Endometrial cancer becomes a three-horse race

Monday's US approval of Imfinzi plus chemo for mismatch repair-deficient (dMMR) endometrial cancer came as a surprise; AstraZeneca had quietly said results of the supporting Duo-E study had been accepted for regulatory review, but didn't reveal anything about the setting it was gunning for.

In fact, in approving this Imfinzi indication the FDA has gone along with a post-hoc subgroup analysis of Duo-E, as well as not backing a regimen that included Lynparza, contained in the third cohort of this study. The approval came on the same day the FDA greenlit a Merck & Co's front-line Keytruda regimen, and both nods will challenge the standing of GSK’s Jemperli in endometrial cancer.

Importantly, the Keytruda approval, based on the NCI-sponsored Keynote-868 study in stage III-IV disease, concerns an all-comers population, irrespective of MMR status. That's wider than the dMMR patients to which Imfinzi will be limited; meanwhile, Jemperli has been carving out a niche in dMMR endometrial cancer, first in second and then in front-line disease.

The next battleground for Jemperli is approval for first-line endometrial cancer irrespective of MMR status, on which the FDA is due to decide by 23 August. Meanwhile, Keytruda recently stumbled in the adjuvant endometrial cancer setting, with the failure of the Keynote-B21 trial of a chemo/radiotherapy combo.

Duo-E

For Astra’s part, Duo-E had already been presented as a win, with data at last year’s ESMO suggesting a positive outcome for both its active treatment arms.

Duo-E gave first-line patients Imfinzi plus chemo, followed by maintenance with Imfinzi monotherapy (cohort B), or with Imfinzi plus Lynparza (cohort C). On its primary endpoint, PFS versus chemo alone, both cohorts B and C scored statistically significant benefits; cohort C also numerically beat cohort B, with a 0.78 hazard ratio for this exploratory analysis.

However, it was apparent that for both active cohorts the benefit was driven by patients whose cancer was dMMR – comprising only some 20% of Duo-E’s roughly 700-strong population. And, looking only at dMMR, cohort C no longer looked better than cohort B, with PFS hazard ratios versus control of 0.41 and 0.42 respectively.

Duo-E summarised

Notes: *exploratory analysis of arm C vs arm B showed HR=0.78; **exploratory analysis of arm C vs arm B showed HR=0.97. Source: JCO & OncologyPipeline.
 

Detailed results of a post-hoc subgroup analysis looking only at dMMR patients were presented in March at the Society of Gynecologic Oncology meeting. It was then that Astra revealed that the Duo-E data had been accepted for review by regulatory authorities in the US, Europe and Japan, without disclosing precisely the setting or combo under review.

Now it seems that the FDA was insufficiently convinced by Astra’s data in MMR-proficient endometrial cancer to grant an all-comers label, and that it didn’t see the addition of Lynparza to the maintenance phase as bringing additional benefit. The PARP inhibitor would logically be expected to bring extra toxicity.

In light of this Keytruda's corresponding approval must seem rather galling for Astra. The Keynote-868 trial yielded a hazard ratio for PFS of 0.30 in patients with dMMR endometrial cancer, but only 0.60 in those whose disease was MMR-proficient; nevertheless, the FDA has granted the Merck drug an all-comers front-line label, perhaps because both subgroup results were strongly statistically significant.

The full importance of the Keytruda and Imfinzi approvals will only become clear after the FDA rules on the filing designed to bring Jemperli to front-line endometrial cancer patients irrespective of MMR status.

Selected trials in endometrial cancer

Notes: MSI-H=microsatellite instability high; dMMR=mismatch repair deficient; MSS=microsatellite stable; pMMR=mismatch repair proficient. Source: OncologyPipeline.

This is an updated version of a story published earlier.