Risk-stratified CA125 screening integrating CA125 trajectories, trajectory-specific progression and transvaginal ultrasound for ovarian cancer.

Abstract

Backgrounds: Cancer antigen 125 (CA125) is widely used for screening ovarian cancer (OC), yet its effectiveness remains debated. Potential factors may include ineffective cut-off value for CA125 in screening, as well as a lack of consideration for CA125 trajectories and trajectory-specific progression.

Methods: Based on data from multiple rounds of CA125 tests and transvaginal ultrasound (TVU) examinations conducted on 28,456 women in the PLCO Trial, time-dependent receiver-operating-characteristic curves (ROCs) and area-under-the-curves (tdAUCs) analyses were employed to identify the optimal CA125 cut-off values for OC screening. Participants were categorized into four CA125 trajectories: stable negative CA125 (CA125_SN), loss of positive CA125 (CA125_LP), stable positive CA125 (CA125_SP), and gain of positive CA125 (CA125_GP). The associations between different CA125 trajectories, trajectory-specific progression indicators, and OC risk were explored. The effectiveness of risk-stratified CA125 screening, incorporating CA125 trajectories, trajectory-specific progression, and TVU, was evaluated using hazard ratio and 95% confidence intervals [HR (95%CIs)], with adjustments for potential confounders.

Results: After a median follow-up of 14.8 years for OC incidence and 23.8 years for OC mortality, 250 OC cases and 218 OC deaths were identified. The tdAUC for 10-year OC incidence with CA125 was 0.663, with an optimal cut-off value of 13.00 U/ml. Trajectory analyses showed that both CA125_SP and CA125_GP were significantly associated with increased risks of OC incidence [HRs (95%CIs): 2.00(1.47-2.73) and 3.06(2.25-4.16)] and mortality [HRs (95%CIs):1.58(1.13-2.21) and 2.60(1.87-3.62)] compared to CA125_SN. Trajectory-specific progression analyses identified relative velocity as the optimal progression indicators for both CA125_SP and CA125_GP (tdAUCs: 0.712 and 0.767), with optimal cut-off values of 9% and 32% per year, respectively. Positive progression was associated with significantly increased risks of OC incidence [HRs (95%CI): 7.26(4.00-13.17) and 3.83(1.96-7.51) CA125_GP and CA125_SP] and mortality [HRs (95%CI): 8.03(4.15-15.56) and 6.04(2.78-13.14)] compared to negative progression. Optimized risk-stratified CA125 screening, which integrated CA125 trajectories, trajectory-specific progression, and TVU, reduced missed OC by 3.6% and improved accuracy compared to traditional screening methods.

Conclusions: Incorporating CA125 trajectories and trajectory-specific progression into screening protocols enhances the identification of the population at high-risk of OC. An optimized screening strategy, which includes these factors along with TVU, is recommended to improve the effectiveness of OC screening.