Prognostic relevance of in vitro response to cell stimulation via surface IgD in binet stage a CLL


Several studies have indicated that stimulation via surface immunoglobulin (sIg) may be an important promoting factor for clonal expansion in chronic lymphocytic leukaemia (CLL) (Chiorazzi & Ferrarini, 2003; Stevenson & Caligaris-Cappio, 2004; Chiorazzi et al, 2005). CLL clones expressing unfavourable prognostic markers, such as ZAP-70 or CD38, have viable sIg-dependent signal transducing pathways, while cells from cases with favourable cellular markers are often anergic to sIg cross-linking. In addition, cases with a more aggressive disease are characterized by the expression of sIg encoded by unmutated IGHV/KV/LV genes. These molecules often have polyspecific activity and bind to a battery of antigens, including self-antigens. sIg from indolent cases are encoded by somatically mutated IGHV/VL genes and do not have polyspecific activity, suggesting that continuous stimulation by self-antigens in vivo may drive the clonal expansion in aggressive CLL cases (Chiorazzi & Ferrarini, 2003; Stevenson & Caligaris-Cappio, 2004; Chiorazzi et al, 2005). Finally, neoplastic cells from a substantial number of patients with aggressive clinical courses share ‘stereotyped receptors’, encoded for by identical IGHV DJ and IGHV JL genes and very similar Complimentary-Determining Region 3, suggesting exposure to selective pressure by special antigens at certain points of lymphomagenesis (Messmer et al, 2004; Widhopf et al, 2004; Murray et al, 2008).

Most CLL clones express both sIgM and IgD and the two isotypes share the same antigen-specific combining site. The available evidence supports the notion that sIgM actively participates in the process of clonal expansion, but information regarding the role of sIgD is scanty (Zupo et al, 2000; Lanham et al, 2003). The present study investigated the effect of stimulating CLL cells in vitro by cross-linking their sIgD in a cohort of patients collected through a cooperative multicenter study (Gruppo Italiano Studio Linfomi). The data were correlated with clinical features and the findings suggested the involvement of sIgD in clonal expansion.

CLL cells, purified from 106 patients, were exposed to anti delta antibody (aδAb) in vitro and apoptosis was measured after 48 h by propidium iodide staining and flow-cytometry (Zupo et al, 2000). Patients were subdivided into three groups, depending upon the type of response observed. In one group (33 cases, I group), inhibition of spontaneous apoptosis was seen, when the apoptosis values following sIgD cross-linking were at least 20% lower than those observed in the absence of cell exposure to aδAb. In a second group (eight cases, D group), exposure to aδAb caused 20% greater cell death than that spontaneously observed in absence of the antibody. In a third group (65 cases, N group), exposure to aδAb failed to modify spontaneous apoptosis below or above 20% of the spontaneous values.

Patient clinical courses were correlated with their classification into one of the three groups defined above. Patients in group D or I were grouped together and classified as ‘responders’. Disease progression was measured as the time elapsed from diagnosis to first treatment (TFS). After a median follow up of 4 years, 36 patients were treated. Cox univariate analysis showed that a cell response to aδAb (group D + I) conferred a hazard ratio (HR) for disease progression greater than the absence of response (Fig 1A). All patients were studied for ZAP-70 and CD38 expression, and IGHV mutational status (Fig 1B). At Cox multivariate analysis, response to aδAb remained an independent variable also in the presence of the three other markers (P = 0·009).

Figure 1.

 Response to sIgD cross-linking by CLL cells purified from 106 patients. (A) Cells were purified and cultured with aδAb [goat antihuman δ-chain antibodies (Sigma, St Louis, MO, USA), at the concentration of 10 μg/ml] for 48 h, as described (Zupo et al, 2000). Apoptosis was determined by propidium iodide staining of permeabilized cells at the end of culture period. Cases were divided into three groups depending upon whether there was no change in apoptosis (N) or whether apoptosis was inhibited (I) or enhanced (D) by more than 20% compared to controls not exposed to aδAb. The data of I and D groups were pooled when TFS was considered. By Cox univariate analysis, the risk of treatment start was significantly higher for I + D group. Cox-derived estimated survival curves according to the response to IgD cross linking is also represented. (B) The risk of therapy requirement for patients whose cells expressed for CD38 and ZAP-70 and utilizing unmutated IGHV genes.

