CA-125 Shows Poor Concordance With Radiologic Progression in Ovarian Cancer

Current treatment guidelines recommend using cancer antigen-125 (CA-125) as a diagnostic marker for ovarian cancer recurrence, but a recent study published in Journal of Clinical Oncology suggests CA-125 has poor concordance with progressive disease (PD) based on RECIST criteria in women with platinum-sensitive relapsed ovarian cancer (PSROC) who responded to chemotherapy and received maintenance PARP inhibitors (PARPi).

Clinical assessment, CA-125, and imaging are typically used to diagnose PSROC, and CA-125 is often utilized both in trials and clinical settings to evaluate tumor response and disease progression. If pretreatment CA-125 is within range or is elevated then normalizes, progression is defined as at least a doubling above the upper limit of the normal range. When a patient has elevated CA-125 on 2 separate readings before treatment, a doubling of the nadir CA-125 value is considered progression. RECIST PD is based on CT imaging, which is typically performed every 8 to 12 weeks in PARPi maintenance clinical trials.

Female reproductive system | Image credit: Aliaksandr Marko - stock.adobe.com

“Accurate detection of cancer progression leading to cessation of PARPi therapy is important to avoid unnecessary adverse effects and the high cost of this treatment,” the authors explained. “Furthermore, cancer progression affects both prognosis and decisions regarding next-line therapy, all of which need to be communicated and discussed with the patient.”

The analysis pooled data from 4 randomized trials of PARPi maintenance therapy vs placebo: Study 19 (NCT00753545), SOLO2 (NCT01874353), ARIEL3 (NCT01968213), and NOVA (NCT01847274). Data on CA-125 PD based on Gynecologic Cancer InterGroup criteria and PD based RECIST v1.1 were extracted and analyzed to determine the concordance between CA-125 and RECIST PD. The researchers assessed both the negative predictive value (NPV) and positive predictive value (PPV) of the measures.

A total of 1262 study participants were included in the analysis, including 818 who received PARPi treatment and 444 who received a placebo. Median follow-up duration was 8.4 months in Study 19, 22.2 months in SOLO2, 41.3 months in ARIEL3, and 16.9 months in the NOVA trial.

CA-125 PD was detected in 403 (32%) patients, 366 of whom had concordant RECIST PD, suggesting a PPV of 91% (95% CI, 88-93). However, the authors noted that the PPV was lower when looking at blinded independent central review assessment (76%). But among 859 patients without CA-125 PD, only 382 did not show RECIST PD, suggesting a NPV of 44% (95% CI, 41-48).

PPV for CA-125 was high in both the placebo (PPV, 91%; 95% CI, 86-95) and treatment (PPV, 91%; 95% CI, 86-94) groups. However, NPV was higher in the treatment group (NPV, 53; 95% CI, 49-57) vs the placebo group (NPV, 25%; 95% CI, 20-31).

Most of the patients (95%) who showed PD based on RECIST criteria but not CA-125 also had normal CA-125 at the start of maintenance therapy. For more than half of those patients (64%), CA-125 remained within normal ranges. Among those who had RECIST-only PD vs RECIST and CA-125 PD, solid organ recurrence without peritoneal disease was more common (36% vs 24%; P < .001).

Study limitations included the derivation of data from studies not prospectively designed for the analysis, as well as a lack of data on CA-125 values at the time of diagnosis. Additionally, excluding women with tumors that do not secrete CA-125 may improve the NPV.

Overall, the findings have significant implications for research and existing clinical practice guidelines, including current National Comprehensive Cancer Network guidelines, which recommend follow-up CA-125 monitoring and imaging if clinically indicated, according to the authors.

“Among patients with PSROC treated with maintenance PARPi therapy, there is poor NPV between CA-125 and imaging, regardless type of PARPi or HRD status,” the authors concluded. “CA-125 monitoring alone is insufficient for surveillance, particularly when CA-125 is not elevated, and periodic imaging should be considered to assess for radiologic progression.”

Reference

Tjokrowidjaja A, Friedlander ML, Ledermann JA, et al. Poor concordance between cancer antigen-125 and RECIST assessment for progression in patients with platinum-sensitive relapsed ovarian cancer on maintenance therapy with a poly(ADP-ribose) polymerase inhibitor. J Clin Oncol. Published online January 12, 2024. doi:10.1200/JCO.23.01182