Abstract
OBJECTIVE To investigate the role of maintenance chemotherapy on stage III ovarian carcinoma.
METHODS A retrospective analysis was conducted of 47 stage III ovarian carcinoma patients with clinical complete remission after first -line chemotherapy. Among these patients, 21 cases were treated with maintenance chemotherapy, while the other 26 cases were free of treatment until progression. The 2 groups were compared with respect to progression-free survival (PFS) and overall survival(OS).
RESULTS The median PFS and OS were not significantly different between the 2 groups. For those patients, in a subgroup of suboptimal surgery (residual disease >2 cm), the median PFS was 110 weeks and 56 weeks and the median OS was 223 weeks and 157 weeks for the maintenance and non-treated respectively. Both PFS and OS values favoured the maintenance group, P=0.004 and P=0.015 respectively. In a subgroup of optimal surgery (residual disease ≤2 cm), the differences were not significant.
CONCLUSION Patients with stage III ovarian carcinoma with clinical complete remission may benefit from maintenance chemotherapy, if the residual disease is >2 cm. To those with a residual disease ≤2 cm, the maintenance chemotherapy maybe of no value. So "individualized maintenance chemotherapy" should be conducted in the clinical setting.
keywords
The prognosis of advanced ovarian cancer is the worst among gynecological neoplasms. A high recurrence rate is one of the important reasons. Even though the patients experience clinical complete remission (cCR) after surgery and first-line chemotherapy, 75% are deemed to recur, and the progression-free-survival (PFS) is usually less than 2 years.[1,2]
Consequently, numerous clinical strategies are being evaluated in an effort to prevent or delay recurrences in patients who achieve a cCR. Maintenance chemotherapy, e.g. continued chemotherapy after cCR maybe beneficial in this setting, but so far it is controversial with some researchers for and some against. Here, using retrospective analysis we investigated the role of maintenance chemotherapy to prolong survival.
MATERIALS AND METHODS
Patient data
Between 1991 and 2001, 47 patients with stage IIIc epithelial ovarian cancer in our institute were included. All patients achived cCR after no less than 4 cycles of adjuvant intravenous platinum-based combination chemotherapy. cCR here was defined as the patients having a normal physical examination, normal radiographic examination and a normal serum CA125 level.
Among the 47 patients, 21 patients received maintenance chemotherapy after cCR, grouped as the MT group (maintenance chemotherapy group) and 26 patients received no treatment until progression, grouped as the FT group (free of treatment group).The patients in the MT group received 3∼8 cycles of platinumbased combination chemotherapy after cCR. The interval between cycles was 4∼12 weeks.
Progression was defined as(a) for the measurable lesions, a 20% increase in the sum of the largest line or appearance of any new lesion (RECIST criteria),[3](b) for assessment of the lesions, a CA125 rise up to above 100 u/ml.[4] PFS was accepted as the time elapsed from surgery to disease progression or last follow-up examination. Overall survival(OS) was accepted as the period from surgery to death or last follow-up examination.
Patient characteristics
For the MT group, the median age was 55. Eleven cases (53%)were optimally resected (defined as residual disease < 2cm), 16 cases(76%) were a serous type and 8 cases (38% ) were poorly-differentiated. Among these, 6 cases (29%) were treated with platinum combined with paclitaxel as first-line chemotherapy and the other 15 cases received platinum combined with other drugs. For the FT group, the median age was 59. Fourteen cases(54%)were optimally resected, 21 cases (81%) were a serous type and 12 cases (46%) were poorly-differentiated. In this group 9 cases(35%) were treated with platinum combined with paclitaxel as first-line chemotherapy. There were no significant differences between the 2 groups based on the above factors (χ2 test). As for the side-effects, no one stoped maintenance chemnotherapy because of non-endurable side-effects, although other side-effects were not statistically analyzed here.
Statistical analyses
Statistical analyses were performed by the statistical programming package SPSS for Window Release 12.0 (SPSS Inc. Chicago, IL). Categorical variables were analyzed by the χ2 test with the Fischer’s exact test. PFS and OS were calculated and curved by the Kalpan- Meier method and compared by the log-rank test. Multiple factor analysis was conducted by the Cox stepwise logistic regression analysis. P-values <0. 05 were regarded as statistically significant.
RESULTS
PFS analysis
The last follow-up examination was on June 1st, 2004. After a median follow-up period of 234 weeks and 162 weeks respectively for the MT group and FT groups, 10 (21%) patients were a live with no evidence of disease, 15 patients were alive with disease and 22 patients died due to disease progression.
