High-dose chemotherapy as adjuvant treatment for high-risk primary breast cancer patients.

In a study recently published in Annals of Oncology, Coombes et al. [1] reported the results of a prospective randomized trial comparing a short course of anthracycline-containing chemotherapy followed by high-dose chemotherapy with stemcell support (HDCT) versus a conventional-dose anthracyclinecontaining chemotherapy (CDCT) as adjuvant treatment in breast cancer patients with four or more positive axillary lymph nodes. In their experience the authors did not observe any significant differences between HDCT and CDCT on diseasefree and overall survival (OS) and they concluded that ‘patients should continue to receive conventional chemotherapy as adjuvant therapy for breast cancer’. However, this study does not seem to provide any clear information for the long-lasting controversy [2–5] on the role of HDCT in high-risk primary breast cancer (HRBC). Indeed, the trial was designed to investigate the benefits deriving from replacing three cycles of conventional anthracycline-containing chemotherapy with a single course of HDCT, and not to compare head-to-head HDCT versus CDCT in HRBC. As discussed by the authors, this trial has some significant limits that should not be undervalued when we look at their conclusions. First of all, the treatment schedule is unsatisfactory in both arms: in the high-dose arm a total dose of epirubicin of 150 mg/m was delivered, which is to be considered totally inadequate for HRBC, whereas in the CDCT arm the total dose of epirubicin was 300 mg/m, lower than the dose used in most other studies. Secondly, 55% of patients in the study had between four and nine positive axillary lymph nodes, but a beneficial effect of HDCT on relapse-free survival has been suggested for patients with 10 or more positive lymph nodes [3]. Thirdly, owing to the low recruitment the trial was closed before the planned number of 300 patients was achieved, and of the 281 enrolled patients, only 80% in the HDCT arm and 76% in the CDCT arm completed the planned therapy because of treatment refusal (10 patients in the HDCT arm), toxicity or other reasons. Of note, treatment-related mortality in the CDCT arm was surprising high (1.5%) considering the dose of anthracycline and in comparison with the HDCT arm (2%). In our experience with HDCT in HRBC, all patients under 65 years of age affected by HRBC with 10 or more involved lymph nodes are evaluated for an adjuvant program including HDCT. From July 1998 to March 2005, 36 patients (Table 1) were treated with four cycles of anthracycline plus taxane followed by a mobilization phase with high-dose cyclophosphamide (4.5 or 7 g/m) and granulocyte colony-stimulating factor. HDCT consisted of melphalan 160 mg/m and thio-tepa 600 mg/m. Patients with hormonereceptor positive tumors were prescribed tamoxifen 20 mg daily for 5 years with or without gonadotrophin-realeasing hormone analogs, according to menopausal status. All patients were evaluable for toxicity and follow-up. After a median follow-up time of 53 months (range 6–83), the actuarial 5-year event-free survival (EFS) and OS rates were 66% and 81%, with a median EFS and OS of 40 and 49 months (range 6–83), respectively. The tolerability of HDCT was good, with no early or late toxic death and only one case of serious treatment-related morbility (brain bleeding). Regarding HDCT, of major concern is why the almost general improvement in EFS after HDCT [3–5] does not translate in an increased OS. Perhaps, the treatment-related mortality could partially explain such difference in some studies [2, 4], but the main limitation is that most trials so far have been underpowered to reveal small differences in the population, therefore being unable to identify some HRBC patient subsets that could particularly benefit from HDCT. Even if HDCT cannot be considered today as standard treatment for HRBC it remains a very interesting field of clinical investigation, and a meta-analysis of all relevant randomized