Skip to content
Bending the Curve Trial by Trial
?

Uterine corpus (endometrium; female reproductive system)

Endometrial cancer

How estrogen findings, molecular classification, sentinel nodes and immunotherapy shaped care for endometrial cancer.

Uterine cancer (SEER category) · 5-year relative survival · United States, SEER 8 registries

88.62%→84.25%

1975 (observed value; first year in the table and the highest observed value) to 2018 (observed value; last observed year in the table). Read these with the caveats under “Then and now”.Source

25 milestones, by decade

  • Two case-control papers in the 4 December 1975 issue of the New England Journal of Medicine reported more estrogen use among women with endometrial carcinoma. Smith and colleagues retrospectively compared 317 women with adenocarcinoma of the endometrium with an equal number of matched controls who had other gynecologic neoplasms: 152 patients and 54 controls had used estrogen, and the authors put the risk at 4.5 times greater with exposure. Ziel and Finkle recorded conjugated estrogen use in 57% of 94 patients and 15% of controls from a twofold age-matched series, with a risk-ratio estimate of 7.6 (one-sided 95% lower confidence limit 4.7) that rose with duration of exposure. Both are observational associations. A later randomised trial, PEPI (596 postmenopausal women, 3 years), reported more endometrial hyperplasia with estrogen alone than with placebo and hyperplasia rates similar to placebo with three estrogen-plus-progestin regimens (P = .16); its outcome was endometrial histology, not cancer incidence. All figures come from the abstracts.
  • A 2013 Nature paper characterised 373 endometrial carcinomas by genomic, transcriptomic and proteomic methods. Its abstract reports that uterine serous tumours and about 25% of high-grade endometrioid tumours had extensive copy-number alterations and frequent TP53 mutations, that a subset of endometrioid tumours had a markedly increased transversion mutation frequency with hotspot mutations in POLE, and that the results classified endometrial cancers into four categories: POLE ultramutated, microsatellite instability hypermutated, copy-number low and copy-number high. The authors stated that the reclassification may affect post-surgical adjuvant treatment for women with aggressive tumours. This is a genomic description from the abstract, not a treatment trial.
  • On the SEER Cancer Stat Facts page for uterine cancer, the US age-adjusted death rate for females is 5.28 per 100,000 in 1975, 4.00 in 1997 (the lowest value in the table) and 5.47 in 2024, while observed 5-year relative survival in the SEER 8 series is 88.62% in 1975, 80.50% in 1981 and 84.25% in 2018. The page text reports death rates rising on average 1.3% each year over 2015-2024 and rates of new cases rising on average 0.7% each year over 2014-2023. These are different series (US mortality data and a limited set of registries), the page attributes the changes to no cause, and the 'Uterine Cancer' heading does not limit the figures to endometrial carcinoma.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 25
Treatment type
Setting

Chemotherapy

  1. 1975 · Understanding · UnderstandingTwo 1975 studies link estrogen use to endometrial cancerIn 1975 two studies in the New England Journal of Medicine reported that women with endometrial cancer had used estrogen medicines far more often than comparison women. In 1996 a randomised trial found that estrogen alone raised rates of endometrial overgrowth, while adding a progestin kept rates close to placebo.Case-control studies (N Engl J Med, 4 Dec 1975): Smith, 317 cases vs 317 controls, risk 4.5 times greater with estrogen; Ziel, 94 cases, risk-ratio estimate 7.6, rising from 5.6 to 13.9 with duration. Mack 1976: risk ratio 8.0 (95% CI 3.5-18.1) for any estrogen. PEPI (randomised, 596 women, 3 years): simple hyperplasia 27.7% with estrogen alone vs 0.8% with placebo; estrogen-plus-progestin regimens similar to placebo (P = .16).

Molecular origins

  1. 1980 · Treatment · RadiationAdjuvant pelvic radiotherapy: less recurrence, no survival gain shownAfter surgery for early endometrial cancer, several randomised trials reported from 1980 onwards tested adding radiotherapy to the whole pelvis. It lowered the chance of the cancer returning in the pelvis or vagina, but the trials did not show that women lived longer, and PORTEC-1 recorded more treatment-related complications.Randomised trials of adjuvant pelvic radiotherapy versus no further treatment after surgery for early-stage disease. PORTEC-1 (stage 1; 714 evaluable): 5-year locoregional recurrence 4% vs 14% (p<0.001); 5-year OS 81% vs 85% (p=0.31). GOG-99 (392 eligible): 2-year cumulative incidence of recurrence 3% vs 12%; 4-year survival 92% vs 86% (p=0.557). ASTEC/EN.5 (905): OS HR 1.05 (95% CI 0.75-1.48). Radiotherapy vs control in each.
  2. 1983 · Understanding · UnderstandingBokhman: two pathogenetic types of endometrial carcinomaIn 1983 J. V. Bokhman proposed, from a study of 366 women, that endometrial cancer arises in two different settings. One was linked to obesity and signs of high estrogen and tended to have better-differentiated tumours and better outcomes; the other lacked those features and tended to have poorly differentiated tumours and a less favourable outlook.Single-author prospective study (366 patients). First pathogenetic type (65%): obesity, hyperlipidaemia, hyperestrogenism signs; 82.3% G1-G2, 80.2% progestogen sensitivity, 85.6% 5-year survival. Second type (35%): 62.5% G3, 27.8% pelvic-node metastases, 42.5% progestogen sensitivity, 58.8% 5-year survival. These are the author's figures from one series, not trial results.
  3. 1987 · Understanding · Understanding⚑ needs reviewGOG-33: surgical-pathologic spread in stage I diseaseA Gynecologic Oncology Group study operated on over six hundred women with stage I endometrial cancer and examined how far the disease had actually spread. More than one in five had disease outside the uterus, which the authors took as support for a primary surgical approach. FIGO moved to surgical staging in 1988.GOG surgical-pathologic study (non-randomised; 621 patients with Stage I carcinoma; hysterectomy, bilateral salpingo-oophorectomy, selective pelvic and paraaortic lymphadenectomy, peritoneal cytology): 144 patients (22% as stated in the abstract) had disease outside the uterus; grade and depth of invasion were related to extrauterine disease. FIGO changed endometrial staging from clinical to surgical/histopathological in 1988 (Creasman and Eddy, 1990).

