Was Bristol-Myers Squibb’s assay to blame for Opdivo’s first-line lung flop?

Bristol-Myers Squibb is testing an Opdivo-Yervoy combo in first-line lung cancer; its monotherapy approach failed last year.

Full results from Bristol-Myers Squibb’s infamous Opdivo monotherapy trial in first-line lung cancer are finally outalongside a theory about why that trial came up short. Some analysts, though, aren’t buying the new hypothesis.

In an editorial published in The New England Journal of Medicine, UCLA's Dr. Edward Garon dismisses a few possible reasons why Opdivo may have failed where Merck’s Keytruda, its archrival, succeeded. The way he sees it, Opdivo’s not an inferior drug; there were no meaningful differences in patient characteristics between the two studies’ populations, and it wasn’t an imbalance in randomization between two trial arms that sank the Opdivo study.

RELATED: BMS loses its lung cancer lead to Merck as Opdivo data come up short

It's more likely, he figures, that the failure is related to the PD-L1 assay Bristol’s been using; it could be that it “discriminates poorly at certain values,” he wrote.

It’s a possibility Bernstein analyst Tim Anderson finds “interesting,” he wrote in a Thursday note to clients, pointing out that BMS “has been quietly working behind the scenes for some time now to bring forward a new PD-L1 assay.”

But Bristol’s management, for its part, has said it’s working on the new assay to pick up a sales edge. Each hospital uses one of two different testing platforms, and by offering both assays, BMS “claims ... it will broaden Opdivo’s commercial appeal.”

And on the flip side, Evercore ISI analyst Umer Raffat isn’t buying Garon’s thesis at all. “I agree that questioning PDL1 assay cutoff would be a satisfying explanation … but we already have an answer to the assay question!” he wrote to clients, citing an analysis of Bristol and Merck’s assays presented at last year’s American Association for Cancer Research annual meeting that concluded the two assays were “strikingly similar.”

“Don’t know if author just hadn’t seen these results, but he doesn’t provide evidence in favor of his view (except that -026 failed and -024 succeeded). In contrast, we have data that clearly suggests the assays are similar,” Raffat wrote. “My conclusion: I don't see an issue in BMY's PDL1 assay.”

RELATED: Another Opdivo setback for Bristol, this time on the combo front, widens Merck's immuno-oncology lead

Whatever the reason behind the monotherapy failure, BMS is still gunning for a spot in the lineup of treatments available for first-line lung cancer, which represents a mammoth patient pool and a key source of sales. It’s working to bring a combination of Opdivo and CTLA4 therapy Yervoy to market in the space, though after rejigging its trial design, the company isn’t expecting results until 2018.

Investors won’t have to wait that long for clues as to whether the pairing will succeed, though. AstraZenecawhich is testing a CTLA4 combo of its ownis expecting a readout later this year.