FOLR1 as a potential novel therapeutic target in low-grade serous carcinoma of the ovary or peritoneum
Yuen Yee Leung1,8, Marta Llaurado-Fernandez1,8, Anna Cameron2, Annalyn Da Anoy3, Chanel Ghesquiere1, Joshua Hoenisch1, Stephanie Gaillard4, Josie Schmid5, Amy Dawson1, Madison Bittner1, Anna V. Tinker6, Hannah Kim1,8, Nelson Wong7,8, Gurdial Dhillon1, Mark S. Carey1,8, Martin Köbel3.
1Department of Obstetrics and Gynecology, University of British Columbia, Vancouver, BC, Canada; 2Department of Obstetrics and Gynecology, University of Calgary, Calgary, AB, Canada; 3Department of Pathology, University of Calgary, Calgary, AB, Canada; 4Division of Medical Oncology, John Hopkins Cancer Center, Baltimore, MD, United States; 5Prince George Surgery Centre, University Hospital of Northern BC, Prince George, BC, Canada; 6Department of Medical Oncology, BC Cancer , Vancouver, BC, Canada; 7Department of Experimental Therapeutics, BC Cancer Research Institute, Vancouver, BC, Canada; 8Department of Clinical Research, BC Cancer, Vancouver , BC, Canada
Objective(s): Mirvetuximab Soravtansine (MIRV) is a novel antibody drug conjugate targeting folate receptor alpha (FOLR1). MIRV has recently shown promising efficacy in chemoresistant tubo-ovarian high-grade serous carcinomas (HGSC) with high FOLR1 expression (32% ORR; SORAYA Trial). Thus, we sought to evaluate FOLR1 expression in other chemoresistant ovarian cancers to assess the feasibility of conducting future clinical trials in less common ovarian cancer types using MIRV.
Study methods: 700 ovarian cancers were examined using 6 independent tissue microarrays (3 from Calgary, 2 from Vancouver and 1 from the COEUR repository). A total of 132 (19%) HGSC, 169 (24%) endometrioid (EC), 171 (24%) low-grade serous (LGSC), 156 (22%) clear cell (CCC) and 72 (10%) mucinous carcinomas (MC) were included. FOLR1 expression was evaluated by immunohistochemistry and scored using the PS2+ system. Clinical and molecular variables were also evaluated. Appropriate statistical analyses was performed, and a p-value of 0.05 was used to establish statistical significance.
Results: High FOLR1 expression, defined as staining in ≥75% of tumor cells with at least moderate intensity, was detected in 44% HGSC, 36% LGSC, 10% CCC, 6% EC and 0% of MC tumors. Among 121 LGSCs with a more comprehensive annotation, the median age was 57 years, 89% were advanced stage (II-IV), 83% expressed high (Allred = 8) ER and 34% had suboptimal (> 1 cm) residual-disease after primary surgery. High FOLR1 levels were associated with a lower proportion of MAPK-mutations (4/33 cases; P = 0.004) and low PR expression (Allred < 2) (49/66 cases; P < 0.025), but not with stage, residuum, or ER expression. Multivariate Cox regression in advanced stage LGSOCs, showed shorter overall survival in patients with residual disease (P = 0.01), abnormal p16 expression (P = 0.002), but not with FOLR1 expression or MAPK mutation status.
Conclusions: Over a third of LGSCs expressed high FOLR1 levels. High FOLR1 expression is associated with a low frequency of MAPK-mutations and low PR expression and may be more effectively targeted with MIRV than either MEKi or antihormone therapy. These findings support the development of a clinical trial to evaluate MIRV in patients with recurrent LGSC.
BC Cancer Foundation. Janet D. Cottrelle Foundation. Gynecologic Cancer Initiative of British Columbia. Ovarian Cancer Canada and Canada Research Society. Vancouver General and UBC Hospital Foundation. MacKenzie, Lawler and Ho families. Society of Gynecologic Oncology of Canada and the Community of Practice Initiative. Canadian Ovarian Experimental Unified Resource (COEUR).