Targeted therapy

  1. 1999 · Treatment · EndocrineHormonal therapy: progestins to letrozole combinationsHormone-based treatment of advanced endometrial cancer began with progestins. A randomised study found oral medroxyprogesterone acetate active, with no benefit from a dose five times higher. Later small phase 2 studies added letrozole to everolimus or palbociclib and reported responses or longer progression-free survival; these are early signals.GOG-81 (299 women): oral MPA 200 vs 1000 mg/day; response 25% vs 15%; median PFS 3.2 vs 2.5 months; median survival 11.1 vs 7.0 months. Phase II everolimus-letrozole (38 enrolled, 35 evaluable): clinical benefit rate 40%, objective response 32%. Randomised phase II PALEO (77 patients): letrozole with palbociclib vs placebo, median PFS 8.3 vs 3.1 months; 12-month landmark HR 0.57 (95% CI 0.32-0.99).
  2. 2004 · Treatment · ChemotherapyChemotherapy for advanced disease: GOG-107, -177 and -209Three Gynecologic Oncology Group trials tested chemotherapy for advanced or recurrent endometrial cancer. Adding cisplatin to doxorubicin shrank more tumours but did not lengthen survival. Adding paclitaxel to doxorubicin and cisplatin lengthened survival but caused more nerve damage. Carboplatin with paclitaxel was later found not to be worse than that three-drug regimen.Randomised phase 3 trials. GOG-107 (299 registered): doxorubicin-cisplatin vs doxorubicin, ORR 42% vs 25%, median OS 9.0 vs 9.2 months. GOG-177 (273 registered): paclitaxel, doxorubicin, cisplatin (TAP) vs doxorubicin-cisplatin, median OS 15.3 vs 12.3 months (P=0.037). GOG-209 (1381 enrolled): carboplatin-paclitaxel vs TAP, noninferior for OS (median 37 vs 41 months; HR 1.002; 90% CI 0.9-1.12).
  3. 2005 · Treatment · ChemotherapyGOG-122: doxorubicin-cisplatin versus whole-abdominal irradiationIn women with stage III or IV endometrial cancer and little cancer left after surgery, two-drug chemotherapy was compared with radiotherapy to the whole abdomen. Women given chemotherapy were predicted to be alive without progression more often at five years (about half versus about four in ten) and to live longer, with more short-term side effects.Randomised phase 3 (422 entered; 396 assessable), stage III or IV disease with at most 2 cm residual: whole-abdominal irradiation (WAI) vs doxorubicin-cisplatin (AP). Stage-adjusted progression HR 0.71 (95% CI 0.55-0.91; p<.01) and death HR 0.68 (0.52-0.89; p<.01), favouring AP. Predicted at 60 months, AP vs WAI: 50% vs 38% alive and disease free; 55% vs 42% alive. Treatment probably contributed to eight deaths (4%) with AP and five (2%) with WAI.
  4. 2008 · Reversal · Surgery⚑ needs reviewSystematic lymphadenectomy: no survival benefit shown in two trialsTwo randomised trials tested adding removal of the pelvic lymph nodes to hysterectomy in women whose cancer seemed to be confined to the uterus. The extra surgery found more nodal spread in one trial, but neither trial showed longer survival with it.Benedetti Panici (514 eligible, preoperative stage I): 5-year DFS 81.0% vs 81.7% and OS 85.9% vs 90.0% (lymphadenectomy vs none); HR for death 1.20 (95% CI 0.70-2.07); nodal metastases found 13.3% vs 3.2%. ASTEC (1,408, ITT): OS HR 1.16 (0.87-1.54), P = 0.31; RFS HR 1.35 (1.06-1.73), P = 0.017 (both above 1 in favour of standard surgery).
  5. 2009 · Treatment · Surgery⚑ needs reviewLAP2 and LACE: laparoscopic vs open surgery for uterine cancerTwo large randomised trials compared keyhole (laparoscopic) surgery with open surgery for uterine cancer. Early reports found fewer complications and shorter hospital stays with keyhole surgery. In the stage I trial, 81 to 82 in 100 women in each group were alive without recurrence or new cancer at 4.5 years. The larger trial fell short of its pre-set recurrence margin.GOG LAP2 (2,616 women, clinical stage I-IIA, 2:1 laparoscopy vs laparotomy): moderate to severe postoperative adverse events 14% vs 21% (P < .0001); recurrence-free interval non-inferiority not met (HR 1.14; 95% upper bound 1.46); 3-year recurrence 11.4% vs 10.2%. LACE (760 women, stage I endometrioid): 4.5-year disease-free survival 81.6% (TLH) vs 81.3% (TAH), equivalence met (7% margin).
  6. 2010 · De-escalation · Radiation⚑ needs reviewPORTEC-2: vaginal brachytherapy in place of pelvic radiotherapyFor women with early-stage endometrial cancer at high-intermediate risk, radiotherapy given only inside the vagina was compared with radiotherapy to the whole pelvis. Vaginal recurrence was rare and similar in both groups, and bowel side effects were less common with vaginal treatment. Ten-year follow-up found more pelvic recurrences after vaginal treatment.Open-label non-inferiority trial (427 women, stage I or IIA, high-intermediate risk): vaginal brachytherapy (VBT) vs pelvic EBRT. Primary endpoint 5-year vaginal recurrence 1.8% (95% CI 0.6-5.9) vs 1.6% (0.5-4.9); HR 0.78 (0.17-3.49); non-inferiority margin 6%. Acute grade 1-2 gastrointestinal toxicity 12.6% vs 53.8%. At 10 years: vaginal recurrence 3.4% vs 2.4% (p=0.55); pelvic recurrence 6.3% vs 0.9% (p=0.004).

Immunotherapy

  1. 2011 · Detection · SurgerySentinel node biopsy tested as an alternative to lymphadenectomyThree prospective studies checked how well sampling only the first draining lymph nodes (the sentinel nodes) finds cancer spread, with removal of many nodes as the reference. In each, most women with lymph-node spread were picked up by the sentinel nodes. None was designed to test survival or recurrence after skipping full node removal.Prospective non-randomised cohorts with lymphadenectomy as the reference and no survival endpoints. SENTI-ENDO (133 enrolled, stage I-II; primary endpoint NPV per hemipelvis 100%, 95% CI 95-100): per patient, sensitivity 84% (62-95), NPV 97% (91-99). FIRES (385 enrolled, indocyanine green): sensitivity 97.2% (85.0-100), NPV 99.6% (97.9-100). SENTOR (156 enrolled, grade 2 endometrioid or high-grade): sensitivity 96% (81-100), NPV 99% (96-100).
  2. 2013 · Understanding · UnderstandingTCGA: four molecular groups of endometrial carcinomaIn 2013 researchers studied the genes of 373 endometrial cancers and sorted them into four groups by their DNA changes. Serous tumours resembled some ovarian and breast cancers. The authors suggested that this could change which women with aggressive tumours receive further treatment after surgery.Integrated genomic, transcriptomic and proteomic study of 373 endometrial carcinomas. Four categories: POLE ultramutated, microsatellite instability hypermutated, copy-number low and copy-number high. Uterine serous tumours and about 25% of high-grade endometrioid tumours had extensive copy-number alterations, few methylation changes and frequent TP53 mutations. The authors reported that the genomic features permit a reclassification that may affect adjuvant treatment of women with aggressive tumours.
  3. 2015 · Understanding · UnderstandingProMisE and PORTEC-3: molecular classes in high-risk diseaseIn 2015 researchers showed that the four genomic groups of endometrial cancer could be approximated with three routine tests on stored tissue. When the tests were applied to PORTEC-3 tissue in 2020, outcomes differed widely between groups, and the apparent benefit of added chemotherapy was clearest in one group, the p53-abnormal tumours.ProMisE (2015): surrogate for the TCGA classes using mismatch repair immunohistochemistry, POLE mutation analysis and p53 immunohistochemistry; tested in an independent cohort of 152 cases and confirmed in 319 further samples (2017). PORTEC-3 tissue (410 classified): 5-year RFS 48% (p53abn), 98% (POLEmut), 72% (MMRd), 74% (NSMP). Subgroup 5-year RFS, chemoradiotherapy vs radiotherapy: p53abn 59% vs 36% (p=0.019); other classes no significant difference.
  4. 2018 · Treatment · TargetedTrastuzumab with chemotherapy in HER2-positive uterine serous cancerIn a small randomised phase 2 trial in advanced or recurrent HER2-positive uterine serous carcinoma, adding trastuzumab to carboplatin and paclitaxel was compared with chemotherapy alone. Cancer took longer to progress with trastuzumab, and an update reported longer survival. The trial was small and used one-sided statistical tests.Randomised phase 2, 61 patients (58 evaluable), HER2 IHC 3+ or 2+/FISH-amplified uterine serous carcinoma; carboplatin-paclitaxel with or without trastuzumab. Median PFS 8.0 vs 12.6 months (HR 0.44; 90% CI 0.26-0.76; one-sided P=0.005). Update: PFS 8.0 vs 12.9 months (HR 0.46); median OS 24.4 vs 29.6 months (HR 0.58; 90% CI 0.34-0.99; P=0.046). One-sided log-rank tests, 90% CIs.
  5. 2018 · Treatment · ChemotherapyPORTEC-3: chemoradiotherapy versus radiotherapy, high-risk diseaseIn women with high-risk endometrial cancer, pelvic radiotherapy alone was compared with radiotherapy plus chemotherapy. At five years survival did not differ significantly, but more women were free of failure with chemotherapy, at the cost of more side effects. A ten-year report found higher survival with chemotherapy.Open-label phase 3 (686 enrolled; 660 eligible): chemoradiotherapy (pelvic RT 48.6 Gy with cisplatin, then carboplatin-paclitaxel) vs pelvic RT alone. Co-primary endpoints, chemoradiotherapy vs RT: 5-year OS 81.8% vs 76.7% (adjusted HR 0.76, 95% CI 0.54-1.06; p=0.11) and FFS 75.5% vs 68.6% (HR 0.71, 0.53-0.95; p=0.022). Grade 3 or worse adverse events during treatment 60% vs 12%. Preplanned 10-year analysis: OS 74.4% vs 67.3% (HR 0.73, 0.54-0.97; p=0.032).
  6. 2019 · Reversal · ChemotherapyGOG-258: adding radiotherapy to chemotherapy did not extend RFSIn stage III or IVA endometrial cancer, chemotherapy plus radiotherapy was compared with chemotherapy alone after surgery. Women given both did not stay free of relapse longer. Radiotherapy lowered recurrence in the vagina and in pelvic and para-aortic lymph nodes, but cancer returned at distant sites more often. Final overall survival results also showed no difference.Randomised phase 3 (813 enrolled; 736 eligible for the primary analysis), stage III or IVA: cisplatin with volume-directed radiation then carboplatin-paclitaxel (chemoradiotherapy) vs carboplatin-paclitaxel alone. 60-month RFS 59% (95% CI 53-65) vs 58% (53-64); HR 0.90 (90% CI 0.74-1.10). 5-year vaginal recurrence 2% vs 7%; pelvic or para-aortic nodal 11% vs 20%; distant 27% vs 21%. Final OS (median follow-up 112 months): HR 1.05 (95% CI 0.82-1.34; p=0.72).
  7. 2019 · Treatment · TargetedLenvatinib plus pembrolizumab after platinum: KEYNOTE-146 and -775After platinum-based chemotherapy, people with advanced endometrial cancer received lenvatinib plus pembrolizumab or doxorubicin or paclitaxel chemotherapy. Cancer took longer to progress and survival was longer with the combination, but severe side effects were more common. A US approval in 2019 rested on an earlier single-group study; the randomised trial was published in 2022.Phase 3, open-label (N=827; 697 pMMR): lenvatinib 20 mg plus pembrolizumab 200 mg vs doxorubicin or paclitaxel after one platinum regimen. pMMR: median PFS 6.6 vs 3.8 months (HR 0.60; 95% CI 0.50-0.72); median OS 17.4 vs 12.0 months (HR 0.68; 0.56-0.84). All-comers OS HR 0.62 (0.51-0.75). Grade 3 or higher adverse events 88.9% vs 72.7%. Earlier single-arm study (N=108): ORR at week 24 38.0%.

