- Email: [email protected]
Avascular Necrosis of Bone after Allogeneic Hematopoietic Cell Transplantation in Children and Adolescents
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
587
Avascular Necrosis of Bone after Allogeneic Hematopoietic Cell Transplantation in Children and Adolescents Xiaxin Li 1, Ruta Brazauskas 2, Zhiwei Wang 3, Amal Al-Seraihy 4, K. Scott Baker 5, Jean-Yves Cahn 6, Haydar A. Frangoul 7, James L. Gajewski 8, Gregory A. Hale 9, Jack W. Hsu 10, Rammurti T. Kamble 11, Hillard M. Lazarus 12, David I. Marks 13, Richard T. Maziarz 8, Bipin N. Savani 7, Ami J. Shah 14, Nirali Shah 15, Mohamed L. Sorror 5,16, William A. Wood 17, Navneet S. Majhail 18,19, * 1
Banner Children’s Specialists Pediatric Hematology/Oncology, Banner Health, Mesa, Arizona Division of Biostatistics, Center for International Blood and Marrow Transplant Research, Medical College of Wisconsin, Milwaukee, Wisconsin 3 Center for International Blood and Marrow Transplant Research, Medical College of Wisconsin, Milwaukee, Wisconsin 4 Hematopoietic Stem Cell Transplantation Program, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia 5 Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, Washington 6 Clinique Universitaire d’Hematologie, University Hospital, Grenoble, France 7 Pediatric Blood and Marrow Transplant Program, Vanderbilt University Medical Center, Nashville, Tennessee 8 Bone Marrow & Stem Cell Transplant Program, Oregon Health and Science University, Portland, Oregon 9 Division of Hematology and Oncology, All Children’s Hospital, St. Petersburg, Florida 10 Division of Hematology & Oncology, Shands HealthCare and University of Florida, Gainesville, Florida 11 Section of Hematology-Oncology, Baylor College of Medicine Center for Cell and Gene Therapy, Houston, Texas 12 Seidman Cancer Center, University Hospitals Case Medical Center, Cleveland, Ohio 13 BMT Unit, Bristol Children’s Hospital, Bristol, United Kingdom 14 Division of Hematology/Oncology, Department of Pediatrics, Mattel Children’s Hospital at UCLA, Los Angeles, California 15 National Cancer Institute, National Institutes of Health, Bethesda, Maryland 16 Division of Medical Oncology, Department of Medicine, University of Washington School of Medicine, Seattle, Washington 17 Division of Hematology and Oncology, University of North Carolina Hospitals, Chapel Hill, North Carolina 18 Center for International Blood and Marrow Transplant Research, Minneapolis, Minnesota 19 Blood & Marrow Transplant Program, Cleveland Clinic, Cleveland, Ohio 2
Article history: Received 18 November 2013 Accepted 23 December 2013 Key Words: Avascular necrosis Late complications Hematopoietic cell transplantation
a b s t r a c t We conducted a nested case-control study within a cohort of 6244 patients to assess risk factors for avascular necrosis (AVN) of bone in children and adolescents after allogeneic transplantation. Eligible patients were 21 years of age, received their first allogeneic transplant between 1990 and 2008 in the United States, and had survived 6 months from transplantation. Overall, 160 patients with AVN and 478 control subjects matched by year of transplant, length of follow-up and transplant center were identified. Patients and control subjects were confirmed via central review of radiology, pathology, and/or surgical procedure reports. Median time from transplant to diagnosis of AVN was 14 months. On conditional logistic regression, increasing age at transplant (5 years), female gender, and chronic graft-versus-host disease (GVHD) were significantly associated with increased risks of AVN. Compared with patients receiving myeloablative regimens for malignant diseases, lower risks of AVN were seen in patients with nonmalignant diseases and those who had received reduced-intensity conditioning regimens for malignant diseases. Children at high risk for AVN include those within the age group where rapid bone growth occurs as well as those who experience exposure to myeloablative conditioning regimens and immunosuppression after hematopoietic cell transplantation for the treatment of GVHD. More research is needed to determine whether screening strategies specifically for patients at high risk for developing AVN with early interventions may mitigate the morbidity associated with this complication. Ó 2014 American Society for Blood and Marrow Transplantation.
INTRODUCTION Avascular necrosis (AVN) of the bone is a debilitating late complication of allogeneic hematopoietic cell transplantation (HCT) that can be associated with significant morbidity [1,2]. The incidence and risk factors for AVN have been well described in adult transplant recipients with an estimated
Financial disclosure: See Acknowledgments on page 591. * Correspondence and reprint requests: Navneet S. Majhail, MD, MS, Blood & Marrow Transplant Program, Cleveland Clinic, 9500 Euclid Ave., R35, Cleveland, OH 44195. E-mail address: [email protected] (S. Kharbanda). 1083-8791/$ e see front matter Ó 2014 American Society for Blood and Marrow Transplantation. http://dx.doi.org/10.1016/j.bbmt.2013.12.567
cumulative incidence of 3% to 10% at 5 years after transplantation [1,3-9]. Graft-versus-host disease (GVHD), exposure to corticosteroids or calcineurin inhibitors, cumulative dose of corticosteroids, older age, female gender, and use of total body irradiation (TBI) as part of conditioning regimen have been identified as risk factors for AVN in adult HCT recipients. Although its pathogenesis is poorly understood, potential mechanisms for development of AVN include local vascular damage that leads to increased marrow edema and ischemia, ineffective osteoblastic repair processes due to metabolic factors, and mechanical stresses [1,10]. Large studies specifically focusing on evaluating risk factors for AVN in pediatric HCT survivors are lacking. Factors such as immaturity and ongoing growth of bones and endocrine dysfunction related to growth and sex hormones
588
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
are exclusive to children and may modulate the risks of AVN in a different way from adults. Hence, extrapolating findings from studies that have only included adults or have combined adults with children can be a challenge. Also, it is not known whether the relatively recent less toxic preparative regimens (nonmyeloablative and reduced-intensity conditioning) are associated with lower risks of AVN than conventional myeloablative regimens in this population. To better understand the risk factors for AVN after allogeneic HCT in children and adolescents, we conducted a case-control study using data from the Center for International Blood and Marrow Transplant Research (CIBMTR). We evaluated risk factors from both “older approaches” (myeloablative regimens, greater use of sibling donors) and “contemporary approaches” (nonmyeloablative/reduced intensity regimens, greater use of unrelated donors) in our analysis. METHODS Data Source The CIBMTR is a working group of more than 450 transplantation centers worldwide that contribute detailed data on HCTs to a Statistical Center at the Medical College of Wisconsin in Milwaukee and the National Marrow Donor Program in Minneapolis. Participating centers are required to report all transplants consecutively, and patients are followed longitudinally. Computerized checks for discrepancies, physicians’ review of submitted data, and onsite audits of participating centers ensure data quality. Data are collected before transplant, 100 days and 6 months after transplant, and annually thereafter, or until death. The follow-up research forms specifically inquire whether a recipient has developed AVN post-transplantation. Observational studies conducted by the CIBMTR are performed under the guidance of the Institutional Review Board of the National Marrow Donor Program and are in compliance with all applicable federal regulations pertaining to the protection of human research participants. Patients For our study, we selected first allogeneic HCT recipients aged 21 years at transplantation who had been reported to the CIBMTR between 1990 and 2008. Because screening and management practices for AVN can vary by region, we restricted our cohort to patients who had received their transplant at a center in the United States. We also limited our cohort to patients who had survived at least 6 months or more after transplantation because our analysis was focused on long-term HCT survivors and on transplant-related risk factors for AVN. Patients with any diagnosis and recipients of both myeloablative and reduced-intensity/nonmyeloablative regimens were eligible. Selection of Patients and Control Subjects Overall, 6244 patients met study eligibility criteria and were the basis for selection of patients and control subjects for our study. Patients were those who had a diagnosis of AVN reported on post-transplant follow-up. AVN of any joint was considered. For all cases identified as potential patients, we requested diagnostic and/or treatment information from centers to ascertain the diagnosis of AVN (eg, copies of radiologic investigations, pathology reports, or surgical operative notes). We