Correspondence

Molecular Profiling of Lymphoma

N Engl J Med 2002; 347:1376-1377October 24, 2002

Article

To the Editor:

With the use of gene profiling in patients with diffuse large-B-cell lymphoma, Rosenwald et al. (June 20 issue)1 found that overall survival after anthracycline-based chemotherapy differed significantly among the three gene-expression subgroups they identified. Was this finding based on a comparison of the germinal-center B-cell–like subgroup with the type 3 subgroup and with the activated B-cell–like subgroup, or on a comparison of the germinal-center B-cell–like subgroup with the other two subgroups combined? Figure 2 of the article by Rosenwald et al. ranks patients in quartiles 1, 2, 3, and 4, with five-year survival rates of 73, 71, 34, and 15 percent, respectively. In comparing these results with the International Prognostic Index (IPI) (Figure 3 of the article), the authors merged quartiles 1 and 2 into one group and quartiles 3 and 4 into a second group, resulting in a falsely worse outcome for quartile 3 and a falsely better outcome for quartile 4.

Saad Akhtar, M.D.
King Faisal Specialist Hospital and Research Center, Riyadh 11211, Saudi Arabia
akhtars18@hotmail.com

1 References

1. 1

   Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 2002;346:1937-1947
   Full Text | Web of Science | Medline

To the Editor:

Rosenwald et al. provide further data on molecular profiling in diffuse large-B-cell lymphoma and separate this entity into three large subgroups. We would like to know whether cytogenetic data were available, and if so, whether they were correlated with the overexpression of the bcl-2 gene in the authors' analysis. Analysis of a larger number of tumor samples with different DNA microarrays, along with a complementary approach of monitoring of single-nucleotide polymorphisms,1 might be needed to identify more accurate gene-expression signatures associated with a good or poor response to therapy.

M. Sitki Copur, M.D.
Peter Ledakis, M.D.
Mark Bolton, M.D., Ph.D.
Saint Francis Cancer Center, Grand Island, NE 68802
mcopur@sfmc-gi.org

1 References

1. 1

   Cargill M, Altshuler D, Ireland J, et al. Characterization of single-nucleotide polymorphisms in coding regions of human genes. Nat Genet 1999;22:231-238[Erratum, Nat Genet 1999;23:373.]
   CrossRef | Web of Science | Medline

Author/Editor Response

Dr. Staudt replies:

To the Editor: In response to Dr. Akhtar: the survival rates among patients assigned to the three gene-expression subgroups of diffuse large-B-cell lymphoma were significantly different (P<0.001) when all three groups were considered together.1 In addition, patients in the germinal-center B-cell–like subgroup had a higher survival rate than did patients in the activated-B-cell–like subgroup (P<0.001) and those in the type 3 subgroup (P=0.01).

The IPI and the gene-expression–based outcome predictor were statistically independent predictors of survival (P<0.001).1 The gene-expression–based outcome predictor was significantly associated with survival among patients assigned to the low-clinical-risk group (IPI score, 0 to 1; P<0.001), the intermediate-clinical-risk group (IPI score, 2 to 3; P<0.001), and the high-clinical-risk group (IPI score, 4 to 5; P=0.02).1 In the Kaplan–Meier plots, the quartile groupings based on the outcome-predictor score were simply a tool for showing the statistical independence of these two methods of outcome prediction. In Figure 3, we chose to combine quartiles 1 and 2 and quartiles 3 and 4 because the numbers of patients in some individual quartiles within an IPI group were quite small. The method of grouping patients in the Kaplan–Meier plots did not alter the finding that the IPI and the gene-expression–based outcome predictor are statistically independent predictors of survival after chemotherapy among patients with diffuse large-B-cell lymphoma.

In response to Copur et al.: Polymerase-chain-reaction assay for the t(14;18) translocation involving bcl-2 was performed, and all instances of this translocation were detected in the germinal-center B-cell–like subgroup of diffuse large B-cell lymphoma. In this subgroup, lymphomas with the t(14;18) translocation had higher bcl-2 messenger RNA (mRNA) levels than those without the t(14;18) translocation. Notably, the activated B-cell–like subgroup had high levels of bcl-2 mRNA without the t(14;18) translocation, possibly because of transcriptional activation of this gene by nuclear factor-κB transcription factors.2 It is certainly possible that either germ-line–encoded or somatically acquired single-base-pair changes in critical genes influence the survival rates among patients with diffuse large-B-cell lymphoma.

Louis M. Staudt, M.D., Ph.D.
National Cancer Institute, Bethesda, MD 20892
lstaudt@mail.nih.gov

for the Lymphoma/Leukemia Molecular Profiling Project

2 References

1. 1

   Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 2002;346:1937-1947
   Full Text | Web of Science | Medline

2. 2

   Davis RE, Brown KD, Siebenlist U, Staudt LM. Constitutive nuclear factor kappaB activity is required for survival of activated B cell-like diffuse large B cell lymphoma cells. J Exp Med 2001;194:1861-1874
   CrossRef | Web of Science | Medline