Skip to main content

Yervoy delivers a surprise in liver cancer

Checkmate-9DW scores a topline win on overall survival, when failure seemed more likely.

There were several reasons why Bristol Myers Squibb’s Checkmate-9DW study in first-line liver cancer wasn’t expected to succeed, but against the odds it delivered a win today. This apparent success, on overall survival and against a realistic comparator, should see Opdivo plus Yervoy head for regulatory review.

What’s still unclear is the magnitude of the benefit, and how this compares against the emerging first-line option of Roche’s Tecentriq plus Avastin; such detail will only become apparent once full data are revealed, as will the safety profile, a key consideration given Yervoy’s toxicity. But perhaps Checkmate-9DW’s most intriguing aspect is that Yervoy’s contribution to efficacy made the difference.

That seems to be the case in light of Bristol’s earlier study of Opdivo monotherapy in first-line liver cancer, Checkmate-459, which failed to show an OS benefit against Nexavar. Yervoy’s toxicity has tended to limit its efficacy, but now Bristol says the Yervoy/Opdivo combo has succeeded in Checkmate-9DW, and not on a surrogate endpoint but according to the gold standard of OS.


That’s just one of the surprising things about Checkmate-9DW. Another is that this trial succeeded even though it didn’t restrict control arm patients to Nexavar, seen by some as a soft comparator.

Instead, Checkmate-9DW’s control cohort comprised physician’s choice of Nexavar or Lenvima. The latter is seen as a somewhat tougher first-line comparator, having beaten Nexavar on PFS (though not on OS) in its registrational Reflect study. Notably, Merck & Co’s Keytruda plus Lenvima failed to beat Lenvima alone in the Leap-002 trial.

It’s also worth pointing out the delays Checkmate-9DW had suffered, something that rarely seems to portend a positive outcome in an event-driven study. The trial had at one point been expected to read out in 2022/23, but Bristol later changed that to 2023 or beyond. This was in spite of its enrolment target being cut from 1,084 to 732 patients, according to

That said, Roche’s Tecentriq plus Avastin casts a long shadow, having convincingly beaten Nexavar in the Imbrave-150 trial, and established itself as a new first-line standard after securing US approval in 2020.

AstraZeneca’s Imfinzi plus Imjudo – notably another PD-1/CTLA-4 blocker combo – also scored a first-line liver cancer approval, based on the Himalaya trial; it might be comforting that, in a finding analogous to the respective failure and success of Checkmate-459 and 9DW, Himalaya’s Imfinzi monotherapy arm didn’t beat Nexavar.

But on a cross-trial basis Himalaya came up short of Imbrave-150, and it’s the Roche result against which Checkmate-9DW data will ultimately be compared.

Selected first-line hepatocellular carcinoma trials

Checkmate-459Opdivo monotherapyFailed to beat Nexavar on mOS: 16.4mth vs 14.7mth (HR=0.85, p=0.075)
Leap-002Keytruda + LenvimaFailed to beat Lenvima on mOS: 21.2mth vs 19.0mth (HR=0.84, p=0.023)
Imbrave-150Tecentriq + AvastinmOS: 19.2mth vs 13.4mth for Nexavar (HR=0.66, p<0.001); US approved Jun 2020
HimalayaImfinzi + ImjudomOS: 16.4mth vs 13.8mth for Nexavar (HR=0.78, p=0.0035); US approved Oct 2022
Imfinzi monotherapymOS: 16.6mth vs 13.8mth for Nexavar (HR=0.86)
Emerald-1Imfinzi + TACE + AvastinmPFS 15.0mth vs 8.2mth for TACE (HR=0.77)
Imfinzi + TACEmPFS 10.0mth vs 8.2mth for TACE (HR=0.94)
Checkmate-9DWOpdivo + Yervoy“Statistically significant & clinically meaningful” OS benefit vs Nexavar or Lenvima
Emerald-3Imfinzi + Imjudo + TACE +/- LenvimaPrimary endpoint PFS vs TACE alone; ends Dec 2025
Skyscraper-14/ Imbrave-152Tiragolumab + Tecentriq + AvastinCo-primary endpoints PFS & OS vs Tecentriq + Avastin; ends Sep 2026

Source: OncologyPipeline.


Molecular Drug Targets