A new synthetic lethal strategy expands the application of PARP inhibitors/cisplatin

Abstract

In a recent study published in Signal Transduction and Targeted Therapy, Zhang et al.¹ identified a panel of genes that served as a novel predictor of response to poly adenosine diphosphate-ribose polymerase (PARP) inhibitors/cisplatin in HR proficient patients, which could guide a broader application of PARP inhibitors/cisplatin in cancer therapy.

Cancer cells differ from normal cells in their ability to repair damaged DNA—most cancer cells lose one or more DNA repair pathways, resulting in greater reliance on the remaining pathways.² Thus, small molecules that can induce DNA damage have been used to treat various cancers. Among them, cisplatin/PARP inhibitors are well established cancer drugs and are used to target tumor cells with homologous recombination (HR) defects.² Platinum salts (carboplatin, cisplatin, and oxaliplatin) are the commonly-used chemotherapeutic agents, which were historically thought to cause cell death by inducing DNA damage.² Recent studies suggest that the mechanisms of action of platinum salts are more diverse³ (Figure 1A). Zhang et al.¹ further showed that cisplatin promotes cell death through DNA damage-induced ribosomal stress, rather than failed DNA repair, in certain tumor cells. PARP inhibitors are approved for the treatment of ovarian and breast cancers with BRCA1/2 mutations, and act through synthetic lethality in DNA repair-deficient tumors.^3-5 However, it is known that some HR-proficient patients also respond well to PARP inhibitors and cisplatin therapy.³ Consistently, Zhang et al.¹ also identified patients who benefited from the treatment of PARP inhibitors, despite their normal HR functions. Therefore, it is necessary to identify biomarkers that can help to stratify the patients so they will benefit most from PARP inhibitors and cisplatin therapy.

To identify these biomarkers, the authors analyzed RNA-Seq data from the Cancer Cell Line Encyclopedia and drug sensitivity data (GDSC) from the extensive and Sanger cell line databases¹ (Figure 1B). They used weighted gene co-expression network analysis to negatively correlate drug signatures with co-expressed gene modules.¹ Through these analyses, the authors found that expression of genes in the ribosome biogenesis pathway could be used to predict cellular drug response to PARP inhibition or cisplatin-based chemotherapy.¹ Ultimately, they obtained a panel of 8 genes involved in ribosome biogenesis for further analysis.¹

In the following studies, the authors provided multiple lines of evidence suggesting that these eight genes could be used to predict PARP inhibitors/cisplatin sensitivity.¹ First, HR-proficient tumor cell lines in which the eight genes were highly expressed were sensitive to PARP inhibitors.¹ Second, the authors verified the results using an organoid library for ovarian cancer treated with PARP inhibitors/cisplatin.¹ Finally, the authors collected a clinical cohort of PARP inhibitor-treated ovarian cancer patients and obtained similar results.¹ Importantly, this may be the first clinical cohort with both complete patient RNA-Seq and WGS data for PARP inhibitor treatment.¹ Based on these studies, the authors proposed a new paradigm to expand the clinical use of PARP inhibitors/cisplatin by combining the gene panel testing and HR status screening¹ (Figure 1C).

In 2005, Helladay and Ashwoth's groups discovered that mutation of BRCA1/2 disrupted the HR pathway of DNA repair in cells, and render the cancer cells particularly sensitive to inhibition of the enzyme PARP.^{4, 5} As PARP is involved in base excision repair, they proposed a synthetic lethal strategy by simultaneously blocking two DNA repair pathways.^{4, 5} Zhang et al.¹ suggested that this model could only apply to cells with low expression of the eight genes; in cells with high expression, PARP inhibitors/cisplatin can induce cancer cell death independent of HR status. The new data explained why a large number of patients with normal HR functions are also sensitive to PARP inhibitors/cisplatin.¹ More importantly, this study suggests that increasing the expression of the 8 genes may represent a novel synergistic lethal strategy, and the authors have already reported three marketed drugs with such functions.¹

Many studies have focused on the DNA repair pathways to understand the sensitivity of PARP inhibitors/cisplatin treatment.^2-5 Zhang et al.¹ went a different direction by searching new features that could predict the response to PARP inhibitors/cisplatin treatment, and made several interesting findindings. First, they discovered a gene panel that could be used to predict PARP inhibitors/cisplatin response in HR proficient patients. In their newly proposed paradigm, patients with high expression of the eight genes could be screened out to undergo PARP inhibitors/cisplatin treatment.¹ Compared to HR status testing, the gene panel testing could be much cheaper and convenient. Secondly, Zhang et al.¹ proposed a new synthetic lethal strategy that is distinct from the classical strategy suggested by Helladay and Ashwoth's groups. Helladay and Ashwoth's strategy relies on inhibiting two different DNA repair pathways, but fails to explain why PARP inhibitor/cisplatin are still effective in treating many tumor patients with normal HR functions.¹ Zhang et al.¹ found that PARP inhibitor/cisplatin could induce lethality in cells with high expression of the eight genes, even their HR functions were proficient.¹ As a result, they proposed a new therapeutic strategy by combining PARP inhibitor/cisplatin with drugs that increase the expression of the eight genes.¹ Last but not least, the authors reported that three marketed drugs could enhance the treatment effect of PARP inhibitors/cisplatin in some tumor cells.¹ Future studies should identity additional drugs with similar activities.

Of course, their work also raises some important questions. First, more rigorous prospective trials are required to validate the authors' conclusions, due to the heterogeneity of ovarian cancer and inherent bias of the retrospective studies. Second, it remains to pinpoint the mechanisms by which these genes regulate ribosomal stress and thus affect cancer drug sensitivity.^{1, 2} In summary, the study by Zhang et al.¹ represents a major milestone in the fields of DNA repair and cancer chemotherapy, and may stimulate additional works that search novel therapeutical approaches by targeting the DNA repair pathways.

Jinrui Wang: writing—original draft (equal); writing—review & editing (equal). Daniel D. Billadeau: writing—original draft (equal); writing—review & editing (equal). Ying Zheng: writing—original draft (equal); writing—review & editing (equal). Da Jia: Funding acquisition (equal); writing—original draft (equal); writing—review & editing (equal). All authors have read and approved the final manuscript.

The authors declare no conflicts of interest.

This study did not involve human participants and/or animals or informed consent. Thus, ethical clearance is not applicable to this article.