Erythropoietic Reconstitution, Macrophages and Reticulin Fibrosis

Home , Erythropoiesis

Leukemia (2000) 14, 1378–1385  2000 Macmillan Publishers Ltd All rights reserved 0887-6924/00 $15.00 www.nature.com/leu Erythropoietic reconstitution, macrophages and reticulin ﬁbrosis in bone marrow specimens of CML patients following allogeneic transplantation J Thiele1, HM Kvasnicka1, DW Beelen2, B Pilgram1, A Rose1, L-D Leder3 and UW Schaefer2

Institutes of Pathology, Universities of 1Cologne, Cologne and 3Essen, Essen; and 2Department of Bone Marrow Transplantation, University of Essen, Essen, Germany

A clinicopathological study was conducted on 351 bone mar- ated or blastic phases,7 some data have been accumulated row trephine biopsies derived from 124 patients with chronic that histopathology may offer supportive evidence. Follow-up myeloid leukemia (CML) at standardized endpoints before and after allogeneic bone marrow transplantation (BMT). The pur- studies on bone marrow histopathology together with clinical pose was to investigate quantitative changes of the nucleated observations were in keeping with the finding that reduction erythroid precursor cell population and other associated fea- in erythropoiesis8 and extent of myelofibrosis9–12 may serve as tures such as resident bone marrow macrophages and myelo- valid parameters for the assessment of disease progression. fibrosis and to elucidate their relevance on engraftment para- Both variables have been shown to be significantly correlated meters. Monoclonal antibodies were applied for the and to exert a predictive value for survival.8,9,11,12 Contrasting identification of erythroid precursors and the labeling of mature the overall progress which has been achieved during the last macrophages; argyrophilic (reticulin-collagen) fibers were 2,13–18 demonstrated by a silver impregnation technique. Following decade concerning BMT in CML, hardly any information morphometric analysis of the pregraft bone marrow specimens exists about bone marrow morphology following marrow- statistical evaluation was in line with an adverse correlation ablative pretransplant regimens and the regeneration of hema- between early to moderate reticulin fibrosis and amount of topoiesis. In particular, this lack of knowledge applies to erythropoiesis. Moreover, a significant relationship was calcu- erythroid precursors in the pre- and post-transplant period and lable between numbers of erythroid precursors and CD68؉ macrophages. After myelo-ablative therapy and BMT a pro- the problem whether a greatly reduced erythropoiesis before nounced decrease in cellularity and in the quantity of erythro- BMT exerts a relevant influence on engraftment. Because poiesis was found. Comparable with the pregraft samples, a there is clear-cut evidence for an important functional role of significant association between erythroid precursors and the central resident (mature) macrophages of erythroblastic macrophages could be determined in the regenerating donor islets in iron turnover, hemoglobin synthesis, erythropoietin bone marrow. A pretransplant relevant reduction of the red cell production and phagocytosis of the expelled erythroblastic lineage and a manifest (reticulin) myelofibrosis indicating an 19 advanced stage of disease were accompanied by a significant nuclei, this cell lineage deserves special attention in the con- delay to reach transfusion independence. This result was text of erythropoietic regeneration. For this reason, the pur- further supported by comparable findings in trephine biopsies pose of the present study was focused on histopathology performed in the early post-transplant period (second month characterizing the reconstitution of erythropoiesis including after BMT). Corresponding examinations revealed an enhance- the macrophage population and associated changes of the ment of fiber density and a decrease in erythropoiesis in those fiber content during the critical phase of marrow engraftment patients who did not conform with the usually accepted criteria for successful engraftment. In conclusion, compelling evidence in CML. To elucidate the putative impact of these features on has been produced that a significantly reduced amount of BMT biology, appropriate analysis was carried out by histoch- erythroid precursors, which is usually associated with myelo- emical and morphometric methods on repeatedly performed fibrosis in the pretransplant bone marrow, exerts an impair- bone marrow trephine biopsies at standardized endpoints ment to undisturbed hematopoietic reconstitution. Moreover, a accompanied by corresponding clinical data. close spatial and numerical relationship between the erythroid lineage and resident (mature) macrophages is observable, in particular in the state of regeneration after BMT. Leukemia (2000) 14, 1378–1385. Materials and methods Keywords: erythropoiesis; macrophages; myelofibrosis; bone marrow biopsy; bone marrow transplantation; CML Patients

