Design: Retrospective single-centre review of consecutive requests
Population: 371 requests for comprehensive genomic profiling in Japan
Primary outcome: Assay quality outcome: cancelled, suspended, qualified or pass
Effect: 14.0% cancelled; 86.2% pass and 13.8% qualified; low tumour content the only independent predictor of failure

A Japanese cancer genomics centre reviewed 371 consecutive requests for a comprehensive genomic profiling panel on formalin-fixed tissue. Pathologists checked tumour area, the number of sections available and residual block volume before submission.

52 requests (14%) were cancelled before submission, mostly for too little tissue. Of 297 reports returned, 86% passed full quality control and 14% came back qualified. Tumour content was lower in the qualified cases on every measure, and in a multivariable model it was the only independent predictor of quality-control failure.

Qualified reports were of limited use for immunotherapy decisions: tumour mutational burden could be assessed in about a third and microsatellite instability in about a fifth.

This is a single-centre retrospective series. Still, the message transfers: the pathologist's assessment of tumour content and block selection decides much of whether a costly test returns an actionable result.

Estimate tumour content on the H&E before sending a block for genomic profiling.
Choose the block with the highest tumour content, not the most convenient one.
Consider macrodissection when tumour content is borderline.
Warn oncologists that low-content samples may not yield TMB or MSI results.
Keep residual tissue in mind before cutting extra sections for other tests.

Why it matters

Block selection by the pathologist is a cheap step that decides whether an expensive test returns usable results.

The statistics, in plain English

Only 31.7% of qualified reports gave a TMB result and 19.5% an MSI result, so a qualified report frequently cannot answer the immunotherapy question it was ordered for.