Genomic scars as biomarkers of homologous recombination deficiency and drug response in breast and ovarian cancers

Abstract

Poly (ADP-ribose) polymerase (PARP) inhibitors and platinum-based chemotherapies have been found to be particularly effective in tumors that harbor deleterious germline or somatic mutations in the BRCA1 or BRCA2 genes, the products of which contribute to the conservative homologous recombination repair of DNA double-strand breaks. Nonetheless, several setbacks in clinical trial settings have highlighted some of the issues surrounding the investigation of PARP inhibitors, especially the identification of patients who stand to benefit from such drugs. One potential approach to finding this patient subpopulation is to examine the tumor DNA for evidence of a homologous recombination defect. However, although the genomes of many breast and ovarian cancers are replete with aberrations, the presence of numerous factors able to shape the genomic landscape means that only some of the observed DNA abnormalities are the outcome of a cancer cell's inability to faithfully repair DNA double-strand breaks. Consequently, recently developed methods for comprehensively capturing the diverse ways in which homologous recombination deficiencies may arise beyond BRCA1/2 mutation have used DNA microarray and sequencing data to account for potentially confounding features in the genome. Scores capturing telomeric allelic imbalance, loss of heterozygosity (LOH) and large scale transition score, as well as the total number of coding mutations are measures that summarize the total burden of certain forms of genomic abnormality. By contrast, other studies have comprehensively catalogued different types of mutational pattern and their relative contributions to a given tumor sample. Although at least one study to explore the use of the LOH scar in a prospective clinical trial of a PARP inhibitor in ovarian cancer is under way, limitations that result in a relatively low positive predictive value for these biomarkers remain. Tumors whose genome has undergone one or more events that restore high-fidelity homologous recombination are likely to be misclassified as double-strand break repair-deficient and thereby sensitive to PARP inhibitors and DNA damaging chemotherapies as a result of prior repair deficiency and its genomic scarring. Therefore, we propose that integration of a genomic scar-based biomarker with a marker of resistance in a high genomic scarring burden context may improve the performance of any companion diagnostic for PARP inhibitors.

Figures

Figure 1

Genomic aberrations in cancer. Three classes of genomic aberration that develop in cancer cells are depicted: mutations of less than 1 Kbp in length (top box), structural copy number aberrations (CNAs) (bottom left box), and structural rearrangements (bottom right box). The initial state in the germline is shown followed by the corresponding change in the tumor. Mutations that affect regions of less than 1 Kbp are of three basic types: substitutions, of which there are transversions and transitions; insertions; and deletions. Insertions and deletions are often collectively termed ‘indels’. Structural CNAs are typically greater than 1 Kbp in size. One of the basic types is copy number gain. The two homologous chromosomes are shown with a gain of two further copies of region A on the paternal chromosome leading to an imbalance in the allelic ratio (1:3, maternal: paternal). The gained region is highlighted by the green bar adjacent to paternal region A. Copy number loss of regions A and B on the paternal chromosome is shown with a red bar highlighting the deleted regions. Three of the commonest types of structural rearrangement are shown, with the letters A to D and X to Z depicting defined chromosomal segments. An inversion on the same chromosome results in a change to the orientation of DNA sequences on the same chromosome either paracentrically (without crossing the centromere) or pericentrically (crossing the centromere). The inverted sequences in the tumor are shown in red. Translocations can be reciprocal or non-reciprocal and typically occur between non-homologous chromosomes (the green and blue chromosomes are non-homologous). A reciprocal translocation is shown with regions A and B exchanged for regions X and Y. Recombinations typically occur between sister chromatids where they are conservative, but can occur between homologous chromosomes (the green and purple chromosomes are homologous with green being the maternal, and purple the paternal) where recombinations at a heterozygous allelic locus can lead to cnLOH. The dotted boxes indicate where these aberrations are detectable by single-nucleotide polymorphism microarrays, whereas the grey dashed line encompasses those that can also be captured by array comparative genomic hybridization (aCGH), which does not distinguish between alleles. All forms of aberration may be interrogated by using sequencing. A, adenine; C, cytosine; cnLOH, copy number-neutral loss of heterozygosity; G, guanine; LOH, loss of heterozygosity; T, thymine.

Figure 2

Scoring by genomic scars of homologous recombination deficiency and drug response. Eight examples of various forms of structural copy number aberrations and rearrangements are given, whereby each box, lettered A to F, represents a genomic segment of approximately 3 Mbp in length. Below the chromosomes, the three genomic scars - homologous recombination defect (HRD), telomeric allelic imbalance score (NtAi), and large-scale transition (LST) - are listed along with the respective integer count for the scar (0 = not seen, 1 = detected once). LOH, loss of heterozygosity.

Figure 3

Workflow for the development of an integrated predictive biomarker of response to homologous recombination (HR) defect directed therapy. The workflow begins with genomics data - either sequence or single-nucleotide polymorphism microarray data - for tumor samples that have been annotated with patient response data to a given HR targeting drug therapy. After development of a genomic scar measure and a cutoff with high negative predictive value (NPV) were shown to identify non-responders but likely poor positive predictive value (PPV) due to inclusion of patients who have developed resistance (for example, 53BP1 loss) subsequent to development of the genomic scar, two groups can be identified: those predicted not to respond and those predicted to respond accepting a poor PPV. Patients in the former group should not be treated with the drug, whereas for patients in the predicted responder group, gene expression or mutation data are collected. Within the latter group, a biomarker excluding those with acquired resistance is constructed that is highly specific for response to the drug, better dichotomizing patients into those who do and those who do not benefit. By combining the genomic scar biomarker with the resistance-refined biomarker, the resultant two-step companion diagnostic should possess both high NPV and high PPV.