TIGIT blows up in a trial that always looked risky
Keyvibe-008 is halted for futility, an outcome that might have been foreseen in 2022.
Keyvibe-008 is halted for futility, an outcome that might have been foreseen in 2022.
Today’s failure of Merck & Co’s Keyvibe-008 study, another setback for TIGIT blockade, raises the question of why this trial was even begun. Just days before Keyvibe-008 was due to commence dosing in March 2022 Roche’s corresponding Skyscraper-02 study blew up in a virtually identical SCLC setting and with the same comparator, but despite this Merck pressed ahead.
Both trials tested whether adding TIGIT blockade (tiragolumab in Skyscraper-02 and vibostolimab in Keyvibe-008) to PD-(L)1 blockade plus chemo could beat Tecentriq plus chemo in first-line SCLC. Skyscraper-02 failed on progression-free and overall survival, and today Merck stopped Keyvibe-008 after interim analysis showed OS futility, with increased adverse events to boot.
Merck might have foreseen such a result in early 2022, and not only because both trials were testing the same mechanistic triplet approach. Crucially, Keytruda plus chemo had itself already failed to beat chemo in first-line SCLC, in the phase 3 Keynote-604 study; conversely, Tecentriq plus chemo did manage a win in Impower-133, a trial on whose basis it was approved.
Therefore, in Keyvibe-008 Merck was trying to show that Tigit, a mechanism that failed in this setting in Skyscraper-02, added to Keytruda, a drug that earlier failed in this setting in Keynote-604, could beat Tecentriq plus chemo – a standard of care.
Given such a precedent the only way Keyvibe-008 might have succeeded was if vibostolimab had meaningfully better efficacy than tiragolumab. Since the Skyscraper-02 setback vibostolimab has also failed in second-line NSCLC (Keyvibe-002) and adjuvant melanoma (Keyvibe-010), and Merck and Roche both await data from front-line NSCLC trials, Keyvibe-003 and Skyscraper-01 respectively.
Another tox signal
While Roche’s TIGIT setbacks have largely been down to lack of efficacy, investors will note an unwelcome similarity between the failures of Keyvibe-010 and 008: in both trials vibostolimab was said to have caused an increase in immune system-mediated adverse events, adding the threat of toxicity to the mix.
A full analysis of both trials will only be possible once they are presented at a scientific congress.
For now Merck says it remains committed to SCLC through other projects. For instance, the pivotal Ideate-Lung02 study of the Daiichi Sankyo-derived ifinatamab deruxtecan has started dosing, while MK-6070, an anti-DLL3 T-cell engager acquired through Harpoon, was yesterday licensed to Daiichi with the aim of testing it in SCLC in combination with ifinatamab-dxd.
As far as TIGIT goes, in addition to the Keyvibe-003 and Skyscraper-01 readouts, GSK and iTeos will reveal full data from the Galaxies Lung-201 study of belrestotug, possibly at ESMO. Tiragolumab has shown promise in liver and oesophageal cancers, but beyond these glimmers of hope time is running out.
Selected studies in 1st-line SCLC
Trial | Active | Comparator | Outcome |
---|---|---|---|
Impower-133 | Tecentriq + chemo | Chemo | Succeeded OS (medians 12.3mth vs 10.3mth) & PFS |
Keynote-604 | Keytruda + chemo | Chemo | Failed for OS (medians 10.8mth vs 9.7mth), succeeded for PFS |
Skyscraper-02 | Tecentriq + chemo + tiragolumab | Tecentriq + chemo | Failed for OS (medians 13.6mth vs 13.6mth) & PFS |
Keyvibe-008 | Keytruda + chemo + vibostolimab | Tecentriq + chemo | Failed for OS, with increase in AEs |
Source: OncologyPipeline.
3134