The best magazine
CON Collaborators. International Collaborative Ovarian Neoplasm Study.
Lancet 352(9140):1571-1576, 1998
A series of meta-analyses of randomized, controlled trials raised the question of whether the three-drug combination of cyclophosphamide, doxorubicin, and cisplatin (CAP) was more or less effective than optimal-dose, single-agent carboplatin for women with advanced ovarian cancer. We carried out an international, multicenter, randomized trial to compare CAP with single-agent carboplatin in women with ovarian cancer requiring chemotherapy. A total of 1526 patients were entered from 132 centers in nine countries. Analyses were by intention to treat; 728 patients have died (368/766 allocated CAP vs 360/760 allocated carboplatin), and the survival curves show no evidence of a difference between CAP and carboplatin (hazard ratio 1.00 [95% CI 0.86-1.16]; P=.98). The results indicate a median survival of 33 months and a 2-year survival of 60% for both groups. We found no evidence that CAP or carboplatin was more or less effective in different subgroups defined by age, stage, residual disease, differentiation, histology, and coordinating center. CAP was substantially more toxic than carboplatin, causing more alopecia, leucopenia, and nausea. More thrombocytopenia occurred with carboplatin. Therefore, we conclude that single-agent carboplatin, with the dose calculated by the area-under-the-curve method, is a safe, effective, and appropriate standard of treatment for women with advanced ovarian cancer.
Shapiro JD, Rothenberg ML, Sarosy GA, et al
Cancer 83(9):1980-1988, 1998
The authors combined cisplatin and carboplatin together with cyclophosphamide to maximize platinum dose intensity in patients with advanced epithelial ovarian cancer (AOC). The authors treated 26 consecutive, newly diagnosed patients with International Federation of Gynecology and Obstetrics (FIGO) Stage III/IV AOC with carboplatin 600mg/m on day 1; cyclophosphamide 250mg/m on day 1; and cisplatin 100mg/m on day 8 every 4 weeks with or without pretreatment with amifostine (range, 740-1140mg/m). Platinum dose intensity was estimated using a 4:1 conversion for equipotent doses of cisplatin and carboplatin for expression as cisplatin dose equivalents (CDE). The mean administered CDE was 49.4mg/m/week, which was 79% of the planned dose. Hematologic toxicity was severe, with FIGO Grade 3-4 anemia in 81% of patients, Grade 3-4 neutropenia in 92% of patients, and Grade 4 thrombocytopenia in 96% of patients. Eleven patients (42%) were admitted to the hospital for febrile neutropenia, and one toxic death occurred. Sensory neuropathy >=Grade 2 occurred in 10 patients (38%), ototoxicity >=Grade 2 occurred in 18 patients (69%), and 6 patients (23%) required long-term hearing aids. Elevations in serum creatinine >=Grade 2 occurred in 7 patients (27%) and >=Grade 2 hypomagnesemia was noted in 23 patients (88%). Other Grade 3 toxicities were nausea (42%), emesis (38%), fatigue (15%), mucositis (4%), and respiratory toxicities (4%). Twenty-two of 26 patients (85%) had a clinical response (19 had a complete response [CR] and 3 had a partial response). Pathologic CR was demonstrated in 10 of 26 patients (38%) and residual microscopic disease in 4 of 26 patients (15%) for a total pathologic response rate of 53%. The median progression-free survival was 13.5 months and the median overall survival was 37.2 months at a median potential follow-up of 79.3 months. Three of 26 patients remained free of disease at 66, 71, and 103 months, respectively. The investigators concluded that, although dose-intensive platinum treatment combined with cyclophosphamide in patients with AOC is active, it also is associated with substantial toxicity.
Aabo K, Adams M, Adnitt P, et al
Br J Cancer 78(11):1479-1487, 1998
The purpose of this systematic study was to provide an up-to-date and reliable quantitative summary of the relative benefits of various types of chemotherapy (nonplatinum vs platinum, single-agent vs combination, and carboplatin vs cisplatin) in the treatment of advanced ovarian cancer. Another aim was to investigate whether well-defined patient subgroups benefit more or less from cisplatin- or carboplatin-based therapy. Meta-analyses were based on updated individual patient data from all available randomized, controlled trials (published and unpublished), including 37 trials, 5667 patients, and 4664 deaths. The results suggest that: (1) platinum-based chemotherapy is better than nonplatinum therapy, (2) there is a trend in favor of platinum combinations over single-agent platinum, and (3) cisplatin and carboplatin are equally effective. There is no convincing evidence that cisplatin is more or less effective than carboplatin in any particular subgroup of patients.
