Objective: We sought to determine whether functional precision medicine (FPM), which combines functional testing of live cancer cells with genomic profiling could utilize patient derived organoids (PDOs) from low-grade serous ovarian carcinomas (LGSOC) to select effective therapies.

Methods: Fresh tumor tissue was obtained from 16 patients undergoing surgery or image-guided biopsy for LGSOC, 11 at recurrence and 5 at primary surgery. Short-term organoids were isolated and evaluated using the SEngine Precision Medicine PARIS test. Tumor origin and driver mutations were confirmed using whole exome sequencing.

Results: Biopsies were obtained from 16 patients with LGSOC with median time to CLIA FPM
test results of 16 days (range 13-45 days) and 100% success rates in generating PDOs. On targeted sequencing, 5 (31.3%) contained potentially targetable gene alterations including pathogenic variants in KRAS in 4 cases and ESR1 in 1 case. Functional drug profiling of the LGSOC PDOs revealed 4/16 (25%) with good response to chemotherapy: gemcitabine in 4 and cisplatin in one. In contrast, significantly more cases demonstrated good or exceptional responses to two or more targeted therapies (15/16, 93.8%, P=0.0002), including most commonly inhibitors of EGFR/HER2, BTK, MEK, and mTOR pathways. In the recurrent setting, of the 6 patients treated with FPM-guided therapy, two had exceptional (defined as 12 months or greater) responses and median PFS was 10.0 months (range 6-16 months). In contrast, of 6 cases treated with non-guided therapy, there was one exceptional response, and the two longest responses were to letrozole, which is now standard of care for frontline therapy of LGSOC. For those receiving non-guided treatment, mean PFS was 6.6 months (range 3-19 months). The two drugs most frequently identified in the FPM profiling as top drugs were lapatinib and ibrutinib in 8 patients each (50%). One patient who initiated therapy with a partial small bowel obstruction had a clinical and biochemical response to ibrutinib lasting 14 months with normalization of GI function. Another patient with LGSOC without HER2 amplification had a partial response on lapatinib for 13 months, followed by lapatinib with low dose trametinib for an additional 1.5 years (ongoing). Most top targeted drug sensitivities were not associated with an identifiable actionable genomic alteration. Two paired LGSOC samples were collected at the
same timepoint from different metastatic sites, with similar response profiles. One patient had two biopsies 22 months apart, with notably different response profiles, likely reflecting tumor evolution during targeted therapy.

Conclusion: FPM using PDOs from LGSOC can identify therapeutic sensitivities to unique targeted agents that would not be identified by standard molecular profiling. We propose a prospective trial of PDO-directed therapy for patients who have recurrent LGSOC that has failed standard therapy.

VIEW THE POSTER