Clinicopathologic correlations of bone marrow angiogenesis in chronic myeloid leukemia: a morphometric study

Article metrics

Abstract

Various morphometric characteristics of microvessels, highlighted by means of anti-CD34 immunohistochemical staining, were evaluated in the bone marrow of 52 patients with chronic myeloid leukemia (CML) in chronic phase, in relation to several clinicopathologic parameters. Twenty control bone marrows and 15 cases of CML in blastic phase were also studied. Microvessel density (MVD), total vascular area (TVA) and several size- and shape-related parameters were quantitated in the region of most intense vascularization using image analysis. Overall, the group of chronic phase CML had higher MVD and size-related parameters and more branching microvessels than controls. Blastic phase was characterized by increased numbers of microvessels with a rounder shape and smaller caliber than chronic phase. A positive correlation emerged between marrow fibrosis and MVD as well as between white blood cell counts and rounder vessel sections. No relationship existed between microvascular parameters and Hasford or Sokal prognostic scores. In univariate analysis, overall and progression-free survival were adversely affected by MVD, size-related parameters, increased platelet count, age and spleen size. Multivariate analysis indicated that microvessel area was related to progression-free survival, whereas both MVD and area were significant prognosticators of overall survival, even when Hasford or Sokal scores are introduced into the model. Our data suggest that changes in angiogenic parameters may participate in the conversion of normal marrow to CML and ultimately to blastic transformation. More importantly, MVD and microvessel caliber are significant predictors of patient survival and progression.

Introduction

The viability and growth of tumors beyond volumes so small that they can be fed and drained by diffusion entails the sprouting of new vessels from preexisting vasculature.1 A plethora of investigations pursued in recent years have verified the presumption that estimation of the angiogenic capability of tumor cells reflects the likelihood of invasion and metastasis in a variety of malignancies.2,3,4,5,6,7

The intimate liaison in physical and conceptual sense between the bone marrow hematopoietic function and its vascular bed has long been appreciated8 strengthening the idea that leukemic cells might depend upon vascular support just like the cells of most solid tumors.9 The common origin of hematopoietic and vascular endothelial cells during embryonic development10 and the production of hematopoietic growth factors by endothelial cells11,12 are in harmony with this concept. Recently, research into angiogenic aspects of hematologic oncology was enlivened by perceived parallels with solid tumor angiogenesis. Although the clinical significance of this phenomenon in most hematologic malignancies is presently an unresolved issue, there is indisputable evidence indirectly implicating bone marrow angiogenesis in the pathophysiology and course of some leukemias and lymphoproliferative disorders.13,14,15,16,17,18 Moreover, intracellular and/or serum levels of vascular endothelial growth factor (VEGF), a potent angiogenic molecule, are reportedly of prognostic significance in acute myeloid leukemia (AML),19 chronic lymphocytic leukemia20 and non-Hodgkin's lymphomas.21 The degree of angiogenesis in chronic myeloid leukemia (CML) has been dealt with in two reports,16,22 in which an increased microvessel density (MVD) has been documented in the bone marrow of a small number of CML patients. Most recently, Tie-1, a molecule involved in the regulation of angiogenesis, has been reported to be an independent predictor of survival in early chronic phase CML.23 The intriguing identification of bcr-abl construct in endothelial cells in CML24 has kindled interest in the area, implying that endothelial cells may be an integrated part in the process of leukemogenesis.

To our knowledge, this study is the first in the literature to evaluate multiple morphometric microvascular characteristics – besides MVD – in CML patients, in relation to clinicopathologic parameters and survival. Such information might prove valuable in clarifying certain aspects in the evolution of the disease and would broaden our understanding of the clinical significance of angiogenesis in these patients.

Material and methods

Patients

Bone marrow paraffin-embedded biopsies obtained at diagnosis from 52 patients with CML in chronic phase, diagnosed at Laikon General Hospital during the years 1984 to 1997, were studied. Control marrows from 20 subjects with no evidence of marrow disease, performed for diagnostic purposes as part of staging procedure for Hodgkin's disease (13 cases), non-Hodgkin's lymphomas (five cases) and solid tumors (two cases), were also evaluated. In addition, we examined trephines from 15 cases of CML in blastic phase, which were age and sex matched with chronic phase CML cases. The same applied to control cases. Diagnosis was established according to the principles laid down in the latest WHO classification25 and was confirmed by three hematologists independently. There were 28 males and 24 females with a median age of 50.5 years (range 15 to 84).