Precision & engineered

  1. 2020 · Treatment · ImmunotherapyGARNET: dostarlimab in recurrent dMMR endometrial cancerIn a single-group study, people whose endometrial cancer had a mismatch repair deficiency and had progressed after platinum chemotherapy received dostarlimab alone. About four in ten had a confirmed tumour response, and most responses were lasting at the time of the report. There was no comparison group.Open-label single-group GARNET cohort A1: dostarlimab 500 mg every 3 weeks for 4 doses, then 1000 mg every 6 weeks, in dMMR endometrial cancer after platinum chemotherapy (104 treated; 71 evaluable). Confirmed ORR by blinded central review 42.3% (95% CI 30.6-54.6); 12.7% complete and 29.6% partial responses. Median DOR not reached at 11.2 months median follow-up; response maintained 96.4% at 6 and 76.8% at 12 months.
  2. 2023 · Treatment · ImmunotherapyDUO-E: durvalumab with or without olaparib, first linePeople with advanced or recurrent endometrial cancer received carboplatin and paclitaxel with placebo, with the immune drug durvalumab, or with durvalumab followed by olaparib. Cancer took longer to progress in both durvalumab groups. Survival data were early. In the United States only durvalumab, for mismatch repair-deficient disease, was approved.Phase 3, double-blind, 1:1:1 (N=718): carboplatin-paclitaxel with placebo, durvalumab 1120 mg, or durvalumab plus maintenance olaparib 300 mg. ITT PFS (investigator): durvalumab HR 0.71 (95% CI 0.57-0.89); durvalumab plus olaparib HR 0.55 (0.43-0.69). Exploratory subgroups: dMMR HR 0.42 and 0.41; pMMR HR 0.77 and 0.57. First interim OS (28% maturity): HR 0.77 (0.56-1.07) and 0.59 (0.42-0.83).
  3. 2023 · Understanding · Understanding⚑ needs reviewFIGO 2023: molecular and histological features enter stagingIn 2023 the international gynaecology federation FIGO updated how endometrial cancer is staged. The new system adds tumour type, growth pattern and molecular subtype to how far the cancer has spread, and encourages molecular testing in all endometrial cancers.FIGO staging of endometrial cancer, 2023 (FIGO Women's Cancer Committee subcommittee; update of the 2009 system) adds histological type, tumour pattern and molecular classification (POLEmut, MMRd, NSMP, p53abn) to staging, with new substages such as IA1-IA3 and IIA-IIC. In stage I-II disease, p53abn or POLEmut status changes the stage (IICmp53abn, IAmPOLEmut). This is a consensus classification, not a trial report.
  4. 2023 · Treatment · ImmunotherapyNRG-GY018: pembrolizumab added to chemotherapy, both MMR cohortsPeople with advanced or recurrent endometrial cancer received carboplatin and paclitaxel with either the immune drug pembrolizumab or a placebo. Tumours with and without a mismatch repair deficiency were studied as two separate groups. In both groups cancer took longer to progress with pembrolizumab. Survival results were still immature.Phase 3, double-blind, placebo-controlled, 816 enrolled, analysed as separate dMMR and pMMR cohorts; primary endpoint investigator-assessed PFS. dMMR: 12-month PFS 74% vs 38% (HR 0.30; 95% CI 0.19-0.48). pMMR: median PFS 13.1 vs 8.7 months (HR 0.54; 0.41-0.71). Later overall survival was immature: pMMR HR 0.79 (0.53-1.17); dMMR HR 0.55 (0.25-1.19); one-sided nominal P values.
  5. 2023 · Treatment · ImmunotherapyRUBY: dostarlimab added to carboplatin-paclitaxel, first linePeople with advanced or recurrent endometrial cancer received carboplatin and paclitaxel with either the immune drug dostarlimab or a placebo, then dostarlimab or placebo alone. Cancer took longer to progress with dostarlimab, most clearly in tumours with a mismatch repair deficiency. A later analysis also showed longer survival in the overall group.Phase 3, double-blind, 1:1 (N=494): carboplatin-paclitaxel with dostarlimab 500 mg or placebo for 6 cycles, then dostarlimab 1000 mg or placebo. Investigator-assessed PFS at 24 months: dMMR/MSI-H 61.4% vs 15.7% (HR 0.28; 95% CI 0.16-0.50); overall 36.1% vs 18.1% (HR 0.64; 0.51-0.80). Second interim OS: overall HR 0.69 (0.54-0.89); dMMR/MSI-H 0.32 (0.17-0.63); pMMR/MSS 0.79 (0.60-1.04), subgroups with nominal P values.
  6. 2024 · Reversal · ImmunotherapyKEYNOTE-B21: adjuvant pembrolizumab, no DFS gain overallIn women with newly diagnosed high-risk endometrial cancer, pembrolizumab or a placebo was added to chemotherapy after surgery. Across all women, disease-free survival did not differ. In a planned subgroup of tumours with mismatch repair deficiency the pembrolizumab group fared better, but this was not a formal test and follow-up is early.Randomised double-blind phase 3 (1095 randomised): pembrolizumab 200 mg or placebo every 3 weeks for six cycles with carboplatin-paclitaxel, then 400 mg or placebo every 6 weeks for six cycles. Interim analysis (data cut-off 4 March 2024): DFS HR 1.02 (95% CI 0.79-1.32; p=0.570); 2-year DFS 75% vs 76%. dMMR (n=281) HR 0.31 (0.14-0.69); pMMR (n=814) HR 1.20 (0.91-1.57). Grade 3 or worse adverse events 71% vs 63%.
  7. 2025 · De-escalation · RadiationPORTEC-4a: molecular profile guides adjuvant treatmentWomen with early endometrial cancer at high-intermediate risk were randomly assigned to standard vaginal radiotherapy or to treatment chosen by the tumour's molecular profile: observation, vaginal radiotherapy or pelvic radiotherapy. Vaginal recurrence was low in both groups at five years, and nearly half the women in the profile group were observed without further treatment.Open-label non-inferiority phase 3 (569 enrolled; 564 analysed: 367 molecular-profile, 197 standard vaginal brachytherapy). Favourable profile observed, intermediate given brachytherapy, unfavourable given pelvic RT. 5-year vaginal recurrence 4.5% (95% CI 2.23-6.76) vs 1.6% (0.00-3.32); HR 2.71 (0.79-9.34); upper bound of one-sided CI of the difference 5.3% vs margin 7.0% (p=0.005). Favourable profile: 4.1% vs 0.9%.
  8. 2026 · Today · ImmunotherapyWhere treatment for advanced endometrial cancer stands in 2026For people with advanced or recurrent endometrial cancer, first-line treatment in the United States now combines carboplatin and paclitaxel with an immune checkpoint antibody, then continues the antibody alone. Which antibody and which tumours are covered differs by drug. After platinum, mismatch repair status and HER2 status shape the options.FDA-labelled first line: dostarlimab or pembrolizumab with carboplatin-paclitaxel for primary advanced or recurrent disease (no biomarker restriction in the labels); durvalumab for dMMR disease only; olaparib has no endometrial indication. After prior systemic therapy: pembrolizumab alone for MSI-H/dMMR, dostarlimab alone for dMMR, pembrolizumab plus lenvatinib for pMMR/not MSI-H, and trastuzumab deruxtecan for HER2 IHC 3+ solid tumours.

Then and now

  • 5-year relative survival · United States, SEER 8 registries

    • 88.62%

      1975 (observed value; first year in the table and the highest observed value)

    Females, all races; cancer under the heading 'Uterine Cancer' on the SEER Cancer Stat Facts page (the SEER 8 five-year relative survival series, 1975-2018, all stages together). The page states no age range, stage mix or histology for this series.

