When Bristol Myers Squibb said last October that its combination of Opdivo, Yervoy and chemo had topped solo chemo at lengthening patients’ lives, analysts and investors wondered whether the company’s three-drug regimen could spur a survival benefit on par with Merck’s market-leading Keytruda-chemo combo.

Spoiler alert: It didn’t, strictly comparing the numbers between trials—but of course, that’s a risky move investigators discourage. And the way Bristol sees it, there are plenty of other factors at play.

With a median follow-up of 8.1 months, the combination of Opdivo, Yervoy and two cycles of chemo cut the risk of death by 31% in previously untreated lung cancer patients.

With longer follow-up—a minimum of 12.7 months—it kept that survival benefit going, regardless of patients’ levels of the PD-L1 biomarker. The cocktail pared down the risk of death by 38% in patients with PD-L1 levels below 1% and by 36% in patients with PD-L1 levels of 1% or more, BMS said ahead of American Society of Clinical Oncology (ASCO) virtual annual meeting. 

The chemo combo also showed it could provoke a response among 38% of patients compared with 25% for chemo alone, and that it could keep 33% of patients progression-free at the one-year mark versus 18% for chemo. Both of those benefits reached the statistical significance threshold.

The overall survival marks are certainly ahead of the 21% death risk reduction Opdivo and Yervoy put up on their own in the Checkmate-227 trial, from which BMS unveiled three-year data on Wednesday. The idea behind adding chemo to the two immuno-oncology drugs was to deliver a “fast initial response” to pair with the “durable longer-term benefit to patients” that Opdivo and Yervoy have already shown they can provide, Wolfe Research analyst Tim Anderson wrote in an October note to clients.

Still, that figure is not exactly the 51% showing that’s helped Merck dominate the lung cancer market with its Keytrudo-chemo pairing.

Of course, cross-trial comparisons are fraught with complications—and there are other things to consider, too, Samit Hirawat, BMS’ chief medical officer, said.

For one, Bristol’s regimen involves less chemo, which is difficult for many patients to tolerate. The three-drug study, called Checkmate-9LA, revealed an opportunity to prolong survival with a “limited amount of chemo,” he said.

“If you think about a community physician sitting in the office in the outpatient setting—I don’t think they think about a hazard ratio when they’re looking at a patient, but rather they’re looking at, ‘What is the best that this patient should do, with what kind of therapy? What is the status of this patient? What can they tolerate in terms of chemo? What kind of side effects can they tolerate?’” Hirawat said. 

But giving three drugs at once may come with its own set of side-effect problems, and Bristol won’t release full safety data until its ASCO presentation later this month. The company did say in a release that “the safety profile of Opdivo plus Yervoy was consistent with previously reported studies in NSCLC, and no new safety signals were observed,” but as Anderson wrote in the October note, “It will only be once full results are presented” that a risk-benefit assessment “will be possible.”

“Opdivo plus Yervoy by itself shows toxicity, and adding chemotherapy into the mix will only increase this. The commercial value of CM-9LA will therefore depend on the balance between the clinical benefit and the toxicity,” he said, adding that bringing a third drug into the mix “also raises the cost of therapy somewhat.”