Enlargement of the spleen was found in 40 patients. The clinical and laboratory characteristics of our patients are summarized in Table 1. For the purposes of statistical analysis in this study, hemoglobin levels and platelet counts were categorized as increased, normal or decreased. Patients were also allocated into one of the three prognostic categories defined by Hasford and Sokal scores.26,27 Cytogenetic data were available in all cases and molecular analysis by reverse transcriptase polymerase chain reaction confirmed the presence of bcr-abl transcript. Additional cytogenetic abnormalities were identified in six patients. All patients at the time of diagnosis had less than 10% circulating blasts. Forty-eight patients were treated only with interferon-alpha-2b (INTRON) 5 × 106 U/m2 three times a week until hematologic response. Maintenance treatment was adjusted as the maximal tolerated dose. Hydroxyurea was given to the remaining four patients because of interferon-alpha intolerance (two patients) or was added to interferon-alpha for rapid reduction of the white cell number in cases with hyperleukocytosis at diagnosis (two patients). Patients in myeloid blast crisis received induction treatment with a combination of Idarubicin (for 3 days) and cytosine arabinoside (for 5 days). In two cases with lymphoid blast crisis vincristine and prednisone were added to the above combination. Response was maintained with oral treatment combining 6-mercaptopurine daily and methotrexate weekly. Hematologic response was defined as 50% reduction in the white cell count from baseline, maintained for at least 2 weeks. A complete hematologic response was defined as a reduction of white cell count to less than 10 × 109/l and in the platelet count to less than 450 × 109/l maintained for at least 4 weeks. Cytogenetic responses, based on the analysis of at least 20 metaphases, were categorized as complete (no Ph-positive metaphases – two patients), major (1–34% Ph-positive metaphases – six patients) or minor (35–65% Ph-positive metaphases – two patients). The 42 remaining patients showed no cytogenetic response (96–100% Ph-positive metaphases). Patients were followed-up for a median period of 53.0 months (range 10–166). During this period 22 patients (42.3%) underwent blastic transformation within 49.0 months (range 28.5–62.3) whereas the 30 remaining patients had not progressed to blast crisis after 50.5 (range 26.8–85.5) months. By the time that this study was undertaken, 23 patients (44.2%) had died from disease-related causes, whereas the remaining 29 patients had survived for 60.0 months (range 33.5–93.0).

Table 1 Clinicopathologic and laboratory characteristics of 52 patients with chronic phase CML at the time of diagnosis

Selection of CD34 as the endothelial marker

During the initial phase of the present investigation, we compared the staining performances of antibodies to the endothelial antigens CD34 and factor VIII in 13 randomly selected cases of CML in chronic phase and in all blastic phase cases. The values for the various angiogenic parameters obtained with both antibodies correlated strongly with each other, the correlation coefficient ranging from 0.9813 to 0.9956. This was apparent in the chronic phase as well as in the blastic phase cases (Figure 1). However, staining for factor VIII was inferior to that obtained with CD34, especially with regard to chronic phase cases, because of increased background staining, strong staining of megakaryocytes and fainter labeling of endothelial cells. On the other hand, although the antibody to CD34 also stained myeloid blasts, their number was sufficiently small in chronic phase CML (less than 5%) and their morphology was quite distinctive so as not to interfere with our analysis. Even so, great care was taken not to overcount immature myeloid cells as miniature vascular structures in blastic phase. CD34 has also been judged by other authors to be a better marker for assessing bone marrow angiogenesis in myelofibrosis,28 myelodysplastic syndromes29 and acute leukemias.18

Figure 1
figure1

Graph illustrating the correlation of MVD (a) and area (b) values obtained with CD34 and FVIII immunohistochemical staining in 13 cases of chronic phase CML and 15 cases of CML in blast crisis. The correlation coefficient is 0.945 for MVD in chronic phase and 0.979 for MVD in blastic crisis. The respective values for area are 0.997 and 0.992.

Processing of bone marrow specimens and immunohistochemical staining

Bone marrow specimens were fixed in buffered formalin, decalcified with EDTA/HCl and embedded in paraffin. Hematoxylin and eosin and reticulin stained slides were reviewed to assess in a uniform fashion the presence of reticulin fibrosis (semiquantitatively graded as absent, focal or extensive) as well as the number of megakaryocytes (increased or not increased). Serial sections (4 μm thick) of each sample were also processed for immunohistochemical identification of microvascular endothelial cells with a mouse monoclonal antibody against CD34 (Clone HPCA-1; Becton Dickinson, San Jose, CA, USA). The antibody was applied at a dilution of 1:80 for 1 h. Before staining, slides were incubated four times for 5 min in citrate buffer pH 6.0, at 750 W, in a microwave oven.30 Application of the primary antibody was followed by the standard three-step streptavidin peroxidase technique.31 All specimens were treated using identical procedures. Negative controls (ie sections in which the primary antibody was substituted with non-immune mouse serum) were also stained in each run.

