The menin fault lines emerge

Syndax and Kura continue to offer simple exposure to the success or failure of menin inhibition and, after the companies squared off at last year’s ASH conference, Syndax today tried to put distance between itself and its rival. The attempt didn’t have the desired effect, however.

Investors will note that at ASH the companies’ duelling datasets each had their strengths and weaknesses. But now Kura has stepped back from one genetically defined niche, AML carrying a KMT2A rearrangement – the subgroup Syndax talked up at an investor event today while also revealing waning efficacy and the emergence of a new toxicity signal.

The battle concerns Syndax’s revumenib and Kura’s ziftomenib, which both inhibit menin. Both had been tested in broad relapsed/refractory AML populations before activity was narrowed down to patients with KMT2A rearrangement and NPM1 mutation, and trials were enriched for these subgroups.

Registrational cohort

Today Syndax presented data on the registrational cohort of its phase 1/2 Augment-101 study, which enrolled 94 AML and ALL patients confirmed to have KMT2A rearrangement.

The good news was that a 23% complete response rate here among 57 patients – the remaining 37 were apparently not evaluable or had insufficient follow-up – was enough to meet the trial’s goal. This sets up a regulatory filing by the end of 2023, Syndax said.

However, investors will recall that at ASH the CR rate among the 46 KMT2Ar AML patients in Augment-101’s phase 1 portion was 33%, or 27% in subjects given the recommended phase 2 dose. Though the phase 2 portion of Augment-101 had set just 10% as the bar for success, 23% is slightly lower than before, and falls short of the 25-30% efficacy Barclays analysts had said would be “positive”.

There was more concerning news on the safety side. Syndax today revealed a 16% rate of grade 3 or higher differentiation syndrome, having reported none in December. Grade 3 QTc prolongation, a known toxicity of revumenib, stood at 14%, versus 13% at ASH.

Syndax said there had been no discontinuations related to either adverse event, but the emerging toxicity picture is is another black mark against revumenib. The fact revumenib hadn’t been associated with severe differentiation syndrome was a positive versus Kura’s ziftomenib, whose ASH dataset had seen high rates of this toxicity – though ziftomenib continues to cause no QTc prolongation.

Duel of the menin inhibitors in r/r AML

Note: *the registrational cohort includes 8 patients with ALL, with the remaining 49 having AML; **across both cohorts and all doses grade 3+ QTc prolongation was seen in 13% of patients at ASH 2022.

However, the twist doesn’t end there. Kura’s dataset, from the phase 1/2 Komet-001 study, showed the majority of differentiation syndrome cases to occur in KMT2Ar patients, where ziftomenib appeared not to be especially active – leading Kura to throw in the towel and focus ziftomenib monotherapy instead on NPM1m disease.

The relevance is in terms of market size: NPM1r represents some 30% of AML patients, versus around 10% for KMT2A-rearranged disease. In theory Kura is therefore limiting itself to a smaller slice of the pie, while Syndax is still ostensibly targeting both subsets.

However, in NPM1m Kura might have the upper hand. Not only is ziftomenib more active in NPM1m AML than revumenib on a cross-trial basis, with a 35% CR rate reported at July’s EHA conference, if severe differentiation syndrome is now seen in revumenib’s dataset then any safety advantage Syndax might have claimed evaporates.

Kura is now enrolling NPM1m patients into Komet-001’s pivotal stage, while Syndax promises an update on Augment-101’s NPM1m cohort by the end of the year. Another cross-trial numbers battle beckons.