Change in blood protein detects more ovarian cancers than fixed threshold, study findsBMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h2367 (Published 05 May 2015) Cite this as: BMJ 2015;350:h2367
Estimating a woman’s risk of ovarian cancer by the speed at which the level of a blood protein increases is more accurate than using a fixed threshold for the protein level, a study has found.1
For many years researchers have been studying levels of a protein linked to ovarian cancer called CA125, which can help to diagnose the disease at an early stage. Previously a fixed cut-off has been used for CA125 (more than 35 U/mL) to identify a possible abnormality. But some women can have much higher levels and still not have the cancer, while others with levels below this threshold could be harbouring the disease.
In the latest study, researchers from University College London compared the accuracy of a statistical calculation to interpret the changing level of CA125 with the conventional screening method that used a fixed value for CA125.
The results came from an analysis of one arm of the United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS), the world’s largest ovarian cancer screening trial. It involved 202 638 post-menopausal women aged 50 or over who were randomly assigned to multimodal screening with serial CA125 testing, transvaginal ultrasound, or no test at all.
The latest study, published in the Journal of Clinical Oncology, evaluated 46 237 women who continued to attend annual multimodal screening after the first screen.1 Their blood was tested once a year for CA125 levels, and a computer algorithm was then used to interpret their risk of ovarian cancer by looking at factors including the woman’s age, the original levels of CA125, and how that level changed over time. The serial pattern was compared with known cases of cancer and controls to estimate the risk of ovarian cancer.
Over 13 years 154 women were found to have primary invasive epithelial ovarian/tubal cancers. The new screening method for detecting these ovarian cancers was found to be 85.8% sensitive (95% confidence interval 79.3 to 90.9) and 99.8% specific (99.8 to 99.8). By comparison, the conventional test using a CA125 cut-off of >35 U/mL would have identified 41% of cancers, and a cut-off of >30 U/mL would have detected 48%.
The researchers concluded that “current findings are of immediate importance as they highlight the need to examine serial change in biomarker levels in the context of screening and early detection of cancer. Reliance on predefined single threshold rules may result in biomarkers of value being discarded.”
Usha Menon, UKCTOCS co-principal investigator and trial coordinator at University College London, said, “The numbers of unnecessary operations and complications were within acceptable limits and we were able to safely and effectively deliver screening for over a decade across 13 NHS trusts.
“While this is a significant achievement, we need to wait until later this year when the final analysis of the trial is completed to know whether the cancers detected through screening were caught early enough to save lives.”
Cite this as: BMJ 2015;350:h2367