The series considered for study was based on a retrospective Introduction evaluation of 124 patients presenting with first chronic (stable) phase of Philadelphia chromosome-positive (Ph1ϩ) CML. Stem cell and bone marrow transplantation (BMT) is generally Patients received unmanipulated marrow grafts from HLA- believed to be the only form of therapy with the option to identical family or alternate donors at a single referral trans- cure chronic myelogenous leukemia (CML).1,2 However, plantation center following standard procedures.14 Details among other factors, outcome determined as capacity for regarding, clinical findings and treatment characteristics are unsupported hematopoietic reconstitution at day 30 following summarized in Table 1. As may be easily derived from Table 1 BMT is significantly influenced by progression of disease at homogeneity of our cohort concerning specifications of cer- onset and thus timing.3–5 Regarding staging of chronic phase tain characteristics is established by showing about 80% sib- CML indicating a certain progress4–6 or definition of acceler- ling donors, in more than 70% an identical HLA constellation and the same conditioning regimen in 90% of patients. Endpo- int of hematological reconstitution with normal graft function was assumed if patients achieved transplant-derived absolute Correspondence: J Thiele, Institute of Pathology, University of × 9 Cologne, Joseph-Stelzmannstr. 9, D-50924 Cologne, Germany; neutrophil counts greater than 0.5 10 /l and self-sustaining 9 Fax: +49–0221–4786360 thrombocyte counts greater than 20.0 × 10 /l until day 30 fol- Received 10 March 2000; accepted 21 March 2000 lowing BMT. Delayed engraftment was diagnosed if these Erythropoiesis and BMT in CML J Thiele et al 1379 Table 1 Characteristics of 123 CML patients with allogeneic BMT for 3–4 days in 10% buffered ethylene-diamine tetra-acetic under study acid (EDTA), pH 7.2, paraffin embedding, and employment of several routine staining techniques, involving Giemsa, PAS Median age (years) 35 (periodic acid Schiff reagent), naphthol-AS-D-chloroacetate range (15–55) Gender (M/F) 73/51 esterase, Perls’ reaction for iron and finally a silver impreg- Sokal index-risk groups, n (%) nation method (Gomori’s technique) for the assessment of low 106 (85) argyrophilic (reticulin-collagen) fiber density. For specific intermediate 11 (9) staining of nucleated erythroid precursor cells the monoclonal high 7 (6) antibody Ret40f (anti-glycophorin C) was used20 and for the Median interval from diagnosis to BMT (months) 20 CD68ϩ mature-resident macrophage population PG-M .21 range 5–133 1 Relation – Donors, n (%) Monoclonal antibodies and other reagents were purchased sibling 97 (79) from Dako-Diagnostica (Hamburg, Germany). Details of stain- unrelated 26 (21) ing procedures (APAAP-method) were reported in a previous Median donor age (years) 37 communication.12,21 range 18–66 Gender of donor (F/M) 64/59 Donor/recipient sex, n M → F34 F → M20Morphometry Predominant pretransplant therapya, n (%) hydroxyurea (HU) 89 (72) interferon-␣ (IFN-␣) 62 (50) Following immunostaining and silver impregnation morpho- busulfan (BU) 43 (35) metric analysis was performed by two manual optic plan- HLA constellation, n (%) HLA identical 90 (73) imeters (MOP-A-MO1-Kontron and VIDAS-Zeiss-Kontron, 1 MM 34 (27) Carl Zeiss, Oberkochen, Germany) with a standard program Conditioning regimen, n (%) set (Kontron software, Kontron, Eching, Germany) on large tre- (Cy 60 mg/kg/day ×2) phine biopsies with an artefact-free mean marrow area of dose of TBI × 2 1 (1) 14.5 4.8 mm . Frequencies of erythroid precursors and 8.6 Gy 60 Colbalt 2 7 (6) macrophages were calculated per square millimeter (mm ) 4 × 2.5 Gy LINAC 116 (93) 4 × 2.5 Gy 60 Cobalt total marrow area by regarding the age- or therapy-related adi- GVHD prophylaxis, n(%) pose tissue and also per hematopoiesis (cellularity). After sil- MTX alone 9 (8) ver impregnation procedures density of argyrophilic (reticulin- sMTX + CyA 98 (79) collagen) fibers was determined by the point-intersection CyA 4 (3) method using an ocular grid. Thus the amount of fibers was + CyA BMA031 12 (10) expressed as intersections (i) per area (mm2) of total bone mar- ABO-blood group constellation, n (%) matched 73 (59) row and also per hematopoietic tissue (cellularity). Reference mismatched to cellularity was necessary to focus on the areas relevant for minor 24 (19) hematopoietic reconstitution and further to avoid the major 27 (22) erroneous impression of a significant reduction of these vari- Marrow cell dose ables due to the expansion of adipose tissue or interstitial × 8 graft size ( 10 /kg nucleated cells) median 2.3 edema. This feature was prominent following marrow-ablative (range) (0.5–12.7) therapy and during the post-transplant period in those patients with delayed hematopoietic recovery. aThese also include various combination regimens. thresholds were not present at that time and graft failure was established if cell counts did not reach these levels at stan- Statistics dardized endpoints (Table 2) and were associated with extreme hypocellularity of bone marrow biopsies. Pretrans- plant busulfan (BU) and hydroxyurea (HU) treatment was gen- Because in our series of 124 patients bone marrow biopsies erally adapted to peripheral leukocyte counts and carried out were performed at significantly different intervals of the pre- using conventional dosages of these agents. Following BMT and post-transplant period, a comprehensively conducted a successful engraftment according to the above mentioned evaluation and comparison of data required a computation generally accepted criteria13–18 was found in 93 patients, a of defined endpoints within standardized intervals. This was delayed hematopoietic reconstitution in 31 including and fail- achieved by selecting a total of 351 examinations which had ure in three patients. During an observation time of more than been performed at certain short periods, in particular follow- 10 years a leukemic relapse occurred in 19 patients. Two ing BMT (Table 2). For this reason, morphometric analysis was patients with transplant-related mortality, ie death within 30 focused on bone marrow trephines at these endpoints. In days after BMT (day 0) were excluded from our calculations. accordance with a widely applied semiquantitative scoring system for myelofibrosis22 a more than three-fold increase in the normal density of argyrophilic (reticulin-collagen) fibers12 Bone marrow biopsies was consistent with an early to moderate myelofibrosis and readily recognizable in corresponding bone marrow samples. Fixation of bone marrow biopsies was carried out in a low- Statistical analysis included computation of the Mann–Whit- concentrated phosphate-buffered formaldehyde solution (2– ney U test to compare histomorphological parameters accord- 3%) for 12–48 h. Further processing included decalcification ing to engraftment status.

