Anthracyclines in non-small-cell lung cancer: Do they have a therapeutic role?

Background: Owing to its low level of activity together with its potential cardiotoxicity, doxorubicin (DXR) has been considered as having a marginal role in the treatment of NSCLC. Its analogue, epirubicin (EPI), has also sh own a poor antitumor activity in the treatment of NSCLC when used at 'stand ard' doses (= 90 mg/m(2)). On the contrary, high-dose epirubicin (HD-EPI) ( > 90 mg/m(2)) has demonstrated antitumor activity as a single agent in the treatment of advanced NSCLC in six small phase II studies (mean 25%, range 17%-36%). Results: A series of consecutive studies on the activity of HD-EPI alone or in combination regimens were carried out at the Division of Medical Oncolo gy of S. Orsola-M. Malpighi Hospital. After activity was confirmed in advan ced disease with doses between 120 and 165 mg/m(2) (PR in 6 of 24 = 25%), a phase II study was carried out on the combination of HD-EPI 120 mg/m(2) cisplatinum (CP) 60 mg/m(2) in stage IIIB-IV NSCLC. PR was achieved in 54% of 35 patients with a median survival of nine months. A subsequent multicen ter phase III trial compared HD-EPI and vinorelbine (VNR), both combined wi th CP. Two hundred twenty-eight patients with locally advanced or metastati c NSCLC were randomized to receive either EPI 120 mg/m(2) plus CP 60 mg/m(2 ) on day 1 or VNR 25 mg/m(2) on day 1 and 8 plus CP 60 mg/m(2) on day 1. Bo th treatments were recycled every 21 days. Eligible patients were 212 and 2 10 patients evaluable for objective response (100 on HD-EPI and 110 on VNR) , respectively. The CR + PR rate was 32% vs. 26% (P = NS) for a median dura tion of nine and eight months, respectively. Median survival was 10 and 9.5 months, respectively. Grade III-IV leucopenia occurred in 38% and 21% on H D-EPI and VNR, respectively (P = 0.01), thrombocytopenia in 6% and 0% (P = 0.02), anemia in 8% and 7% (NS). Non-hematological toxicity was moderate an d the only difference between the treatments was alopecia (88% vs. 33% on H D-EPI and VNR, respectively). Supraventricular arrhythmia occurred in three patients on HD-EPI; a > 15% LVEF decrease by MUGA scan was observed in 22. 5% and 14% patients on HD-EPI and VNR, respectively (NS). No congestive hea rt failure was observed. Conclusions: EPI can be safely administered at a dose of 120-135 mg/m(2) in non-pretreated patients showing a significant antitumor activity in NSCLC. If the cumulative dose of 800-900 mg/m(2) is not exceeded, clinical manife stations of cardiotoxicity are very rare. However, grade 3-4 myelotoxicity and alopecia are very common and can limit the use of this drug in the pall iative treatment of this disease. Interesting results are observed in an on going pilot study that employed HD-EPI + CP + VNR + G-CSF in the induction therapy of locally advanced NSCLC.