CTNI-73. LAT1 EXPRESSION SCALE AND ADAPTIVE DESIGN FOR A PHASE IIA TRIAL OF QBS72S IN BREAST CANCER BRAIN METASTASES

Abstract Breast cancer is the leading cause of cancer-related deaths in women and often spreads to the brain. As many as 30% of patients with breast-to-brain metastases will develop leptomeningeal disease (LMD), with poor survival, rapid neurologic decline, and a lack of effective long-term treatment options. The novel agent QBS72S (also known as QBS10072S) is designed to exploit the high expression of L-type amino acid transporter 1 (LAT1) on breast cancer cells and the blood-brain barrier. Through conjugation of an amino acid analogue to a DNA alkylating moiety, QBS72S can utilize LAT1 for targeted delivery into the brain and metastatic tumor cells. We developed a single-arm, phase IIa study to evaluate the preliminary efficacy and safety of QBS72S for breast-to-brain metastasis, focusing on two distinct cohorts: intraparenchymal metastasis (Cohort 1; without LMD) and LMD (Cohort 2; with or without intraparenchymal metastasis). The primary endpoint is the overall response rate among evaluable participants in Cohort 1, while secondary endpoints encompass progression-free survival, overall survival, duration of response, and treatment-related adverse events in Cohort 1. Exploratory endpoints include correlation of LAT1 expression in formalin-fixed paraffin-embedded samples with treatment response, for which we developed a novel immunohistochemical grading scale, and identification of novel CSF-based LMD biomarkers. While evidence of LAT1 expression is not required for enrollment, patients who are later found to have no LAT1 expression will not be considered evaluable. This adaptive clinical trial design may facilitate rapid enrollment and allow for tailored endpoints for patient cohorts with disparate baseline outcomes. Our LAT1 staining protocol will support ongoing trials in glioblastoma (NCT02977780) and future studies in additional brain-trophic metastatic cancers with high LAT1 expression to correlate drug efficacy, while our exploratory endpoints may facilitate identification of more rapid and reliable biomarkers of LMD treatment response and resistance.