Abstract 3267: Identification of a novel biomarker response in a prospective clinical trial of immune checkpoint blockade in recurrent ovarian carcinoma

Abstract

Abstract The role of immune checkpoint inhibitors (ICPIs) in ovarian cancer continues to be refined. The presence of intratumoral T cells in ovarian tumors correlates with improved survival and progression, suggesting that immune modulation, such as by ICPIs, may be beneficial. However, a better understanding of the immune effects following ICPI use, particularly in those with a long progression free survival, is needed. Thus, we used imaging mass cytometry (IMC) to compare the tumor immune landscape of patients with ovarian cancer before and after a CTLA4 immune checkpoint inhibitor treatment to determine the association between alterations in immune landscapes and progression free survival. Fine needle biopsies were obtained from 8 patients with recurrent platinum-resistant ovarian carcinoma enrolled in a phase 2 randomized trial evaluating the efficacy of Tremelimumab and Durvalumab (CTLA4 and PDL1 checkpoint inhibitors, respectively) in recurrent ovarian cancer treatment. Each patient had a pre-treatment biopsy and on-treatment biopsy after 3 cycles of CTLA4 inhibitor therapy. Progression free time (PFS) was recorded with 3 patients with a long PFS (>180 days) and 5 with short PFS (<60 days). Formalin fixed paraffin embedded (FFPE) tissue sections were stained for IMC analysis via Fluidigm protocol with 34 metal-tagged antibodies to detect various cell and immune related markers. The IMC data was obtained using the Fluidigm Helios CyTOF instrument and Hyperion Imaging System laser ablation module. A novel image informatics pipeline through Matlab was used to assess the image intensity of each marker and cell location. Using the developed image informatics pipeline, cell types and locations were analyzed for 16 samples from 8 patients. In all patients, mean CD8+ T cell densities had an increased trend in on-treatment vs pre-treatment samples (205.6 v 129.9 cells/mm2, p=0.086). An analysis of cell location found that in all patients on-treatment, the mean number of CD8+ T cells adjacent to M2 macrophages increased significantly (0.09 v. 0.17 cells, p = 0.0460) as well as the mean number of CD8+ T cells adjacent to B7H4+ tumor cells (0.035 v. 0.01 cells, p = 0.046). Furthermore, patients with a long PFS had a significantly higher number of CD8+ cells neighboring B cells on-treatment than short PFS (0.12 v. 0.27 cells, p=0.028). Using IMC and novel image informatics pipeline, we found that patients with recurrent ovarian cancer treated with Tremelimumab with a long PFS mounted a significant immune response compared to patients with a short PFS, with a higher number of CD8+ T cells neighboring B cells after treatment. For all patients, the immune landscape altered after treatment. To our knowledge, this is the first known use of IMC to demonstrate immune response in ovarian cancer following ICPI therapy. Overall, IMC and image informatics pipeline are robust tools that simultaneously analyze multiple biomarkers and spatial location of cells to better assess the tumor immune microenvironment and cellular interactions. Citation Format: Han Cun, Emily M. Hinchcliff, Ying Zhu, Sammy Ferri-Borgogno, Rita Cheng, Jared K. Burks, Stephen T. Wong, Amir A. Jazaeri, Samuel C. Mok. Identification of a novel biomarker response in a prospective clinical trial of immune checkpoint blockade in recurrent ovarian carcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3267.