New Predictive Epidemiologic Model May Identify Individuals at High Risk for Endometrial Cancer

Get Permission

Investigators have developed a predictive model that may be used in research and eventually clinical settings to identify individuals at high risk of developing endometrial cancer who would benefit from screenings, according to a report published by Shi et al in the Journal of the National Cancer Institute. The investigators noted that it may be one of the first predictive models to assess the utility of genetic factors in endometrial cancer risk prediction.


Endometrial cancer is the most common gynecologic malignancy in the United States, yet current guidelines fail to recommend that individuals already at high risk—such as those with Lynch syndrome—undergo screening.

“Given the rising incidence and mortality rate of endometrial cancer, population screening to identify high-risk [individuals] is an attractive public health strategy,” highlighted first study author Joy Shi, PhD, Professor of Epidemiology at the Harvard T.H. Chan School of Public Health at Harvard University. “I was interested in leveraging the incredibly rich questionnaire and genetic data available from the Epidemiology of Endometrial Cancer Consortium (E2C2) and combining it with other nationally representative data sources to assess whether we could improve on our ability to predict endometrial cancer risk in the general population,” she added.

The only two previous predictive models for determining endometrial cancer risk were focused on much more selective study populations and did not include risk factors such as education level, history of diabetes, history of hypertension, or use of hormonal therapy.

Study Methods and Results

The model was trained on pooled data from an international consortium spanning 19 studies across different countries before being validated in three large U.S.-based cohorts. The 19 studies involved participants aged 45 to 85 years who were White and postmenopausal.

The investigators used a Least Absolute Shrinkage and Selection Operator model to predict which individuals would be at a higher risk of endometrial cancer relative to other participants. Then, they translated the relative risks into an absolute risk prediction to determine the probability that an individual might develop endometrial cancer within the next 10 years. To conduct this analysis, the investigators combined estimates from the E2C2 data with nationally representative data from the Surveillance, Epidemiology, and End Results Program and the National Health and Nutrition Examination Survey.

The investigators further validated the predictive model using data from three additional cohorts—121,700 registered nurses aged 30 to 55 years in the Nurses’ Health Study (NHS) study; 116,430 registered nurses aged 25 to 42 years in the NHS II; and 78,232 individuals aged 55 to 74 years in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Participants were followed for 10 years in all three validation cohorts. The investigators used questionnaires, death records, and medical records to confirm endometrial cancer diagnoses made by pathologists and physicians.

“[The] NHS, NHS II, and PLCO cohorts [have] uniquely rich datasets, with data on tens of thousands of participants over decades of follow-up,” explained senior study author Immaculata De Vivo, PhD, MPH, Professor of Epidemiology at the Harvard T.H. Chan School of Public Health at Harvard University as well as Professor of Medicine at Brigham and Women’s Hospital and Harvard Medical School. “There is also genetic data available for many of these study participants, which allowed us to investigate the potential contributions of genetic factors in predicting endometrial cancer risk,” she noted.

The investigators matched 18 previously identified genome-wide significant single-nucleotide polymorphisms for endometrial cancer from 12 different genome-wide association studies of various disease outcomes with the participants from the NHS.

They found that the average cumulative risk of developing endometrial cancer for individuals aged 45 to 85 years was 5.4%—though individuals in the lowest decile of risk exhibited a risk of 1.4% to 1.8%; whereas individuals in the highest decile of risk exhibited a risk of 13.7% to 15.01%.

The investigators revealed that the predictive epidemiologic model demonstrated moderate accuracy with only slight improvements when genetic factors were included—and was well-calibrated when applied to the NHS II data, but poorly calibrated when applied to the NHS data. However, both the epidemiologic and the epidemiologic plus genetic models were well calibrated in the PLCO study.

“We found that epidemiologic risk factors alone, which could be quickly, cheaply, and easily collected in clinical or public health settings, could accurately distinguish between [individuals] at high and low risk for endometrial cancer,” Dr. Shi underscored.

When the risk model was applied to a more recent and representative U.S. population of White individuals, the predictive model identified 2.5% of those with more than a 20% cumulative risk of developing endometrial cancer between the ages of 40 and 85. Those in the 97th percentile of risk had a predicted lifetime risk comparable to that in patients with Lynch syndrome, who are recommended to undergo annual screening. The findings suggested that clinical guidance regarding endometrial cancer screenings may need updating, and that predictive models will need to continually be recalibrated in new geographic, temporal, or population settings to account for rarer subtypes of endometrial cancer and make predictions in more diverse populations.

Since 95% of the training cohort and 93% to 95% of the validation cohorts consisted of White individuals, the predictive model is currently not capable of generating and validating race-specific estimates for non-White individuals. However, the E2C2 has been awarded a grant to understand disparities in endometrial cancer mortality across racial subgroups.

The investigators concluded that although their prediction models based on epidemiologic factors alone successfully identified individuals who were most at risk of developing endometrial cancer, additional studies will be needed to refine the predictive models for clinical and public health practice across multiethnic populations.

Disclosure: For full disclosures of the study authors, visit

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.