Objectives: To explore associations between histology and outcome in advanced or recurrent endometrial cancer patients participating in Gynecologic Oncology Group chemotherapy trials.
Methods: Age, race, performance status, histologic type (serous=S; clear cell=CC; endometrioid=E), disease stage, and prior radiation were evaluated using various analytic methods to evaluate the probability of response and identify independent predictors of progression-free survival (PFS) and overall survival (OS).
Results: Single agent or combination chemotherapy regimens including doxorubicin (A) (12%), doxorubicin/cisplatin (AP) (63%), doxorubicin/paclitaxel (AT) (13%), and paclitaxel/doxorubicin/cisplatin (TAP) (11%) were used among 1203 patients treated on 4 randomized clinical trials. Breakdown of disease stage was 7.8% stage III, 22.8% stage IV, and 69.4% recurrent disease. Histologic distribution was 18% S, 3.7% CC, 8.5% mixed, 51.7% E and 18.1% other. More S/CC patients enrolled on trials with advanced stage (III-IV) disease (as opposed to recurrent disease) compared to E patients (45% vs. 24%, p<0.05). Overall response rate was 42% (E=44%, S=44%, CC=32%). Histologic type was not an independent predictor of response. Independent predictors of PFS included race, performance status, disease stage, and CC histology. Histology was also an independent predictor of OS; the relative hazard ratio for S histology was 1.2 (1.02-1.4; p=0.03), and for CC was 1.51 (1.1-2.07; p=0.01).
Conclusion: In patients with advanced/recurrent endometrial cancer treated with A, P and/or T, response was not associated with histology. This exploratory analysis does not support exclusion of S tumors in future trials. Poorer PFS and OS were observed in CC compared to other types, but a lack of benefit from chemotherapy was not shown, and as this histology represents such a small fraction, it does not seem feasible to have separate chemotherapy trials for CC.