BRCA1/2-deficient tumor cells are more sensitive to PARP inhibitors (PARPi) through a synthetic lethal mechanism. Currently, several targeted PAPRi have been approved for ovarian and breast cancer indications. However, PARPi resistance is widespread in the clinic. More than 40% of patients with BRCA1/2 deficiency do not respond to PARPi.

In addition, many patients develop PARPi resistance after long-term oral administration of PARPi. Homologous recombination repair deficiency (HRD), a necessary prerequisite for synthetic lethality, plays a crucial role in killing tumor cells. Therefore, homologous recombination repair restoration (HRR) becomes the main cause of PARPi resistance. DNA replication fork protection has also been reported to contribute to PARPi resistance in BRCA1/2-deficient cells and patients.

Since HRD is a major prerequisite for PARPi to exert its anticancer effect, therefore understanding the HR repair pathway is crucial. DNA damage response (DDR) is activated when DNA double-strand breaks (DSBs) occur in mammalian cells. Cells coordinately use two typical mechanisms to repair DSBs: homologous recombination repair (HR) and non-homologous end joining repair (NHEJ). Generally, NHEJ is the primary repair mechanism for joining broken DNA ends in a non-homologous end-joining manner and occurs throughout the cell cycle, especially in the G0/G1 phase. HR has an advantage in the S/G2 period.

During HR, DSB ends are first cleaved by the MRN complex along with CtIP and nucleases, resulting in the formation of single-stranded DNA (ssDNA) and submitting cells to HR. The excised DNA ends are then coated with hyperphosphorylated single-stranded DNA-binding protein A (RPA). Variant H2AX is activated and phosphorylated by apical kinases. Diffusion of γH2AX along chromosomes facilitates recruitment and accumulation of additional DDR proteins, including p53-binding protein (53BP1) and BRCA1.

Supported by PALB2, BRCA2 binds to BRCA1 and promotes the loading of the recombinase RAD51 on ssDNA. RAD51 mediates the invasion of homologous sequences and the formation of nucleoprotein filaments and D-loops by eliminating secondary structure formation and protecting DNA ends from degradation. Therefore, restoration of the HR pathway by inducing the process of DNA end resection and formation of nucleoprotein filaments and D-loops may lead to PARPi resistance.

01

DNA end resection in PARPi resistance

DNA end resection is the key to the selection of different DNA repair pathways, and multiple reports suggest that DNA end resection is involved in PARPi resistance.

The cell cycle controls the selection of DSB repair pathways. During G1 phase, 53BP1 and RIF1 proteins localize to DSB sites, leading to inhibition of BRCA1 recruitment, blocking DNA excision, and promoting the NHEJ repair pathway. Otherwise, DNA end excision is stimulated in S/G2 phase and HR repair is promoted. DNA end resection is dependent on the activity of cyclin-dependent kinases (CDKs), and loss of function of CDK12 disrupts HR repair and sensitizes ovarian cancer cells to veliparib.

In triple-negative breast cancer (TNBC), loss of CDK12 reversed primary and secondary PARPi resistance regardless of whether BRCA was wild-type or mutant. Furthermore, CDK18 promotes HR and PARPi resistance in glioblastoma stem-like cells by interacting with ATR and regulating ATR-Rad9/ATR-ETAA1 interactions. These evidences suggest that CDKs block DNA end resection leading to PARPi resistance, and their inhibitors may overcome PARPi resistance. Combination therapy of PARPi and CDKs inhibitors is expected to be used in clinical practice

In addition to cell cycle and CDKs, 53BP1 and RIF1 also contribute greatly to DNA end resection and PARPi resistance. 53BP1 is a chromatin-binding protein, and loss of 53BP1 induces DNA end resection and HR restoration, leading to PARPi resistance in various cancers, such as breast, glioblastoma, and ovarian cancer. The protective function of 53BP1 requires the interaction of PTIP and RIF1. Therefore, the interaction between 53BP1 and RIF1 plays a key role in DNA end resection and PARPi resistance.

02

D-loop formation in PARPi resistance

RAD51-ssDNA filaments perform central functions in DNA strand exchange and HR repair, suggesting RAD51 foci as functional biomarkers for HR repair and PAPRi resistance in addition to BRCA mutations. EMI1 was identified as constitutively targeting RAD51 for degradation and as a regulator of PARPi sensitivity. Downregulation of EMI1 enhanced RAD51 accumulation, leading to restoration of HR and development of PARPi resistance in BRCA1-deficient TNBC cells.

