Nearly half of CIN2 diagnoses change on expert review

Design

historical cohort study with blinded independent regrading by three expert pathologists using haematoxylin-eosin and p16, modified Poisson regression

Population

437 women aged 23-40 under active surveillance for CIN2, Aarhus University Hospital 2000-2010, median age 27, followed 28 months

Primary outcome

subsequent record of CIN3 or worse by expert-verified histological grade

Effect

expert review upgraded 12.8% and downgraded 27.5%; CIN3+ in 22.5% (expert CIN1/normal), 42.2% (CIN2) and 64.3% (CIN3); adjusted RR 1.37 (1.07-1.75)

Active surveillance for CIN2 in younger women rests on high regression rates and on the preterm birth risk that follows excision. It also rests on a histological diagnosis known to be poorly reproducible. This historical cohort took 437 Danish women aged 23 to 40 diagnosed with CIN2 at Aarhus University Hospital between 2000 and 2010 and had three expert pathologists independently regrade the original slides with p16 staining, then followed the women for 28 months.

Expert review moved 40% of diagnoses: 12.8% up to CIN3, 27.5% down to CIN1 or normal, with 59.7% confirmed. Subsequent CIN3 or worse was recorded in 39.6% overall, but that headline conceals a gradient — 22.5% where experts said CIN1 or normal, 42.2% where they confirmed CIN2, and 64.3% where they said CIN3 (adjusted relative risk 1.37, 95% CI 1.07-1.75 versus confirmed CIN2). The gradient was steeper in women aged 31 to 40 and in those with high-grade cytology. Crucially, women with high-grade index cytology or HPV16 remained at high risk whatever the expert said about the histology.

That last finding is the practical one. Much of what looks like progression under active surveillance is initial misclassification, so the answer is not to abandon surveillance but to stop deciding on histology alone. Cytology grade and HPV genotype carry independent information, and a woman with CIN2, high-grade cytology and HPV16 is a different proposition from a woman with CIN2 and low-grade cytology — even though the biopsy report reads the same.

• Do not select for active surveillance on the histology report alone
• Weight high-grade index cytology and HPV16 positivity heavily — they predicted risk regardless of expert grade
• Request p16 staining and, where the decision is close, a second pathologist's opinion
• Explain to the woman that CIN2 is a reproducibility problem as much as a biological one
• Age matters: the gradient was steeper in women aged 31 to 40, where regression is less likely

The statistics, in plain English

An adjusted relative risk of 1.37 (95% CI 1.07-1.75) for expert-verified CIN3 against confirmed CIN2 is a modest and imprecise estimate; the raw progression rates — 22.5%, 42.2% and 64.3% across the three expert grades — carry the message more clearly than the regression does. Outcomes are recorded diagnoses of CIN3 or worse, which depend on how often each woman was biopsied, and women whose lesions looked worse were probably followed more closely, which inflates the gradient. Slides came from 2000 to 2010 and were reviewed with p16, a stain not routinely used at the time, so the reclassification rate partly measures the improvement in the test rather than pathologist disagreement alone.

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.