Personalized genomic analyses for cancer mutation discovery and interpretation

Area of Science:

• Genomics
• Cancer Biology
• Personalized Medicine

Background:

• Massively parallel sequencing is increasingly used for tumor characterization and therapy selection.
• A key question is whether tumor tissue alone is sufficient or if matched normal DNA is needed to identify somatic alterations.

Purpose of the Study:

• To evaluate the necessity of matched normal DNA in comprehensive genomic profiling of tumors.
• To assess the accuracy and clinical utility of tumor-only versus tumor-normal sequencing.

Main Methods:

• Whole exome or targeted gene sequencing of 815 tumor-normal paired samples across 15 cancer types.
• Validation of genomic alterations with high sensitivity and specificity.
• Analysis of somatic mutations and germline alterations in cancer-predisposing genes.

Main Results:

• An average of 140 somatic mutations per exome and 4.3 per targeted analysis were identified.
• Over 75% of cases showed somatic alterations linked to therapies or clinical trials.
• Matched normal DNA analysis revealed germline mutations in 3% of sporadic cancer patients and identified false positives in tumor-only approaches.

Conclusions:

• Matched tumor-normal sequencing is crucial for accurate identification and interpretation of somatic and germline alterations.
• This approach has significant implications for cancer patient diagnosis and therapeutic management.
• Tumor-only sequencing can lead to misinterpretation of actionable alterations.