Cell blocks, not smears, are the safer cytology material for hybrid-capture sequencing

Design

Retrospective single-laboratory analysis of sequencing quality

Population

455 cytology specimens and 2,759 FFPE controls, within 10,900 sequenced specimens

Primary outcome

Percentage of target bases covered at 100× or more

Effect

Cell blocks median 98% or more, similar to FFPE; Papanicolaou smears 74.7%; combined preparations 89.2%

A retrospective analysis covered 10,900 specimens sequenced on one custom hybrid-capture panel. It compared 455 consecutive cytology specimens with 2,759 FFPE surgical controls. Preparations were Diff-Quik and Papanicolaou smears, cell-block sections and FFPE tissue. Quality was the percentage of target bases covered at 100× or more.

Cell blocks performed like FFPE tissue, with median target coverage of 98% or more. Smears gave DNA of adequate quantity and quality, but coverage was less uniform: median 74.7% for Papanicolaou smears and 89.2% for combined preparations. GC bias and pairwise variability were also higher (P < 2.2 × 10⁻¹⁶). The differences did not trace to DNA input or tumour purity but to the way smears are prepared and stained.

This is one laboratory's panel, so thresholds may not transfer. It argues for validating smears separately, with cytology-specific normalisation, and for making a cell block when material allows.

• Make a cell block whenever the cytology specimen is the only tissue and sequencing may follow.
• Validate smear-based sequencing as its own specimen type rather than reusing FFPE cut-offs.
• Record preparation and stain type with the specimen so quality problems can be traced.
• Interpret low-coverage regions on smear results with caution before reporting a negative.
• Discuss with your molecular laboratory what each specimen type has been validated for.

Why it matters

Cytology is often all that exists for advanced cancer, so assuming it behaves like tissue risks false negative sequencing results.

Don't overread it

The data are retrospective and from one laboratory and one panel, so the coverage figures will differ elsewhere.

The statistics, in plain English

With thousands of specimens, even small differences get extremely small P values. The meaningful number is the coverage gap: about 75% of targets on Papanicolaou smears against 98% or more for cell blocks, which means more of the panel may be unreadable on smears.

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.