At week 24, the LS mean change from baseline in mTSS was similarly low among patients in all treatment groups who achieved CDAI remission at 24 weeks (n = 285; range 0.00–0.11) or LDA (n = 575; range 0.18–0.35; Fig. 2A). In patients with higher clinical activity at 24 weeks, smaller LS mean changes from baseline in mTSS were seen in the FIL200 group vs. the placebo group (0.05 vs. 0.56 in the 471 patients with MDA; − 0.26 vs. 1.12 in the 157 patients with HDA; P < 0.05 for both; Fig. 2B). Small numerical differences in LS mean change from baseline in mTSS were observed in the ADA group vs. the placebo group (0.34 vs. 0.56 in patients with MDA; 0.39 vs. 1.12 in patients with HDA) as well as in the FIL100 group vs. the placebo group (0.41 vs. 0.56 in patients with MDA; 0.72 vs. 1.12 in patients with HDA). Additionally, among patients with HDA, LS mean mTSS decreased from baseline by 0.26 in the FIL200 group, with an increase from baseline of 0.39 in the ADA group. The mTSS values at baseline and at week 24 are shown in Supplementary Table 2.

LS mean change from baseline in mTSS by remission and LDA (A) and MDA and HDA (B) at week 24. ADA adalimumab, CDAI Clinical Disease Activity Index, FIL100 filgotinib 100 mg, FIL200 filgotinib 200 mg, HDA high disease activity, LDA low disease activity, LS least squares, MDA medium disease activity, mTSS modified total sharp score, PBO placebo. *P < 0.05 vs. PBO

Among patients who did not achieve Boolean 1.0 or Boolean 2.0 remission, the LS mean changes from baseline in mTSS were significantly lower in the FIL200, FIL100, and ADA groups (0.09 [95% CI − 0.28, 0.45], 0.25 [95% CI − 0.12, 0.61], and 0.21 [95% CI − 0.18, 0.60], respectively, for Boolean 1.0, and − 0.01 [95% CI − 0.52, 0.50], 0.22 [95% CI − 0.28, 0.71], and 0.20 [95% CI − 0.32, 0.73], respectively, for Boolean 2.0) compared with the placebo group at week 24 (0.61; P < 0.05 for each; Fig. 3A and B). LS mean changes from baseline in mTSS were small in all groups among patients who achieved remission.

Change from baseline in mTSS by the achievement of Boolean 1.0 (A) and Boolean 2.0 (B) remission at week 24. ADA adalimumab, FIL100 filgotinib 100 mg, FIL200 filgotinib 200 mg, LS least squares, mTSS modified total Sharp score, PBO placebo. *P < 0.05 vs. PBO

At week 24, the LS mean change from baseline in erosion score was similarly low among patients in all groups who achieved CDAI remission (0.00 [95% CI − 0.1, 0.1], 0.00 [95% CI − 0.1, 0.1], 0.00 [95% CI − 0.1, 0.1], and 0.00 [95% CI − 0.1, 0.2] for the FIL200, FIL100, ADA, and placebo groups, respectively) or LDA (0.10 [95% CI − 0.2, 0.4], 0.10 [95% CI − 0.2, 0.4], 0.10 [95% CI − 0.3, 0.4], and 0.2 [95% CI 0.1, 0.5] for the respective groups; Fig. 4A). Small numerical differences in the LS mean change from baseline in erosion score were observed between the placebo group and the FIL200, FIL100, and ADA groups among patients who achieved MDA and HDA, but these differences were not statistically significant (Fig. 4B). Erosion also decreased from baseline by 0.20 in patients with HDA who received FIL200 but decreased by only 0.10 in those who received ADA.

LS mean change from baseline in erosion score (A, B) and JSN score (C, D) among patients by achievement of CDAI remission, LDA, MDA, and HDA at week 24. ADA adalimumab, CDAI Clinical Disease Activity Index, FIL100 filgotinib 100 mg, FIL200 filgotinib 200 mg, HDA high disease activity, JSN joint space narrowing, LDA low disease activity, LS least squares, MDA medium disease activity, PBO placebo

The LS mean change from baseline in JSN score was similarly low among patients from all groups who achieved CDAI remission (0.00 [95% CI − 0.2, 0.1], 0.00 [95% CI − 0.2, 0.2], 0.00 [95% CI − 0.2, 0.2], and 0.00 [95% CI − 0.2, 0.3] for the FIL200, FIL100, ADA, and placebo groups, respectively) or LDA (0.10 [95% CI − 0.1, 0.4], 0.10 [95% CI − 0.2, 0.3], 0.10 [95% CI − 0.2, 0.3], and 0.10 [95% CI − 0.1, 0.4] for the FIL200, FIL100, ADA, and placebo groups, respectively) at week 24 (Fig. 4C). Among patients who achieved MDA, the LS mean change from baseline in JSN score was similar among the FIL100, ADA, and placebo groups (0.20 [95% CI − 0.1, 0.6], 0.20 [95% CI − 0.2, 0.6], and 0.20 [95% CI − 0.1, 0.6], respectively) but numerically lower for the FIL200 group (0.00 [95% CI − 0.4, 0.3]; Fig. 4D). Differences in the LS mean change from baseline in JSN score for patients who achieved HDA were observed between all groups (− 0.30 [95% CI − 1.2, 0.6], 0.20 [95% CI − 0.7, 1.2], and 0.10 [95% CI − 0.8, 1.1] for the FIL200, FIL100, and ADA groups, respectively) and placebo (0.40 [95% CI − 0.5, 1.3]), but these differences were not statistically significant. Notably, JSN score decreased from baseline by 0.30 in patients with HDA who received FIL200 but increased by 0.10 in those who received ADA. The proportion of patients who had CDAI remission or HDA and achieved structural remission at week 24 (defined by a change from baseline in mTSS ≤ 0.5) was significantly greater in those who received FIL200 vs. placebo (Supplementary Table 3).

When stratified by the achievement of varying levels of pain relief, the LS mean change from baseline in mTSS was smaller for patients in the FIL200 and ADA groups with pain VAS ≤ 10 and for patients in the FIL200, FIL100, and ADA groups with pain VAS > 10 than for patients in the placebo group (P < 0.05 vs. placebo; Fig. 5A). The LS mean change from baseline in mTSS was smaller for the FIL200 group than for the placebo group, regardless of whether patients achieved pain VAS ≤ 20 (P < 0.05 vs. placebo; Fig. 5B).

LS mean change from baseline in mTSS at week 24 among patients by achievement of pain VAS ≤ 10 or > 10 (A) and pain VAS ≤ 20 or > 20 (B). ADA adalimumab, FIL100 filgotinib 100 mg, FIL200 filgotinib 200 mg, LS least squares, mTSS modified total Sharp score, PBO placebo, VAS visual analog scale. *P < 0.05 vs. PBO

Among patients who achieved remission, the LS mean changes from baseline in CDAI were significantly lower in the FIL200, FIL100, and ADA groups than in the placebo group beginning as early as week 2 and at multiple later time points through week 24 (Supplementary Table 4). Conversely, the LS mean change from baseline in CDAI was lower in the placebo group than in the FIL200, FIL100, and ADA groups only sporadically between weeks 14 and 24 among patients with LDA, MDA, or HDA.