Bone mineral decrements in survivors of childhood acute ... - Nature

Leukemia (2001) 15, 728–734  2001 Nature Publishing Group All rights reserved 0887-6924/01 $15.00 www.nature.com/leu

Bone mineral decrements in survivors of childhood acute lymphoblastic leukemia: frequency of occurrence and risk factors for their development SC Kaste1,2, D Jones-Wallace3, SR Rose4, JM Boyett3, RH Lustig4, GK Rivera5, C-H Pui5,6 and MM Hudson5,6 Departments of 1Diagnostic Imaging, 5Hematology-Oncology, and 3Biostatistics and Epidemiology, St Jude Children’s Research Hospital, Memphis; Departments of 2Radiology, 4Endocrinology, and 6Pediatrics, University of Tennessee, Memphis, TN, USA

We assessed the clinical and treatment factors that predispose survivors of childhood acute lymphoblastic leukemia (ALL) to low bone mineral density (BMD). Using quantitative computed tomography, we determined the frequency of low BMD (defined as ⬎1.645 standard deviations (SD) below the mean) in leukemia survivors treated with multiagent chemotherapy including prednisone and antimetabolite. All participants had completed therapy at least 4 years earlier, remained in continuous complete remission, and had no second malignancies. We statistically correlated BMD results with patient characteristics and treatment histories. Among 141 survivors (median age, 15.9 years; median time after diagnosis, 11.5 years), median BMD z score was −0.78 SD (range, −3.23 to 3.61 SDs). Thirty participants (21%; 95% confidence interval, 15% to 29%) had abnormally low BMD, a proportion significantly (P ⬍ 0.0001) greater than the expected 5% in normal populations. Risk factors for BMD decrements included male sex (P = 0.038), Caucasian race (P ⬍ 0.0001), and cranial irradiation (P = 0.0087). BMD inversely correlated with cranial irradiation dose. BMD z scores of patients who received higher doses of antimetabolites were lower than those of other patients. Childhood ALL survivors are at risk to have low BMD, especially males, Caucasians, and those who received cranial irradiation. Leukemia (2001) 15, 728–734. Keywords: acute lymphoblastic leukemia; osteopenia; osteoporosis; late effects

Introduction With childhood acute lymphoblastic leukemia (ALL) survival rates now approaching 80%,1–3 there is a need to assess the nature of, prevalence of and risk factors for long-term treatment sequelae. Contemporary cancer treatment regimens include agents known to adversely affect bone mineralization. When administration of these agents is combined with nutritional deficits and inactivity, bone mass accretion can be impaired and patients can be predisposed to altered timing of skeletal maturation and diminished bone mineral density (BMD), with the potential for osteoporotic complications. Patients treated for ALL, the most common childhood cancer, are at particular risk of impaired bone mass accretion, because the peak age of disease onset corresponds to a period of rapid growth and bone mass accumulation. Therefore, survivors of ALL, who have a highly curable cancer, are an optimal group in whom to evaluate the long-term metabolic effects of antineoplastic therapy on BMD studies. Prompted by clinical and radiographic observations of diminished bone density in patients with ALL, we undertook a study to evaluate BMD in survivors of childhood ALL. Few of the children in whom this abnormality was observed had clinical signs of diminished BMD. Thus, in the absence of sys-

Correspondence: SC Kaste, Department of Diagnostic Imaging, St Jude Children’s Research Hospital, 332 N Lauderdale, Memphis, TN 38105, USA; Fax: 901/495–4398 Received 7 August 2000; accepted 29 November 2000

tematic imaging studies, physicians are likely to be unaware of this silent bone loss; a decrement that may be exacerbated when coupled with the bone mineral loss that occurs during senescence.4–6 BMD accounts for 75% to 85% of the variance in bone strength, and can be determined only by direct measurement.7 Prospective studies have shown that a decrease in BMD by 1 SD is associated with a 1.5- to 3-fold increase in the relative risk of fracture.7 Low BMD can be determined many years in advance of fracture, when preventive measures may be effective.8 We reasoned that early detection of diminished BMD by a sensitive imaging technique (ie quantitative computed tomography (QCT)) would allow corrective intervention during the most active period of bone mineral deposition. The resulting protocol, opened to enrollment in July 1997, was designed to estimate the frequency, severity and risk factors of diminished BMD in survivors of childhood ALL who were treated on a single institutional protocol.

Patients and methods

Eligibility and recruitment Study participants, recruited from 248 long-term survivors, were initially enrolled on Total Therapy Study XI between 1984 and 1988 and received antileukemic therapy for 2.5 years.9 Patients living in the continental United States were contacted by letter, inviting them to participate or recruited from the After Completion of Therapy (ACT) Clinic at the time of their annual evaluation. We limited recruitment to those able to participate within a 2-year time span (1 July 1997 to 1 July 1999). Patients had completed therapy at least 4 years prior to enrollment in the present study. Patients were eligible to participate in the study if they remained in complete continuous remission and had no second malignancies. Exclusion criteria included a history of spinal irradiation, allogeneic bone marrow transplantation, pregnancy or lactation, disease recurrence, development of a second malignant neoplasm, or absence of written, informed consent to participate. The study protocol was approved by the Institutional Review Board, and informed consent was obtained from parents or guardians and, as appropriate, from patients. Participants were evaluated annually in the ACT Clinic. The objective of the ACT Clinic is to provide, through a multidisciplinary approach, periodic follow-up of the primary disease and to ensure adequate monitoring of late cancer-treatment sequelae. The Clinic also facilitates the transition of the care of adult long-term childhood cancer survivors provided by our staff to that provided by community physicians.

