| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Department of Endocrinology (L.T., G.L., A.C.) and Divisions of Rheumatology (A.D.P., A.E.) and Hematology (B.S., A.M.R., G.D.R., B.R., C.S.), Federico II University of Naples, 80131 Naples, Italy
Address all correspondence and requests for reprints to: Carmine Selleri, M.D., Division of Hematology, Federico II University, Via S. Pansini 5, 80131 Napoli, Italy. E-mail: selleri{at}unina.it.
Abstract
Bone complications after allogeneic stem cell transplant (allo-SCT) include osteoporosis, fractures, and osteonecrosis. We investigated bone abnormalities in long-term survivors after busulfan cyclophosphamide-conditioning regimen, followed by human leukocyte antigen-identical sibling SCT. Bone density was measured by dual-energy x-ray absorptiometry at lumbar spine (LS) and femoral neck (FN) and phalangeal osteosonogrammetry (OSG) in 41 patients 1–10 yr after allo-SCT. Using colony-forming units-fibroblast (CFU-F) assay, we analyzed the repopulating capacity of clonogenic fibroblast progenitors belonging to the osteogenic stromal lineage. LS and FN bone mineral density (BMD) and phalangeal densitometric values were significantly reduced, compared with 188 healthy controls (P < 0.001). Decrease in T-score less than 1 SD was documented in 29% and 52% of patients at the LS and FN, respectively. OSG detected densitometric values with a T-score less than 1 SD in 68% of transplanted patients. The patients examined within the first 3 yr after transplant showed low BMD, which remained stable at FN and improved at LS. Phalangeal densitometry was low up to 10 yr after transplant. CFU-F was found permanently depressed and unable to give rise to a confluent stroma. Low serum osteocalcin levels were present throughout the whole follow-up period. A significant correlation was found between densitometric values detected by both techniques and CFU-F growth in vitro. Osteonecrosis was associated with lower FN BMD, and phalangeal densitometry correlated inversely with duration of amenorrhea and chronic graft vs. host disease requiring long-lasting steroid therapy. In conclusion, dual-energy x-ray absorptiometry and phalangeal OSG may provide complementary information on bone density after allo-SCT. Prolonged severe impairment of femoral BMD and phalangeal densitometry suggest that bone loss may persist for many years after transplant. Inability to regenerate a normal number of osteoblastic precursors in the stromal stem cell compartment may in part account for severe long-lasting posttransplant decrease in bone mass.
IN THE PAST FEW years, the use of allogeneic stem cell transplant (allo-SCT) in the treatment of hematological diseases has grown exponentially along with a progressive improvement in patients’ outcome because of a fall in transplant-related mortality (1). Because the population of posttransplant long-term survivors is rapidly growing, attention is now focused on early and late complications of this procedure. Osteoporosis, fractures, and avascular necrosis (AVN) are considered common complications leading to pain and disability, which negatively affect lifestyle and quality of life of transplanted patients (2, 3). A significant decrease in bone mass has been shown in heart-, kidney-, lung-, and liver-transplanted patients as well as survivors after autologous and allo-SCT (4, 5, 6, 7, 8, 9, 10, 11, 12). The mechanisms of bone metabolism alteration caused by allo-SCT are multiple and not completely understood (13). Major risk factors for transplant-related osteoporosis include myeloablative conditioning regimens; cytokine storm at the time of transplantation; posttransplant long-lasting high-dose steroids and cyclosporin-A (CsA) therapy; immobilization; and decreased kidney, liver, and bowel function; these result in reduced intake and altered metabolism of calcium and vitamin D (8, 13, 14, 15). In addition, most women experience ovarian failure after allo-SCT (16).
Osteoporosis is defined as compromised bone strength and increased susceptibility to fractures, and osteopenia represents a less severe bone abnormality (17). Altered biochemical markers of bone metabolism have been found early after allo-SCT, suggesting a transplant-induced increase in bone turnover (18, 19, 20). In line with this finding, a high prevalence of osteoporosis has been found at trabeculae-rich sites within 12 months after allo-SCT (9, 11, 21). Dual-energy x-ray absorptiometry (DEXA) is currently the best well-referenced method to diagnose transplant-related bone loss (22); quantitative computed tomography has also been adopted in one study (10). Although bone mineral density (BMD), a parameter determined by DEXA and tomography, accounts for about 70% of bone strength, there is increasing evidence that other structural aspects that largely contribute to fracture risk are not detected by these techniques (23, 24, 25). Multiple risk factors related to stem cell transplant (SCT) influence both mineral content and cellular components of bone, which cause modification of bone microarchitecture and mechanical properties (13). Phalangeal osteosonogrammetry (OSG), as detected by quantitative ultrasonometry, is a recent noninvasive and radiation-free method that provides information on bone density and mechanical properties of bone including density, elasticity, and width of trabeculae by assessing amplitude and speed of ultrasound signals crossing the bone (26, 27, 28, 29). Phalanges show an equal proportion of compact and cancellous bone and a high metabolic turnover, which make them an important skeletal site for early identification of bone changes (30).