Leukemia Erythropoiesis and BMT in CML J Thiele et al 1380 Table 2 Quantiﬁcation of erythropoiesis, macrophages and reticulin ﬁbers (mean ± s.d.) per total marrow area and also per hematopoiesis (in brackets) according to different standardized endpoints of the transplantation procedure in CML. Patients may have up to six biopsies

Endpoints Biopsy intervals (days) No. of Cellularity Erythroid Macrophages Fiber density Hemoglobin g/dl patients % precursors mm2 i × 102/mm2 (hematocrit %) 2 range mean ± s.d. mm

(1) Pretreatment – – 35 98 ± 3 285 ± 136 374 ± 128 37 ± 25 13.6 (41) (322 ± (150) (438 ± 154) (45 ± 26) (2) Before BMT 1–120 24 ± 22 124 88 ± 17 279 ± 200 358 ± 100 37 ± 20 13.7 (40) (314 ± 223) (407 ± 120) (43 ± 20) BMT 0→ After BMT (3) 0–30 22 ± 62566± 23 183 ± 149 359 ± 125 29 ± 19 9.8 (28) (292 ± 199) (385 ± 264) (47 ± 23) (4) 31–45 38 ± 45264± 17 219 ± 144 329 ± 122 26 ± 12 10.8 (30) (312 ± 190) (489 ± 167) (41 ± 15) (5) 46–119 64 ± 17 56 55 ± 18 191 ± 140 321 ± 107 21 ± 13 10.6 (30) (295 ± 199) (531 ± 192) (55 ± 19) (6) 120–365 179 ± 61 18 50 ± 21 255 ± 140 322 ± 177 15 ± 9 11.0 (31) (387 ± 211) (555 ± 188) (31 ± 14)

