Purpose: We described previously a prognostic model for high-risk primary breast cancer patients receiving highdose chemotherapy (HDC). Such model included nodal ratio (no. involved nodes:no. dissected nodes), tumor size, hormone receptors, and HER2. In the present study we intended to test this model prospectively in a second patient cohort. In addition, we analyzed the long-term overall outcome of our HDC trials.Experimental Design: We analyzed all 264 patients enrolled since 1990 in our prospective trials for 4 -9؉, >10؉ nodes, or inflammatory disease. Patients of the second cohort (treated since 1997) had their prognostic score estimated prospectively before receiving HDC.Results: Fourteen patients (5.3%) died from HDCrelated complications. At median follow-up of 7.1 years, relapse-free survival and overall survival of the whole group were 69.8% and 73%, respectively. Median time to relapse was 14 months (63.5% relapses within the first 2 years, 6.7% after year 5). The model was validated in the second cohort, establishing the following pretransplant risk categories: low risk (low score, HER2؊), 44% patients, 87% freedom from relapse (FFR); intermediate risk (low score, HER2؉), 29% patients, 68% FFR; and high risk (high score, any HER2), 27% patients, 49% FFR.Conclusions: Few relapses are seen after year 5 of follow-up, which indicates the need for mature results of the randomized trials before their final interpretation or metaanalysis. Our prospectively validated prognostic model, if additionally confirmed in the randomized trial populations, may provide an insight into the relative benefit of HDC in different risk patient subsets.