Evaluation of molecular analysis in challenging ovarian sex cord-stromal tumours: a review of 50 cases

Colin J.R. Stewart, Benhur Amanuel, Leanne De Kock, Maria Apellaniz-Ruiz, Amerigo Carrello, Tino Giardina, Fabienne Grieu-Iacopetta, Marc A. Thomas, William D. Foulkes

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)


Molecular profiling was performed in 50 problematic ovarian sex cord-stromal tumours (SCSTs) most of which were seen in consultation. Following analysis, 17 were classified as adult granulosa cell tumour (AGCT), 16 of which showed a FOXL2 sequence variant (mutation); the initial favoured diagnosis in five of the cases was benign thecoma/fibrothecoma. Thirteen tumours ultimately classified as cellular fibroma or thecoma were FOXL2 sequence variant negative which was helpful in excluding AGCT. All six Sertoli–Leydig cell tumours (SLCTs) demonstrated DICER1 ‘hot spot’ sequence variants, and one case each of AGCT and SLCT showed high grade histological transformation associated with a concurrent TP53 sequence variant. All eight unclassified SCSTs were negative for FOXL2 mutations and the six tested cases were DICER1 wild type; however, three tumours demonstrated MET, CTNNB1 or TP53 sequence variants. Four cases were classified as juvenile granulosa cell tumour, and one of these harboured a GNAS sequence variant. The single gynandroblastoma and microcystic stromal tumours in the series demonstrated FOXL2 and CTNNB1 alterations, respectively. In summary, molecular analysis aids in accurate classification of challenging ovarian SCSTs and sometimes leads to revision of the favoured provisional diagnosis. TP53 sequence variants may be associated with dedifferentiation in both SLCTs and AGCTs.

Original languageEnglish
Pages (from-to)686-693
Number of pages8
Issue number6
Publication statusPublished - Oct 2020


Dive into the research topics of 'Evaluation of molecular analysis in challenging ovarian sex cord-stromal tumours: a review of 50 cases'. Together they form a unique fingerprint.

Cite this