Rapid analysis and response evaluation of combination anti-neoplastic agents in rare tumors (RARE CANCER) trial (RARE 3): Tiragolumab + Atezolizumab

Background: Rare tumors (incidence <6 per 100,000 per year in the US) have limited therapeutic options. Combined immune checkpoint inhibition (ICI) with anti-PD-L1 (atezolizumab) and anti-TIGIT (tiragolumab) demonstrated activity in early clinical trials and enhanced CD8⁺ T-cell-mediated tumor rejection in preclinical models. This trial will assess activated intratumoral CD8⁺ T cells and clinical activity of this combination in rare tumors. Methods: RARE-3 (NCT05715281) is a phase II pharmacodynamic (PD) trial with a primary objective of assessing the increase in proportion of activated CD8⁺ T cells in tumor biopsies at baseline and prior to cycle 3, day 1 (±3 days). Secondary objectives included overall response rate (RECIST v1.1), progression-free survival (PFS), and the proportion of patients (pts) with a clinically meaningful increase in CD8⁺ T-cell infiltration. Eligible pts had histologically confirmed tumors, age ≥18 years, ECOG ≤2 and adequate organ function. Prior ICI therapy (excluding anti-TIGIT) was permitted. Pts received atezolizumab (1200mg IV) plus tiragolumab (600mg IV) every 21 days until progression or unacceptable toxicity. Ten evaluable paired biopsies would provide 95% probability and 88% power (alpha=0.05) to detect a >1.25 standard deviation increase from baseline. Results: Twelve pts were enrolled from 9/26/23 to 8/18/25. Median age was 60.5 years (range 19-71); pt population was 75% female, 83.3% White, and median number of prior systemic therapies was 3 (range 0-9). As of 11/3/25, the highest treatment-emergent adverse events (AEs), reported in 9 pts, were grade 3. Treatment-related AEs included lymphopenia, Guillain-Barré syndrome, pruritus, maculopapular rash, myositis, and diarrhea. Nine pts were assessed for response including stable disease (SD) in 6 pts (50%) (3 with SD ≥6 cycles; including adrenocortical carcinoma, chromophobe renal carcinoma, perivascular epithelioid cell tumor), and progressive disease in 3 pts. One pt discontinued treatment for clinical progression; 2 were not evaluable due to not starting treatment and consent withdrawal after 1 cycle. Median PFS was 4.29 months (interquartile range, 2.29-8.97). Eight paired biopsies are currently being assessed. Conclusion: The study closed early due to sponsor decision. There was no unexpected toxicity. Ongoing analyses are assessing the impact of treatment on intratumoral CD8⁺ T cells and other PD markers, and their association with clinical activity to inform future studies. Funded by NCI Contract No. HHSN261201500003I.