excluded 2 patients from our analysis for whom we were not able to confirm the diagnosis of AVN from their transplant center. We established a pool of control subjects using eligible patients who had received their transplant at the same centers as patients and did not have a diagnosis of AVN reported to the CIBMTR. For each patient we chose a control subject that was matched by year of transplantation (1 year) and follow-up duration (follow-up post-transplant no less than the interval from HCT to onset of AVN for the corresponding patient). Control subjects were selected from the same center as patients, if available. If a control subject could not be identified for a patient from the same center, control subjects were selected from another center that had patients with AVN included in this study. Each patient was matched with up to 3 control subjects. For patients with several matched control subjects, 3 were selected randomly for the analysis. For each selected control subject, we contacted transplant centers and requested them to review medical records and confirm that the patient did not have AVN. On this review, 4 control subjects were identified to have AVN post-transplantation and were subsequently considered as patients. Control subjects were excluded from the analysis if they had a pre-HCT diagnosis of AVN (n ¼ 1) or if centers were not able to confirm the absence of AVN diagnosis (n ¼ 26). We identified 160 confirmed patients with AVN and 478 matched control subjects. Among these case-control pairs, 407 (85%) were matched
within the same center as the patient. One hundred fifty nine patients had 3 matched control subjects, and 1 patient had 1 matched control subject. Study Definitions and Statistical Analysis Conditioning regimens were defined as myeloablative, reduced intensity, and nonmyeloablative using established guidelines [11]. Because no clear guidelines exist for classifying conditioning regimens for nonmalignant diseases, these diseases were considered as a separate category when describing conditioning regimen intensity. Disease status for malignant diseases was assigned as early, intermediate, or advanced [12]. Early disease included acute myeloid leukemia or acute lymphoblastic leukemia in first complete remission, chronic myeloid leukemia in first chronic phase, myelodysplastic syndrome refractory anemia, or refractory anemia with ringed sideroblasts or unspecified myelodysplastic syndrome with <5% marrow blasts. Patients with acute myeloid leukemia or acute lymphoblastic leukemia in second or greater remission, chronic myeloid leukemia in second or greater chronic phase, or chronic myeloid leukemia in accelerated phase were classified as intermediate-risk disease. All other patients, including patients with lymphoma, were classified as advanced disease. The National Marrow Donor Program’s classification of HLA matching status was used for unrelated donor transplant recipients (well matched, partially matched, or mismatched) [13]. The goal of our case-control study was to assess potential risk factors for developing AVN in children and adolescents after allogeneic HCT. For comparing characteristics between patients and control subjects, we used the chi-square or Fisher’s test (as applicable) for categorical variables and Wilcoxon 2-sample test for continuous variables. To evaluate risk factors, we performed multivariable analyses using conditional logistic regression on all matched sets. The following variables were considered in this analysis: age at transplantation, gender, diagnosis, disease status, conditioning regimen intensity, dose of TBI, donor source, and history of GVHD before AVN. If feasible, categories with a small number of patients were combined with related categories. Patients receiving transplant from HLA-mismatched related donors (n ¼ 23), patients with unknown conditioning regimen intensity (n ¼ 2), and patients with unknown date of GVHD onset (n ¼ 14) were excluded from the risk factor analysis. All P-values are 2-sided. All analyses were carried out using SAS statistical software (SAS Institute Inc., Cary, NC).
RESULTS Characteristics of Patients and Control Subjects Table 1 shows the characteristics of the 160 AVN patients and 478 control subjects. The median age at transplantation was 15 years for patients and 8 years for control subjects. The primary diagnosis of nonmalignant disorder was higher in the control group (32%) compared with AVN patients (13%). A greater proportion of AVN patients had received TBI containing myeloablative regimen compared with control subjects (65% versus 48%). Related, unrelated, and umbilical cord blood donors were used in 21%, 64%, and 16% of AVN patients and 11%, 69%, and 21% of control subjects, respectively. Fiftysix percent of patients had a history of chronic GVHD before the onset of AVN compared with 49% in the control group. Among AVN patients, the median time from HCT to the onset of AVN was 14 months (range, <1 to 172 months). In 37% of patients AVN occurred within 1 year of HCT, in 59% AVN occurred 1 to 5 years after HCT, and in 4% it occurred more than 5 years after transplantation. Detailed information was available for 59 patients to completely characterize the extent of joint involvement by AVN. Among these patients, collectively 119 joints were affected by AVN with a median of 2 joints (range, 1 to 6). Femoral head (82%) was the most common site of involvement and was followed by the knee joint (78%), the vertebral column (12%), and the ankle joint (10%). AVN of the shoulder joint was rare (5% of patients). Pathologic fracture was the initial presentation of AVN in 3 patients. We also evaluated the characteristics of patients included in our study by donor source (related ¼ 83 patients [41% patients, 59% control subjects], unrelated ¼ 432 patients [23% patients, 77% control subjects], umbilical cord blood ¼ 123 [20% patients, 80% control subjects]). There were notable differences among related, unrelated, and umbilical
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
Table 1 Characteristics of AVN Patients and Their Control Subjects (Matched by Transplant Center, Year of Transplant, and Duration of Follow-Up) Characteristics
Patients
Control Subjects
Number of patients Number of centers Age at transplant, yr, median (range) Age at transplant, yr <5 5-9 10-14 15-21 Patient gender Male Female Lansky/Karnofsky score before transplant <90 90 Missing Patient race White Non-White Disease Acute myeloid leukemia Acute lymphoblastic leukemia Chronic myeloid leukemia Myelodysplastic syndrome Non-Hodgkin lymphoma Hodgkin lymphoma Severe aplastic anemia Inherited abnormality of erythrocyte differentiation SCID and other immune system disorders Inherited disorder of metabolism Histiocytic disorders Disease risk before transplant Early Intermediate Advanced Nonmalignant disease Unknown Interval from diagnosis to transplant, mo <6 6-11 12 Unknown Year of transplant* 1990-1994 1995-1999 2000-2004 2005-2008 Total TBI dose, cGy No TBI <1200 cGy 1200 cGy Missing Conditioning regimen intensity Myeloablative (with TBI) Myeloablative (no TBI) Nonmyeloablative/reduced intensity Nonmalignant diseases Missing Graft type Bone marrow Peripheral blood bone marrow Umbilical cord bloody Donor type HLA-matched sibling Other related Well-matched unrelated Partially/mismatched unrelated
160 54 15 (2-21)
478 52 8 (<1-21) <.01 <.01 154 (32) 112 (23) 94 (20) 118 (25) <.01 289 (60) 189 (40) .99
5 16 56 83
(3) (10) (35) (52)
78 (49) 82 (51)
22 (15) 121 (85) 17
66 (15) 362 (85) 50
128 (80) 32 (20)
385 (81) 93 (19) <.01
(29) (36) (9) (11) (1) (1) (7) (3)
119 128 31 32 15 0 42 36
(25) (27) (6) (7) (3) (9) (8)
2 (1)
42 (9)
2 (1) 1 (1)
20 (4) 13 (3) <.01
62 55 22 21
(39) (34) (14) (13)
0
130 147 43 153 5
(27) (31) (9) (32) (1) .02
60 31 66 3
(38) (19) (41) (2)
135 100 205 38
(28) (21) (43) (8)
5 13 60 82
(3) (8) (38) (51)
19 40 215 204
(4) (8) (45) (43)
Table 1 (continued) Characteristics
Patients
P
.88
47 57 14 18 2 1 11 5
589
Unrelated, match unknown Umbilical cord blood GVHD prophylaxis FK506 þ MMF others FK506 þ MTX others (except MMF) FK506 others (except MTX, MMF) CSA þ MMF others CSA þ MTX others (except MMF) CSA others (except MTX, MMF) T cell depletion Other/unknown History of GVHDz No GVHD Acute GVHD only Chronic acute GVHD Interval from transplant to AVN, mo, median (range) Follow-up of survivors, mo, median (range)*
Control Subjects
1 (1) 25 (16)
2 (<1) 98 (21)
13 (8) 28 (18)
26 (5) 86 (18)
P
.51
9 13 58 24 10 5
(6) (8) (36) (15) (6) (3)
18 41 161 72 56 18
(4) (9) (34) (15) (12) (4)
43 28 89 14
(27) 179 (37) (18) 66 (14) (56) 233 (49) (<1-172) d
.05
61 (11-194)
63 (7-225)
SCID, severe combined immunodeficiency; MMF, mycophenolate mofetil; MTX, methotrexate; CSA, cyclosporine. Values are total number of incidences with percents in parentheses, unless otherwise noted. * Variable used for matching patients and control subjects. y Includes 6 related umbilical cord blood transplants (1 patient, 5 control subjects). z History of GVHD before AVN for patients; for control subjects, any history of GVHD within the corresponding follow-up time period for the matched patient.
