: CD Figure 2. Kaplan-Meier curves for Crohn s disease treatment outcomes stratified by previous TNFa exposure. A, cumulative rates of clinical remission; B, cumulative rates of mucosal healing; C, cumulative rates of deep remission; D, cumulative rates of colectomy. Plevris N and Lees CW (2018; submitted)

35 Scottish Vedolizumab Cohort: UC Figure 2. Kaplan-Meier curves for ulcerative colitis treatment outcomes stratified by previous TNFa exposure. A, cumulative rates of clinical remission; B, cumulative rates of mucosal healing; C, cumulative rates of deep remission; D, cumulative rates of colectomy. Plevris N and Lees CW (2018; submitted)

36 Figures adapted from Coskun M, et al. Trends Pharmacol Sci. 2017;38: and Benson J, et al. MAbs. 2011;3: UK/EYV/1803/0027k April

37 This non-promotional meeting has been organised and funded by AbbVie 42

38 * * * *products not licensed for use in IBD in the UK Figures adapted from Coskun M, et al. Trends Pharmacol Sci. 2017;38: and Benson J, et al. MAbs. 2011;3: UK/EYV/1803/0027k April 2018

39 Patients with PASI90 (%) Risankizumab* Phase II psoriasis PASI Risankizumab 18 mg Risankizumab 180 mg Risankizumab 90 mg Ustekinumab Week Grey arrow: risankizumab 18mg single dosing; black arrows: rizankizumab (90mmg or 180mg) and ustekinumab multiple dosing Papp KA, et al. N Engl J Med 2017;376: UK/EYV/1803/0027k April 2018 *Risankizumab not licensed for use in IBD in the UK

40 Proportion of patients (%) Risankizumab* Phase II Crohn s disease Patients in clinical remission (CDAI <150) at week 12 primary endpoint Adjusted = 9.0 p= Adjusted = 20.9 p= Primary endpoint Pooled risankizumab vs. placebo Adjusted = 15.1 p= n/n 6/39 10/41 15/41 25/82 Placebo 200 mg risankizumab 600 mg risankizumab Pooled risankizumab Full analysis set was used for this analysis, using non-response imputation for missing values and stratified Cochran-Mantel-Haenszel test. Adjusted and p-values are for comparisons vs. placebo. CDAI: Crohn s disease activity index. Feagan B, et al. Lancet 2017;389: UK/EYV/1803/0027k April 2018 *Risankizumab not licensed for use in IBD in the UK

41 Proportion of patients (%) Risankizumab* Phase II Crohn s disease Patients with endoscopic remission at week Adjusted = 12.1 p= Adjusted = 16.8 p= Adjusted = 14.5 p= n/n 3.0 1/39 6/41 8/41 14/82 Placebo 200 mg risankizumab 600 mg risankizumab Pooled risankizumab Endoscopic remission is a CDEIS score of 4 at Week 12 for patients with initial isolated ileitis a score of 2. Full analysis set was used for this analysis, using non-response imputation for missing values and stratified Cochran-Mantel-Haenszel tests. Adjusted and p-values are for comparisons vs. placebo. CDEIS, Crohn s Disease Endoscopic Index of Severity. Feagan B, et al. Lancet 2017;389: UK/EYV/1803/0027k April 2018 *Risankizumab not licensed for use in IBD in the UK

42 JAK inhibitors * * * * *products not licensed for use in IBD in the UK UK/EYV/1803/0027k April 2018 Figures adapted from Coskun M, et al. Trends Pharmacol Sci. 2017;38: and Pedersen J, et al. World J Gastroenterol. 2014;20:

