Could a Simple Biomarker Panel Improve Ovarian Cancer Detection?

August 23, 2019

Using two validation methods, researchers found that a simple biomarker panel may have strong predictive ability for detecting ovarian cancer.

A simple biomarker panel may be able to detect epithelial ovarian cancer (EOC) one to two years earlier than existing methods, according to a new study published online ahead of print on August 7 in the British Journal of Cancer. The study included a small number of cases and thus the method does still require further validation.

“The non-specific nature of early EOC symptoms combined with the rarity of the disease presents a major barrier to increasing rates of detection of pre-clinical disease through the primary care setting,” wrote study authors led by Matthew R. Russell, PhD, of the University of Manchester in the United Kingdom. “Achieving a mortality reduction in EOC will require a screening strategy capable of triggering intervention early enough to alter the natural history of ovarian cancer.”

The new panel includes measurement of four biomarkers: CA125, phosphatidylcholine-sterol acyltransferase, vitamin K-dependent protein Z, and C-reactive protein. It involves assigning a score based on the dysregulation of each protein from baseline for each tested individual.

In total, the researchers included 49 patients with EOC who were enrolled in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS), along with 31 control individuals, representing a total of 482 serial samples. In that trial, serum samples were collected over up to a 7-year span, allowing the discovery of biomarkers in serum prior to EOC diagnosis.

The panel yielded an area-under-the-curve (AUC) for EOC samples less than one year prior to diagnosis of 0.971 compared to the control patients. For those less than 2 years from diagnosis, the AUC was 0.920 compared to controls. For those between 1 and 2 years from diagnosis, the AUC was 0.848 compared to the control patients.

The authors noted that CA125 made the largest contribution to the panel’s sensitivity, while the other three biomarkers combined to match CA125’s total contribution.

They also used two validation methods (leave-one-out cross-validation and k-fold cross-validation) and found the panel did have strong predictive ability. They determined using a simulated screening program that women could be assigned classifications including severe, elevated, intermediate, or normal based on panel testing, which could help guide treatment decisions.

A direct comparison with existing screening algorithms was not possible given the retrospective nature of the analysis.

“However,” the authors wrote, “the signal identified in the protein panel is certainly real and warrants further investigation in independent sample sets both in Type II and Type I EOCs.”

Rachel Shaw, PhD, a research information manager at Cancer Research UK, noted that approximately half of ovarian cancer cases are detected at a late stage, when treatment is less likely to be effective. “developing simple tests like these that could help detect the disease sooner is essential,” she said in a press release. " It's really exciting to see these encouraging results for this type of ovarian cancer."

Currently, the U.S. Preventive Services Task Force recommends against screening for ovarian cancer in asymptomatic adults. Their review found that screening with transvaginal ultrasound, as well as testing for dysregulation of CA-125, does not reduce ovarian cancer mortality, and that screening could result in “important harms” including substantial false positives.