Association between conditioning intensity and height growth after allogeneic hematopoietic stem cell transplantation in children

Abstract Background The present study aimed to examine the association between the conditioning intensity and height growth in pediatric patients who underwent allogeneic hematopoietic stem cell transplantation (allo‐HSCT). Methods We reviewed the clinical records of 89 children with malignant diseases who underwent initial allo‐HSCT between 2003 and 2021. Height measurements were standardized using standard height charts prepared by the Japanese Society for Pediatric Endocrinology to calculate standard deviation score (SDS). We defined short stature as a height SDS less than −2.0 in that reference. Myeloablative conditioning (MAC) comprised total‐body irradiation at more than 8 Gy and busulfan administration at more than 8 mg/kg (more than 280 mg/m2). Other conditioning regimens were defined as reduced intensity conditioning (RIC). Results A total of 58 patients underwent allo‐HSCT with MAC, and 31 patients received allo‐HSCT with RIC. There were significant differences in the height SDS at 2 and 3 years after allo‐HSCT between MAC and RIC group (−1.33 ± 1.20 vs. −0.76 ± 1.12, p = 0.047, −1.55 ± 1.28 vs. −0.75 ± 1.11, p = 0.022, respectively). Multivariate logistic regression analysis with the adjustments for potential confounding factors of patients less than 10 years of age at allo‐HSCT and chronic graft‐versus host disease demonstrated that MAC regimen was associated with a markedly increased risk of a short stature at 3 years after allo‐HSCT (adjusted odds ratio, 5.61; 95% confidence interval, 1.07–29.4; p = 0.041). Conclusion The intensity of conditioning regimen may be associated with short statures after allo‐HSCT.


| INTRODUCTION
][5] Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is one of the effective therapies for high-risk hematological diseases, some solid tumors, and primary immunodeficiency. 6High-dose chemotherapy and total body irradiation (TBI), which have been illustrated to increase the antitumor effect, are often used as conditioning regimens for allo-HSCT. 7][10] Although long-term survival is expected by many pediatric patients who receive allo-HSCT, they have a substantially increased burden of serious late effects. 11,12rowth disorder is one of the most common late effects of allo-HSCT in children and often influences the final adult height among CCS. 13,14Growth disorder in patients who have received allo-HSCT is caused by the growth hormone (GH) deficiency, which is induced by the irradiation to the pituitary-hypothalamus system. 15he irradiation to the body is also associated with impaired bone growth. 16The irradiation as a conditioning regimen or systemic steroids for the treatment of graftversus host disease (GVHD) directly inhibit the bone growth. 174][15][16][17][18] Therefore, there was a need for the establishment of the optimal strategy for allo-HSCT to reduce the endocrine complications and improve the QOL of CCS who have received allo-HSCT.
0][21] The aim of RIC is to decrease short-and long-term complications induced by myeloablative conditioning (MAC) with promoting adequate engraftment and stable donor-derived hematopoiesis. 22,23owever, few previous reports have investigated the effect of conditioning intensity on growth after allo-HSCT in children. 24n the present study, we examined the association between the conditioning intensity and height growth in the pediatric patients who received allo-HSCT.

| Study population
In this study, a total of 120 consecutive patients who underwent initial allo-HSCT for malignant disease at our institute between April 1, 2003, and December 31, 2021 were included.Patients who relapsed (n = 11) or died (n = 9) within 1 year of allo-HSCT, or experienced primary graft failure (n = 10) were excluded.A patient with trisomy 21 was also excluded because of the innate short stature compared to patients without trisomy 21.Finally, data on 89 patients were analyzed retrospectively.We reviewed the medical records to obtain the patients' characteristics including clinical information.The last clinical follow-up date in the present study was on December 31, 2022.
The present study was approved by the ethics review committees of Kobe Children's Hospital (R4-142).

| Conditioning regimen
The conditioning regimen was determined by the physician's intent considering patient's disease, disease status, and donor type.MAC comprised TBI at more than 8 Gy and busulfan administration at more than 8 mg/kg (more than 280 mg/m 2 ). 25 Other conditioning regimens were defined as RIC.

| Height evaluations
Height was routinely measured at the time of allo-HSCT and then annually until the age of 18 years or final height, as part of medical examinations.Final height was defined as the tallest height measured at age 18 years or older, and when height velocity was inferior to 1 cm per year for consecutive 2 years.Endocrinological disorders after allo-HSCT were also extracted from medical records.The height standard deviation score (SDS) was calculated using the growth charts with the reference to Japanese children's age and sex prepared by the Japanese Society for Pediatric Endocrinology.The growth charts consisted of the datasets obtained from the Japanese national survey in the year 2000, including 18,550 anthropometric measurements of babies and infants surveyed by the Ministry of Health, Labour and Welfare and 695,600 anthropometric measurements of school children surveyed by the Ministry of Education, Culture, Sports, Science, and Technology, as established by the lambda-mu-sigma method. 26,27hort stature was defined as a height SDS less than −2.0 for individuals of the same sex and chronologic age in that reference.