Outdated Studies, Clinically Irrelevant Results
Editorial Comment by Jubilee B. Robinson, MD and William P. McGuire, III, MD
The three studies abstracted here investigated the role of chemotherapy in the treatment of advanced epithelial ovarian cancer. The ICON2 trial was a large, randomized, prospective trial that compared patient outcomes after treatment with cyclophosphamide, doxorubicin (Adriamycin), and cisplatin (CAP) versus single-agent carboplatin. This study was based on previous meta-analyses that had suggested that platinum compounds in combination with other agents are more effective than single-agent platinum compounds, and that cisplatin and carboplatin are equally effective. This study sought to determine whether carboplatin can be used at a sufficient dose as a single agent to be equally effective as and less toxic than CAP. The authors demonstrated that carboplatin is not significantly different from CAP combination chemotherapy, and that carboplatin is less toxic overall.
These findings appear to be valid based on the power of the study (1526 patients), but the purpose of the study is outdated, and the authors' conclusions are erroneous. Since GOG 111 demonstrated a clear advantage in progression-free interval and overall survival with the use of paclitaxel and cisplatin over cyclophosphamide and cisplatin, the pursuit of outcome data excluding the use of paclitaxel is clinically irrelevant. In fact, the ICON2 trial was stopped prematurely in July 1996 because of interest in paclitaxel. The authors suggest that single-agent carboplatin is "an appropriate standard of treatment" based on the results of this study, but without comparison of one arm or the other against the new standard of paclitaxel and a platinum compound, such a conclusion is unfounded. Such conclusions must await maturity of the data from ICON3, which compares each regimen from ICON2 to the paclitaxel/carboplatin doublet.
Additionally, the investigators of ICON2 discussed long-term survival but did not disclose the percentage of patients subsequently treated with paclitaxel or the timing of such treatment. Late application of salvage treatment had little effect on survival in one multinational study, while early application of salvage treatment blurred survival differences.
The second study, by Aabo and associates, also investigated the role of platinum-based therapy in patients with advanced epithelial ovarian cancer. This meta-analysis was inherently weaker than the ICON2 trial because of its design, comparing multiple different and unrelated regimens. As expected, the results demonstrated the superiority of platinum-containing regimens over nonplatinum regimens. The trial compared single-agent platinum to platinum combinations. In contrast to ICON2, the investigators stated that the results "favor the use of combination chemotherapy." However, they then stated that the results are "inconclusive" with a P-value of .21. They suggest that cisplatin-based regimens favor combination therapy but that carboplatin-based regimens favor single-agent therapy, despite no difference in overall outcome between cisplatin- and carboplatin-based regimens. These are counter-intuitive statements, and the association between cisplatin/carboplatin and single-agent/combination chemotherapy is weak, with borderline if any statistical significance. Although these findings differ from those of the ICON2 trial, the lack of clear statistical significance in this less powerful meta-analysis does not provide convincing evidence for a true advantage to combination chemotherapy. It is likely that the ICON2 trial is more accurate in this regard. Aabo's trial also suffers from the same clinical irrelevance as the ICON2 trial, in that the use of paclitaxel was not evaluated.
The third study, by Shapiro and colleagues, reiterates the well-supported findings that no benefit exists to dose intensity of platinum therapy. Indeed, no survival advantage is seen, and toxicity is increased.
The findings of these three articles support the continued role of platinum-based therapy but do not impact the treatment of ovarian cancer at present.
McGuire WP, Hoskins WJ, Brady MF, et al: Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 334(t):I-6, 1996.
Stuart G, Bertelsen K, Mangioni C, et al: Updated analysis shows a highly significant improved over all survival (OS) for cisplatin-paclitaxel as first line treatment of advanced ovarian cancer: Mature results of the EORTC-GCCG, NOCOVA, NCIC CTG and Scottish intergroup trial. ASCO 17:At394, 1998.
Muggia FM, Braly PS, Brady MF, et al: Phase III trial of cisplatin (P) or paclitaxel M, versus their combination in suboptimal stage III and stage IV epithelial ovarian cancer (EOC): Gynecologic Oncology Group study #132. ASCO 16:Al257, 1997.
Source: ...