cord blood HCT recipients with respect to median age at HCT (14 years versus 12 years versus 6 years, P < .001), time from diagnosis to transplant (HCT within 6 months of diagnosis in 53% versus 25% versus 35% patients, P < .001), use of TBI as part of conditioning (45% versus 68% versus 57%, P < .001), and history of chronic GVHD (37% versus 55% versus 43%, P < .001), respectively. There were also differences in GVHD prophylaxis regimens by donor source; for example, 2% of related donor, 15% of unrelated donor, and 0% of umbilical cord blood HCT recipients received ex vivo or in vivo T cell depletion to prevent GVHD.
d
<.01 45 (28) 14 (9) 100 (63) 1
191 (40) 53 (11) 234 (49) 0
102 (65) 27 (17) 8 (5)
229 (48) 68 (14) 28 (6)
21 (13)
153 (32) 0
93 (58) 42 (26) 25 (16)
268 (56) 112 (23) 98 (21)
23 11 60 40
37 12 194 135
<.01
2
.38
.01 (14) (7) (38) (25)
(8) (3) (41) (28) (Continued)
Risk Factor Analysis Table 2 shows the results of conditional logistic regression analysis. Risk factors independently associated with increased risks for AVN were older age at HCT (>5 years), female gender, and a history of chronic GVHD. The use of nonmyeloablative/reduced-intensity regimens for conditioning patients with malignant disease and the diagnosis of nonmalignant disease, regardless of conditioning intensity, were associated with statistically significantly lower risks for AVN when compared with the use of myeloablative regimens for conditioning patients with malignant diseases. The risks of AVN were similar in patients who received and did not receive TBI as part of myeloablative conditioning for malignant diseases. Interestingly, the use of unrelated donors was observed to be associated with a lower risk of AVN. We tested for and found no significant interactions between donor source and other variables (including GVHD) considered in multivariate analyses. DISCUSSION Our study identifies important risk factors for AVN in a large cohort of pediatric allogeneic HCT recipients and lays
590
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
Table 2 Risk Factors Identified to be Independently Associated with Post-Transplant AVN on Conditional Logistic Regression Analysis Risk Factor
Category*
Patients n (%)
Control Subjects n (%)
Hazard Ratio (95% CI)
P
Age
<5 yr 5-9 yr 10-21 yr Male Female Malignant disease, MA regimen, TBI Malignant disease, MA regimen, no TBI Malignant disease, NMA/RIC regimen Nonmalignant disease HLA-identical sibling Unrelated Umbilical cord blood None Acute GVHD only Chronic GVHD acute GVHD
5 14 128 71 76 97 22 7 21 23 100 24 41 23 83
148 103 201 271 181 215 64 25 148 36 321 95 175 63 214
1.00 3.40 19.83 1.00 1.65 1.00 .64 .31 0.30 1.00 .26 .41 1.00 1.45 1.88
<.01* .03 <.01
Gender Diagnosis and conditioning
Donor
GVHD
(3) (10) (87) (48) (52) (66) (15) (5) (14) (16) (68) (16) (28) (16) (56)
(33) (23) (44) (60) (40) (48) (14) (6) (33) (8) (71) (21) (39) (14) (47)
(1.17-9.89) (7.23-54.42) (1.05-2.58) (.32-1.26) (.11-.88) (0.14-0.63) (.12-.56) (.17-1.03) (.71-2.97) (1.09-3.23)
.03 <.01y .20 .03 <.01 <.01y <.01 .06 .08y .31 .02
CI, confidence interval; MA, myeloablative; NMA/RIC, nonmyeloablative/reduced intensity. * Patients receiving transplant from HLA-mismatched related donors (n ¼ 23), patients with unknown conditioning regimen intensity (n ¼ 2), and patients with unknown date of GVHD onset (n ¼ 14) were excluded from the risk factor analysis. y Overall P value.
the foundation for further research on screening and prevention of AVN in this population. We observed that risks increased with recipient age at transplantation. This has been reported by other relatively smaller studies in pediatric allogeneic HCT recipients [14,15] and suggests that children within the age group when rapid bone growth occurs are most susceptible to AVN. Other identified risk factors, including female gender and history of chronic GVHD, were similar to that previously reported by studies that have primarily included adults (Table 3) [3,5,9,16-18]. Of note, the observations from relatively smaller published studies are inconsistent when describing the gender-related differences in risk of AVN. Some studies have reported an increase risk among females, whereas others have demonstrated an increased risk in males, and some have failed to show a gender preference. The risks of AVN with nonmyeloablative/reducedintensity conditioning have not been previously described, and an important finding from our study is our observation that use of less-intense conditioning regimens is associated with lower risks for AVN. Compared with recipients of myeloablative conditioning, patients with malignant diseases receiving nonmyeloablative/reduced-intensity regimens had a 70% lower risk of developing AVN. Similarly, patients with nonmalignant diseases, who tend to receive conditioning of comparable intensity as reduced-intensity preparative regimens, had significantly lower risks of developing AVN. The decision to pursue myeloablative or nonmyeloablative/reduced-intensity conditioning for transplantation is frequently complex and has to take into account various factors such as diagnosis, disease status, performance status, and presence of comorbidities. Risk for complications is also a consideration, and our study will inform the decision process for patients who are at high risk for developing AVN. The lower risk of AVN in patients receiving unrelated donor HCT is in contrast to what has been previously reported in the literature [1]. We did not find any significant interactions between donor source and GVHD. This may be due to different exposures; for example, regimens for prevention or treatment of GVHD in unrelated donor HCT recipients may contain no corticosteroids or lesser doses of corticosteroids in combination with other agents. Indeed, a greater proportion of unrelated donor recipients had received T cell depletion for GVHD prophylaxis in our study.
It is also possible that ascertainment bias might partly explain this finding. Diagnostic testing for AVN is more likely to occur among patients who are followed up long term at transplant centers, where providers are more aware of this complication. Centers may preferentially follow more unrelated donor over HLA-identical sibling donor HCT recipients, especially because the former have a greater likelihood of developing chronic GVHD and need ongoing specialized care at the transplant center. Indeed, we had more control Table 3 Summary of Contemporary Studies Investigating Risk Factors for AVN after Allogeneic HCT in Children Reference AVN Patients Socie et al. (1997) [8] Kaste et al. (2004) [23] Faraci et al. (2006) [24] Leung et al. (2007) [14] Campbell et al. (2009) [7]
Median Risk Factors Age (yr)
25 77 (21 patients aged < 20 yr) 19 (all 10 children)
Female gender
43 (all children)
13*
Chronic GVHD, TBI, older age at HCT
20 (all children)
10
Female gender, older age at HCT
34 75 (46 patients aged < 35 yr; 73% allogeneic HCT) 29 McAvoy 66 (15 patients et al. (2010) aged 18 yr) [6] 44 (all 11 Sharma children) et al. (2012) [15] Present 160 (all 15 study children)
*
Age 16 yr at HCT, diagnosis of aplastic anemia or acute leukemia, acute GVHD, chronic GVHD
Mean age.
Male gender, chronic GVHD, exposure to calcineurin inhibitors and prednisone
Cumulative dose of corticosteroids
Age 10 yr at HCT, pre-HCT history of AVN
Older age at HCT, female gender, chronic GVHD, myeloablative conditioning regimen
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
subjects who were unrelated donor HCT recipients than HLAidentical sibling donor HCT recipients (control subjects included 11% related donor, 69% unrelated donor, and 21% umbilical cord blood recipients). Other risk factors that are not captured by the CIBMTR may also explain this observation and need further evaluation in future studies. Some limitations of our study have to be considered. AVN is frequently an under-diagnosed and hence under-reported late complication of transplantation. Exposure to calcineurin inhibitors and corticosteroids and the cumulative dose of corticosteroids have been shown to be an important risk factor for AVN [6,7,19]. Similarly, pretransplant chemotherapy, radiation, and corticosteroid exposures and posttransplant events such as endocrine late effects may also modulate the risks of post-transplant AVN. The CIBMTR does not routinely collect data on these variables, and these risk factors could not be evaluated in our study. Even though our study represents the largest analysis of AVN in pediatric HCT recipients, we were not able to evaluate some risk factors because of the small number of patients (eg, specific diagnoses where pre- and post-transplant exposures may be different). One such notable risk factor is the diagnosis of acute lymphoblastic leukemia, which is a well-established risk factor for AVN in pediatric patients who do not receive a transplant. Centers may have different practices for followup of patients post-transplantation and for diagnosis and screening for AVN. To account for this, we tried to identify control subjects from the same centers as patients or from other centers that had reported a case (85% of case-control pairs were from the same center). We excluded patients with a pre-HCT diagnosis of AVN. However, subclinical AVN may begin pretransplantation and manifest clinically posttransplantation. In conclusion, older age at HCT, female gender, exposure to myeloablative conditioning regimen, and chronic GVHD are risk factors for AVN after allogeneic HCT in children and adolescents. Screening strategies specifically for patients at high risk for developing AVN with early initiation of interventions (eg, physical therapy, limited surgery) might mitigate the morbidity associated with this complication. Our study, by highlighting important risk factors, lays the foundation for future research in this area. Clinicians taking care of pediatric allogeneic HCT survivors should maintain vigilance for this complication and have a high index of suspicion for AVN in patients with risk factors who present with new bone-related symptoms. Until more research is available, clinicians should follow published consensus guidelines for long-term follow-up that recommend early screening for AVN with magnetic resonance imaging in patients with such symptoms [20-22].
