Title: A Peripheral Blood T Cell Clone is a Prognostic Marker in Mycosis Fungoides
Abstract: To the Editor: DrMuche et al., 2000Muche M. Lukowsky A. Alnhudt C. Gellrich S. Sterry W. Peripheral blood T cell clonality in mycosis fungoides–an independent prognostic marker?.J Invest Dermatol. 2000; 115: 504-505Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar have raised several criticisms of our recent article (Fraser-Andrews et al., 2000Fraser-Andrews E. Woolford A. Russell Jones R. Seed P.T. Whittaker S.J. Detection of a peripheral blood T cell clone is an independent prognostic marker in mycosis fungoides.J Invest Dermatol. 2000; 114: 117-121Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar). Specifically they question our findings that a peripheral blood T cell clone is an independent prognostic feature in mycosis fungoides (p = 0.03 with a hazards ratio of 2.6 after adjusting for age, skin stage, and lymph node stage). Our results are based on multivariate analysis of 66 patients with up to 10 y follow-up and crucially includes analysis of data at diagnosis for the majority of patients. Whilst we acknowledge that further prospective studies of larger numbers with longer follow-up are required, their comment that ‘‘the survival or time to progression in stage of the reported mycosis fungoides (MF) patients with or without periphreal blood T cell clonality’' is not relevant if data are analyzed from diagnosis. In other words the results at diagnosis give a measure of tumor burden that can then be entered into a multivariate analysis. The reason for the authors differing conclusions is because their study is not based on data at the time of diagnosis and predominanlty consists of patients with T1 and T2 MF and only four patients with T3 stage disease. In addition the follow-up in their study is much shorter than ours. They state that the presence of a peripheral blood T-cell clone fluctuated during the course of the disease in some of their patients, which would be expected and might reflect a partial response to different therapies. Once again this emphasises the importance of analysing data from diagnosis. Inevitably they cannot perform a multivariate analysis on these results and therefore cannot make any conclusoins regarding the prognostic implications. We agree that the proportion of peripheral blood T cell clones detected will reflect the sensitivity of the method and certainly the use of oligonucleotides for the tumor specific V(D)J clonal TCR gene rearrangement would be too sensitive. Intriguingly the relative proporitons of their T1/T2 patients with preipheral blood T cell clones are similar to our findings in early stage disease. This suggests that the sensitivity of the two techniques are broadly similar, which would be expected given that both methods depend on distinguishing clonal rearrangements on the basis of sequence and size. In fact our method employed monoplex rather than multiplex polymerase chain reaction and radioactively labelled products which should increase the sensitivity. There is certainly no basis for the authors contention that our approach is less sensitive but comparative studies would be worthwhile. The study by Laetsch et al. has reveaeled similar findings to our study, namely that almost 75% of CTCL patients with stage III–IVb disease have peripheral blood T cell clones. The essential aspect of this study is that both the presence of a peripheral blood T cell clone and the CD4+/CD7-cell count are independently correlated with stage. They have not drawn any conclusions regarding the prognostic significance of their data as suggested by Muche et al. Finally Muche et al. speculate about physiologic recirculation of tumor cells. We agree that tumor cells will recirculate physiologically but the detection of these cells can still represent a measure of tumor burden and these two concepts are not mutually exclusive.