The Journal of Pathology: Clinical Research (Jan 2022)

Selection of endometrial carcinomas for p53 immunohistochemistry based on nuclear features

  • Eun Young Kang,
  • Nicholas JP Wiebe,
  • Christa Aubrey,
  • Cheng‐Han Lee,
  • Michael S Anglesio,
  • Derek Tilley,
  • Prafull Ghatage,
  • Gregg S Nelson,
  • Sandra Lee,
  • Martin Köbel

DOI
https://doi.org/10.1002/cjp2.243
Journal volume & issue
Vol. 8, no. 1
pp. 19 – 32

Abstract

Read online

Abstract The World Health Organization endorses molecular subclassification of endometrial endometrioid carcinomas (EECs). Our objectives were to test the sensitivity of tumor morphology in capturing p53 abnormal (p53abn) cases and to model the impact of p53abn on changes to ESGO/ESTRO/ESP (European Society of Gynaecological Oncology/European Society for Radiotherapy and Oncology/European Society of Pathology) risk stratification. A total of 292 consecutive endometrial carcinoma resections received at Foothills Medical Centre, Calgary, Canada (2019–2021) were retrieved and assigned to ESGO risk groups with and without p53 status. Three pathologists reviewed the representative H&E‐stained slides, predicted the p53 status, and indicated whether p53 immunohistochemistry (IHC) would be ordered. Population‐based survival for endometrial carcinomas diagnosed during 2008–2016 in Alberta was obtained from the Alberta Cancer Registry. The cohort consisted mostly of grade 1/2 endometrioid carcinomas (EEC1/2; N = 218, 74.6%). One hundred and fifty‐two EEC1/2 (52.1% overall) were stage IA and 147 (50.3%) were low risk by ESGO. The overall prevalence of p53abn and subclonal p53 was 14.5 and 8.3%, respectively. The average sensitivity of predicting p53abn among observers was 83.6%. Observers requested p53 IHC for 39.4% with 98.5% sensitivity to detect p53abn (99.6% negative predictive value). Nuclear features including smudged chromatin, pleomorphism, atypical mitoses, and tumor giant cells accurately predicted p53abn. In 7/292 (2.4%), p53abn upgraded ESGO risk groups (2 to intermediate risk, 5 to high risk). EEC1/2/stage IA patients had an excellent disease‐specific 5‐year survival of 98.5%. Pathologists can select cases for p53 testing with high sensitivity and low risk of false negativity. Molecular characterization of endometrial carcinomas has great potential to refine ESGO risk classification for a small subset but offers little value for approximately half of endometrial carcinomas, namely, EEC1/2/stage IA cases.

Keywords