PARP Inhibitors: The Journey From Research Hypothesis to Clinical Approval

Kishan AT Naipal; Dik C van Gent


Personalized Medicine. 2015;12(2):139-154. 

In This Article

Resistance Mechanisms

One important problem that may arise during PARP inhibitor treatment is the development of resistance. Even BRCA-deficient mouse tumors were reported to acquire resistance to PARP inhibitors. For human tumors insufficient data are available on resistance to PARP inhibitors. However, PARP inhibitor resistance mechanisms may be deduced from results obtained from other DNA damaging agents that are highly effective in HR-deficient tumors, such as platinum-based chemotherapy. Using this rationale, several resistance mechanisms have been proposed for PARP inhibitors.

Secondary Mutations

Platinum drugs and PARP inhibitors have cross sensitivity in BRCA1 and BRCA2 mutated cancers. However, despite initial high sensitivity to platinum-based chemotherapy and PARP inhibitors, they can develop resistance by acquiring secondary mutations in the BRCA1 or BRCA2 gene itself. These secondary mutations restore the open reading frame of the BRCA1/2 genes, thereby restoring protein expression.[90–92] These newly transcribed proteins are often different from the normal proteins, often shorter forms containing the specific functional domains that are essential for HR.[90] In a clinical setting these secondary mutations have also been identified in ovarian tumors from patients who had acquired resistance to platinum-based chemotherapy. This appears to be a sizeable fraction, reaching more than 40% of tumors with acquired resistance.[90–93] As no secondary mutations have been reported for tumors showing primary resistance to platinum drugs, this resistance mechanism appears to be associated with acquired resistance. Whether these secondary mutations are also a cause of acquired resistance to PARP inhibitors in human tumors remains to be investigated. Research in mouse tumor models shows that secondary BRCA mutations can cause de novo PARP inhibitor resistance.[94] Although this resistance mechanism has also been reported in some limited clinical data, the frequency of this phenomenon after PARP inhibitor treatment has not yet been investigated sufficiently to draw conclusions about its clinical impact.[93,95]

In addition, secondary BRCA mutations might be the connection between resistance to PARP inhibitors and platinum drugs and also explain the rationale for clinical trials with PARP inhibitors in platinum sensitive cancers as tumors with acquired resistance to platinum also do not respond to PARP inhibitors, most likely due to acquired additional mutations.

Rewiring of DDR Pathways

In addition to reactivation of BRCA function, PARP inhibitor resistance can also be caused by compensatory mutations in other genes. As this specific concept is the opposite of synthetic lethality, the current literature often refers to this phenomenon as 'synthetic viability'.[96] In general, these compensatory mutations do not have a major impact on cellular survival by themselves, but in combination with the BRCA1 gene deficiency they cause PARP inhibitor resistance.

The best studied example of such a compensatory mutation is the loss of 53BP1 in BRCA1-deficient cells.[94,97–100] Further investigation of the molecular mechanism revealed that 53BP1 and BRCA1 compete with each other at DNA ends, where 53BP1 favors Non-Homologous End-Joining (NHEJ) and BRCA1 promotes HR. This can be explained by regulating the formation of a single-stranded DNA tail (also called DNA end resection), which is stimulated by BRCA1 (Figure 3). This step is most probably the main determinant of DSB repair pathway choice, which is influenced by positive and negative regulators. The loss of BRCA1 in cells and animals causes growth arrest, senescence and embryonic lethality.[97,98] However, combination of this genetic defect with 53BP1 loss restores HR, as well as embryonic viability and reduced PARP inhibitor sensitivity.[61,98,100] This mechanism has also been found to explain PARP inhibitor resistance in a BRCA1-deficient mouse tumor model: resistance development after a primary response to PARP inhibitors was explained by loss of 53BP1 expression in at least 20% of tumors.[94] The clinical importance of 53BP1 loss is less clear: the loss of 53BP1 was not associated with resistant BRCA mutated tumors nor with de novo resistance to platinum drugs in patients.[101]

Other examples that restore HR capacity and cause PARP inhibitor resistance have been reported, as well, for example, loss of Rif1 expression that appears to be epistatic with 53BP1.[102–105] These data reveal that sensitivity or resistance to PARP inhibitors might depend on changes of the balance between NHEJ and HR in tumor cells; however, the clinical impact of these mechanisms still has to be evaluated in patient tumor samples.

Upregulated Drug Efflux Pumps

As a completely different mechanism of drug resistance, upregulation of membrane bound drug transporters has been shown to cause resistance to many types of anticancer drugs, including platinum agents. These pumps enhance the transport of drugs from intracellular locations to the extracellular space, thereby reducing the therapeutic effect of the drug.[106] In the case of PARP inhibitors, the Abcref-1a and Abcref-1b genes have been reported to be upregulated in mouse BRCA1 mutated tumor models after long term treatment with Olaparib.[107] These genes encode the P-glycoprotein efflux pump (also known as P-gp or MDR1) that removes this PARP inhibitor from the cells, thus causing resistance.[61,107] To cope with this, novel experimental PARP inhibitors were developed by AstraZeneca (AZD2461) that are poor substrates to P-gp pumps. In mouse tumor models less resistance was noted after prolonged treatment with this specific PARP inhibitor and also no increased expression of P-gp was noticed.[94] Moreover, from experimental data Veliparib was found to be still effective in P-gp overexpressing chemotherapy resistant cells compared with Olaparib, thus suggesting that this PARP inhibitor is not a substrate of P-gp.[108] In conclusion, PARP inhibitors that are not pumped out of the cell by P-gp may be favorable for prolonged treatment or treatment of chemotherapy resistant tumors. However, it is not clear what the contribution of this mechanism is to clinical resistance against PARP inhibitors.