A systematic review of chemotherapy trials in several tumour types was performed by The Swedish Council of Technology Assessment in Health Care (SBU). The procedures for the evaluation of the scientific literature are described separately (Acta Oncol 2001; 40: 155-65). This synthesis of the literature on chemotherapy for B-cell chronic lymphocytic leukaemia (B-CLL) is based on data from 20 randomised controlled trials and one meta-analysis. Moreover, data from 19 prospective studies, one retrospective study and four other articles were used. Totally 44 scientific articles are included, involving 11,289 patients. The conclusions reached can be summarized into the following points: Primary treatment of patients with symptomatic B-CLL is recommended to be an oral alkylating agent such as chlorambucil. This drug induces tumour remission and symptomatic relief in a majority of patients with progressive disease. Response may be long-lasting, but cure is not obtained. Optimum dose and schedule of administration of chlorambucil or other alkylating agents have not been defined. It is recommended to defer initial therapy until required by disease progression. Large randomised trials have demonstrated that early treatment with chlorambucil in a continuous or an intermittent schedule does not prolong survival in B-CLL patients with low tumour burden (Binet stage A). The addition of corticosteroids to alkylator regimens has not been proven to give any benefit. Combination chemotherapy as primary treatment has not shown any advantage compared with single drugs. Early inclusion of anthracyclines to the therapy does not convincingly add to the activity of alkylating agents. The purine analogues fludarabine and 2-chlorodeoxyadenosine are active in B-CLL. However, like other drugs, they do not appear to be curative. In randomised multicentre trials a benefit from fludarabine as primary therapy compared with polychemotherapy (CHOP or CAP) has been observed in terms of tolerance and treatment response but not yet in survival. No randomised studies have been performed to show whether one of the purine analogues should be preferred. At relapse after single drug treatment, retreatment with the same drug often induces new remissions. However, the proportion of patients responding declines each time chlorambucil or any other single agent is readministered. At progression on single alkylating agents, the purine analogues or various combinations, mostly CHOP, frequently induce tumour remissions. For patients with advanced B-CLL failing to respond to fludarabine or CHOP, the prognosis is poor. None of the salvage regimens reported has produced durable remissions. High-dose chemo-radiotherapy with stem cell transplantation has been evaluated for young patients with B-CLL. A long survival has been shown in some patients following allogeneic and autologous transplantation. However, the risk of transplantation-related mortality is still high with allo-transplants and relapse is common after auto-transplantation. A benefit of purging autologous stem cells has been proposed but evidence is lacking. Thus, transplantation remains experimental; more patients and a longer follow-up are needed to assess if cure can be achieved. In the future an individual risk-adapted therapy will be required. The clinical heterogeneity of the disease has pointed to the necessity of new predictors for prognosis evaluated in prospective trials.
