Bristol Myers Squibb just broke into previously untreated non-small cell lung cancer less than a year ago with an FDA go-ahead in PD-L1-positive patients for the dual checkpoint inhibitor regimen of Opdivo and Yervoy. But rival Merck & Co. is already casting doubt on its place in the lucrative market.
Adding Yervoy to Merck’s Keytruda didn’t extend the lives of first-line NSCLC patients whose tumors express biomarker PD-L1 at a tumor proportion score of at least 50%; instead, it increased the incidence of serious side effects, according to data unveiled at the World Conference on Lung Cancer.
Specifically, patients on the Keytruda-Yervoy combo in the phase 3 Keynote-598 study lived a median 21.4 months, compared with 21.9 months for the solo Keytruda group. The combo receivers also lived a median 8.2 months without their disease worsening, shorter than the 8.4 months for those in the Keytruda monotherapy arm.
“The monotherapy arm performed as expected,” Roy Baynes, chief medical officer at Merck Research Laboratories, said in an interview ahead of the data presentation, ruling out the scenario in which the control group outperformed and made the experimental regimen look less effective.
RELATED: Merck scraps Keytruda lung cancer trial adding Bristol Myers’ Yervoy. What about the Opdivo combo?
The long-held hypothesis is that targeting CTLA-4 would have additional benefits for certain cancer patients when used on top of PD-1 blockade, but there was no well-designed head-to-head trial to make that direct comparison—until Keynote-598.
As Baynes sees it, the answer to the PD-1/CTLA-4 question is now “pretty clear.” At least in this population, “the addition of [Yervoy] did not do anything to enhance efficacy, and it did introduce additional toxicities,” he said. “It does point out to clinicians that there’s really no need, at least in PD-L1-positive patients, to add a CTLA-4 to a PD-1.”
About 27.7% of patients on the Keytruda-Yervoy cocktail experienced serious side effects, while the rate was lower at 13.9% for solo Keytruda. The discontinuation rate was also higher for the combo than the monotherapy.
RELATED: Bristol Myers Squibb finally breaks into first-line lung cancer with Opdivo-Yervoy nod
While the trial directly showed that the addition of Yervoy did more harm and no good with Keytruda, it also indirectly raised questions around Opdivo.
The thing is, Opdivo only has a place in front-line NSCLC because of Yervoy. Back in 2016, Opdivo as a monotherapy failed to top chemotherapy in treatment-naïve patients in the CheckMate-026 trial. The drug only entered the setting last May after showing that paired with Yervoy, it could reduce the risk of death by 21% over chemotherapy in PD-L1-positive patients in the phase 3 CheckMate-227 trial.
To industry watchers, it was already clear at the time that Opdivo and Yervoy’s showing was no match to a Keytruda-chemo combo’s 51% death risk reduction versus solo chemo, regardless of PD-L1 expression.
But that still left a chance that Opdivo-Yervoy might make a better chemo-free option compared with solo Keytruda. After all, in the Keynote-042 trial, Keytruda monotherapy cut the risk of death by just 19% among first-line patients with a PD-L1 score of at least 1%—a figure that was dragged down by a lackluster performance in low PD-L1 expressers. But now that Keytruda-Yervoy has failed to top solo Keytruda, the bull case for Opdivo-Yervoy has diminished.
Since Keytruda’s chemo-combo monster showing, the Merck drug has become the standard of care in front-line NSCLC, squeezing Opdivo’s market in the second-line setting.
In a statement, Sabine Maier, Bristol Myers’ head of oncology clinical development, said the benefit of the Opdivo-Yervoy combo “has been well established in phase 3 trials across lung cancer, melanoma, renal cell carcinoma and mesothelioma,” that these studies “clearly demonstrate that combining Yervoy with Opdivo can help patients achieve durable, long-term survival outcomes.”
Editor’s Note: This story has been updated with a statement from Sabine Maier of Bristol Myers Squibb.
