Poly(adenosine diphosphate [ADP]–ribose) polymerase (PARP) inhibitors are designed to exploit the “synthetic lethality” concept for cancer therapy. The success of PARP inhibitors in BRCA-mutated tumors in the initial clinical trials provided a proof-of-concept for the validity of this approach. PARP inhibitors have attracted particular attention in the field of breast cancer because of their potential to be the first targeted therapy for the difficult-to-treat triple negative disease (TNBC). However, the excitement over the positive findings associated with iniparib treatment in TNBC in the phase II study[1] was later dampened by the recent disappointment at the phase III data[2] and the confusion over the actual mechanisms of action of iniparib; meanwhile, investigations involving more potent PARP inhibitors in sporadic breast cancers are still in too early a stage to inform the conversation. The manuscript by Rios and Puhalla provides a needed summary of this topic and a perspective on the future of the development of PARP inhibitors in breast cancer.
The efficacy of PARP inhibitors used as single agents in advanced BRCA-related breast cancers has been impressive in early-phase clinical trials of olaparib[3] and MK-4827.[4] We agree that although phase III studies are needed, and while data on other PARP inhibitors and combinations of PARP inhibitors with various chemotherapy agents are pending, these agents are likely going to have a significant impact in this patient population.
What about the adjuvant or prevention setting, as Rios and Puhalla suggest? It makes sense to eradicate micrometastatic disease via the synthetic lethality approach in the adjuvant setting, and “maintenance” hopefully would not be needed because cell death would have occurred as a result of the “induction” therapy. In addition, the data from Fong et al demonstrating the accumulation of DNA breaks in the normal cells of hair follicles of patients treated with PARP inhibitors raise concerns regarding the safety of long-term PARP inhibition.[5] Disruption of the adprt1 gene (homologue for PARP1) in mice resulted in telomere shortening and chromosomal instability.[6] It is therefore a significant concern that PARP inhibition in the prevention setting might actually exacerbate genomic instability. Interestingly, there have been no reports of PARP mutations being detected in hereditary cancer syndromes. We agree with Rios and Puhalla that caution must be exerted with regard to the use of PARP inhibitors in otherwise healthy individuals.
In the case of TNBC, the complexity of the genomic structure and the molecular heterogeneity present a significant challenge in the evaluation of targeted therapeutics, and convincing data have yet to be generated proving the utility of PARP inhibitors. If synthetic lethality is to be exploited in TNBC with PARP inhibitors, a way is needed to identify those patients who have defects in the BRCA pathway or the homologous recombination–directed DNA repair (HRR) pathway in order to narrow down the target population. However, to date there is not a validated assay that can detect HRR deficiency using archival tumor specimens. Functional assays, such as that which detects the formation of DNA repair protein foci following radiation in ex vivo breast cancer biopsy specimens have been explored, but these assays are difficult to employ in the clinical setting.[7] Rios and Puhalla have reviewed the possibility of using PARP expression level, array CGH patterns, gene expression signature or BRCA1 mRNA expression, and PTEN loss, but none of these approaches has been validated. Perhaps another way to attack the problem is at the genomic level. McCabe et al reported that deficiencies in HRR pathway genes—including RAD51, RAD54, DSS1, RPA1, NBS1, ATR, ATM, CHK1, CHK2, FANCD2, FANCA, and FANCC—also induced sensitivity to PARP inhibition.[8] In another study by Turner et al, a synthetic lethal SiRNA screen identified additional genes that mediate sensitivity to a PARP inhibitor, including cyclin-dependent kinase 5 (CDK5) and other genes.[9] Perhaps a gene mutation panel could be created to predict PARP inhibitor sensitivity?
We argue that, in addition to the enhancement of the cytotoxicity of chemotherapy agents, combination therapy with other targeted agents is another potentially rational approach. Perhaps a synthetic lethality can be generated by the combination of inhibitors against PARP and against another HRR pathway target in an otherwise DNA repair–proficient cellular background. This could greatly broaden the utility of PARP inhibitors.
In summary, PARP inhibitors have great therapeutic potential for the treatment of breast cancer patients, particularly those who have germ-line BRCA mutations. Although their role in sporadic breast cancer has not been defined, clinical trials that capture the subset of TNBC that possess a true BRCA-like phenotype is crucial.
Financial Disclosure: Dr. Ellis serves as a consultant to sanofi-aventis. Dr. Roop and Dr. Ma have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
