Cytological analysis of voided urine and vaginal fluid has good accuracy for detecting endometrial cancer in women with postmenopausal bleeding, as shown in the DETECT trial.
The combined urine and vaginal cytology had a detection sensitivity of 80.8 percent and specificity of 92.6 percent for endometrial cancer. The negative predictive value was 98.8 percent, while the positive predictive value was 38.1 percent. [Lancet Obstet Gynaecol Womens Health 2026;doi:10.1016/S3050-5038(26)00222-0]
Notably, the combined cytology detected 95.8 percent of aggressive histologies and 96.4 percent of locally advanced or metastatic diseases. Detection sensitivity was higher in high-grade than low-grade disease (95.8 percent vs 66.7 percent).
Diagnostic accuracy of the combined cytology was good overall, with an area under the receiver operating characteristic curve (AUC) of 0.89. However, the test performed best for high-grade (AUC 0.95) and late-stage (stage ≥2; AUC 0.98) cancers than for low-grade (AUC 0.74) and early-stage (AUC 0.85) tumours.
For detecting any pelvic cancer, the sensitivity and specificity of the combined cytology were 80 percent and 92.6 percent, respectively.
“The findings of our diagnostic accuracy study show that cancer cells that are shed from endometrial tumours can be detected by cytology in urine and vaginal samples in women with postmenopausal bleeding,” said first author Dr Eleanor Jones from the University of Manchester, Manchester, UK, and colleagues.
“The high specificity of urogenital cytology makes it especially attractive as a rule-in test and, as such, it could be of particular benefit to older and frail patients as well as those who cannot tolerate routine gynaecological procedures without anaesthesia,” Jones and colleagues pointed out.
The authors also acknowledged that the test performance for low-grade, early-stage endometrial cancer was insufficient to recommend urogenital cytology as a definitive rule-out test. “However, repeat sampling or full diagnostic workup in the event of ongoing symptoms could provide appropriate safety-netting for these clinically indolent tumours,” they said.
Jones and colleagues argued that urogenital cytology could “support rapid diagnosis, provide quick reassurance, and considerably reduce the number of women requiring invasive tests for suspected endometrial cancer,” if incorporated into the current diagnostic pathway.
“Sampling was simple, minimally invasive, and well tolerated. Sample collection does not need specialist training, lending itself to a community-based triage tool to expedite or avoid rapid access clinic referrals,” they added.
DETECT involved 1,864 consecutive women (median age 57 years, 85.2 percent White) who presented with postmenopausal bleeding. Of these, 115 (6.2 percent) had malignancies, with 99 having endometrial cancer and 16 having other pelvic cancers. Other pelvic cancers included cervical cancer (n=7), ovarian cancer (n=3), leiomyosarcoma (n=2), colorectal cancer (n=2), bladder cancer (n=1), and metastatic pancreatic cancer (n=1).
All women provided a self-collected voided urine sample and a Delphi screener-collected vaginal fluid sample before undergoing routine clinical investigations for suspected endometrial cancer. The urogenital samples were independently assessed by two of three cytopathologists unaware of participant cancer status, and discrepancies were settled by consensus review.