A network meta-analysis of therapies for previously untreated chronic lymphocytic leukemia

Background: Several therapy options are available for symptomatic, treatment-naive chronic lymphocytic leukemia (CLL). Many of these therapies have been compared against chlorambucil, but have not been directly compared against each other. There is currently no agreed upon standard therapeutic regimen for treatment-naive CLL. Methods: We performed a systematic literature review to identify randomized controlled trials (RCTs) published prior to November 2011 of therapies for previously untreated CLL. We conducted a network meta-analysis using fixed and random effect statistical models to estimate differences between shape and scale parameters of progression-free survival (PFS) curves for each competing therapy. We used the parameter estimates and a Weibull distribution to project mean PFS for each therapy option. Results: Five RCTs were included in our comparison network. Overall, patients were younger (59-65 years), had good performance status based on the Eastern Cooperative Oncology Group scale (ECOG 0-1), and earlier stage disease (Rai 0-II or Binet A or B). The combination regimen fludarabine with cyclophosphamide and rituximab (FCR) was estimated to yield mean PFS of 76 months (95% CrI: 60, 91), FC 60 months (46, 73), fludarabine 38 months (27, 49), alemtuzumab 24 months (15, 32). and chlorambucil 23 months (15, 32). Conclusion: Our results suggest that FCR has relatively higher potential of preventing disease progression in younger, healthier, treatment-naive CLL patients and should be considered an optimal initial treatment strategy for this patient population. However, because estimates are based on model simulation, additional studies of FCR are necessary to clinically validate its therapeutic potential. (C) 2012 Elsevier Ltd. All rights reserved.