Epirubicin/paclitaxel/carboplatin (TEC) vs paclitaxel/carboplatin (TC) in first‐line treatment of ovarian cancer FIGO stages IIB–IV. Results of a randomized AGO‐GINECO GCIG Intergroup phase III trial

. A randomized phase III trial was conducted in patients with advanced ovarian cancer FIGO IIB–IV in order to determine whether epirubicin (E) in addition to paclitaxel (T) and carboplatin (C) will increase survival in comparison to TC alone. The rationale of this study was based not only on the evidence of activity of free and liposomal anthracylins in patients who have been pretreated with the platinum–paclitaxel regimen but also on the evidence of meta-analyses performed before the taxane era of a 7% improvement in survival when doxorubicin was added as the third agent in first-line treatment (Ovarian Cancer Meta-analysis Project. J Clin Oncol 1991;9:1668–74). In addition, there were some reports suggesting a synergistic activity between paclitaxel and anthracylins (Ray-Coquard I et al. Gynecol Oncol 2003;88:351–7) in patients with ovarian cancer in relapse. Between November 1997 and February 2000, 1282 patients were randomized to receive six cycles of either T 175 mg/m2 3 h iv + C area under curve 5 (according to the Calvert formula) + E 60 mg/m2 iv (TEC) or TC at same doses, both in 3-week intervals. Patients were stratified per centre and into one of two strata according to FIGO stage and residual tumor size: stratum I includes patients with FIGO IIB–IIIC and residual tumor ≤1 cm and stratum II patients with FIGO IIB–IIIC and residual tumor >1 cm or FIGO IV. The primary end point was overall survival (OS). Secondary end points included toxicity, response to treatment, quality of life, and progression-free survival (PFS). Median follow-up at the time of analysis was 51.9 months. The whole population is available for PFS and OS analysis. A total of 1264 patients received at least one cycle of treatment and are evaluable for toxicity. Patient characteristics are well balanced between the two arms with respect to median age, performance status, FIGO stage, histology, and stratification, with less than one third of the patients suboptimally debulked. The three-drug combination induced a markedly higher myelotoxicity resulting in increased demand on supportive care. Grade 3/4 febrile neutropenia occurred in 5.5% patients in TEC and in 1.3% patients in TC (P < 0.0001) and infection occurred in 8.7% in TEC and 3.1% in TC (P < 0.0001). Addition of E induced more grade 3/4 nausea (6.9% vs 3.3%, P= 0.004), emesis (6.4% vs 2.8%, P= 0.002), and mucositis (1.8% vs 0.2%, P= 0.004) but did not increase cardiac or other nonhematologic toxicity. The increased toxicity of TEC compared to TC was associated with a higher rate of course delay (12.7% vs 8.9%) and dose reduction (4.7% vs 2.0); mean dose effectively received by patients of each of the drugs included in the TEC triplet, however, was close to that planned (epirubicin, 58.8 mg/m2; paclitaxel, 171.2 mg/m2; carboplatin, area under curve 4.9). In addition, the percentage of patients receiving at least the six planned cycles was similar in the TEC and TC arms (86.6% vs 88.8%). Of 353 patients with measurable disease, response data were available from 295 patients, 137 in the TEC arm and 158 in the TC arm. The TEC regimen was associated with 73.7% clinically complete and partial responses, the TC regimen with 70.3% (ns). At the time of this analysis, 968 progressive diseases have been diagnosed and 732 patients have died. Median PFS for patients with/without E was 18.4 months (95% CI, 16.2–20.2 months) vs 17.9 months (95% CI, 16.3–19.7 months), corresponding to a hazard ratio of 0.94 (95% CI, 0.83–1.07). Patients in stratum I showed a slightly but still not significantly better PFS after TEC as compared to TC, whereas patients in stratum II showed virtually no difference. Median OS for all patients was 42.7 months (95% CI: 39.6–47.0) with a median OS of 41 months (95% CI, 38.2–46.1 months) for patients treated with TC and 45.8 months (95% CI, 39.9–49.6 months) for those treated with TEC, corresponding to a hazard ratio of 0.93 (95% CI, 0.83–1.07). The addition of E to TC did not result in significantly superior PFS and OS but induced more toxicity and costs.