Ovarian Cancer

Epidemiology of folate receptor alpha expression in ovarian cancer by age, race/ethnicity, and histotype in a real-world, US-based clinicogenomic database

2026
Rebecca C. Arend, MD, MSPH - University of Alabama at Birmingham, Birmingham, Alabama, USA; Kent F. Hoskins, MD - University of Illinois Chicago, Chicago, Illinois, USA; Jenny S. Guadamuz, PhD - University of California, Berkeley, Berkeley, California, USA; Gregory S Calip, PharmD, MPH, PhD - AbbVie, Inc., North Chicago, Illinois, USA; Megan A Clarke, PhD, MHS - AbbVie, Inc., North Chicago, Illinois, USA; Yookyung Christy Choi, PharmD, PhD - AbbVie, Inc., North Chicago, Illinois, USA; Qu Zhang, PhD - AbbVie, Inc., North Chicago, Illinois, USA; Ricardo R. Lastra, MD - The University of Chicago, Chicago, Illinois, USA; Amanda L. Strickland, MD - University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA; Sarah K. Lynam, MD - University Hospitals Seidman Cancer Center, Cleveland, Ohio, USA
Read more

Summary

Ovarian cancer (OC) is molecularly heterogeneous, requiring biomarkers to guide personalized therapy. Folate receptor alpha (FR⍺) is an actionable biomarker with therapeutic implications.

Methods

We conducted a cross-sectional analysis of patients diagnosed with ovarian, fallopian tube, or primary peritoneal cancer from 2019 to 2025 using the ConcertAI RWD360™ database linked to genomic data from Caris Life Sciences. FR⍺ testing was done using the VENTANA FOLR1 (FOLR1-2.1) RxDx Assay (Roche Diagnostics); FR⍺-high expression was defined as ≥75% of tumor cells with ≥2+ membrane staining. Stratified prevalence rates and multivariable prevalence rate ratios (RRs) with 95% CIs were estimated using modified Poisson regression. Sensitivity and exploratory analyses evaluated associations with FR⍺ expression in the full, unselected OC clinicogenomic database, applying a validated machine learning classifier trained on whole transcriptome RNA-seq to predict FR⍺-high expression (N=3045).

Results

Among the 1155 patients with OC who received FR⍺ testing, median age was 66 years (y) (IQR, 58–73); 71% were non-Hispanic White and 61% had high-grade serous histology. FR⍺-high expression was observed in 33% of patients, with higher prevalence in high-grade serous (43%) compared to less common histotypes (low-grade serous, 24%; endometrioid, 6%; clear cell, 2%; carcinosarcoma, 9%) (P< 0.001). There were no significant associations between FR⍺-high prevalence and age after adjusting for histotype (< 50y, Reference; 50–64y, adjusted RR [aRR] 0.95 [95% CI 0.69–1.31]; 65–74y, aRR 1.15 [95% CI 0.84–1.57]; ≥75y, aRR 1.18 [95% CI 0.85–1.63]). FR⍺-high prevalence did not differ by race/ethnicity (P=0.933) including after adjustment for age and histotype: non-Hispanic Black, 34% (aRR 1.03 [95% CI 0.78–1.35]); non-Hispanic Asian, 29% (aRR 0.97 [95% CI 0.57–1.65]); Hispanic, 30% (aRR 0.95 [95% CI 0.64–1.40]); non-Hispanic White (32%, Reference). In sensitivity analyses restricted to high-grade serous histology, consistent associations with age and race/ethnicity were observed. In exploratory analyses, similar associations were observed in the broader database.

Conclusions

FR⍺-high expression does not significantly differ by age or racial/ethnic group and is most prevalent in high-grade serous OC. These findings further support FR⍺ as an OC biomarker with broad clinical applicability across demographics and demonstrate the importance of accounting for potential confounding by histotype in studies of FR⍺-high prevalence.