Overall, these observations suggest that stimulation via sIgD, perhaps by self-antigens, may have pathogenic relevance by facilitating clonal expansion in vivo. Stimulation via sIgD may prevent cell apoptosis in the majority of cases, although this hypothesis may be somewhat speculative, because the outcome of cell stimulation may be related to the number of apoptotic/antiapoptotic signals delivered in vivo. From the standpoint of our experimental design, the outcome of cell stimulation (apoptosis inhibition or induction) following sIgD cross-linking represents a readout system for cell susceptibility to the stimulus.

Next, we investigated possible correlations between protein tyrosine phosphorylation following sIgD cross ligation and in vitro response. Preliminary data showed that the maximum tyrosine phosphorylation was observed after 1 min of cell exposure to aδΑb (P = 0·003) (Fig 2A). Tyrosine phosphorylation at 1 min was significantly different in 10 N and 14 I + D cases (Fig 2B). As stimulation may be influenced by sIgD density, we measured sIgD density as mean fluorescence intensity (MFI) by flow-cytometry in N (65 cases) and I + D (38 cases) patients (Fig 2C). The mean MFI value ± SEM was 27 ± 6·7 (median 9) in the whole patient cohort (106). When cases with an MFI ≥ 9 were analysed separately from those with an MFI < 9, it was clear that there was more N cases with lower density surface IgD (42/56 [75%] vs. I + D 14/56, [25%]). Among cases with higher density sIgD, the N cases were 23/47 (49%) and I + D cases were 24/47 (51%). This difference was statistically significant (P = 0·014).

Figure 2.

 (A, B) Protein tyrosine phosphorylation following exposure to aδAb. (A) Cells from 24 cases were exposed to aδAb for various times and protein tyrosine phosphorylation determined as described in (Cutrona et al, 2008). The total amount of phosphorylation was calculated by summing up the phosphorylation of the single bands (1D Image Analysis Software version 3.5; Kodak, Rochester, NY USA) (Cutrona et al, 2008). The Friedman test, which compares the distributions of more related variables, demonstrated significant differences in protein tyrosine phosphorylation after 1 and 10 min or 5 and 10 min of stimulation. (B) Protein tyrosine phosphorylation bands following exposure to aδAb for 1 min was measured in CLL cells from 14 patients from I + D and 10 patients from N group. (C) Correlation between stimulation via surface IgD capacity and cell surface IgD density measured as mean fluorescence intensity (MFI), in N (65 cases) and I + D (38 cases) patient groups. (D) Patients with high and low surface IgD were studied for their TFS. The values recorded were significantly different.

Finally, patients with high cellular IgD expression (MFI ≥ 9) had a shorter TFS (Fig 2D) and a 2·6-fold higher HR, determined by Cox univariate analysis, than the group with IgD MFI < 9. sIgD density maintained an independent predictive value when analysed together with ZAP-70, CD38 and IGHV status (P = 0·045).

Studies on the same cohort revealed a correlation between an apoptotic response of the CLL cells in vitro to sIgM cross-linking and an unfavourable clinical course. Again, sIgM density correlated with both cellular response and clinical course. Although the issue goes beyond the scope of this letter, it is of note that response to IgM cross-linking is not independent of other prognostic factors (CD38, ZAP-70 and IGHV status) at multivariate analysis. In addition, comparative analysis of the cellular responses to stimulation via sIgM and sIgD revealed heterogeneous cases capable of responding to both, one or none of the stimuli.

sIgD density and response to sIgD cross-linking may represent an additional predictor of disease progression, although these parameters may not be of great practical value for everyday clinical practice. Nevertheless, both studies of sIgD density and particularly of the response to stimulation via sIgD appear to provide useful information on the mechanisms that potentially lead to disease progression.


Supported from Associazione Italiana Ricerca sul Cancro (AIRC) (to FM, AN and MF), FIRB (Grant no RBIP06LCA9, to MF) MIUR, CIPE (2006-8, CBA project, to SZ), ISS (to SZ), progetto Regione Liguria (to SZ). Progetti Progetti Strategici – Ricerca Finalizzata Ministero Italiano della Salute ‘RFPS_2006_3_33_99_60’ (to GC) and ‘RFPS_2006_340196’ (to FM and MF); progetto ordinario ricerca finalizzata Ministero Italiano della salute-2007 (to GC), progetto Compagnia San Paolo (to GC), the Fondazione Internazionale Ricerche Medicina Sperimentale (FIRMS) provided financial and administrative assistance, Fondazione ‘Amelia Scorza’ onlus, Cosenza, Italy; GISL (Gruppo Italiano Studio Linfomi); and Associazione Italiana contro le Leucemie -Milano. SM is supported by fellowship from the Fondazione Italiana Ricerca sul Cancro (FIRC). We thank Laura Veroni and Brigida Gulino for their valuable secretarial assistance.