Results areas of reconstituting erythropoiesis and centrally localized macrophages (Figure 2c). Depending on the standardized In the diagnostic pretreatment biopsies of the CML bone mar- selection of patients with short intervals between the post- row erythroblastic islets were dispersed among the prevalent transplant biopsies, a continuously occurring increase in the neutrophil granulopoiesis (Figure 1a) and were usually very amount of regenerating erythropoiesis was evident and, small (Figure 1b). Following various pretransplant therapeutic although supported by intensive transfusion therapy with regimens (Table 1) examinations performed shortly before packed cells, was accompanied by a corresponding rise of myelo-ablative therapy and BMT revealed no significant hemoglobin/hematocrit values (Table 2). Dynamics of regen- changes in the total amount of erythroid precursors or the erating erythropoiesis were closely linked with the other cell hemoglobin/hematocrit level in the majority of patients. Ther- lineages which has been explicitly calculated for megakary- apy with IFN-␣ resulted in a slight to moderate increase in opoiesis (Figure 1d). Statistical evaluation of the amount of erythropoiesis. However, probably due to the high standard CD61ϩ megakaryocytes after BMT revealed a comparable deviation in comparison to HU this improvement was not sig- capacity for recovery including the finding that a significantly nificant. Statistical analysis was in keeping with an adverse reduced number was also in keeping with delayed correlation (r =−0.506, P Ͻ 0.01) between manifest early to engraftment in the early post-transplant period. Some patients moderate reticulin myelofibrosis (Figure 2a) found in 33 mostly with delayed engraftment showed clusters of large patients of our series and the amount of erythropoiesis and macrophages (scavenger macrophages) shortly after BMT splenomegaly. In 18 patients the Sokal index revealed an (Figure 2e). Probably depending on the small number of intermediate or high pretransplant risk profile (Table 1) . Con- patients for this comparative study and the relatively high stan- trasting the low risk group these patients displayed a tendency dard deviation, the amount of macrophages revealed only a for a reduced amount of erythropoiesis and an enhanced tendency for increase per hematopoietic area in those patients reticulin content before BMT, thus signalling a more advanced with successful engraftment. In several patients four or even stage of disease. Moreover, in the pregraft bone marrow a sig- more sequential bone marrow biopsies were performed during nificant relationship was determinable between the quantity the post-transplant period demonstrating the progression of of nucleated erythroid precursors and the CD68ϩ macrophage erythropoietic recovery (Figure 3). population (r = 0.515, P Ͻ 0.01) calculated per area of Significant correlations were calculable between pretrans- hematopoiesis. plant number of erythroid precursor cells and fiber density After BMT (day 0) histopathology was characterized by a with delayed (31 patients) or successful (93 patients) conspicuous decrease in cellularity accompanied by an engraftment (Table 3). From these data it may be derived that expansion of the adipose tissue (Figure 1c). Starting at approxi- a relevant pretransplant reduction in the quantity of erythropo- mately day 10 there was a proteinaceous interstitial edema iesis and early to moderate myelofibrosis, which are indicative containing a variable number of tiny to small-sized erythro- for a more advanced stage of CML, exert an unfavorable poietic islets and a few dispersed megakaryocytes (Figure 1d) impact on transfusion-independent hematopoietic recovery. or fat tissue with groupings of erythroid precursors (Figure 1e). Progressive changes in erythropoiesis and fiber density In a number of patients with pretransplant manifest myelo- became most prominent when dividing our patients in two fibrosis (Figure 2a) an insidious increase in reticulin fibers was cohorts according to engraftment status. Thus additional detectable in the regenerating areas of hematopoiesis after evaluations of bone marrow features performed at stan- about 3–4 months (Figure 2b). Again a significant relationship dardized endpoints (endpoints 3 to 6 in Table 2) in the early (r = 0.550, P Ͻ 0.01) between quantity of erythroblasts and post-transplant period produced confirmative evidence for resident macrophages could be calculated in the post-trans- alterations of regenerating hematopoiesis in accordance with plant specimens immediately after BMT corresponding with completed or unsuccessful reconstitution. This finding is endpoint (3) in Table 2. This finding was not only demon- exemplarily shown about 2 weeks after day 30 at endpoint (4) strable by morphometric-statistical computation, but also by in Table 2. In comparison with the larger group of 41 patients histochemistry exhibiting a close spatial relationship between already displaying an unsupported recovery of hematopoiesis

Leukemia Erythropoiesis and BMT in CML J Thiele et al 1381

a c

d e

Figure 1 Erythropoiesis before and after BMT in CML. (a) Pretreatment CML bone marrow with dispersed and inconspicuous islets of erythroid precursors (arrowheads). (b) Erythroid islets are very small and contain erythroblasts at various stages of maturation. (c) Following BMT a decrease in cellularity is observable with groupings of the engrafted marrow between fat cells. (d) At about day 10 after BMT nests of regenerating erythroblasts (hot spots) besides CD61+ megakaryocytes (arrowheads) are embedded in a proteinaceous edema. (e) Erythroblastic islets cover a relatively extensive area between adipocytes in the reconstituting hematopoiesis shortly after BMT. (a) ×370; (b, d, e) ×570; (c) ×170; (a–c, e) Ret40f immunostaining; (d) CD61 immunostaining of megakaryocytes. significant differences regarding enhanced density of argyro- and therefore this cohort was not significantly different from philic fibers and retarded erythropoietic regeneration were the other patients. Analysis of the distinctive influence of detectable in the remaining 11 patients who presented with donor–recipient relationship, HLA constellation, conditioning the clinical findings of an delayed engraftment at that time regimen and GVHD prophylaxis failed to disclose significant (Figure 4). On the other hand, statistical analysis revealed that differences regarding the impact of erythropoiesis on this adverse effect regarding erythropoietic reconstitution was engraftment. This result probably depends on the relative limited to the first 2 months after BMT and could not be ascer- homogeneity of our series of patients derived from one trans- tained at the later endpoints shown in Table 2. Finally, of the plantation center (Table 1). Furthermore, similar findings were 19 patients with a leukemic relapse during the observation encountered regarding the ABO-blood group constellation. time (about 10 years) only six showed a delayed engraftment Patients with minor or major mismatch revealed no significant

