Cisplatin-epirubicin-paclitaxel weekly administration in advanced breast cancer: A phase I study of the Southern Italy Cooperative Oncology Group

Purpose: Both cisplatin and epirubicin have been shown to enhance the antit umor activity of paclitaxel in vitro. Weekly administration could result in a substantial improvement in the therapeutic index of cisplatin and paclit axel. This study was aimed at determining the MTDs of epirubicin and paclit axel given weekly with a fixed dose of cisplatin. Patients and methods: Sixty-three breast cancer patients with advanced dise ase (24 locally advanced and 39 metastatic), who had not received prior che motherapy (except adjuvant), received weekly cisplatin (CDDP) doses of 30 m g/m(2) together with escalating doses of paclitaxel (PTX) and epirubicin (E PI) for a minimum of six cycles. The dose escalation was stopped if DLT occ urred during the first six treatment cycles in > 33 of patients of a given cohort. Results: Nine different dose levels were tested, for a total of 506 weekly cycles delivered. G-CSF support on days 3-5 of each week was also given in the last four cohorts (24 patients). An overall 11 patients showed DLT in t he first six cycles. EPI and PTX doses up to 40 and 85 mg/m(2)/week, respec tively, were safely delivered without G-CSF support. However, the actually delivered mean dose intensity was only 64% in this cohort. Therefore, the d ose escalation continued with the addition of filgrastim from day 3 to day 5 each week. Doses of EPI and PTX up to 50 and 120 mg/m(2)/week were admini stered without observing DLT in the first six cycles in more than one third of the patients enrolled. No toxic deaths were observed. Only two patients had to be hospitalized because of sepsis. Grade 3-4 neutropenia, thrombocy topenia, and anemia occurred in 25, 9, and 16 patients, respectively. Alope cia was almost universal. Other nonhematologic toxicities were generally mi ld, being of grade 3-4 in only eight patients (fatigue and loss of appetite in two cases, diarrhoea in four cases, peripheral neuropathy and mucositis in one case). Fifteen complete and 37 partial responses have been register ed for an 82% (95% CI = 71-91) overall clinical response rate (ORR). Eight complete and 14 partial responses occurred in the 24 patients with locally advanced disease, for a 92% (95% CI = 73-99) ORR, as compared to seven comp lete and 23 partial responses in the 39 women with metastatic disease, 77% (95% CI = 61-89). A clear dose-response relationship was not observed, sinc e an overall response rate of at least 70% was achieved at all dose levels. However, the ORR increased to 92% in the last four cohorts which included patients who received higher doses of EPI and PTX with G-CSF support. All o f the 24 patients with locally advanced disease underwent modified radical mastectomy with axillary dissection. Three of them showed no invasive cance r on pathologic examination, and in another five patients a tumor smaller t han 1 cm was found in the surgical specimen of the breast. At a nine-month median follow-up (range 2-14), 11 patients have progressed and three have d ied. Twenty-three out of 24 patients who underwent surgery are still free f rom progression. The one-year projected progression-free survival is 77% fo r the whole population. Conclusions: The CDDP/EPI/PTX weekly administration is a well tolerated and very effective approach in advanced breast cancer patients. Full doses of all the three drugs can be delivered even in absence of G-CSF support. A ve ry impressive increment of the dose-intensity can be obtained, however, by adding filgrastim. A phase II study is under way to better define the thera peutic efficacy of this regimen in patients with advanced breast cancer.