Suffering two separate clinical trial blows, Bristol Myers Squibb and a collaboration between Merck KGaA and Pfizer learned the hard way that head and neck cancer is difficult to treat—even though the experimental therapies may not entirely be to blame.
Despite Opdivo’s success in previously treated squamous cell carcinoma of the head and neck (SCCHN), Bristol Myers found that adding Yervoy to the PD-1 inhibitor couldn’t significantly outdo the standard-of-care EXTREME therapy containing Eli Lilly’s Erbitux in prolonging lives for newly diagnosed patients.
Among front-line patients with recurrent or metastatic head and neck cancer whose tumors expressed the PD-L1 biomarker at a combined positive score of 20 or above, Opdivo and Yervoy pared down the risk of death by 22% over EXTREME. However, the improvement failed to meet the statistical significance bar. The results were published at the European Society for Medical Oncology 2021 virtual meeting.
Patients with PD-L1 scores of at least 20 lived a median 17.6 months while on Opdivo-Yervoy, versus 14.6 months for EXTREME, results from the CheckMate-651 trial showed.
During an interview, BMS’ head of oncology development, Jonathan Cheng, noted that patients in the EXTREME group performed better than expected, which he attributed to a high rate of subsequent use of immunotherapy after disease progression. To be specific, 46% patients in the EXTREME arm and 8% in the Opdivo-Yervoy arm received subsequent immunotherapy.
Previously in Keytruda’s Keynote-048 trial—which in 2019 earned the Merck PD-1 drug two simultaneous FDA go-aheads in front-line recurrent or metastatic SCCHN—Keytruda monotherapy cut the risk of death by 39% and a Keytruda-chemo combo the same risk by 40% over EXTREME in patients with PD-L1 levels at 20 or above. The median overall survival lengths among those patients were 14.9 months for Keytruda monotherapy, 14.7 months for Keytruda-chemo, and 10.7 months for chemo alone.
“We do think it’s not uncommon in oncology for the standard of care to continue to improve as time goes on with better supportive care and better treatments,” Cheng said. “In this case particularly, subsequent therapy clearly has a large impact as the checkpoint inhibitors … get more and more used in later lines of therapy.”
Cheng declined to comment on BMS’ regulatory plan but said that investigators and other physicians who’ve tried this regimen are “enthusiastic about the potential benefit of checkpoint inhibitors,” including the Opdivo-Yervoy combo, for head and neck cancer patients. The immunotherapy doublet has previously failed to top Opdivo monotherapy in previously treated patients.
As for a strategy for coping with this continued outperformance of a control arm, Cheng said the power of immuno-oncology agents as later-line treatments must be taken into account in study designs. “One just has to acknowledge that and be able to predict what the improvement of standard of care might be moving forward” and plan appropriately, he said.
Cheng also suggested that measuring tumor progression might be of increasing value. In CheckMate-651, patients with PD-L1 scores of at least 20 were at similar risk of disease progression for Opdivo-Yervoy and EXTREME; in fact, the median progression-free survival for the BMS therapy was shorter at 5.4 months, versus 7 months for EXTREME. But more patients on the dual immunotherapy (26%) were alive without disease progression at the two-year mark than on EXTREME (16%).
BMS wasn’t the only one that hit a head and neck cancer setback at ESMO 2021. Pfizer and Merck KGaA’s PD-L1 inhibitor Bavencio, used alongside Erbitux-containing standard treatment, failed to show a survival advantage in locally advanced SCCHN.
In the Gortec-Reach trial, the Bavencio regimen showed a favorable—but not statistically significant—trend toward staving off tumor progression in patients who were unfit to receive the chemotherapy cisplatin. And the two-year overall survival rates were similar between the two groups at 58% for Bavencio and 54% for control. As for the cisplatin-fit cohort, progression-free survival instead favored standard of care over Bavencio.
The Erbitux combo trial flop followed Pfizer and Merck KGaA’s own Javelin Head and Neck 100 defeat, where Bavencio was paired with just chemoradiotherapy. The alliance currently has no ongoing trials exploring Bavencio in head and neck cancer, a spokesperson at Merck KGaA’s EMD Serono told Fierce Pharma.
Merck KGaA has other plans for the hard-to-treat disease. The German company recently put down €188 million upfront to in-license Debiopharm’s xevinapant, a potentially first-in-class antagonist of Inhibitor of Apoptosis Proteins. The drug is being studied in combination with chemoradiotherapy in cisplatin-eligible patients with locally advanced SCCHN in a phase 3 called TrilynX.