Reply to 'CD49d in B-cell chronic lymphocytic leukemia ... - Nature

2 downloads 0 Views 66KB Size Report
Jan 19, 2006 - We are gratified to see the additional data of Zucchetto and colleagues that further support the association of a4 integrin. (CD49d) and CD38 ...
Letters to the Editor

528

Reply to ‘CD49d in B-cell chronic lymphocytic leukemia: correlated expression with CD38 and prognostic relevance’ by Zucchetto et al. Leukemia (2006) 20, 528. doi:10.1038/sj.leu.2404088; published online 19 January 2006

We are gratified to see the additional data of Zucchetto and colleagues that further support the association of a4 integrin (CD49d) and CD38 expression in B-CLL.1–3 Our recently published study discovered the link between CD38 expression and CD49d on leukemic CLL B cells as a result of gene expression profiling.2 Using a novel method of surface antigen expression profiling to define the immunophenotype of specific B-CLL subsets, Zucchetto et al.3 also recently reported that overexpression of CD49d is frequently observed in association with the above average expression of CD38, CD29, and CD49e. The additional data included in the Letter to the Editor extend both those studies2,3 by showing that CD49d expression correlates with the clinically important parameters and outcomes including advanced Rai stage at the time of diagnosis, unfavorable IgVH mutational status, time to progression, and overall survival. Of note, we defined CD38 positivity according to the mean fluorescence intensity (MFI) ratio of CD38-staining in relation to control staining2 whereas Zucchetto et al.1,3 used a cutoff value of 30%. It would, therefore, be informative to compare results with these investigators to determine how many of their patients who are discordantly CD38low/CD49dhigh would have been considered CD38 þ using the methods of our study, that is, displaying CD38 staining at levels above control staining. Moreover, it would also be of interest to see if the association of CD38 and/or CD49d expression has an even greater positive correlation with the Rai stage at the time of blood sampling as compared to the Rai stage at the time of diagnosis. When reviewing our data in the context of Zucchetto et al. report,1 it could be suggested that CD38 and CD49d may act synergistically to confer aggressive disease behavior. It is known that integrins such as CD49d require Ca þ þ to function as adhesion molecules in their role promoting homing and migration of lymphocytes. It is also known that CD38 functions as an ectoenzyme, catalyzing the conversion of NAD þ to cADPR, a potent Ca þ þ mobilizer.4 Therefore, it is plausible that their coexpression could facilitate the homing of CLL B cells to tissue sites with a more favorable environmental milieu that in

turn promotes more rapid leukemic B-cell growth and longer survival, a notion congruent with the finding that the level of a4 integrin expression is associated with the presence of clinically detectable lymphadenopathy.5 It would also be interesting to determine if the function of integrins expressed on CLL B cells is impaired in CD38 but not in CD38 þ CLL B cells. Given the provocative association of these two antigens on CLL B cells now demonstrated by both groups,1–3 and the demonstration that CD38 and CD49d are complementary adverse prognostic factors in B-CLL,1 it is tempting to speculate that these molecules play an important role in more progressive clinical disease. In summary, these reports suggest simultaneous assessment of both antigens may refine the usefulness of CD38 as a useful prognostic tool in B-CLL.

BT Pittner1, TD Shanafelt2, NE Kay2 and DF Jelinek1 Department of Immunology, Mayo Clinic College of Medicine, Mayo Graduate School, Rochester, MN, USA and 2 Division of Hematology, Department of Internal Medicine, Mayo Clinic College of Medicine, Mayo Graduate School, Rochester, MN, USA E-mail [email protected] 1

References 1 Zucchetto A, Bomben R, Dal Bo M, Bulien P, Benedetti D, Nanni P et al. CD49d in B-cell chronic lymphocytic leukemia: correlated expression with CD38 and prognostic relevance. Leukemia 2006 (this volume and edition). 2 Pittner BT, Shanafelt TD, Kay NE, Jelinek DF. CD38 expression levels in chronic lymphocytic leukemia B cells are associated with activation marker expression and differential responses to interferon stimulation. Leukemia 2005, Oct 6; [E-pub ahead of print]. 3 Zucchetto A, Sonego P, Degan M, Bomben R, Dal Bo M, Russo S et al. Signature of B-CLL with different prognosis by shrunken centroids of surface antigen expression profiling. J Cell Physiol 2005; 204: 113–123. 4 Deaglio S, Capobianco A, Bergui L, Durig J, Morabito F, Duhrsen U et al. CD38 is a signaling molecule in B-cell chronic lymphocytic leukemia cells. Blood 2003; 102: 2146–2155. 5 Till KJ, Lin K, Zuzel M, Cawley JC. The chemokine receptor CCR7 and alpha4 integrin are important for migration of chronic lymphocytic leukemia cells into lymph nodes. Blood 2002; 99: 2977–2984.