Microvessel counting and image analysis

Immunohistochemically stained slides for CD34 were examined using a Zeiss Axiolab microscope (Carl Zeiss Jena, Jena, Germany) with a mechanical stage. According to Padrö et al,17 to identify the area showing the most intense vascularization (the ‘hot spot’), the entire bone marrow section was simultaneously scanned, field per field, by two experienced investigators at ×100 magnification. The selection of hot spot was based on observing restricted areas within a ×100 field with an impression of a higher density of CD34-positive single endothelial cells or clusters of endothelial cells. The magnification was then changed to ×250 and the investigators were allowed to reposition the slide until the highest number of microvessels was included within the ×250 field. This area was defined as the ‘hot spot’.32 These vascular ‘hot spots’ were suitable for analysis provided they were within cellular areas of the marrow (bone labellae, fat and connective tissue were excluded). The hot spot was photographed and printed on high quality photographic paper. Photographs were scanned by a Hewlett Packard ScanJet 3400C (Hewlett Packard Company, Palo Alto, CA, USA) to become digital and were stored as BMP files (1480 × 1070 pixels, 16.7 million colors (24-bit)). The reason for adopting this technique instead of directly digitalizing the hot spot was the higher resolution obtained in this way. The whole system was calibrated with the use of a micrometer slide. In cases in which the most vascularized area was not obvious by visual impression, two to four optical fields with the highest density of capillaries and small vessels were photographed, although only the one with the highest MVD was finally taken into account for further evaluation. Conforming to published consensus,17,33 any brown stained endothelial cell or cluster, with or without a lumen, that was clearly separated from adjacent microvessels and other bone marrow cells, was considered as a single countable microvessel. The presence of red blood cells or fibrin without any detectable endothelial cells was not sufficient to define a microvessel. Vessels with muscular walls were not counted; however, there was no restriction regarding the size of the countable vessels, so as not to underestimate longitudinal sections or bifurcations of microvessels.6 In addition to endothelial cells, scattered myeloid blasts were also strongly positive for CD34. These cells were easily recognized by their morphology and served as internal controls to verify the adequacy of staining.

Quantification on digital images was performed by one investigator by using the SigmaScan Software (Jandel Scientific, Erkrath, Germany) on a Pentium III personal computer. For each countable microvessel the outline was identified and traced (Figure 2) and the following morphometric parameters were estimated: major axis length (the distance between the two points along the vessel periphery that are furthest apart), minor axis length (the longest axis perpendicular to major axis formed by two points along the vessel periphery), perimeter, area, compactness (Perimeter2/Area), shape factor (4π*Area/Perimeter2) and Feret diameter √\(\overline{(4*Area)/π}\).34

Figure 2
figure2

Immunohistochemical staining of bone marrow endothelial cells for CD34 in a case of chronic phase CML (a, b) and a case of CML in blastic phase (c, d). The outline of each vessel is traced; the red layer represents the section area of each vessel. Green arrows point towards branching vessels (×250).

The variables entered into the statistical analysis were the mean values of the above seven morphologic indices, the total count of microvessels per optical field (MVD), as well as the total area occupied by them (total vascular area, TVA) and the number of vascular ramifications per 100 vessel sections, as an expression of the complexity of the microvascular network. These points, at which a new branch emerged from a parent vessel, were recognized on the histological sections as Y-shaped, T-shaped or E-shaped.6 The whole procedure took place without any knowledge of the patients’ data.

Statistical analysis

To confirm that the counting technique above was representative of bone marrow angiogenesis, we performed additional analysis in 10 randomly selected CML cases. In these cases, we counted the microvessels in three random fields ×250. Intrarater and interrater reliability were assessed by Pearson's correlation coefficient in the same 10 cases. Concerning the interobserver and intraobserver variability the correlation coefficient was very satisfactory for all the examined parameters ranging from 0.873 to 0.996. Reproducibility of measurements of the same optical field or of the same individual vessel was examined by the coefficient of variation between repeated measurements (values ranged from 0.8% to 4.1%). Furthermore, the angiogenic parameters in the hot spot strongly correlated with the mean values in the three examined fields, the correlation coefficient fluctuating between 0.885 and 0.995. The normality of distributions was tested with the Kolmogorov–Smirnov test. Differences in the values of microvascular parameters between groups were examined by the Mann–Whitney U test. For the analysis of variance we used Kruskal–Wallis test. Spearman's rank correlation coefficient was calculated to determine associations between numerical variables.

Survival analysis was carried out in the 52 patients with chronic phase CML. The effect of various parameters (age, gender, the degree of bone marrow fibrosis, the presence of increased megakaryocytes, clinical and laboratory findings, Hasford score, Sokal score and the microvascular parameters) on clinical outcome (evolution to blastic phase or death) was assessed by plotting survival curves according to Kaplan–Meier method and groups were compared using the log-rank test. White blood cell count was categorized on the basis of the median value. Age was trichotomized as follows: <50, 50–70, >70. Microvascular parameters were categorized on the basis of the median value rounded at its nearest 10%. Multivariate analysis was performed using the stepwise Cox's regression model to evaluate the predictive power of each variable independently of the others. Separate multivariate analyses were also performed to test the prognostic significance of microvascular parameters against Hasford or Sokal scores and the remaining clinicopathologic parameters. Statistical analysis was performed using the SPSS for Windows Software (SPSS, Chicago, IL, USA) on an IBM-compatible computer. Differences were considered statistically significant when P value (two-sided) was ≤0.05.

