Abstract
NSCLC is the most lethal tumor, with little improvement in survival. A plethora of new targeted agents are being developed but their efficacy is still rather negligible when used in non-customized settings. In patients with EGFR mutations, landmark clinical outcomes are being attained with EGFR tyrosine kinase inhibitors (TKIs), yet patients often relapse. In patients without EGFR mutations, DNA damage response is central for proper management of NSCLC patients, yet to date, techniques have not been available for use in daily clinical practice. DNA double-strand breaks (DSBs) are the most detrimental form of DNA damage1 . BRCA1 induces a 10–1000-fold increase in resistance to a range of DNA-damaging agents, in particular those that give rise to DSBs but a >100-fold increase in sensitivity to spindle poisons, paclitaxel and vinorelbine2 . BRCA1 has also been identified as a prognostic marker in resected non-small-cell lung cancer (NSCLC) patients3,4 .
A variety of experimental models in breast and ovarian cancer cells show that inducible expression of BRCA1 enhanced paclitaxel sensitivity5 . A short interfering RNA-mediated inactivation of endogenous BRCA1 led to paclitaxel and docetaxel resistance2,6,7 , and reconstitution of BRCA1-deficient cells with wild-type BRCA1 enhanced sensitivity to paclitaxel and vinorelbine2 . This differential modulating effect of BRCA1 mRNA expression was also observed in tumor cells isolated from malignant effusions of NSCLC and gastric cancer patients, where BRCA1 mRNA levels correlated negatively with cisplatin sensitivity and positively with docetaxel sensitivity8 . Four retrospective studies – two in NSCLC9,10 and two in ovarian cancer7,11 patients – found that low or intermediate BRCA1 mRNA levels correlated with a significantly longer survival following platinum-based chemotherapy7,9 , while survival in patients with higher BRCA1 expression increased following taxane-based chemotherapy7 .
RAP80 targets the BRCA1-BARD1 E3 ligase to MDC1-H2AX-dependent lysine 63-linked ubiquitin proteins at double-strand breaks12 . Three studies have shown that the abrogation of RAP80 reduced the formation of BRCA1-induced foci to 28%13 , 2%14 and 0%15 . Moreover, Abraxas and RAP80 foci formation is BRCA1-independent13 . A phase II study16 tested whether high levels of RAP80 could predict resistance to cisplatin in the presence of low BRCA1 levels. In the group of patients receiving customized gemcitabine plus cisplatin based on low BRCA1 levels, progression-free survival (PFS) was 14 months in those with low RAP80 levels, but when RAP80 was elevated, PFS declined abruptly. Patients with intermediate BRCA1 levels, treated with cisplatin plus docetaxel, had an overall median survival of 9 months, which increased to 13–16 months when RAP80 levels were intermediate or high. These results can be explained by pre-clinical findings that RAP80 is able to translocate to irradiation-induced foci in HCC1937 cells which express a truncated BRCA1 that is unable to migrate to nuclear foci17 . This indicates that RAP80 could replace the BRCA1 DNA repair function in cells lacking BRCA1. Thus, although different platinum doublets show the same18 – or slightly different19 – survival overall, differences could be found when customizing chemotherapy based on a model of BRCA1 and RAP80.
Based on these findings, an international phase III trial (BRCA1-RAP80 Expression Customization [BREC]) has been initiated in advanced NSCLC to include customization based on RAP80 and BRCA1 mRNA expression. Patients in the control arm receive cisplatin plus docetaxel and those in the experimental arm receive chemotherapy based on RAP80 and BRCA1 mRNA levels: low RAP80 levels, cisplatin plus gemcitabine; intermediate-high RAP80 and low-intermediate BRCA1, cisplatin plus docetaxel; intermediate-high RAP80 and high BRCA1, docetaxel alone.
In contrast to the BRCA1-RAP80 model, in NSCLC patients harboring EGFR mutations treated with the EGFR inhibitor erlotinib, PFS and overall survival were strikingly influenced by BRCA1 – but not by RAP80 – levels. PFS was 14 months overall, while it was 27 months in patients with low BRCA1 levels, 18 months in those with intermediate levels, and 10 months in those with high levels; no influence of RAP80 levels was observed (data derived from patients included in the Spanish Lung Adenocarcinoma Data Base20 ). The T790M mutation, which predicts resistance to erlotinib, was present at baseline in 35% of the patients. PFS was still 27 months in patients with the T790M mutation and low BRCA1 levels; however, it plummeted to three months in those with the T790M mutation and intermediate or high BRCA1 levels. Depending on the type of DNA damage and recognition of DSBs, the H2AX-dependent signaling pathway could be unnecessary for BRCA1-dependent DSB repair21 . A parallel pathway has been proposed that acts cooperatively with the H2AX/MDC1/RNF8/RAP80 pathway to regulate the localization and function of BRCA1 in response to specific DNA damage21 .
Recently, it has been shown that BRCA1 SUMOylation could have a central role in predicting response to radiotherapy and chemotherapy. SUMO E3 ligases PIAS1 and PIAS4 are responsible for SUMOylation events at DSBs. Depletion of PIAS1 and PIAS4 causes defects in DSB repair, and these cells are highly sensitive to DSBs caused by ionizing radiation. Importantly, recruitment of BRCA1 depends on both PIAS1 and PIAS422,23 , which regulate RPA, and thus the function of the NHEJ pathway23 . mRNA levels of PIAS, CtIP and other genes are now being explored in the SLADB20 patients, with the aim of elucidating the role of BRCA1 SUMOylation. Translational studies could lead to the creation of a gene expression toolkit for personalized treatment with specific chemotherapy or targeted therapy, to be implemented in large-scale European-wide studies for advanced NSCLC patients.
Citation Information: Clin Cancer Res 2010;16(7 Suppl):PL2-2
RSS Feeds