BMD decrements in childhood ALL survivors SC Kaste et al

Table 1 Clinical and demographic correlates of BMD z scores in the 141 survivors of ALL

Sex Male Female Age at diagnosis ⬍4 years ⬎4 years Tanner stageb I/II/III IV/V Race White Non-white

na (%)

Median z scores (SD)

Range

68 (49) 72 (51)

−1.04 −0.70

−3.06 to 3.34 −3.23 to 3.61

P value

0.038 0.85

70 (50) 70 (50)

−0.74 −0.85

−3.23 to 2.04 −2.74 to 3.61

19 (14) 121 (86)

−1.32 −0.73

−3.23 to 0.9 −2.72 to 3.61

125 (89) 15 (11)

−0.93 0.51

−3.23 to 3.61 −1.05 to 3.34

0.038 ⬍0.0001

a Because one participant in Treatment Arm 1 received treatment different from the other groups (chemotherapy and 2400 cGy), this patient was excluded from further analysis. b At the time of QCT examination.

CNS leukemia was not a criterion for risk stratification. CNS-directed therapy included intrathecal chemotherapy with methotrexate, hydrocortisone, and cytarabine during induction, consolidation treatment, and every 8 weeks during the first year of continuation treatment. Cranial irradiation was given to 63% of patients, either with higher risk leukemia (1800 cGy) or CNS leukemia at the time of diagnosis (2400 cGy). The overall 5-year event-free survival estimate (±1 SE) was 71.8 ± 2.4% at 5 years and 69.3 ± 2.4% at 10 years.9 The protocol prescribed a cumulative prednisone dose of 9520 mg/m2 for all patients. The dose of antimetabolite in Treatment Arm 1 differed from that in Treatment Arms 2 and 3. The planned cumulative dose for each chemotherapy agent was the same in Treatment Arms 2 and 3. The planned cumulative MTX dose was 7600 mg/m2 for patients in Treatment Arm 1 and was 5200 mg/m2 for the remaining patients. Because we found no significant difference in either the prevalence or the severity of BMD decrements among patients who were treated on Treatment Arms 2 and 3 without radiation therapy (P = 0.61 and P = 0.87, respectively), we combined the data from these groups for additional evaluation.

729

Evaluation of bone mineral density Treatment history This Study XI treatment9 is summarized in Figure 1. Briefly, remission induction therapy featured a six-drug multiagent regimen of prednisone, vincristine, daunorubicin, l-asparaginase, etoposide and cytarabine. High-dose methotrexate (MTX) and 6-mercaptopurine (6-MP) were given as consolidation treatment upon attaining complete remission. Patients were stratified into low and high risk groups according to presenting leukocyte count, age, race, and blast cell cytogenetics. One third of the low risk patients were randomly assigned to receive antimetabolite-based therapy comprising daily 6-MP and weekly MTX for 3 weeks, alternating with a fourth week of daily prednisone and weekly vincristine (Treatment Arm 1). Two thirds of the low-risk patients and two-thirds of the highrisk patients were randomly assigned to receive weekly rotational combination chemotherapy with four drug pairs (etoposide and cyclophosphamide, 6-MP and MTX, teniposide and cytarabine, prednisone and vincristine; Treatment Arm 2). Finally, one third of the high risk patients received the same four drug pairs, rotated every 6 weeks (Treatment Arm 3; Figure 1). All patients were treated for 120 weeks.9

Figure 1

QCT evaluation: Vertebral trabecular BMD was determined by using a Siemens Somatom-Plus spiral CT scanner (Siemens, Iselin, NY, USA) and Mindwaves QCT Calibration Phantoms and software (Mindwaves Software, South San Francisco, CA, USA).10–13 To measure BMD, we acquired direct axial images of the midbodies of the first and second lumbar vertebrae (L1 and L2, respectively) as determined from sagittal or coronal scout images.10,14–17 If either or both of the L1 and L2 bodies showed evidence of fracture or deformity, then the affected vertebral body was excluded from analysis and an alternative vertebral body from T11 to L4 was chosen. Between 28 and 32 3-mm-thick contiguous-slice images were obtained and provided complete assessment of the target vertebral bodies. The topogram (lateral scout image) was acquired at 60 mA and 140 kV. The axial images were acquired at 360 mA and 80 kV. The dose, approximating half that of a chest X-ray,10–12 was localized to approximately 10 cm in the upper abdomen. The entire scanning procedure, from entry into the CT suite to exit was completed in about 8 min; data processing after the scan required approximately 5 min.