Results

Comparison of morphometric variables among CML, control and blastic phase groups

As shown in Table 2, the values of MVD and size-related parameters (area, minor and major axis length, TVA and Feret diameter) were higher in chronic phase CML than in controls. Within the control group, microvascular parameters did not correlate with cellularity (P > 0.10). Controls also showed a lesser degree of vessel branching compared to CML. These differences translated to a large number of large caliber, often arborizing vessels in CML bone marrow, as opposed to few simple and straight microvessels in controls (Figure 3).

Table 2 Evaluated vascular morphometric parameters (median, range) in control, chronic phase CML (CML) and blast crisis bone marrows
Figure 3
figure3

Microvessel density (MVD) and microvessel area in controls, chronic phase CML and blastic phase CML. MVD increases from controls through chronic phase to blastic phase. Area increases from controls to chronic phase but decreases in blastic phase.

Bone marrows in blast crisis were characterized by increased MVD but lower values of size-related parameters compared to chronic phase CML (Figure 2, Table 2). These differences were observed not only on CD34 stained slides but also on those stained for factor VIII (P < 0.001 for MVD and P = 0.046 for area). Furthermore, shape factor values were higher and compactness values were lower in blastic phase indicating the prevalence of rounder vessel sections. Finally, TVA and branching values were increased in blastic phase, although the difference failed to reach statistical significance (Table 2). There were no significant differences in the distribution of microvascular parameters among the three Hasford and Sokal prognostic subgroups.

Association of morphometric variables with clinicopathologic characteristics in patients with chronic phase CML

None of the morphometric variables was related to patients’ age, gender, spleen size, karyotype, cytogenetic response, hemoglobin level and platelet, blast, eosinophil or basophil counts (for all P > 0.10). White blood cell count correlated positively with shape factor (r = 0.312, P = 0.024) and negatively so with compactness (r = −0.320, P = 0.021). Thus, patients with higher white blood cell counts tended to display rounder vessel sections in their bone marrows than those with lower white blood cell counts. Extensive marrow fibrosis was accompanied by higher MVD (P = 0.023, Kruskal–Wallis ANOVA).

Correlations among the various morphometric variables in CML

It is to be noted that a negative (although not strong) correlation existed between vessel number and vessel size parameters (ie area, major axis length, minor axis length, perimeter, Feret diameter), evidenced by the presence of cases with relatively many but delicate microvessels on one end of the spectrum and cases with few but larger caliber vessels on the other. Also, branching values positively correlated with MVD and TVA.

Prognostic relevance of morphometric variables in CML

To investigate whether bone marrow angiogenesis at diagnosis may predict progression to blastic phase, we examined the impact of the microvascular parameters on the incidence as well as on time to progression. Patients who progressed were characterized by higher values of area (P = 0.036), perimeter (P = 0.047) and TVA (P = 0.010) than those who did not (Mann–Whitney U test). When all parameters were entered into logistic regression analysis, increased values of Feret diameter (P = 0.035, odd ratio 1.081) emerged as the independent predictors of higher likelihood of progression to blastic phase. In univariate analysis, patient's age (>70 years), splenomegaly (>3 cm), MVD, minor axis length, area, TVA and platelet count were the only parameters adversely affecting time to progression whereas marrow fibrosis attained a borderline significance in this regard (Table 3, Figure 4). Multivariate analysis indicated that the duration of chronic phase was independently related to platelet count, patient's age, karyotype and area. As evidenced from the values of hazard ratio in Cox's model, the presence of additional cytogenetic abnormality was an adverse prognostic indicator. When Hasford or Sokal score were included in Cox's model, only area remained significant (Table 4).

Table 3 Univariate analysis of overall and progression free survival of patients with chronic phase CML (log-rank test)
Figure 4
figure4

Kaplan–Meier curves. Progression-free survival in relation to (a) microvessel density (MVD) and (b) area.

Table 4 Cox's proportional hazard estimation of overall and progression-free survival. Models B and C are derived after Hasford (B) and Sokal (C) scores are incorporated into multivariate analysis

With regard to survival, univariate analysis identified seven significant parameters, namely MVD, minor axis length, area, TVA, platelet count, patient's age and spleen size (Table 3). Patients older than 70 years at diagnosis and with splenomegaly, thrombocytosis and increased values of the above angiogenic parameters tended to fare worse (Figure 5). Cox's model selected only MVD, area and spleen size. MVD and area retained their significance in the presence of Hasford or Sokal scores, though only the latter appeared in Cox's model (Table 4).

Figure 5
figure5

Kaplan–Meier curves. Overall survival in relation to (a) microvessel density (MVD) and (b) area.

Discussion

The present investigation clearly confirms the elevated bone marrow vessel counts in CML compared to normal controls.16,22,29,35 In this regard, CML appears to join the chorus of other hematologic malignancies with a rich vascular network.13,14,15,16,17,18,22,29,36 In earlier studies16,29 MVD was higher in CML than in AML. Our observation of a higher degree of vascular proliferation in blastic phase CML compared to chronic phase indicates that the former may have a different angiogenic basis from primary AML.