Bone is constantly regenerated by a process of osteoclast-mediated resorption and osteoblast-induced replacement; alterations of this equilibrium are the primary causes of osteoporosis. Osteoclasts are multinucleated hemopoietic cells derived from the monocyte/macrophage lineage, whereas osteoblasts are mesenchymal-derived marrow stromal cells that allow the formation and mineralization of secreted bone matrix (31). Estimation of bone remodeling includes quantification of marrow colony-forming units-fibroblast (CFU-F) cells in vitro. Currently, CFU-F cells are believed to be the best in vitro surrogate for the most primitive precursors for osteoblasts (32). Chemotherapy used for hematological malignancies may damage the function of microenvironment precursors, whereas myeloablative-conditioning regimens, followed by autologous or allo-SCT, seem to delay regeneration of bone microenvironment (33, 34).
The main aim of this cross-sectional study was to assess bone damage in adults with functioning donor marrow graft lasting more than 1 yr. In particular, we focused on bone densitometry performed 1–10 yr after allo-SCT at three skeletal sites by two different methods: DEXA and OSG. Finally, densitometric results were compared with in vitro recovery of clonogenic fibroblast progenitors, which represents a pivotal step of bone remodeling.
Subjects and Methods
Patients and transplantation procedures
We evaluated bone abnormalities in 41 consecutive patients who had been successfully allotransplanted at least 1 yr before entering the study, with unmanipulated marrow from a human leukocyte antigen-identical sibling. Reasons for transplant included patients with acute myeloid leukemia (n = 18), chronic myeloid leukemia (n = 17), acute lymphoblastic leukemia (n = 4), and Hodgkin’s lymphoma (n = 2). There were 20 women and 21 men, with a median age of 28.5 yr (range, 14–50) at SCT, and 32 yr (range, 20–51) at time of bone evaluation. All patients had been conditioned with the regimen of 16 mg/kg busulfan and 120 mg/kg cyclophosphamide. All patients had also received CsA (1 mg/kg·d by continuous iv infusion from d -1 to d +20 and then 8 mg/kg·d orally) plus short-course methotrexate as prophylaxis for graft vs. host disease (GVHD). No patient had received prophylactic therapy with growth factors after SCT. Twenty-three patients developed acute GVHD (aGVHD) of global grade I–IV and were successfully treated with methylprednisolone at doses ranging from 2–10 mg/kg and then tapered as tolerated. Twenty-nine patients had been or were affected by chronic GVHD (cGVHD) (10 limited and 19 extensive form), treated with prednisolone at doses of 1–2 mg/kg, associated with CsA at doses ranging from 1–8 mg/kg·d. The median cumulative dose of steroids given to this cohort of patients was equivalent to 5.8 g prednisone (range, 0.8–24) for a period ranging 3–36 months. Calcium intake was estimated by description of the usual diet and ranged from 500–800 mg/d; daily supplemental calcium intake recommended for patients receiving steroid therapy was 1000 mg. Body mass index (BMI) was calculated for each patient as weight/height squared (kilograms per square meter). No patient smoked or drank alcohol or more than three cups of coffee per day. The clinical characteristics of the patients are summarized in Table 1
. The control group consisted of 188 healthy individuals, including family members and donors of stem cells, matched for age, sex, and BMI. Informed consent was obtained from all patients and controls in accordance with institutional guidelines, and the study design was made in accordance with the Helsinki II declaration.