ACKNOWLEDGMENTS Financial disclosure: The CIBMTR is supported by Public Health Service Grant/Cooperative Agreement U24-CA76518 from the National Cancer Institute (NCI), the National Heart, Lung and Blood Institute (NHLBI), and the National Institute of Allergy and Infectious Diseases; Grant/Cooperative Agreement 5U01HL069294 from the NHLBI and NCI; contract HHSH234200637015 C from the Health Resources and Services Administration; 2 grants (N00014-06-1-0704 and N00014-08-1-0058) from the Office of Naval Research; and grants from Allos, Inc.; Amgen, Inc.; Angioblast; Anonymous donation to the Medical College of Wisconsin; Ariad; Be the
591
Match Foundation; Blue Cross and Blue Shield Association; Buchanan Family Foundation; CaridianBCT; Celgene Corporation; CellGenix, GmbH; Children’s Leukemia Research Association; Fresenius-Biotech North America, Inc.; Gamida Cell Teva Joint Venture Ltd.; Genentech, Inc.; Genzyme Corporation; GlaxoSmithKline; HistoGenetics, Inc.; Kiadis Pharma; The Leukemia & Lymphoma Society; The Medical College of Wisconsin; Merck & Co, Inc.; Millennium: The Takeda Oncology Co.; Milliman USA, Inc.; Miltenyi Biotec, Inc.; National Marrow Donor Program; Optum Healthcare Solutions, Inc.; Osiris Therapeutics, Inc.; Otsuka America Pharmaceutical, Inc.; RemedyMD; Sanofi; Seattle Genetics; Sigma-Tau Pharmaceuticals; Soligenix, Inc.; StemCyte, A Global Cord Blood Therapeutics Co.; Stemsoft Software, Inc.; Swedish Orphan Biovitrum; Tarix Pharmaceuticals; Teva Neuroscience, Inc.; THERAKOS, Inc.; and Wellpoint, Inc. The views expressed in this article do not reflect the official policy or position of the National Institutes of Health, the Department of the Navy, the Department of Defense, or any other agency of the U.S. Government. Conflict of interest statement: There are no conflicts of interest to report. Authorship statement: X.L., R.B., Z.W., and N.M. designed the study and analyzed the results. X.L. and N.M. wrote the manuscript. R.B. and Z.W. performed statistical analysis. All authors contributed to the study design, interpreted data, critically reviewed the manuscript, and approved the final version of the manuscript. REFERENCES 1. McClune B, Majhail NS, Flowers ME. Bone loss and avascular necrosis of bone after hematopoietic cell transplantation. Semin Hematol. 2012;49: 59-65. 2. Hautmann AH, Elad S, Lawitschka A, et al. Metabolic bone diseases in patients after allogeneic hematopoietic stem cell transplantation: report from the Consensus Conference on Clinical Practice in chronic graft-versus-host disease. Transplant Int. 2011;24:867-879. 3. Schulte CM, Beelen DW. Avascular osteonecrosis after allogeneic hematopoietic stem-cell transplantation: diagnosis and gender matter. Transplantation. 2004;78:1055-1063. 4. Fink JC, Leisenring WM, Sullivan KM, et al. Avascular necrosis following bone marrow transplantation: a case-control study. Bone. 1998;22:67-71. 5. Enright H, Haake R, Weisdorf D. Avascular necrosis of bone: a common serious complication of allogeneic bone marrow transplantation. Am J Med. 1990;89:733-738. 6. McAvoy S, Baker KS, Mulrooney D, et al. Corticosteroid dose as a risk factor for avascular necrosis of the bone after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2010;16:1231-1236. 7. Campbell S, Sun CL, Kurian S, et al. Predictors of avascular necrosis of bone in long-term survivors of hematopoietic cell transplantation. Cancer. 2009;115:4127-4135. 8. Socie G, Cahn JY, Carmelo J, et al. Avascular necrosis of bone after allogeneic bone marrow transplantation: analysis of risk factors for 4388 patients by the Societe Francaise de Greffe de Moelle (SFGM). Br J Haematol. 1997;97:865-870. 9. Tauchmanova L, De Rosa G, Serio B, et al. Avascular necrosis in longterm survivors after allogeneic or autologous stem cell transplantation: a single center experience and a review. Cancer. 2003;97: 2453-2461. 10. Gangji V, Hauzeur JP. Cellular-based therapy for osteonecrosis. Orthop Clin North Am. 2009;40:213-221. 11. Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning regimens: working definitions. Biol Blood Marrow Transplant. 2009;15:1628-1633. 12. Pasquini MC, Wang Z. Current use and outcome of hematopoietic stem cell transplantation: CIBMTR Summary Slides, 2012. Available at http:// www.cibmtr.org. 13. Weisdorf D, Spellman S, Haagenson M, et al. Classification of HLAmatching for retrospective analysis of unrelated donor transplantation: revised definitions to predict survival. Biol Blood Marrow Transplant. 2008;14:748-758. 14. Leung W, Ahn H, Rose SR, et al. A prospective cohort study of late sequelae of pediatric allogeneic hematopoietic stem cell transplantation. Medicine (Baltimore). 2007;86:215-224.
592
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
15. Sharma S, Leung WH, Deqing P, et al. Osteonecrosis in children after allogeneic hematopoietic cell transplantation: study of prevalence, risk factors and longitudinal changes using MR imaging. Bone Marrow Transplant. 2012;47:1067-1074. 16. Schulte CM, Beelen DW. Low pretransplant bone-mineral density and rapid bone loss do not increase risk for avascular osteonecrosis after allogeneic hematopoietic stem cell transplantation. Transplantation. 2005;79:1748-1755. 17. Torii YMD, Hasegawa YMD, Kubo TMD, et al. Osteonecrosis of the femoral head after allogeneic bone marrow transplantation. Clin Orthop Relat Res. 2001;382:124-132. 18. Wiesmann A, Pereira P, Bohm P, et al. Avascular necrosis of bone following allogeneic stem cell transplantation: MR screening and therapeutic options. Bone Marrow Transplant. 1998;22:565-569. 19. Jagasia S, Misfeldt A, Griffith M, et al. Age and total-body irradiation in addition to corticosteroid dose are important risk factors for avascular necrosis of the bone. Biol Blood Marrow Transplant. 2010;16: 1750-1751.
20. Majhail NS, Rizzo JD, Lee SJ, et al. Recommended screening and preventive practices for long-term survivors after hematopoietic cell transplantation. Bone Marrow Transplant. 2012;47:337-341. 21. Majhail NS, Rizzo JD, Lee SJ, et al. Recommended screening and preventive practices for long-term survivors after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2012;18:348-371. 22. Pulsipher MA, Skinner R, McDonald GB, et al. National Cancer Institute, National Heart, Lung and Blood Institute/Pediatric Blood and Marrow Transplantation Consortium First International Consensus Conference on late effects after pediatric hematopoietic cell transplantation: the need for pediatric-specific long-term follow-up guidelines. Biol Blood Marrow Transplant. 2012;18:334-347. 23. Kaste SC, Shidler TJ, Tong X, et al. Bone mineral density and osteonecrosis in survivors of childhood allogeneic bone marrow transplantation. Bone Marrow Transplant. 2004;33:435-441. 24. Faraci M, Calevo MG, Lanino E, et al. Osteonecrosis after allogeneic stem cell transplantation in childhood. A case-control study in Italy. Haematologica. 2006;91:1096-1099.
