Correspondence
2360 may previously exist and its clinical manifestation mimics MDS, that is, myelodysplastic features with cytopenias, but its biologic nature is closely associated with myeloproliferative disorders. Unfortunately, we did not detect its mutation before MF in our MDS/MF patients. Although MDS patients with MF have an unfavorable prognosis, the current study demonstrates the genotypic heterogeneity of such patients.
Acknowledgements Thanks are due to Professor J Patrick Barron for his review of this manuscript and Mr Kunio Hori and Tohru Makino, NovusGene, Tokyo, for their technical assistance. This work was supported in part by a Grant-in-Aid for ‘Intractable Hematopoietic Diseases’ from the Ministry of Health, Welfare, and Labor, Japan (to KO), the ‘High-Tech Research Center’ Project from the Ministry of Education, Culture, Sports, Science and Technology: MEXT) (to KO, JHO), and by the ‘University-Industry Joint Research Project’ from MEXT (to KO, JHO).
K Ohyashiki1 Y Aota1 D Akahane1 A Gotoh1 K Miyazawa1 Y Kimura1 JH Ohyashiki2
1 The First Department of Internal Medicine, Tokyo Medical University, Tokyo, Japan; and 2 Intractable Immune System Research Center, Tokyo Medical University, Tokyo, Japan
References 1 Kralovics R, Passamonti F, Buser AS, Teo SS, Tiedt R, Passweg JR et al. A gain-of-function mutation of Jak2 in myeloproliferative disorders. N Engl J Med 2005; 352: 1779–1790. 2 Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJ et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and agnogenic myeloid metaplasia. Cancer Cell 2005; 7: 387–397. 3 James C, Ugo V, Le Couedic JP, Staerk J, Delhommeau F, Lacout C et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature 2005; 484: 1144–1148. 4 Baxter EJ, Scott LM, Campbell PJ, East C, Fourouclas N, Swanton S et al. Cancer genome project. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 2005; 365: 1054–1061. 5 Steensma DP, Dewald GW, Lasho TL, Powell HL, McClure RF, Levine RL et al. The JAK2 V617F activating tyrosine kinase mutation is an infrequent event in both ‘atypical’ myeloproliferaive disorders and myelodysplastic syndromes. Blood 2005; 106: 1207–1209. 6 Ohyashiki K, Sasao I, Ohyashiki JH, Murakami T, Iwabuchi A, Tauchi T et al. Clinical and cytogenetic characteristics of myelodysplastic syndromes developing myelofibrosis. Cancer 1991; 68: 178–183. 7 Ohyashiki K, Yokoyama K, Kimura Y, Ohyashiki JH, Ito Y, Kuratsuji T et al. Myelodysplastic syndrome evolving into a myeloproliferative disorder: one disease or two? Leukemia 1993; 7: 338–340.
Angiogenesis and mast cells in Hodgkin lymphoma
Leukemia (2005) 19, 2360–2362. doi:10.1038/sj.leu.2403992;
published online 13 October 2005 TO THE EDITOR
Hodgkin lymphoma (HL) differs from other lymphomas because the malignant cells, the Hodgkin and Reed–Sternberg (HRS) cells, are in minority and the majority of the tissue consists of surrounding benign cells, for example, eosinophilic granulocytes and mast cells, fibrosis and a varying number of microvessels. It has recently been reported that angiogenesis correlates to poor prognosis in HL.1 We have previously reported that HL patients with many mast cells in their tumour tissue have a worse prognosis.2 Mast cells produce functionally active CD30 ligand (CD30L) and the poorer prognosis has been proposed to be caused by a stimulation of HRS by CD30L.3 Furthermore, we have shown that mast cells, upon stimulation with CD30, release cytokines and chemokines, among which interleukin-8 (IL-8) is known to have angiogenic properties (manuscript in preparation). In other lymphomas, mast cells are proposed to contribute to angiogenesis.4 In order to increase our understanding of inflammatory cells, their importance in tumour progression and especially angiogenesis in HL, we investigated the possible relation between the Correspondence: Dr I Glimelius, Department of Oncology, Radiology and Clinical Immunology, Uppsala University Hospital, Rudbeck laboratory C11, Uppsala S-751 85, Sweden; Fax: þ 46 18 611 34 32; E-mail:
[email protected] Received 30 August 2005; accepted 15 September 2005; published online 13 October 2005 Leukemia