Leukemia Erythropoiesis and BMT in CML J Thiele et al 1382

a b

cde

Figure 2 Fibers and macrophages in CML following BMT. (a) Mild reticulin ﬁbrosis in the pretreatment diagnostic biopsy. (b) Regeneration of reticulin ﬁbers about 4 months after BMT in the area of regenerating hematopoiesis. (c) Close contact of irregular stellate macrophages with reconstituted erythropoiesis following successful engraftment. (d) Retarded development of erythroid islets in a patient with delayed engraftment. (e) Large cluster of ‘scavenger macrophages’ shortly after BMT and delayed hematopoietic reconstitution. (a, b) ×370; (c–e) ×570; (a, b) silver

impregnation; (c, e) PGM1 immunostaining; (d) Ret40f immunostaining.

differences in the amount or regenerative capacity of various conditioning regimens used to prepare the patients nucleated erythroid precursors in the post-transplant bone represent a major determinant of transplant outcome,1,2,13–18 marrow samples. depending significantly on the number and types of cells that survive.23 The myelo-ablative therapy not only creates ‘microenvironmental niches’ for the differentiation and matu- Discussion ration of the donor hematopoietic stem cells, but also reduces the number of cells that may cause rejection.24 In our series A variety of host and donor factors has been repeatedly dis- comprising about 80% sibling donors with identical HLA con- cussed as exerting a significant influence on allogeneic mar- stellation from a single transplant center identical methods of row engraftment. Amongst others donor–recipient relation- myelo-ablative treatment were applied in 116 of the 124 ships and histocompatibility, pretransplant therapy and the patients (Table 1). Following pregraft conditioning regimens in

Leukemia Erythropoiesis and BMT in CML J Thiele et al 1383 quence of these changes which are expressed both by a reduced capacity of stem cells self-renewal and commitment towards erythroid-granulocytic and megakaryocytic lineages. A wealth of data has been gathered by the International Bone Marrow Transplant Registry (IBMTR) suggesting that advanced stages of CML may be determined by certain labora- tory parameters (degree of anemia, blasts in the peripheral blood, increasing splenomegaly, persistent thrombocytosis).4 In addition to these clinical variables, histological features such as myelofibrosis8–12 and the amount of erythroid precur- sors11 in the bone marrow have been previously postulated as indicators for progression. In this context convincing evidence has been produced that these parameters are characterized by significant mutual relationships and consequently exert their prognostic impact at first presentation of patients with CML.12 Concerning this study which comprises a considerable number of patients with pretransplant chemo- and interferon therapy (Table 1), it may be questioned whether these parameters Figure 3 Quantification and dynamics of erythroid precursors dur- of predictive value4,8–12 also exert an influence on ing allogeneic BMT in two patients with CML presenting with different engraftment. As could be shown in our series a significant pre- levels of hemoglobin (HB) at onset and four sequential post-transplant transplant reduction in erythropoiesis was usually associated bone marrow biopsies. with myelofibrosis and in keeping with an intermediate to high risk profile according to the Sokal index and thus suggested a Table 3 Significant associations between pretransplant (endpoint more advanced disease stage. This adverse constellation had 2, Table 1) amount of erythroid precursor cells and fiber density with an unfavorable impact on hematopoietic reconstitution accord- engraftment (platelet and granulocyte transfusion independence). His- ing to the generally accepted criteria (Table 3).13–18 Confirma- tological parameters are listed per total bone marrow area tive evidence for the validity of these features were addition- (mean ± s.d.) and also per hematopoiesis (in brackets) ally derived from bone marrow examinations performed shortly after the cut-off point of 30 days in the second month Engraftment status Statistical analysis of the post-transplant period (Figure 4). Regarding pretransplant myelofibrosis and its effect on platelet and red blood Successful Delayed P value (р30 days) (Ͼ30 days) cell transfusion independence after BMT divergent results were consecutively reported from one study group.26,27 The No. of patients 93 31 failure of the authors to assess definitely an adverse influence Erythroid 306 ± 213 201 ± 131 0.004 of myelofibrosis on hematopoietic recovery may be signifi- precursors mm2 (347 ± 232) (218 ± 159) 0.001 cantly related to the striking heterogeneity of the series of Fiber density 33 ± 16 47 ± 26 0.003 patients investigated. Besides a variety of pretreatment i × 102/mm2 (40 ± 18) (53 ± 25) 0.005 modalities, conditioning regimens and donor–recipient relationships these also included leukemic disorders at very different stages of the disease process in addition to malignant lymphomas and CML.27 Bone marrow stromal cells, in particular macrophages establish an intimate spatial contact and functional relationship with developing hematopoietic cells.28–36 This feature plays a key role in the regulation and differentiation of progenitors of all lineages including erythroid precursors. Here macrophages create a peculiar anatomic unit, the erythroblastic islets.31,32 In recent years the concept of the central- resident macrophages providing essential mediators for the complex mechanism of generating mature erythrocytes has gained direct support from experimental findings. Despite much detailed speculation the critical function of the central mononuclear phagocytes characterizing the erythroblastic islets has been limited to their involvement in the degradation Figure 4 Relationship between myelofibrosis and erythropoiesis of the expelled nuclei and the presumed turnover of iron and with engraftment parameters in allogeneic BMT for CML. Box and other metabolites that are generated by this breakdown.33–37 whisker plots were constructed for two groups of patients examined at day 38 Ϯ 4 after BMT (endpoint 4 in Table 2) according to their In addition to other putative mediators bone marrow macro- engraftment status (successful 41, delayed 11 patients). phages produce and secrete erythropoietin (EPO) which exerts a key function by promoting late stage erythroid maturation.19,35 The significant correlation between the number of vitro studies are in keeping with the finding that the microen- erythroid precursor cells and CD68ϩ macrophages calculated vironment remains damaged not only quantitatively but also per hematopoietic tissue in the post-transplant marrow speci- qualitatively several months after BMT, irrespective of any mens provides persuasive evidence for this distinctive func- allo-immune reaction or postgraft immunosuppressive ther- tional association. Moreover, in this context it is noticeable apy.25 Post-transplant hematopoietic impairment is the conse- that according to their myelo-monocytic progeny the pretrans-