|
Blood samples were obtained from all patients between 0800 and 1030 h every 3–6 months after SCT. Serum FSH, LH, 17ß-estradiol, and testosterone were monitored to assess gonadal function; measurements were performed by commercially available kits: FSH and LH with RIA (RIA; Biodata, Rimini, Italy), testosterone and estradiol using solid-phase chemoluminescent enzyme immunoassay (Diagnostic Products, Los Angeles, CA). At study entrance, serum calcium (Ca), phosphorus (P), creatinine, alkaline phosphatase (ALP), albumin, intact molecule PTH (iPTH), and osteocalcin were determined after at least 8 h fasting, and urinary Ca and hydroxyproline excretion were measured in a 24-h urine collection and corrected for creatinine excretion. The iPTH and serum osteocalcin levels were measured by RIA (Nichols Institute Diagnostics, San Juan Capistrano, CA); the detection limit of the latter was 0.35 mg/liter. Hydroxyproline excretion was measured by HPLC. Blood chemistry profile, including levels of Ca, P, ALP, 24-h urinary Ca excretion, and creatinine, was analyzed using a standard autoanalyzer.
Bone density evaluation, fracture, and AVN assessment
Bone density was determined simultaneously at three different skeletal sites by two methods: DEXA and phalangeal OSG. Lumbar spine (LS) (L1-L4) and femoral neck (FN) BMD were measured by DEXA, using the QDR 1000 densitometer (Hologic, Inc., Waltham, MA). Individual BMD values are expressed as grams per square centimeter and T- and Z-scores. Quality control was maintained by daily scanning of an anthropomorphic spine phantom. The coefficient of variation for the DEXA technique was less than 1% for the LS and 1.5% for the FN. The reference population adopted in this study was the international pooled sample provided by the manufacturer; their data, however, did not differ significantly from those obtained on a local sample in a study performed when the device was set up (35).
Phalangeal OSG evaluation was performed using DBM sonic bone profiler (Igea, Carpi, Mo, Italy) as previously described (36). The amplitude-dependent speed of sound (AD-SoS) and ultrasound bone profile index (UBPI) were measured (27). Measurements were carried out on the second to fifth proximal phalanges of the nondominant hand, and the device automatically averaged the AD-SoS values of the four fingers. AD-SoS results were reported as meters per second and T- and Z-scores. The latter two parameters were calculated with the software provided by the manufacturer using measurements obtained in a sample of Italian population as reference database. UBPI is an optimum mathematical combination of three signal parameters developed to better discriminate fracture risk (27). It represents the probability of a single subject to belong to the nonfractured group; its values are normalized and range from 0 to 1, 1 being attributed to the highest value obtained (37). Measurements were always performed by the same skilled operator and the coefficient of variation was 0.8%, determined by repeated measurements in a subgroup of 30 subjects (three measurement per person on three different days within 1 wk). According to the World Health Organization criteria, osteopenia and osteoporosis were defined by a T-score below -1 and -2.5 SD, respectively, for the DEXA technique (22). Using phalangeal OSG, osteopenia and osteoporosis were defined by AD-SoS T-score less than -1 and less than -3.2 SD, according to a recent epidemiological study on more than 10,000 women (27). Potential asymptomatic vertebral compression fractures were investigated by standard spinal radiographs in all patients with BMD T-score less than -2.5 SD. AVN was detected by computed tomography or magnetic resonance imaging.
CFU-F assay and stromal layer cultures
Bone remodeling was investigated by in vitro CFU-F study in 30 transplanted patients and 20 bone marrow donors. Mononuclear cells from bone marrow (BMMNCs) were isolated by density gradient centrifugation using lymphocyte separation medium. BMMNCs were washed by centrifugation at 1200 x g for 10 min at 20 C with PBS containing 2% fetal bovine serum (FBS) for CFU-F assay and stromal cell cultures. For CFU-F assay, BMMNCs were resuspended at a concentration of 2 x 106/ml in McCoy’s 5A modified medium containing 10% FBS with L-glutamine (Mesencult; StemCell Technologies, Vancouver, CA) supplemented with 1 x 10-8 mol/liter dexamethasone (Sigma, St. Louis, MO), which allows the recruitment of bone marrow mesenchymal cells to the osteoblastic lineage, and plated in 25-cm2 tissue culture flasks. Fibroblast colony growth was evaluated after incubation at 37 C, 5% CO2 for 14 d in a humidified atmosphere. Characteristic fibroblastoid cell aggregates of more than 50 cells were scored in situ as CFU-F under an inverted microscope. When needed, osteoblastic differentiation of the colonies was defined by their ability to express ALP activity, and tissue cultures were stained with crystal violet for scoring CFU-F. For marrow stromal layer, 5 x 106/ml BMMNCs were resuspended in a culture medium that consisted of long-term stem cell medium (Myelocult; StemCell Technologies) supplemented with 1 x 10-6 mol/liter hydrocortisone sodium hemisuccinate (Sigma) plated into 25-cm2 tissue culture flasks and incubated in a humidified atmosphere (37 C, 5% CO2). On a weekly basis, the stromal layer cultures were fed by complete replacement of the medium and analyzed for stromal confluence after 4–5 wk. All cultures were performed in duplicate. Lymphocyte separation medium, Hank’s balanced salt solution, and FBS were purchased from Life Technologies, Inc. (Gaithersburg, MD).
Statistical analysis
Data are expressed as mean ± SD and ±SEM as appropriate throughout the text and in tables. Analysis of risk factors was performed using Pearsons’s correlation coefficient for data expressed by parametric values and using paired Student’s t test for nonparametric variables. After grouping the patients with normal or pathological BMD parameters, 
2 test was used to assess association with specific clinical features. The linear regression was used to detect correlation between densitometer values and risk factors for bone loss. Statistical significance was considered for P less than 0.05.
Results
Biochemical parameters at the time of bone density evaluation
Serum Ca, P, creatinine, albumin, urinary Ca excretion, and iPTH were within the normal range in all patients and did not differ significantly from control values (Table 2). ALP was higher in patients than in controls, likely because several patients were affected by liver cGVHD. Mean serum testosterone and LH were normal in all males, whereas FSH was elevated in nine men, suggesting spermatogenesis impairment (Table 2). Ovarian insufficiency occurred in all but two (90%) women. Sixteen of 18 amenorrheic women were receiving hormone replacement therapy (HRT) at the time of testing, and gonadotropin levels were within the normal range in all except three, who were undertreated, likely because of reduced intestinal absorption caused by mild intestinal cGVHD. Two women were not treated by HRT because of severe liver cGVHD. Osteocalcin levels were significantly lower in patients than in controls (P = 0.004), but hydroxyproline excretion did not differ significantly between these two groups.
|
The median interval between SCT and bone status analysis was 36.5 months (range, 12–120). At L1-L4 vertebrae, lumbar BMD and Z-scores were significantly lower in patients than in controls (P < 0.001). According to World Health Organization criteria, seven patients (17%) had osteopenia (three males and four females), and five (12%) had osteoporosis (two females and three males) at this site. In addition, femoral BMD values and Z-scores differed significantly between patients and controls (P < 0.001), 15 patients (7 males and 8 females) (37%) having osteopenia and six (three males and three females) (15%) osteoporosis at this site. Phalangeal AD-SoS, Z-scores, and UBPI were significantly lower in patients than in controls (P < 0.001, all); T-score values were within the range of osteopenia in 25 patients (10 males and 15 females) (60%) and osteoporosis in three (two males and one female) (7%). When female and male patients were compared separately with their gender-matched controls, a significant decrease in bone density was detected at all skeletal sites evaluated (Table 3). OSG detected a significant difference in both Ad-SoS and UBPI. No significant difference in densitometric values was found between male and female patients (Table 3). Women with amenorrhea lasting more than 3 months had significantly lower AD-SoS values (1993 ± 80 vs. 2073 ± 53 m/sec, P = 0.04) and Z-scores (-1.81 ± 1.13 vs. -0.68 ± 0.77 SD; P = 0.04) than women with amenorrhea for less than 3 months, but the difference in UBPI levels (0.59 ± 0.21 vs. 0.76 ± 0.13; P = 0.09) did not reach significance. No difference was found regarding amenorrhea by the DEXA technique.
|
When bone mass was evaluated in relation to the time elapsed since transplantation, different findings were obtained depending on the analysis method (Fig. 1). Patients who were evaluated more than 36 months after SCT had significantly higher (P < 0.001) lumbar BMD than patients evaluated less than 36 months. BMD values of the group with longer follow-up period overlapped with those of normal controls (P = 0.16). Moreover, lumbar BMD showed a linear correlation with the time elapsed since SCT (P = 0.008). On the other hand, no difference was seen in patients evaluated before or after the 36th month since SCT at the FN and phalanxes. No correlation was found between OSG values and months from transplant, and AD-SoS remained low even after more than 6 yr. Twelve patients belonging to the group evaluated less than 36 months after SCT were reevaluated after an additional period of 12 months: Bone loss was 7% and 8% of the baseline value at the LS and FN, respectively, but it exceeded 10% at the phalanxes. No significant difference was found in osteocalcin levels after less than 3 or more than 3 yr from SCT.
|
Eight patients (4 females and 4 males) experienced bone AVN 11–120 months after allo-SCT. Sites included both femoral head in all patients with concomitant humeral involvement in three. Clinical features and densitometric values of the subjects with aseptic necrosis are summarized in Table 4. Groups with or without AVN were similar in terms of age, follow-up period, and BMI. All patients with AVN suffered from extensive cGVHD, having received CsA (P = 0.079) and steroid (P < 0.001) treatments for significantly longer periods than patients without this complication. The steroid dose received before AVN occurrence was also significantly higher (P < 0.005) than that of patients without this complication. Lumbar and femoral BMD were significantly lower in the group with AVN, but mean AD-SoS value and UBPI did not differ between patients with or without AVN (Table 4). However, six of eight patients who developed AVN showed AD-SoS T-score value less than -1 SD. In this study in a small number of patients with AVN, DEXA measurement was superior to OSG in identifying patients with AVN.
|
Marrow CFU-F and stromal layer in transplanted patients
All transplanted patients showed complete engraftment at the hemopoietic and molecular level at the time of analysis. Marrow compartment of stromal cells, measured as CFU-F cells, was decreased 2- to 3-fold in transplanted patients, compared with normal donors (22.3 ± 3/105 mononuclear cell plated vs. 55 ± 4; P < 0.0001) (Fig. 2). As some of us recently documented for the marrow long-term culture-initiating cell compartment (38), CFU-F cell number in transplanted patients was not influenced by the number of myeloid progenitors (CFU-GM) infused. Using the median value (2.93 x 104/kg CFU-GM) as cut-off, there was no difference in CFU-F frequency between the groups of patients who had received more or less CFU-GM (P = 0.74) (38). To analyze marrow microenvironment at a functional level, we studied the capacity of stromal cells to generate a confluent stromal layer, which is essential to support allogeneic hematopoietic progenitors growth in vitro. After 4- to 5-wk long-term cultures, marrow stromal cells produced a confluent marrow stroma only in 20% of patients, compared with 80% of normal controls (data not shown). Analyzing the effect of time elapsed since transplantation on marrow CFU-F number, we found marrow CFU-F compartment markedly depleted during the first 6 yr after transplant (Fig. 2). Between the 6th and 10th years, the mean marrow CFU-C cell number tended to increase (13.28 ± 3.48 vs. 33.30 ± 5.36 before and after 67 months, respectively; P = 0.004), although most of the patients showed CFU-F numbers permanently below those observed in normal controls. The cGVHD was significantly related to decreased number of CFU-F colonies in vitro (46.4 ± 9.31 vs. 13.7 ± 12.85; P = 0.01), but no relationship was found in patients with or without AVN (16.4 ± 18 vs. 20.2 ± 17; P = 0.06).
|
To gain insight into the cause of SCT-related bone abnormalities, a univariate analysis was performed to assess the relationship among potential risk factors, densitometric values, AVN occurrence, and CFU-F growth in vitro. The following risk factors were considered: age, gender, BMI, cumulative dose and length of steroid and CsA received, duration of amenorrhea, and aGVHD/cGVHD occurrence. Age at SCT correlated inversely with AD-SoS (r = -0.43; P = 0.006), UBPI (r = -0.39; P = 0.01), and lumbar BMD (r = -0.5; P = 0.01). A similar correlation was also found with age at bone status evaluation (P = 0.001, P = 0.01, and P = 0.02, respectively). Length of amenorrhea correlated with AD-SoS (r = -0.38; P = 0.03). On the other hand, no correlation was shown between densitometric parameters at any site and sex, BMI, immunosuppressive treatments, and cGVHD grading. Osteocalcin levels were inversely correlated with UBPI (r = -0.7; P = 0.03) and slightly with AD-SoS (r = -0.64; P = 0.06). Only extensive cGVHD was strongly related to the onset of AVN, but steroid treatment length and dose had borderline results (P = 0.05, both) (Table 4
).
CFU-F growth correlated with Ad-SoS, its Z-score (r = 0.48; P = 0.05; r = 0.48; P = 0.04, respectively) and with Z-scores of lumbar and FN BMD (r =0.39; P = 0.03 and r = 0.55; P < 0.04, respectively). The grade of aGVHD did not affect CFU-F number after SCT (P = 0.38, P = 0.06, and P = 0.43 for grade I, II, and III-IV). The cGVHD alone was related to a significantly lower number of CFU-F colonies in vitro (r = 0.55; P = 0.002).
Discussion
In the last few years, it has became clear that osteoporosis and osteonecrosis represent frequent and serious complications of allo-SCT. As already suggested by other cross-sectional-based studies, our results document a frequent decrease in BMD (T-score less than -1 SD) after allo-SCT at the lumbar spine (29% of patients) and even more at the femoral neck (51%) (10, 21, 39, 40). A significant decrease in BMD appears early after transplant and seems to continue over the first 3 yr with no deterioration afterward. Only two prospective studies have attempted to establish the exact time when bone loss occurs after SCT. A study by Välimäki et al. (9) revealed posttransplant bone loss within 6 months after allogenic SCT at both spine and FN. A more recent and larger study by Stern et al. (11) found a significant decrease in lumbar and femoral BMD within 3 months, the decrease in femoral BMD continuing (-2.5%) between the 3rd and the 12th month after transplantation. Other smaller studies (10, 21, 39, 40) have described bone loss, prevalent at FN, within a few months after SCT.
Bone strength is determined by two main features: bone density and bone quality. Although bone density accounts for most of the strength, bone quality refers to architecture, turnover, damage accumulation, and mineralization (41). The DEXA technique measures bone density and mineralization but does not provide information on architectural damage and bone formation. OSG may detect more physical properties of bone tissue and accounts for more structural changes than traditional methods (26, 27, 43). Ultrasound velocity depends on bone density and elasticity, trabecular orientation, and cortical to trabecular ratio, all of which are influenced by mineral content and organic matrix (26). By phalangeal OSG (27), 61% of patients could be classified as having osteopenia and 7.3% as having osteoporosis. Although a positive correlation was found in healthy controls between BMD and phalangeal OSG results, we failed to show any correlation between these two methods after allo-SCT. Such a lack of correlation likely represents further evidence that DEXA and OSG measure different parameters associated with bone loss. In contrast to lumbar densitometry, femoral BMD and ultrasonometric parameters (AD-SoS and UBPI) were persistently decreased in patients suggesting that bone loss may persist for many years after transplantation or may be irreversible.
Multiple atraumatic spine fractures were detected by screening in two patients who were osteoporotic at LS. AVN occurred in eight patients after 11–120 months since allografting. Men and women were equally affected, all of them having been previously treated with prolonged high-dose steroid therapy for extensive cGVHD. Significant statistical association was found by the 2 test between AVN occurrence and extensive cGVHD, but dose and length of steroid treatment resulted just at the significance limit (P = 0.05).
Discussion on the pathophysiology of posttransplant bone loss always rely on the main contributing factors: age, BMI, immunosuppressive treatment, and hypogonadism. As documented both in vitro and in vivo, steroids may reduce formation and increase resorption of bone, whereas CsA has been shown to increase both bone formation and resorption (44, 45). In addition, the onset of hypogonadism leads to rapid bone loss because of osteoclast overactivity (22).
A possible recovery in lumbar BMD with time after transplant was suggested by the Pearson’s correlation analysis between time elapsed since SCT and lumbar BMD. Among different risk factors for bone loss tested by univariate analysis, age at SCT correlated inversely with spine BMD and phalangeal AD-SoS. The latter was inversely correlated also to amenorrhea length. By the 2 test, longer CsA treatment (
199 d) and aGVHD were associated with lower lumbar BMD values, whereas longer steroid treatment (>210 d) and amenorrhea period (>3 months) were associated with lower OSG values. No association was found between densitometric values and BMI, cumulative doses of steroids and CsA, and cGVHD.
Mesenchymal stem cells residing in bone marrow are progenitors for osteoblasts and other mesenchymal cell lineages. CFU-Fs represent clonogenic mesenchymal progenitors leading to precursors for stromal microenvironment and osteogenic compartment able to support hemopoiesis and form bone tissue in vitro and in vivo (31, 32, 33). The origin of posttransplant CFU-F is still controversial; a recent large study (34) suggests a recipient origin. Under growth conditions known to promote the maturation of primitive osteogenic precursors in the CFU-F fraction of human bone marrow, the numbers of marrow CFU-F in transplanted patients remains permanently below those observed in normal donors, although some recovery has been observed after 6 yr after SCT. In agreement with a recent report in a small group of patients evaluated for osteoblastic precursors within the first year after SCT (46), we documented that impairment of osteogenic progenitors was severe and persistent after allo-SCT. Furthermore, bone marrow stromal cells of almost all transplanted patients were unable to give rise to confluent stroma in long-term cultures, implying also a long-lasting functional damage. These findings suggest that in transplanted patients the decrease in bone mass could in part reflect a decreased number and function of osteoblasts because of transplant-related loss of osteogenic progenitors. This hypothesis is further supported by the correlation found between low densitometric values and marked decrease in CFU-F growth and is in line with the finding of persistently low osteocalcin values.
In conclusion, our data confirm and expand the results of previous studies showing a significant decrease in BMD and a reduced functional capacity of osteoblastic precursors after allo-SCT (11, 46). Early bone loss may consist of both demineralization and organic matrix deficit, the first detectable by DEXA and the second by OSG. Although mineralization seems to improve at trabecular-rich sites (LS), no significant change was detected at cortical bone (FN). No improvement was revealed by OSG even after a prolonged follow-up. Phalangeal OSG and DEXA are not interchangeable methods for bone loss estimation. FN and phalangeal OSG may be preferable sites for investigation of transplant-related bone loss in patients who survive more than 3 yr from allografting. Whether combined measurements from these anatomical sites improve bone loss detection still needs to be determined. Although long-lasting steroid therapy for cGVHD and hypogonadism play major roles in determining bone loss, our study also shows a severe and permanent deficit in number and function of osteoblastic precursors within the stromal stem cell compartment, suggesting that inability to regenerate a normal osteogenic cell compartment may in part account for severe bone damage after allo-SCT.
Acknowledgments
Footnotes
This work was supported in part by grants from the Ministero dell’Università e della Ricerca Scientifica e Tecnologica (MURST) and the Consiglio Nazionale delle Ricerche (CNR).
Abbreviations: Ad-SoS, Amplitude-dependent speed of sound; aGVHD, acute graft vs. host disease; allo-SCT, allogeneic stem cell transplant; ALP, alkaline phosphatase; AVN, avascular necrosis; BMD, bone mineral density; BMI, body mass index; BMMC, mononuclear cells from bone marrow; Ca, calcium; CFU-F, colony forming units-fibroblast; CFU-GM, colony forming units-myeloid progenitor; cGVHD, chronic graft vs. host disease; CsA, cyclosporin-A; DEXA, dual-energy x-ray absorptiometry; FBS, fetal bovine serum; FN, femoral neck; GVHD, graft vs. host disease; HRT, hormone replacement therapy; iPTH, intact molecule PTH; LS, lumbar spine; OSG, osteosonogrammetry; P, phosphorus; SCT, stem cell transplant; UBPI, ultrasound bone profile index.
Received May 23, 2002.
Accepted August 9, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  L. Tauchmanova, A. Colao, G. Lombardi, B. Rotoli, and C. Selleri REVIEW: Bone Loss and Its Management in Long-Term Survivors from Allogeneic Stem Cell Transplantation J. Clin. Endocrinol. Metab., December 1, 2007; 92(12): 4536 - 4545. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Petryk, T. L. Bergemann, K. M. Polga, K. J. Ulrich, S. K. Raatz, D. M. Brown, L. L. Robison, and K. S. Baker Prospective Study of Changes in Bone Mineral Density and Turnover in Children after Hematopoietic Cell Transplantation J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 899 - 905. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Syrjala, S. L. Langer, J. R. Abrams, B. E. Storer, and P. J. Martin Late Effects of Hematopoietic Cell Transplantation Among 10-Year Adult Survivors Compared With Case-Matched Controls J. Clin. Oncol., September 20, 2005; 23(27): 6596 - 6606. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kananen, L. Volin, K. Laitinen, H. Alfthan, T. Ruutu, and M. J. Valimaki Prevention of Bone Loss after Allogeneic Stem Cell Transplantation by Calcium, Vitamin D, and Sex Hormone Replacement with or without Pamidronate J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 3877 - 3885. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Tauchmanova, P. Ricci, B. Serio, G. Lombardi, A. Colao, B. Rotoli, and C. Selleri Short-Term Zoledronic Acid Treatment Increases Bone Mineral Density and Marrow Clonogenic Fibroblast Progenitors after Allogeneic Stem Cell Transplantation J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 627 - 634. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
