In this issue of the journal, Corbigni et al. present a large retrospective multicentric study evaluating serum immunogloblins and paraproteins in patients (pts) with newly diagnosed chronic lymphocytic leukaemia (CLL).1 This study is so far the largest (n = 1505) published on this topic. Its main findings include: hypogammaglobulinaemia of any class was detected in 13·2% of patients, while IgM and IgG paraproteins were present in 4·8 and 10%. Presence of IgM paraprotein significantly correlated with more advanced clinical stages, TP53 aberrations, and inferior treatment-free survival (TFS). Likewise, pts with decreased IgG or paraprotein of the same class experienced shorter TFS. Using receiver operator curve analysis for identification of an optimal threshold, serum concentration of IgA ≤ 0·82 g/l was also associated with inferior TFS. Complex perturbation of the immune system including both humoral and cellular mechanisms has long been recognized as one of the hallmarks of CLL.2 Hypogammaglobulinaemia, or decreased serum concentrations of immunoglobulins (Ig), is by far the most studied phenomenon of immune dysfunction in CLL, with first publications appearing more than 50 years ago.3, 4 Even these early studies in limited numbers of pts noted a significantly higher occurrence of infections in connection with hypogammaglobulinaemia. Results of studies published between 1975 and 2010 have been often conflicting regarding the frequency of hypogammaglobulinaemia at diagnosis (IgG, 5–39%, IgA, 3–37%, IgM, 4–64%) and its prognostic significance, most likely due to relatively small (usually n < 100) patient groups, heterogeneous proportion of early stages, different thresholds for hypogammaglobulinaemia, variable duration of follow-up and selection bias.5-9 After 2010, there have been four major publications dealing with the topic.10-14 Despite inevitable heterogeneity, some recurring patterns could be identified in the literature, namely, association of lower Ig concentrations with advancing disease or antileukaemic therapy and the unfavourable prognostic value of IgA hypogammaglobulinaemia in terms of shorter treatment-free and/or overall survival. From the clinical point of view, combination of significant IgG hypogammaglobulinaemia (the arbitrary cut-off is usually set to ≤ 5g/l) and repeated or severe bacterial infections represents a generally accepted indication for Ig replacement therapy15, 16 in the form of intravenous infusion (IVIG) or more recently, subcutaneous injection (SCIG) which is more cost-effective, more convenient for the patient and yields comparable results.16, 17 The research of monoclonal immunoglobulins/paraproteins in CLL has attracted much less attention than that of hypogammaglobulinaemia. First publications emerged at the end of 1970s and during the 1980s,18, 19 Pangalis et al. 1988,20 and several larger studies reported the frequency of paraproteins in newly diagnosed CLL pts between 7 and 35%;21-23 results regarding their prognostic value of were again inconclusive. In a recent Italian study, the presence of IgG or IgM paraproteins was linked with several unfavourable prognostic factors such as unmutated IGHV genes or TP53 aberrations and correlated with shorter TFS.22 The present study by Corbigni et al. seems to confirm some of the previously published findings, namely association of IgM paraprotein with advanced clinical stages and TP53 abnormalities, as well as significantly shorter TFS in pts with paraprotein or IgG/IgA deficiency (including multivariate analysis). None of these parameters was significant for shorter overall survival in the multivariate analysis; probably due to recent developments of the therapeutic armamentarium, namely novel targeted oral inhibitors such as ibrutinib, idelalisib, and venetoclax, whose introduction has led to marked survival improvement especially in pts with relapsed/refractory CLL. Finally, it is important to point out that the present study has several limitations: first, incidence and severity of infectious complications was not recorded; furthermore, there is no information on the cause of death (infection versus CLL progression versus other); subclasses of IgG and IgA have not been tested; also data regarding Ig replacement therapy is not available. Despite these constraints, the study by Corbigni et al. adds another valuable piece to the puzzle of the area investigating humoral immune dysfunction in CLL. Supported by programme PROGRES Q40/08 and by DRO (UHHK, 00179906). I declare no conflict of interest related to this publication.