Improving Accuracy of Somatic Mutation Profiling in Large Epidemiologic Studies: Addressing Cases without Matched...

Area of Science:

• Genomics
• Cancer Research
• Bioinformatics

Background:

• Accurate somatic mutation detection is crucial for cancer research and personalized medicine.
• Patient-matched normal DNA is ideal for distinguishing somatic from germline mutations.
• Archival tissue studies often lack matched normal samples, posing challenges for mutation analysis.

Purpose of the Study:

• To develop and validate a protocol for improved somatic mutation detection in tumor samples without matched normal DNA.
• To enhance the accuracy of mutation calling in large-scale genomic studies, such as the InterMEL study.

Main Methods:

• Developed a novel bioinformatic approach utilizing variant allele frequency (VAF) and tumor purity estimation from tumor-only samples.
• Leveraged the principle that germline mutations typically present at ~50% VAF, while somatic mutations in impure tumors have lower VAFs.
• Validated the method on 137 melanoma samples from the InterMEL Study with matched normal tissue as a gold standard.

Main Results:

• The conventional pipeline using unmatched normal samples exhibited a 15.6% false positive rate and a 3.5% false negative rate.
• The new technique significantly reduced error rates, achieving a 6.4% false positive rate and a 2.1% false negative rate.
• Demonstrated improved distinction between somatic and germline variants in tumor-only samples.

Conclusions:

• The proposed method offers a robust solution for accurate somatic mutation detection in the absence of matched normal samples.
• This approach enhances the reliability of genomic profiling in large retrospective and multi-institutional cancer studies.
• The improved accuracy has significant implications for cancer genomics research and clinical applications.