Assessment of predictive biomarker prevalence in molecularly defined adult-type ovarian granulosa cell tumors.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 5567-5567
Author(s):  
Robert Tyler Hillman ◽  
Douglas I. Lin ◽  
David Marc Gershenson

5567 Background: The FDA has separately approved pembrolizumab for all advanced solid tumors with either microsatellite instability or high tumor mutational burden (≥10 mutations per megabase), and homologous recombination deficiency is an established biomarker for response to poly ADP ribose polymerase inhibitors in epithelial ovarian cancer. The detection of these predictive biomarkers in adult-type ovarian granulosa cell tumors could identify novel treatment strategies in this rare disease. The primary objective of this study was to determine the prevalence of established predictive biomarkers among molecularly defined adult-type ovarian granulosa cell tumors. Methods: With institutional review board approval, we performed a cross-sectional study examining de-identified FoundationOne companion diagnostic molecular profiles for 423 women with molecularly defined (FOXL2 c.C402G positive) adult type ovarian granulosa cell tumors. The dataset was comprised of coding variants for up to 406 genes as well as genomic signatures including microsatellite instability, tumor mutational burden, and genome wide loss of heterozygosity. PD-L1 expression by immunohistochemistry was also available for a subset of tumors. Descriptive statistics were used for comparison between groups and all statistical tests were two-sided. Results: Women in this cohort had a mean age of 57 years (range 24-87) at the time of sample submission for molecular profiling. The median tumor mutational burden was 1.3 mutations per megabase [mut/Mb] (range 0-8.8 mut/Mb). TP53-mutated aGCT had a higher tumor mutation burden than TP53 non-mutated tumors (median 2.4 mut/Mb, 95% CI 1.7-3.0 mut/Mb vs median 1.3 mut/Mb, 95% CI 1.5-1.9 mut/Mb; P=.02). All 384 tumors with available microsatellite instability testing were microsatellite stable. Sixty-seven tumors had PD-L1 expression measured and of these 94% (63/67) were negative with the remainder “low positive.” No tumors were positive for genome wide loss of heterozygosity. Apart from FOXL2 c.C402G, the most frequent short variants were TERT promoter mutations (-124C>T: 190/423, 45.0%; -146C>T: 39/423, 9.2%). Other frequently observed variants included truncating mutations in KMT2D/MLL2 (71/423, 16.8%), pathogenic TP53 mutations (35/423, 8.3%), CDKN2A/B deletions (43/423, 10.2%), and activating PIK3CA mutations (23/423, 5.4%). Conclusions: No women with molecularly defined adult-type ovarian granulosa cell tumors in this large cross-sectional study would be eligible for FDA-approved pembrolizumab based on either microsatellite instability or high tumor mutational burden. No tumors exhibited evidence of homologous recombination deficiency and molecularly targetable mutations were rare. The development of novel precision treatment options remains a critical unmet need for this rare disease.

2016 ◽  
Vol 143 (3) ◽  
pp. 571-577 ◽  
Author(s):  
Saara Bryk ◽  
Anniina Färkkilä ◽  
Ralf Bützow ◽  
Arto Leminen ◽  
Johanna Tapper ◽  
...  

2012 ◽  
Vol 124 (2) ◽  
pp. 244-249 ◽  
Author(s):  
Hsu-Dong Sun ◽  
Hao Lin ◽  
Mei-Shan Jao ◽  
Kung-Liahng Wang ◽  
Wen-Shiung Liou ◽  
...  

2018 ◽  
Vol 36 (5_suppl) ◽  
pp. 157-157 ◽  
Author(s):  
Donna Nichol ◽  
Siân Jones ◽  
Samuel V. Angiouli ◽  
Laurel Keefer ◽  
Monica Nesselbush ◽  
...  

157 Background: Checkpoint inhibitors (CPIs) have been approved for frontline or subsequent therapies in several indications over the last few years. While patient response can be remarkably durable, many patients do not benefit. Current clinical biomarkers of response to CPIs include microsatellite instability (MSI) and PD-L1 expression. While a proportion of many solid tumors display microsatellite instability, the prevalence is often very low. Similarly, while clinically informative, PD-L1 expression alone is not sufficient to predict therapeutic outcomes with high accuracy. The lack of predictive biomarkers for response highlights the need for improved biomarkers with greater prevalence across tumor types to predict response to CPIs. Multiple clinical studies have revealed that high tumor mutational burden (TMB) is associated with improved clinical response. Methods: Here, we describe the development of a method that can be used to accurately infer mutational burden from a discrete set of targeted regions of interest across the exome. Initially, we performed an assessment of the accuracy across multiple bioinformatics methods for identification of individual sequence mutations (SBS/indels) using orthogonally validated data together with publicly available TCGA whole-exome sequencing data. The targeted regions were then isolated from these datasets to demonstrate analytical performance across several different solid tumor types. Finally, we evaluated independent non-small cell lung cancer (NSCLC) and colorectal carcinoma (CRC) cohorts to demonstrate the analytical accuracy of the assay and bioinformatics approach for determination of mutational burden when compared to whole exome sequencing. Results: In summary, high concordance was observed across a large dynamic range of mutations per megabase of coding sequence. Conclusions: Our data indicate that the assay can be used to accurately determine mutational burden in a range of tumor types, across a spectra of potential mutational burden cut-offs using automated, complex mutation identification algorithms.


2006 ◽  
Vol 101 (1) ◽  
pp. 18-23 ◽  
Author(s):  
Sebastian Leibl ◽  
Koppany Bodo ◽  
Margit Gogg-Kammerer ◽  
Andelko Hrzenjak ◽  
Edgar Petru ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document