We further noted a morphologic variability of the vascular network among controls, CML and blastic phase of CML. Size-related parameters presented a peak in chronic phase but round vessel sections were most frequent in blastic phase. These findings are in agreement with the notion that angiogenesis is primarily initiated in chronic phase CML as an outburst in the formation of wide sinusoidal spaces.37 Entrance into blastic phase is heralded by an increase in vessel caliber, probably as a consequence of intussusceptive mechanism of neovascular growth,38 while the rate of new vessel formation is sustained, endowing bone marrow with excess blood flow to satisfy the increased metabolic demands of rapidly growing blast cells. The high complexity of the vascular network in blastic phase obstructs blood flow and causes intraluminal pressure to rise, which explains the prevalence of round vessel sections at this phase of CML. These qualitative and quantitative changes in the vascular pattern in association with disease progression are further illustrated in the correlation of elevated white blood cell counts with the presence of round vessel sections and of marrow fibrosis with increased MVD, paralleling recent observations in myelodysplastic syndromes.36

It could be argued that the elevated vessel counts recorded in CML may be attributable to reactivation of preexisting functionally dormant sinusoid endothelium for endothelial markers.39 However, the distinctly abnormal architecture of the microvessels with increased tortuosity speaks in favor of a truly neoangiogenic phenomenon elicited in response to increased VEGF production from elevated leukocyte numbers in the peripheral blood as well as in the bone marrow.34 Increased marrow cellularity is unlikely to be a major factor accounting for the increased vascularity, since we did not find any significant differences between reactive hypercellular bone marrows and their normocellular counterparts. In this context, it is worth noting that serum VEGF levels are much higher in CML than in any other type of acute or chronic leukemia16 and that VEGF expression is particularly prominent in megakaryocytes in CML.29 Taking into account that increased numbers of megakaryocytes are associated with marrow fibrosis, it is not surprising that severely fibrotic bone marrows in CML display higher microvessel counts. On the other hand, myeloid blasts and immature myeloid cells are known to generate VEGF by either a paracrine or an autocrine mechanism.12,40 Hence, it might be speculated that myeloid precursors and megakaryocytes are among the principal providers of VEGF in the bone marrow environment in CML.

The recent demonstration of bcr-abl construct in CML bone marrow-derived endothelial cells24 seems to shed additional light upon the mechanisms underlying angiogenesis in this disease. This observation shows that endothelial cells in CML are clonal and therefore proliferate faster and have a prolonged life span owing to suppression of apoptosis. Furthermore, new endothelial cells are produced from clonal bone marrow-derived cells, although the origin of these endothelial cells (ie a bipotent hemangioblast, a hematopoietic stem cell or a multipotent stem cell) remains elusive.41 What emerges from the aforementioned data is that bone marrow angiogenesis in CML is a complex, not fully elucidated as yet, phenomenon in which excessive production of angiogenic factors and mechanisms of leukemogenesis cannot be disentangled. The clonality of endothelial cells and hematopoietic cells producing VEGF may also explain why polycythemia vera and essential thrombocythemia display lower microvessel counts despite the increased numbers of megakaryocytes in the bone marrow.12,18

The prognostic relevance of neovascularization in most hematologic malignancies remains contentious, having so far been appreciated in AML,17 multiple myeloma,42 myelofibrosis28 and myelodysplastic syndromes.36 To the best of our knowledge, the current study is the first to deal with the impact of angiogenesis on the progression and clinical outcome of patients with CML, against several clinicopathologic parameters. Our data provide evidence for the prognostic value of the degree of angiogenesis and the caliber of microvessels and advocate that they could serve as additional variables in clinical prognostic models. The fact that these parameters retain their prognostic value in the presence of Hasford and Sokal scores and the absence of any significant differences in microvascular parameters among the three risk groups underline the independence of angiogenesis as a prognostic factor for CML patients. The adverse effect of increased vessel caliber and MVD may reflect a sequence of events analogous to that of solid tumor angiogenesis: aggressive rapidly growing neoplastic populations soon lead their environment to hypoxia, which, being an important stimulus for the production of VEGF (through overexpression of hypoxia-inducible factor 1α), will result in excessive angiogenesis in terms of both vessel size and number to restore tissue oxygen concentration.1,43 Notwithstanding, neoplastic cells adjacent to neovessels may still be hypoxic, because of an architecturally deficient microcirculation.43 Of course, it is possible that additional hypoxia-independent pathways may also contribute to bone marrow angiogenesis in CML.

