How we approached the harmonization of TMB | Friends of Cancer Research

How we approached the harmonization of TMB

How we approached the harmonization of TMB

In the first post of this series, we covered why Friends of Cancer Research (Friends) decided to create the Tumor Mutational Burden (TMB) Harmonization Project. Due to a previous lack of understanding of the different ways to measure TMB and apply it as a biomarker for treatment determination, harmonizing definitions of TMB was critical in giving patients access to the best possible care.


That post covered the why of the project, but this week we are covering how the TMB Harmonization Project approached aligning TMB panels. To begin, Friends convened a working group consisting of 15 laboratories with TMB panels, as well as key leaders in the FDA, NCI, and industry to discuss how to align panel TMB for clinical use. The team quickly realized that before the clinical applications of TMB could be understood, they first had to understand how the panels themselves worked. So, the project was divided into phases. In Phase I, the project team analyzed publicly available data from The Cancer Genome Atlas (TCGA), looking at the entire genome through whole exome sequencing (WES) to estimate TMB. This served as the gold standard. All other panels used the same data to estimate TMB and compared it to the gold standard to understand how the panel TMB values differed from the WES standard and from one another.

The Friends of Cancer Research TMB Harmonization Project is an outstanding example of how collaboration among stakeholders can facilitate efficient advancement of quality, reliable diagnostics to the clinic.”   - Lisa McShane, National Cancer Institute

Only until the group understood how each panel compared to the WES gold standard could they proceed to assess the clinical implications of TMB. Phase II of the project applied the same technique as Phase I but in a clinically applicable setting, with cells derived from human tumors. This is how TMB would be measured in an oncology clinic, using samples taken directly from patients. Thus, the results of this project will most directly benefit patients since by using TMB as a biomarker a patient’s treatment decision can be better tailored.


This project was a collaborative effort between numerous stakeholders. Friends was key in promoting collaboration across laboratories to understand how the panels detected mutations and estimated TMB and facilitating agreement on analytical standards that would ensure consistent results across panels. Moreover, this work determined a standard reference that would help determine how a panel’s TMB compared to another panel’s TMB, ensuring that patients are receiving accurate information and treatment.


TMB as a biomarker is also beginning to redefine how treatments are prescribed to patients. Tune in next week to find out how the results from Phase III of the project helped inform the use of TMB as a tissue-agnostic biomarker, and how this new application of TMB is starting to shift the paradigm on how treatments are determined.

Return to blog
Back to Top