Leukemia Erythropoiesis and BMT in CML J Thiele et al 1384 plant resident bone marrow macrophages in CML patients are up study in 186 patients with chronic myeloid leukemia. Virchows Phϩ (bcr/ablϩ).38,39 These malignant macrophages are sup- Arch A Pathol Anat 1992; 420: 473–478. posed to play a crucial role in the expansion of the leukemic 12 Thiele J, Kvasnicka HM, Fischer R. Bone marrow histopathology in chronic myelogenous leukemia (CML) – evaluation of distinctive cell clone and obviously contribute to the disturbances of the features with clinical impact. Histol Histopathol 1999; 14: 40 microenvironment in CML. 1241–1256. In conclusion, by an elaborate analysis of bone marrow fea- 13 Gratwohl A, Hermans J, Niederwieser D, Frassoni F, Arcese W, tures compelling evidence has been produced that in CML a Gahrton G, Bandini G, Carreras E, Vernant JP, Bosi A, de Witte significant reduction of erythropoiesis usually accompanied T, Fibbe WE, Zwaan F, Michallet M, Ruutu T, Devergie A, Iriondo by myelofibrosis before allogeneic transplantation is in keep- A, Apperley J, Reiffers J, Speck B, Goldman JM for the Chronic Leukemia Working Party of the European Group for Bone Marrow ing with a more advanced stage of the disease process and Transplantation. Bone marrow transplantantion for chronic thus greatly impairs engraftment. Following myelo-ablative myeloid leukemia: long-term results. Bone Marrow Transplant therapy and BMT, a rapid reconstitution of erythroid islets is 1993; 12: 509–516. observable in patients with successful engraftment. In keeping 14 Beelen DW, Graeven U, Elmaagacli AH, Niederle N, Kloke O, with their important functional role during erythropoiesis a Opalka B, Schaefer UW. Prolonged administration of interferon-␣ significant correlation between the amount of erythroid in patients with chronic-phase Philadelphia chromosome-positive precursors and number of mature-resident macrophages is chronic myelogenous leukemia before allogeneic bone marrow transplantation may adversely affect transplant outcome. Blood detectable in the pre- and also post-transplant period of 1995; 36: 2981–2990. hematopoietic recovery. 15 Lamparelli T, Van Lint MT, Gualandi F, Occhini D, Barbanti M, Sacchi N, Ficai G, Ghinatti C, Ferrara GB, Delfino L, Pozzi S, Morabito A, Zikos P, Vitale V, Corvo R, Frassoni F, Bacigalupo A. Acknowledgements Bone marrow transplantation for chronic myeloid leukemia (CML) from unrelated and sibling donors: single center experience. Bone Marrow Transplant 1997; 20: 1057–1062. We are greatly indebted to Mrs B Rosenbach, Mrs M Wons- 16 Van Rhee F, Szydlo RM, Hermans J, Devergy A, Frassoni F, Arcese chick and Mrs H Tilemann for their excellent technical assist- W, de Witte T, Kolb HJ, Niederwieser D, Jacobson N, Gahrton G, ance. This work was supported by a grant from the Dr M Bandini G, Carreras E, Bacigalupo A, Michallet M, Ruutu T, Reif- Scheel Foundation for Cancer Research (70–2134-Th1). fers J, Goldman JM, Apperley J, Gratwohl A for the Chronic Leuke- mia Working Party of the European Group for Blood and Marrow Transplantation. Long-term results after allogeneic bone marrow References transplantation for chronic myelogenous leukemia in chronic phase: a report from the Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation. Bone 1 Appelbaum FR, Clift R, Radich J, Anasetti C, Buckner CD. Bone Marrow Transplant 1997; 20: 553–560. marrow transplantation for chronic myelogenous leukemia. Semin Oncol 1995; 22: 405–441. 