Immunovant: The Data Arrived Early; the FDA Didn't
A surprise p-value in refractory RA hands Roivant a third pillar — and no regulatory path — while LISRAYA's first five weeks book $31 million and management quietly shelves Sjögren's.
IMVT · Earnings Call · 2026-11-09
A p-value nobody planned for
Immunovant's quarter arrived through its parent's microphone — this is Roivant's fiscal Q2 call — and Matthew Gline opened by conceding the biggest item on the tape was not on his script: “I want to jump now to another update that I wasn't totally expecting to provide today, but it's a great update” — Matthew Gline, CEO · 2026-11-09. That update was Period 2 of the IMVT-1402 difficult-to-treat RA study. The design is a randomized withdrawal trial: 118 responders from the open-label period re-randomized to 600 mg, 300 mg, or placebo, with 117 analyzed. Roughly 92.5% of patients on the high dose held an ACR20 response at week 28 versus about 75% on placebo, and the 600 mg arm cleared a p-value. ACR50 separated too, at 80% versus 56%. The genuinely surprising part is the middle arm: 300 mg behaved like placebo. As Gline put it on the call, “the fact that we were able to preserve north of 90% of those responders on drug is pretty exciting” — Matthew Gline, CEO · 2026-11-09 — particularly in patients who had already failed TNFs, JAKs and, in some cases, IL-6s. That dose dissection is the whole commercial thesis in miniature. Management has spent several quarters arguing that deeper IgG suppression is what separates the winners in FcRn biology, and Period 2 is the cleanest clinical proof yet that a shallower dose will not do: “it's clear to us that we have a competitive advantage by being able to more deeply suppress IgG and autoantibody levels, and we're trying to preserve that advantage” — Matthew Gline, CEO · 2026-11-09. It is also why they refused to show IgG or autoantibody data — a refusal that reads as strategic rather than evasive. The hole in the story is regulatory. Management had previously promised to deliver the data and the FDA conversation as one package:They hit a p-value unexpectedly, disclosed it alone, and left the design questions — one Phase III or two, what primary endpoint, parallel versus withdrawal — wide open: “we don't know on either of those questions for sure” — Matthew Gline, CEO · 2026-11-09. The company says the discussions will resolve "later this year or early next year." There is a real asset here and no filing calendar attached to it, which is a strange combination to hand a market that just took the stock down 21.8% from its August peak.My dream, our dream for the second half of this year is that we can come back with everything tied as nicely into a bow as possible, which is to say the period 2 data, the FDA conversation, maybe some further analysis of patient experience... all shared at the same time.