Patients with intermediate grade non-Hodgkin's lymphoma (NHL) who relapse or fail to achieve a complete remission after anthracycline-containing induction regimens have a poor outcome with conventional-dose salvage treatment. This outcome may be improved with intensive therapy and autologous transplantation (ABMT) but even in patients with proven chemotherapy-sensitive disease, relapse rates of up to 60% are observed. Reliable and powerful prognostic indicators are needed to identify appropriate patients for this expensive procedure and those subjects to whom alternative or additional treatment should be offered. We were interested in testing the hypothesis that tumour burden, and hence remission status immediately prior to transplant, is an important prognostic indicator of survival. We aggressively treated patients with conventional-dose salvage chemotherapy to maximum tumour response, and tested, by multivariate regression analysis, predictors of outcome post-transplant. We studied 81 consecutive patients with intermediate grade and immunoblastic NHL who achieved either a partial (PR) or complete remission (CR) following repetitive cycles of conventional-dose salvage therapy. Intensive therapy consisted of etoposide (60 mg/kg) and intravenous melphalan (160-180 mg/m2) with or without total body irradiation (TBI) followed by infusion of autologous unpurged bone marrow and/or blood cells. The predicted 4-year survival and progression-free survival (PFS) with a median follow-up of 37 months was 58% and 48% (95% confidence interval (CI) 37-55%), respectively. The only factor predictive of outcome was remission status at transplant (P=0.0001). The PFS at 4 years for the CR group was 61% (95% CI 53-75%). In contrast, only 25% (95% CI 11-40%) of patients undergoing autotransplant in PR were progression free at 4 years. We conclude that remission status at transplant after maximum tumour reduction is a powerful prognostic indicator.

A total of 10 -40% of patients with Hodgkin's disease relapse following initial curative therapy. Intensive follow-up is resource intensive and may identify false relapses. We performed a retrospective review of all patients with Hodgkin's disease treated at our centre between 1990 and 1999 to evaluate the utility of the components of follow-up. A total of 107 patients met the inclusion and exclusion criteria. The median age was 33 years and the median duration of follow-up 38 months. The total number of follow-up visits was 1209 and total number of CT scans 283. There were 109 suspected relapses of which 22 proved to be true relapses. Of the latter, 14 were identified clinically, six radiologically and two via lab testing. The routine CT scan detected only two relapses (9%), yet accounted for 29% of the total follow-up costs. Based on data from our centre, the cost per true relapse was $6000 US, 49% incurred by radiological tests. The majority of the cost of follow-up was incurred by routine follow-up (84%) as opposed to the investigation of suspected relapses (16%). We conclude that most true relapses are clinically symptomatic and that the routine CT is an expensive and inefficient mode of routine follow-up.