number of mast cells and the microvessel count in primary diagnostic HL tissue. We also wanted to further elucidate the prognostic implication of microvessel count in HL. Patient samples and clinical data were acquired from the database of the National Health Care Programme for HL in Sweden. A total of 120 patients treated with curative intention, according to the principles of the Health Care Programme2 in ¨ rebro health care region between 1989 and 1994, the Uppsala/O were included. The paraffin-embedded tissue samples were from HL involved lymph nodes from the primary diagnosis. The clinical characteristics are presented in Table 1. Progression free survival (PFS) and HL specific survival (HLS) were analysed. The mean follow-up of living patients was 11 years (range 6–15 years). The estimation of the number of microvessels immunohistochemically stained for CD31 (Figure 1), was done by one of the authors using the Chalkley technique.5 Three to five fields with the highest concentration of vessels (a hot spot) were counted and an average of the highest three countings in every case was used. In all, 20 cases were recounted independently by another author and the counts correlated with an R-value of 0.75, (P ¼ 0.0002). All evaluations were done without knowledge of patient data. The counts varied from 1 to 12 vessels/hot spot. The median was 3 and the 75th percentile was 4.3 vessels/hot spot. Nonbulky disease correlated to high microvessel count (Table 1) and there was a lower proportion of patients with WBC415 in the upper quartile group (data not shown), but there were no other correlations to histology, laboratory parameters, stage, B-symptoms or sex. In univariate analyses, HL patients with a high microvessel count, cutoff at the 75th percentile (n ¼ 33), have a worse PFS
Correspondence
2361 Table 1
Clinical characteristics in relation to microvessel count
Clinical characteristics
All CD31 stained cases Number
44.3 CD31 stained vessels/hot spot Number (% of all CD31 stained cases)
o4.3 CD31 stained vessels/hot spot Number (% of all CD31 stained cases)
Age
Mean 42, Median 38, Range: 12–79
Mean 51, Median 56, Range: 6–76
Mean 39, Median 31, Range: 12–79
Stage I–IIA Stage IIB–IV
71 53
20 (28) 13 (25)
51 (72) 40 (75)
NS
A B-symptoms
82 42
22 (27) 11 (26)
60 (73) 31 (74)
NS
Bulky disease Nonbulky disease
25 99
2 (8) 31 (31)
23 (92) 68 (69)
0.02
410 mast cells/HPF o10 mast cells/HPF Missing mast cell data
70 52 2
18 (26) 15 (29) 0
52 (74) 37 (71) 2
124
33 (27)
91 (73)
Total
P-value: cases with many microvessels vs few
0.002
NS
NS ¼ not significant.
Figure 1 Hodgkin lymphoma tissue immunohistochemically stained with the monoclonal antiendothelial antibody, anti-CD31, at magnification 100; showing an area with many microvessels.
compared with those with a lower number of microvessels (Figure 2). The 75th percentile resulted in the best separation in outcome compared to the 25th and 50th percentiles. The HLS was not significantly influenced by the number of vessels (P ¼ 0.17). When tested in a multivariate analysis against age and low s-albumin, the most powerful prognostic factors, the microvessel count was not statistically significantly related to PFS (P ¼ 0.10), probably due to the very strong influence of age on prognosis. The observed correlations between high number of microvessels and high age on the one hand (Table 1) and worse PFS, on the other hand (Figure 2) are in line with the recent study by Korkolopoulou et al.1 However, in that study a morphometric approach was used where they investigated different features of the microvessels.1 It might be speculated that the higher number of microvessels in elderly patients points towards a biological difference between HL in younger and older patients.
Figure 2 Kaplan–Meier curve showing that patients (n ¼ 120) with high microvessel count, have a worse progression-free survival compared to patients with low microvessel count (P ¼ 0.03).
Immunostaining and grouping according to the number of the tryptase positive mast cells (Figure 3) were done in an earlier report on the same material and has thus been previously described.2 Five additional cases were stained and counted. Mast cell data was missing for two patients because of lack of tumour material. Unexpectedly, a high microvessel count did not correlate to high number of mast cells (Spearman R ¼ 0.05, P ¼ 0.62). Neither were there any correlations between the cutoff level for vessels (4.3 vessels/hot spot) and the number of mast cells (Mann–Whitney U-test P ¼ 0.55), between the number of mast cells grouped into more or less than 10 and the absolute number of vessels (Mann–Whitney U-test P ¼ 1.00), or, as shown in Table 1, between the groups of more or less than 10 mast cells, compared to the groups of few and many microvessels (w2 test P ¼ 0.86). Leukemia
Correspondence
2362 however, not possible to discriminate between newly formed and old microvessels. In conclusion, this study adds further proof that high microvessel count is of importance for the prognosis in HL but there is no correlation to mast cell count. Instead, the correlation between mast cells and poor prognosis is probably due to, for example, CD30/CD30L interaction with the HRS cells.
Acknowledgements We thank Lila Shokohideh and Majlis Book for their expert technical assistance. This study was supported by The Swedish Cancer Society, Lions Cancer research foundation in Uppsala and FoU-forum in Ga¨vle.