Optimal therapy for CML in chronic phase remains controversial. Available treatments have so far been of limited effectiveness and are associated with a considerable risk of side-effects. Clinical trials indicate that anti-angiogenic therapy is effective in certain hematologic neoplastic diseases such as multiple myeloma,44 myelofibrosis45 and probably also in MDS,46 inhibiting neoplastic growth by a mechanism of ‘tumor stabilization’.47 In view of our observation that the degree of angiogenesis is independently linked to clinical outcome, anti-angiogenic therapy warrants a trial as an adjunct to the currently available treatment options in CML patients. A novel promising line of approach to CML treatment is the inactivation of bcr-abl gene product thereby reversing the leukemic phenotype.48 If in fact endothelial cells take part in the leukemic proliferation,24 it might be speculated that the efficacy of this treatment may partly relate to the inhibition of anarchic endothelial cell expansion.

In summary, our data support the hypothesis that the generation of new vessels may be essential to the multistep process of conversion of normal bone marrow to CML and ultimately to blast crisis. More importantly, our study demonstrates for the first time that the prognostic significance of angiogenesis in CML is better assessed by the size of microvessels, whereas MVD becomes influential only with regard to survival. While of obvious biologic interest, the clinical worth of counting microvessels in the bone marrow from CML patients should await validation in controlled prospective investigations with a larger number of patients. The same applies to the potential value of anti-angiogenic therapy in delaying or even preventing disease progression in these patients.

References

  1. 1

    Folkman, J. Clinical applications of research on angiogenesis. N Engl J Med, (1995). 333, 1757–1763.

  2. 2

    Schnitt, SJ The real value of angiogenesis research in breast cancer. Adv Anat Path, (1997). 4, 23–27.

  3. 3

    Srivastava, A, Laidler, P, Davies, RP, Horgan, K & Hughes, LE The prognostic significance of tumor vascularity in intermediate-thickness (0.76–4.0 mm thick) skin melanoma: a quantitative histologic study. Am J Pathol, (1988). 133, 419–423.

  4. 4

    Korkolopoulou, P, Konstantinidou, AE, Kavantzas, N, Patsouris, E, Pavlopoulos, PM, Christodoulou, P, Thomas-Tsagli, E & Davaris, P Morphometric microvascular characteristics predict prognosis in superficial and invasive bladder cancer. Virch Arch, (2001). 438, 603–611.

  5. 5

    Brawer, MK, Deering, RE, Brown, M, Preston, SD & Bigler, SA Predictors of pathologic stage in prostatic carcinoma. The role of neovascularity. Cancer, (1994). 73, 678–687.

  6. 6

    Pavlopoulos, PM, Konstantinidou, AE, Agapitos, E, Kavantzas, N, Nikolopoulou, P & Davaris, P A morphometric study of neovascularization in colorectal carcinoma. Cancer, (1998). 83, 2067–2075.

  7. 7

    Yamazaki, K, Abe, S, Tagekawa, H, Sukoh, N, Watanabe, N, Ogura, S, Nakajima, I, Isobe, H, Inoue, K & Kawakami, Y Tumor angiogenesis in human lung adenocarcinoma. Cancer, (1994). 74, 2245–2250.

  8. 8

    Bruyn, PPH, Breen, PC & Thomas, TB The microcirculation of the bone marrow. Anat Rec, (1970). 168, 55–68.

  9. 9

    Mangi, MH Angiogenesis and angiogenic mediators in haematological malignancies. Br J Haematol, (2000). 111, 43–51.

  10. 10

    Choi, K, Kennedy, M, Kazarov, A, Papadimitriou, JC & Keller, G A common precursor for hematopoietic and endothelial cells. Development, (1998). 125, 725–732.

  11. 11

    Yamaguchi, H, Ishii, E, Saito, S, Tashiro, K, Fujita, I, Yoshidani, S, Ohtubo, M, Akazawa, K & Miyazaki, S Umbilical vein endothelial cells are an important source of c-kit and stem cell factor which regulate the proliferation of haemopoietic progenitor cells. Br J Haematol, (1996). 94, 606–611.

  12. 12

    Fiedler, W, Graeven, U, Ergn, S, Verago, S, Kilie, N, Stockschlader, M & Hossfeld, DK Vascular endothelial growth factor, a possible paracrine growth factor, in human acute myeloid leukemia. Blood, (1997). 89, 1870–1875.

  13. 13

    Vacca, A, Ribatti, D, Roncali, L, Ranieri, G, Serio, G, Silvestris, F & Dammacco, F Bone marrow angiogenesis and progression in multiple myeloma. Br J Haematol, (1994). 87, 503–508.

  14. 14

    Ribatti, D, Vacca, A, Nico, B, Fanelli, M, Roncali, L & Dammacco, F Angiogenesis spectrum in the stroma of B-cell non-Hodgkin's lymphomas. An immunohistochemical and structural study. Eur J Haematol, (1996). 56, 45–53.

  15. 15

    Perez-Atayde, AR, Sallan, SE, Tedrow, V, Connors, S, Allred, E & Folkman, J Spectrum of tumor angiogenesis in the bone marrow of children with acute lymphoblastic leukemia. Am J Pathol, (1997). 150, 815–821.