587
Avascular Necrosis of Bone after Allogeneic Hematopoietic Cell Transplantation in Children and Adolescents Xiaxin Li 1, Ruta Brazauskas 2, Zhiwei Wang 3, Amal Al-Seraihy 4, K. Scott Baker 5, Jean-Yves Cahn 6, Haydar A. Frangoul 7, James L. Gajewski 8, Gregory A. Hale 9, Jack W. Hsu 10, Rammurti T. Kamble 11, Hillard M. Lazarus 12, David I. Marks 13, Richard T. Maziarz 8, Bipin N. Savani 7, Ami J. Shah 14, Nirali Shah 15, Mohamed L. Sorror 5,16, William A. Wood 17, Navneet S. Majhail 18,19, * 1
Banner Children’s Specialists Pediatric Hematology/Oncology, Banner Health, Mesa, Arizona Division of Biostatistics, Center for International Blood and Marrow Transplant Research, Medical College of Wisconsin, Milwaukee, Wisconsin 3 Center for International Blood and Marrow Transplant Research, Medical College of Wisconsin, Milwaukee, Wisconsin 4 Hematopoietic Stem Cell Transplantation Program, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia 5 Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, Washington 6 Clinique Universitaire d’Hematologie, University Hospital, Grenoble, France 7 Pediatric Blood and Marrow Transplant Program, Vanderbilt University Medical Center, Nashville, Tennessee 8 Bone Marrow & Stem Cell Transplant Program, Oregon Health and Science University, Portland, Oregon 9 Division of Hematology and Oncology, All Children’s Hospital, St. Petersburg, Florida 10 Division of Hematology & Oncology, Shands HealthCare and University of Florida, Gainesville, Florida 11 Section of Hematology-Oncology, Baylor College of Medicine Center for Cell and Gene Therapy, Houston, Texas 12 Seidman Cancer Center, University Hospitals Case Medical Center, Cleveland, Ohio 13 BMT Unit, Bristol Children’s Hospital, Bristol, United Kingdom 14 Division of Hematology/Oncology, Department of Pediatrics, Mattel Children’s Hospital at UCLA, Los Angeles, California 15 National Cancer Institute, National Institutes of Health, Bethesda, Maryland 16 Division of Medical Oncology, Department of Medicine, University of Washington School of Medicine, Seattle, Washington 17 Division of Hematology and Oncology, University of North Carolina Hospitals, Chapel Hill, North Carolina 18 Center for International Blood and Marrow Transplant Research, Minneapolis, Minnesota 19 Blood & Marrow Transplant Program, Cleveland Clinic, Cleveland, Ohio 2
Article history: Received 18 November 2013 Accepted 23 December 2013 Key Words: Avascular necrosis Late complications Hematopoietic cell transplantation
a b s t r a c t We conducted a nested case-control study within a cohort of 6244 patients to assess risk factors for avascular necrosis (AVN) of bone in children and adolescents after allogeneic transplantation. Eligible patients were 21 years of age, received their first allogeneic transplant between 1990 and 2008 in the United States, and had survived 6 months from transplantation. Overall, 160 patients with AVN and 478 control subjects matched by year of transplant, length of follow-up and transplant center were identified. Patients and control subjects were confirmed via central review of radiology, pathology, and/or surgical procedure reports. Median time from transplant to diagnosis of AVN was 14 months. On conditional logistic regression, increasing age at transplant (5 years), female gender, and chronic graft-versus-host disease (GVHD) were significantly associated with increased risks of AVN. Compared with patients receiving myeloablative regimens for malignant diseases, lower risks of AVN were seen in patients with nonmalignant diseases and those who had received reduced-intensity conditioning regimens for malignant diseases. Children at high risk for AVN include those within the age group where rapid bone growth occurs as well as those who experience exposure to myeloablative conditioning regimens and immunosuppression after hematopoietic cell transplantation for the treatment of GVHD. More research is needed to determine whether screening strategies specifically for patients at high risk for developing AVN with early interventions may mitigate the morbidity associated with this complication. Ó 2014 American Society for Blood and Marrow Transplantation.
INTRODUCTION Avascular necrosis (AVN) of the bone is a debilitating late complication of allogeneic hematopoietic cell transplantation (HCT) that can be associated with significant morbidity [1,2]. The incidence and risk factors for AVN have been well described in adult transplant recipients with an estimated
Financial disclosure: See Acknowledgments on page 591. * Correspondence and reprint requests: Navneet S. Majhail, MD, MS, Blood & Marrow Transplant Program, Cleveland Clinic, 9500 Euclid Ave., R35, Cleveland, OH 44195. E-mail address: [email protected] (S. Kharbanda). 1083-8791/$ e see front matter Ó 2014 American Society for Blood and Marrow Transplantation. http://dx.doi.org/10.1016/j.bbmt.2013.12.567
cumulative incidence of 3% to 10% at 5 years after transplantation [1,3-9]. Graft-versus-host disease (GVHD), exposure to corticosteroids or calcineurin inhibitors, cumulative dose of corticosteroids, older age, female gender, and use of total body irradiation (TBI) as part of conditioning regimen have been identified as risk factors for AVN in adult HCT recipients. Although its pathogenesis is poorly understood, potential mechanisms for development of AVN include local vascular damage that leads to increased marrow edema and ischemia, ineffective osteoblastic repair processes due to metabolic factors, and mechanical stresses [1,10]. Large studies specifically focusing on evaluating risk factors for AVN in pediatric HCT survivors are lacking. Factors such as immaturity and ongoing growth of bones and endocrine dysfunction related to growth and sex hormones
588
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
are exclusive to children and may modulate the risks of AVN in a different way from adults. Hence, extrapolating findings from studies that have only included adults or have combined adults with children can be a challenge. Also, it is not known whether the relatively recent less toxic preparative regimens (nonmyeloablative and reduced-intensity conditioning) are associated with lower risks of AVN than conventional myeloablative regimens in this population. To better understand the risk factors for AVN after allogeneic HCT in children and adolescents, we conducted a case-control study using data from the Center for International Blood and Marrow Transplant Research (CIBMTR). We evaluated risk factors from both “older approaches” (myeloablative regimens, greater use of sibling donors) and “contemporary approaches” (nonmyeloablative/reduced intensity regimens, greater use of unrelated donors) in our analysis. METHODS Data Source The CIBMTR is a working group of more than 450 transplantation centers worldwide that contribute detailed data on HCTs to a Statistical Center at the Medical College of Wisconsin in Milwaukee and the National Marrow Donor Program in Minneapolis. Participating centers are required to report all transplants consecutively, and patients are followed longitudinally. Computerized checks for discrepancies, physicians’ review of submitted data, and onsite audits of participating centers ensure data quality. Data are collected before transplant, 100 days and 6 months after transplant, and annually thereafter, or until death. The follow-up research forms specifically inquire whether a recipient has developed AVN post-transplantation. Observational studies conducted by the CIBMTR are performed under the guidance of the Institutional Review Board of the National Marrow Donor Program and are in compliance with all applicable federal regulations pertaining to the protection of human research participants. Patients For our study, we selected first allogeneic HCT recipients aged 21 years at transplantation who had been reported to the CIBMTR between 1990 and 2008. Because screening and management practices for AVN can vary by region, we restricted our cohort to patients who had received their transplant at a center in the United States. We also limited our cohort to patients who had survived at least 6 months or more after transplantation because our analysis was focused on long-term HCT survivors and on transplant-related risk factors for AVN. Patients with any diagnosis and recipients of both myeloablative and reduced-intensity/nonmyeloablative regimens were eligible. Selection of Patients and Control Subjects Overall, 6244 patients met study eligibility criteria and were the basis for selection of patients and control subjects for our study. Patients were those who had a diagnosis of AVN reported on post-transplant follow-up. AVN of any joint was considered. For all cases identified as potential patients, we requested diagnostic and/or treatment information from centers to ascertain the diagnosis of AVN (eg, copies of radiologic investigations, pathology reports, or surgical operative notes). We excluded 2 patients from our analysis for whom we were not able to confirm the diagnosis of AVN