17 Dini G, Lamparelli T, Rondelli R, Lanino E, Barbanti M, Costa C, 2 Clift RA, Anasetti C. Allografting for chronic myeloid leukaemia. Manfredini L, Guidi S, Rosti G, Alessandrino EP, Locatelli F, Mar- Baillie`re’s Clin Haematol 1997; 10: 319–336. enco P, Soligo D, Di Bartolomeo P, Aversa F, La Nasa G, Busca 3 Goldman JM, Szydlo R, Horowitz MM, Gale RP, Ash AC, Atkinson A, Majolino I, De Laurenzi A, Bacigalupo A. Unrelated donor mar- K, Dicke KA, Gluckman E, Herzig RH, Marmont A, Masaoka T, row transplantation for chronic myelogenous leukaemia. Br J McGlave PB, Messner H, O’Reilly RJ, Reiffers J, Rimm AA, Speck Haematol 1998; 102: 544–552. B, Veum-Stone JA, Wingard JR, Zwaan FE, Bortin MM. Choise of 18 Hansen JA, Gooley TA, Martin PJ, Appelbaum F, Chauncey TR, pretransplant treatment and timing of transplants for chronic mye- Clift AR, Petersdorf EW, Radich J, Sanders JE, Storb RF, Sullivan logenous leukemia in chronic phase. Blood 1993; 82: 2235–2238. KM, Anasetti C. Bone marrow transplants from unrelated donors 4 Savage DG, Szydlo RM, Chase A, Apperley JF, Goldman JM. Bone for patients with chronic myeloid leukemia. New Engl J Med 1998; marrow transplantation for chronic myeloid leukaemia: the effects 338: 962–968. of differing criteria for defining chronic phase on probabilities of 19 Vogt C, Pentz S, Rich IN. A role for the macrophage in normal survival and relapse. Br J Haematol 1997; 99: 30–35. hematopoiesis. III. In vitro and in vivo erythropoietin gene 5 Gale RP, Hehlmann R, Zhang MJ, Hasford J, Goldman JM, Heim- expression in macrophages detected by in situ hybridization. Exp pel H, Hochhaus A, Klein JP, Kolb HJ, McGlave PB, Passweg J, Hematol 1989; 17: 391–397. Rowlings PA, Sobocinski KA, Horowitz MM, and the German CML 20 Gatter KC, Cordell JL, Turley H, Heryet A, Kiefer N, Anstee DJ, Study Group. Survival with bone marrow transplantation versus Mason DJ. The immunohistological detection of platelet, megakar- hydroxyurea or interferon for chronic myelogenous leukemia. yocytes and thrombi in routinely processed specimens. Histo- Blood 1998; 91: 1810–1819. pathology 1988; 13: 257–267. 6 Cortes J, Kantarjian HM, Giralt S, Talpaz M. Natural history and 21 Thiele J, Braeckel C, Wagner S, Falini B, Dienemann D, Stein H, staging of chronic myelogenous leukaemia. Baillie`re’s Clin Fischer R. Macrophages in normal human bone marrow and Haematol 1997; 10: 277–290. chronic myeloproliferative disorders: an immunohistochemical 7 Muehleck SD, McKenna RW, Arthur DC, Parkin JL, Brunning RD. and morphometric study by a new monoclonal antibody (PG-M1) Transformation of chronic myelogenous leukemia: clinical, mor- on trephine biopsies. Virchows Arch A Pathol Anat 1992; 421: phologic and cytogenic features. Am J Clin Pathol 1984; 82: 1–14. 33–39. 8 Thiele J, Hoefer M, Kvasnicka HM, Bertsch HP, Zankovich R, 22 Bauermeister DE. Quantitation of bone marrow reticulin: a normal Fischer R. Erythropoiesis in CML – immunohistomorphometric range. Am J Clin Pathol 1971; 56: 24–31. quantification, PCNA-reactivity, and influence on survival. Hema- 23 Martin PJ. Determinants of engraftment after allogeneic marrow tol Pathol 1993; 7: 239–249. transplantation. Blood 1992; 79: 1647–1650. 9 Lazzarino M, Morra E, Castello A, Inverardi D, Coci A, Pagnuco 24 Salomon O, Lapidot T, Terenzi A, Lubin I, Rabi I, Reisner Y. Induc- G, Magrini U, Bernasconi C. Myelofibrosis in chronic granulocytic tion of donor-type chimerism in murine recipients of bone marrow leukemia. Clinical pathologic correlations and prognostic signifi- allografts by different radiation regimens currently used in treat- cance. Br J Haematol 1986; 64: 227–240. ment of leukemia patients. Blood 1990; 76: 1872–1878. 10 Dekmezian R, Kantarjian HM, Keating M, Talpaz M, McCredie 25 Domenech J, Roingeard F, He´rault O, Truglio D, Desbois I, Col- KB, Freireich EJ. The relevance of reticulin stain-measured fibrosis ombat P, Binet C. Changes in functional capacity of marrow stro- at diagnosis in chronic myelogenous leukaemia. Cancer 1987; 57: mal cells after autologous bone marrow transplantation. Leuk 1739–1743. Lymphoma 1998; 29: 533–546. 11 Buhr T, Choritz H, Georgii A. The impact of megakaryocyte pro- 26 Rajantie J, Sale GE, Deeg HJ, Amos D, Appelbaum F, Storb R, liferation for the evolution of myelofibrosis. Histological follow- Clift RA, Buckner CD. Adverse effect of severe marrow fibrosis