  16. 16

    Aguayo, A, Kantarjian, H, Manshouri, T, Gidel, C, Estey, E, Thomas, D, Koller, C, Estror, Z, O’Brien, S, Keating, M, Freireich, E & Albitar, M Angiogenesis in acute and chronic leukemias and myelodysplastic syndromes. Blood, (2000). 96, 2240–2245.

  17. 17

    Padrö, T, Ruiz, S, Bieker, R, Burger, H, Steins, M, Kienast, J, Buchner, T, Berdel, WE & Mesters, RM Increased angiogenesis in the bone marrow of patients with acute myeloid leukemia. Blood, (2000). 95, 2637–2644.

  18. 18

    de Bont, ESJM, Rosati, S, Jacobs, S, Kamps, WA & Vellenga, E Increased bone marrow vascularization in patients with acute myeloid leukemia: a possible role for vascular endothelial growth factor. Br J Haematol, (2001). 113, 296–304.

  19. 19

    Aguayo, A, Estey, E, Kantarjian, H, Manshouri, T, Gidel, C, Keating, M, Giles, F, Estror, Z, Barlogie, B & Albitar, M Cellular vascular endothelial growth factor is a predictor of outcome in patients with acute myeloid leukemia. Blood, (1999). 94, 3717–3721.

  20. 20

    Aguayo, A, O’Brien, S, Keating, M, Manshouri, T, Gidel, C, Barlogie, B, Beran, M, Keller, C, Kantarjian, H & Albitar, M Clinical relevance of intracellular vascular endothelial growth factor levels in B-cell chronic lymphocytic leukemia. Blood, (2000). 96, 768–770.

  21. 21

    Salven, P, Orpana, A, Teerenhovi, L & Joensuu, H Simultaneous elevation in the serum concentrations of the angiogenic growth factors VEGF and βFGF is an independent predictor of poor prognosis in non-Hodgkin's lymphoma: a single-institution study of 200 patients. Blood, (2000). 96, 3712–3718.

  22. 22

    Lundberg, LG, Lerner, R, Sundelin, P, Rogers, R, Folkman, J & Palmblad, J Bone marrow in polycythemia vera, chronic myelocytic leukemia and myelofibrosis has an increased vascularity. Am J Pathol, (2000). 157, 15–19.

  23. 23

    Verstovsek, S, Kantarjian, H, Manshouri, T, O’Brien, S, Faderl, S, Talpaz, M, Cortes, J & Albitar, M Prognostic significance of Tie-1 protein expression in patients with early chronic phase chronic myeloid leukemia. Cancer, (2002). 94, 1517–1521.

  24. 24

    Gunsilius, E, Duba, HC, Petzer, AL, Kahler, CM, Grunewald, K, Stockhammer, G, Gabl, C, Dirnhofer, S, Clausen, J & Gastl, G Evidence from a leukemia model for maintenance of vascular endothelium by bone-marrow-derived endothelial cells. Lancet, (2000). 355, 1688–1691.

  25. 25

    Harris, NL, Jaffe, ES, Diebold, J, Flandrin, G, Muller-Hermenlink, HK, Vardiman, J, Lister, TA & Bloomfield, CD The World Health Organization classification of hematological malignancies report of the Clinical Advisory Committee Meeting, Airlie House, Virginia, November 1997. Histopathology, (2000). 36, 69–87.

  26. 26

    Hasford, J, Pfirrmann, M, Hehlmann, R, Allan, NC, Baccarani, M, Johanna, C, Nelemans, K, Alimena, G, Steegmann, JL & Ansari, H A new prognostic score for survival of patients with chronic myeloid leukemia treated with interferon alfa. J Natl Cancer Inst, (1998). 90, 850–858.

  27. 27

    Sokal, J, Cox, E, Baccarani, M, Tura, S, Gomez, GA, Robertson, JE, Tso, CY, Braun, TJ, Clarkson, BD, Cervantes, F & Rozman, C the Italian Cooperative CML Study Group Prognostic discrimination in ‘good risk’ chronic granulocytic leukemia. Blood, (1984). 63, 1352–1357.

  28. 28

    Mesa, RA, Hanson, CA, Rajkumar, V, Schroeder, G & Tefferi, A Evaluation and clinical correlations of bone marrow angiogenesis in myelofibrosis with myeloid metaplasia. Blood, (2000). 96, 3374–3380.

  29. 29

    Pruneri, G, Bertolini, F, Soligo, D, Carboni, N, Cortelezzi, A, Ferucci, PF, Buffa, R, Lambertenghi-Deliliers, G & Pezzela, F Angiogenesis in myelodysplastic syndromes. Br J Cancer, (1999). 81, 1398–1401.

  30. 30

    Shi, SR, Key, ME & Kalra, K Antigen retrieval in formalin-fixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections. J Histochem Cytochem, (1991). 39, 741–748.

  31. 31

    Hsu, SM, Raine, L & Fanger, H Use of avidin–biotin–peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem, (1981). 29, 577–580.

