Ovarian Cancer Biomarkers

Indications for Testing

Laboratory testing for ovarian cancer biomarkers is indicated in patients with an adnexal/pelvic mass, other signs of ovarian cancer (eg, ascites), or symptoms of ovarian cancer (eg, abdominal pain, increased urinary frequency or urgency, feeling full quickly). 

Germline genetic testing for hereditary ovarian cancer syndromes should be offered to individuals who are diagnosed with and/or have a family history of ovarian cancer. For more information on when to perform testing for hereditary ovarian cancer syndromes, see Familial Risk Genetic Testing.

Laboratory Testing

Serum Biomarker Testing in Diagnosis, Prognosis, and Monitoring

Diagnosis, staging, and prognosis of ovarian cancer are largely determined through surgical pathology.  However, a preoperative assessment that includes certain biomarkers (described in the following sections), in addition to imaging, physical examination, and some basic laboratory tests such as CBC, chemistry profile, and liver function tests, may provide a gynecologic oncologist with information useful in determining if surgery is the right option for a particular patient.  These biomarkers may also have prognostic value and can be used to monitor treatment response and to monitor for recurrence.  If serial testing is required (eg, for monitoring disease course or for recurrence), all testing should be conducted by the same method at the same laboratory to avoid confounding clinical evaluation and treatment decisions.

Cancer Antigen 125

Testing for cancer antigen 125 (CA-125) concentrations in serum is recommended as part of the preoperative workup for ovarian cancer because CA-125 levels typically correlate with the extent of disease and may be useful in treatment planning. 

This preoperative assessment of CA-125 may be useful in determining prognosis. Higher CA-125 concentrations have been associated with worse survival rates in some studies,  and CA-125 concentrations have generally been shown to correlate with disease course.  Thus, a preoperative assessment of CA-125 can provide useful prognostic information. 

CA-125 is useful in monitoring treatment response in patients with ovarian cancer.   In patients with initially elevated CA-125, concentrations generally correlate with disease activity,  decreasing during treatment response and increasing in cases of disease progression and drug resistance.  However, CA-125 concentrations may return to normal even in the presence of persistent disease, so measuring CA-125 is not an optimal approach to determine if therapy should be terminated. 

If initially elevated, CA-125 may be useful in monitoring for recurrence.  In patients with epithelial tumors, a CA-125 measurement should be considered every 2-4 months for 2 years, every 3-6 months for years 3-5, and then annually thereafter.  In borderline tumors, a CA-125 measurement should be considered every 3-6 months for up to 5 years and then annually thereafter. 

Human Epididymis Protein 4

Some evidence suggests that human epididymis protein 4 (HE4) concentrations may have prognostic value for epithelial ovarian cancer,   and some studies have indicated that high HE4 levels may be a stronger marker of worse prognosis than high CA-125 levels.  However, results are not consistent across studies, and HE4 testing is not recommended as a routine element of the workup for ovarian cancer.  

Serum HE4 concentrations decrease in patients who respond to treatment and may be useful for monitoring treatment, either in conjunction with CA-125 or in patients without elevated CA-125.   Additionally, HE4 concentrations may be useful in monitoring for recurrence in individuals without elevated CA-125 at initial diagnosis.  

Other Biomarkers

Testing for other biomarkers may be useful in certain clinical scenarios, specifically in the assessment of certain nonepithelial ovarian cancers.

­­Biomarker Testing for Treatment Decision-Making

The presence of certain markers can help inform treatment decisions, and after diagnosis and staging, testing for these markers should be performed on tumor tissue. 

Somatic BRCA1 and BRCA2 Variants

Although assessment for germline pathogenic variants in BRCA1 and BRCA2 can be used to determine cancer susceptibility (see Familial Risk Genetic Testing), detection of somatic variants in these genes may also assist in treatment decision-making.   If no germline variants in BRCA1 and BRCA2 are detected, then somatic tumor testing for BRCA1 and BRCA2 variants should be performed.  BRCA1 and BRCA2 mutations, both germline and somatic, cause homologous recombination deficiencies (HRDs), which are associated with higher rates of response to treatment by poly (ADP-ribose) polymerase (PARP) inhibitors.  Next generation sequencing (NGS) can be used to detect somatic variants in BRCA1 and BRCA2. 

Other genetic alterations in the homologous recombination (HR) pathway can also result in HRDs.  An assessment for variants in other HR genes (eg, RAD51C/D, PALB2, and BRIP1) may be considered  ; however, apart from BRCA1 and BRCA2 testing, HRD testing does not adequately differentiate patient response to PARP inhibitors, and its routine use is not recommended.  

Mismatch Repair/Microsatellite Instability

Mismatch repair (MMR) or microsatellite instability (MSI) testing can be used to assist in treatment decisions and screen for Lynch syndrome/hereditary nonpolyposis colorectal cancer (HNPCC). Patients with MMR-deficient (dMMR) or MSI-high (MSI-H) tumors are eligible for treatment with pembrolizumab.   Somatic testing for MMR deficiency should be offered to patients with clear cell, mucinous, or endometrioid ovarian cancer, and may be offered to patients diagnosed with other subtypes of epithelial ovarian cancer.  Immunohistochemistry (IHC) can be used to detect DNA MMR protein, and polymerase chain reaction (PCR) testing can be used to test for MSI. 

Familial Risk Genetic Testing

Germline Genetic Testing in Ovarian Cancer Workup

When a diagnosis of ovarian cancer is confirmed, the patient should undergo a genetic risk evaluation, including germline testing for hereditary ovarian cancer syndromes.   Approximately 25% of ovarian cancers are the result of a heritable condition,  and at least 10% of epithelial ovarian cancers are due to specific pathogenic germline variants in the BRCA1 or BRCA2 genes. 

Although BRCA1 and BRCA2 are the best characterized and most prevalent ovarian cancer susceptibility genes, other genes (eg, ATM, BRIP1, NBN, PALB2, RAD51C, RAD51D, and STK11) are also associated with an increased susceptibility to ovarian cancer.  Additionally, evidence suggests that Lynch syndrome confers an increased risk of ovarian cancer in women, especially in those with pathogenic variants in the MLH1, MSH2, or MSH6 genes. 

Germline Genetic Testing in Other Situations

Individuals considered to be at risk for hereditary ovarian cancer should be offered genetic counseling, even if they are not currently being assessed for ovarian cancer.    According to the National Comprehensive Cancer Network (NCCN), genetic testing is clinically indicated for individuals:

• With any blood relative with a known pathogenic/likely pathogenic variant in a cancer susceptibility gene 
• With a personal or family history of breast, ovarian, pancreatic, and/or metastatic prostate cancer who meet additional specific criteria 
• Who have tested negative with limited testing (eg, single-gene testing and/or absent deletion/duplication analysis) but are interested in pursuing multigene testing 
• In whom a pathogenic variant is identified on tumor genomic testing that has clinical implications if present in the germline 
• Who are affected with cancer, to aid in making decisions concerning systemic therapy 

Testing may be considered in individuals:

• With multiple primary breast cancers first diagnosed between the ages of 50 and 65 years
• With Ashkenazi Jewish ancestry
• Individuals (either affected or unaffected) who do not meet any of the criteria for genetic testing but who have a 2.5-5% probability of having a BRCA1 or BRCA2 pathogenic variant based on probability models (eg, Tyrer-Cuzick, CanRisk, BRCAPro)

Refer to the Hereditary Cancer Testing Criteria in the NCCN guidelines for the complete criteria.