Abstract
Aim: To correlate CA125 levels after platinum- and paclitaxel-based chemotherapy with progression-free survival (PFS) and overall survival (OS) in advanced ovarian carcinoma following primary cytoreduction. Materials and Methods: The study was conducted on 247 patients. Results: By log-rank test, PFS and OS were related to performance status (PS) (p=0.04 and p=0.00002), residual disease (p=0.00002 and p=0.001), ascites (p=0.00001 and p=0.0003) and post-chemotherapy CA125 using 10.8 U/ml as cut-off (p=0.0001 and p=0.01). PFS was related to post-chemotherapy CA125 assay (cut-off values of 7.1 U/ml (p=0.02), 18.5 U/ml (p<0.000001) and 35.0 U/ml (p=0.0001)). OS was related to FIGO stage (p=0.02). On multivariate analysis, residual disease, ascites and post-chemotherapy CA125 with any selected cut-off were independent prognostic variables for PFS, whereas residual disease, PS and post-chemotherapy CA125 (10.8 U/ml as cut-off) were independent prognostic variables for OS. Conclusion: Post-operative CA125 using 10.8 U/ml as cut-off was an independent prognostic variable for both PFS and OS.
Ovarian carcinoma represents the sixth most common cancer among women worldwide and it causes more deaths per year than any other cancer of the female genital tract (1). Primary cytoreductive surgery followed by platinum- and paclitaxel-based chemotherapy is the standard therapeutic approach for advanced ovarian carcinoma, able to achieve a clinical complete response rate of approximately 50%, a pathological complete response rate of 25-30%, a median progression-free survival (PFS) of 15.5-22 months and a median overall survival (OS) of 31-44 months (2-6). Almost 75% of the patients who achieve a clinical complete response and 50% of those who obtain a pathological complete response will relapse after a median time of 18-24 months (7). Phase III studies have evaluated multi-drug combinations, dose-dense weekly scheduling, intraperitoneal delivery, neo-adjuvant chemo therapy, maintenance therapy and targeting of angiogenesis (8). Incremental gains in median PFS or OS have been sometimes achieved, but without an impact on overall mortality.
The persistence of grossly visible residual disease after primary cytoreduction is the strongest poor prognostic variable, whereas FIGO stage IV disease, high histological grade, presence of ascites and poor performance status (PS), are other significant unfavorable factors in most, but not in all, studies (5, 6, 9-12).
Whereas it is commonly accepted that serum CA125 kinetics during early chemotherapy is a strong predictor of clinical outcome (9, 12, 13), the prognostic value of CA125 level after the end of chemotherapy is still debated (12, 14-19).
In the present investigation, serum CA125 levels after the sixth cycle of chemotherapy were related to PFS and OS of patients with advanced ovarian carcinoma treated with primary cytoreduction and platinum- and paclitaxel-based chemotherapy.
Materials and Methods
We reviewed the hospital records, including surgical notes and pathological reports, of 247 patients with FIGO stage III-IV ovarian carcinomas who underwent primary cytoreductive surgery followed by six cycles of platinum- and paclitaxel-based chemotherapy at our Department between March 1996 and March 2013. Patients who underwent neoadjuvant chemotherapy followed by interval de-bulking surgery, as well as those with borderline ovarian tumours, were excluded from the analysis.
The tumor stage and histological diagnosis of each case were determined according to FIGO criteria and the histological typing system of the World Health Organization, respectively (11). Tumours were graded as: well (G1), moderately (G2) or poorly (G3) differentiated.
After the sixth cycle of chemotherapy, patients with no evidence of disease at clinical, sonographic and radiological examination and with serum CA125 <35 U/ml were defined in clinical complete response. Three to five weeks after the end of chemotherapy, a second-look laparotomy or laparoscopy was often proposed to clinically complete responders until 2006, mostly to patients enrolled onto clinical trials. A pathological complete response at second-look was defined as the disappearance of all macroscopic tumor deposits with negative peritoneal washing and negative multiple random biopsies. A microscopic residual disease was defined as the disappearance of all macroscopic tumor deposits with positive peritoneal washing and/or positive peritoneal biopsies. Patients who had a macroscopic residuum, which was completely resected during second-look, were defined as converted complete responders. All patients with clinically or surgically detectable persistent disease, all converted complete responders, as well as some clinically or pathologically complete responders, received additional chemotherapy as consolidation treatment.
All patients were periodically followed-up until they died or until February 2014. An asymptomatic patient with rising CA 125 levels and negative clinical and imaging examination results was not still considered to have a recurrent disease and underwent a more stringent follow-up program. The median follow-up of survivors was 43 months (range, 7-183 months).
The serum levels of CA125 were determined with immunoradiometric assays or with enzyme immunoassays using commercially available kits. The manufacturer's reagents were used with standard quality control procedures to ensure comparability. The upper limit of normal was 35 U/ml for all assays used.