Distribution of patients according to treatment. Leukemia


730

QCT data analysis: Vertebral BMD was determined by using an off-line personal computer equipped with Mindways QCT PRO software (Mindways Software). Weekly completion of a quality assurance monitor examination was performed as described by the software manufacturer. Accurate analysis of the trabecular BMD of the midvertebral body was performed with the interactive software, which allows for three-dimensional rotation of the image of the spine to optimally position the regions of interest. The coefficient of variation for the calculated BMD simultaneously calibrated to the phantom with the QCT PRO method (using three-dimensional positioning of the axial image) is approximately 2.8% (positioning error, approximately 1.5% in serial studies; scanner variability, 1.3%).15

Data from the QCT studies: BMD was defined as the average of the values for the two studied vertebral bodies. A standardized z score, expressed as the number of SDs above or below the mean, was calculated to indicate the difference between the patient’s BMD value and the mean value for ageand sex-appropriate controls (derived from data for normal children as included in a QCT-based database). Results of participants’ studies were maintained automatically in the QCT PRO database management system.

Definition of diminished BMD: We conservatively estimated that 5% of the normal population has diminished BMD; thus, 95% of the normal population has BMD z scores above −1.645 SDs. To compare the prevalence of diminished BMD in our cohort with that expected in the normal population, we considered a BMD value of 1.645 SDs below the mean (5th percentile) or lower to be abnormal. Patients whose BMD values were more than 2 SDs below the mean for age- and sex-matched controls were referred to endocrinologists for further evaluation and possible treatment. Patients with normal or low-normal BMD z scores were given an educational brochure outlining dietary and lifestyle practices to optimize bone mineralization.

Assessment of growth and pubertal status: To correlate the effects of growth and pubertal progression on BMD, we abstracted data regarding height and pubertal status from the medical records. The time of attainment of near-adult height was defined as the first year in which longitudinal growth slowed to 2 cm or less per year.18 The estimated median age for achieving this criterion in the general population is 14.5 ± 1 years for adolescent girls and 17 ± 1 years for adolescent boys.18 We used this growth parameter and evaluated growth velocities of the study participants to retrospectively estimate the ages at which skeletal growth was complete. We excluded from the analysis of pubertal growth those patients who had not reached near-final height by the time of the QCT examination (n = 58). The longitudinal growth timing of participants was then coded as normal, advanced, or delayed; this information was correlated with BMD z scores. We also recorded the Tanner stage19 of each participant at the time of study enrollment to determine whether there was a relationship between pubertal status and BMD. Leukemia

Statistical analysis Differences in the distribution of patient characteristics between the cohort participating in this study (n = 141) and the nonparticipating eligible cohort (n = 107) were assessed with ␹2 tests for categorical factors and with Wilcoxon rank sum tests for continuous variables. A one-sample t-test was used to compare the mean BMD SDs of the entire cohort with the expected mean SD of the normal population. The binomial procedure for one-sample rates and proportions was used to construct exact 95% confidence intervals for the estimated proportion of survivors with diminished BMD and to compare the estimated proportion with the expected proportion in the normal population. Fisher’s exact tests were used to test differences between treatment groups in the prevalence of diminished BMD z scores, and exact Wilcoxon-rank sum tests were used to measure differences between the groups in median BMD z scores. Significance levels (P ⭐ 0.05) were adjusted for multiple comparisons among the radiation dose groups by the Bonferroni method. Because of the differences among treatment groups, Wilcoxon rank sum analyses addressing the impact of sex, pubertal status at the time of study enrollment, age at the time of diagnosis, and race were stratified by treatment: Treatment Arm 1 (n = 19), Treatment Arms 2 and 3 with no irradiation (n = 27), Treatment Arms 2 and 3 with 1800 cGy of irradiation (n = 79), Treatment Arms 2 and 3 with 2400 cGy of irradiation (n = 15). Correlations between age at the time of attainment of near-final height and BMD z scores were estimated with Spearman’s correlation coefficient. Median BMD z scores were compared among the growth groups by exact Wilcoxonrank sum tests separately for males and females. Results

Patient characteristics Between 1 July 1997 and 1 November 1998, 141 patients from a pool of 248 eligible patients were studied. Most participants were Caucasian (89%); 69 (49%) were male. The median age at the time of diagnosis of ALL was 4.0 years (range, 0.9 to 18.7 years) and at the time of study enrollment 15.9 years (range, 10.6 years to 30.1 years). The median time from diagnosis to study enrollment was 11.5 years (range, 8.9 to 14.6 years). We found clinically unsuspected compression abnormalities of lumbar vertebral bodies in five participants, four of whom were male. These abnormalities were wedging of a single lumbar vertebral body in two patients and wedging of more than one vertebral body in three patients. In these five patients, BMD z scores were 0.18, −1.10, −1.74, −2.57, and −2.72 SDs. Thus, BMD z scores were abnormal (ie more than 1.645 SDs below the mean) in three of these five cases. Evaluation of the characteristics of the participating cohort and the nonparticipating eligible cohort, revealed a significant difference only in the median age at the time of diagnosis (4 years vs 5.7 years, P = 0.016), due to the fact that older patients have been placed on alumni status and were followed only by our tumor registry. The median BMD z score of the participating cohort was 0.78 SD below the mean (range, 3.23 SDs below the mean to 3.61 SDs above the mean; Figure 2), which is significantly below the expected median BMD z score (0 SDs) of the normal population (P ⬍ 0.0001). The BMD values of 21% of the long-term survivors studied were


731

Figure 2 Distribution of BMD standardized z scores. The number on each bar indicates the percent of studied cases. Note that the median BMD z score of the population lies between 1 and 2 SDs below the expected median.