43 JAK inhibitors What are they? Intracellular small molecules that inhibit the JAK/STAT signalling pathway 1,2 Why they may work in IBD JAKs play an essential role in inflammatory signalling 1 JAK/STAT pathway is implicated in the pathogenesis of immune-mediated inflammatory conditions, including IBD 1 What are the potential candidates? Tofacitinib* (CD: discontinued; UC: Phase III) 2 Filgotinib* (CD: Phase III; UC: Phase III) 2 Upadacitinib* (CD: Phase II; UC: Phase II) 1 Peficitinib* (UC: Phase II) 1 Distinct profiles of JAK inhibitors JAK3 Tofacitinib 3 > JAK1 > JAK2 > TYK2 Peficitinib 4 JAK3 > JAK1 > TYK2 = JAK2 Filgotinib 5 Upadacitinib 6 JAK1 1. Olivera P, et al. Expert Rev Clin Immunol. 2017; doi: / X ; 2. Coskun M, et al. Trends Pharmacol Sci. 2017;38:127 42; 3. Meyer D, et al. J Inflamm. 2010;7:41 53; 4. Ito M, et al. J Pharmacol Sci. 2017;133:25 33; 5. Van Rompaey L, et al. J Immunol. 2013;191: ; 6. Voss J, et al. Arthritis Rheumatol. 2013;65(Suppl 10):2374. *products not licensed for use in IBD in the UK UK/EYV/1803/0027k April 2018

44 Randomization Randomization Tofacitinib UC Phase 3 Program Assessment Assessment Re-randomization Assessment Enrollment Assessment OCTAVE Induction 1 A mg BID Nonresponders Placebo 8 weeks* N=598 (4:1) OCTAVE Induction 2 A Responders OCTAVE Sustain A mg BID 5 mg BID Placebo Completers & Treatment Failures OCTAVE Open A mg BID 10 mg BID 10 mg BID Responders 52 weeks* N=593 (1:1:1) Up to 1st approval N=944 Placebo 8 weeks* N=541 (4:1) Nonresponders *Final complete efficacy assessment at Week 8/52. Treatment continued up to Week 9/53. Subjects in remission at OLE baseline: 5 mg BID; all others:10 mg BID. BID=twice daily; OLE=open-label extension. 1. Clinicaltrials.gov. NCT Clinicaltrials.gov. NCT Clinical trials.gov. NCT Clinicaltrials.gov. NCT Pfizer data on file. 50

45 Patients, % Tofacitinib: Induction and maintenance results from three Phase III trials in UC OCTAVE Induction 1 Inadequate response to IMMs, CS or TNFs Tofacitinib 10 mg BID (n=476) Placebo (n=122) 18.5 p= Remission Week 8 results OCTAVE Induction 2 Inadequate response to IMMs, CS or TNFs Tofacitinib 10 mg BID (n=429) Placebo (n=112) 16.6 p< Remission 40.6 Week 52 results OCTAVE Sustain OCTAVE 1 and 2 completers Tofacitinib 10 mg BID (n=197) Tofacitinib 5 mg BID (n=198) Placebo (n=198) p<0.001 p< Remission 45.7 p<0.001 p< Mucosal healing Mayo score 2 with no subscore >1 and a rectal bleeding score of 0; Mayo endoscopic subscore 1. Sandborn WJ, et al. N Engl J Med. 2017;376:

46 Percent of Patients With Remission Primary Endpoint: Remission at Week 8 By Prior TNFi Treatment OCTAVE Induction 1 OCTAVE Induction 2 60 Placebo Tofacitinib 10 mg BID Δ=9.4 Δ= Δ= * Δ= * 15.8 * TNFi-treated TNFi-naive TNFi-treated TNFi-naive n N Efficacy data are full analysis set with nonresponder imputation. * P 0.05 vs placebo. P values based on Cochran-Mantel-Haenszel chi-square test stratified by prior treatment with TNFi, corticosteroid use at baseline, and geographic region. Sandborn W et al NEJM 2017; 376:

47 Percent of Patients With Mucosal Healing Key Secondary Endpoint: Mucosal Healing at Week 8 By Prior TNFi Treatment 60 OCTAVE Induction 1 OCTAVE Induction 2 Placebo Tofacitinib 10 mg BID 50 Δ=13.3 Δ= Δ= Δ=15.6 * * * TNFi-treated TNFi-naive TNFi-treated TNFi-naive n N * P 0.05 vs placebo. P values based on Cochran-Mantel-Haenszel chi-square test stratified by prior treatment with TNFi, corticosteroid use at baseline, and geographic region. Sandborn W et al NEJM 2017; 376:

