De rol van genetische veranderingen in de behandeling van het ovarium carcinoom

Ovarian cancer has the highest mortality of all gynaecologic malignancies in the developed world. Following the standard first-line treatment of cytoreductive surgery and platinum-based chemotherapy, there are large inter-individual differences in therapy-related toxicities and outcome. Most patients benefit from initial treatment, but progressively develop platinum resistance during subsequent lines. Advanced genetic technologies have recently improved our understanding of the genetic landscape, characterizing the different clinical phenotypes. However, few predictive or prognostic biomarkers have been identified and consequently less therapeutic stratification has been introduced. In view of the persistent need for a more effective targeted approach for this poor prognostic disease, we aimed to further decipher underlying genetic markers that potentially could tailor therapeutic strategies. Reviewing the current knowledge on the genetic characteristics underlying the most frequent high grade serous (HGSOC) subtype (chapter 3), we remarked the striking heterogeneity with few recurrent mutations and a high number of copy number alterations causing genomic instability. We undertook a first study (chapter 4) to identify germline variants in genes involved in platinum metabolism (ABCB1/MDR1, ABCC1, ABCA1, CYP3A4, GSTP1 and ERCC2) that may determine side-effects and resistance following platinum-based therapy. However, independent validation of these associations is needed prior to their potential clinical application. We contributed this study to the international Ovarian Cancer Association Consortium (OCAC) evaluating associations between outcome and candidate polymorphisms in a maximally powered population, but these studies failed to confirm previously reported associations. To understand the clonal evolution of genetic characteristics towards platinum resistance, we applied in a next step (chapter 5) whole-exome sequencing, targeted resequencing and SNP array profiling to paired samples from diagnosis and recurrence. As such, we identified branched clonal evolutionary patterns causing a considerable degree of heterogeneity between these temporally divided biopsies. This advocates the need for taking biopsies at each recurrence, to optimize clinical trial design when testing targeting therapies. Deficiency of the homologous recombination pathway (HRD) characterized the recurrence biopsies at least to the same extend as the primary biopsies, both for resistant as for sensitive recurrences, concluding that previously described secondary events restoring HRD function are not the major mechanism driving platinum resistance. These findings suggest the potential efficacy of PARP (poly-ADP ribose polymerase) inhibitors for platinum-resistant relapses. While we failed to detect recurrence-specific or resistance-specific aberrations, a ‘platinum score’ of 13 somatic copy number alterations (SCNAs) was developed. This score includes focal amplifications among interesting cancer genes such as CCNE1, ERBB2 and TERT, and correlated with the platinum-free interval (PFI) following the time point the biopsy was taken. An increase of this score in recurrent tumors predicted the change in PFI during subsequent therapy, indicating that a gradual increase in accumulating CNAs during treatment renders the tumor more oncogenic. This way, the adaptability of the HGSOC genome provides the molecular basis of treatment resistance. To improve therapeutic outcomes, a combined approach targeting the different mechanisms included in the ‘platinum score’ might be the step forward. In the future our research group will further focus on the validation of the identified genetic hallmarks of resistance, both for platinum and PARP inhibitors. The application of less invasive technologies such as ‘liquid biopsies’ containing circulating tumor-DNA, might bypass the need for large collections of fresh frozen relapse-biopsies, that limited the current research in validating low prevalent mechanisms as potential drivers of therapy response.status: publishe