from their transplant center. We established a pool of control subjects using eligible patients who had received their transplant at the same centers as patients and did not have a diagnosis of AVN reported to the CIBMTR. For each patient we chose a control subject that was matched by year of transplantation (1 year) and follow-up duration (follow-up post-transplant no less than the interval from HCT to onset of AVN for the corresponding patient). Control subjects were selected from the same center as patients, if available. If a control subject could not be identified for a patient from the same center, control subjects were selected from another center that had patients with AVN included in this study. Each patient was matched with up to 3 control subjects. For patients with several matched control subjects, 3 were selected randomly for the analysis. For each selected control subject, we contacted transplant centers and requested them to review medical records and confirm that the patient did not have AVN. On this review, 4 control subjects were identified to have AVN post-transplantation and were subsequently considered as patients. Control subjects were excluded from the analysis if they had a pre-HCT diagnosis of AVN (n ¼ 1) or if centers were not able to confirm the absence of AVN diagnosis (n ¼ 26). We identified 160 confirmed patients with AVN and 478 matched control subjects. Among these case-control pairs, 407 (85%) were matched
within the same center as the patient. One hundred fifty nine patients had 3 matched control subjects, and 1 patient had 1 matched control subject. Study Definitions and Statistical Analysis Conditioning regimens were defined as myeloablative, reduced intensity, and nonmyeloablative using established guidelines [11]. Because no clear guidelines exist for classifying conditioning regimens for nonmalignant diseases, these diseases were considered as a separate category when describing conditioning regimen intensity. Disease status for malignant diseases was assigned as early, intermediate, or advanced [12]. Early disease included acute myeloid leukemia or acute lymphoblastic leukemia in first complete remission, chronic myeloid leukemia in first chronic phase, myelodysplastic syndrome refractory anemia, or refractory anemia with ringed sideroblasts or unspecified myelodysplastic syndrome with <5% marrow blasts. Patients with acute myeloid leukemia or acute lymphoblastic leukemia in second or greater remission, chronic myeloid leukemia in second or greater chronic phase, or chronic myeloid leukemia in accelerated phase were classified as intermediate-risk disease. All other patients, including patients with lymphoma, were classified as advanced disease. The National Marrow Donor Program’s classification of HLA matching status was used for unrelated donor transplant recipients (well matched, partially matched, or mismatched) [13]. The goal of our case-control study was to assess potential risk factors for developing AVN in children and adolescents after allogeneic HCT. For comparing characteristics between patients and control subjects, we used the chi-square or Fisher’s test (as applicable) for categorical variables and Wilcoxon 2-sample test for continuous variables. To evaluate risk factors, we performed multivariable analyses using conditional logistic regression on all matched sets. The following variables were considered in this analysis: age at transplantation, gender, diagnosis, disease status, conditioning regimen intensity, dose of TBI, donor source, and history of GVHD before AVN. If feasible, categories with a small number of patients were combined with related categories. Patients receiving transplant from HLA-mismatched related donors (n ¼ 23), patients with unknown conditioning regimen intensity (n ¼ 2), and patients with unknown date of GVHD onset (n ¼ 14) were excluded from the risk factor analysis. All P-values are 2-sided. All analyses were carried out using SAS statistical software (SAS Institute Inc., Cary, NC).
RESULTS Characteristics of Patients and Control Subjects Table 1 shows the characteristics of the 160 AVN patients and 478 control subjects. The median age at transplantation was 15 years for patients and 8 years for control subjects. The primary diagnosis of nonmalignant disorder was higher in the control group (32%) compared with AVN patients (13%). A greater proportion of AVN patients had received TBI containing myeloablative regimen compared with control subjects (65% versus 48%). Related, unrelated, and umbilical cord blood donors were used in 21%, 64%, and 16% of AVN patients and 11%, 69%, and 21% of control subjects, respectively. Fiftysix percent of patients had a history of chronic GVHD before the onset of AVN compared with 49% in the control group. Among AVN patients, the median time from HCT to the onset of AVN was 14 months (range, <1 to 172 months). In 37% of patients AVN occurred within 1 year of HCT, in 59% AVN occurred 1 to 5 years after HCT, and in 4% it occurred more than 5 years after transplantation. Detailed information was available for 59 patients to completely characterize the extent of joint involvement by AVN. Among these patients, collectively 119 joints were affected by AVN with a median of 2 joints (range, 1 to 6). Femoral head (82%) was the most common site of involvement and was followed by the knee joint (78%), the vertebral column (12%), and the ankle joint (10%). AVN of the shoulder joint was rare (5% of patients). Pathologic fracture was the initial presentation of AVN in 3 patients. We also evaluated the characteristics of patients included in our study by donor source (related ¼ 83 patients [41% patients, 59% control subjects], unrelated ¼ 432 patients [23% patients, 77% control subjects], umbilical cord blood ¼ 123 [20% patients, 80% control subjects]). There were notable differences among related, unrelated, and umbilical
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
Table 1 Characteristics of AVN Patients and Their Control Subjects (Matched by Transplant Center, Year of Transplant, and Duration of Follow-Up) Characteristics
Patients
Control Subjects
Number of patients Number of centers Age at transplant, yr, median (range) Age at transplant, yr <5 5-9 10-14 15-21 Patient gender Male Female Lansky/Karnofsky score before transplant <90 90 Missing Patient race White Non-White Disease Acute myeloid leukemia Acute lymphoblastic leukemia Chronic myeloid leukemia Myelodysplastic syndrome Non-Hodgkin lymphoma Hodgkin lymphoma Severe aplastic anemia Inherited abnormality of erythrocyte differentiation SCID and other immune system disorders Inherited disorder of metabolism Histiocytic disorders Disease risk before transplant Early Intermediate Advanced Nonmalignant disease Unknown Interval from diagnosis to transplant, mo <6 6-11 12 Unknown Year of transplant* 1990-1994 1995-1999 2000-2004 2005-2008 Total TBI dose, cGy No TBI <1200 cGy 1200 cGy Missing Conditioning regimen intensity Myeloablative (with TBI) Myeloablative (no TBI) Nonmyeloablative/reduced intensity Nonmalignant diseases Missing Graft type Bone marrow Peripheral blood bone marrow Umbilical cord bloody Donor type HLA-matched sibling Other related Well-matched unrelated Partially/mismatched unrelated
160 54 15 (2-21)
478 52 8 (<1-21) <.01 <.01 154 (32) 112 (23) 94 (20) 118 (25) <.01 289 (60) 189 (40) .99
5 16 56 83
(3) (10) (35) (52)
78 (49) 82 (51)
22 (15) 121 (85) 17
66 (15) 362 (85) 50
128 (80) 32 (20)
385 (81) 93 (19) <.01
(29) (36) (9) (11) (1) (1) (7) (3)
119 128 31 32 15 0 42 36
(25) (27) (6) (7) (3) (9) (8)
2 (1)
42 (9)
2 (1) 1 (1)
20 (4) 13 (3) <.01
62 55 22 21
(39) (34) (14) (13)
0
130 147 43 153 5
(27) (31) (9) (32) (1) .02
60 31 66 3
(38) (19) (41) (2)
135 100 205 38
(28) (21) (43) (8)
5 13 60 82
(3) (8) (38) (51)
19 40 215 204
(4) (8) (45) (43)
Table 1 (continued) Characteristics
Patients
P
.88
47 57 14 18 2 1 11 5
589
Unrelated, match unknown Umbilical cord blood GVHD prophylaxis FK506 þ MMF others FK506 þ MTX others (except MMF) FK506 others (except MTX, MMF) CSA þ MMF others CSA þ MTX others (except MMF) CSA others (except MTX, MMF) T cell depletion Other/unknown History of GVHDz No GVHD Acute GVHD only Chronic acute GVHD Interval from transplant to AVN, mo, median (range) Follow-up of survivors, mo, median (range)*
Control Subjects
1 (1) 25 (16)
2 (<1) 98 (21)
13 (8) 28 (18)
26 (5) 86 (18)
P
.51
9 13 58 24 10 5
(6) (8) (36) (15) (6) (3)
18 41 161 72 56 18
(4) (9) (34) (15) (12) (4)
43 28 89 14
(27) 179 (37) (18) 66 (14) (56) 233 (49) (<1-172) d
.05
61 (11-194)
63 (7-225)
SCID, severe combined immunodeficiency; MMF, mycophenolate mofetil; MTX, methotrexate; CSA, cyclosporine. Values are total number of incidences with percents in parentheses, unless otherwise noted. * Variable used for matching patients and control subjects. y Includes 6 related umbilical cord blood transplants (1 patient, 5 control subjects). z History of GVHD before AVN for patients; for control subjects, any history of GVHD within the corresponding follow-up time period for the matched patient.
