A whole-genome approach to finding tumors with a DNA-repair defect may widen access to PARP inhibitor treatments and refine use of platinum-based chemotherapy, based on new research from Weill Cornell Medicine and NewYork-Presbyterian. In a study published Jan. 12 in Communications Medicine, investigators trained and validated a machine-learning algorithm on whole-genome sequencing (WGS) data to detect homologous recombination deficiency (HRD)-a vulnerability that can make cancers more responsive to PARP inhibition and some DNA-damaging chemotherapies.
“A comprehensive analysis of the entire genome has advantages compared with traditional, targeted detection strategies for predicting homologous recombination deficiency,” said study senior author Dr. Juan Miguel Mosquera, professor of pathology and laboratory medicine and director of research pathology at the Englander Institute for Precision Medicine at Weill Cornell and a pathologist at NewYork-Presbyterian/Weill Cornell Medical Center. Dr. Mosquera is also a member of the Sandra and Edward Meyer Cancer Center at Weill Cornell.
Dr. Juan Miguel Mosquera
Genome-wide signal over single-gene testing
Clinical practice has often centered on BRCA1/2 mutations to guide PARP inhibitor use in breast, ovarian, pancreatic, and prostate cancers. The study team reports that a genome-wide signature of HRD captured with WGS can identify additional patients whose tumors lack BRCA1/2 mutations but still harbor repair defects, potentially reclassifying patients who would be considered ineligible for targeted therapies under current panel-based approaches. Their WGS-derived HRD score builds on prior work showing that genomic “scars” across the entire genome can serve as a functional readout of DNA-repair failure rather than a single-gene snapshot. [1][2][3]
The algorithm was developed in collaboration with a diagnostics company and tested against a commercial method using real-world tumor samples and outcomes, positioning it as a candidate decision-support tool rather than a replacement for existing tests at this stage.
Study design and cohorts
The authors structured the research to move from technical training to clinical comparison, with three predefined phases designed to mirror how new diagnostics are typically evaluated before broad adoption.
| Phase | Sample size | Purpose |
|---|---|---|
| Training | 305 tumors (multiple cancer types) | Train WGS-based algorithm to recognize HRD signals using known HRD-positive and HRD-negative reference cases |
| Validation | 556 cancers | Validate performance across a broader cohort and assess robustness across tumor types and sequencing conditions |
| Head-to-head comparison | 212 tumors | Compare algorithm with an existing commercial method using patient outcomes and therapy exposure |
Detection rates reported in the study
Across tumor types, the WGS-based approach identified clinically meaningful rates of HRD, including in cancers where testing is less standardized today.
| Tumor type | HRD detected | Notes |
|---|---|---|
| Breast | 21% | Includes cases beyond BRCA1/2 mutations, suggesting potential expansion of PARP inhibitor eligibility |
| Pancreatic & bile duct | 20% | Findings may intersect with existing use of platinum-based chemotherapy in subsets of these cancers |
| Gynecologic | 17% | Complementary to existing HRD testing practices in ovarian and related cancers |
| Across detected cases | 24% lacked BRCA1/2 mutations | Supports the idea that structural genomic changes and other pathway alterations can create HRD beyond classic germline mutations |
For health systems and payers, these prevalence estimates offer an early indication of how many additional patients might be captured if WGS-based HRD testing were incorporated into routine oncology workflows.
Signals that challenge existing commercial calls
The head-to-head comparison between the new algorithm and a widely used commercial assay highlights areas where different testing strategies could lead to divergent treatment decisions.
- Several tumors flagged as HRD by the algorithm had treatment responses consistent with HRD-such as durable responses to PARP inhibition or platinum agents-despite a negative call by a commercial method.
- Conversely, some tumors called HRD by the commercial method lacked outcomes that would typically align with HRD sensitivity, and were not flagged by the algorithm.
- These discordances highlight the need for prospective, outcomes-linked studies before any shift in standard-of-care testing, including formal clinical trials designed to adjudicate which assay best predicts benefit in defined settings.
Clinical and system implications under consideration
If validated prospectively, the WGS-based HRD signal could influence a series of upstream and downstream decisions-from tumor board deliberations to payer coverage policies.