Statistical analysis. The SPSS ver.13 Inc (Chicago, IL, USA) was used for computations. The time from first surgery to clinical or radiological diagnosis of progression was defined as PFS. The time from first surgery to death or last observation was defined as OS.
The analysed prognostic variables were FIGO stage, patient age, PS, histological type, tumor grade, presence or absence of ascites, residual disease after primary surgery, serum post-chemotherapy CA 125, second-look surgery and consolidation treatment.
The cumulative probability of PFS and OS were estimated by the product-limit method. The log-rank test was used to compare the homogeneity of PFS and OS functions across strata defined by categories of prognostic variables. A multiple regression analysis based on the Cox proportional hazard model was used to test jointly the relative importance of variables as predictors of PFS and OS times.
Results
The median age of patients was 59 years (range=26-78 years). One hundred and seventy-five patients (70.8%) were <65 years old and 72 were >65 years old. PS was 0 in 174 patients (70.4%). According to the FIGO classification, tumor stage was IIIa in 14 patients (5.7%), IIIb in 17 (6.9%), IIIc in 198 (80.1%) and IV in 18 (7.3%). Histologically, 200 (81.0%) carcinomas were serous, 15 (6.1%) endometrioid, 11 (4.5%) mixed, 10 (4.0%) mucinous, 6 (2.4%) clear cell and 5 (2.0%) undifferentiated. Tumor grade was G1 in 8 patients (3.2%), G2 in 62 (25.1%) and G3 in 177 (71.7%). Ascites was detected in 126 patients (51.0%). After initial surgery, 89 patients (36.0%) had no macroscopic residual disease, 42 (17.0%) had macroscopic residual disease ≤1 cm and 116 (47.0%) had a larger residual tumor.
Combination chemotherapy consisted of: i) paclitaxel 175 mg/m2+carboplatin area under the curve (AUC) 5-6 every 3 weeks in 224 patients; ii) epidoxorubicin 80 mg/m2+paclitaxel 175 mg/m2+carboplatin AUC 5 every 4 weeks in 7 patients; iii) paclitaxel 175 mg/m2+carboplatin AUC 5+bevacizumab 15 mg/kg every 3 weeks in 11 patients; iv) paclitaxel 175 mg/m2+carboplatin AUC 5 every 3 weeks+vergatef 400 mg/die or placebo daily in 3 patients (enrolled in the AGO OVAR 12 trial); v) and paclitaxel 175 mg/m2+carboplatin AUC 5 every 3 weeks+weekly trebanabib 15 mg/kg or placebo in 2 patients (enrolled in the TRINOVA trial).
After the sixth cycle of chemotherapy, 181 (73.3%) patients were in clinical complete response, 57 (23.1%) were in clinical partial response and 9 (3.6%) had stable or progressive disease. Out of the 181 clinically complete responders, 73 (40.3%) underwent a second-look laparotomy/laparoscopy. Second-look findings showed a pathological complete response in 43 (58.9%), a microscopic residual disease or a converted complete response in 21 (28.8%) and a macroscopically persistent residual disease in 9 (12.3%).
One hundred and thirty-three out of the 247 patients (53.8%) received further treatment after the sixth cycle of chemotherapy, consisting of: i) 2 to 3 cycles of the same induction chemotherapy in 26 patients; ii) 6 cycles of three-weekly paclitaxel 175 mg/m2 in 39; iii) 21 cycles of weekly paclitaxel 60 mg/m2 in 54; iv) up to 22 cycles of three-weekly bevacizumab 15 mg/kg in 9; v) and other agents in 5.
CA 125 levels three to four weeks after the sixth cycle of chemotherapy (post-chemotherapy CA125) were available for 229 patients (92.7%) and ranged from 2.0 to 357.0 U/ml. The 25%, 50% and 75% quantiles of post-chemotherapy CA 125 were 7.1 U/ml, 10.8 U/ml and 18.5 U/ml, respectively. Antigen values were >35 U/ml in 19 patients (8.2%).
In the entire cohort 2-year, 3-year, 5-year and 7-year PFS were 47.8%, 37.1%, 26.6% and 21.8%, respectively. Median PFS was 22.7 months. Two-year, 3-year, 5-year and 7-year OS were 89.2%, 75.4%, 55.5% and 41.7%, respectively. Median OS was 69.4 months.
By the log-rank test, PFS was related to PS (0 versus ≥1, p=0.04), residual disease (0 cm versus >0 cm and ≤ 1 cm versus >1 cm, p=0.00002), ascites (absent versus present, p=0.00001), post-chemotherapy CA 125 taking 7.1 U/ml as a cut-off (p=0.02), post-chemotherapy CA 125 taking 10.8 U/ml as cut-off (p=0.0001) (Figure 1a), post-chemotherapy CA125 taking 18.5 U/ml as cut-off (p=0.000001) and post-chemotherapy CA125 taking 35.0 U/ml as cut-off (p=0.0001) (Table I).