48 Percent of Patients With Remission Primary Endpoint: Remission at Week 52 Based on Prior TNFi Failure Status (FAS, NRI, Central Read) Placebo (N=198) Tofacitinib 5 mg BID (N=198) Tofacitinib 10 mg BID (N=197) 60 Δ= Δ=12.9 Δ= Δ= Prior TNFi Failure: Yes Prior TNFi Failure: No n N1* Diff. from PBO (95% CI) (1.6, 24.2) 25.3 (13.5, 37.1) (20.0, 41.5) 33.2 (22.0, 44.4) * N1=number of subjects in each group at baseline, and used as denominator in percentage calculation. BID=twice daily; CI=confidence interval; FAS=full analysis set; NRI=non-responder imputation; PBO=placebo; TNFi=tumor necrosis factor inhibitor. 58

49 Percent of Patients With Mucosal Healing Key Secondary Endpoint: Mucosal Healing at Week 52 (FAS, NRI) Mucosal healing: Mayo endoscopic subscore of 0 or 1 Placebo (N=198) Tofacitinib 5 mg BID (N=198) Tofacitinib 10 mg BID (N=197) Δ=32.6 Δ=24.2 P< P< Δ=38.2 Δ=29.3 P< P< Central Read Local Read n Diff. from PBO (95% CI) (16.0, 32.5) 32.6 (24.2, 41.0) (20.7, 37.9) 38.2 (29.5, 46.8) P values based on Cochran-Mantel-Haenszel chi-square test stratified by induction treatment and remission at baseline. BID=twice daily; CI=confidence interval; FAS=full analysis set; NRI=non-responder imputation; PBO=placebo. 59

50 Most Common Treatment-Emergent AEs by Decreasing Frequency - Maintenance TEAEs listed were those occurring with 5% frequency in either tofacitinib group MedDRA Preferred Term Placebo N=198 Tofacitinib 5 mg BID N=198 Tofacitinib 10 mg BID N=196 Colitis ulcerative, n (%) 71 (35.9) 36 (18.2) 29 (14.8) Nasopharyngitis, n (%) 11 (5.6) 19 (9.6) 27 (13.8) Arthralgia, n (%) 19 (9.6) 17 (8.6) 17 (8.7) Headache, n (%) 12 (6.1) 17 (8.6) 6 (3.1) Upper respiratory tract infection, n (%) 7 (3.5) 13 (6.6) 12 (6.1) Rash, n (%) 8 (4.0) 6 (3.0) 11 (5.6) Blood creatine phosphokinase increased, n (%) 4 (2.0) 6 (3.0) 13 (6.6) Hypercholesterolaemia, n (%) 2 (1.0) 4 (2.0) 11 (5.6) Herpes zoster, n (%) 1 (0.5) 2 (1.0) 10 (5.1) AE=adverse event; BID=twice daily; TEAE=treatment-emergent adverse event. 60

51 Most Common Treatment-Emergent AEs by Decreasing Frequency - Maintenance TEAEs listed were those occurring with 5% frequency in either tofacitinib group MedDRA Preferred Term Placebo N=198 Tofacitinib 5 mg BID N=198 Tofacitinib 10 mg BID N=196 Colitis ulcerative, n (%) 71 (35.9) 36 (18.2) 29 (14.8) Nasopharyngitis, n (%) 11 (5.6) 19 (9.6) 27 (13.8) Arthralgia, n (%) 19 (9.6) 17 (8.6) 17 (8.7) Headache, n (%) 12 (6.1) 17 (8.6) 6 (3.1) Upper respiratory tract infection, n (%) 7 (3.5) 13 (6.6) 12 (6.1) Rash, n (%) 8 (4.0) 6 (3.0) 11 (5.6) Blood creatine phosphokinase increased, n (%) 4 (2.0) 6 (3.0) 13 (6.6) Hypercholesterolaemia, n (%) 2 (1.0) 4 (2.0) 11 (5.6) Herpes zoster, n (%) 1 (0.5) 2 (1.0) 10 (5.1) AE=adverse event; BID=twice daily; TEAE=treatment-emergent adverse event. 61

