Screening method frequently misses endometrial cancer in Black women
Transvaginal ultrasound used to determine appropriateness of endometrial cancer biopsy missed nearly five times more cases of the disease among Black women vs. white women, according to a simulated cohort study published in JAMA Oncology.
“Black women have more than a 90% higher mortality rate after a diagnosis of endometrial cancer compared with white women in the U.S.,” Kemi M. Doll, MD, MS, researcher in the department of obstetrics and gynecology at University of Washington School of Medicine, told Healio. “This is a longstanding disparity that we have yet to make meaningful progress to address. Although we have focused before on evaluating access to care — whether and how Black women receive guideline-adherent care — in this study we sought to evaluate the guidelines themselves.”
Doll and colleagues constructed a simulated cohort — using SEER registry data from 2012 to 2016, the U.S. Census and published estimates of endometrial thickness distribution and prevalence of fibroids — of 367,073 Black (12.2%) and white (87.8%) women with postmenopausal bleeding to determine the accuracy of transvaginal ultrasonography endometrial thickness thresholds in identifying cases of endometrial cancer.
“One clinical care pathway is to perform a diagnostic biopsy for endometrial cancer only when the transvaginal ultrasonography screening shows the thickness of the endometrium is 4 mm or greater. But not all endometrial cancers increase endometrial thickness, and uterine fibroids can make the endometrium harder to measure,” Doll said.
Researchers also noted that the transvaginal ultrasound testing strategy was developed based upon large population-based studies from Scandinavia, Italy and Hong Kong that did not include Black women.
Among the women in the simulated cohort, 36,708 had endometrial cancer.
The 4 mm-or-greater endometrial thickness threshold prompted biopsy for less than half of Black women with endometrial cancer, for a sensitivity of 47.5% (95% CI, 46-49). Among Black women referred for biopsy, 13.1% were diagnosed with endometrial cancer, for a positive predictive value of 13.1% (95% CI, 12.5-13.6) and an area under the curve for the 4 mm-or-more threshold of 0.57 (95% CI, 0.56-0.57).
Conversely, the 4mm-or-more threshold led to biopsy among most white women with endometrial cancer, for a sensitivity of 87.9% (95% CI, 87.6-88.3). Among white women referred for biopsy, 14.6% were diagnosed with endometrial cancer, for a positive predictive value of 14.6% (95% CI, 14.4-14.7) and an AUC of 0.73 (95% CI, 0.73-0.74).
“This startling difference is due to the higher prevalence of fibroids and nonendometrioid histology type among Black women, both of which decrease the accuracy of the transvaginal ultrasound strategy,” Doll said. “This places Black women at higher risk for false-negative results, which is unacceptable in a group that already is most vulnerable to the worst outcomes in endometrial cancer. A false-negative result could easily lead to delays in diagnosis and treatment, allowing more time for cancer growth and spread, ultimately making surviving an endometrial cancer diagnosis much less likely.”
Doll and colleagues cautioned against immediate practice change and recommend their results be confirmed prospectively by longer follow-up and with community engagement as a central component.
“However, we urge all clinicians to recognize the potential for missed diagnoses among Black women, especially those with fibroids, and counsel patients on the relative benefit of endometrial biopsy for definitive diagnosis vs. relying on the less-invasive transvaginal ultrasound screening,” Doll said. “The next important step we are undertaking is using direct clinical data from Black women to confirm these results and understand what additional factors may lead to misdiagnosis of early signs of endometrial cancer. We currently have a national study investigating knowledge and practice patterns of health care providers when caring for people with postmenopausal bleeding and the GUIDE-EC Study, including more than 4,000 symptomatic Black women, to evaluate transvaginal ultrasound performance using real-world patient data.”
Taken together, the data from this study suggest transvaginal ultrasound clinical guidelines fall short in detecting endometrial cancer when simulated for Black women, according to an accompanying editorial by Eloise Chapman-Davis, MD, researcher in the department of obstetrics and gynecology at Weill Cornell Medicine, and Dineo Khabele, MD, researcher in the department of obstetrics and gynecology at Washington University School of Medicine in St. Louis.
“These findings may partially explain failures and delays in diagnosis that lead to advanced-stage disease and excess mortality for Black women,” Chapman-Davis and Khabele wrote. “New risk-based clinical algorithms should be developed to account for the higher prevalence of nonendometrioid endometrial cancer and fibroids among Black women. New risk-based algorithms could also include a lower threshold for endometrial sampling in Black women who experience postmenopausal bleeding. Involvement of Black women in designing clinical trials and developing new algorithms will be necessary for appropriate intervention.”
For more information:
Kemi M. Doll, MD, MS, can be reached at University of Washington, 1939 NE Pacific St., Box 356460, Seattle, WA 98195; email: firstname.lastname@example.org.