Quality Control Failure in a Clinical Exome Laboratory: Impact on Patient Care
A clinical exome sequencing laboratory has a monthly quality control (QC) review. They notice that the average coverage for a batch of 50 exomes is lower than usual: mean coverage 80x (previously 100x), and the percentage of bases covered at >20x is 92% (previously 95%). The laboratory director investigates and finds that a new lot of capture probes was used. The manufacturer's certificate of analysis claims equivalent performance, but the lab's internal validation for the new lot was not performed. The lab has a policy to validate each new lot of reagents. The affected batch includes samples from 10 patients, including one with a suspected Mendelian disorder where the exome identified a likely pathogenic variant in a gene with low coverage in that sample. The variant was confirmed by Sanger. The lab is CAP-accredited and CLIA-certified. The laboratory director must decide whether to report the results, repeat the test, or take other action.
1. What are the regulatory and quality management implications of using an unvalidated reagent lot? What should the laboratory director do immediately?
Read the worked answer
Using an unvalidated reagent lot violates CAP/CLIA regulations and the laboratory's own quality management procedures. CAP requires that the laboratory validate new reagent lots before use, especially for critical reagents like capture probes that affect test performance. The implications include:
- Regulatory non-compliance: Could lead to citations, loss of accreditation, or sanctions.
- Patient safety risk: Lower coverage may lead to missed variants, especially in GC-rich regions or genes with low coverage.
- Ethical and legal risk: If a patient is harmed due to a missed diagnosis, the laboratory could be liable.
Immediate actions:
- Stop reporting results from the affected batch until further review.
- Notify the laboratory director and quality manager.
- Assess the impact: Review the coverage data for each patient, especially for genes relevant to their phenotype. For the patient with the likely pathogenic variant, ensure that the variant is confirmed by Sanger (already done) and that other relevant genes had adequate coverage.
- Repeat testing for patients with inadequate coverage or where the clinical question requires higher confidence. Consider re-sequencing with a validated lot.
- Perform validation of the new lot retrospectively to determine if it meets specifications. If not, switch back to a validated lot.
- Document the event and perform a root cause analysis. Implement corrective actions: retraining staff, updating SOPs, and ensuring lot validation is done.
- Communicate with clinicians about the potential need for repeat testing and the impact on patient care.
- Report the event to CAP as required (if applicable).
The director should err on the side of caution and repeat testing for any patient where coverage was insufficient, especially if it could affect diagnosis.
2. How should the laboratory handle the specific patient with the likely pathogenic variant? What additional steps are needed before reporting?
Read the worked answer
For the patient with the likely pathogenic variant, the laboratory must ensure that the variant is true and that no other clinically relevant variants were missed due to low coverage. Steps:
- Confirm the variant: Already done by Sanger, so the variant is real.
- Assess coverage of the gene: Check the coverage of the gene containing the variant; if it was low, the variant might have been missed if it were in a different region. But since it was found, coverage was adequate at that locus. However, other genes may have had low coverage.
- Evaluate other genes: Review the coverage of all genes relevant to the patient's phenotype. If any key genes had <20x coverage, consider re-sequencing or targeted testing.
- Consider re-sequencing the sample with a validated lot to ensure comprehensive coverage. If the variant is confirmed and other genes are adequately covered, the result can be reported with a comment about the lot issue? Actually, better to repeat to ensure full coverage.
- Report the variant with appropriate caveats if re-sequencing is not possible. But the laboratory should aim to provide a complete and accurate result.
- Document the QC issue in the report? Not typically, but the laboratory should have a policy for reporting with deviations.
The director should decide to repeat the exome for this patient using a validated lot to ensure no other variants are missed. If repeat is not feasible, targeted testing of other genes of interest could be done. The variant can be reported as likely pathogenic, but the report should note that the exome had lower coverage and that additional testing may be considered. However, best practice is to repeat.