    Joinpoint modeled-trend value for 1975: 88.74%. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the yearly value, as opposed to the Joinpoint modeled-trend value in the adjacent column. The column is headed 'Year' and the chart note says only 'SEER 8 5-Year Relative Survival Percent from 1975-2018'; the page does not say the year is the year of diagnosis. SEER 8 covers a limited set of registries. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition. The page does not report how stage mix, histology mix, staging rules or imaging changed over the period and attributes the change to no cause. Read it beside the mortality values. Compare only with other values from this same SEER 8 series, not with the SEER 21 values.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 80.5%

      1981 (observed value; lowest observed value in the table)

    Females, all races; cancer under the heading 'Uterine Cancer' on the SEER Cancer Stat Facts page (the SEER 8 five-year relative survival series, 1975-2018, all stages together). The page states no age range, stage mix or histology for this series.

    Included because the series is not monotonic: it falls from 88.62% (1975) to 80.50% (1981) and later returns to 84.25% (2018); the modeled-trend value for 1981 is 81.03%. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the yearly value, as opposed to the Joinpoint modeled-trend value in the adjacent column. The column is headed 'Year' and the chart note says only 'SEER 8 5-Year Relative Survival Percent from 1975-2018'; the page does not say the year is the year of diagnosis. SEER 8 covers a limited set of registries. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition. The page does not report how stage mix, histology mix, staging rules or imaging changed over the period and attributes the change to no cause. Read it beside the mortality values. Compare only with other values from this same SEER 8 series, not with the SEER 21 values.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 84.25%

      2018 (observed value; last observed year in the table)

    Females, all races; cancer under the heading 'Uterine Cancer' on the SEER Cancer Stat Facts page (the SEER 8 five-year relative survival series, 1975-2018, all stages together). The page states no age range, stage mix or histology for this series.

    Joinpoint modeled-trend value for 2018: 83.61%; the entries for 2019-2023 are modeled-trend values only (83.56% to 83.35%) and are not used. Over 1975-2018 the observed series ranges from 80.50% to 88.62% while the death rate below fell and then rose. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the yearly value, as opposed to the Joinpoint modeled-trend value in the adjacent column. The column is headed 'Year' and the chart note says only 'SEER 8 5-Year Relative Survival Percent from 1975-2018'; the page does not say the year is the year of diagnosis. SEER 8 covers a limited set of registries. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition. The page does not report how stage mix, histology mix, staging rules or imaging changed over the period and attributes the change to no cause. Read it beside the mortality values. Compare only with other values from this same SEER 8 series, not with the SEER 21 values.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 80.9%

      2016-2022

    Females, all races; cancer under the heading 'Uterine Cancer' on the SEER Cancer Stat Facts page; all stages together; the headline figure on the page.

    Different registry set (SEER 21, excluding IL) and a multi-year period, so it is not a continuation of the SEER 8 series and must not be subtracted from it. The page labels the period '2016-2022' and does not say whether it is year of diagnosis. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 94.9%

      2016-2022; localized stage at diagnosis (67% of cases)

    Females, all races; cancer under the heading 'Uterine Cancer' on the SEER Cancer Stat Facts page; localized (confined to primary site) by SEER Combined Summary Stage.

    Stage-specific value from the same SEER 21 table as the 80.9% figure. Stage mix on the page: 67% localized, 18% regional, 11% distant, 4% unstaged. Unstaged cases have a five-year relative survival of 58.3%. Stage at diagnosis, not histology or molecular class, is the only split shown. Different registry set (SEER 21, excluding IL) and a multi-year period, so it is not a continuation of the SEER 8 series and must not be subtracted from it. The page labels the period '2016-2022' and does not say whether it is year of diagnosis. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 70.1%

      2016-2022; regional stage at diagnosis (18% of cases)

    Females, all races; cancer under the heading 'Uterine Cancer' on the SEER Cancer Stat Facts page; regional (spread to regional lymph nodes) by SEER Combined Summary Stage.

    Same SEER 21 table and caveats as the localized value. Different registry set (SEER 21, excluding IL) and a multi-year period, so it is not a continuation of the SEER 8 series and must not be subtracted from it. The page labels the period '2016-2022' and does not say whether it is year of diagnosis. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 19.9%

      2016-2022; distant stage at diagnosis (11% of cases)

    Females, all races; cancer under the heading 'Uterine Cancer' on the SEER Cancer Stat Facts page; distant (cancer has metastasized) by SEER Combined Summary Stage.

    Same SEER 21 table and caveats as the localized value. Different registry set (SEER 21, excluding IL) and a multi-year period, so it is not a continuation of the SEER 8 series and must not be subtracted from it. The page labels the period '2016-2022' and does not say whether it is year of diagnosis. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition.

    Source

  • Age-standardised mortality · United States

    • 5.28per 100,000

      1975 (observed value; first year in the table)

    Females, all races; cancer under the heading 'Uterine Cancer'; age-adjusted death rate from U.S. Mortality data (the 'Death Rate - U.S.' column; the page states that deaths come from U.S. Mortality, not SEER registries).

    Death rates describe the whole US female population and are not measured from the date of diagnosis, so they are not affected by lead time as survival from diagnosis can be; they are shown beside survival for that reason. The page does not say how deaths are assigned across uterine sites or whether the denominator excludes women who have had a hysterectomy, and it attributes the change to no cause. Joinpoint modeled-trend value for 1975: 5.38. The page does not list the histologies or site codes it covers. SEER's separate site recode definition for 'Corpus and Uterus, NOS' lists ICD-O-3 topography C540-C549 for corpus uteri (excluding only histology codes 9050-9055, 9140 and 9590-9993) and C559 for uterus NOS (no histology exclusion listed), so a group defined that way is not limited to endometrial carcinoma (uterine sarcomas and carcinosarcoma would be included); the Stat Facts page does not say that it uses that definition.

    Source

  • Age-standardised mortality · United States

    • 4per 100,000

      1997 (observed value; lowest value in the table)

    Females, all races; cancer under the heading 'Uterine Cancer'; age-adjusted death rate from U.S. Mortality data (the 'Death Rate - U.S.' column; the page states that deaths come from U.S. Mortality, not SEER registries).

    Death rates describe the whole US female population and are not measured from the date of diagnosis, so they are not affected by lead time as survival from diagnosis can be; they are shown beside survival for that reason. The page does not say how deaths are assigned across uterine sites or whether the denominator excludes women who have had a hysterectomy, and it attributes the change to no cause. Included to show that the rate fell and then rose again: observed values are 4.15 in 1995, 4.00 in 1997 and 4.12 in 2005. Modeled-trend value for 1997: 4.07.

    Source

  • Age-standardised mortality · United States

    • 5.47per 100,000

      2024 (observed value; last year in the table)

    Females, all races; cancer under the heading 'Uterine Cancer'; age-adjusted death rate from U.S. Mortality data (the 'Death Rate - U.S.' column; the page states that deaths come from U.S. Mortality, not SEER registries).

    Death rates describe the whole US female population and are not measured from the date of diagnosis, so they are not affected by lead time as survival from diagnosis can be; they are shown beside survival for that reason. The page does not say how deaths are assigned across uterine sites or whether the denominator excludes women who have had a hysterectomy, and it attributes the change to no cause. The page headline gives 5.4 per 100,000 for 2020-2024 deaths; use one or the other, not both. The page text states that age-adjusted death rates have been rising on average 1.3% each year over 2015-2024 and that rates of new cases have been rising on average 0.7% each year over 2014-2023 (statistical models); the highest observed value in the table is 5.56 (2023). Rising mortality sits beside survival that has not risen: show both.

    Source

  • 10-year relative survival · United Kingdom

    • 55.6%

      1970s (as labelled on the page)

    Females diagnosed with uterine cancer in the UK; the page states that almost 6 in 10 (55.6%) survived their disease beyond ten years. Age range and the exact diagnosis period are not stated beside the figure.

    The page does not name the measure (net, relative or observed survival), the data source or the age standardisation; the metric label is the nearest available. The page says survival trends reflect a combination of changes in treatment and stage distribution. Not comparable with the US SEER series; keep the two separate.

    Source

  • 10-year relative survival · United Kingdom

    • 71.7%

      2018 (as labelled on the page; described as predicted)

    Females diagnosed with uterine cancer in the UK; the page states that it is predicted that more than 7 in 10 (71.7%) survive their disease for ten years or more (2018). Aged 15-44 the page gives 80.1% and aged 75-99 it gives 54.8% (2018).

    Described on the page as predicted, so a modelled estimate rather than an observed ten-year follow-up. The page says ten-year survival has increased in all broad age groups since the early 1970s. Same series and measure caveats as the 1970s value.

    Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

Early-stage endometrial cancer, adjuvant radiotherapy and de-escalation

Women after hysterectomy for stage I-II endometrial carcinoma at intermediate or high-intermediate risk: PORTEC-1 (stage I), GOG-99 (stages IB, IC and occult II) and PORTEC-2 (stage I-IIA, high-intermediate risk), and PORTEC-4a (high-intermediate risk, molecular profile-based treatment) · Adjuvant · endpoint: Vaginal recurrence

Results are not comparable across trials.

Risk definitions, eras, and the recurrence measure differ. The axis shows 5-year recurrence percentages. PORTEC-2 and PORTEC-4a report vaginal recurrence as the primary endpoint; PORTEC-1 is plotted with its 5-year locoregional recurrence (a broader measure, so it is drawn with a distinct marker). GOG-99 reports a 2-year cumulative incidence of recurrence and is listed but not plotted. The comparisons also change: PORTEC-1 and GOG-99 compare pelvic radiotherapy with no further treatment; PORTEC-2 compares vaginal brachytherapy with pelvic radiotherapy; PORTEC-4a compares molecular profile-based treatment with vaginal brachytherapy, with non-inferiority margins in the last two. PORTEC-2 also reported higher 5-year locoregional recurrence (5.1% vs 2.1%) and 10-year pelvic recurrence (6.3% vs 0.9%) with vaginal brachytherapy; those values are in the trial record, not on this axis. Compare only each trial's own two arms; the chart is not a pooled or network estimate.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100Vaginal recurrence at 60 months (%)198020102025PORTEC-1PORTEC-2PORTEC-4a
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
Vaginal recurrence at 60 months, rate, across 3 trials in this lane. PORTEC-1 (1980): experimental 4 % vs control 14 %; PORTEC-2 (2010): experimental 1.8 % vs control 1.6 %; PORTEC-4a (2025): experimental 4.5 % vs control 1.6 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
PORTEC-11980Locoregional recurrence (differs from lane)14 %4 %
PORTEC-22010Vaginal recurrence1.6 %1.8 %
PORTEC-4a2025Vaginal recurrence1.6 %4.5 %
GOG-991980Cumulative incidence of recurrence (differs from lane) (not plotted: different measure or time point)12 %3 %

High-risk endometrial cancer, adjuvant chemotherapy, chemoradiotherapy and immunotherapy

Women after surgery for high-risk or stage III-IV endometrial carcinoma: GOG-122 (stage III-IV, at most 2 cm residual disease), PORTEC-3 (high-risk stage I-III), GOG-258 (stage III-IVA) and KEYNOTE-B21 (high-risk, newly diagnosed) · Adjuvant · endpoint: RFS

Results are not comparable across trials.

Populations, eras and control arms differ. GOG-122 compared doxorubicin-cisplatin with whole-abdominal irradiation (its 60-month figures are percentages predicted alive and disease free, adjusted for stage, and its endpoint is labelled progression-free survival). PORTEC-3 compared chemoradiotherapy with pelvic radiotherapy alone and is plotted with failure-free survival. GOG-258 compared chemoradiotherapy with chemotherapy alone and reports relapse-free survival. KEYNOTE-B21 added pembrolizumab or placebo to chemotherapy and reports 2-year disease-free survival from an interim analysis; it is listed but not plotted because its landmark differs, and its mismatch repair subgroup results are not shown here. Endpoint definitions differ, so non-matching entries are drawn with a distinct marker. Compare only each trial's own two arms; the chart is not a pooled or network estimate.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100RFS at 60 months (%)200520182019GOG-122PORTEC-3GOG-258
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
RFS at 60 months, rate, across 3 trials in this lane. GOG-122 (2005): experimental 50 % vs control 38 %; PORTEC-3 (2018): experimental 75.5 % vs control 68.6 %; GOG-258 (2019): experimental 59 % vs control 58 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
GOG-1222005PFS (differs from lane)38 %50 %
PORTEC-32018FFS (differs from lane)68.6 %75.5 %
GOG-2582019RFS58 %59 %
KEYNOTE-B212024DFS (differs from lane) (not plotted: different measure or time point)76 %75 %

Advanced or recurrent endometrial cancer, first-line cytotoxic chemotherapy

Women with stage III, IV or recurrent endometrial carcinoma treated with cytotoxic chemotherapy in three Gynecologic Oncology Group trials: GOG-107, GOG-177 and GOG-209 · Metastatic · endpoint: OS

Results are not comparable across trials.

Three trials share arms in a chain: GOG-107 compared doxorubicin-cisplatin with doxorubicin; GOG-177 compared doxorubicin-cisplatin plus paclitaxel (TAP) with doxorubicin-cisplatin; GOG-209 compared carboplatin-paclitaxel with TAP as a non-inferiority trial, so its lower median is not a deficit in the trial's own terms. Eras, supportive care, imaging, eligibility (prior radiotherapy or hormonal therapy) and the number of patients differ. Medians are as published for each trial. Compare only each trial's own two arms; the chart is not a pooled or network estimate.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

01020304050OS, median (months)200420042004GOG-107GOG-177GOG-209
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 3 trials in this lane. GOG-107 (2004): experimental 9 months vs control 9.2 months; GOG-177 (2004): experimental 15.3 months vs control 12.3 months; GOG-209 (2004): experimental 37 months vs control 41 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
GOG-1072004OS9.2 months9 months
GOG-1772004OS12.3 months15.3 months
GOG-2092004OS41 months37 months

dMMR endometrial cancer, first-line chemotherapy with or without a checkpoint antibody

Primary advanced (stage III-IV) or recurrent endometrial cancer with mismatch repair-deficient (dMMR) tumours, in the dMMR cohort of NRG-GY018 and the dMMR subgroups of RUBY and DUO-E; control arm carboplatin-paclitaxel with placebo · Metastatic · endpoint: PFS

Results are not comparable across trials.

This lane holds dMMR results only. NRG-GY018 analysed dMMR as a separate stratified cohort (12-month progression-free survival); RUBY's dMMR/MSI-H population was a prespecified subgroup (24-month progression-free survival); DUO-E's dMMR result is a prespecified exploratory subgroup reported as a median. The entries therefore use different landmarks or measures and only the first is plotted; the others are listed. Eligibility, the antibody and its dose, partner drugs (DUO-E also tested olaparib; its durvalumab-alone dMMR median was not reached and is not shown), review method and follow-up differ. Compare only each trial's own two arms; the chart is not a pooled or network estimate.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100PFS at 12 months (%)2023NRG-GY018
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS at 12 months, rate, across 1 trial in this lane. NRG-GY018 (2023): experimental 74 % vs control 38 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
NRG-GY0182023PFS38 %74 %
RUBY2023PFS (not plotted: different measure or time point)15.7 %61.4 %
DUO-E2023PFS (not plotted: different measure or time point)7 months31.8 months

pMMR endometrial cancer, first-line chemotherapy with or without a checkpoint antibody

Primary advanced (stage III-IV) or recurrent endometrial cancer with mismatch repair-proficient (pMMR) tumours, in the pMMR cohort of NRG-GY018 and the pMMR subgroup of DUO-E; control arm carboplatin-paclitaxel with placebo · Metastatic · endpoint: PFS

Results are not comparable across trials.

This lane holds pMMR results only. NRG-GY018 analysed pMMR as a separate stratified cohort; DUO-E's pMMR results are prespecified exploratory subgroup analyses, with durvalumab alone and durvalumab plus olaparib each compared with the same control arm. RUBY's pMMR/MSS result is a hazard ratio for overall survival without arm medians in this record, so RUBY is listed on the lane but has no plotted entry. Eligibility, partner drugs, review method and follow-up differ. Compare only each trial's own two arms; the chart is not a pooled or network estimate.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

05101520PFS, median (months)202320232023NRG-GY018DUO-EDUO-E
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS, median, across 3 trials in this lane. NRG-GY018 (2023): experimental 13.1 months vs control 8.7 months; DUO-E (2023): experimental 9.9 months vs control 9.7 months; DUO-E (2023): experimental 15 months vs control 9.7 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
NRG-GY0182023PFS8.7 months13.1 months
DUO-E2023PFS9.7 months9.9 months
DUO-E2023PFS9.7 months15 months

Advanced or recurrent endometrial cancer after platinum chemotherapy

Women with advanced, recurrent or metastatic endometrial cancer that had progressed after a platinum-based regimen: KEYNOTE-775 (randomised, all randomised patients) and GARNET (single-group, dMMR) · Relapsed / refractory · endpoint: PFS

Results are not comparable across trials.

