Endometrial Cancer: Molecular Types, Risk and Radiotherapy

Endometrial cancer is no longer staged and treated on anatomy and histology alone. A three-test molecular panel — POLE sequencing, mismatch repair (MMR) immunohistochemistry and p53 immunohistochemistry — sorts every tumour into one of four types, and the type can change the risk group, the FIGO stage, and the radiotherapy decision.

4
molecular types
3
tests — POLE sequencing, MMR IHC, p53 IHC
FIGO 2023
staging integrates molecular type
ESGO 2025
ESGO–ESTRO–ESP guideline update

The four molecular types

POLEmut — polymerase epsilon ultramutated (about 5–10%)

  • Pathogenic POLE exonuclease-domain mutation → ultramutated tumour
  • Often looks high-grade under the microscope, behaves indolently — very high neoantigen load, strong immune response
  • Best prognosis of the four types

MMRd — mismatch repair deficient (about 25–30%)

  • Loss of MLH1, MSH2, MSH6 or PMS2 — sporadic MLH1 methylation, or germline Lynch syndrome
  • Hypermutated; intermediate prognosis
  • The immunotherapy-responsive type in advanced disease
  • Every MMRd result → Lynch pathway: germline testing and cascade screening

NSMP — no specific molecular profile (about 40–50%)

  • Copy-number-low remainder; mostly endometrioid
  • The most heterogeneous group — the 2025 guidelines split it by grade and oestrogen receptor (ER) status
  • Low-grade ER-positive → favourable end · High-grade or ER-negative → behaves as high-risk
  • CTNNB1 mutation and L1CAM overexpression mark the worse end (used in the PORTEC-4a profile)

p53abn — p53 abnormal (about 15–25%)

  • Copy-number-high group — almost all serous carcinomas, plus about a quarter of high-grade endometrioid cancers
  • Worst prognosis; failure is predominantly distant
  • Test HER2 — a subset is HER2-amplified and trastuzumab-responsive in advanced disease

Multiple-classifier tumours (~3–5%)

  • Carry more than one classifying feature — classify by the driver
  • POLEmut or MMRd with secondary abnormal p53 staining behaves as POLEmut/MMRd, not p53abn — do not escalate on the p53 result alone
~98%
POLEmut — 5-year recurrence-free survival (high-risk PORTEC-3 cohort)
~72–74%
MMRd — 5-year recurrence-free survival
~74–76%
NSMP — 5-year recurrence-free survival
~48–50%
p53abn — 5-year recurrence-free survival

How the molecular type moves the risk group

Low <8%
estimated 5-year recurrence risk (ESGO–ESTRO–ESP 2025)
Intermediate 8–14%
estimated 5-year recurrence risk
High-intermediate 15–24%
estimated 5-year recurrence risk
High ≥25%
estimated 5-year recurrence risk
  • Stage I–II POLEmut → low risk, whatever the grade or histotype — FIGO 2023 stage IAm POLEmut
  • p53abn with myometrial invasion → high risk, even confined to the uterus — FIGO 2023 stage IICm p53abn
  • MMRd and NSMP → allocated by stage, grade, LVSI — and, for NSMP, ER status
  • Worked example: one stage IB, grade 3 endometrioid cancer — POLEmut = low risk (no adjuvant treatment) · NSMP = high-intermediate · p53abn = high risk (chemoradiation). Same stage, same grade — three different treatments.

Implications for radiotherapy

  • POLEmut, stage I–II → omit adjuvant treatment. 97–100% recurrence-free survival in both PORTEC-3 arms — chemotherapy and radiotherapy added nothing measurable. De-escalation here is evidence-based, not undertreatment. (Stage III POLEmut: data thin; RAINBO-BLUE is testing omission.)
  • Low risk (any type) → no adjuvant radiotherapy.
  • Intermediate risk → vaginal brachytherapy (e.g. HDR 21 Gy in 3 fractions), or observation in selected patients. Pelvic EBRT is overtreatment at this level of risk.

