Oral Presentation ESA-SRB-ANZOS 2025 in conjunction with ENSA

The genomic landscape of juvenile granulosa cell tumours of the ovary (127988)

Tianzhi Yu 1 2 , Maria Alexiadis 2 , William Stocker 3 , Craig Harrison 3 , Peter Fuller 2 , Stuart Archer 4 , Simon Chu 2
  1. Monash University, Melbourne, VIC, Australia
  2. Centre for Endocrinology and Reproductive Health, Hudson Institute of Medical Research, Melbourne, VIC, Australia
  3. Department of Physiology, Monash University, Melbourne, VIC, Australia
  4. Monash Genomics and Bioinformatics Platform, Monash University, Melbourne, VIC, Australia

Juvenile granulosa cell tumour (jGCT) is a rare, hormonally-driven ovarian tumour that primarily affects girls and young woman. While early-stage disease often has favourable outcomes, prognosis is poor for patients with advanced or recurrent disease. Unlike adult GCT, which is uniformly characterised by the FOXL2C134W mutation, the genomic drivers of jGCT remain poorly defined, limiting the development of targeted therapies.

To address this, we performed whole genome sequencing (WGS) on tumour and matched germline DNA from six patients with histologically confirmed jGCT. Sequencing data was processed using the nf-core/Sarek pipeline with two variant callers (Mutect2 and Strelka2), followed by stringent multi-layered filtering, and variant prioritisation through a customised variant ranking framework.

Consistent with previous reports, the pathognomonic FOXL2C134W mutation, was absent in all cases. The known jGCT-associated somatic GNAS hotspot mutation was identified in two patients, and a pathogenic IDH1R132 mutation was observed in one tumour. Mutations in AKT1 and DICER1 were not detected. Each of three cancer-related genes (PLEC, TGFBR1/ALK5, TP53) harboured somatic variants, with each variant detected in two different patients.

Of particular interest, a previously unreported heterozygous missense mutation in TGFBR1/ALK5, a valine substitution at a highly conserved glycine residue, was found in both GNAS-mutated tumours. This variant predicted to activate SMAD2/3 signalling, was functionally validated using a TGFBR1-SMAD transactivation assay. The TGFBR1/ALK5 mutation induced constituent receptor signalling and then enhanced further following ligand binding. Strikingly, an equivalent variant in orthologues ACVR1B/ALK4 gene was also identified in a separate tumour, suggesting activation of ALK/SMAD signalling as a potential oncogenic mechanism in jGCT.

These findings highlight the genetic heterogeneity of jGCT and uncover ALK/SMAD signalling as a recurrently altered pathway. Our study underscores the need for personalised molecular profiling in jGCT and lays the groundwork for exploring novel targeted therapeutic strategies for this rare and understudied tumour.