Name Shoko Vos
Date defended October 3, 2018
(Co-) Supervisors Prof. Dr P.J. van Diest, Prof. Dr A.L. Bredenoord
Title of thesis In search of tissue biomarkers for BRCA1/2-related breast cancer: Investigating immunohistochemical, molecular, and ‘gen-ethical’ aspects

Read the thesis online

Thesis summary
Hereditary breast cancer, accounting for 5-10% of all breast cancer cases, is mainly caused to BRCA1 and BRCA2 germline mutations. Current eligibility for genetic testing is mostly based on clinical characteristics such as young age at diagnosis and positive family history. However, potential germline-mutation carriers may still be missed using these criteria. This thesis’ primary aim was to identify clues in the tumor tissues themselves for underlying BRCA1/2 germline mutations that could help in selecting patients for genetic testing. We developed the most accurate and comprehensive immunohistochemistry (protein expression)-based prediction model to date for BRCA1/2-related breast cancer. The prediction model includes 14 predictors (age, mitotic activity index, cyclinD1, ERα, ERβ, FGFR2, FGFR3, FGFR4, GLUT1, IGFR, Ki67, MLH1, p120, and TOP2A) with excellent discriminative performance (AUC = 0.943). BRCA promoter methylation and germline mutations are thought to be mutually exclusive, a feature which could be exploited in clinical practice. From our experimental MS-MLPA study and a systematic review, we can conclude that BRCA methylation is indeed rare in BRCA1/2 germline mutation-related breast carcinomas (at least 0.0-5.3%), but methylation frequencies vary between CpG sites. However, the selected studies show large differences in methodology, quality, and risk of bias. The occurrence of BRCA methylation in breast or ovarian carcinomas from BRCA carriers may therefore be an underestimation. We also discovered many deregulated miRNAs in BRCA1/2-related breast carcinomas, which may be potentially specific to BRCA1/2-related breast carcinogenesis and to the chromosomal instability characteristic of these tumors. The second part of thesis was focused on the ‘gen-ethical’ aspects that should be taken into account for a responsible advancement of cancer research and clinical cancer care. In cancer research and modern pathology practice, large-scale DNA-sequencing techniques are increasingly being used, generating large amounts of potentially sensitive data with an increasing risk of unsolicited findings. We showed that many moral duties have been attributed to genomics researchers, related to disclosure, consent, privacy, and social responsibilities. However, awareness and concrete implementation of these duties in research practice is lagging behind. Moreover, we showed that ethical issues are present in every stage of the modern molecular pathology analysis procedure. These include issues related to consent, disclosure as well as future use of genetic data. Current molecular pathology guidelines are focused on technical issues and attention let alone consensus on ethical issues has been lagging behind. Pathologists should be encouraged to take responsibility for the adequate and responsible use of molecular analyses their laboratories perform. To conclude, this thesis shows that pathology may play a role in the early detection of BRCA1/2 germline mutation carriers among breast cancer patients. Technical advances alone are however not enough for the (responsible) advancement of personalized cancer care. It calls for an active role of researchers and pathologists, who should be aware of their moral duties and responsibilities with respect to participants, patients and society.