Targeting DNA Homologous Repair Proficiency With Concomitant Topoisomerase II and c-Abl Inhibition.

Critical DNA repair pathways become deranged during cancer development. This vulnerability may be exploited with DNA-targeting chemotherapy. Topoisomerase II inhibitors induce double-strand breaks which, if not repaired, are detrimental to the cell. This repair process requires high-fidelity functional homologous recombination (HR) or error-prone non-homologous end joining (NHEJ). If either of these pathways is defective, a compensatory pathway may rescue the cells and induce treatment resistance. Consistently, HR proficiency, either inherent or acquired during the course of the disease, enables tumor cells competent to repair the DNA damage, which is a major problem for chemotherapy in general. In this context, c-Abl is a protein tyrosine kinase that is involved in DNA damage-induced stress. We used a low-dose topoisomerase II inhibitor mitoxantrone to induce DNA damage which caused a transient cell cycle delay but allowed eventual passage through this checkpoint in most cells. We show that the percentage of HR and NHEJ efficient HeLa cells decreased more than 50% by combining c-Abl inhibitor imatinib with mitoxantrone. This inhibition of DNA repair caused more than 87% of cells in G2/M arrest and a significant increase in apoptosis. To validate the effect of the combination treatment, we tested it on commercial and patient-derived cell lines in high-grade serous ovarian cancer (HGSOC), where chemotherapy resistance correlates with HR proficiency and is a major clinical problem. Results obtained with HR-proficient and deficient HGSOC cell lines show a 50-85% increase of sensitivity by the combination treatment. Our data raise the possibility of successful targeting of treatment-resistant HR-proficient cancers.

Constitutional mosaicism for a BRCA2 mutation as a cause of early-onset breast cancer.

Germline mutations in the BRCA1 and BRCA2 genes cause hereditary breast and ovarian cancer syndrome (HBOC). Mutations in these genes are usually inherited, and reports of de novo BRCA1/2 mutations are rare. To date, only one patient with low-level BRCA1 mutation mosaicism has been published. We report on a breast cancer patient with constitutional somatic mosaicism of a BRCA2 mutation. BRCA2 mutation c.9294C>G, p.(Tyr3098Ter) was detected in 20% of reads in DNA extracted from peripheral blood using next-generation sequencing (NGS). The BRCA2 mutation was subsequently observed at similar levels in normal breast tissue, adipose tissue, normal right fallopian tube tissue and ovaries of the patient, suggesting that this mutation occurred early in embryonic development. This is the first case to report constitutional mosaicism for a BRCA2 mutation and shows that BRCA2 mosaicism can underlie early-onset breast cancer. NGS for BRCA1/2 should be considered for patients whose tumors harbor a BRCA1/2 mutation and for individuals suggestive of genetic predisposition but without a family history of HBO.

Molecular alterations in pediatric brainstem gliomas.

BACKGROUND: Diffuse intrinsic pontine gliomas (DIPGs) have a dismal prognosis. Previously, diagnosis was based on a typical clinical presentation and magnetic resonance imaging findings. After the start of the era of biopsies, DIPGs bearing H3 K27 mutations have been reclassified into a novel entity, diffuse midline glioma, based on the presence of this molecular alteration. However, it is not well established how clinically diagnosed DIPG overlap with H3 K27-mutated diffuse midline gliomas, and whether rare long-term survivors also belong to this group. METHODS: We studied tumor samples obtained at diagnosis or upon autopsy from 23 children, including two long-term survivors. Based on clinical, radiological, and histological findings, all tumors were previously diagnosed as DIPGs. All samples were analyzed for genetic alterations by next-generation sequencing (NGS) and for protein expression by immunohistochemistry (IHC). RESULTS: H3 K27 was mutated in NGS or IHC in 20 patients, excluding both long-term survivors. One of these long-term survivors harbored a mutation in IDH1, formerly considered to be an alteration absent in pediatric diffuse brainstem gliomas. Other altered genes in NGS included TP53 (10 patients), MET and PDGFRA (3 patients each), VEGFR and SMARCA4 (2 patients each), and PPARγ, PTEN and EGFR in 1 patient, respectively. IHC revealed cMYC expression in 15 of 24 (63%) of all samples, exclusively in the biopsies. CONCLUSIONS: Eighty-seven percent of the tumors formerly diagnosed as DIPGs could be reclassified as H3 K27-mutated diffuse midline gliomas. Both long-term survivors lacked this alteration. Contrary to former conceptions, IDH1 mutations may occur also in pediatric brainstem gliomas.

EGFR gene copy number predicts response to anti-EGFR treatment in RAS wild type and RAS/BRAF/PIK3CA wild type metastatic colorectal cancer.

Anti-EGFR antibodies are used for the treatment of RAS wild type metastatic colorectal cancer. We previously showed that EGFR gene copy number (GCN) predicts response to anti-EGFR therapy in KRAS exon 2 wild type metastatic colorectal cancer. The aim of our study was to analyse the predictive role of EGFR GCN in RAS/BRAF/PIK3CA wild type metastatic colorectal cancer. The material included 102 patients with KRAS exon 2 wild type metastatic colorectal cancer treated with anti-EGFR ± cytotoxic therapy. Next generation sequencing was used for KRAS, NRAS, BRAF and PIK3CA gene mutation analyses. EGFR GCN was analysed by EGFR immunohistochemistry guided automated silver in situ hybridisation. Increased EGFR GCN (≥4.0) predicted a better response and prolonged progression free survival in anti-EGFR treated RAS/BRAF/PIK3CA wild type patients (Log-rank test, p = 0.0004). In contrast, survival of RAS/BRAF/PIK3CA wild type, EGFR GCN below 4.0 patients did not differ from patients with mutant RAS, BRAF or PIK3CA. Our study indicates that EGFR GCN predicts anti-EGFR treatment efficacy in patients with RAS/BRAF/PIK3CA wt metastatic CRC. Tumours with EGFR GCN below 4.0 appear to be as refractory to anti-EGFR treatment as tumours with mutation in any of the RAS/RAF/PIK3CA pathway genes.