Leukemia Erythropoiesis and BMT in CML J Thiele et al 1385 on hematologic recovery after chemoradiotherapy and allogeneic 34 Rich IN. A role for the macrophage in normal hemopoiesis. II. bone marrow transplantation. Blood 1986; 67: 1693–1697. Effect of varying physiological oxygen tensions on the release of 27 Soll E, Massumoto C, Clift RA, Buckner CD, Appelbaum FR, Storb hemopoietic growth factors from bone-marrow-derived macro- R, Sale G, Hackman R, Martin P. Relevance of marrow fibrosis in phages in vitro. Exp Hematol 1986; 14: 746–751. bone marrow transplantation: a retrospective analysis of 35 Wang CQ, Udupa KB, Lipschitz DA. The role of macrophages in engraftment. Blood 1995; 86: 4667–4673. the regulation of erythroid colony growth in vitro. Blood 1992; 28 Liesveld JL, Winslow JM, Kempski MC, Ryan DH, Brennan JK, 80: 1702–1709. Abboud CN. Adhesive interactions of normal and leukemic 36 Kelley LL, Koury MJ, Bondurant MC, Koury ST, Sawyer ST, Wick- human CD34ϩ myeloid progenitors: role of marrow stromal, rema A. Survival or death of individual proerythroblasts results fibroblast, and cytomatrix components. Exp Hematol 1991; 19: from differing erythropoietin sensitivities: a mechanism for con- 63–70. trolled rates of erythrocyte production. Blood 1993; 82: 2340– 29 Wilson JG, Tavassoli M. Microenvironmental factors involved in 2352. the establishment of erythropoiesis in bone marrow. Ann NY Acad 37 Ponka P, Beaumont C, Richardson DR. Function and regulation Sci 1994; 718: 271–284. of transferrin and ferritin. Semin Hematol 1998; 35: 35–54. 30 Obinata M, Okuyama R, Matsuda KI, Koguma M, Yanai N. Regu- 38 Golde DW, Burgaleta C, Sparkes RS, Cline MJ. The Philadelphia lation of myeloid and lymphoid development of hematopoietic chromosome in human macrophages. Blood 1977; 49: 367–370. stem cells by bone marrow stromal cells. Leuk Lymphoma 1998; 39 Thiele J, Schmitz B, Fuchs R, Kvasnicka HM Lorenzen J, Fischer 29: 61–69. R. Detection of the bcr/abl gene in bone marrow macrophages in 31 Breton-Gorius J, Vuillet-Gaugler MH, Coulombel L, Guichard J, CML and alterations during interferon therapy – a fluorescence in Teillet F, Vainchenker W. Association between erythroid progeni- situ hybridization study on trephine biopsies. J Pathol 1998; 186: tors and bone marrow macrophages. Blood Cells 1991; 17: 331–335. 127–146. 40 Bhatia R, McGlave BM, Dewald GW, Blazar BR, Verfaillie CM. 32 Hanspal M. Importance of cell–cell interactions in regulation of Abnormal function of the bone marrow microenvironment in erythropoiesis. Curr Opin Hematol 1997; 4: 142–147. chronic myelogenous leukemia: role of malignant stromal macro- 33 Gregory CJ, Eaves AC. Three stages of erythropoietic progenitor phages. Blood 1995; 85: 3636–3645. cell differentiation distinguished by a number of physical and bio- logic properties. Blood 1978; 51: 527–537.

Leukemia