DRUM Team (DNA Repair and Uveal Melanoma)

Marc-Henri Stern Chef d'équipe Tel:

Our team DNA Repair and Uveal Melanoma (D.R.U.M.) is mainly interested (i) in genomic instabilities in breast/ovarian cancers and uveal melanoma and (ii) in genetics and genomics of uveal melanoma, using innovative bioinformatics and data mining approaches.

1- Genomic characterization of breast and ovarian carcinoma.

• Deleterious germline mutations of the two major susceptibility genes BRCA1 or BRCA2 lead to an increased risk of developing breast and ovarian carcinomas. BRCA1 and BRCA2 encode key actors of the DNA repair by homologous recombination (HR) pathway and behave like classical tumor suppressor genes in approximately half of high grade so called triple negative (hormone receptors negative, HER2 not amplified) breast carcinoma and high grade ovarian carcinoma. We have identified a signature of genomic instability related to the inactivation of HR (HR deficiency or HRD, also known as BRCAness), most often by inactivation of BRCA1 or BRCA2 (1); published patents: US20150140122A1, US20170260588A1; exclusive licensing with Myriad Genetics, USA). This signature measures Large Scale State Transition (LST), the large number of which sign HRD. We have shown that LSTs correspond to inter-chromosomal translocations, developed a predictive signature of BRCA2 mutations, and confirmed the diagnostic interest of the LST signature in independent series of breast and ovarian cancers (2-10). Recent developments include adapting the LST signature from shallow Whole Genome Sequencing, for affordable and robust clinical diagnosis (Eeckhoutte, in preparation).
• However, the identification of unusual genomic patterns in ovarian carcinoma led us to identify a new genomic instability related to inactivation of CDK12 (11). This instability is characterized by numerous giant tandem duplications distributed over the tumor genome. CDK12 encodes a Cyclin K dependent kinase that activates RNA polymerase II. Its inactivation has for main effect to stimulate intronic premature polyadenylation, downregulating the expression of large genes, including DNA repair genes.
• As many breast cancer prone families are not explained by germline mutations of BRCA1, BRCA2 or other HR genes, we characterized in depth a family with an unusual predisposition to breast and kidney cancers. This led us to identify BAP1 as a new predisposing gene for clear cell renal cell carcinoma, while its role in breast cancer predisposition was excluded (12). We sought to understand the functions of BAP1 in cellular models and showed profound metabolic and cellular changes related to BAP1 expression (13). BAP1 is a deubiquitinase, mainly of histone H2A (H2AK119ub1), antagonizing the Polycomb Repressive Complex 1 (PRC1). We are now involved in international studies on BAP1 germline mutations (14).
• DNA repair related diseases include rare pediatric complex syndromes, often associating immuno-deficiency and cancer proneness. We characterized unusual forms of such diseases, including ataxia telangiectasia, Nijmegen disease and ATLD, sometimes discovered in adults, and we developed assays to assist diagnosis and to understand the consequences of gene mutations on DNA repair (15-18).

2- Genetic and genomic characterization of uveal melanoma (UM).

Choroidal UM is the most common form of intraocular primary malignancies in adults, but is a rare tumor with an incidence rate of 5.6 cases per million person-years (~500 new cases a year in France). Prognosis is dismal when the disease spreads, frequently to the liver. A better understanding of the disease is an urgent need, considering the rapidly unfavorable evolution and lack of effective chemotherapy in its metastatic form.