Only KEYNOTE-775 is a randomised comparison and is plotted, using its all-randomised-patients median (lenvatinib plus pembrolizumab versus doxorubicin or paclitaxel); its pMMR results are reported separately in the trial record. GARNET is a single-group study in dMMR disease with no comparison arm, so it is listed but not plotted. KEYNOTE-146 (single-arm) supported the 2019 approval and is also not plotted. Compare only the trial's own two arms; the chart is not a pooled or network estimate.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

0246810PFS, median (months)2019KEYNOTE-775
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS, median, across 1 trial in this lane. KEYNOTE-775 (2019): experimental 7.2 months vs control 3.8 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
KEYNOTE-7752019PFS3.8 months7.2 months

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Adjuvant treatment after surgery for endometrial cancer

Women after hysterectomy for endometrial carcinoma, by stage and risk group as defined in each cited trial · Adjuvant

  1. Radiation1Pelvic radiotherapy after surgeryAdjuvant pelvic radiotherapy: less recurrence, no survival gain shown (1980)
  2. Chemotherapy2Doxorubicin-cisplatin, stage III-IVGOG-122: doxorubicin-cisplatin versus whole-abdominal irradiation (2005)
  3. Radiation3Vaginal brachytherapy, high-intermediatePORTEC-2: vaginal brachytherapy in place of pelvic radiotherapy (2010)
  4. Chemotherapy4Chemotherapy + radiotherapy, high riskPORTEC-3: chemoradiotherapy versus radiotherapy, high-risk disease (2018)
  5. Chemotherapy5Chemotherapy with or without radiotherapyGOG-258: adding radiotherapy to chemotherapy did not extend RFS (2019)
  6. Understanding6Molecular classification of tumoursProMisE and PORTEC-3: molecular classes in high-risk disease (2015)
  7. Radiation7Molecular profile-guided treatmentPORTEC-4a: molecular profile guides adjuvant treatment (2025)
2026. Based only on the cited trials; practice varies. PORTEC-4a tested treatment chosen by molecular profile against standard vaginal brachytherapy at high-intermediate risk.
Text version
  • 2005: Pelvic radiotherapy after surgery → Doxorubicin-cisplatin, stage III-IV
  • 2010: Pelvic radiotherapy after surgery → Doxorubicin-cisplatin, stage III-IV → Vaginal brachytherapy, high-intermediate
  • 2020: Pelvic radiotherapy after surgery → Doxorubicin-cisplatin, stage III-IV → Vaginal brachytherapy, high-intermediate → Chemotherapy + radiotherapy, high risk → Chemotherapy with or without radiotherapy → Molecular classification of tumours
  • 2026: Pelvic radiotherapy after surgery → Doxorubicin-cisplatin, stage III-IV → Vaginal brachytherapy, high-intermediate → Chemotherapy + radiotherapy, high risk → Chemotherapy with or without radiotherapy → Molecular classification of tumours → Molecular profile-guided treatment

Advanced or recurrent endometrial cancer: systemic therapy

Women with stage III-IV or recurrent endometrial carcinoma, first line and after platinum, as defined in each cited trial · Metastatic

  1. Endocrine1Medroxyprogesterone acetate (progestin)Hormonal therapy: progestins to letrozole combinations (1999)
  2. Chemotherapy2Doxorubicin-cisplatin-paclitaxel (TAP)Chemotherapy for advanced disease: GOG-107, -177 and -209 (2004)
  3. Chemotherapy3Carboplatin-paclitaxel (non-inferior to TAP)Chemotherapy for advanced disease: GOG-107, -177 and -209 (2004)
  4. Targeted4Lenvatinib + pembrolizumab after platinumLenvatinib plus pembrolizumab after platinum: KEYNOTE-146 and -775 (2019)
  5. Immunotherapy5Dostarlimab + carboplatin-paclitaxelRUBY: dostarlimab added to carboplatin-paclitaxel, first line (2023)
  6. Immunotherapy6Pembrolizumab + carboplatin-paclitaxelNRG-GY018: pembrolizumab added to chemotherapy, both MMR cohorts (2023)
  7. Immunotherapy7Durvalumab + carboplatin-paclitaxelDUO-E: durvalumab with or without olaparib, first line (2023)
2026. Based only on the cited trials; practice varies. DUO-E added durvalumab, with or without olaparib, to first-line chemotherapy. Coverage differs by drug, label and tumour type.
Text version
  • 1999: Medroxyprogesterone acetate (progestin)
  • 2004: Medroxyprogesterone acetate (progestin) → Doxorubicin-cisplatin-paclitaxel (TAP)
  • 2020: Medroxyprogesterone acetate (progestin) → Doxorubicin-cisplatin-paclitaxel (TAP) → Carboplatin-paclitaxel (non-inferior to TAP) → Lenvatinib + pembrolizumab after platinum
  • 2023: Medroxyprogesterone acetate (progestin) → Doxorubicin-cisplatin-paclitaxel (TAP) → Carboplatin-paclitaxel (non-inferior to TAP) → Lenvatinib + pembrolizumab after platinum → Dostarlimab + carboplatin-paclitaxel → Pembrolizumab + carboplatin-paclitaxel
  • 2026: Medroxyprogesterone acetate (progestin) → Doxorubicin-cisplatin-paclitaxel (TAP) → Carboplatin-paclitaxel (non-inferior to TAP) → Lenvatinib + pembrolizumab after platinum → Dostarlimab + carboplatin-paclitaxel → Pembrolizumab + carboplatin-paclitaxel → Durvalumab + carboplatin-paclitaxel

Reversals and dead ends

Approaches that were abandoned or withdrawn, and trials whose results were later questioned, are part of the story.

  • 2008

    Systematic lymphadenectomy: no survival benefit shown in two trials

    Two randomised trials tested adding removal of the pelvic lymph nodes to hysterectomy in women whose cancer seemed to be confined to the uterus. The extra surgery found more nodal spread in one trial, but neither trial showed longer survival with it.

    Benedetti Panici (514 eligible, preoperative stage I): 5-year DFS 81.0% vs 81.7% and OS 85.9% vs 90.0% (lymphadenectomy vs none); HR for death 1.20 (95% CI 0.70-2.07); nodal metastases found 13.3% vs 3.2%. ASTEC (1,408, ITT): OS HR 1.16 (0.87-1.54), P = 0.31; RFS HR 1.35 (1.06-1.73), P = 0.017 (both above 1 in favour of standard surgery).

  • 2019

    GOG-258: adding radiotherapy to chemotherapy did not extend RFS

    In stage III or IVA endometrial cancer, chemotherapy plus radiotherapy was compared with chemotherapy alone after surgery. Women given both did not stay free of relapse longer. Radiotherapy lowered recurrence in the vagina and in pelvic and para-aortic lymph nodes, but cancer returned at distant sites more often. Final overall survival results also showed no difference.

    Randomised phase 3 (813 enrolled; 736 eligible for the primary analysis), stage III or IVA: cisplatin with volume-directed radiation then carboplatin-paclitaxel (chemoradiotherapy) vs carboplatin-paclitaxel alone. 60-month RFS 59% (95% CI 53-65) vs 58% (53-64); HR 0.90 (90% CI 0.74-1.10). 5-year vaginal recurrence 2% vs 7%; pelvic or para-aortic nodal 11% vs 20%; distant 27% vs 21%. Final OS (median follow-up 112 months): HR 1.05 (95% CI 0.82-1.34; p=0.72).

  • 2024

    KEYNOTE-B21: adjuvant pembrolizumab, no DFS gain overall

    In women with newly diagnosed high-risk endometrial cancer, pembrolizumab or a placebo was added to chemotherapy after surgery. Across all women, disease-free survival did not differ. In a planned subgroup of tumours with mismatch repair deficiency the pembrolizumab group fared better, but this was not a formal test and follow-up is early.

    Randomised double-blind phase 3 (1095 randomised): pembrolizumab 200 mg or placebo every 3 weeks for six cycles with carboplatin-paclitaxel, then 400 mg or placebo every 6 weeks for six cycles. Interim analysis (data cut-off 4 March 2024): DFS HR 1.02 (95% CI 0.79-1.32; p=0.570); 2-year DFS 75% vs 76%. dMMR (n=281) HR 0.31 (0.14-0.69); pMMR (n=814) HR 1.20 (0.91-1.57). Grade 3 or worse adverse events 71% vs 63%.

