Fig. 2
Prognostic effect of non-malignant B-cell clusters. (A) Kaplan-Meier (K-M) plot and log-rank test of progression free survival (PFS) comparing patients with FlowSOM-defined NMP clusters above or below the median value (1.52%) at the early post-CIT timepoint*. (B) Multivariable Cox proportional hazard analysis assessing the association between PFS and baseline characteristics, treatment allocation, post-chemoimmunotherapy (CIT) MRD status (using a cut-off of 10−3 since only five patients achieved 10−4), and NMP cluster size at the early post-CIT timepoint*. (C) K-M plot and log-rank test of overall survival (OS) comparing patients with NMP clusters above or below the median (1.52%) at the early post-CIT timepoint. (D) Multivariable Cox proportional hazard analysis assessing the association between OS and baseline characteristics, treatment allocation, post-CIT MRD status (using a cut-off of 10−3), and NMP cluster size at the early post-CIT timepoint. NMP cluster frequency was analysed as a continuous variable in (B) and (D). *One patient progressed prior to the post-treatment blood sample and was excluded.