Topoisomerase IIβ-binding protein 1 (TOPBP1) is indispensable for the phosphorylation of RAD51 at serine and for the recruitment of RAD51 to chromatin and the formation of RAD51 foci. Deletion of TOPBP1 abolished HR and increased ovarian cancer cell sensitivity to olaparib. Bromodomain 4 (BRD4) is a key chromosomal regulator of genome stability. BRD4 is amplified in various cancers. There is increasing evidence that BRD4 inhibitors are PARPi-sensitive and expand the clinical application of PARPi. APRIN and PALB2 preferentially bind to the D-loop structure and directly interact with RAD51 to stimulate chain invasion and promote HR. Deletion of APRIN and PALB2 has been shown to induce "cancer" and sensitize cells to PARPi. In addition, Pol δ plays an important role in D-loop extension, and inhibition of Pol δ also enhances the sensitivity of HR-cancer cells to PARPi.

03

Back mutations for PARPi resistance

The strongest rationale for the clinical development of PARPis stems from the responses observed in BRCA1/2-deficient (and HR-deficient) cells. Therefore, one resistance mechanism of PARPi involves secondary somatic reversion of BRCA1/2 genes in tumor cells, which essentially restores HR. This may be due to chemotherapy-induced tumor heterogeneity and clonal expansion, known as "Darwinian escape". Combined with a growing body of data, it appears that there is a reversion mutation in PARPi resistance, which appears to be the most effective mechanism of PARPi resistance in BRCAm cancer patients. However, whether the reverse mutation is induced by PARPi itself or by other anticancer drugs, or even spontaneously, remains unclear. After all, cancer cells with BRCA mutations are more likely to be repaired by NHEJ, which leads to the accumulation of genetic variants and an increased risk of reversion.

Recently, the incidence of BRCA reversion in metastatic castration-resistant prostate cancer (mCRPC) was assessed. Using a large genomic database, 24 gBRCAm carriers were screened from 1534 mCRPC patients who underwent ctDNA testing. At the time of testing, five of these 24 patients had received either a PARP inhibitor or platinum-based chemotherapy. Two patients, one receiving olaparib and one receiving carboplatin, developed BRCA2 reverse mutations. Thus, in this germline mutation-positive, platinum- or PARP-exposed cohort, the frequency of BRCA2 reversion was 40%. However, results from another clinical trial showed that of 97 HGSOC patients with gBRCAm or sBRCAm, 8 (8.2%) were identified as having a BRCA reverse mutation prior to rucaparib treatment. After treatment with rucaparib, only 8 of the 78 patients developed BRCA reverse mutations, and the incidence of reverse mutations was only 10.3%. All these results reflect that BRCA reverse mutations may be different in different cancers. Due to the small sample size, additional studies are needed in more patients and various cancers.

Over the past few decades, PARPi has been successfully developed to treat patients with BRCA mutations, providing the notion that synthetic lethal interactions can be translated into cancer therapy. However, preclinical and clinical studies of PRARi are far from complete. In terms of PARPi resistance, multiple potential resistance mechanisms have been identified, such as HR restoration and protection of DNA replication forks. A comprehensive understanding of the function of PARPi, and in particular how PARPi's role in processes unrelated to DNA repair affects PARPi's anticancer activity, will provide insights into the development of drug resistance.

Reference

[1].Liu VF, Weaver DT. The ionizing radiation-induced replication protein a phosphorylation response differs between ataxia telangiectasia and normal human cells. Mol Cell Biol. 1993;13:7222–31.

[2].Polo SE, Jackson SP. Dynamics of DNA damage response proteins at DNA breaks: a focus on protein modifications. Genes Dev. 2011;25:409–33.

[3].Bunting SF, Callén E, Wong N, Chen HT, Polato F, Gunn A et al (2010) 53BP1 inhibits homologous recombination in Brca1-deficient cells by blocking resection of DNA breaks. Cell 141:243–254

[4].Hurley RM, Wahner Hendrickson AE, Visscher DW, Ansell P, Harrell MI, Wagner JM et al (2019) 53BP1 as a potential predictor of response in PARP inhibitor-treated homologous recombination-deficient ovarian cancer. Gynecol Oncol 153:127–134

[5].Bouwman P, Aly A, Escandell JM, Pieterse M, Bartkova J, van der Gulden H et al (2010) 53BP1 loss rescues BRCA1 deficiency and is associated with triple-negative and BRCA-mutated breast cancers. Nat Struct Mol Biol 17:688–695