Today and tomorrow

Standard of care today

Localised

Surgery and staging, evidence base only; guideline text was not opened, so no recommendation is attributed to any guideline. The FIGO 2023 paper (an update of the 2009 system) adds histological type, tumour pattern and molecular classification to staging and encourages complete molecular classification (POLEmut, MMRd, NSMP, p53abn) in all endometrial cancers; the abstract of the ESGO-ESTRO-ESP 2025 guideline update states that it incorporates the FIGO 2023 system and covers diagnosis, management, follow-up and patient education. Surgical approach:

  • in GOG LAP2 (clinical stage I to IIA uterine cancer
  • 1,696 laparoscopy, 920 laparotomy) moderate to severe postoperative adverse events were 14% versus 21%, and the recurrence-free interval did not meet the protocol-specified non-inferiority definition (hazard ratio 1.14
  • 95% upper bound 1.46) while 5-year overall survival was almost identical in both arms (89.8%)
  • in LACE (760 women, stage I endometrioid) 4.5-year disease-free survival was 81.6% after total laparoscopic and 81.3% after total abdominal hysterectomy, meeting the 7% equivalence margin. Lymph nodes: two randomised trials of pelvic lymphadenectomy in presumed early-stage disease did not show a survival benefit (ASTEC, 1,408 women: overall survival hazard ratio 1.16, 95% CI 0.87 to 1.54
  • Benedetti Panici, 514 women: 5-year overall survival 85.9% with lymphadenectomy versus 90.0% without)
  • sentinel-node biopsy has been tested as an alternative in prospective non-randomised cohorts (FIRES, SENTOR) whose abstracts report diagnostic accuracy against lymphadenectomy (sensitivity 97.2% and 96%), not recurrence or survival. The LAP2 recurrence paper (2012) and the FIGO 2023 paper each have a published correction that could not be read, so their figures are given as the abstracts state them. Guideline identifiers only: ESGO-ESTRO-ESP update 2025 (Lancet Oncol, PMID 40744042), ESGO/ESTRO/ESP 2021 (Radiother Oncol, PMID 33712263), ESMO 2022 (Ann Oncol, PMID 35690222), NCCN Guidelines Insights, Uterine Neoplasms, Version 3.2025 (J Natl Compr Canc Netw, PMID 40763788)
  • whether these are the current editions was not checked
Berek JS, Matias-Guiu X, Creutzberg C, Fotopoulou C, Gaffney D, Kehoe S, et al 2023Concin N, Matias-Guiu X, Cibula D, Colombo N, Creutzberg CL, Ledermann J, et al 2025Concin N, Matias-Guiu X, Vergote I, Cibula D, Mirza MR, Marnitz S, et al 2021Oaknin A, Bosse TJ, Creutzberg CL, Giornelli G, Harter P, Joly F, et al 2022Guideline: Abu-Rustum NR, Campos SM, Amarnath S, Arend R, Barber E, Bradley K, et al 2025Walker JL, Piedmonte MR, Spirtos NM, Eisenkop SM, Schlaerth JB, Mannel RS, et al 2009Walker JL, Piedmonte MR, Spirtos NM, Eisenkop SM, Schlaerth JB, Mannel RS, et al 2012Janda M, Gebski V, Davies LC, Forder P, Brand A, Hogg R, et al 2017ASTEC study group, Kitchener H, Swart AM, Qian Q, Amos C, Parmar MK 2009Benedetti Panici P, Basile S, Maneschi F, Alberto Lissoni A, Signorelli M, Scambia G, et al 2008Rossi EC, Kowalski LD, Scalici J, Cantrell L, Schuler K, Hanna RK, et al 2017Cusimano MC, Vicus D, Pulman K, Maganti M, Bernardini MQ, Bouchard-Fortier G, et al 2021
Adjuvant

Adjuvant therapy after surgery; trial abstracts and the regulatory documents below were read, guideline text was not, so no guideline recommendation is attributed. PORTEC-2 (high-intermediate risk):

  • the authors of the primary report concluded that vaginal brachytherapy should be the adjuvant treatment of choice. PORTEC-3 (high-risk disease, chemoradiotherapy versus pelvic radiotherapy alone): the primary analysis did not show a significant overall survival difference (5-year overall survival 81.8% versus 76.7%, adjusted hazard ratio 0.76, 95% CI 0.54 to 1.06)
  • a post hoc update with one more year of follow-up reported 81.4% versus 76.1% (adjusted hazard ratio 0.70, 95% CI 0.51 to 0.97), and its authors wrote that the schedule should be discussed and recommended, especially for stage III or serous cancers. GOG-258 (stage III to IVA, chemoradiotherapy versus chemotherapy alone): 60-month relapse-free survival 59% versus 58% (hazard ratio 0.90, 90% CI 0.74 to 1.10) and final overall survival hazard ratio 1.05 (95% CI 0.82 to 1.34). KEYNOTE-B21 (adjuvant pembrolizumab with chemotherapy, 1,095 randomised): interim disease-free survival hazard ratio 1.02 (95% CI 0.79 to 1.32), with a hazard ratio of 0.31 (95% CI 0.14 to 0.69) in the dMMR subgroup, a preplanned subgroup analysis for a stratification factor
  • the authors concluded no improvement in the all-comer population. PORTEC-4a (high-intermediate risk, molecular-profile-based versus standard vaginal brachytherapy): 5-year vaginal recurrence 4.5% versus 1.6% (hazard ratio 2.71, 95% CI 0.79 to 9.34
  • the upper bound of the one-sided confidence interval of the difference, 5.3%, was below the pre-set 7.0% non-inferiority margin)
  • the authors concluded that profile-based treatment is safe and effective and that 46% of the profile group, with a favourable profile, were observed. Regulatory: the endometrial indications on the current Keytruda, Imfinzi and Jemperli labels read (label versions effective 10 July 2026, 31 August 2026 and 2 September 2025) are for advanced or recurrent disease
  • none is an adjuvant indication. Guideline documents are cited by identifier only (see the surgery entry)
Nout RA, Smit VT, Putter H, Jürgenliemk-Schulz IM, Jobsen JJ, Lutgens LC, et al 2010de Boer SM, Powell ME, Mileshkin L, Katsaros D, Bessette P, Haie-Meder C, et al 2018de Boer SM, Powell ME, Mileshkin L, Katsaros D, Bessette P, Haie-Meder C, et al 2019Matei D, Filiaci V, Randall ME, Mutch D, Steinhoff MM, DiSilvestro PA, et al 2019Matei DE, Enserro DM, Randall ME, Mutch D, Small W, DiSilvestro PA, et al 2025Van Gorp T, Cibula D, Lv W, Backes F, Ortaç F, Hasegawa K, et al 2024Slomovitz BM, Cibula D, Lv W, Ortaç F, Hietanen S, Backes F, et al 2025van den Heerik ASVM, Horeweg N, Haverkort MAD, Kuijsters N, Kommoss S, Koppe FLA, et al 2026Concin N, Matias-Guiu X, Cibula D, Colombo N, Creutzberg CL, Ledermann J, et al 2025Guideline: Abu-Rustum NR, Campos SM, Amarnath S, Arend R, Barber E, Bradley K, et al 2025FDA 2026FDA 2026FDA 2025
Metastatic

Primary advanced or recurrent disease, first line; regulatory documents read. FDA:

  • dostarlimab with carboplatin and paclitaxel, then dostarlimab alone, 31 July 2023 for dMMR or MSI-H disease (letter for BLA 761174/S-006) and 1 August 2024 for primary advanced or recurrent disease regardless of mismatch repair status (FDA notice and label
  • the approval letter was not available)
  • durvalumab with carboplatin and paclitaxel, then durvalumab alone, 14 June 2024 for dMMR disease (BLA 761069/S-045)
  • pembrolizumab with carboplatin and paclitaxel, then pembrolizumab alone, 17 June 2024 for primary advanced or recurrent endometrial carcinoma, with no mismatch repair restriction in the letter (BLA 125514/S-155). None of the three letters read contains accelerated-approval wording, and in the current label texts read (Keytruda, Imfinzi, Jemperli) no accelerated-approval statement accompanies the endometrial indications (the Jemperli label attaches one to its dMMR solid-tumour indication)
  • the approval type 'regular' remains an inference from that absence. For RUBY, the FDA notice of 1 August 2024 gives, in the overall population (494 patients), median overall survival of 44.6 months with dostarlimab and 28.2 months with placebo (hazard ratio 0.69, 95% CI 0.54 to 0.89). The Jemperli label (revised 08/2024) states that analyses in the mismatch repair proficient/microsatellite stable subgroup (n = 372
  • 76% of patients) were pre-specified exploratory, with an overall survival hazard ratio of 0.82 (95% CI 0.62 to 1.08), and gives 0.34 (95% CI 0.18 to 0.62) in the dMMR/MSI-H subgroup, which was not tested formally. EMA: the Jemperli procedural-steps document records an extension of indication to first-line treatment of dMMR/MSI-H disease (CHMP opinion 12 October 2023, Commission decision 7 December 2023, with the switch from conditional to standard authorisation) and a further extension to first-line treatment of primary advanced or recurrent disease, whose scope text names no dMMR/MSI-H restriction (CHMP opinion 12 December 2024, Commission decision 15 January 2025)
  • the EPAR overview page was last updated 12-2023 and still describes dMMR/MSI-H disease. EU decisions for pembrolizumab and durvalumab were not read. The Lynparza prescribing information read (label version effective 10 July 2025) has no endometrial indication, so olaparib with durvalumab has no FDA endometrial indication in the documents read. Guideline identifiers only: ESGO-ESTRO-ESP update 2025 (PMID 40744042)
  • the ESMO 2022 guideline (PMID 35690222) was published before the 2023 first-line immunotherapy reports
FDA 2023FDA 2024FDA 2024FDA 2025FDA 2024FDA 2026FDA 2024FDA 2026FDA 2025EMA 2025Concin N, Matias-Guiu X, Cibula D, Colombo N, Creutzberg CL, Ledermann J, et al 2025Oaknin A, Bosse TJ, Creutzberg CL, Giornelli G, Harter P, Joly F, et al 2022Mirza MR, Chase DM, Slomovitz BM, dePont Christensen R, Novák Z, Black D, et al 2023
Relapsed / refractory