Figure 3 Hodgkin lymphoma tissue immunohistochemically stained with the monoclonal antibody G3, recognising the mast cellspecific protein tryptase, at magnification 400; showing mast cells in brown. The arrow shows a Hodgkin cell.
The poor prognosis, previously described by us, when many mast cells are found in the tumours2 could thus not be explained by a higher microvessel count in the tumours. However, mast cells are the predominant CD30L expressing cells in HL3 and CD30-CD30L activation is of importance for tumorgenesis in HL.3 A plausible explanation for the poor prognosis in mast cell rich HL is therefore that the proliferation and activation of the tumour cells are enhanced by CD30L expressed by the mast cells.3 The lack of correlation between high microvessel count and many mast cells in our study might indicate a different pathway for the proliferation of microvessels in HL compared to other lymphomas.4 It is, however, still possible that mast cells contribute to microvessel stimulation, but this is not reflected in mast cell numbers. Mast cells express several factors that can affect angiogenesis both directly and indirectly, including vascular endothelial growth factor (VEGF), basic fibroblast growth factor, transforming growth factor-b, tumour necrosis factor-a, IL-8, histamine, tryptase, matrix metalloproteinase-9 and heparine.6 Variations in the production and release of these factors by mast cells in HL, may be of greater relevance than the actual mast cell numbers. However, there are other possible pathways for vascularization in HL, for example, VEGF is produced by the HRS cells.7 Eotaxin, expressed in HL, but not from the mast cells,8 induces the formation of blood vessels. There are also other surrounding cells in the HL tumour tissue, which may influence vascularization, for example, macrophages.6 Thus, several possible mechanisms for angiogenesis exist in HL and further elucidation is needed. The best method to discriminate between blood and lymphatic vessels in lymph nodes is debated. Some authors claim that CD31 is the best discriminator whereas others favour CD34.9,10 In our study, the antiendothelial antibody CD31 was used to verify microvessels. In the study by Korkolopoulou et al1 CD34 was used. With any of the methods so far used, it is,
Leukemia
I Glimelius1 A Edstro¨m1 M Fischer1 G Nilsson2 C Sundstro¨m2 D Molin1 R-M Amini1 G Enblad1
1 Department of Oncology, Radiology and Clinical Immunology, Rudbeck Laboratory, Uppsala University, Uppsala, Sweden; and 2 Department of Genetics and Pathology, Rudbeck laboratory, Uppsala University, Uppsala, Sweden
References 1 Korkolopoulou P, Thymara I, Kavantzas N, Vassilakopoulos TP, Angelopoulou MK, Kokoris SI et al. Angiogenesis in Hodgkin’s lymphoma: a morphometric approach in 286 patients with prognostic implications. Leukemia 2005; 19: 894–900. 2 Molin D, Edstrom A, Glimelius I, Glimelius B, Nilsson G, Sundstrom C et al. Mast cell infiltration correlates with poor prognosis in Hodgkin’s lymphoma. Br J Haematol 2002; 119: 122–124. 3 Molin D, Fischer M, Xiang Z, Larsson U, Harvima I, Venge P et al. Mast cells express functional CD30 ligand and are the predominant CD30L-positive cells in Hodgkin’s disease. Br J Haematol 2001; 114: 616–623. 4 Ribatti D, Crivellato E, Roccaro AM, Ria R, Vacca A. Mast cell contribution to angiogenesis related to tumour progression. Clin Exp Allergy 2004; 34: 1660–1664. 5 Fox SB, Leek RD, Weekes MP, Whitehouse RM, Gatter KC, Harris AL. Quantitation and prognostic value of breast cancer angiogenesis: comparison of microvessel density, Chalkley count, and computer image analysis. J Pathol 1995; 177: 275–283. 6 Norrby K. Mast cells and angiogenesis. Apmis 2002; 110: 355–371. 7 Doussis-Anagnostopoulou IA, Talks KL, Turley H, Debnam P, Tan DC, Mariatos G et al. Vascular endothelial growth factor (VEGF) is expressed by neoplastic Hodgkin-Reed-Sternberg cells in Hodgkin’s disease. J Pathol 2002; 197: 677–683. 8 Jundt F, Anagnostopoulos I, Bommert K, Emmerich F, Muller G, Foss HD et al. Hodgkin/Reed-Sternberg cells induce fibroblasts to secrete eotaxin, a potent chemoattractant for T cells and eosinophils. Blood 1999; 94: 2065–2071. 9 Hattori H. Caution should be taken in using CD31 for distinguishing the vasculature of lymph nodes. J Clin Pathol 2003; 56: 638–639. 10 Vermeulen PB, Gasparini G, Fox SB, Toi M, Martin L, McCulloch P et al. Quantification of angiogenesis in solid human tumours: an international consensus on the methodology and criteria of evaluation. Eur J Cancer 1996; 32A: 2474–2484.
