With currently available combination chemotherapy regimens, the outcome of the patients newly diagnosed with aggressive non-Hodgkin's lymphoma (NHL) identified as 'high' and 'high-intermediate' risk groups according to the international prognostic index (IPI) is still unsatisfactory and a more innovative therapy is urgently required to improve the survival of the patients. The purpose of this study was to compare the efficacy of rituximab given in combination with CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) and ESHAP (etoposide, methylprednisolone, high-dose Ara-C, cisplatin) vs CHOP-ESHAP and upfront high-dose therapy (HDT) and autologous stem cell transplantation (ASCT) vs standard CHOP in patients aged < or = 65 years old newly diagnosed with 'high' and 'high-intermediate' risk aggressive lymphoma enrolled onto two consecutive treatment trials at the institute. Between May 1995 - July 2002, 84 patients, aged 15 - 65 years old, with newly diagnosed aggressive NHL and an age-adjusted IPI of 2 or 3 were enrolled. The median age of the patients was 38 years (range 15 - 65). The baseline demographic features, in particular the major prognostic variables, were similar between the treatment groups. Patients treated with rituximab-CHOP-ESHAP received eight cycles of rituximab (375 mg m(-2) on day 1 of cycles 1 - 6 and days 21 and 28 of cycle 7) plus CHOP (day 3 of cycles 1, 3 and 5) and ESHAP (day 3 of cycles 2, 4 and 6 and day 1 of cycle 7) at 21-day intervals. Patients enrolled onto the CHOP-ESHAP-HDT arm (n = 23) were treated with three courses of CHOP and then switched to two or four cycles of ESHAP followed by HDT. Patients treated with CHOP alone (n = 25) were treated with the standard eight cycles of CHOP. The rate of complete remission was significantly improved with rituximab-CHOP-ESHAP compared with either CHOP-ESHAP-HDT or CHOP alone (67% compared with 44% and 36%, respectively; p = 0.043). With a median follow-up time of 53 months, the 5-year overall survival (OS) was improved by the addition of rituximab-61% with rituximab-CHOP-ESHAP, compared with 43% for CHOP-ESHAP-HDT and 24% for CHOP alone (p = 0.088). Significant increases in failure-free survival (FFS) and disease-free survival (DFS) (61% and 96%), compared with CHOP-ESHAP-HDT (34% and 90%) and CHOP (16% and 44%; p = 0.002 and p < 0.001, respectively) were observed. Compared to CHOP, rituximab-CHOP-ESHAP yielded significantly superior OS (p = 0.014), FFS (p < 0.001) and DFS (p < 0.001). The survivals, however, were not significantly different from patients treated with CHOP-ESHAP-HDT. It is concluded that rituximab-ESHAP-CHOP is superior over standard CHOP and fares comparably to upfront HDT/ASCT in previously untreated patients with aggressive lymphoma. A prospective randomized controlled trial is warranted to confirm these results.