By log-rank test, OS was related to PS (p=0.00002), FIGO stage (p=0.02), residual disease (p=0.001), ascites (p=0.0003), post-chemotherapy CA125 taking 10.8 U/ml as cut-off (p=0.01) (Figure 1b) (Table II).
On multivariate analysis, residual disease, ascites and post-chemotherapy CA 125 with any selected cut-off were independent prognostic variables for PFS (Table III), whereas residual disease, PS and post-chemotherapy CA 125 using 10.8 U/ml as cut-off were independent prognostic variables for OS (Table IV).
Discussion
In the last decades, several authors have investigated the clinical relevance of serum CA125 nadir or CA125 value after the end of first-line platinum-based or platinum- and paclitaxel-based chemotherapy in ovarian carcinoma patients (12, 14-19). Crawford et al. (15), who have assessed 79 complete responders with serum CA125 <30 U/ml after chemotherapy, found that the median time to biochemical progression was 2,436 days for the patients with CA125 nadir <10 U/ml, 182 days for those with CA 125 nadir between 11 and 20 U/ml and 90 days for those patients with nadir between 21 and 30 U/ml (p<0.001). The corresponding median survivals were 2,968 days, 537 days and 537 days, respectively. The Cox model revealed that CA125 nadir was the strongest independent prognostic variable for both time to biochemical progression (p<0.001) and survival (p<0.001). A retrospective investigation on 631 patients from eight French Cancer Centers showed that median PFS and median OS were 26.4 and 49.1 months for women with CA125 nadir ≤20 U/ml and, respectively, 11.9 and 18.7 months for those with higher antigen nadir (p<0.0001 for both) (12). CA125 nadir was an independent predictor for both PFS (p=0.0002) and OS (p=0.0001). Other authors reported that CA125 nadir ≤10 U/ml was an independent favorable prognostic variable for PFS (16, 18) and OS (18) of complete responders after chemotherapy. Juretzka et al. (18) analyzed 241 complete responders who underwent second-look surgery and intra-peritoneal consolidation chemotherapy. Median PFS was 2.8 years for women with serum CA125 ≤12 U/ml at the end of primary chemotherapy and 1.7 years for those with higher antigen levels, while the corresponding OS was 5.8 years and 3.7 years, respectively. CA125 level at the end of primary therapy was an independent predictor of PFS and OS. Markman et al. (19) assessed 384 patients who had CA125 ≤35 U/ml at the end of chemotherapy and who were included in two trials of maintenance treatment (20, 21). Median PFS was 24 months, 17 months and 7 months, respectively, for women who had pre-maintenance CA125 <10 U/ml, 11-20 U/ml and 21-35 U/ml. Serum CA125 before starting maintenance chemo therapy predicted the risk of subsequent relapse either as a categoric variable (p<0.001) or as a continuous variable (p<0.0001).
In the present investigation, serum CA 125 after the sixth cycle of chemotherapy was significantly related to PFS and OS of patients with advanced ovarian carcinoma treated with primary cytoreduction and platinum- and paclitaxel-based chemotherapy. Seventy-three per cent of the patients were in clinical complete response, 23.1% were in clinical partial response and 3.6% had stable or progressive disease after primary treatment. The 25%, 50% and 75% quantiles of post-chemotherapy CA125 levels were 7.1 U/ml, 10.8 U/ml and 18.5 U/ml, respectively. The Cox model showed that post-chemotherapy CA125 levels >7.1 U/ml, >10.8 U/ml, >18.5 U/ml and >35 U/ml were poor independent prognostic variables for PFS, in addition to large residual disease and presence of ascites. Moreover, post-chemotherapy CA125 >10.8 U/ml was an unfavorable independent prognostic variable for OS, in addition to large residual disease and PS>0. Patients with CA125> 10.8 U/ml had a 1.805-fold higher risk of progression and a 1.570-fold higher risk of death, respectively, when compared to those with lower antigen levels.
In conclusion, post-chemotherapy CA125 using 10.8 U/ml as cut-off was an independent prognostic variable for both PFS and OS. Usually, the achievement of serum CA 125 value <35 U/ml at the end of primary chemotherapy is considered to be a biochemical predictor of good prognosis in patients with advanced ovarian carcinoma. Conversely, only the achievement of a lower antigen value, i.e. 10.8 U/ml corresponding to the 50% quantile of post-chemotherapy levels, is associated with a significantly better long-term clinical outcome.
- Received October 3, 2014.
- Revision received October 30, 2014.
- Accepted November 4, 2014.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved