Hematopoietic stem cell transplantation for infantile ... - Nature

Bone Marrow Transplantation, (1998) 22, 941–946  1998 Stockton Press All rights reserved 0268–3369/98 $12.00 http://www.stockton-press.co.uk/bmt

Hematopoietic stem cell transplantation for infantile osteopetrosis M Eapen, SM Davies, NKC Ramsay and PJ Orchard Department of Pediatrics and the Bone Marrow Transplant Program, University of Minnesota, MN USA

Summary: Infantile osteopetrosis is a lethal disorder resulting from a severe defect in the ability of osteoclasts to resorb bone. The only therapy shown to be capable of providing lasting benefit is allogeneic hematopoietic stem cell transplantation (HCT). We report the outcome of 10 patients with infantile malignant osteopetrosis treated with HCT from an HLA A, B, DRB1 matched (n = 6) or A or B locus mismatched (n = 4) family member or unrelated donor at the University of Minnesota between 1978 and 1997. Eight of 10 patients achieved primary engraftment; secondary graft failure was seen in two patients. Five of 10 patients survive; three with full or partial donor chimerism and two with autologous hematological recovery. Transient or partial donor chimerism can be sufficient to correct the hematological manifestations of osteopetrosis. We recommend early referral for consideration of HCT with a related or unrelated donor as neurosensory manifestations of osteopetrosis are generally not reversible. Donor engraftment may be easier to achieve early in the course of the disease. Keywords: osteopetrosis; transplantation; engraftment; chimerism

been described in these patients, and this appears to be associated with osteoclast dysfunction.4 Children with osteopetrosis are severely impaired if left untreated and only 30% survive to the age of 6 years.1 Reported treatments of infantile osteopetrosis include supportive care measures, dietary regimens to reduce calcium and increase phosphate intake, vitamin D, parathyroid hormone and calcitonin infusions to induce bone resorption.3 These measures have all been largely unsuccessful. The use of gamma interferon enhances bone resorption and leukocyte function,4 but the effects on morbidity and survival remain unclear. Allogeneic transplantation from an HLA-matched family member or unrelated donor has been shown in a small number of cases to be effective in reconstituting normal hematopoiesis. This can lead to immune recovery, bone remodeling, improved growth and survival.5–12 In this study we report the outcomes of 10 patients with infantile osteopetrosis, transplanted at the University of Minnesota between 1978 and 1997. The data indicate that allogeneic HCT can be an effective therapeutic option for these children.

Patients and methods Patients

Osteopetrosis is an inherited skeletal disease in which osteoclast dysfunction results in excessive bone deposition.1 Osteoclasts are bone-resorbing cells derived from hemopoietic stem cells and osteoblasts are bone-forming cells of mesenchymal origin derived from marrow stromal cells.2 In osteopetrosis, failure of osteoclasts to resorb and remodel bone in the presence of normal bone formation by osteoblasts results in deposition of excessive mineralized osteoid and cartilage.2,3 Osteopetrosis is a heterogeneous disease which can be inherited as an autosomal dominant characteristic and tends to result in a milder form of the disorder. In contrast, the malignant infantile form is inherited as an autosomal recessive characteristic and is characterized by dense, sclerotic bone, extramedullary hematopoiesis, bone fractures, growth failure, delayed tooth eruption and sensory and neurological impairment.1,3 Defects in neutrophil and monocyte function have been observed, as well as abnormalities in natural killer cell function.3 A defect in leukocyte superoxide formation has Correspondence: M Eapen, Box 484 UMHC, University of Minnesota, Minneapolis, MN 55455, USA Received 28 May 1998; accepted 8 July 1998

Clinical and laboratory data were retrieved from the University of Minnesota HCT database, which systematically and prospectively collects data on all consecutively transplanted patients. The marrow transplant protocols and consent forms are approved by the Institutional Review Board of the University of Minnesota. Ten patients (median age 9.5 months) with infantile malignant osteopetrosis were treated with allogeneic HCT between 1978 and 1997. The diagnosis of osteopetrosis was based on skeletal radiologic changes of uniformly dense sclerotic bone, absent medullary cavities on long bone X-rays, extramedullary hematopoiesis, anemia with leukoerythroblastosis, neurosensory defects and bone biopsies consistent with osteopetrosis. Table 1 shows the characteristics of patients with infantile osteopetrosis. UPN 0070 has been described previously.6 Seven patients were male. All patients had evidence of radiologic skeletal changes, bone marrow fibrosis and extramedullary hematopoiesis. Eight patients had abnormal visual evoked responses with varying degrees of clinical visual impairment. Three patients required optic nerve decompression prior to transplant. Four patients presented with abnormal auditory brain stem evoked responses and impaired hearing. All patients showed varying degrees of

Transplantation for osteopetrosis M Eapen et al

942

Table 1

Patient characteristics

UPN No.

Sex

Age at DX months

Age at HCT months

Clinical features at HCT

0070

F

3

5

0205 0285b 0306a,b 0784

M M F M

11 2 3 2.5

24 4 8 7

1990b

M

3

19

2122c 2270a,d 2496c 2665

F M M M

36 Birth Birth 6

43 2 7 25

Abnormal BAER, abnormal VER, bilateral optic atrophy, multiple cranial nerve palsies Normal BAER, normal VER Abnormal BAER, abnormal VER, pseudotumor cerebri Abnormal BAER, absent VER Normal BAER, abnormal VER, blind left eye, bilateral optic atrophy Normal BAER, absent VER, bilateral dysconjugate gaze, multiple cranial nerve palsies, global developmental delay Abnormal BAER, absent VER, bilateral optic atrophy Normal BAER, normal VER Normal BAER, absent VER Normal BAER, absent VER, bilateral optic atrophy, generalized hypotonia

All patients presented with hepatosplenomegaly, anemia with leukoerythroblastosis and thrombocytopenia. a Biological full siblings. b Bilateral narrowing of optic foramina requiring decompression. c Hydrocephalus with ventriculoperitoneal shunt. d Hypocalcemia at birth. BAER = brainstem audio evoked responses; VER = visual evoked responses.

developmental delay and all were noted to be macrocephalic; two required ventriculo-peritoneal shunt placement prior to transplant. One patient presented with generalized hypotonia and global developmental delay. Two patients had multiple cranial nerve palsies and one patient had pseudotumor cerebri of unknown etiology. UPNs 0306 and 2270 are full biological siblings. Preparative regimen Preparative regimens are shown in Table 2. Nine patients received cyclophosphamide 60 mg/kg for 2 days and the remaining patient, 50 mg/kg for 4 days. Three patients received busulfan 40 mg/m2/dose every 6 h for 2 days and one received procarbazine 12.5 mg/kg for 3 days, in addition to cyclophosphamide. All patients received TBI following chemotherapy. Eight patients received TBI in a single dose and two fractionated TBI. Two patients received splenic irradiation for massive splenomegaly. GVHD prophylaxis Patients received GVHD prophylaxis as shown in Table 2. Four patients received methotrexate, ATG, prednisone13 and two received ‘short-course’ methotrexate and cyclosporin A. Two patients received marrow ex vivo T cell depleted by counterflow elutriation, cyclosporin A and prednisone14 and one patient received cyclosporin A and prednisone.15 The remaining patient received erythrocyte rosette-depleted marrow and no additional prophylaxis.16

serologic DR typing (Table 2). Five patients received stem cells from family members and five received unrelated donor (URD) stem cells of which one was umbilical cord blood (UPN 2270). Potential unrelated donors were located through a network of national and international bone marrow donor registries as previously described.17 Three patients received HLA A,B DR matched sibling donor marrow, one received A,B DR matched marrow from an aunt, and one received marrow from a father (5/6 match GVHD vector; 4/6 match rejection vector). Two received URD marrow matched at HLA A,B and DRB1, one URD marrow with an A locus mismatch and one URD marrow with a B locus mismatch. One patient received URD cord blood with an A locus mismatch, supplied by the New York Blood Center.18 Analysis of outcomes Graft-versus-host disease was diagnosed and graded by standard criteria.19 The day of neutrophil engraftment was the first of 3 consecutive days on which the absolute neutrophil count was ⬎0.5 × 109/l. Secondary graft failure was defined as an absolute neutrophil count ⬍0.5 × 109/l for 3 consecutive days after achieving initial engraftment, in the absence of an infective or drug-related cause of cytopenia, or return of autologous hematopoiesis. Donor origin of reconstituted cells were documented using molecular markers20 and chromosome analysis.6

Results Donor and recipient compatibility Patients and donors were typed for HLA-A and B using serological techniques identifying all WHO-recognized specificities current at the time of transplant. Five pairs of patients were typed for DRB1 by high resolution typing. The remaining pairs, transplanted in previous years had

Engraftment and chimerism Post HCT outcomes are summarized in Table 3. Eight of 10 patients engrafted at a median of 25 days (range 16–28 days). Two patients (UPN 0306 and 2122) failed to achieve engraftment by the time of death, 52 days and 34 days,


Table 2 UPN No.

943

Transplant characteristics Donor

0070 sibling 0205 maternal aunt 0285 sibling 0306 father

Donor/Recipient compatibility

(R) HLA A2 A29 B44 B44 (D) HLA A2 A29 B44 B44 non-reactive MCL (R) HLA A2 A3 B7 B44 (D) HLA A2 A3 B7 B44 non-reactive MLC HLA (R) A2 A28 B62 B17 DR4 DR7 (D) A2 A28 B62 B17 DR4 BR7 HLA (R) A1 A29 B44 B44 DR51 DR7 (D) A1 A28 B44 B35 DR51 DR7

0784 sibling

HLA (R) A2 A3 B49 B27 DR76 DR5 (D) A2 A3 B49 B27 DR76 DR5 1990 unrelated HLA (R) A3 A28 B27 B35 DRB1 01 01 (D) A3 A11 B27 B35 DRB1 01 01 2122 unrelated HLA (R) A24 A31 B51 B35 DRB1 15 01 (D) A24 A31 B51 B35 DRB1 15 01 2270 unrelated HLA cord blood (R) A1 A29 B44 BW4 DRB1 07 01 (D) A23 A29 B44 BW4 DRB1 07 01 2496 unrelated

HLA (R) A1 A11 B58 B61 DRB1 07 01 (D) A1 A11 B44 B61 DRB1 07 01

2665 unrelated

HLA (R) A2 A11 B27 B62 DRB1 09 01 (D) A2 A11 B27 B62 DRB1 09 01

Preparative regimen

Cyclophosphamide 200 mg/kg TBI 400 cGyi Cyclophosphamide 120 mg/kg TBI 750 cGyi Cyclophosphamide 120 mg/kg TBI 750 cGyi Cyclophosphamide 120 mg/kg TBI 750 cGyi Procarbazine 37.5 mg/kg ATGb Cyclophosphamide 120 mg/kg TBI 750 cGyi Cyclophosphamide 120 mg/kg TBI 1320 cGyj Cyclophosphamide 120 mg/kg TBI 1320 cGyj Splenic irradiation 600 cGy Busulfan 320 mg/m2 Cyclophosphamide 120 mg/kg TBI 750 cGyi ATGc Busulfan 320 mg/m2 Cyclophosphamide 120 mg/kg TBI 750 cGyi ATGc Busulfan 320 mg/m2 Cyclophosphamide 120 mg/kg TBI 760 cGyi ATGc Splenic irradiation 500 cGy

GVHD prophylaxis

Methotrexatee ATGa Prednisoneg Methotrexatee ATGa Prednisoneg Methotrexatee ATGa Prednisoneg E-rosette T cell depletion

Nucleated cell dose

11 × 108/kg 3.2 × 108/kg 3.2 × 108/kg 0.41 × 108/kgd (post processing)

Methotrexatee ATGa Prednisoneg Cyclosporin A Methotrexatef

4.4 × 108/kg

Cyclosporin A Methotrexatef

3.0 × 108/kg

Cyclosporin A Methylprednisoloneh

2.81 × 108/kg

3.0 × 108/kg

Ex vivo T cell depletion 1.02 × 108/kg Cyclosporin A (post processing) Methylprednisoloneb CD34+ cells 6.8 × 106/kg CD3+ cells 3.2 × 105/kg Ex vivo T cell depletion 0.44 × 108/kg (post Cyclosporin A processing) Methylprednisoloneh CD34+ cells 1.5 × 106/kg CD3+ cells 2.6 × 105/kg

R = recipient; D = donor. ATG = antithymocyte globulin: a15 mg/kg i.v. (days 8, 10, 12, 14, 16, 18, 20) post HCT; b15 mg/kg i.v. for 3 days; c30 mg/kg i.v. for 4 days. dCD34+, CD3+ cell counts not available. Methotrexate: e15 mg/m2 (day 1), 10 mg/m2 i.v. (days 3, 6, 11, 18 and weekly to day 100) post HCT; f15 mg/m2 (day 1), 10 mg/m2 i.v. (days 3, 6, 11) post HCT; gPrednisone 40 mg/m2/day (days 7–20) post HCT; hMethylprednisolone 2 mg/kg/day (days 5–19) post HCT. TBI = total body irradiation: isingle dose; jfractionated dose (165 cGy twice a day for 4 days).

respectively, from transplant. Secondary graft failure with autologous marrow recovery was seen in two patients (UPN 0070 and 1990), 251 and 607 days after transplant. Both patients showed mixed donor chimerism in the immediate post-transplant period with progressive loss of donor engraftment until full autologous recovery. Following autologous recovery, UPN 1990 received 6 months of gamma interferon and 3 months of alpha interferon therapy. UPNs 0784 and 2496 currently have stable partial donor chimerism. Both initially had ⬎50% donor chimerism but showed a steady decline in the proportion of hematopoietic cells that are of donor origin, and both are currently stable with 1–10% donor cells by RFLP. Donor chimerism was complete and engraftment was sustained in three patients. However, two of three patients expired from transplant-related complications 208 and 237 days from transplant. One patient continues with complete donor chimerism 2 years post HCT. Hypercalcemia following transplantation was not seen in any of our patients.

GVHD Grades I–III acute GVHD occurred in four of eight patients who achieved donor engraftment (Table 3). One patient developed extensive chronic GVHD involving the respiratory and gastrointestinal tract (UPN 0205), which contributed to death. The four patients without GVHD all received HLA-identical marrow from siblings. Survival At the time of analysis (April 1998), five of 10 patients are alive, a median of 4 years post HCT (range 2–18 years). Five patients died between 34 and 209 days post HCT from transplant-related complications (Table 3). Four patients died of interstitial pneumonitis leading to respiratory failure, in one case associated with extensive cGHVD. One patient died from disseminated Aspergillus infection without engraftment (UPN 2122).


944

Table 3

Outcome of HCT

UPN No.

Time to myeloid Acute GVHD engraftment max grade

0070

Day 25

0

0205

Day 16

Skin 1 Liver 2

0285

Day 24

0

0306

Not achieved

N/E

0784

Day 27

0

1990

Day 25

0

2122 2270

Not achieved Day 23

N/E Skin 3

2496

Day 26

Skin 2

2665

Day 28

Skin 3 Rectum 1

Chronic GVHD

Major complications

0

Chimerism

Short stature, sclerotic bones, blind Mixed donor chimerism by left eye, impaired vision right eye cytogenetics until D251 when all cells host by cytogenetics Extensive Died of interstitial pneumonitis and 100% donor (cytogenetics) unresolved GVHD infection secondary to extensive involving chronic GVHD pulmonary and gastrointestinal tract 0 Died of interstitial pneumonitis and N/E bacterial infection N/E Died of non-interstitial pneumonitis N/E and bacterial infection 0 Short stature, growth hormone and 1–10% donor (RFLP) thyroid deficiency, impaired vision, bilateral cataracts, optic nerve decompression 0 Severe progressive osteopetrosis 0% donor Secondary graft failure D607 (RFLP) N/E Died from disseminated aspergillus N/E 0 Pneumonitis → IPPV and HD 100% donor steroids, short stature, hypothyroid, 2.5 years post transplant bilateral optic nerve decompression, visually impaired (stable) 0 Visually impaired (light perception 1–10% donor bilaterally), developmental delay 1 year post transplant 0 Died from interstitial pneumonitis 100% donor and alveolar hemorrhage 3 months post transplant

Survival

Alive 18 year + Dead

Dead Dead Alive 10 years + Alive 4 year + Dead Alive 2 year +

Alive 1 year + Dead

N/E = not evaluable.

Of the five surviving patients, one currently has full and two partial donor chimerism. Two patients with initial partial donor chimerism subsequently had full autologous reconstitution 251 days and 607 days from transplant (UPN 0070 and 1990). UPN 1990 showed no resolution of symptoms following transient initial engraftment and now has severe progressive disease. The patient is anemic and thrombocytopenic and transfusion dependent. In contrast, UPN 0070 who also had only transient donor engraftment, survives more than 18 years after transplantation with normal blood counts. Bone marrow biopsies up to August 1995 show focal areas of osteopetrosis. She is blind in her left eye, but has normal development, intelligence, hearing and is post pubertal. She is short and her bones are extremely sclerotic. At last contact (1996) she was attending college. The two patients with continuing partial donor chimerism show no evidence of progressive disease. UPN 0784 is blind in the left eye, and visual acuity in the right is markedly decreased but considered stable. Recently, he has developed bilateral cataracts. He attends fifth grade, uses specialized equipment for reading within the main classroom, is learning braille and is doing well academically. He hears well. Height and weight are below the 5th percentile for age, and he receives thyroid and growth hormone supplements. He suffers from malocclusion and multiple dental erosions. Hematological indices and pulmonary function tests are normal. UPN 2496 is also visually impaired, with some light perception bilaterally. He hears well, height and weight fall below the 5th percentile for age, and although he continues to gain milestones, develop-

ment is significantly delayed by comparison to chronological age. His blood counts and bone marrow biopsies are normal. Skeletal radiological changes consistent with disease are no longer present. UPN 2270, with full donor chimerism, has shown no signs of disease progression after transplantation. His height and weight are below the 5th percentile for age. He has had surgery for optic canal decompression and his visual acuity is 20/380 for each eye and stable. He has mild posterior subcapsular cataracts bilaterally. He hears well and is on Synthroid (Knoll Pharmaceuticals, USA). His development is 9 months delayed for chronological age of 36 months. His blood counts and bone marrow biopsies are normal. Discussion Autosomal recessive osteopetrosis presents during infancy with variable clinical expression as a result of osteoclast dysfunction.1 Bone remodeling occurs continuously through the resorption of old bone by osteoclasts and the laying down of new bone by osteoblasts and in osteopetrosis, imbalance between bone resorption and new bone formation lead to increased skeletal mass.2,3,21 Osteoclasts and osteoblasts are derived from marrow progenitor cells – osteoclasts from hemopoietic stem cells and osteoblasts from marrow stromal cells.2 While numerous molecular defects have been described in rodent models of osteopetrosis, the human genetic defect(s) remains unclear. It is thought that the defect could be intrinsic to the osteoclasts


or to the mesenchymal cells that constitute the microenvironment supporting the development and activation of the osteoclasts.2 Some patients have associated renal tubular acidosis and cerebral calcification; and in this group of patients, the defect has been identified as carbonic anhydrase II inhibitor deficiency.2,3 In another subgroup of patients, osteoclast dysfunction has been associated with craniometaphyseal dysplasia.2 A defect in leukocyte superoxide formation related to defective bone resorption is present in patients with osteopetrosis.4 In the osteopetrotic murine model (op/op), macrophage colony-stimulating factor is absent, however there is no evidence that this occurs in humans.2,22 Taken together these data suggest heterogeneity in the genetic basis for the clinical phenotype of osteopetrosis. In untreated patients, the risk of developing impaired hematopoiesis and visual impairment secondary to optic atrophy is about 75%.1 Early visual and hematological impairment, especially before 3 months of age, carry a very poor prognosis for survival.1 HCT replaces recipient osteoclasts with those of donor origin, although osteoblasts remain of recipient origin.23 HCT is the only potentially curative treatment for osteopetrosis. We describe HCT in 10 children with osteopetrosis treated at a single institution. Our analysis shows that eight of 10 patients achieved primary engraftment. In the two patients that failed to engraft, one received T cell-depleted marrow from a haploidentical parent and the other was transplanted late in the disease course when bone sclerosis and loss of marrow cavities were far advanced. In addition, two patients showed secondary graft failure with autologous reconstitution of hematopoiesis. One of these had received a relatively lower dose of TBI; the second patient was transplanted relatively late and had severe disease at the time of transplantation. These features may have impaired engraftment. Transient engraftment improved hematological indices in one patient but, a second patient with transient engraftment showed no benefit and has severe progressive osteopetrosis. This difference may reflect heterogeneity of the disease itself, as some murine models of osteopetrosis also do not respond to transplantation.2 Patients with stable partial and full donor chimerism have no clinical evidence of progressive disease. They have normal blood counts, bone marrow biopsies and complete resolution of radiological features of osteopetrosis. These data indicate that achieving sustained engraftment is a significant problem in children with osteopetrosis, particularly in the setting of increased HLA disparity and/or advanced disease. Quality of life post transplant was significantly influenced by the severity of symptoms at presentation. Visual impairment at presentation was irreversible. Neuropsychological development was variable, with younger children showing delay, but the two older patients attend mainstream school and college. Growth generally continued along the percentile achieved at the time of presentation, with very little catch-up growth, similar to earlier reports.24 Age at transplantation and the quality and durability of engraftment all play a role in determining clinical outcome. In this study of children receiving HCT for osteopetrosis, we show that engraftment is a significant problem,

especially in patients with advanced disease. Increased intensity of conditioning therapy may improve engraftment as seen in three patients who received cyclophosphamide, busulfan and TBI, although only partial donor chimerism was achieved in one of the three patients. Achieving donor chimerism remains a major limitation on the success of HCT in osteopetrosis and difficulty is likely due to disease progression with loss of marrow space and progressive hepatosplenomegaly. Early referral and transplantation is probably the best approach to this problem. Interestingly, transient or partial donor chimerism can be sufficient to correct the hematological complications of osteopetrosis, although disease heterogeneity may influence eventual outcome. We conclude that HCT offers an effective therapeutic option for this disease which otherwise carries a very poor prognosis. In the absence of a matched family donor, unrelated marrow or cord blood donors are a suitable alternative. Early transplantation offers the best option for limiting neurosensory defects and impairment of growth and development. Additionally, engraftment may be easier before disease becomes advanced. We recommend early referral of infants with osteopetrosis to a transplant center. Acknowledgements This work was supported in part by grants from the Children’s Cancer Research Fund and the Bone Marrow Transplant Research Fund.

References 1 Gerritsen EJA, Vossen JM, van Loo IHG et al. Autosomal recessive osteopetrosis: variability of findings at diagnosis and during the natural course. Pediatrics 1994; 93: 247–253. 2 Felix R, Hofstetter W, Cecchini MG. Recent developments in the understanding of the pathophysiology of osteopetrosis. Eur J Endocrinol 1996; 134: 143–156. 3 Coccia PF. Bone marrow transplantation for ostoepetrosis. In: Forman SJ (ed). Bone Marrow Transplant. Blackwell Scientific: Oxford, 1994, pp 874–882. 4 Key LL, Ries WL, Rodrigiz RM et al. Recombinant human interferon gamma therapy for osteopetrosis. J Pediatr 1992; 121: 119–124. 5 Ballet JJ, Griscelli C, Coutris C et al. Bone marrow transplantation in osteopetrosis. Lancet 1977; 2: 1137. 6 Coccia PF, Krivit W, Cervenka J et al. Successful bone marrow transplantation for infantile malignant osteopetrosis. New Engl J Med 1980; 302: 701–708. 7 Sorell M, Kapoor N, Kirkpatrick D et al. Marrow transplantation for juvenile ostoepetrosis. Am J Med 1981; 70: 1280– 1287. 8 Sieff CA, Levinsky R, Rogers D et al. Allogeneic bone marrow transplantation in infantile malignant osteopetrosis. Lancet 1983; 1: 437–441. 9 Fischer A, Friedrich W, Levinsky R et al. Bone marrow transplantation for immunodeficiencies and osteopetrosis: European survey. Lancet 1986; 1: 1080–1084. 10 Orchard PJ, Dickerman JD, Mathews CHE et al. Haploindentical bone marrow transplantation for osteopetrosis. Am J Hematol Oncol 1987; 9: 335–340. 11 Solh H, Da Cunha AM, Giri N et al. Bone marrow transplan-

945


946

12 13 14 15

16

17

tation for infantile malignant osteopetrosis. J Pediatr Hematol Oncol 1995; 17: 350–355. Taylor GM, Dearden SP, Will AM et al. Infantile osteopetrosis; bone marrow transplantation from a cousin donor. Arch Dis Child 1995; 73: 453–455. Ramsay NKC, Kersey JH, Robison LL et al. A randomized study of the prevention of acute graft versus host disease. New Engl J Med 1982; 306: 392–397. Wagner JE, Santos GW, Noga SJ et al. Bone marrow graft engineering by counterflow centrifugal elutriation: results of a phase I–II clinical trial. Blood 1990; 75: 1370–1377. Wagner JE, Rosenthal J, Sweetman R et al. Successful transplantation of HLA-matched and HLA-mismatched umbilical cord blood from unrelated donors: analysis of engraftment and acute graft versus host disease. Blood 1996; 88: 795–802. Filipovich AH, Ramsay NKC, McGlave P et al. Mismatched bone marrow transplantation at the University of Minnesota: use of related donors other than HLA MLC identical siblings and T cell depletion. In: Gale RP (ed). Rec Adv Bone Marrow Transplant. Alan R Liss: New York, 1983, pp 769–783. Davies SM, Shu XO, Blazar BR et al. Unrelated donor bone marrow transplantation: influence of HLA A and B incompatability on outcome. Blood 1995; 86: 1636–1642.

18 Rubinstein P, Dobrill L, Rosenfield RE et al. Processing and cryopreservation of placental/umbilical cord blood for unrelated bone marrow reconstitution. Proc Natl Acad Sci USA 1995; 92: 10119–10122. 19 Weisdorf D, Haake R, Blazar B et al. Treatment of moderate/severe acute graft versus host disease after allogeneic bone marrow transplantation: an analysis of clinical risk features and outcome. Blood 1990; 75: 1024–1030. 20 Blazar BR, Orr HT, Arthur DC et al. Restriction fragment length polymorphisms as markers of engraftment in allogeneic marrow transplantation. Blood 1985; 66: 1436–1444. 21 Manolagas SC, Jilka RL. Bone marrow, cytokines, and bone remodeling. New Engl J Med 1995; 332: 305–310. 22 Orchard PJ, Dahl N, Aukerman SL et al. Circulating macrophage colony-stimulating factor is not reduced in malignant osteopetrosis. Exp Hematol 1992; 20: 103–105. 23 Lajeunesse D, Busque L, Menard P et al. Demonstration of an osteoblast defect in two cases of malignant osteopetrosis. Correction of the phenotype after bone marrow transplant. J Clin Invest 1996; 98: 1835–1842. 24 Gerritsen JA, Vossen JM, Fasth A et al. Bone marrow transplantation for autosomal recessive osteopetrosis. J Pediatr 1994; 125: 896–902.