52 Upadacitanib: CELEST: Co-primary endpoints: Clinical remission based on PRO & endoscopic remission (NRI) *p<0.1; **p<0.05; ***p<.01 Clinical Remission at Week * N= N= Clinical Remission (Week 16) 35 PBO 3 mg BID 6 mg BID 12 mg BID 24 mg BID 24 mg QD Endoscopic remission: SES-CD 4 and 2 point reduction vs. baseline (BL) and no subscore > 1 in any individual variable Clinical remission: average daily very soft or liquid SF 1.5 and not worse than baseline AND average daily AP 1.0 and not worse than Baseline Sandborn WJ, et al. Digestive Disease Week; Presentation 874h UK/EYV/1803/0027k April 2018 Upadacitanib not licensed for use in IBD in the UK

53 Patients in clinical remission (%) FITZROY Primary Endpoint: Clinical remission (Crohn s Disease Activity Index <150) Filgotinib * 200 mg Placebo (N=37) (N=111) p= Weeks Vermiere et al. Lancet 2016 UK/EYV/1803/0027k April 2018 *Filgotinib not licensed for use in IBD in the UK

54 * * * *products not licensed for use in IBD in the UK Figures adapted from Degagné E, Saba JD. Clin Exp Gastroenterol. 2014;7: and Coskun M, et al. Trends Pharmacol Sci. 2017;38: UK/EYV/1803/0027k April 2018

55 Imprint tissue specific lymphocyte trafficking

56 Lymphocyte egress

57 Gut specific lymphocyte trafficking

58 Patients, % * Ozanimod: Induction and maintenance results from a Phase II trial Week 8 Week 32 p= p= Response 51.0 p<0.001 p= p= Ozanimod 1.0 mg (n=67) Ozanimod 0.5 mg (n=65) Placebo (n=65) Response Remission Mucosal healing Reduction in Mayo score 3 and 30% from baseline, with a decrease in the rectal bleeding subscore of 1 or a subscore of 1; Mayo score 2 with no subscore >1; Endoscopy subscore 1. The primary outcome was clinical remission (Mayo Clinic score 2, with no subscore >1) at 8 weeks p=0.01 p=0.005 p=0.006 Differences in the primary outcome between the group that received 0.5 mg of ozanimod and the placebo group were not significant; therefore, the hierarchical testing plan deemed the analyses of secondary outcomes exploratory Sandborn WJ, et al. N Engl J Med. 2016;374: UK/EYV/1803/0027k April 2018 *Ozanimod not licensed for use in IBD in the UK

59 Blood vessel Pathogen Intestinal lumen Lamina propria Dendritic cell AMG 139 Risankizumab LY IEL Etrolizumab Laquinimod PF Tofacitinib Filgotinib Upadacitinib Peficitinib Ustekinumab IL-6R JAKs STATs IL-12/23R SMADs SMAD7 TGF-βRI/II Mongersen IL-10 IL-6 TNF-α IL-17A/F IL-12 (p40/p35) TGF-β1 IL-23 (p40/p19) Ozanimod T E Estrasimod Amiselimod S1P 1 S1P gradient Lymph node TNFR NF- B Transcription of target genes Macrophages Infliximab Adalimumab Golimumab Certolizumab pegol PF α4β1 MAdCAM-1 Natalizumab AJM 300 α4β7 Vedolizumab AMG 181 Etrolizumab αeβ7 VCAM-1 S1P T-cells Endothelial cell 70 Figure adapted from Coskun M, et al. Trends Pharmacol Sci. 2017;38:

60 IBD_Treat-to-target: optimised Identify risk factors for progressive disease Stratify therapy by RISK NOD2; FOXOa1 Microbial signature Smoking/diet/environment NORMAL quality of life COMPLETE MUCOSAL HEALING PREVENTION OF BOWEL DAMAGE NO TOXICITY FROM DRUGS Identify dominant biological pathway Stratify therapy by BIOLOGY Anti-TNF Anti-integrin Anti-p40/p19 Anti-SP1 JAK inhibition PDE4 inhibition FMT Optimize with TDM & pharmacogenomics TREAT TO TARGET MONITORING = symptoms + inflammation Adjust therapy to reach target NB. This slide contains details of pipeline therapies that are not approved by regulatory authorities. Safety and efficacy have not been established FMT: faecal microbiota transplantation; FOXO: forkhead box O; IBD: inflammatory bowel disease; JAK: janus kinase; NOD: nucleotide-binding oligomerization domain-containing protein; PDE: phosphodiesterase; SP: service pack; TNF: tumour necrosis factor Image source: Lees C

61 QUESTIONS?