The PFS was not significantly (P=0.526) different between the MT and FT groups (Fig.1).The median PFS was 156 weeks (range 73∼603 weeks) for the MT group and 121 weeks for the FT group.
Furthermore, the 47 patients were divided into 2 subgroups based on the residual disease. In the subgroup with residual disease ≤ 2cm,the PFS was not significantly different (P=0.600) between the 2 groups (Fig.2), the median PFS was 197 weeks for the MT group and 167 weeks for the FT group. However, in the subgroup with residual disease >2cm, the PFS was significantly (P=0.004) better for the MT group than the FT group (Fig.3). The median PFS was 110 weeks for the MT group and 56 weeks for the FT group.
PFS comparison between the MT and FT groups.
PFS comparison between the MT and FT groups in the subgroup with residual disease < 2 cm.
PFS comparison between the MT and FT groupa in the subgroup with residual disease >2 cm.
For other prognostic factors, such as residual disease, age, pathological differentiation and first-line chemotherapy protocols, the only significant factor was residual disease by means of the log-rank test. The median PFS was 448 weeks for patients with residual disease ≤ 2 cm and 85 weeks for the patients with residual disease >2 cm(P<0.001). Taking all the above factors into account combined with the treatment (MT or FT) altogether in the Cox regression model, the residual disease was the only prognostic factor of PFS (P<0.001).
OS analysis
The OS was not significantly (P=0.272) different between the MT group and the FT group (Fig.4).The median OS was 312 weeks (range 157∼808 weeks) for the MT group and 204 weeks for the FT group.
OS comparison between the MT and FT groups.
Furthermore, in the subgroup of residual disease ≤ 2 cm, the OS was not significantly different (P=0.629) between the 2 groups (Fig.5), as the median OS was 278 weeks for the MT group and 261 weeks for the FT group. However, in the subgroup with residual disease >2 cm, the OS was significantly (P=0.015) better for the MT group than for the FT group (Fig.6), as the median OS was 223 weeks for the MT group and 157 weeks for the FT group.
OS comparison between the MT and FT groups in the subgroup with residual disease ≤ 2 cm.
OS comparison between the MT and FT groups in the subgroup with residual disease >2 cm.
For other prognostic factors, residual disease was the only significant factor by means of the log-rank test. The median OS was 600 weeks for patients with residual disease ≤ 2 cm and 186 weeks for the patients with residual disease >2 cm (P=0.002). Taking all above factors combined with treatment considered altogether in the Cox regression model, the residual disease was the only prognostic factor of OS (P=0.002).
DISCUSSION
Maintenance chemotherapy performed after clinical complete remission is aimed at prolonging remission and reducing or delaying progression. Although this benefit has not been clearly established yet, a recent study by the Gynecologic Oncology Group (GOG)pro- vides evidence favoring this approach. In this study, investigators looked at 2 different schedules of maintenance chemotherapy with intravenous paclitaxel in patients with clinical complete remission after primary chemotherapy. This study was closed early with a significant improvement in PFS favoring the prolonged regimen compared to the shorter program[5] However, so far this study is the only positive report on maintenance chemotherapy. Other large-scale prospective randomized controlled clinical trials were all negative, whenever the drug was cisplatin or topotecan[6-[8] Also, it should be pointed out that the GOG study had some faulty features, one of which was the lack of a placen- bo-controlled group, which enormously weakened the study’s strength. These results may indicate that maintenance chemotherapy would not improve the survival of all cCR group patients. In the present study, we also supported this point of view by comparison of the entire MT group and FT group with negative results. But, was the maintenance chemotherapy of no value to everyone? To some individuals, especially the high-risk individuals, did maintenance chemotherapy come up into the same conclusion? In the present study, we found that those patients with residual disease more than 2 cm benefited from maintenance chemotherapy, whereas those with residual disease less than 2 cm did not benefit. So here, we propose that the clinical application of maintenance chemotherapy should be individualized, with preference given to those high-risk patients.
However, this outcome should be evaluated with caution, as large-scale randomized trials are required to determine the value of "individualized maintenance chemotherapy".
- Received August 3, 2005.
- Accepted September 20, 2005.
- Copyright © 2005 by Tianjin Medical University Cancer Institute & Hospital and Springer