cord blood HCT recipients with respect to median age at HCT (14 years versus 12 years versus 6 years, P < .001), time from diagnosis to transplant (HCT within 6 months of diagnosis in 53% versus 25% versus 35% patients, P < .001), use of TBI as part of conditioning (45% versus 68% versus 57%, P < .001), and history of chronic GVHD (37% versus 55% versus 43%, P < .001), respectively. There were also differences in GVHD prophylaxis regimens by donor source; for example, 2% of related donor, 15% of unrelated donor, and 0% of umbilical cord blood HCT recipients received ex vivo or in vivo T cell depletion to prevent GVHD.
d
<.01 45 (28) 14 (9) 100 (63) 1
191 (40) 53 (11) 234 (49) 0
102 (65) 27 (17) 8 (5)
229 (48) 68 (14) 28 (6)
21 (13)
153 (32) 0
93 (58) 42 (26) 25 (16)
268 (56) 112 (23) 98 (21)
23 11 60 40
37 12 194 135
<.01
2
.38
.01 (14) (7) (38) (25)
(8) (3) (41) (28) (Continued)
Risk Factor Analysis Table 2 shows the results of conditional logistic regression analysis. Risk factors independently associated with increased risks for AVN were older age at HCT (>5 years), female gender, and a history of chronic GVHD. The use of nonmyeloablative/reduced-intensity regimens for conditioning patients with malignant disease and the diagnosis of nonmalignant disease, regardless of conditioning intensity, were associated with statistically significantly lower risks for AVN when compared with the use of myeloablative regimens for conditioning patients with malignant diseases. The risks of AVN were similar in patients who received and did not receive TBI as part of myeloablative conditioning for malignant diseases. Interestingly, the use of unrelated donors was observed to be associated with a lower risk of AVN. We tested for and found no significant interactions between donor source and other variables (including GVHD) considered in multivariate analyses. DISCUSSION Our study identifies important risk factors for AVN in a large cohort of pediatric allogeneic HCT recipients and lays
590
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
Table 2 Risk Factors Identified to be Independently Associated with Post-Transplant AVN on Conditional Logistic Regression Analysis Risk Factor
Category*
Patients n (%)
Control Subjects n (%)
Hazard Ratio (95% CI)
P
Age
<5 yr 5-9 yr 10-21 yr Male Female Malignant disease, MA regimen, TBI Malignant disease, MA regimen, no TBI Malignant disease, NMA/RIC regimen Nonmalignant disease HLA-identical sibling Unrelated Umbilical cord blood None Acute GVHD only Chronic GVHD acute GVHD
5 14 128 71 76 97 22 7 21 23 100 24 41 23 83
148 103 201 271 181 215 64 25 148 36 321 95 175 63 214
1.00 3.40 19.83 1.00 1.65 1.00 .64 .31 0.30 1.00 .26 .41 1.00 1.45 1.88
<.01* .03 <.01
Gender Diagnosis and conditioning
Donor
GVHD
(3) (10) (87) (48) (52) (66) (15) (5) (14) (16) (68) (16) (28) (16) (56)
(33) (23) (44) (60) (40) (48) (14) (6) (33) (8) (71) (21) (39) (14) (47)
(1.17-9.89) (7.23-54.42) (1.05-2.58) (.32-1.26) (.11-.88) (0.14-0.63) (.12-.56) (.17-1.03) (.71-2.97) (1.09-3.23)
.03 <.01y .20 .03 <.01 <.01y <.01 .06 .08y .31 .02
CI, confidence interval; MA, myeloablative; NMA/RIC, nonmyeloablative/reduced intensity. * Patients receiving transplant from HLA-mismatched related donors (n ¼ 23), patients with unknown conditioning regimen intensity (n ¼ 2), and patients with unknown date of GVHD onset (n ¼ 14) were excluded from the risk factor analysis. y Overall P value.
the foundation for further research on screening and prevention of AVN in this population. We observed that risks increased with recipient age at transplantation. This has been reported by other relatively smaller studies in pediatric allogeneic HCT recipients [14,15] and suggests that children within the age group when rapid bone growth occurs are most susceptible to AVN. Other identified risk factors, including female gender and history of chronic GVHD, were similar to that previously reported by studies that have primarily included adults (Table 3) [3,5,9,16-18]. Of note, the observations from relatively smaller published studies are inconsistent when describing the gender-related differences in risk of AVN. Some studies have reported an increase risk among females, whereas others have demonstrated an increased risk in males, and some have failed to show a gender preference. The risks of AVN with nonmyeloablative/reducedintensity conditioning have not been previously described, and an important finding from our study is our observation that use of less-intense conditioning regimens is associated with lower risks for AVN. Compared with recipients of myeloablative conditioning, patients with malignant diseases receiving nonmyeloablative/reduced-intensity regimens had a 70% lower risk of developing AVN. Similarly, patients with nonmalignant diseases, who tend to receive conditioning of comparable intensity as reduced-intensity preparative regimens, had significantly lower risks of developing AVN. The decision to pursue myeloablative or nonmyeloablative/reduced-intensity conditioning for transplantation is frequently complex and has to take into account various factors such as diagnosis, disease status, performance status, and presence of comorbidities. Risk for complications is also a consideration, and our study will inform the decision process for patients who are at high risk for developing AVN. The lower risk of AVN in patients receiving unrelated donor HCT is in contrast to what has been previously reported in the literature [1]. We did not find any significant interactions between donor source and GVHD. This may be due to different exposures; for example, regimens for prevention or treatment of GVHD in unrelated donor HCT recipients may contain no corticosteroids or lesser doses of corticosteroids in combination with other agents. Indeed, a greater proportion of unrelated donor recipients had received T cell depletion for GVHD prophylaxis in our study.
It is also possible that ascertainment bias might partly explain this finding. Diagnostic testing for AVN is more likely to occur among patients who are followed up long term at transplant centers, where providers are more aware of this complication. Centers may preferentially follow more unrelated donor over HLA-identical sibling donor HCT recipients, especially because the former have a greater likelihood of developing chronic GVHD and need ongoing specialized care at the transplant center. Indeed, we had more control Table 3 Summary of Contemporary Studies Investigating Risk Factors for AVN after Allogeneic HCT in Children Reference AVN Patients Socie et al. (1997) [8] Kaste et al. (2004) [23] Faraci et al. (2006) [24] Leung et al. (2007) [14] Campbell et al. (2009) [7]
Median Risk Factors Age (yr)
25 77 (21 patients aged < 20 yr) 19 (all 10 children)
Female gender
43 (all children)
13*
Chronic GVHD, TBI, older age at HCT
20 (all children)
10
Female gender, older age at HCT
34 75 (46 patients aged < 35 yr; 73% allogeneic HCT) 29 McAvoy 66 (15 patients et al. (2010) aged 18 yr) [6] 44 (all 11 Sharma children) et al. (2012) [15] Present 160 (all 15 study children)
*
Age 16 yr at HCT, diagnosis of aplastic anemia or acute leukemia, acute GVHD, chronic GVHD
Mean age.
Male gender, chronic GVHD, exposure to calcineurin inhibitors and prednisone
Cumulative dose of corticosteroids
Age 10 yr at HCT, pre-HCT history of AVN
Older age at HCT, female gender, chronic GVHD, myeloablative conditioning regimen
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
subjects who were unrelated donor HCT recipients than HLAidentical sibling donor HCT recipients (control subjects included 11% related donor, 69% unrelated donor, and 21% umbilical cord blood recipients). Other risk factors that are not captured by the CIBMTR may also explain this observation and need further evaluation in future studies. Some limitations of our study have to be considered. AVN is frequently an under-diagnosed and hence under-reported late complication of transplantation. Exposure to calcineurin inhibitors and corticosteroids and the cumulative dose of corticosteroids have been shown to be an important risk factor for AVN [6,7,19]. Similarly, pretransplant chemotherapy, radiation, and corticosteroid exposures and posttransplant events such as endocrine late effects may also modulate the risks of post-transplant AVN. The CIBMTR does not routinely collect data on these variables, and these risk factors could not be evaluated in our study. Even though our study represents the largest analysis of AVN in pediatric HCT recipients, we were not able to evaluate some risk factors because of the small number of patients (eg, specific diagnoses where pre- and post-transplant exposures may be different). One such notable risk factor is the diagnosis of acute lymphoblastic leukemia, which is a well-established risk factor for AVN in pediatric patients who do not receive a transplant. Centers may have different practices for followup of patients post-transplantation and for diagnosis and screening for AVN. To account for this, we tried to identify control subjects from the same centers as patients or from other centers that had reported a case (85% of case-control pairs were from the same center). We excluded patients with a pre-HCT diagnosis of AVN. However, subclinical AVN may begin pretransplantation and manifest clinically posttransplantation. In conclusion, older age at HCT, female gender, exposure to myeloablative conditioning regimen, and chronic GVHD are risk factors for AVN after allogeneic HCT in children and adolescents. Screening strategies specifically for patients at high risk for developing AVN with early initiation of interventions (eg, physical therapy, limited surgery) might mitigate the morbidity associated with this complication. Our study, by highlighting important risk factors, lays the foundation for future research in this area. Clinicians taking care of pediatric allogeneic HCT survivors should maintain vigilance for this complication and have a high index of suspicion for AVN in patients with risk factors who present with new bone-related symptoms. Until more research is available, clinicians should follow published consensus guidelines for long-term follow-up that recommend early screening for AVN with magnetic resonance imaging in patients with such symptoms [20-22].
