Most genomic profiling failures traced to too little tumour in the block

Design

Retrospective single-centre review of consecutive requests

Population

371 requests for comprehensive genomic profiling on FFPE tissue in Japan

Primary outcome

Assay cancellation, suspension, qualified or pass status

Effect

14.0% cancelled; 13.8% of results qualified; low tumour content the only independent predictor

A Japanese cancer genomic medicine hospital reviewed 371 consecutive requests for a commercial comprehensive genomic profiling assay (FoundationOne CDx) on formalin-fixed paraffin-embedded tissue.

Fifty-two requests (14.0%) were cancelled before submission, mostly for insufficient specimen. Of 319 submitted, 22 (6.9%) were temporarily suspended. Of 297 with results, 86.2% passed and 13.8% were only qualified. Lower tumour content — whether estimated by the hospital pathologist, the laboratory or sequencing — was the only independent predictor of quality failure. In qualified cases, tumour mutational burden could be assessed in only 31.7% and microsatellite instability in 19.5%.

The practical lesson is that the pathologist's block selection and tumour content estimate largely determine whether an expensive test returns a usable answer. This is single-centre and retrospective, but the pattern is familiar.

• Choose the block with the highest tumour content for genomic profiling, not simply the most recent one.
• Estimate and record tumour cellularity on the marked H&E for every molecular request.
• Consider macrodissection or a different block when tumour content is borderline.
• Warn clinicians that a 'qualified' result may lack tumour mutational burden and microsatellite status.
• Keep a log of cancelled and failed requests to audit block selection.

Why it matters

Where these tests are paid for privately, as often in India, a failed run is a real cost to the patient.

Don't overread it

This is a single-institution retrospective review of one assay; failure rates elsewhere may differ.

The statistics, in plain English

Of every hundred requests, about 14 never went to the laboratory and, of those reported, about 14 came back with reduced reliability. 'Independent predictor' means tumour content still mattered after accounting for the other factors measured.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.