Representative Sequencing: Unbiased Sampling of Solid Tumor Tissue

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Thousands of solid tumors are sequenced, but current methods have inherent under-sampling bias.
• Fixed-dimension tissue samples (e.g., 6 mm biopsy) are under-powered as tumor volume increases.
• This bias leads to misclassification of tumor characteristics like tumor mutation burden (TMB).

Purpose of the Study:

• To introduce and validate Representative Sequencing (Rep-Seq) as a novel method for unbiased tumor tissue sampling.
• To assess the impact of Rep-Seq on the accuracy of intratumor TMB variability analysis.
• To demonstrate the potential of Rep-Seq for improved clinical utility in cancer research.

Main Methods:

• Rep-Seq involves homogenizing fixed residual tumor material for next-generation sequencing.
• Analysis focused on intratumor TMB variability and misclassification rates compared to single-biopsy methods.
• Evaluated misclassification in lung and bladder tumors using both current and Rep-Seq methodologies.

Main Results:

• Current single-biopsy methods misclassify TMB in a significant percentage of tumors (20% lung, 52% bladder).
• Rep-Seq dramatically reduced misclassification rates to 2% for lung and 4% for bladder tumors.
• Demonstrated high intratumor TMB variability that is poorly captured by conventional biopsies.

Conclusions:

• Rep-Seq provides a more representative sampling method for comprehensive tumor profiling.
• This improved methodology significantly enhances the accuracy of TMB assessment.
• Rep-Seq holds substantial potential for advancing clinical utility and precise clonal structure deconvolution in oncology.