Figure 3 Distribution of the BMD z scores according to radiation therapy.

Effect of other factors on BMD more than 1.645 SDs below the mean for age- and sexmatched controls (exact 95% confidence interval, 15% to 29%). This percentage is significantly greater than the expected 5% in the normal population (P ⬍ 0.0001). BMD z scores more than 2 SD below the mean were found in 18 patients (13%); 12 patients met the World Health Organization criterion for osteopenia (z scores of 2 to 2.5 SDs below the mean), and six patients met the criteria for osteoporosis (z scores more than 2.5 SDs below the mean).20

Effect of antimetabolite therapy on BMD Of note, a comparison of the BMD z scores of the 19 nonirradiated patients on Treatment Arm 1 with those of the 27 nonirradiated patients on Treatment Arms 2 and 3 showed that those who received higher doses of 6-MP and MTX (Treatment Arm 1) had lower BMD z scores. This difference did not reach statistical significance (P = 0.10), perhaps because of the small sample size, but suggests possible added effect of these commonly used antimetabolite agents on bone metabolism.

Effect of cranial radiation therapy on BMD Because all patients in Treatment Arms 2 and 3 received the same type and dose of chemotherapy but received different amounts of cranial irradiation, we combined the BMD z scores of these patients to evaluate the effect of cranial irradiation on BMD. The 15 patients treated with 2400 cGy irradiation had significantly lower median BMD z scores than the 79 who received 1800 cGy (P = 0.006), or the 27 who received no radiation therapy (n = 27; P = 0.056). There was no significant difference between BMD z scores of those who received 1800 cGy and those who received no radiation therapy (P = 1.00; Figure 3). Fifty-three percent of those receiving 2400 cGy had diminished BMD, defined as more than 1.645 SDs below the mean for age- and sex-matched controls. Further, those receiving 2400 cGy irradiation were 3.6 times more likely to have diminished BMD than the combined groups of 0 cGy and 1800 cGy (P = 0.016; 95% CI: 1.1–12.0).

Demographic factors: After stratifying the cohort by treatment method, males (P = 0.038) and Caucasians (P ⬍ 0.0001) had lower BMD z scores than females and non-Caucasians, respectively (Table 1). The severity of BMD decrement among males was even more apparent when the cohort was limited to Caucasians. This finding was true throughout all strata. BMD z scores did not correlate with age at the time of diagnosis. The BMD z scores of children who were younger than 4 years at diagnosis of ALL (median z score = 0.74 SD below the mean; range, 3.23 SDs below the mean to 3.61 SDs above the mean) were not different (P = 0.85) than those of older patients (median z score, 0.85 SD below the mean; range, 2.74 SDs below the mean to 3.61 SDs above the mean). However, there was a difference in BMD z scores with respect to Tanner stage. In all strata, participants with Tanner stages 1, 2, or 3 at the time of BMD measurement had lower BMD z scores than did those with Tanner stages 4 or 5 (P = 0.038). Hormonal supplementation: Twelve study participants were receiving thyroid hormone replacement therapy, and 17 were receiving growth hormone replacement therapy. We found no significant difference (P = 0.091, two-sided exact Wilcoxon rank sum test) in the BMD between patients receiving growth hormone supplementation and those who did not. However, since referral for growth hormone deficiency evaluation was not consistent in the early years of study, some families declined evaluation for growth hormone deficiency, and some chose not to pursue growth hormone therapy, we could not accurately evaluate the effect of growth hormone replacement in our study cohort. Attainment of near-adult height: Assessment of growth velocity of all participants indicated that 83 (27 males and 56 females) had completed growth before their QCT examination; 47 (57%) of these achieved near-adult height within 12 months of the estimated median age for the general population (ie 14.5 years for adolescent girls and 17 years for adolescent boys). Completion of linear growth was premature in 24 adolescent girls (43%) and nine adolescent boys (33%) and Leukemia


732

was delayed in two girls (2%) and two boys (8%). We found a weak negative correlation between BMD and estimated age of attainment of near-adult height (r = 0.31 for females and r = 0.25 for males). The negative correlation suggests that as age at the time of attainment of near-final height decreases, the BMD z score increases. Those who completed linear growth at the expected age had lower BMD z scores (females, −0.66 SD; males, −0.96 SD) than did those who completed their growth at an older age (females, 0.09 SD (P = 0.031); males, −0.79 (P = 0.56). We could not perform an analysis of those patients with delayed attainment of near-final height because the sample size was too small. Discussion At a time when our study cohort should have achieved peak BMD, we found that 68% had BMD below the mean. The degree of additional loss that is expected during middle age leads to the possibility of severe osteoporosis and its costly sequelae in this cohort of survivors. Bone mineral accumulation normally occurs during the period corresponding to the onset of most childhood ALLs.21–23 Most of bone mass is usually achieved between puberty and the age of 25 years, after which there is little, if any, subsequent accretion. Achievement of normal BMD appears to be largely genetically determined24–28 but may also be modified by such factors as cranial irradiation,21,29–35 chemotherapy,22,23,36–50 nutritional status,38–40,42–45,52,53–55 and physical activity levels38,40,51,53,54,56,57 as well as age at the onset of puberty.38,47–49,58,59 The relative contribution of these factors to diminished BMD in survivors of childhood cancer has hitherto not been well studied. Abnormalities of bone mineralization have been previously reported in survivors of ALL.21–23,36,37,60,61 However, the impact of different risk factors for this complication in children treated with contemporary therapy was unclear. In two previous studies, severe abnormalities in BMD (ie BMD z scores more than 2 SDs below the age- and sex-appropriate mean) were seen exclusively in patients treated with cranial irradiation.21,39 We found a significant radiation effect only in those who received 2400 cGy of radiation therapy. The lack of a significant effect with 1800 cGy suggests a dose-dependent effect on hypothalamic-pituitary function which in turn affects bone mineral accrual. Although 1800 cGy is sufficient to cause growth hormone deficiency in some patients, the effect is more profound and more common at higher doses.30–35 We hypothesize that the diminished BMD in patients receiving 2400 cGy is a consequence of subclinical endocrine alteration (growth hormone deficiency and central hypothyroidism). It should be noted that the data from the cohort in this study are skewed, because 70% of the survivors received cranial radiation. Accrual of a larger proportion of participants who did not receive cranial irradiation will permit a more informative analysis of the role of cranial irradiation on the development of diminished BMD. Contrary to prior findings implicating cranial irradiation as the sole or primary risk factor for diminished BMD in survivors of ALL, our results demonstrate that BMD deficits are not limited to patients treated with cranial irradiation. Antimetabolites and glucocorticoids – key components in the treatment of childhood ALL – have detrimental effects on bone metabolism. In previous studies, Ragab et al37 reported a high incidence of skeletal growth disturbance, severe osteoporosis, and fractures in children with ALL receiving MTX therapy (with or

Leukemia

without concurrent steroid administration). Others have described fractures and delayed fracture healing in patients receiving MTX continuation therapy.22,41 The mechanism by which methotrexate induces bone mineral loss is unknown. An increased excretion of urinary and fecal calcium in children treated with MTX suggests that increased bone resorption and excretion, rather than diminished bone formation, may be the causal mechanism.41,42 In clinical trials at St Jude Children’s Research Hospital, children at higher risk of treatment failure receive higher doses of MTX. This risk-directed therapy facilitated evaluation of the contribution of MTX to diminished BMD in our study population. Our study shows that chemotherapy can affect BMD: those patients who received higher doses of 6-MP and MTX have lower BMD z scores. The osteoporotic effects of glucocorticoids, a standard component of ALL therapy, are well established. Boot et al36 implicated corticosteroids and cytostatic therapy as contributors to diminished BMD in survivors of ALL. Children, like postmenopausal women, are particularly at risk of steroidinduced osteoporosis because of their more rapid bone turnover.43,61 Glucocorticoids prevent 1␣-hydroxylation of vitamin D in the kidneys to form the active metabolite 1,25dihydroxyvitamin D.44 When this metabolite is not produced, the intestinal absorption of calcium is impaired. Glucocorticoids directly inhibit the expression or function of many hormones and factors important in calcium accretion of bone.43,46–48,62 For example, glucocorticoids inhibit the expression of the vitamin D receptor in bone49 and the production of osteocalcin (the principal bone matrix protein), and they decrease local production of cytokines, which normally inhibit bone resorption.43,46–48,50,62 Hence, the mechanisms of glucocorticoid-induced osteoporosis are multifactorial, with the greatest impact observed on the vertebral trabecular bone.63 One surprising finding of our study is that, among survivors of ALL, BMD decrements occur more often and with greater severity in boys than in girls. One hypothesis is that skeletal maturation is more delayed in boys than in girls and that progression through puberty is accelerated in adolescent girls, particularly after radiation therapy.35 This accelerated maturation results in an earlier and potentially greater estrogen effect on BMD accretion. In addition, estrogen is thought to have a greater effect on bone mineral accretion than does androgen.19 Adolescent girls may also receive oral contraceptives (estrogen) for birth control, which will increase BMD. Finally, some of our teen-aged boys may have partial hypogonadism (due to cyclophosphamide treatment), which may adversely affect BMD accretion. The cause of the inverse correlation between longitudinal growth and BMD decrement is unclear. Accelerated bone accretion occurs during puberty as a result of the direct sex hormone effect on both bone and cartilage, coincident with the epiphyseal fusion process. Indeed, our patients with Tanner stages I, II, or III had lower BMD z scores than did our patients with Tanner stages IV or V (Table 1). Many children treated for ALL exhibit earlier entry into puberty, and sometimes exhibit accelerated tempos of pubertal progression.34,35 Patients who rapidly progress through puberty would, by definition, have undergone more rapid bone accretion during this finite time period because of the sex hormone effect. However, a person’s BMD does not reach its maximum level until the age of approximately 25 years.64 Conceivably, the BMD of patients undergoing normal pubertal progression might ‘catch up’ as they reach the second and third decades of life. Longitudinal evaluation of our cohort will be required