Advanced or recurrent disease after prior systemic therapy; regulatory documents read. FDA, pembrolizumab with lenvatinib for disease that is not MSI-H or dMMR:

  • accelerated approval on 17 September 2019 (BLA 125514/S-065, based on tumour response rate and durability of response), and on 21 July 2021 a supplement (S-105) whose letter states that it fulfils the commitments made under 21 CFR 601.41 and covers prior systemic therapy in any setting. FDA, single-agent pembrolizumab for MSI-H or dMMR disease: 21 March 2022 (S-110
  • the letter has no accelerated-approval wording). FDA, single-agent dostarlimab for dMMR disease after platinum-containing treatment: accelerated approval on 22 April 2021, and regular approval on 9 February 2023 (S-003/S-004), the letter stating that the supplement fulfils commitments under 21 CFR 601.41. EMA, dostarlimab monotherapy: conditional marketing authorisation issued 21 April 2021 (opinion 25 February 2021), switched to standard authorisation on 7 December 2023 (EPAR page). FDA, fam-trastuzumab deruxtecan: accelerated approval on 5 April 2024 for unresectable or metastatic HER2-positive (IHC 3+) solid tumours after prior systemic treatment without satisfactory alternatives
  • the letter is tumour-agnostic and does not name endometrial cancer. Trastuzumab with carboplatin-paclitaxel in HER2-positive uterine serous carcinoma rests on a randomised phase 2 trial, and no regulatory action for it was read. Ongoing phase 3 trials enrol people after platinum chemotherapy and immunotherapy (see the horizon list), and the documents read do not settle the sequence of therapy after first-line immunotherapy
FDA 2019FDA 2021FDA 2022FDA 2021FDA 2023EMA 2021FDA 2024Fader AN, Roque DM, Siegel E, Buza N, Hui P, Abdelghany O, et al 2018

Trials to watch

  • TroFuse-005 NCT06132958

    After platinum-based chemotherapy and immunotherapy for endometrial cancer, does sacituzumab tirumotecan (MK-2870) differ from treatment of physician's choice in progression-free survival by blinded independent central review and in overall survival (the registry's two primary outcomes)? Registry record: randomised, open-label phase 3, 710 estimated, sponsor Merck Sharp & Dohme LLC. Registry status: active, not recruiting; start 2023-12-06; registry primary completion date 2028-01-10 (estimated); the record was last updated 2026-08-28; no results posted.

  • ASCENT-GYN-01 NCT06486441

    After platinum-based chemotherapy and anti-PD-1/PD-L1 immunotherapy for endometrial cancer, does sacituzumab govitecan differ from treatment of physician's choice in progression-free survival by blinded independent central review and in overall survival (the registry's two primary outcomes)? Registry record: randomised, open-label phase 3, 640 estimated, sponsor Gilead Sciences. Registry status: active, not recruiting; start 2024-08-28; registry primary completion date 2029-06 (estimated); the record was last updated 2026-07-29; no results posted. A design paper (Int J Gynecol Cancer 2026, PMID 41073155) exists; no results were read.

  • DESTINY-Endometrial01 NCT06989112

    As first-line treatment of HER2-expressing (IHC 3+/2+), mismatch repair proficient, primary advanced (stage III/IV) or recurrent endometrial cancer, how do trastuzumab deruxtecan plus rilvegostomig, or trastuzumab deruxtecan plus pembrolizumab, compare with carboplatin-paclitaxel plus pembrolizumab in progression-free survival by blinded independent central review (the registry's primary outcome)? Registry record: randomised phase 3 with three arms, open-label and sponsor-blinded, 600 estimated, sponsor AstraZeneca. Registry status: recruiting; start 2025-03-27; registry primary completion date 2029-01-19 (estimated); the record was last updated 2026-10-13; no results posted.

  • KEYNOTE-C93 NCT05173987

    In dMMR advanced or recurrent endometrial carcinoma with no prior systemic therapy for advanced or recurrent disease (one prior platinum-based adjuvant line is allowed if recurrence came at least 6 months after it), does first-line pembrolizumab alone differ from carboplatin-paclitaxel in progression-free survival by blinded independent central review and in overall survival (the registry's two primary outcomes)? Registry record: randomised, open-label phase 3, 280 estimated, sponsor Merck Sharp & Dohme LLC. Registry status: active, not recruiting; start 2022-02-03; registry primary completion date 2027-05-27 (estimated); the record was last updated 2026-05-05; no results posted.

  • DOMENICA NCT05201547

    In mismatch repair deficient relapsed or advanced endometrial cancer, does first-line dostarlimab alone differ from carboplatin-paclitaxel in progression-free survival (the registry's primary outcome)? Registry record: randomised phase 3, 260 estimated, sponsor ARCAGY/ GINECO GROUP. Registry status: active, not recruiting; start 2022-04-15; registry primary completion date 2026-04 (listed as estimated and already passed; the record was last updated 2025-12-04); no results posted; PubMed and a web search on 2026-10-21 found only a design paper (Future Oncol 2025, PMID 40323277).

  • NRG-GY026 NCT05256225

    In a phase II/III design (the registry phase field lists phase 3 only), in HER2-positive, stage I to IV, chemotherapy-naive endometrial serous carcinoma or carcinosarcoma, does adding trastuzumab and hyaluronidase-oysk, or pertuzumab, trastuzumab and hyaluronidase-zzxf, to paclitaxel-carboplatin differ from paclitaxel-carboplatin alone in progression-free survival, overall survival and incidence of dose-limiting toxicities (the registry's three primary outcomes)? Registry record: randomised, three arms, 360 estimated, sponsor National Cancer Institute (NCI); the official title says phase II/III. Registry status: active, not recruiting; start 2022-11-16; registry primary completion date 2027-10-31 (estimated); the record was last updated 2026-10-02; no results posted.

References

Show all 98 references

Loading references… Open the list as a page

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.

2019ReversalChemotherapy

majorAdjuvant

GOG-258: adding radiotherapy to chemotherapy did not extend RFS

In stage III or IVA endometrial cancer, chemotherapy plus radiotherapy was compared with chemotherapy alone after surgery. Women given both did not stay free of relapse longer. Radiotherapy lowered recurrence in the vagina and in pelvic and para-aortic lymph nodes, but cancer returned at distant sites more often. Final overall survival results also showed no difference.

Randomised phase 3 (813 enrolled; 736 eligible for the primary analysis), stage III or IVA: cisplatin with volume-directed radiation then carboplatin-paclitaxel (chemoradiotherapy) vs carboplatin-paclitaxel alone. 60-month RFS 59% (95% CI 53-65) vs 58% (53-64); HR 0.90 (90% CI 0.74-1.10). 5-year vaginal recurrence 2% vs 7%; pelvic or para-aortic nodal 11% vs 20%; distant 27% vs 21%. Final OS (median follow-up 112 months): HR 1.05 (95% CI 0.82-1.34; p=0.72).

Standard before

Six cycles of carboplatin and paclitaxel were the control arm; the trial tested whether adding radiation to platinum-based chemotherapy lengthened relapse-free survival.

What changed

The authors concluded that chemoradiotherapy was not associated with longer relapse-free survival than chemotherapy alone in stage III or IVA disease. The 2025 report concluded that it did not increase overall survival or relapse-free survival.

Why it mattered

The randomised comparison found that adding radiation to chemotherapy shifted where recurrences appeared (fewer in the vagina and in pelvic and para-aortic nodes, more at distant sites) without improving relapse-free or overall survival.

Caveats

  • The hazard ratio for relapse-free survival is published with a 90% confidence interval (0.74-1.10), not a 95% interval.
  • Distant recurrence was more common with chemoradiotherapy (27% vs 21%; HR 1.36, 95% CI 1.00-1.86).
  • Grade 3, 4 or 5 adverse events were reported in 58% with chemoradiotherapy and 63% with chemotherapy alone.
  • The trial enrolled stage III to IVA disease (the 2025 report also lists stage I or II serous or clear-cell cancer with positive cytology) and compared chemoradiotherapy with chemotherapy alone. PORTEC-3 compared chemoradiotherapy with radiotherapy alone and included stage I and II disease; the results are not comparable.