DNA ploidy in curettage specimens identifies high-risk patients and lymph node metastasis in endometrial cancer.

BACKGROUND: Preoperative risk stratification is essential in tailoring endometrial cancer treatment, and biomarkers predicting lymph node metastasis and aggressive disease are aspired in clinical practice. DNA ploidy assessment in hysterectomy specimens is a well-established prognostic marker. DNA ploidy assessment in preoperative curettage specimens is less studied, and in particular in relation to the occurrence of lymph node metastasis. METHODS: Curettage image cytometry DNA ploidy in relation to established clinicopathological variables and outcome was investigated in 785 endometrial carcinoma patients prospectively included in the MoMaTEC multicentre trial. RESULTS: Diploid curettage status was found in 72.0%, whereas 28.0% were non-diploid. Non-diploid status significantly correlated with traditional aggressive postoperative clinicopathological features, and was an independent predictor of lymph node metastasis among FIGO stage I-III patients in multivariate analysis (OR 1.94, P=0.033). Non-diploid status was related to shorter disease-specific survival (5-year DSS of 74.4% vs 88.8% for diploid curettage, P<0.001). When stratifying by FIGO stage and lymph node status, the prognostic effect remained. However, in multivariate regression analysis, preoperative histological risk classification was a stronger predictor of DSS than DNA ploidy. CONCLUSIONS: Non-diploid curettage is significantly associated with aggressive clinicopathological phenotype, lymph node metastasis, and poor survival in endometrial cancer. The prognostic effect was also observed among subgroups with (presumably) less aggressive traits, such as low FIGO stage and negative lymph node status. Our results indicate curettage DNA ploidy as a possible supplement to existing parameters used to tailor surgical treatment.

A discordant histological risk classification in preoperative and operative biopsy in endometrial cancer is reflected in metastatic risk and prognosis.

INTRODUCTION: In endometrial cancer, tissue for histological evaluation is obtained preoperatively (endometrial biopsy) and operatively (hysterectomy specimen). We investigated if a discordant risk classification based on preoperative and operative biopsy is reflected in metastatic risk and prognosis. PATIENTS AND METHODS: One thousand three hundred and seventy-four patients were prospectively included in a multicentre setting (Molecular Markers for Treatment of Endometrial Cancer (MoMaTEC) study). Preoperative and operative specimens were classified as high risk if non-endometrioid histology or endometrioid grade 3; otherwise low risk. Disease specific survival differences were calculated by means of Kaplan-Meier and Cox proportional hazard models. RESULTS: Discordant risk was found in 207 (16%) cases. Lymph node metastases were detected in 7% and 23% of patients with concordant low and high risk respectively versus 14% and 20% in the discordant groups (p<0.001). Five-year disease specific survival in the discordant groups proved intermediate (75-80%) to concordant low (94%) or high (58%) risk. Both operative and preoperative biopsy high-risk results have independent prognostic impact on disease specific survival with adjusted hazard ratios of 2.4 (95% confidence interval (95% CI) 1.5-3.9) and 2.1 (95% CI 1.3-3.2) respectively by Cox analysis. CONCLUSIONS: Discordant risk in preoperative biopsy and hysterectomy identifies an intermediate group with respect to disease spread and prognosis. Preoperative biopsy results remain important also with the hysterectomy histology available.