to evaluate this possibility. The inverse correlation between longitudinal growth and BMD suggests that as age at the time of attainment of near-final height decreases, the BMD increases. This finding is expected because the sex hormones promote increased BMD.58,64 In our preliminary analyses, growth hormone therapy (analyzed as an indicator of growth hormone deficiency severe enough to require intervention) does not appear to be predictive of decreased BMD analysis. Likewise, Brennan et al65 recently reported a lack of correlation between growth hormone status and lumbar trabecular bone using QCT in a small (n = 31) cohort of childhood ALL survivors who received cranial irradiation. However, the results of both studies differ from those of other reports21,61,66 in which more severe deficits were found in patients with growth hormone deficiency. This discrepancy may reflect the small number of patients receiving growth hormone therapy in our sample and that of Brennan et al. It is possible that some patients in our cohort have subclinical or untreated growth hormone deficiency that has not yet been identified. During the era in which many of these children were treated, routine evaluation of growth velocity was inconsistent. Because numerous confounding factors may be related to the impact of growth hormone deficiency and replacement therapy, further study is clearly needed. Osteopenia in survivors of childhood ALL constitutes an important health risk in adulthood. The prevalence of this adverse late effect is significant, even at a median of 10 years after therapy. Further, we have shown that males, Caucasians, and those treated with 2400 cGy of cranial irradiation are at particularly high risk of osteopenia and, potentially, osteoporosis. The information obtained from this study should provide impetus for similar studies of patients with other pediatric malignancies. Acknowledgements We thank Ms Sherri Patterson and Margie Zacher for medical record reviews and data entry, Drs Russell Chesney and Laura Carbone for advice on study design and data interpretation, Ms Myra Hazard for patient recruitment, and Dr Julia Cay for editorial advice and manuscript preparation. This work was supported in part by grants P30 CA-21765 and P01 CA-23099 from the National Cancer Institute, by the American Lebanese Syrian Associated Charities (ALSAC), by the Society for Pediatric Radiology and by a Center of Excellence Grant from the State of Tennessee. References 1 Pui C-H, Evans WE. Acute lymphoblastic leukemia. N Engl J Med 1998; 339: 605–615. 2 Gaynon PS, Trigg ME, Heerema NA, Sensel MG, Sather HN, Hammond GD, Bleyer WA. Children’s Cancer Group Trials in Childhood acute lymphoblastic leukemia: 1983–1995. Leukemia (in press). 3 Schrappe M, Reiter A, Ludwig W-D, Harbott J, Zimmermann M, Hiddemann W, Niemeyer C, Henze G, Feldges A, Zintl F, Kornhuber B, Ritter J, Welte K, Gadner H, Riehm H. Improved outcome in childhood acute lymphoblastic leukemia despite reduced use of anthracyclines and of cranial radiotherapy: results of trial ALL-BFM 90. German–Austrian–Swiss ALL-BFM Study Group. Blood 2000; 95: 3310–3322. 4 Chestnut CH III. Report from the NIH consensus conference, 1984, and NIH/NOF workshop, 1987. In: Genant HK (ed.). Osteoporosis Update 1987. Radiology Research and Education Foundation: San Francisco, 1987, pp 3–6.

