Abstract

5561 Background: Targeted therapy in folate receptor alpha (FOLR1)-positive high grade serous ovarian carcinoma (HG) is now a mainstay for platinum-resistant disease, though the rate of FOLR1-positivity in low grade serous ovarian carcinoma (LG) is unknown. We compared the genomic and transcriptomic landscapes in FOLR1-positive/negative LG in comparison to its HG counterpart. Methods: LG (N = 281) and HG (N = 5086) tumors were tested at Caris Life Sciences (Phoenix, AZ) with NextGen Sequencing on DNA (592 genes or whole exome) and RNA (whole transcriptome). PD-L1+ (22C3, TPS > 1%) and FOLR1 (Positive [F+], ⩾ 2+, ⩾75%) expression was assessed by IHC. Mutations were defined as pathogenic SNVs/indels (-Mt). Transcriptomic signatures associated with response to immunotherapy (T cell-inflamed) or with MAPK pathway activation (MPAS) were applied. Fisher’s exact/χ 2 and Mann-Whitney U tests were applied as appropriate ( p < .05, adjusted for multiple comparisons). Real-world overall survival (OS) was obtained from insurance claims and Kaplan-Meier estimates were calculated for molecularly defined patients. This study was reviewed by the Johns Hopkins Medicine IRB and determined to qualify as exempt human subjects research. Results: HG tumors had a higher prevalence of F+ tumors (43.5%) as compared to LG (24.6%). HG tumors had a higher prevalence of TP53-Mt compared to LG (LG F+: 11.1%, LG F-: 4.1, HG F+: 97.1, HG F-: 95.6, p < .001). Conversely, KRAS-Mt and NRAS-Mt were enriched in LG tumors (KRAS [LG F+: 22.2%, LG F-: 21.6, HG F+: 0.4, HG F-: 3.3, p < .001], NRAS [LG F+: 0%, LG F-: 10.2, HG F+: 0.2, HG F-: 0.1, p < .001]). BRAF-Mt were also almost exclusively present in LG tumors although more prevalent in LG F- as compared to LG F+ (LG F+: 7.4%, LG F-: 14.3, HG F+: 0.2, HG F-: 0.3, p < .001). There was a higher prevalence of PDL1+ tumors in HG vs LG (LG F+: 39.3%, LG F-: 35.9, HG F+: 70.0, HG F-: 72.4, p < .001). LG tumors had a significantly higher MPAS than HG tumors (LG F+: 1.38 arbitrary units, LG F-: 1.66, HG F+: -0.35, HG F-: -0.25, p < .001). LG F- and HG F- tumors had a similar proportion of T cell-inflamed tumors (LG F-: 36 vs HG F-: 33, p > .05), whereas LG F+ tumors have a significantly lower proportion of T cell-inflamed tumors as compared to HG F+ (LG F+: 18% vs HG F+: 42, p < .001). Amongst tumors that had not received mirvetuximab soravtansine, no difference in OS was observed between HG F- v HG F+ (HR 1.3, p = .072; median OS HG F-: 87 months [N = 1092], HG F+: 98 [N = 756]). No difference in OS was observed when comparing LG F- vs LG F+ (HR 1.0 , p = .93; median OS LG F-: 98 months [N = 116], LG F+: not reached [N = 37]). Conclusions: A notable portion of LG tumors were FOLR1+, which suggests that FOLR1 expression in LG could be a viable target for this rare histology, particularly in the recurrent setting. MAPK activation was significantly higher in LG tumors when compared to HG, yet no difference between LG F+ and F- tumors was observed.