ACKNOWLEDGMENTS Financial disclosure: The CIBMTR is supported by Public Health Service Grant/Cooperative Agreement U24-CA76518 from the National Cancer Institute (NCI), the National Heart, Lung and Blood Institute (NHLBI), and the National Institute of Allergy and Infectious Diseases; Grant/Cooperative Agreement 5U01HL069294 from the NHLBI and NCI; contract HHSH234200637015 C from the Health Resources and Services Administration; 2 grants (N00014-06-1-0704 and N00014-08-1-0058) from the Office of Naval Research; and grants from Allos, Inc.; Amgen, Inc.; Angioblast; Anonymous donation to the Medical College of Wisconsin; Ariad; Be the
591
Match Foundation; Blue Cross and Blue Shield Association; Buchanan Family Foundation; CaridianBCT; Celgene Corporation; CellGenix, GmbH; Children’s Leukemia Research Association; Fresenius-Biotech North America, Inc.; Gamida Cell Teva Joint Venture Ltd.; Genentech, Inc.; Genzyme Corporation; GlaxoSmithKline; HistoGenetics, Inc.; Kiadis Pharma; The Leukemia & Lymphoma Society; The Medical College of Wisconsin; Merck & Co, Inc.; Millennium: The Takeda Oncology Co.; Milliman USA, Inc.; Miltenyi Biotec, Inc.; National Marrow Donor Program; Optum Healthcare Solutions, Inc.; Osiris Therapeutics, Inc.; Otsuka America Pharmaceutical, Inc.; RemedyMD; Sanofi; Seattle Genetics; Sigma-Tau Pharmaceuticals; Soligenix, Inc.; StemCyte, A Global Cord Blood Therapeutics Co.; Stemsoft Software, Inc.; Swedish Orphan Biovitrum; Tarix Pharmaceuticals; Teva Neuroscience, Inc.; THERAKOS, Inc.; and Wellpoint, Inc. The views expressed in this article do not reflect the official policy or position of the National Institutes of Health, the Department of the Navy, the Department of Defense, or any other agency of the U.S. Government. Conflict of interest statement: There are no conflicts of interest to report. Authorship statement: X.L., R.B., Z.W., and N.M. designed the study and analyzed the results. X.L. and N.M. wrote the manuscript. R.B. and Z.W. performed statistical analysis. All authors contributed to the study design, interpreted data, critically reviewed the manuscript, and approved the final version of the manuscript. REFERENCES 1. McClune B, Majhail NS, Flowers ME. Bone loss and avascular necrosis of bone after hematopoietic cell transplantation. Semin Hematol. 2012;49: 59-65. 2. Hautmann AH, Elad S, Lawitschka A, et al. Metabolic bone diseases in patients after allogeneic hematopoietic stem cell transplantation: report from the Consensus Conference on Clinical Practice in chronic graft-versus-host disease. Transplant Int. 2011;24:867-879. 3. Schulte CM, Beelen DW. Avascular osteonecrosis after allogeneic hematopoietic stem-cell transplantation: diagnosis and gender matter. Transplantation. 2004;78:1055-1063. 4. Fink JC, Leisenring WM, Sullivan KM, et al. Avascular necrosis following bone marrow transplantation: a case-control study. Bone. 1998;22:67-71. 5. Enright H, Haake R, Weisdorf D. Avascular necrosis of bone: a common serious complication of allogeneic bone marrow transplantation. Am J Med. 1990;89:733-738. 6. McAvoy S, Baker KS, Mulrooney D, et al. Corticosteroid dose as a risk factor for avascular necrosis of the bone after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2010;16:1231-1236. 7. Campbell S, Sun CL, Kurian S, et al. Predictors of avascular necrosis of bone in long-term survivors of hematopoietic cell transplantation. Cancer. 2009;115:4127-4135. 8. Socie G, Cahn JY, Carmelo J, et al. Avascular necrosis of bone after allogeneic bone marrow transplantation: analysis of risk factors for 4388 patients by the Societe Francaise de Greffe de Moelle (SFGM). Br J Haematol. 1997;97:865-870. 9. Tauchmanova L, De Rosa G, Serio B, et al. Avascular necrosis in longterm survivors after allogeneic or autologous stem cell transplantation: a single center experience and a review. Cancer. 2003;97: 2453-2461. 10. Gangji V, Hauzeur JP. Cellular-based therapy for osteonecrosis. Orthop Clin North Am. 2009;40:213-221. 11. Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning regimens: working definitions. Biol Blood Marrow Transplant. 2009;15:1628-1633. 12. Pasquini MC, Wang Z. Current use and outcome of hematopoietic stem cell transplantation: CIBMTR Summary Slides, 2012. Available at http:// www.cibmtr.org. 13. Weisdorf D, Spellman S, Haagenson M, et al. Classification of HLAmatching for retrospective analysis of unrelated donor transplantation: revised definitions to predict survival. Biol Blood Marrow Transplant. 2008;14:748-758. 14. Leung W, Ahn H, Rose SR, et al. A prospective cohort study of late sequelae of pediatric allogeneic hematopoietic stem cell transplantation. Medicine (Baltimore). 2007;86:215-224.
592
X. Li et al. / Biol Blood Marrow Transplant 20 (2014) 577e592
15. Sharma S, Leung WH, Deqing P, et al. Osteonecrosis in children after allogeneic hematopoietic cell transplantation: study of prevalence, risk factors and longitudinal changes using MR imaging. Bone Marrow Transplant. 2012;47:1067-1074. 16. Schulte CM, Beelen DW. Low pretransplant bone-mineral density and rapid bone loss do not increase risk for avascular osteonecrosis after allogeneic hematopoietic stem cell transplantation. Transplantation. 2005;79:1748-1755. 17. Torii YMD, Hasegawa YMD, Kubo TMD, et al. Osteonecrosis of the femoral head after allogeneic bone marrow transplantation. Clin Orthop Relat Res. 2001;382:124-132. 18. Wiesmann A, Pereira P, Bohm P, et al. Avascular necrosis of bone following allogeneic stem cell transplantation: MR screening and therapeutic options. Bone Marrow Transplant. 1998;22:565-569. 19. Jagasia S, Misfeldt A, Griffith M, et al. Age and total-body irradiation in addition to corticosteroid dose are important risk factors for avascular necrosis of the bone. Biol Blood Marrow Transplant. 2010;16: 1750-1751.
20. Majhail NS, Rizzo JD, Lee SJ, et al. Recommended screening and preventive practices for long-term survivors after hematopoietic cell transplantation. Bone Marrow Transplant. 2012;47:337-341. 21. Majhail NS, Rizzo JD, Lee SJ, et al. Recommended screening and preventive practices for long-term survivors after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2012;18:348-371. 22. Pulsipher MA, Skinner R, McDonald GB, et al. National Cancer Institute, National Heart, Lung and Blood Institute/Pediatric Blood and Marrow Transplantation Consortium First International Consensus Conference on late effects after pediatric hematopoietic cell transplantation: the need for pediatric-specific long-term follow-up guidelines. Biol Blood Marrow Transplant. 2012;18:334-347. 23. Kaste SC, Shidler TJ, Tong X, et al. Bone mineral density and osteonecrosis in survivors of childhood allogeneic bone marrow transplantation. Bone Marrow Transplant. 2004;33:435-441. 24. Faraci M, Calevo MG, Lanino E, et al. Osteonecrosis after allogeneic stem cell transplantation in childhood. A case-control study in Italy. Haematologica. 2006;91:1096-1099.