Analysis of Prediagnostic Metabolic Profiles of Endometrial Carcinoma Patients in the Obese Population

Abstract

Abstract Endometrial carcinoma has been traditionally divided into type 1 and type 2 carcinomas. Excess estrogen is believed to be the pathologic mechanism of type 1 carcinoma. The WHO Classification of Tumors of Female Reproductive Organs lists obesity as a risk factor for type 1 endometrial carcinoma. On the contrary, type 2 carcinoma is generally not associated with estrogen or obesity. Endometrial endometrioid carcinoma is the most common type 1 carcinoma. Endometrial serous carcinoma is a typical type 2 carcinoma. In our community, with a high prevalence of obesity, we investigated whether serous carcinoma, a type 2 carcinoma, could also occur in the obese population, and whether there are identifiable metabolic differences between endometrioid carcinoma patients and serous carcinoma patients in the obese population prior to cancer diagnosis, which might provide clues to the different pathogenesis of the two types of carcinoma. Two cohorts of postmenopausal patients with serous carcinoma or endometrioid carcinoma between 2006 and 2016 were established. Then each cohort was subdivided into four categories based on patients’ average BMI (within 5 years before cancer diagnosis): nonobese (<30), moderately obese (30-35), severely obese (35-40), and morbidly obese (>40). Patients’ average triglycerides, HDL, blood pressures, and HbA1c levels (within 5 years before cancer diagnosis) were obtained. In total, 304 patients were in the endometrioid carcinoma cohort, while 135 patients were in the serous carcinoma cohort. In the morbidly obese category (BMI >40), serous carcinoma patients had significantly lower triglycerides and HbA1c levels than endometrioid carcinoma patients while their BMIs were comparable. For HDL and systolic and diastolic blood pressures, no significant difference was observed between the two groups. In each of the other BMI categories, serous carcinoma patients also had lower triglycerides and HbA1c levels than endometrioid carcinoma patients. In the serous carcinoma patient cohort, there was a moderately negative correlation between average BMI and triglycerides/HbA1c levels, with a Pearson’s correlation coefficient of –0.25 for triglycerides level and –0.19 for HbA1c level. On the contrary, there was no correlation between average BMI and triglycerides/HbA1c levels in the endometrioid carcinoma patient cohort (Pearson’s coefficient close to 0). In summary, the serous carcinoma patients in the obese population lacked the metabolic profiles generally associated with high BMI. These findings show that there are identifiable prediagnostic metabolic differences between endometrioid carcinoma patients and serous carcinoma patients. The metabolic differences between the two groups in each BMI category suggest that BMI should be combined with metabolic markers, rather than used alone, in the evaluation of risk factors for endometrial carcinoma. Future epidemiologic and experimental studies focusing on the association between metabolic syndrome and endometrial carcinoma should separate different histologic subtypes of endometrial carcinoma in the study design due to their distinct metabolic profiles.