| Statistical analysis
In our study, the primary outcome measure was the height SDS.The secondary outcome measure was the development of short stature.The patients who relapsed or died of any cause more than 1 year after allo-HSCT were censored.The chi-squared or Fisher's exact test was used to compare categorical variables in patients' characteristics between the MAC and RIC groups.The unpaired t-test or Mann-Whitney U test was also used to compare continuous variables in patients' characteristics between the MAC and RIC groups.By Spearman's rank correlation test, the correlation between height SDS and TBI dose or age at allo-HSCT was assessed.A multivariate logistic regression model was used to calculate the adjusted odds ratio (OR) for short stature at 3 years after allo-HSCT with a 95% confidence interval (CI).The development of chronic GVHD (cGVHD) or systemic steroids for the treatment of cGVHD, and age less than 10 years at allo-HSCT were included in the model as potential confounding factors.For all models, the number of examined covariates was determined according to the number of outcome events, with 10 events required for one covariate.Reported p-values of less than 0.05 were considered statistically significant.All statistical analyses were conducted using EZR (Saitama Medical Center, Jichi Medical University), which is a modified version of R commander (The R Foundation for Statistical Computing). 28

| Clinical characteristics
The median follow-up period after allo-HSCT was 2338 days (range, 375-6835 days).Median age at allo-HSCT was 8.3 years (range, 0.5-16.2years).The primary diagnoses were as follows: acute lymphoblastic leukemia (ALL, n = 39), acute myeloid leukemia (AMZL, n = 20), myelodysplastic syndrome/myeloproliferative disease (n = 17), malignant lymphoma (n = 4), and solid tumors (n = 9) (Table 1).ALL was the most common primary disease in the MAC group (58.6%), but not in the RIC group (16.1%).There was a significant difference between the MAC and RIC groups in the distribution of the primary disease (p < 0.001).There was also a significant difference in donor source between the MAC and RIC groups (p = 0.041).There were no statistically significant differences in gender, donor type, HLA compatibility, use of TBI, craniospinal irradiation (CSI) prior to allo-HSCT, acute GVHD, chronic GVHD (cGVHD), or use of systemic steroids for the treatment of cGVHD between the MAC and RIC groups.Six patients, all in the MAC group, were treated with GH or thyroid hormone (TH) replacement after allo-HSCT.However, there was no significant difference in the number of patients who were treated with GH or TH replacement after allo-HSCT between the MAC and RIC groups (p = 0.088).
The comparison between male and female patients revealed no significant differences in height SDS on allo-HSCT, and 1, 2, 3, 4, and 5 years after allo-HSCT (Table S1).
Patients with and without ALL were also compared, because corticosteroids are key drugs for ALL and often suppress GH secretion and the function of insulin-like growth factor (IGF)-I and other factors. 29No significant differences were found between the ALL and non-ALL groups in height SDS on allo-HSCT, and 1, 2, 3, 4, and 5 years after allo-HSCT (Table S2).

| Effects of age at allo-HSCT and TBI dose on height SDS
Table 2 shows the correlation between the height SDS at 2 and 3 years after allo-HSCT and age at allo-HSCT.There was no statistically significant correlation between the height SDS at 2 and 3 years after allo-HSCT and age at allo-HSCT and (r = 0.030, p = 0.81 and r = 0.027, p = 0.85, respectively).
The correlation between the height SDS at 2 and 3 years after allo-HSCT and TBI dose are also shown in Table 2.The height SDS at 2 and 3 years after allo-HSCT did not show statistically significant correlations with total doses of TBI (r = 0.13, p = 0.23 and r = 0.020, p = 0.88, respectively).T A B L E 1 Clinical characteristics of the studied cohort.

| Association between conditioning intensity and the development of short stature
after allo-HSCT.The MAC regimen was associated with a marked increase in risk of the short stature at 3 years after allo-HSCT (adjusted OR, 5.61; 95% CI, 1.07-29.4).Age less than 10 years at allo-HSCT was not associated with an increased risk of the short stature at 3 years after allo-HSCT (adjusted OR, 0.93; 95% CI, 0.25-3.55).The development of cGVHD was not associated with an increased risk of the short stature at 3 years after allo-HSCT (adjusted OR, 1.14; 95% CI, 0.27-4.76).The use of systemic steroids for the treatment of cGVHD demonstrated no association with an increased risk of short stature 3 years after allo-HSCT (adjusted OR, 1.77; 95% CI, 0.25-12.5)(Table S3).2).A significant reduction in the height SDS on allo-HSCT, one and 2 years after allo-HSCT in the group with TBI of ≤8 Gy, compared with the group with TBI of >8 Gy (p = 0.001, 0.002, and 0.032, respectively) (Figure 2A-C).