- Patient selection: A genome-wide HRD signal could expand eligibility for PARP inhibitors and inform use of platinum-based regimens in select settings, particularly in tumors that have not historically undergone comprehensive genomic profiling.
- Guideline alignment: Any transition from single-gene markers to genome-wide signatures would require assessment by clinical guideline bodies and multidisciplinary tumor boards, and would need to fit within existing frameworks for biomarker-driven therapy recommendations.
- Laboratory readiness: WGS adoption depends on sequencing capacity, bioinformatics pipelines, quality controls, and clinically validated reporting, as well as the ability to integrate HRD scores alongside other genomic findings in unified reports.
- Turnaround and cost: Health systems would need predictable turnaround times and clear total-cost frameworks for WGS compared with panel tests, including consideration of how a single whole-genome assay might replace multiple targeted tests over the course of a patient’s disease.
Regulatory and reimbursement landscape
Because HRD status can directly determine access to specific drugs, any WGS-based algorithm that informs treatment would sit within a tightly regulated environment and be scrutinized by payers and regulators.
- Clinical laboratory standards: Tests used to guide treatment decisions must be performed in certified laboratories with robust analytical validation, in line with federal quality requirements for clinical labs such as those set out under the Clinical Laboratory Improvement Amendments.
- Companion diagnostics: Where drug labels require a specific test, FDA-reviewed companion diagnostics or equivalent validated approaches may be necessary in practice, influencing how quickly new algorithms can be adopted outside research settings.
- Coverage: Payment policies for next-generation sequencing in oncology vary by payer and indication, emphasizing the importance of clear clinical utility and outcomes data to justify WGS-based HRD testing, especially when used beyond cancers where PARP inhibitors are already standard.
- Data governance: Broader WGS use heightens requirements for patient consent, data privacy, and secure data sharing within health systems, particularly as whole-genome data are repurposed for research, quality improvement, or algorithm refinement.
Equity and access considerations
The move from panel tests to WGS-based decision tools risks widening existing gaps in cancer care unless questions of access and representation are addressed in parallel.
- Availability of WGS: Access can differ by geography and health system resources, potentially widening disparities if comprehensive genomic profiling is concentrated in academic and urban centers.
- Representativeness: Ensuring diverse tumor types and patient populations in validation cohorts is essential for equitable performance, so that HRD predictions hold across ancestries, care settings, and socioeconomic groups.
- Financial barriers: Out-of-pocket costs and prior authorization processes can limit uptake even when tests demonstrate clinical value, making payer policies a determining factor in real-world availability.
Methodological cautions for decision-makers
For policymakers, guideline committees, and institutional review bodies, the study underscores that analytic innovation must be matched by operational rigor.
- Tumor purity and sample quality can influence WGS signal detection and should be controlled in routine workflows, including clear pre-analytic standards for tissue handling and sequencing depth.
- False positives/negatives remain possible; confirmatory evidence from treatment response and longitudinal outcomes is needed before the assay is used as the sole arbiter of eligibility for high-cost targeted therapies.
- Standardized definitions of HRD and cross-platform calibration are important to compare assays fairly, particularly when health systems are weighing procurement decisions between competing commercial offerings and in-house tools.
Timeline and next evidence steps
The current publication represents a technical and translational milestone rather than an immediate practice change. The authors and collaborating institutions outline a path that will increasingly involve regulators, payers, and professional societies.
- Jan. 12: Study publication outlines training, validation, and comparison of the WGS-based HRD algorithm, establishing an evidence base for future clinical trials.
- Planned studies: The team indicates intent to launch larger investigations to evaluate the algorithm as a general decision-support tool across cancers, including in tumor types where HRD testing is not yet routine.
- Key milestones to watch:
- Prospective trials linking WGS-HRD calls to PARP inhibitor and platinum outcomes, with predefined endpoints such as progression-free survival and response rate.
- Head-to-head evaluations against commercial assays with predefined clinical endpoints, enabling payers and health systems to compare performance and value.
- Health economic analyses addressing cost-effectiveness and budget impact for payers and hospitals, including scenarios where WGS replaces or complements existing targeted panels.
