Skip to main navigation
Skip notifications
Access to POEMs and Essential Evidence Plus will no longer be included in CMA membership as of Dec. 1, 2023.
Search
Algorithm to predict malignancy in a pelvic mass in female patients
Clinical Question
Is the use of an algorithm of genes and biomarkers accurate enough to predict malignancy in a pelvic mass in female patients?
Bottom line
A combination of 8 genes and 4 serum biomarkers for use as testing to predict malignancy in a pelvic mass in females could be accurate enough for use in practice. This combination needs to be validated in a larger patient population before its routine use in clinical practice.
Reference
Study design: Diagnostic test evaluation
Funding: Industry
Setting: Outpatient (specialty)
Synopsis
This was an industry-sponsored trial (ANGLE Europe Limited, United Kingdom) to develop a testing algorithm of genes and serum biomarkers for use to predict malignancy of female pelvic masses. The rationale for the study was to identify a more accurate approach than what is currently in use to identify patients most likely to have ovarian cancer and refer them to gynecologic oncologists for surgery. The authors enrolled 200 female participants with pelvic masses who were scheduled to undergo biopsy or surgical excision. Patients with cancers diagnosed within the previous 5 years, other than squamous or basal cell skin cancers, were excluded. The authors obtained peripheral blood from each patient to gather and test cancer-related circulating tumor cells. Only patients with a confirmed histologic diagnosis were deemed evaluable (n = 183). Of these, 104 (57%) patients had benign disease, 17 (9%) had low malignant potential tumors, 42 (23%) had ovarian cancers, 14 (8%) had nonovarian gynecologic cancers, and 6 (3%) had nongynecologic metastatic cancers. The authors used a gene identification system (Parsortix R) and serum biomarker testing to develop a testing algorithm of the most accurate combination of the 72 genes and 7 biomarkers available. No individual genes or biomarkers were useful in predicting malignancy. By using logistic regression and plotting receiver operating curves, the authors determined area under the curve (AUC) to identify the most accurate combinations. The most accurate was a combination of 8 genes and 4 biomarkers with an AUC of 95.1% (95% CI 92.0% - 98.2%), which included 98.9% (96.7% - 100%) of patients with stage III-IV epithelial ovarian cancer. This study population should be considered a derivation population; the algorithm needs to be further tested in a larger study population.
Reviewer
Linda Speer, MD
Professor and Chair, Department of Family Medicine
University of Toledo
Toledo, OH
Comments
Markers for pelvic masses
Hopefully this testing can become screening for things like ovarian cancer, where diagnosis is often made too late for cure. Hopefully the cost of this screening won't be too high.
Pelvic mass malignancy
As above in the problem with information.
Predicting malignancy in women with pelvic mass
8 genes and 4 biomarkers mean a lot of money to spend on a dubious way to investigate a pelvic mass. Biopsy would be more cost effective and will not just predict but diagnose.