| DISCUSSION
Herein, we demonstrate that MAC regimen increased the risk of a short stature after allo-HSCT in the present study.Conditioning intensity may be associated with growth disorder for pediatric patients after allo-HSCT.Few studies investigated the effect of conditioning intensity on short stature. 24Myers, et al. reported that the even RIC regimen was associated with significant late endocrine effects in children and young adults. 24This discrepancy with our results may be attributable to the cohort bias.Most subjects in the study by Myers et al. had non-malignant diseases including immune disorders, metabolic, and genetic disorders, in contrast to the present study.Furthermore, some cases of these disorders are associated with short stature. 24lthough there were significant differences in the height SDS at 2 and 3 years after allo-HSCT between the MAC group and RIC group, the significant differences in the height SDS at 4 and 5 years after allo-HSCT between the two groups were not observed.This result may be attributable to the effect of small number of patients with height measurements available at 4 and 5 years after allo-HSCT.
Growth disorder after TBI as a conditioning regimen has been well documented. 15,30,31Patients who receive the irradiation to central nervous system (CNS) prior to allo-HSCT develop in GH deficiency more frequently than those who do not receive irradiation to the CNS. 15,31In our cohort, three patients received CSI.One of these patients underwent CSI followed by allo-HSCT at 7 years of age with MAC regimen containing TBI at a dose of 12 Gy and required the GH therapy after allo-HSCT.
Couto-Silva et al. found that the prevalence of short stature was greater in the patients who received TBI at a younger age. 32Patients who received TBI before 4 years of age developed skeletal lesions leading to substantial height loss, despite GH treatment. 32The previous study comparing 12 Gy of TBI against a total Bu dose 16 mg/kg or 480 mg/m 2 as a conditioning regimen after allo-HSCT revealed that the patients who received TBI showed a significant decrease in their height SDS compared to Bu. 30 In the present study, there were no correlations between height SDS and age at allo-HSCT or the total dose of TBI.In the MAC group, median age at allo-HSCT of patients who received >8 Gy of TBI was 9.3 years (range, 1.8-16.2years), while the median age at allo-HSCT among the patients who received ≤8 Gy of TBI was 2.0 years (range, 6 months-14.9years).Between the two MAC regimens, there was a significant difference in age at allo-HSCT (p = 0.005).This was predominantly due to physician's bias such as the avoidance of TBI containing regimens in children who underwent allo-HSCT at a younger age.As a matter of fact, a significant reduction was observed in the height SDS at allo-HSCT in the group with TBI of <8 Gy (Figure 2).
Corticosteroids are key drugs for ALL and often suppress GH secretion and the function of IGF-I and other factors. 29Moreover, corticosteroids are used for the treatment of cGVHD and prolonged use is expected to be at high risk for growth.However, there were no significant differences between the ALL and non-ALL groups in height SDS on allo-HSCT, and 1, 2, 3, 4, and 5 years after allo-HSCT in the present study (Table S2).The use of systemic steroids for the treatment of cGVHD revealed no association with an increased risk of short stature 3 years after allo-HSCT (Table S3).A previous report by Sanders et al revealed that patients with ALL or Non-Hodgkin lymphoma (NHL) had lower average height SDS at the diagnosis of GH deficiency and were less effective for GH replacement therapy, compared with non-ALL such as AML. 14However, the proportion of patients who received cranial irradiation before allo-HSCT was significantly higher in ALL or NHL than in others.Another report from European Bone Marrow Transplantation Group on the final height of patients who received allo-HSCT revealed no significant difference was in the delta-SDS, which is defined the difference between the height SDS at allo-HSCT and that at the final height between the ALL, AML, or chronic myeloid leukemia groups. 33Further studies are needed to identify the risk of steroids for the longitudinal height growth of the patients with ALL who received allo-HSCT following chemotherapy, including steroids.Some previous reports revealed male gender as one of the factors associated with decreased the height SDS. 14,33owever, our cohort demonstrated no significant differences between males and females in height SDS on allo-HSCT, and 1, 2, 3, 4, and 5 years after allo-HSCT (Table S1).The small sample size might affect the difference between our results and that of the previous reports.
The present study had some limitations.First, due to the nature of retrospective analysis, we were unable to completely rule out the physician's bias such as the avoidance of TBI containing regimens or MAC regimens for patients who received allo-HSCT at a younger age.Second, because of the heterogeneity in the primary diseases and the significant difference in the primary diseases between the MAC and RIC groups, the treatments for different primary diagnoses provided prior to allo-HSCT might affect height growth.Third, many patients in our cohort did not reach their final heights before the end of the study period.In the present study, median age at allo-HSCT was 8.3 years with 7.3 years of median follow-up period.Therefore, longer follow-up periods are required to determine the effect of condition intensity on height growth after allo-HSCT in children.
In conclusion, conditioning intensity may be associated with a short stature after allo-HSCT in children.Further studies are needed to validate this observed association between conditioning intensity and height growth of children after allo-HSCT.

Table 4
shows the clinical characteristics of the 58 patients in the MAC group.The median age at allo-HSCT of patients in the MAC group who received >8 Gy of TBI was 9.3 years (range, 1.8-16.2years), whereas that of the patients who received ≤8 Gy of TBI was 2.0 years (range, 6 months-14.9years).A significant difference Multivariate logistic regression analysis for the development of short stature at 3 years after allo-HSCT.
T A B L E 3 Clinical characteristics of the studied cohort in the MAC group.