Background: The use of the Physician Global Assessment (PGA) subscore, a component of the full Mayo Clinic Score (MCS), has been discouraged as a primary endpoint by the US Food and Drug Administration (FDA) and the European Medicines Agency in 2016 draft guidance. Instead, a composite, clinical remission primary endpoint based on stool frequency (SF), rectal bleeding (RB), and endoscopic subscores (ES) (i.e. MCS w/o PGA) is recommended by the FDA. Etrolizumab, a humanized anti-β7 mAb, showed greater clinical remission at week (wk) 10 based on the full MCS (w/PGA) compared with placebo (PBO) in the phase 2 EUCALYPTUS trial (Vermeire et al. Lancet 2014). The percentage of patients (pts) achieving clinical remission was numerically lower with higher vs lower etrolizumab dose. This treatment effect was more prominent in anti-tumour necrosis factor-a (aTNF)-naive pts. We evaluated overall symptomatic and endoscopic remission endpoints in aTNF-naive pts enrolled in EUCALYPTUS.
Methods: EUCALYPTUS was an international, multicentre, double-blind, PBO-controlled, randomised, phase 2 study (NCT01336465) in 124 pts with moderate-to-severe UC who had not responded to conventional therapy. Eligible pts were randomly assigned (1:1:1) to subcutaneous etrolizumab (100 mg at wks 0, 4 and 8, with PBO at wk 2; or 420-mg loading dose at wk 0, followed by 300 mg at wks 2, 4 and 8), or matching PBO. The primary endpoint was clinical remission (full MCS ≤2, with no individual subscore >1 at wk 10). Post-hoc analyses assessed SF remission (SF ≤1 and ≥1-point decrease from baseline), RB remission (RB =0), symptomatic remission (SF and RB remission — SFRB), endoscopic remission (ES ≤1), and a composite of both SFRB and ES remission in aTNF-naive pts at wk 10.
Results: PBO-adjusted treatment differences observed with symptomatic and endoscopic remission assessments were similar (30–37%) to those of the primary full MCS remission endpoint (36%), except for RB remission, which had the highest PBO rate and the smallest treatment effect size (24%). Remission rates based on symptomatic remission (SF, SFRB), endoscopic remission (ES) or ES + SFRB were similar between the 100 and 300 mg etrolizumab arms.
Conclusions: When PGA is removed from the MCS-defined remission assessment, aTNF-naive pts experienced similar rates of remission whether treated with low- or high-dose etrolizumab. Treatment effects observed with SFRB remission and ES + SFRB remission were of similar magnitude to that observed for the primary MCS remission endpoint.