CA-125 Test Accuracy for Ovarian Cancer Prediction Varies by Ethnicity

The cancer antigen (CA)–125 test's accuracy in predicting ovarian cancer varies by ethnicity, mainly due to factors like age and comorbidities rather than ethnicity itself, according to a study published in the Journal of Ovarian Research.¹

Ovarian cancer is the sixth most common cancer in female patients in the United Kingdom (UK), with about 7500 new cases and over 4000 deaths annually.² Clinical features of suspected ovarian cancer include increased urinary frequency, the presence of ascites, and an abdominal/pelvic mass.¹ Also, levels of CA-125, a protein biomarker, are often elevated in patients with ovarian cancer.

General practitioners (GPs) can offer a CA-125 test to those with suspected ovarian cancer symptoms, which measures CA-125 levels in the blood. Guidelines recommend further ovarian cancer investigation if the CA-125 levels are 35 units (U)/mL or above.³

Although there have been several studies on the diagnostic performance of CA-125 for the detection of ovarian cancer, little is known about its performance in women from diverse ethnic backgrounds.¹ Therefore, the researchers conducted a study to examine the association between patient ethnicity and ovarian cancer diagnosis following a CA-125 test.

The accuracy of the cancer antigen (CA)-125 test in predicting ovarian cancer varies by ethnicity, primarily due to factors like age and comorbidities rather than ethnicity itself. | Image Credit: jarun011 - stock.adobe.com

To do so, they used electronic health records from the Clinical Practice Research Datalink (CPRD) Aurum dataset, which contains information on patient demographics, blood test results, diagnoses, and prescriptions; as of May 2022, it contained 13,300,067 currently registered patients, representing 19.8% of the UK population.

Eligible patients were women 40 years or older registered with a GP practice in England that contributed to CPRD between 2010 and 2017. They also needed an ethnicity record and a valid CA-125 test result recorded during the study period. The researchers categorized patient ethnicities into 5 broad groups: White, Asian, Black, mixed, and other.

The study population consisted of 328,201 women. Most patients were White (90%), followed by Asian (5%) and Black (3%). Conversely, women from the other and mixed groups represented less than 1% of the study population. Similarly, 2434 (< 1%) patients were included with records of unknown ethnicity.

The 1-year ovarian cancer incidence was 0.8%, with 2756 women receiving a diagnosis within 1 year of their CA-125 test (2589 White patients, 43 Black patients, 10 other patients, 10 mixed patients, and 23 patients with unknown ethnicity). However, there were too few ovarian cancer cases among the mixed and other groups to draw meaningful ovarian cancer incidence comparisons; as a result, the researchers dropped the women of these groups from the analysis.

Overall, White women were almost twice as likely to be diagnosed with ovarian cancer than Asian and Black women (0.9% vs 0.5% for both Asian and Black women). In terms of ovarian cancer stage at diagnosis, Black women were more likely to be diagnosed at an advanced stage than White or Asian women (67% vs 58% for both White and Asian women).

Also, the researchers determined that a high CA-125 result had the highest sensitivity in ovarian cancer detection among Black women (90.7%; 95% CI, 90.1%-91.2%) vs White (79.4%; 95% CI, 79.3%-79.6%) or Asian (71.6%; 95% CI, 70.9%-72.3%) women. Conversely, the positive predictive values (PPVs) of a high CA-125 result were approximately twice as high for White women (10.2%; 95% CI, 10.1%-10.3%) than for Asian (4.9%; 95% CI, 4.6%-5.3%) or Black (5.6%; 95% CI, 5.2%-6.1%) women.

However, when adjusted for all covariates, the estimated PPVs were similar among White (9.31%; 95% CI, 8.83%-9.82%), Black (8.39%; 95% CI, 6.12%-11.4%), and Asian (7.84%; 95% CI, 6.06%-10.1%) women. Similarly, the adjusted PPVs of an advanced stage diagnosis following a raised CA-125 result were similar among White (75.9%; 95% CI, 72.6%-78.9%), Black (66.8%; 95% CI, 41.4%-85.1%), and Asian (69.6%; 95% CI, 72.3%-93.7%) women. Therefore, the degree of association between an elevated CA-125 result and being diagnosed with an advanced ovarian cancer stage did not differ by ethnicity.

Lastly, the overall estimated ovarian cancer incidence at a CA-125 level of 35 U/mL was 2.4% (95% CI, 2.3%-2.6%), and the estimated CA-125 level that corresponded to a 3% ovarian cancer incidence was 43 U/mL (95% CI, 40-45). The researchers reported that the figures for both the estimated ovarian cancer incidence at a CA-125 level of 35 U/mL (2.5%; 95% CI, 2.3%-2.7%) and the estimated CA-125 level that corresponded to a 3% ovarian cancer incidence (41 U/mL; 95% CI, 39-44 U/mL) were similar for White women.

However, the predicted ovarian cancer incidence for Asian women following a CA-125 level of 35 U/mL was half the predicted incidence for White women (1.8%; 95% CI, 0.8%-1.8%). Similarly, Asian women reached the estimated 3% ovarian cancer incidence rate at higher CA-125 levels (69 U/mL; 95% CI, 52-89).

The researchers acknowledged their limitations, one being that they could not determine whether women from different ethnic groups were offered or accepted CA-125 tests at equal rates. Also, they did not account for interethnic differences, like those between East and South Asian ethnic groups. Despite their limitations, the researchers suggested areas for future research based on their findings.

“Future research should consider whether other diagnostic tests perform equally well for all ethnic groups and should also consider the socioeconomic challenges that some patients may face when accessing health care,” the authors concluded.

References

1. Barlow M, Down L, Mounce LTA, et al. The diagnostic performance of CA-125 for the detection of ovarian cancer in women from different ethnic groups: a cohort study of English primary care data. J Ovarian Res. 2024;17(1):173. oi:10.1186/s13048-024-01490-5
2. Ovarian cancer statistics. Cancer Research UK. Accessed August 28, 2024. https://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/ovarian-cancer
3. Ovarian cancer: recognition and initial management. National Institute for Health and Care Excellence (NICE). April 27, 2011. Accessed August 28, 2024. https://www.nice.org.uk/guidance/cg122/