5 Cummings SR. Epidemiology of osteoporotic fractures. In: Genant HK (ed.). Osteoporosis Update 1987. Radiology Research and Education Foundation: San Francisco, 1987, pp 7–12. 6 Prince RL, Smith M, Dick IM, Price RI, Webb PG, Henderson NK, Harris NM. Prevention of postmenopausal osteoporosis. N Engl J Med 1991; 325: 1189–1195. 7 Melton LJ III, Chrischilles EA, Cooper C, Lane AW, Riggs BL. Perspective: how many women have osteoporosis? J Bone Miner Res 1992; 7: 1005–1010. 8 Johnston CC Jr, Slemenda CW, Melton LJ III. Clinical use of bone densitometry. N Engl J Med 1991; 324: 1105–1109. 9 Pui C-H, Boyett JM, Rivera GK, Hancock ML, Sandlund JT, Ribeiro RC, Rubnitz JE, Behm FG, Raimondi SC, Gajjar A, Razzouk B, Campana D, Kun LE, Relling MV, Evans WE. Long-term results of Total Therapy Studies 11, 12, and 13A for childhood acute lymphoblastic leukemia at St. Jude Children’s Research Hospital. Leukemia (in press). 10 Cann CE. Low-dose CT scanning for quantitative spinal mineral analysis. Radiology 1981; 140: 813–815. 11 Cann CE. A rational approach to radiation exposure in bone densitometry. Radiology 1987; 165: 184. 12 Kalendar WA. Effective dose values in bone mineral measurements by photon absorptiometry and computed tomography. Osteoporosis Int 1992; 2: 82–87. 13 ICRP Publication 60. 1990 Recommendations of the International Commission on Radiological Protection. Ann ICRP 1991; 21: 1–3. 14 Cann CE. Quantitative CT for determination of bone mineral density: a review. Radiology 1988; 166: 509–522. 15 Cann CE, Genant HK. Precise measurement of vertebral mineral content using computed tomography. J Comput Assist Tomogr 1980; 4: 493–500. 16 Cann CE, Genant HK, Kolb FO, Ettinger B. Quantitative computed tomography for prediction of vertebral fracture risk. Bone 1985; 6: 1–7. 17 Cann CE. Quantitative CT for determination of bone mineral density: a review. Radiology 1988; 166: 509–522. 18 Greulich WW, Pyle SI. Radiographic Atlas of Skeletal Development of the Hand and Wrist, 2nd edn. Stanford University Press: Stanford, 1992, pp 231–251. 19 Styne D. Puberty. In: Greenspan FS, Strewler GJ (eds). Basic and Clinical Endocrinology, 5th edn. Appleton and Lange: Stamford, 1997, pp 521–546. 20 National Osteoporosis Foundation. Physician’s Guide to Prevention and Treatment of Osteoporosis. National Osteoporosis Foundation: Washington, DC, 1998. 21 Gilsanz V, Carlson ME, Roe TF, Ortega JA. Osteoporosis after cranial irradiation for acute lymphoblastic leukemia. J Pediatr 1990; 117: 238–244. 22 Stanisavljevic S, Babcock AL. Fractures in children treated with methotrexate for leukemia. Clin Orthop 1977; 125: 139–144. 23 Arikoski P, Komulainen J, Voutilainen R, Riikonen P, Parvianen M, Tapanainen P, Knip M, Kroger H. Reduced bone mineral density in long-term survivors of childhood acute lymphoblastic leukemia. J Pediatr Hematol Oncol 1998; 20: 234–240. 24 Ferrari S, Rizzoli R, Chevalley T, Slosman D, Eisman JA, Bonjour JP. Vitamin-D-receptor-gene polymorphisms and change in lumbar-spine bone mineral density. Lancet 1995; 345: 423–424. 25 Sainz J, Van Tornout JM, Loro ML, Sayre J, Roe TF, Gilsanz V. Vitamin D–receptor gene polymorphisms and bone density in prepubertal American girls of Mexican descent. N Engl J Med 1997; 337: 77–82. 26 Howard G, Nguyen T, Morrison N, Watanabe T, Sambrook P, Eisman J, Kelly PJ. Genetic influence on bone density: physiological correlates of vitamin D receptor gene alleles in premenopausal women. J Clin Endocrinol Metab 1995; 80: 2800–2805. 27 Krall EA, Parry P, Lichter JB, Dawson-Hughes B. Vitamin D receptor alleles and rates of bone loss: influences of years since menopause and calcium intake. J Bone Miner Res 1995; 10: 978–984. 28 Tokita A, Matsumoto H, Morrison NA, Tawa T, Miura Y, Fukamauchi K, Mitsuhashi N, Irimoto M, Yamamori S, Miura M, Watanabe T, Kuwabara T, Yakuta K, Eisman JA. Vitamin D receptor alleles, bone mineral density and turnover in premenopausal Japanese women. J Bone Miner Res 1996; 11: 1003–1009. 29 Dacou-Voutetakis C, Xypolyta A, Haidas S, Constantinidas M, Papavasiliou C, Zannos-Mariolea L. Irradiation of the head.

733

Leukemia


734 30 31

32 33

34

35 36 37 38

39

40 41 42 43 44 45 46 47 48

Leukemia

Immediate effect on growth hormone secretion in children. J Clin Endocrinol Metab 1977; 44: 791–794. Shalet SM, Beardwell CG, Pearson D, Jones PH. The effect of varying doses of cerebral irradiation on growth hormone production in childhood. Clin Endocrinol 1976; 5: 287–290. Oberfield SE, Nirenberg A, Allen JC, Cohen H, Donahue B, Prasad V, Schiff R, Pang S, Ghavimi F, David R, Chrousos G, Sklar C. Hypothalamic-pituitary-adrenal function following cranial irradiation. Horm Res 1997; 47: 9–16. Wara WM, Richards GE, Grumbach MM, Kaplan SL, Sheline GE, Conte FA. Hypopituitarism after irradiation in children. Int J Radiat Oncol Biol Phys 1977; 2: 549–552. Lustig RH, Schriock EA, Kaplan SL, Grumbach MM. Effect of growth hormone-releasing factor on growth hormone release in children with radiation-induced growth hormone deficiency. Pediatrics 1985; 76: 274–279. Blatt J, Lee P, Suttner J, Finegold D. Pulsatile growth hormone secretion in children with acute lymphoblastic leukemia after 1800 cGy cranial radiation. Int J Radiat Oncol Biol Phys 1988; 15: 1001–1006. Rappaport R, Brauner R. Growth and endocrine disorders secondary to cranial irradiation. Pediatr Res 1982; 25: 261–267. Boot AM, vd Heuvel MM, Ha¨hlen K, deMuinck Keizer SMPF. Bone mineral density of children with acute lymphoblastic leukemia. Med Pediatr Oncol 1996; 27: 292 (Abstr P-21). Ragab AH, Frech RS, Vietti TJ. Osteoporotic fractures secondary to methotrexate therapy of acute leukemia in remission. Cancer 1970; 25: 580–585. Kanis JA, Delmas P, Burckhardt P, Cooper C, Torgerson D. Guidelines for diagnosis and management of osteoporosis. The European Foundation for Osteoporosis and Bone Disease. Osteoporos Int 1997; 7: 390–406. Hoorweg-Nijman JJG, Kardos G, Roos JC, van Dijk HJ, Netelenbos C, Popp-Snijders C, deRidder CM, Delemarre-van de Waal HA. Bone mineral density and markers of bone turnover in young adult survivors of childhood lymphoblastic leukaemia. Clin Endocrinol 1999; 50: 237–244. Schwartz AM, Leonidas JC. Methotrexate osteopathy. Skeletal Radiol 1984; 11: 13–16. Nesbit M, Krivit W, Heyn R, Sharp H. Acute and chronic effects of methotrexate on hepatic, pulmonary, and skeletal systems. Cancer 1976; 37: (Suppl. 2) 1048–1057. Nevinney HB, Krant MJ, Moore EW. Metabolic studies of the effects of methotrexate. Metabolism 1965; 14: 135–139. Joseph JC. Corticosteroid-induced osteoporosis. Am J Hosp Pharm 1994; 51: 188–197. Chesney RW, Mazess RB, Hamstra AJ, DeLuca HF. Reduction of serum 1,25-dihydroxyvitamin-D3 in children receiving glucocorticoids. Lancet 1978; 2: 1123–1125. Gaynon PS, Lustig RH. The use of glucocorticoids in acute lymphoblastic leukemia of childhood. J Pediatr Hematol Oncol 1995; 17: 1–12. Lukert BP, Raisz LG. Glucocorticoid-induced osteoporosis: pathogenesis and management. Ann Intern Med 1990; 112: 352–364. Bockman RS, Weinerman SA. Steroid-induced osteoporosis. Orthop Clin North Am 1990; 21: 97–107. Reid IR. Pathogenesis and treatment of steroid osteoporosis. Clin Endocrinol (Oxf) 1989; 30: 83–103.

