New blood test shows promise in early detection of ovarian cancer

By Dr. Liji Thomas, MDReviewed by Benedette Cuffari, M.Sc.Apr 28 2024

Epithelial ovarian cancer (EOC) accounts for the most significant number of deaths from female reproductive system cancers.

Although effective treatments are available at early stages of EOC, timely diagnosis is difficult due to nonspecific presentation and signs. Thus, there has been intense interest in developing systems that can identify patients at high risk for a diagnosis of EOC before the disease spreads and becomes advanced.

A new paper published in the British Journal of Cancer explores the potential utility of glycoproteins as biomarkers of EOC. In the future, these types of models might allow the diagnosis of EOC to be made from a blood test.

Study: Diagnosing and staging epithelial ovarian cancer by serum glycoproteomic profiling. Image Credit: Shiflovski / Shutterstock.com

Existing diagnostic guidelines

Currently, a sequence of diagnostic tests is used to stage EOC, including imaging and blood tests for CA-125, an epithelial surface protein shed from EOC cells. However, these approaches lack specificity, and their sensitivity remains low in early EOC.

Moreover, about 20% of EOCs do not have increased CA-125. In fact, CA-125 levels can be raised in other non-cancerous gynecological conditions like endometriosis and uterine fibroids. Thus, CA-125 can be a more reliable marker for monitoring cancer progression and treating EOCs.

Early EOC is frequently missed, with non-cancerous conditions that present with a pelvic mass often prioritized for surgery. Ultimately, only about 20% of all pelvic masses are malignant, whereas up to 33% of early EOCs are more advanced following histological examination.

There is an unmet clinical need for sensitive and specific tests for early detection, staging of ovarian cancer, and for assessing malignant versus benign pelvic masses.”

The current study used a case-control design to analyze serum glycoproteins by mass spectrometry, along with artificial intelligence (AI), to identify those that can distinguish EOC from other conditions and differentiate late from early EOC. A classification system using these markers was subsequently constructed and used to explore the underlying mechanisms that might account for these changes.

EOC vs. non-cancerous masses

A total of 27 differential biomarkers were identified, including several associated with ovarian cancer, such as alpha-1-antichymotrypsin, alpha-1-acid glycoprotein 1, and immunoglobulin G (IgG).

Incorporating these glycoproteins into a novel classification system for EOS showed high accuracy, sensitivity, and specificity of over 85% each, which increased with the cancer stage. This model also performed creditably when applied to healthy patients, as it presented values similar to those observed with benign tumors.

A prospective study on women with pelvic masses using this model showed similar expression patterns of these biomarkers. In this study, the accuracy, specificity, and sensitivity of the model was reduced to 72%, about 50%, and 74%, respectively.

Fucose residues and EOC progression

Fucosylation appears to be increased in EOC, with tri- and tetra-antennary fucosylated N-glycans exhibiting the greatest association with EOC in a linear manner from early to late stages. As the tumor advanced in stage, there was a progressive shift from non-fucosylated to fucosylated forms of these glycopeptides.

These specific markers could help diagnose late-stage EOC, as a model classifier based on these proteins achieved over 90% accuracy and 100% sensitivity despite low specificity. With metastatic EOC, fucosylation continued to rise, whereas observed differential gene expression suggested highly branched fucosylated N-glycans.

Cytokines and EOC

Late-stage and metastatic EOC was accompanied by changes in serum cytokines, including interleukin 6 (IL-6), IL-8, IL-10, and monocyte chemoattractant protein 1 (MCP-1).

These biomarkers, which typically originate from the liver and circulating immune cells, have been observed in both late-stage EOC and the peripheral tissues of affected individuals. Thus, common factors that are triggered by a change in cytokine levels underlie the common glycosylation processes at these sites.

Conclusions

As noninvasive triaging and staging can influence treatment protocols for women with pelvic masses, these data represent an important step towards the advancement of EOC diagnostics.”

For early diagnosis, blood glycoproteins modified by the presence of the disease that evokes a systemic response are preferable to cell-free or circulating tumor glycoproteins like CA-125.

Even biomarkers that are generated in the very early stages of disease can be efficiently detected by our analytical platform, allowing the diagnosis of disease in the early stage.”

Further research is needed to validate these biomarkers and the models using larger, prospective, and independent cohorts. The role of these markers in the development and progression of these tumors should also be elucidated.