Prevalence of somatic SF3B1 mutations in lactotroph tumours

J. Simon*^a (Ms), LG. Perez-Rivas^a (Dr), J. Flitsch^b (Prof), M. Buchfelder^c (Prof), J. Thorsteinsdottir^d (Dr), H. Lasolle^e (Dr), C. Cortet^f (Dr), F. Chasseloup^g (Dr), D. Maiter^h (Prof), P. Chanson^g (Prof), G. Raverot^e (Prof), M. Theodoropoulou^a (Prof)

^a LMU Klinikum, Munich, GERMANY ; ^b Universitätskrankenhaus Hamburg-Eppendorf, Hamburg, GERMANY ; ^c University of Erlangen-Nürnberg, Erlangen, GERMANY ; ^d Klinikum der LMU München, Munich, GERMANY ; ^e Université Lyon 1, Lyon, FRANCE ; ^f CHRU de Lille, Lille, FRANCE ; ^g Bicêtre Hospital, Le Kremlin-Bicêtre, FRANCE ; ^h UCLouvain Cliniques Universitaires Saint-Luc, Bruxelles, BELGIUM

* Julia.Simon@med.uni-muenchen.de

Introduction. A somatic mutational hotspot in the exon 14 of SF3B1 gene was described in ~20% of lactotroph tumours in an Asiatic patient cohort (Li et al., 2020; PMID: 32427851). The aim of the study was to determine the prevalence of SF3B1 variants in that hotspot in a Caucasian cohort of patients with lactotroph tumours.

Material and Methods. This retrospective study included 101 patients: 98 patients with lactotroph tumours (54 male) and 3 patients with lactotroph carcinomas (3 male). The mean age at surgery 42±14 years; range 17-72 months). Mean follow up since diagnosis was 112±82 months (range 0-401 months). Most tumours (80) were macroadenomas. Cavernous sinus invasion was reported in 43 cases.

Pituitary tumours and one metastasis were snap frozen and DNA was extracted using the Maxwell tissue purification kit (Promega). SF3B1 exon 14 amplification and Sanger sequencing were performed using specific primers. Sequences were analysed using Mutation Surveyor V4.0.9 and the Ensembl database.

Results. We identified the SF3B1 c.1874G>A (p.Arg625His) variant in 4 out of 101 cases (prevalence 3.96%): 2 carcinomas (one primary tumour, one metastasis) and 2 lactotroph macroadenomas (both French-Five tiered classification 2b). Patients with SF3B1 mutant tumours underwent significantly higher number of pituitary surgeries prior to the sample collection compared to wild-type tumours (Fisher exact test, p=0.016). Radiotherapy was used more frequently in patients with SF3B1 mutant tumours (Chi-square test, p=0.028). Furthermore Ki-67 proliferation index was significantly higher in tumours with SF3B1 mutations (Mann Whitney U-test, p=0.002).

One of the patient with the SF3B1 mutant macroadenoma was treated with radiotherapy after the pituitary surgery, while the other -diagnosed at the age of 17 and operated recently- has not achieved remission after surgery. Both carcinoma patients underwent ≥3 pituitary surgeries, radiotherapy, and received temozolomide and other treatments (bevacizumab, cisplatin/etoposide and pembrolizumab). One of them died due to disease complications.

Conclusion. These data from a multicentre study reveal that the SF3B1 c.1874G>A (p.Arg625His) variant is uncommon (<4%) in lactotroph tumours. In our group of patients with lactotroph tumours SF3B1 c.1874G>A (p.Arg625His) is strongly linked to aggressive lactotroph tumours and carcinomas.

The author has declared no conflict of interest.