Efficacy and Safety of Filgotinib as Induction and Maintenance Therapy for Crohn's Disease: Results From the Phase 3 Randomized, Double-Blind, Placebo-Controlled DIVERSITY1 Study

Introduction: We evaluated the efficacy and safety of filgotinib (FIL), an oral, once-daily, Janus kinase 1 preferential inhibitor, for the treatment of Crohn’s disease (CD). Methods: DIVERSITY1 (NCT02914561) was a phase 3 double-blind, placebo (PBO)-controlled trial in adults aged 18–75 years with moderately to severely active CD. Patients were randomized 1:1:1 to FIL 200 mg (FIL200), FIL 100 mg (FIL100) or PBO once daily for 10 weeks in Induction (IND) Study A (biologic [bio]-naive and bio-experienced) or B (bio-experienced). Non-EU-specific co-primary endpoints at IND Study W10 and Maintenance (MNT) Study W58 were CDAI clinical remission (score < 150) and endoscopic response assessed by SES-CD score (≥ 50% reduction from baseline). FIL-treated patients were rerandomized 2:1 to their IND dose or PBO in the MNT Study if at W10 they had endoscopic response or were in PRO2 clinical remission (the EU-specific co-primary endpoint). Results: In IND Studies A and B, 707 and 665 patients were randomized to FIL200 (n=223; n=204), FIL100 (n=245; n=230) or PBO (n=239; n=231), respectively. In the MNT Study, 481 patients were rerandomized to continue FIL or PBO treatment (FIL200–FIL200, n=118; FIL200–PBO, n=56; FIL100–FIL100, n=105; FIL100–PBO, n=56; PBO–PBO, n=146). Among treated patients, 89.3% (629/704), 83.8% (552/659) and 48.1% (230/478) completed the IND A, IND B and MNT studies, respectively. In the IND Studies at W10, a significantly higher proportion of patients in the FIL200 group vs PBO group were in CDAI clinical remission (IND A: 32.9% vs 19.8%; IND B: 26.7% vs 14.8%; both p < 0.05). The proportions of patients who had endoscopic response at W10 were not statistically significantly different for either FIL dose vs PBO (Figure 1). In the MNT Study at W58, the proportions of patients in CDAI clinical remission were not statistically significantly different for either FIL dose vs PBO. A significantly greater proportion of patients in the FIL200–FIL200 group vs FIL200–PBO group had endoscopic response at W58 (30.4% vs 9.4%; p = 0.0038). Secondary endpoints are summarized in the Table 1. FIL had a safety profile in CD that was consistent with that in UC. Conclusion: In the DIVERSITY1 study in CD, FIL200 was not superior to PBO based on both co-primary endpoints in IND. Nonetheless, FIL200 was efficacious in inducing CDAI clinical remission at W10. Among patients who had endoscopic response or were in PRO2 clinical remission at W10, FIL200 was efficacious in achieving endoscopic response at W58 in MNT.Figure 1.: Co-primary Endpoints (CDAI Clinical Remission and Endoscopic Response) at W10 of A) IND Study A and B) IND Study B, and at W58 of C) MNT Study of FIL in CD (Full Analysis Sets). a: The Full Analysis Set for the MNT study includes all rerandomized patients who met the protocol definition of PRO2 clinical remission or endoscopic response at W10 and received at least one dose of FIL or PBO during the MNT study. The Full Analysis Set for the MNT study excluded 19 rerandomized patients (FIL200–FIL200, n=6; FIL200–PBO, n=3; FIL100–FIL100, n =7; FIL100–PBO, n=3). Each co-primary endpoint comparison of FIL200 vs PBO was tested at the 2-sided 0.05 significance level first. If the difference was statistically significant for both co-primary endpoints, the secondary endpoints and comparing FIL100 vs PBO for the co-primary endpoints were tested at the 2-sided 0.025 significance level using hierarchical testing within each sequence. CDAI clinical remission was defined as a CDAI score of less than 150. Endoscopic response was defined as a reduction of at least 50% in Simple Endoscopic Score for CD from IND baseline. CD, Crohn’s Disease; CDAI, Crohn’s Disease Activity Index; CI, confidence interval; FIL, filgotinib; FIL100, filgotinib 100 mg; FIL200, filgotinib 200 mg; IND, Induction; MNT, Maintenance; PBO, placebo; PRO2, 2-item patient-reported outcomes; W10, week 10; W58, week 58. Table 1. - Selected Secondary Endpoints in IND and MNT Studies of FIL in CD (Full Analysis Sets) IND Study, A or B (W10) AFIL200n=222 AFIL100n=245 APBOn=237 A∆% FIL200 vs PBO (95% CI); P valueand∆% FIL100 vs PBO (95% CI); P value BFIL200n=202 BFIL100n=228 BPBOn=229 B∆% FIL200 vs PBO (95% CI); P valueand∆% FIL100 vs PBO (95% CI); P value PRO2 clinical remission, n (%) 73/222 (32.9) 75/245 (30.6) 61/237 (25.7) 6.9 (−1.4, 15.2); 0.0963and4.2 (−3.9, 12.2); 0.3050 60/202 (29.7) 43/228 (18.9) 41/229 (17.9) 11.9 (3.7, 20.2); 0.0039and1.1 (−6.2, 8.5); 0.7556 CDAI clinical response, n (%) 116/222 (52.3) 113/245 (46.1) 94/237 (39.7) 12.0 (3.2, 20.8); 0.0074and5.5 (−3.2, 14.1); 0.2159 78/202 (38.6) 81/228 (35.5) 63/229 (27.5) 11.6 (2.6, 20.6); 0.0110and8.2 (−0.4, 16.7); 0.0593 MNT Study (W58) FIL200–FIL200n=112 FIL200–PBOn=53 ∆% FIL200–FIL200 vs FIL200–PBO (95% CI); P value FIL100–FIL100 n=98 FIL100–PBOn=53 ∆% FIL100–FIL100 vs FIL100–PBO (95% CI); P value PRO2 clinical remission, n (%) 49/112 (43.8) 14/53 (26.4) 16.8 (2.0, 31.6); 0.0382 29/98 (29.6) 13/53 (24.5) 5.8 (−8.5, 20.0); 0.4263 CDAI clinical response, n (%) 51/112 (45.5) 18/53 (34.0) 10.9 (−4.7, 26.4); 0.1827 33/98 (33.7) 16/53 (30.2) 4.0 (−11.1, 19.2); 0.5944 6-month CS-free CDAI clinical remission, n (%) 17/50 (34.0) 5/25 (20.0) 14.0 (−5.1, 33.1); 0.1900 2/41 (4.9) 3/25 (12.0) −5.5 (−22.6, 11.6); 0.4248 In line with the hierarchical testing strategy, any statistical significance for the secondary endpoints is nominal owing to the co-primary endpoints not being met in IND Study A, IND Study B or MNT Study. Therefore, no conclusions about the statistical significance of the secondary endpoints can be made. PRO2 clinical remission was defined as an abdominal pain score of not more than 1 and a stool frequency score of not more than 3. CDAI clinical response was defined as a reduction from IND baseline by at least 100 points or a CDAI score of less than 150. Six-month CS-free CDAI clinical remission was defined as a CDAI score of less than 150, with no use of CS indicated for CD for at least 6 months prior to MNT Study W58 among patients with CS use at MNT baseline. CD, Crohn’s Disease; CDAI, Crohn’s Disease Activity Index; CI, confidence interval; CS, corticosteroid; FIL, filgotinib; FIL100, filgotinib 100 mg; FIL200, filgotinib 200 mg; IND, Induction; MNT, Maintenance; PBO, placebo; PRO2, 2-item patient reported outcome; W10, week 10; W58, week 58.