An oral drug held platelet gains for about a quarter of refractory ITP patients

An oral drug called rilzabrutinib produced platelet responses that lasted, and that sometimes turned up in people who had not responded at first, in the open-label phase of the phase 3 LUNA3 trial in hard-to-treat immune thrombocytopenia [s1]. The honest scale of the benefit matters as much as its existence: in the trial's controlled phase, a durable response was reached by 23% of patients on the drug versus none on placebo — real and statistically clear, but a minority [s2].

Immune thrombocytopenia (ITP) is an autoimmune disorder in which the immune system destroys platelets and blunts their production, leaving too few of the cells that stop bleeding. People with persistent or chronic ITP who have already failed several treatments have few good options, and the ones that work often do so by broadly suppressing the immune system. Rilzabrutinib is a covalent, reversible inhibitor of Bruton tyrosine kinase (BTK), a signalling enzyme; blocking it is thought to act on several immune mechanisms at once — dampening antibody-driven platelet destruction while sparing the platelets themselves [s2].

What the earlier, controlled phase showed

LUNA3 was a multicentre, randomised, phase 3 trial in adolescents and adults, funded by Sanofi [s1]. Its double-blind period, reported separately, is the part that carries the strongest evidence because it compared the drug against placebo [s2]. Previously treated adults were randomly assigned 2:1 to oral rilzabrutinib 400 mg twice daily (133 patients) or placebo (69 patients) for 24 weeks; the group was a difficult one, with a median ITP duration of 7.7 years and 28% having already had their spleen removed [s2].

The primary endpoint — a durable platelet response, defined as a count of at least 50 x 10^9 per litre for most of the last 12 of 24 weeks without rescue treatment — was met by 31 of 133 patients (23%) on rilzabrutinib and none on placebo (P<0.0001) [s2]. Responses came quickly in those who had them, with a median of 15 days to a first platelet response, and the drug cut the use of rescue therapy by 52% [s2]. That is the comparison the chart shows: 23% against zero.

What the open-label phase adds

After the blinded period, patients could enter a 28-week open-label phase in which everyone received rilzabrutinib, including those who had been on placebo and those who had not initially responded [s1]. Of 202 patients randomised in the double-blind period, 180 entered the open-label phase — 115 who had been on rilzabrutinib and 65 who had been on placebo; their median age was 48 years and 112 (62%) were female [s1].

Two findings stand out. First, responses held up over time: 46 (23%) of 198 patients exposed to rilzabrutinib had a stable platelet response by the data cutoff, and complete responses were recorded in 42 (23%) of 180 [s1]. Second, some patients who had been on placebo, or who had not met the response bar in the blinded phase, did respond once on the drug — 14 (22%) of the 65 former-placebo patients reached a durable response during the open-label phase [s1]. Patient-reported physical fatigue and bleeding scores were sustained or improved [s1].

Safety in the open-label phase looked manageable. Treatment-related adverse events occurred in 46 (26%) of 180 patients, most commonly mild-to-moderate diarrhoea (17, 9%) and nausea (17, 9%); treatment-related grade 3 events occurred in four (2%) patients and serious treatment-related events in two (1%) — a case of interstitial lung disease, and a case of bronchopulmonary aspergillosis with cytomegalovirus infection — and no deaths occurred during that period [s1]. In the earlier blinded phase, one patient on the drug had a serious blood clot in the leg and another died of a pneumonia judged unrelated to treatment [s2].

How to read it

The useful way to hold these numbers together is that rilzabrutinib helps a defined minority of people with otherwise refractory ITP, and when it helps, the effect is durable rather than fleeting — an important property for a chronic condition treated for years [s1][s2]. The open-label data are genuinely informative about durability and about late responders, but they are not controlled: once everyone is taking the drug, there is no placebo group to separate a real effect from the natural ups and downs of platelet counts, so the strongest efficacy claim still rests on the 23%-versus-zero comparison from the blinded phase [s1][s2]. An oral option that avoids broad immunosuppression is a meaningful addition for this group; it is not a response for everyone, and the trial does not claim to be.

Related coverage has examined a B-cell-depleting antibody tested in lupus, a small-interfering-RNA drug for the antibody-driven disease myasthenia gravis, and a European regulator's backing of the first drug for non-CF bronchiectasis.

What to watch

The company is running a further trial, LUNA4, to test rilzabrutinib earlier in the course of ITP and to see whether it can produce sustained responses that hold after treatment stops [s1]. The question that will decide its place in practice is not whether it works — the blinded trial settled that for a minority — but for whom, and whether earlier use widens the share of patients who benefit [s1][s2].

This article describes research and is not medical advice. Treatment of immune thrombocytopenia is a decision for patients and their haematologists.

Sources

• Clinical activity and safety of rilzabrutinib in patients with previously treated immune thrombocytopenia (LUNA3): results from the open-label period of a phase 3 trial — The Lancet Haematology, 17 September 2026
• Safety and efficacy of rilzabrutinib vs placebo in adults with immune thrombocytopenia: the phase 3 LUNA3 study — Blood, 12 June 2025