Abstract

TO THE EDITOR: The article by Sehn et al analyzed the prognostic significance of the concordant versus discordant bone marrow involvement (BMI) in a cohort of 795 patients with diffuse large B-cell lymphoma treated with rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone. On the basis of this large series, the authors found that both types of BMI produced a negative impact in terms of progression-free survival, whereas only the concordant involvement was a negative prognostic factor for overall survival that was also independent of the International Prognostic Index score at multivariate analysis. With this report, the first step toward the understanding of this topic seemed to be completed, and it is time to look further. Nevertheless, other important clues could have been retrieved from this representative series with relatively small additional effort. For example, among the group of patients with concordant BMI, a subset analysis of outcomes according to the cell of origin (COO) (ie, the germinal center B-cell v non–germinal center B-cell categories) could potentially further discriminate variables with prognostic relevance. Because categorization according to the COO could be performed retrospectively by means of the algorithm of Hans et al or other similar algorithms that are based on conventional immunohistochemistry, it still could be explored in a follow-up report of this series. Ideally, gene-expression profiling should also be analyzed in future prospective or retrospective studies to determine whether the different COO signatures are associated with different risk of BMI or different outcomes within the category of concordant BMI, but thus far, this has been limited by the availability of frozen tissue. Concerning the group of patients with discordant BMI, it still remains to be determined whether those low-grade infiltrates that are clonally related with the diffuse large B-cell lymphoma (which define a possible transformation from a low-grade to a higher-grade histology) lead to similar biologic effects in terms of progression-free survival and overall survival compared with clonally unrelated cases (which define the presence of two different lymphomas in the same patient). This delineation could be also explored in a retrospective series by determining B-cell receptor rearrangements by polymerase chain reaction– based methods applied on paraffin-embedded tissue samples. As a result of the difficulties in gathering sufficiently large series of patients with BMI and adequate tissue availability, it would be of major interest for the clinical community to maximize the analysis of all variables that could bear a potential predictive or prognostic impact and are assessable by widely available techniques or tools.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call