pyTMB

pyTMB calculates tumor mutational burden (TMB) scores from next-generation sequencing (NGS) panel data to quantify somatic mutation load as a biomarker for immune checkpoint inhibitor response.

Key Features:

• Customizable NGS panel compatibility: Supports TMB calculation from various in-house and targeted NGS panels and multiple sample types.
• Variant allele frequency (VAF) optimization: Implements VAF cut-offs of 10% for formalin-fixed paraffin-embedded (FFPE) samples and 5% for frozen samples to tailor mutation calling to preservation method.
• Comprehensive TMB calculation with manual curation: Integrates biological manual curation to reclassify tumors as MSI or POLE mutated and to identify true TMB-high cases.
• Comparative algorithm analysis: Enables comparison of the Institut Curie (IC) algorithm outputs with other algorithms such as FoundationOne®CDx to assess methodological differences in TMB estimates.

Scientific Applications:

• Predictive biomarker evaluation: Provides TMB scores to support assessment of likely response to immune checkpoint inhibitors, including pembrolizumab, in unresectable or metastatic tumors.
• Cancer subtype analysis: Identifies high TMB occurrences across cancer types such as lymphoma, lung, endometrial, and cervical cancers for subtype-specific investigations.
• Research and molecular tumor board support: Supplies mutation burden metrics and curated classifications to inform research studies and clinical decision-making in molecular tumor boards.

Methodology:

Sequencing of FFPE or frozen tumor samples with an in-house NGS panel followed by the IC algorithm applying VAF cut-offs (10% FFPE, 5% frozen) and manual biological curation to compute TMB and reclassify MSI/POLE-mutated tumors.