Pathogenic germline variants in SMARCA4 and further cancer predisposition genes in early onset ovarian cancer

To assess the role of germline pathogenic variants (PVs) in SMARCA4 and further established ovarian cancer (OC) predisposition genes in early onset OC, we investigated a clinical cohort of 206 unrelated OC index patients with an age at diagnosis of OC ≤40 years using an extended panel of 24 (candidate) cancer predisposition genes. PVs in established OC predisposition genes were most frequent in patients with high grade serous OC (21/62, 33.9%), comparatively rare in patients with epithelial OC other than high grade serous (5/74, 6.8%) or borderline ovarian tumours (2/39, 5.1%) and absent in mucinous OC (0/27). We demonstrate that germline PVs in SMARCA4 unlikely predispose for early onset OC other than SCCOHT.

SMARCA4, a gene associated with the small cell carcinoma of the ovary, hypercalcemic type (SCCOHT), appears to be the sole ovarian cancer (OC) predisposition gene associated with a very early age at diagnosis (≤40 years). To assess the role of germline pathogenic variants (PVs) in SMARCA4 and further established OC predisposition genes in early onset OC, we investigated a clinical cohort of 206 unrelated index patients using an extended panel of 25 (candidate) cancer predisposition genes. PVs in established OC predisposition genes were most frequent in patients with high grade serous OC (21/62, 33.9%), rare in patients with other epithelial OC (5/74, 6.8%) or borderline ovarian tumors (2/39, 5.1%). Their absence in mucinous OC (0/27) supports the notion that this entity represents a tumor phenotype not associated with PVs in established OC predisposition genes. PVs in SMARCA4 were restricted to SCCOHT and unlikely predispose for early onset OC other than SCCOHT.
According to the recommendations of the German Consortium for Hereditary Breast and Ovarian Cancer (GC-HBOC), all women diagnosed with epithelial ovarian cancer (OC) before the age of 80 years should have germline genetic testing of BRCA1/2 and other OC predisposition genes, irrespective of their tumor histology or family cancer history. 1 The American Society of Clinical Oncology (ASCO) recommends genetic testing for all women diagnosed with epithelial OC, irrespective of their age at diagnosis (AAD), tumor histology, or family cancer history. It was suggested by the ASCO that multigene panel testing should cover, at minimum, ten selected OC predisposition genes (BRCA1, BRCA2, BRIP1, MLH1, MSH2, MSH6, PMS2, PALB2, RAD51C, and RAD51D). 2 The probability to detect germline pathogenic variants (PVs) in the ten selected OC predisposition genes is largely dependent on the family cancer history, tumor histology, and AAD. Two large clinical cohort investigations revealed an overall mean AAD of OC of approximately 60 years, while patients with PVs in BRCA1 and the mismatch repair genes MSH2 and MSH6 demonstrate a statistically significant earlier mean AAD than OC patients without. 3,4 Germline PVs in the selected OC predisposition genes other than BRCA1, however, were rare in OC patients with an early AAD of OC ≤40 years. 4 In addition to the ten selected OC predisposition genes, germline PVs causing a loss of SMARCA4 gene function predispose for the rare small cell carcinoma of the ovary, hypercalcemic type (SCCOHT), primarily affecting females under 40 years with an average age of onset of 24-30 years. [5][6][7] Based on the published literature, SMARCA4 appears to be the sole OC predisposition gene associated with an AAD ≤40 years. Germline PVs in OC predisposition genes rarely associate with one tumor phenotype only. For example, germline PVs in BRCA1 associate primarily with high grade serous OC tumor phenotype, but also with high grade endometrioid and clear cell OC, though with a lower PV prevalence. 4 Thus, our study has two objectives. First, we aimed to assess the potential role of germline PVs in SMARCA4 along with an extended panel of 24 additional (candidate) cancer predisposition genes in a clinical cohort of 206 OC patients (including three SCCOHT index patients) with an AAD ≤40 years. Second, we aimed to characterize families with SMARCA4 PVs to gain more insights into SMARCA4-dependent cancer predisposition and progression.
