Roche annuncia lo stop dello studio di fase II/III con tiragolumab (TIGIT) + atezolizumab (PD-L1) nel trattamento del tumore al polmone per inferiorità rispetto al braccio di controllo

Roche announced that P2/3  SKYSCRAPER-06: tiragolumab (TIGIT) + atezolizumab (PD-L1) + chemotherapy vs pembrolizumab (PD-1) + chemotherapy in 1L nsq mNSCLC did not meet 1EP of PFS at primary analysis and 1EP of OS at the first interim analysis.

Based on these results, the study will be discontinued.

• The failure of SKY-06 reinforces the challenge 1L regimens face to surpass the benchmark set by pembro + chemo in 1L NSCLC

  • Although SKY-06 and STAR-121 have similar trial designs, they are differentiated by patient selection (nsq vs. nsq & sq) and sample size (N=542 vs. N=720)

• Tira + atezo + chemo showed detrimental efficacy compared to control arm in the intent-to-treat population, demonstrating PFS HR=1.27 and OS HR=1.33 (immature data set)

  • The overall safety profile remains consistent with the safety profile previously observed for the combination, and no new or unexpected findings were identified

• The reduced efficacy of Roche’s regimen may be a result of the underperformance of atezo (anti-PD-L1) relative to pembro (anti-PD-1) in which case tira’s potential activity was insufficient to offset atezo’s underperformance

  • Although cross-trial comparison is cautioned, atezo’s underperformance is noted in previous 1L NSCLC studies with the same chemo backbone as SKY-06 (pemetrexed + carboplatin or cisplatin). Pembro + chemo has demonstrated an mOS of 22.0 mos (KEYNOTE-189) while atezo + chemo has demonstrated an mOS of 18.2 mos (IMPOWER-132) and did not meet statistical significance (Please see table 2 below)

  • Atezo’s underperformance also casts doubts on the highly anticipated final analysis results for SKYCRAPER-01 (tira + atezo in 1L NSCLC with high PD-L1 expression) which utilizes atezo as opposed to pembro in the control arm has generated mixed results – missed PFS at IA1; leaked IA2 data demonstrated numerically improved OS (HR=0.81)

• SKY-06’s negative result represents another setback for the anti-TIGIT class as a whole but data from Merck’s P3 KEYVIBE-007 evaluating vibostolimab/pembrolizumab (TIGIT/PD-1) coformulation + chemo vs pembro + chemo in 1L PD-L1+ mNSCLC (STAR-121’s highest threat competitor) is highly anticipated in 2025 and will elucidate the therapeutic potential of the TIGIT/PD-1 regimen in 1L NSCLC.

• A phase 3 trial of MSD’s anti-TIGIT antibody vibostolimab and PD-1 inhibitor pembrolizumab (KeyVibe-010 study was looking at a co-formulation of experimental vibostolimab and pembrolizumab as an adjuvant treatment for patients with high-risk melanoma treated with surgery) has been abandoned after a high rate of discontinuations mainly due to “immune-mediated” side effects rendered the study futile.