Introduction Immune checkpoint inhibitors are active in advanced cervical cancer. SKYSCRAPER-04 (NCT04300647) evaluated dual blockade with tiragolumab (anti-TIGIT) and atezolizumab (anti-PD-L1) (tira+atezo), an approach hypothesized to overcome immune suppression and restore immune response.

Methods Eligible patients had measurable (per investigator assessment) PD-L1-positive recurrent/persistent cervical cancer after 1–2 prior chemotherapy lines (including ≥1 platinum-based regimen). Patients were randomized 3:1 to atezolizumab 1200 mg with or without tiragolumab 600 mg q3w until unacceptable toxicity/progression. Crossover to tira+atezo was permitted after unequivocal progression during single-agent atezolizumab. Stratification factors were ECOG PS, prior (chemo)radiotherapy, and disease status. The primary endpoint was independent review committee (IRC)-assessed confirmed objective response rate (ORR) per RECIST v1.1 in all treated patients randomized to tira+atezo. An ORR ≥21% (1-sample z-test p≤0.0245) was required to demonstrate statistically significant improvement versus a 14.6% historical reference [Chung, 2019]. Secondary endpoints included IRC-assessed progression-free survival, overall survival, and pre-crossover safety.

Results Prior therapy in 171 treated patients included bevacizumab in 35%, (chemo)radiotherapy in 80%, and paclitaxel in 93%. IRC-assessed ORRs were 19.0% with tira+atezo and 15.6% with atezo alone (table 1). In post hoc exploratory analyses of patients with measurable disease per IRC assessment, ORRs were 21.6% (tira+atezo) and 15.8% (atezo). There were no new safety signals. In a post hoc follow-up analysis, 15% of patients remained on treatment and 15/45 initially randomized to atezo had crossed over to tira+atezo.

Conclusion/Implications The ORR with tira+atezo was numerically but not significantly higher than the historical benchmark. This is the first reported phase 2 cervical cancer trial targeting TIGIT and PD-L1 concurrently.