High-intermediate risk → the molecular-integrated profile decides (PORTEC-4a, 569 women):

  • Favourable (POLEmut, or NSMP CTNNB1 wild-type) → observation
  • Intermediate (MMRd, or NSMP CTNNB1-mutant) → vaginal brachytherapy
  • Unfavourable (p53abn, substantial LVSI, or L1CAM overexpression) → pelvic EBRT
  • Vaginal recurrence 4.5% vs 1.6% vs standard brachytherapy — within the trial’s non-inferiority conclusion; no significant difference in recurrence-free or overall survival
46%
of high-intermediate patients had a favourable profile and safely received no adjuvant treatment (PORTEC-4a)
~8% vs >30%
5-year locoregional recurrence in the unfavourable profile — pelvic EBRT vs vaginal brachytherapy
  • High risk, especially p53abn → chemoradiation, not radiotherapy alone (PORTEC-3) — see the tiles below
  • MMRd → chemotherapy added nothing (5-year RFS 68% chemoradiation vs 76% radiotherapy alone). The systemic lever for MMRd is immunotherapy; radiotherapy decisions rest on stage and LVSI.
  • NSMP → in between. A non-significant trend favouring chemoradiation, which the 10-year PORTEC-3 analysis suggests is modulated by ER status.
59% vs 36%
p53abn 5-year recurrence-free survival — chemoradiation vs radiotherapy alone (PORTEC-3)
52.7% vs 36.6%
p53abn 10-year overall survival — chemoradiation vs radiotherapy alone (PORTEC-3)

Practical points

  • Order all three tests at diagnosis (POLE sequencing covering the 11 pathogenic exonuclease-domain variants, MMR IHC, p53 IHC). The 2025 guidelines allow POLE testing to be omitted only in low-risk stage I disease where the result could not change the adjuvant decision.
  • Report the FIGO 2023 stage with the molecular annotation (m) — IAm POLEmut and IICm p53abn are staging categories in their own right.
  • MMRd → Lynch pathway; p53abn/serous → HER2 testing.
  • Counselling point for de-escalation: in PORTEC-4a, the small excess of vaginal recurrences in untreated favourable-profile patients were mostly isolated vault recurrences, which are usually salvageable.
  • Fertility-sparing and ovarian-preservation decisions are also refined by molecular type in the 2025 guidelines — see the guideline algorithms.

Where this is heading: the RAINBO programme

Four international trials testing class-directed adjuvant treatment in high-risk endometrial cancer (results expected from 2028):

  • p53abn-RED: chemoradiation + 2 years of olaparib vs chemoradiation alone
  • MMRd-GREEN: radiotherapy + concurrent and adjuvant durvalumab (1 year) vs radiotherapy alone
  • NSMP-ORANGE: radiotherapy + 2 years of progestin vs chemoradiation (ER-positive disease — a de-escalation question)
  • POLEmut-BLUE: phase II de-escalation — no adjuvant therapy for lower-risk, none or radiotherapy alone for higher-risk POLEmut disease

References

  1. Cancer Genome Atlas Research Network. Integrated genomic characterization of endometrial carcinoma. Nature 2013;497:67–73.
  2. León-Castillo A, de Boer SM, Powell ME, et al. Molecular classification of the PORTEC-3 trial for high-risk endometrial cancer: impact on prognosis and benefit from adjuvant therapy. J Clin Oncol 2020;38:3388–3397. doi:10.1200/JCO.20.00549
  3. PORTEC-3 investigators. Adjuvant chemoradiotherapy versus radiotherapy alone in women with high-risk endometrial cancer (PORTEC-3): 10-year clinical outcomes and post-hoc analysis by molecular class. Lancet Oncol 2025;26(10).
  4. van den Heerik ASVM, et al. Molecular profile-based adjuvant treatment for women with high-intermediate risk endometrial cancer (PORTEC-4a): a randomised, phase 3, non-inferiority trial. Lancet Oncol 2026;27(1):23–35.
  5. Concin N, et al. ESGO–ESTRO–ESP guidelines for the management of patients with endometrial carcinoma — update 2025. Lancet Oncol 2025;26:e423–e435.
  6. Berek JS, et al. FIGO staging of endometrial cancer: 2023. Int J Gynaecol Obstet 2023;162:383–394.
  7. RAINBO research programme (Refining Adjuvant treatment IN endometrial cancer Based On molecular profile). ClinicalTrials.gov NCT05255653.

Summary prepared for clinician readers; not a substitute for the full guidelines or multidisciplinary discussion.