Endometrial cancer is the most common female genital tract malignancy in developed countries, and it occurs predominantly in postmenopausal women - approximately 75-80% of patients are postmenopausal at diagnosis. The disease is strongly linked to estrogen exposure: endometrioid adenocarcinoma (the most common type) is classified as estrogen-dependent, meaning prolonged or excessive estrogen stimulation of the uterine lining promotes cancer development.
Despite a generally favorable prognosis compared to other gynecological cancers, approximately 20% of endometrial cancer patients die from their disease within 5 years. Established prognostic factors include FIGO surgical stage, tumor grade, histological type, and lymph node status. However, there is growing interest in whether menopausal status itself - beyond simply being pre- or postmenopausal - provides independent prognostic information.
Specifically, this study asked whether the duration of time since the last menstrual period (a continuous measure of how long a woman has been postmenopausal) and postoperative FSH (follicle-stimulating hormone) concentration - a hormonal marker of ovarian inactivity - contribute to overall survival after endometrial cancer diagnosis. These aspects of menopausal status are rarely included in prognostic models, and no prior study had used exact time since last menstruation as a continuous prognostic variable.
The study enrolled 92 women with endometrioid endometrial adenocarcinoma treated at the Medical University in Lublin, Poland, between September 2013 and April 2016. All patients underwent standard surgical treatment (total hysterectomy with bilateral salpingo-oophorectomy, with or without pelvic lymph node dissection) and received adjuvant treatment per European Society of Gynaecological Oncology guidelines when indicated. The median follow-up extended to June 2022 - up to 9 years after diagnosis.
Blood samples were collected within one day of surgery for baseline measurements of FSH, LH (luteinizing hormone), CA125, and HE4 (a protein marker used in gynecological cancer assessment). FSH and LH were measured again 4 weeks after surgery to capture the hormonal response to surgical removal of the ovaries. Demographic data including age, body mass index (BMI), parity (number of pregnancies), and precise dates of first and last menstruation were collected from records and patient interviews.
The primary statistical tool was the Cox proportional hazards regression model, which simultaneously assesses the contribution of multiple variables to survival time. The model included eight variables: age at diagnosis, FIGO stage, BMI, adjuvant treatment (brachytherapy), parity, time since last menstruation (in years), and postoperative FSH concentration. Model quality was verified using Akaike's Information Criterion and residual diagnostics, confirming the model was statistically sound.
The strongest predictor of worse survival was advanced FIGO stage. Patients with FIGO II-IV disease had a hazard ratio (HR) of 10.77 compared to FIGO IA patients (p<0.0001), meaning their risk of death at any given time was nearly 11 times higher when all other factors were held constant. This confirms the dominant role of surgical stage in endometrial cancer prognosis. Notably, FIGO IB stage was not significantly different from IA (HR=1.47, p=0.59), reflecting the generally good prognosis of fully confined early-stage disease.
Two menopausal status variables were statistically significant. Time since last menstruation had an HR of 1.151 per year (p=0.0478), meaning each additional year of postmenopausal life increased the mortality hazard by 15.1%. This compounds substantially: 5 additional years more than doubled the hazard (x2.02), 10 years quadrupled it (x4.08), and 15 years increased it more than 8-fold (x8.24). Postoperative FSH had an HR of 1.020 per mIU/mL (p=0.0493); at 30 mIU/mL the hazard was 1.81x baseline, and at 70 mIU/mL it was 4.0x baseline.
Importantly, several commonly assumed prognostic factors were not significant in this model: age at diagnosis (HR=0.906, p=0.146), BMI (HR=1.041, p=0.358), adjuvant brachytherapy (HR=0.954, p=0.924), and parity (HR=0.894, p=0.533) all failed to reach statistical significance. Kaplan-Meier survival analysis confirmed the menopausal duration finding in both younger (age 45-64) and older (age 65+) patient subgroups independently.
The finding that longer postmenopausal duration predicts worse survival is novel and biologically plausible. As women remain postmenopausal for longer, the hormonal milieu of the body shifts progressively - estrogen levels continue to decline, while FSH rises further. This prolonged low-estrogen state may be associated with changes in immune function, metabolic health, and cellular senescence that collectively influence how aggressively endometrial cancer behaves and how well patients tolerate and respond to treatment.
High postoperative FSH levels reflect the complete absence of ovarian estrogen production, which occurs both in natural menopause and in women who undergo surgical removal of the ovaries (as part of endometrial cancer treatment). The FSH measurement 4 weeks after surgery captures the body's hormonal state following surgery. Previous studies showed conflicting results about simple menopausal status and endometrial cancer prognosis, possibly because grouping all postmenopausal women together obscures the gradient effect that this study demonstrated with continuous time-since-menstruation analysis.
This is the first study to show that the exact duration of time since last menstruation - rather than simply being pre- or postmenopausal - has independent prognostic value. The authors suggest this variable should be considered for inclusion in future AI-based prognostic models for endometrial cancer, where collecting precise menstrual history data could improve the accuracy of survival predictions without requiring molecular testing.
A key strength of this study is that both identified prognostic factors - time since last menstruation and postoperative FSH - are easy to collect in routine clinical practice. Menstrual history is obtained during standard patient interviews, and FSH measurement is a simple blood test available at virtually any hospital laboratory. Neither requires molecular tumor profiling or expensive genetic testing.
This simplicity makes the findings particularly relevant for settings where advanced molecular diagnostics are unavailable or cost-prohibitive. The authors note that the molecular TCGA classification system - while powerful - is used in only 3-5% of published endometrial cancer studies due to logistical and financial barriers. Simple clinical variables like menopausal duration and FSH could complement existing staging systems and potentially improve risk stratification for the majority of patients who do not have access to molecular testing.
The study has limitations: it is single-center and retrospective, involving 92 patients from a single Polish institution with predominantly endometrioid histology. The cohort predates wide adoption of TCGA molecular classification, meaning the prognostic value of menopausal variables within each molecular subtype is unknown. Future larger, prospective, multi-center studies incorporating both menopausal variables and molecular classification are needed to fully validate and contextualize these findings.
This study demonstrates for the first time that the precise duration of time since last menstruation is an independent prognostic factor for overall survival in endometrial cancer patients. Combined with postoperative FSH concentration, these menopausal status variables add meaningful information beyond FIGO staging - with each additional decade of postmenopausal life multiplying mortality hazard by more than 4-fold.
The unexpected non-significance of age, BMI, and adjuvant brachytherapy in this multivariate model challenges some assumptions in the field. While these factors show univariate associations, they appear to share predictive overlap with the menopausal status variables and FIGO stage in this analysis. The finding that brachytherapy showed no survival benefit likely reflects the well-established pattern from PORTEC trials, where adjuvant radiotherapy reduces local recurrence but does not significantly improve overall survival in early-stage disease.
Collecting precise menstrual history data as part of routine cancer registry protocols could contribute to more accurate AI-based prognostic models in the future. The authors advocate for updating and revising current endometrial cancer prognostic frameworks to incorporate complementary clinical indicators like menopausal duration, ultimately enabling more truly personalized cancer counseling and care.