Page 137 - 2019_10 resto del Mondo_web

Meta-analysis of the impact of MRD in adult B-ALL
the heterogeneity in the studies in terms of design, popu- lations and treatment protocols. Reassuringly, a prede- fined analysis of key subgroups of interest showed a con- sistent improvement in OS in patients who achieved MRD negativity (Figure 5). As for RFS, no significant dif- ferential subgroup effects were seen. The prognostic value of MRD appeared to be greater in patients who received HSCT before MRD assessment (HR=8.02; 95% CI: 2.32– 27.7) than those assessed after chemotherapy (HR=3.01; 95% CI: 2.08–4.37) or targeted therapy only (HR=1.65; 95% CI: 1.24–2.20). The prognostic value of MRD was less notable in patients who received HSCT treatment after MRD assessment (HR=1.24; 95% CI: 0.86–1.78) than those who received alternative interventions (HR=2.50; 95% CI: 1.88–3.33). As for RFS, the number of studies in some subgroups was small, and many of the confidence intervals overlapped.
Meta-regression
Meta-regression was explored to further investigate the heterogeneity between studies. The analysis was limited
by the small number of studies that provided information for certain covariates (i.e. those used to define the sub- groups), and for this reason we focused on RFS for which more studies were available. Heterogeneity remained even after adjustment for almost all covariates, as reflected by a significant within-group Q-statistic (Qw) (Online Supplementary Table S9). The exception was risk group; however, interpretation of the relevance of this covariate is limited by its inclusion in only three studies.
Of note, the Tierney method30 used to calculate the haz- ard ratio for each study had no significant effect on the treatment difference.
Discussion
This systematic review and meta-analysis found that in adults with B-ALL, achieving MRD negativity was consis- tently associated with better survival outcomes than those of patients with MRD-positive status. In addition, the benefit of achieving MRD negativity was evident in all the
Figure 2. Forest plot of relapse-free survival hazard ratios for all studies included in the primary analysis. CI: confidence interval; HR: hazard ratio; MRD: minimal residual disease; MRD neg: minimal residual disease-negative status; MRD pos: minimal residual disease-positive status.
haematologica | 2019; 104(10)
2033