Reply to Pittner et al. Leukemia (2006) 20, 528–529. doi:10.1038/sj.leu.2404089; published online 12 January 2006

In their published study, Pittner et al.,1 by comparing the gene expression profiling (GEP) of B-cell chronic lymphocytic leukemia (B-CLL) cells either expressing or not expressing the Leukemia

negative prognosticator CD38, were able to find a striking correlation between the overexpression of CD38 and that of the integrin a-chain CD49d. Interestingly, the same conclusion was reached in a recent study of ours, in which the expression of surface antigens, instead of genes, was analyzed.2 So the two studies corroborate each other. In addition, we provided evidences on the clinical value of CD49d expression in B-CLL

Letters to the Editor

529 by demonstrating the association between worse prognosis and overexpression of the molecule, as well as the complementary value of CD38 and CD49d as adverse prognosticators.3 By replying to this report,4 Pittner et al. made, among others, some comments regarding how to compare their own results on CD38 and CD49d expression, evaluated as mean fluorescence intensity (MFI) ratio (i.e. a ratio between the total MFI of the surface marker divided by the total MFI of the corresponding isotypic control), with those reported in our studies,2,3 in which expression values were computed as per cent of positive cells. To address this issue, we selected 61 cases of our series in which expression values for CD38 and CD49d were computed by using both the per cent of positive cells and MFI values, these latter values expressed in linear scale from 0 to 10 000. Median value of MFI ratios for CD38 and CD49d, computed exactly as in Pittner et al.,1 were 1.8 (range 0–36) and 3.8 (range 0–37), respectively. By comparing per cent of positive cell and MFI ratio values, direct correlations yielding significant r-values (0.92 for CD38 and 0.77 for CD49d) were obtained. In our original studies, a cutoff set at 30% of positive cells was chosen to identify cases expressing both CD38 and CD49d at low vs high intensity. Noteworthy, the value of 30% of positive cells was selected as the optimal cutoff point capable to discriminate patient subsets with different survival probabilities by applying for the expression levels of each antigen a specific statistical algorithm, that is the maximally selected log-rank statistics.5 The use of this algorithm in our series basically means to test B-CLL cases for the null hypothesis that the event ‘expression level for a given prognosticator’ (antigen in our series) if less or equal to a given ‘cutoff point’ (% of positive cells in our series), has no influence on the distribution of the chosen response variable (survival in our series).5 The same algorithm has been already successfully applied by us and other groups to identify the optimal percentage of IgVH mutations capable to split patients into subsets with different survivals,6,7 as well as to identify alternative cutoff points of prognostic relevance for CD38.7 Of note, in the case of CD38, the choice of the 30% of positive cells cutoff was also motivated to be in keeping with most part of the current literature8 since, according to the trend of the standardized log-rank statistic plot found by us for the expression values of this marker in our series, basically all the values ranging from 10 to 30% of positive cells yielded optimal separations of patients into two subgroups with different survivals. Similar findings have been also reported in the above quoted study by Krober et al.,7 in which the best cutoff point for CD38 was identified at a value as low as 7% of positive cells. According to the correlation made by us between MFI ratios and per cent of positive cells, the value of 30% of positive cells corresponded to MFI ratio values of 7.5 and 6.5 for CD38 and CD49d, respectively; however, the confidence intervals associated with these figures were highly wide, making in practice extremely difficult a straightforward comparison of our data with those by Pittner et al.1 and the selection of a reliable MFI ratio value corresponding with that of 30% of positive cells. There-

fore, the application of a procedure similar to that employed by us, i.e. the maximally selected log-rank statistics, also in the case of the MFI ratios as utilized by Pittner et al.1 could be of help for identifying the best cutoff level associated with different prognosis also for these parameters. In conclusion, studies of both groups concurrently demonstrated the association between CD38 and CD49d in B-CLL cells;1–4 we are pleased to read that Pittner et al. agree with our suggestion to include CD49d as additional marker in the prognostic assessment of B-CLL patients, due to its independent prognostic value for this disease.

Acknowledgements This work was supported in part by Ministero della Salute (Ricerca Finalizzata IRCCS and ‘Alleanza Contro il Cancro’), Rome, Italy.

A Zucchetto1, R Bomben1, M Dal Bo1, P Bulian1, D Benedetti , P Nanni1, G Del Poeta2, M Degan1 and V Gattei1 1 Clinical and Experimental Hematology Research Unit, Centro di Riferimento Oncologico, IRCCS, Aviano (PN), Rome, Italy and 2 Hematology section, S Eugenio Hospital, University of Tor Vergata, Rome, Italy E-mail: [email protected] 1

References 1 Pittner BT, Shanafelt TD, Kay NE, Jelinek DF. CD38 expression levels in chronic lymphocytic leukemia B cells are associated with activation marker expression and differential responses to interferon stimulation. Leukemia 2005, [E-pub ahead of print]. 2 Zucchetto A, Sonego P, Degan M, Bomben R, Dal Bo M, Russo S et al. Signature of B-CLL with different prognosis by Shrunken centroids of surface antigen expression profiling. J Cell Physiol 2005; 204: 113–123. 3 Zucchetto A, Bomben R, Dal Bo M, Bulian P, Benedetti D, Nanni P et al. CD49d in B-cell chronic lymphocytic leukemia: correlated expression with CD38 and prognosis relevance. Leukemia, (this volume). 4 Pittner BT, Shanafelt TD, Kay NE, Jelinek DF. Reply to ‘CD49d in B-cell chronic lymphocytic leukemia: correlated expression with CD38 and prognosis relevance’. Leukemia, (this volume). 5 Hothorn T, Lausen B. On the exact distribution of maximally selected rank statistics. Computational Statistics & Data Analysis 2003; 43: 121–137. 6 Bomben R, Dal Bo M, Zucchetto A, Zaina E, Nanni P, Sonego P et al. Mutational status of IgV(H) genes in B-cell chronic lymphocytic leukemia and prognosis: percent mutations or antigen-driven selection? Leukemia 2005; 19: 1490–1492. 7 Krober A, Seiler T, Benner A, Bullinger L, Bruckle E, Lichter P et al. VH mutational status, CD38 expression level, genomic aberrations, and survival in chronic lymphocytic leukemia. Blood 2002; 1000: 1410–1416. 8 Damle RN, Wasil T, Fais F, Ghiotto F, Valetto A, Allen SL et al. Ig V gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood 1999; 94: 1840–1847.

Leukemia