Eight randomized trials of HDC in MBC have been reported to date. Of these, six have compared high-dose therapy to conventional-dose chemotherapy, and two have evaluated the timing of high-dose therapy: early versus late.
Stadtmauer et al.17 compared a single cycle of highdose therapy to maintenance conventional-dose chemotherapy in 184 patients responding to a long course of conventional-dose chemotherapy. At a median follow-up of 67 months, an intent-to-treat analysis showed no differences between the two arms in EFS (4% vs 3%) or OS (14% vs 13%). Surprisingly, the partial remission (PR) to CR conversion rate, while small, was higher in the maintenance conventional-dose chemotherapy arm than in the HDC arm (9% vs 6%). This trial has been widely criticized because of its 45% dropout rate.
Using a similar approach, a group of Canadian investigators randomized 224 patients with chemore-sponsive MBC to receive either additional conventional-dose chemotherapy or a single cycle of highdose therapy with conventional-dose chemotherapy. In its first intent-to-treat analysis at a median follow-up of 19 months, significant differences in favor of HDC were observed in EFS (38% vs 24%, P = 0.01), but not OS (median 2 vs 2.3 years, P = 0.9). This trial suffered high attrition (21%) and transplant-related mortality (TRM) (7.7%) rates in the transplant arm.18
In the French National PEGASE-03 study, Biron et al.19 randomized 180 responding patients to one cycle of high-dose therapy with conventional-dose chemotherapy or observation. High-dose therapy was well tolerated, with a TRM of 1% and an increase in the CR rate from 11 to 24% (P = 0.0002). Interestingly, in this study with a median follow-up of 48 months, there was a significant improvement in EFS in favor of high-dose therapy (27% vs 10%, P = 0.0005).
Crown et al.20 looked at initial therapy with tandem high-dose therapy compared with conventional-dose doxorubicin/docetaxel followed by maintenance chemotherapy in 110 patients with MBC. The response rate on the high-dose arm was significantly better, with overall response (71% vs 29%) and CR rates (44% vs 6%) better than the control arm. Using EFS as the primary endpoint, the high-dose arm was superior (16% vs 9%, P = 0.01) at a median follow-up of 42 months.
Following a similar design, Schmid et al.21 compared two sequential cycles of high-dose therapy to six to nine cycles of another modern conventional-dose chemotherapy regimen of doxorubicin and paclitaxel in 92 untreated patients. The CR rate and EFS were significantly superior in the transplant arm, without significant differences in OS at a follow-up of 14 months.
In the other French trial, Lotz et al.22 randomized 61 responding MBC patients to additional conventional-dose chemotherapy or one HDC cycle. The apparently large differences seen in favor of the transplant arm in EFS (median 35 months vs 20 months) and OS (median 43 months vs 20 months, with 5-year OS rates
30% vs 18%) did not reach statistical significance in this study.
Finally, investigators at Duke University conducted two small trials with a crossover design, comparing early versus late use of HDC in MBC patients in CR,23 and with bone-only disease,24 respectively. In those studies, patients were randomized to immediate HDC with one cycle of cyclophosphamide/cisplatin/BCNU or observation, with the same HDC offered to patients in the control arm upon relapse or progression. In both trials the immediate transplant arm had significantly superior EFS (25% vs 10%, and 17% vs 9%, respectively) with no significant benefit in OS, compared to late transplant (33% vs 38%, and 28% vs 22%, respectively).
In summary, eight randomized trials in MBC, enrolling 1020 patients, have been reported. None are particularly large with half enrolling 100 or fewer patients. EFS differences in favor of high-dose therapy were seen in seven of those eight trials,18-24 with the only exception being the Philadelphia study.17 Longer follow-up is needed to see if the EFS advantage translates into an OS benefit.
Was this article helpful?