49 Godschalk M, Levy JR, Downs RW Jr. Glucocorticoids decrease vitamin D receptor number and gene expression in human osteosarcoma cells. J Bone Miner Res 1992; 7: 21–27. 50 Halton JM, Atkinson SA, Fraher L, Webber C, Gill GJ, Dawson S, Barr RD. Altered mineral metabolism and bone mass in children during treatment for acute lymphoblastic leukemia. J Bone Miner Res 1999; 11: 1774–1783. 51 Munnings F. Osteoporosis: what is the role of exercise? Phys Sports Med 1992; 20: 127–138. 52 Cooper C, Cawley M, Bhalla A, Egger P, Ring F, Morton L, Baker D. Childhood growth, physical activity, and peak bone mass in women. J Bone Miner Res 1995; 10: 940–947. 53 McCulloch RG, Bailey DA, Houston CS, Dodd BL. Effects of physical activity, dietary calcium intake and selected lifestyle factors on bone density in young women. Can Med Assoc J 1990; 142: 221–227. 54 Sentipal JM, Wardlaw GM, Mahan J, Matkovic V. Influence of calcium intake and growth indexes on vertebral bone mineral density in young females. Am J Clin Nutr 1991; 54: 425–428. 55 Gilsanz V, Roe TF, Antunes J, Carlson M, Duarte ML, Goodman WG. Effect of calcium on bone density in growing rabbits. Am J Physiol 1991; 260: E471–E476. 56 Johnston CC, Miller JZ, Slemenda CW, Reister TK, Hui S, Christian JC, Peacock M. Calcium supplementation and increases in bone mineral density in children. N Engl J Med 1992; 327: 82–87. 57 Matkovic V, Fontana D, Tominac C, Goel P. Chesnut CH III. Factors that influence peak bone mass formation: a study of calcium balance and the inheritance of bone mass in adolescent famales. Am J Clin Nutr 1990; 52: 878–888. 58 Gilsanz V, Gibbens DT, Roe TF, Carlson M, Senac MO, Boechat MI, Huang HK, Schulz EE, Libanati CR, Cann CC. Vertebral bone density in children: effect of puberty. Radiology 1988; 166: 847–850. 59 Finkelstein JS, Neer RM, Biller BM, Crawford JD, Klibanski A. Osteopenia in men with a history of delayed puberty. N Engl J Med 1992; 326: 600–604. 60 Warner JT, Evans WD, Dunstan FDJ, Bell W, Gregory JW. Relative osteopenia following treatment for childhood acute lymphoblastic leukemia. Med Pediatr Oncol 1996; 27: 241 (Abstr O-121). 61 Henderson RC, Madsen CD, Davis C, Gold SH. Bone density in survivors of childhood malignancies. J Pediatr Hematol Oncol 1996; 18: 367–371. 62 Southard RN, Morris JD, Mahan JD, Hayes JR, Torch MA. Bone mass in healthy children: measurement with quantitative DXA. Radiology 1991; 179: 735–738. 63 Clinical indications for bone mass measurements. A report from the Scientific Advisory Board of the National Osteoporosis Foundation. J Bone Miner Res 1989; 4 (Suppl. 2): 1–28. 64 Gilsanz V, Gibbens DT, Carlson M, Boechat MI, Cann CE, Schulz EE. Peak trabecular vertebral density: a comparison of adolescent and adult females. Calcif Tissue Int 1988; 43: 260–262. 65 Brennan BMD, Rahim A, Adams JA, Eden OB, Shalet SM. Reduced bone mineral density in young adult following cure of acute lymphoblastic leukaemia in childhood. Br J Cancer 1999; 79: 1859–1863. 66 Nussey SS, Hyer SL, Brada M, Leiper AD, Pazianas M. Bone mineralization after treatment of growth hormone deficiency in survivors of childhood malignancy. Acta Paediatr Suppl 1994; 399: 9–14.