Key characteristics of the overall study sample according to the tumor classification of the WHO (https:// tumou rclas sific ation.iarc.who.int) are given in Table 1.  SCCOHT and germline PVs present at a significantly younger age than those without. 14 In line with this observation, we described here three SMARCA4 germline PV carriers out of two families (patient #1, patient #2 and her sister) showing an AAD of 25, 26, and 27 years, respectively, while a SCCOHT patient with an exclusively somatic SMARCA4 inactivation (patient #3) had an AAD of 35 years. It has been reported that regardless of tumor stage, most SCCOHT patients relapse and die of disease within a comparatively short period of time, with an estimated one-year overall survival rate of 50% and a fiveyear overall survival rate of less than 10%. 15 A review of 47 cases revealed a median overall survival of 14.9 months only, which increased to 35.3 months when SCCOHT was diagnosed at tumor stage I. 5 Young et al. reported that 14 of 42 patients with stage IA disease remained well and free of disease 1-13 (average 5.7) years postsurgery. 7 In our investigation, two patients with a stage IA disease and a germline PV in SMARCA4 remained cancer-free for 7 years (patient #1) and 5 years (patient #2), respectively, following platinum-based multi-agent regimen. Future clinical trials may assess whether SCCOHT patients with germline PVs in SMARCA4 have a better therapy outcome compared with SCCOHT patients without germline PVs in SMARCA4.
The by far highest germline PV prevalence in established OC predisposition genes was observed in patients with high grade serous OC (21/62, 33.9%). In contrast, germline PVs in established OC predisposition genes were rare or even absent in patients with borderline (2/39, 5.1%), low grade serous (1/26, 3.8%), and mucinous (0/27) ovarian tumors. The mucinous tumor phenotype was highly enriched in our study sample compared with study samples unselected for AAD, in which the proportion of this tumor phenotype was reported to be approximately 3%. 16  Thus, the proportion of the mucinous tumor phenotype appeared to be inversely correlated with the mean AAD of the overall study sample, which is consistent with the observation that 26% of all patients with mucinous OC are younger than 44 years at presentation. 16 An intriguing finding is that in our study sample none of the 27 patients with mucinous OC carried a germline PV in the established OC predisposition genes. The same holds true for the 16 patients with mucinous OC described by Norquist et al., and the nine patients with mucinous OC described in our previously reported observational AGO-TR1 trial. 1 Thus, our findings support the notion that mucinous OC represents a separate disease entity not associated with PVs in BRCA1/2 16 or other established OC predisposition genes. 4 AUTHOR CONTRIBUTIONS Natalie Herold: Data curation (equal); investigation (equal); writing -original draft (equal); writing -review and editing (equal). Johanna Schmolling: Data curation (equal); resources (equal); writing -review and editing (equal). Corinna Ernst: Data curation (lead); writingreview and editing (equal). Beyhan Ataseven: Resources (equal); writing -review and editing (equal). Britta Bluemcke: Data curation (equal); writing -review and editing (equal). Birgid Schömig-Markiefka: Resources (equal); writing -review and editing (equal). Sebastian Heikaus: Resources (equal); writing -review and editing (equal). Uwe-Jochen Goehring: Resources (equal); writing -review and editing (equal). Christoph Engel: Resources (equal); software (equal); writing -review and editing (equal). Bjoern Lampe: Resources (equal); writing -review and editing (equal). Kerstin Rhiem: Resources (equal); writing -review and editing (equal). Philipp Harter: Resources (equal); writing -review and editing (equal). Jan Hauke: Resources (equal); writing -review and editing (equal). Rita K. Schmutzler: Funding acquisition (equal); resources (equal); supervision (equal); writing -review and editing (equal). Eric Hahnen: Conceptualization (lead); data curation (equal); formal analysis (equal); funding acquisition (equal); investigation (equal); methodology (equal); project administration (equal); supervision (equal); writing -review and editing (equal).

ACKNOWLEDGMENTS
Open Access funding enabled and organized by Projekt DEAL.

FUNDING INFORMATION
The GC-HBOC is supported by the German Cancer Aid (grant no 110837 and grant no 70114178, coordinator: Rita K. Schmutzler, Cologne) and the Federal Ministry of Education and Research (BMBF), Germany (grant no 01GY1901). Genetic analyses were supported by the Köln Fortune Program, Faculty of Medicine, University of Cologne, Germany. The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; nor in the decision to submit the manuscript for publication.
We acknowledge support for the Article Processing Charge from the German Research Foundation (DFG, 491454339).

DATA AVAILABILITY STATEMENT
Data available in article supplementary material.