• In collaboration with R. Marais (Cancer Research UK), we have shown the low mutational burden and lack of ultra-violet radiation signature in UM, in contrast to cutaneous melanoma. We also described recurrent mutations of the splicing gene SF3B1 (19).
• We then deciphered the consequences of SF3B1 mutations on splicing (20-22). More recently, a pan-cancer analysis using cloud computing (in collaboration with Seven Bridges, USA) allowed us to identify SUGP1 mutations as a genocopy of SF3B1 (Alsafadi, submitted); the analysis of the oncogenic consequences of these SF3B1 mutations; and the exploitation of splicing abnormalities as a potential source of tumor immunogenicity (collaboration with O. Lantz).
• UM has an unusual epidemiology, as the disease occurs mostly in individuals of European ancestry. In the hypothesis of predisposing alleles in this population, we initiated the first pan-genomic association study (GWAS) for uveal melanoma. We identified risk variants in the TERT/CLPTM1L region, and the HERC2 pigmentation gene region (23). This study is currently being expanded to search for new risk loci (collaboration with CeRePP, CNG, IARC and CLB (Mobuchon et al, in preparation).
• We participated in the TCGA project to describe primary UM at the –omics level (24) and we are key players in the Horizon2020 project aiming to better understand and target metastatic UM (25). We analyzed the mechanisms of metastatic progression and resistance to chemotherapy of these metastases, and showed that mutational heterogeneity is very limited and does not explain the therapeutic resistance, whereas metastatic progression is mainly associated with the acquisition of recurrent genomic gains and losses (26).

3- DNA repair and uveal melanoma.

• Exploring a UM metastatic patient with an outlier response to immune checkpoint inhibitors, we discovered the role of MBD4 inactivation in a new genetic instability (27). We further explored UM progression in MBD4 deficient tumors, showed the continuous acquisition of mutations in the course of the disease, and used this biological clock to reconstruct the natural history of the disease (26).
• Most MBD4deficient UM cases are associated with deleterious germline MBD4 mutations, and we recently demonstrated the predisposing role of this gene in UM (Derrien, submitted).

4- Other translational activities.

In collaboration with O. Lantz, we initiated the analysis of tumor DNA circulating in UM. We developed several methods in different cancers (28-37), and 3 patents: US20190256921A1, WO2019011971A1, WO2019175323A12017,).

1. Popova et al. (2012) Cancer Research, 72:5454
2. Natrajan et al. (2012) J Pathol, 227:29
3. Pecuchet et al. (2013) Int J Cancer, 133:2834
4. Gruel et al. (2014) Breast Cancer Res, 16:R46
5. Goundiam et al. (2015) Int J Cancer, 137:1890
6. Curtit et al. (2015) Oncotarget, 6:35616
7. Manie et al. (2016) Int J Cancer, 138:891
8. Weigelt et al. (2015) Modern Pathology, 28:607
9. Jdey et al. (2017) Cancer Res, 77:4207
10. Gentric et al. (2019) Cell Metab, 29:156
11. Popova et al. (2016) Cancer Res, 76:1882
12. Popova et al. (2013) Am J Hum Genet, 92:974
13. Hebert et al. (2017) Oncotarget, 8:72513
14. Walpole et al. (2018) J Natl Cancer Inst, 110:1328
15. Meneret et al. (2014) Neurology, 83:1087
16. Rieunier et al. (2017) Methods Mol Biol, 1599:25
17. Fievet et al. (2019) Hum Mutat, 40:1690
18. Fievet et al. (2019) Hum Mutat, 40:1713
19. Furney et al. (2013) Cancer Discov, 3:1122
20. Alsafadi et al. (2016) Nat Commun, 7:10615
21. Gentien et al. (2014) Leukemia, 28:1355
22. Bondu et al. (2019) Sci Transl Med, 11:
23. Mobuchon et al. (2017) NPJ Genom Med, 2:
24. Robertson et al. (2017) Cancer Cell, 32:204
25. Rodrigues et al. (2019) Cancers (Basel), 11:1032
26. Rodrigues et al. (2019) Clin Cancer Res, 25:5513
27. Rodrigues et al. (2018) Nat Commun, 9:1866
28. Bidard et al. (2019) Cells, 8:
29. Decraene et al. (2018) Clin Chem, 64:317
30. Cabel et al. (2018) Nat Rev Clin Oncol, 15:639
31. Riva et al. (2017) Clinical Chemistry, 63:691
32. Riva et al. (2016) Molecular Oncology, 10:481
33. Saliou et al. (2015) Expert Review of Molecular Diagnostics, 16:39
34. Madic et al. (2015) Int J Cancer, 136:2158
35. Lebofsky et al. (2015) Molecular Oncology, 9:783
36. Bidard et al. (2014) Int J Cancer, 134:1207
37. Madic et al. (2012) Clin Cancer Res, 18:3934