Early Decline in Cancer Antigen 125 as a Surrogate for Progression-Free Survival in Recurrent Ovarian Cancer

• Published in: JNCI : Journal of the National Cancer Institute 2011-09, Vol.103 (17), p.1338-1342
• Main Authors: Lee, Chee K., Friedlander, Michael, Brown, Chris, Gebski, Val J., Georgoulopoulos, Alexander, Vergote, Ignace, Pignata, Sandro, Donadello, Nicoletta, Schmalfeldt, Barbara, Delva, Rémy, Mirza, Mansoor Raza, Sauthier, Philippe, Pujade-Lauraine, Eric, Lord, Sarah J., Simes, R. John
• Format: Article
• Language: English
• Subjects: Adult; Aged; Antigens; Antineoplastic Combined Chemotherapy Protocols - therapeutic use; Biological and medical sciences; Biomarkers, Tumor - blood; CA-125 Antigen - blood; Carboplatin - administration & dosage; Chemotherapy; Clinical outcomes; Confidence Intervals; Disease-Free Survival; Doxorubicin - administration & dosage; Doxorubicin - analogs & derivatives; Female; Female genital diseases; Gynecology. Andrology. Obstetrics; Humans; Kaplan-Meier Estimate; Medical sciences; Middle Aged; Multivariate Analysis; Odds Ratio; Oncology; Ovarian cancer; Ovarian Neoplasms - drug therapy; Ovarian Neoplasms - immunology; Ovarian Neoplasms - pathology; Paclitaxel - administration & dosage; Polyethylene Glycols - administration & dosage; Predictive Value of Tests; Prognosis; Proportional Hazards Models; Recurrence; Reproducibility of Results; Time Factors; Tumors
• ISSN: 0027-8874; 1460-2105
• DOI: 10.1093/jnci/djr282

We used data from 886 patients from the CAELYX in Platinum Sensitive Ovarian Patients (CALYPSO) trial, recruited between April 2005 and September 2007, to examine the role of early decline in cancer antigen 125 (CA125) and early tumor response as prognostic factors and surrogates for superiority of treatment with carboplatin-pegylated liposomal doxorubicin (CPLD) compared with carboplatin-paclitaxel (CP) in a landmark analysis. Progression-free survival (PFS) was estimated by Kaplan-Meier analyses. We used univariate and multivariable Cox proportional hazards analyses to assess early decline and early response as surrogates for CPLD treatment benefit compared with CP. All statistical tests were two-sided. Early decline (defined as rate of CA125 decrease of at least 50% per month) was associated with improved PFS (adjusted hazard ratio [HR] for progression = 0.81, 95% confidence interval [CI] = 0.67 to 0.97, P = .02) but early response (complete or partial responses) was not. CPLD was associated with improved PFS compared with CP (HR = 0.82, 95% CI = 0.69 to 0.96, P = .01). However, fewer CPLD patients had an early decline (161 [37.4%] vs 233 [51.2%], P < .001) or an early response (146 [33.9%] vs 176 [38.7%], P = .14) compared with CP patients. The PFS for CPLD patients did not change statistically significantly after adjustment for early decline (adjusted HR = 0.80, 95% CI = 0.68 to 0.94, P = .007). These findings are opposite to what would be expected if these markers were good surrogates for treatment benefit.