JGOG2033: randomized phase III trial of whole pelvic radiotherapy vs cisplatin-based chemotherapy in patients with intermediate risk endometrial carcinoma

Reviewer: Walter Sall, MD
Last Modified: May 16, 2005

Presenter: S. Sagae
Presenter's Affiliation: Sapporo Medical University
Type of Session: Scientific


  • Optimal adjuvant therapy for endometrial carcinoma has not been clearly defined.  Radiation therapy is generally used for earlier stages, while more advanced disease is generally treated with chemotherapy. 
  • A GOG study comparing adjuvant whole pelvis radiation (WPI) and adjuvant chemotherapy in early stage disease closed due to poor accrual.

Materials and Methods

  • Patients underwent surgical staging with TAH/BSO, pelvic and para-aortic lymph node sampling and peritoneal fluid cytology.
  • Patients had T1C disease or worse, age < 75, and PS 0-3.
  • Patients were randomized to receive either adjuvant WPI or chemotherapy.
  • WPI was given with open AP/PA fields.  45-50 Gy in 180-200cGy daily fractions was given.  11 patients also received para-aortic radiation.
  • 3 cycles of CAP chemotherapy were given.  Each cycle consisted of cyclophosphamide, doxorubicin and cisplatin given every 4 weeks.
  • Endpoints included Progression Free Survival (PFS), Overall Survival (OS) and toxicity.


  • From 1/1994 to 12/2000, 475 patients were entered on this study.  238 were randomized to WPR and 237 to chemo.  Of these, 193 and 192 patients, respectively, were eligible for this analysis.
  • Patients were well balanced for known prognostic factors.  60% of patients had IC disease, 17% had stage II.  55% of patients had grade I disease
  • 1.6% of WPI patients and 4.7% of chemo patients had Gr III/IV toxicity.  No treatment related deaths occurred.
  • With a median follow-up of 60 months, 5 year PFS was 84.0% and 82.1% and 5 year OS was 85.9% and 87.1% for WPI and chemotherapy, respectively.  There was no significant difference for both endpoints.
  • An unplanned subgroup analysis of 1C patients only, showed no significant difference in PFS and OS between the two treatment arms.
  • Unplanned subgroup analysis of stage II and IIIA (positive cytology only) showed a significant PFS and OS benefit for chemotherapy over WPI.
  • Overall risk of recurrence was 15-16%.  30% of recurrences were in the pelvis and 70% distant in both treatment arms.

Author's Conclusions

  • There is no difference in risk of recurrence, PFS or OS between WPR and chemotherapy as adjuvant treatment for intermediate risk endometrial cancer.
  • Subgroup analysis of 1C patients showed difference between WRT and chemotherapy.
  • Subgroup analysis of stage II and IIIa (positive cytology) patients showed an OS and PFS benefit to chemotherapy over WPR.

Clinical/Scientific Implications

In this trial, a very heterogeneous group of patients was treated with a standardized technique.  Radiation and/or chemotherapy technique was not customized for patient stage and known risk factors as has become common practice.  However, the overall result of the trial, showing no difference between WPR and chemotherapy for intermediate risk disease, is the first randomized phase III evidence that chemotherapy alone may be adequate adjuvant treatment, without WPR, in this group of patients.  Because of the limitations of the study, it should be considered hypothesis-generating, rather than as defining a new standard of care.  Further study of risk factors for local vs distant recurrence and pattern of failure is sorely needed and will help us to custom tailor adjuvant treatment regimes to meet each patient's needs.

The unplanned subgroup analyses may indeed support the hypothesis that more advanced and high risk endometrial cancer patients benefit from chemotherapy because of the high risk of metastatic disease.  Whether or not high-intermediate risk patients benefit from the local control benefit of radiation cannot be discerned from this trial.  Results from unplanned subgroup analysis must be taken with a grain of salt.