  32. 32

    Vermeulen, PB, Gasparini, G, Fox, SB, Toi, M, Martin, L, McCulloch, P, Pezzela, F, Viale, G, Weidner, N, Harris, AL & Dirix, LY Quantification of angiogenesis in solid human tumours: an international consensus on the methodology and criteria of evaluation. Eur J Cancer, (1996). 32, 2474–2484.

  33. 33

    Weidner, N, Folkman, J, Pozza, F & Bevilacqua, P Tumor angiogenesis: a new significant and independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst, (1992). 84, 1875–1887.

  34. 34

    Fox, E & Ulrich, C Sigma Scan Image Measurement Software, User's Electronic Manual. (1995). Sigma Scan, Jandel Scientific: Erkrath, Germany

  35. 35

    Branemark, P-I Bone marrow microvascular structure and function. Adv Microcirc, (1968). 1, 1–65.

  36. 36

    Korkolopoulou, P, Apostolidou, E, Pavlopoulos, PM, Kavantzas, N, Vyniou, N, Thymara, I, Terpos, E, Patsouris, E, Yataganas, X & Davaris, P Prognostic evaluation of the microvascular network in myelodysplastic syndromes. Leukemia, (2001). 15, 1369–1376.

  37. 37

    Benjamin, LE & Keshet, E Conditional switching of vascular endothelial growth factor (VEGF) expression in tumors: induction of endothelial cell shedding and regression of hemangioblastoma-like vessels by VEGF withdrawal. Proc Natl Acad Sci USA, (1997). 94, 8761–8766.

  38. 38

    Patan, S, Munn, LL & Jain, PK Intussusceptive microvascular growth in a human colon adenocarcinoma xenograft: a novel mechanism of tumor angiogenesis. Microvasc Res, (1996). 51, 260–272.

  39. 39

    Garcia-Monzon, C, Sanchez-Madrid, F, Garcia-Buey, L, Garcia-Arroyo, A, Garcia-Sanchez, A & Moreno-Otero, R Vascular adhesion molecule expression in viral chronic hepatitis: evidence of neoangiogenesis in portal tracts. Gastroenterology, (1995). 108, 231–241.

  40. 40

    Bellamy, WT, Richter, L, Sirjani, D, Roxas, C, Glinsmann-Gibson, B, Frutiger, Y, Grogan, TM & List, AF VEGF is an autocrine promoter of abnormal localized immature myeloid precursors and leukemic progenitor formation in myelodysplastic syndromes. Blood, (2001). 97, 1427–1434.

  41. 41

    Green, AR Hemangioblast origin of chronic myeloid leukaemia?. Lancet, (2000). 355, 1659–1660.

  42. 42

    Rajkumar, SV, Leong, T, Roche, PC, Fonseca, R, Dispenzieri, A, Lacy, M, Lust, JA, Witzig, TE, Kyle, RA, Gertz, MA & Greipp, PR Prognostic value of bone marrow angiogenesis in multiple myeloma. Clin Cancer Res, (2000). 6, 3111–3116.

  43. 43

    Zhong, H, De Marzo, AM, Laughner, E, Lim, M, Hilton, DA, Zagzag, D, Buechler, P, Isaacs, WB, Scmenza, GL & Simons, JW Overexression of hypoxia-inducible factor 1α in common human cancers and their metastases. Cancer Res, (1999). 59, 5830–5835.

  44. 44

    Rajkumar, SV, Dispenzieri, A, Fonseca, R, Lacy, MR, Geyer, S, Lust, JA, Kyle, RA, Greipp, PR, Gertz, MA & Witzig, TE Thalidomide for previously untreated indolent or smoldering multiple myeloma. Leukemia, (2001). 15, 1274–1276.

  45. 45

    Barosi, G, Grossi, A, Comotti, B, Musto, P, Gamba, G & Marchetti, M Safety and efficacy of thalidomide in patients with myelofibrosis with myeloid metaplasia. Br J Haematol, (2001). 114, 78–83.

  46. 46

    Raza, A, Meyer, P, Dutt, D, Zorat, F, Lisak, L, Nascimben, F, du Randt, M, Kaspar, C, Goldberg, C, Loew, J, Dar, S, Gezer, S, Venugopal, P & Zeldis, J Thalidomide produces transfusion independence in long-standing refractory anemias of patients with myelodysplastic syndromes. Blood, (2001). 98, 958–965.

  47. 47

    Harris, AL Antiangiogenesis for cancer therapy. Lancet, (1997). 349, 13–15.

  48. 48

    Druker, BJ & Lydon, NB Lessons learned from the development of an abl tyrosine kinase inhibitor for chronic myelogenous leukemia. J Clin Invest, (2000). 105, 3–7.

Download references

Acknowledgements

We thank Miss Evie Delicha for the statistical analysis of results.

Author information

Correspondence to P Korkolopoulou.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • chronic myeloid leukemia
  • microvessel morphometry
  • angiogenesis

Further reading