In European journal of haematology
OBJECTIVE : In chronic lymphocytic leukaemia, growing evidence has accumulated about long-term outcomes of first-line treatments. Our objective was to perform indirect comparisons across first-line treatments.
METHODS : We applied the Shiny method, an artificial intelligence technique that analyses Kaplan-Meier curves and reconstructs patient-level data. Reconstructed patient data were then evaluated through standard survival statistics and indirect head-o-head comparisons. The endpoint was progression-free survival (PFS).
RESULTS : Seven first-line treatments were studied (1,983 patients). Three treatments based on either ibrutinib or venetoclax (i.e. ibrutinib monotherapy, ibrutinib+ rituximab/obinutuzumab and venetoclax+obinutuzumab) showed a very similar survival pattern. The PFS for these 3 treatments was significantly better than that of the remaining 4 treatments (fludarabine+cyclophosphamide+rituximab, chlorambucil+obinutuzumab, bendamustine+rituximab, and chlorambucil monotherapy). Regarding chlorambucil+ obinutuzumab, a significant between-trial variability was found.
CONCLUSIONS : Long-term results are particularly favorable to ibrutinib (alone or in combination) and discourage further use of chlorambucil. As in other studies based on the Shiny method, the multi-treatment Kaplan-Meier graph summarized the available evidence in comparative terms. The evidence generated this way contributes to define the place in therapy of individual agents.
chlorambucil, chronic lymphocytic leukaemia, cyclophosphamide rituximab, fludarabine, ibrutinib, obinutuzumab, progression-free survival, rituximab, venetoclax