T cell receptor and immune gene expression pharmacodynamics for durvalumab monotherapy and in combination with tremelimumab or bevacizumab in unresectable hepatocellular carcinoma (uHCC).

person Robin Kate Kelley Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA info_outline Robin Kate Kelley, Young S Lee, James Conway, John F Kurland, Alejandra Negro, Patricia McCoon

Authors person Robin Kate Kelley Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA info_outline Robin Kate Kelley, Young S Lee, James Conway, John F Kurland, Alejandra Negro, Patricia McCoon Organizations Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA, Oncology R&D, AstraZeneca, Gaithersburg, MD, Oncology R&D, AstraZeneca, Waltham, MA Abstract Disclosures Research Funding AstraZeneca Background: Study 22 (NCT02519348), a Phase 2 trial of immune checkpoint inhibitor (ICI) monotherapy and combination regimens in uHCC, showed higher rates of objective response (ORR) with STRIDE (Single Tremelimumab [T] Regular Interval Durvalumab [D]) or D plus bevacizumab (B) than with D (1,2). Pharmacodynamic analyses showed that improved efficacy with STRIDE versus D was associated with increased expansion of T cell clones at the end of cycle 1 (C1) versus D (3). We performed further T cell clone analyses to include D+B and compared gene expression signatures (GES) associated with each treatment arm to explore commonalities and differences in mechanistic biomarkers underpinning the efficacies associated with STRIDE and D+B. Methods: Participants (pts) with uHCC and no prior ICI therapy received STRIDE (T 300 mg + D 1500 mg [1 dose during C1] then D 1500 mg once every 4 weeks [Q4W]), T75+D (T 75 mg + D 1500 mg Q4W [4 doses] then D 1500 mg Q4W, D (1500 mg Q4W), T (750 mg Q4W [7 doses] then Q12W), or D+B (D 1120 mg + B 15 mg/kg Q3W). DNA and RNA were isolated from PAXgene-preserved whole blood collected at baseline and at the end of C1. CDR3 sequencing of T cell receptor β using the immunoSEQ Assay (Adaptive Biotechnologies, Seattle, WA) and RNA sequencing (Q Squared Solutions, Morrisville, NC) were performed. ImmunoSeq results and immunologic gene expression changes were reported with ORR across treatment arms. Results: ImmunoSeq did not show significant differences in baseline T cell clonality across treatment arms. In contrast to STRIDE versus D, the number of expanded T cell clones at the end of C1 for D+B versus D was unchanged, and no association between expanded T cell clones and ORR was observed (Table). While increases in Type I and II interferon and mitotic GES were seen for both STRIDE and D+B, more changes in immune-related peripheral GES, with greater magnitude of change, were associated with STRIDE. Significant changes in CD8 T cell, M1 macrophage, and Treg GES were only observed for STRIDE. Conclusions: While STRIDE and D+B both had higher ORRs than D monotherapy in uHCC, only STRIDE was associated with an increased expansion of T cell clones versus D. This, in addition to the differences in magnitude and type of GES changes, suggest that STRIDE and D+B have distinct and potentially complementary mechanisms of action. 1. Kelley, et al. J Clin Oncol 2021. 2. Lim, et al. J Clin Oncol 2022, A436. 3. McCoon, et al. J Clin Oncol 2021, A4087. Clinical trial information: NCT02519348. D * (n=104) D+B † (n=47) STRIDE * (n=75) T75+D * (n=84) T * (n=69) Pts with ImmunoSeq paired samples, n 31 37 28 26 17 Pts with RNASeq paired samples, n 35 37 38 28 21 Median expanded T cell clones at end of C1, n (25 th , 75 th percentile) 32 (13, 49) 18 (5, 37) 56 (33, 148) 36 (15, 77) 100 (50, 160) ORR, % 10.6 21.3 24.0 9.5 7.2 * Pts who had progressed on, were intolerant to, or refused sorafenib; † Pts who had not received prior systemic therapy.

Clinical