|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 26, No 18 (June 20), 2008: pp. 3031-3037 © 2008 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.14.6399 CRHR1 Polymorphisms Predict Bone Density in Survivors of Acute Lymphoblastic Leukemia
From the Colleges of Pharmacy and Medicine and the Department of Radiology, University of Tennessee; and the Departments of Pharmaceutical Sciences, Radiological Sciences, and Biostatistics and Oncology, St Jude Children's Research Hospital, Memphis TN; and the Channing Laboratory and Pulmonary Division, Brigham and Women's Hospital and Harvard Medical School, Boston, MA Corresponding author: Mary V. Relling, PharmD, Department of Pharmaceutical Sciences, St Jude Children's Research Hospital, 332 North Lauderdale, Memphis, TN 38105-2794; e-mail: mary.relling{at}stjude.org
Purpose Corticosteroids are a critical component of therapy for acute lymphoblastic leukemia (ALL) but are associated with late effects, such as osteoporosis. Risk factors remain poorly defined. Because CRHR1 polymorphisms have been associated with other corticosteroid effects, our goal was to define whether CRHR1 polymorphisms predict which patients with ALL are likely to develop bone mineral deficits. Patients and Methods The mean bone mineral density z scores of 309 long-term survivors of ALL were determined by quantitative computed tomography of the trabecular lumbar spine. We analyzed whether CRHR1 genotypes, adjusted for sex, ALL treatment regimen, and weight, could predict bone density. Results We found that three single nucleotide polymorphisms (SNPs), all in linkage disequilibrium, were associated with bone density in a sex-specific manner. Bone density was lower in males (P = .001), in nonblack patients (P < .08), in those who were not overweight (P < .001), and in those who received intensive antimetabolites and glucocorticoids (P < .001). After adjustment for these features, the G allele at the rs1876828 SNP was associated with lower z scores (P = .02) in males but tended to have the opposite association in females (P = .09). Conclusion CRHR1 polymorphisms may impact the risk of bone density deficits in patients treated with corticosteroids and antimetabolites in a sex-specific manner.
Corticosteroids are associated with a significant number of adverse effects. For decades, antileukemic therapy has included corticosteroids and antimetabolites (eg, mercaptopurine and methotrexate) as the backbone of postremission therapy. Children treated for acute lymphoblastic leukemia (ALL) are at high risk for bone mineral decrements long after the completion of antileukemic therapy.1-3 As many as 75% of long-term survivors of ALL have bone mineral densities less than the population mean,1 and fractures can occur in as many as 39% of patients during therapy.4 Corticosteroids and antimetabolites can cause deleterious effects on bone metabolism.5-9 Corticosteroids induce bone loss by stimulating osteoclastic activity, preventing osteoblastic activity, inhibiting the 1-alpha hydroxylation of vitamin D (which impairs intestinal absorption and renal resorption of calcium), and inhibiting expression of the vitamin D receptor in bone.8,10,11 The mechanism by which methotrexate induces bone loss is not well understood but may be related to increased bone resorption and increased urinary and fecal calcium excretion.7,12,13 Interindividual differences in corticosteroid-associated toxicity and efficacy have been observed in other disease states, such as nephrotic syndrome and asthma and in patients who undergo transplantation.14-19 Germline variations in the corticotrophin-releasing hormone receptor-1 (CRHR1) gene, including the rs1876828 single nucleotide polymorphism (SNP), have been associated with variable lung function responses to inhaled corticosteroids in patients with asthma.15 Because there is tremendous heterogeneity in bone mineral density among patients who have completed glucocorticoid-containing therapy for ALL, we studied whether CRHR1 polymorphisms were associated with bone loss in children treated for ALL.1
Patients All patients who were treated on one of three front-line protocols for ALL (Total XI-XIII) between 1984 and 1997 at St Jude Children's Research Hospital, who were in complete and continuous remission for at least 4 years, and who subsequently had undergone a standardized bone density study as part of one of two institutional protocols (BONE I and II) were eligible for this study.1 This study was approved by the institutional review board, informed consent was obtained from all patients or their representatives, and the study complied with the Health Information Portability and Accountability Act of 1996. All patients received similar remission induction therapy containing prednisone, but primary differences existed in the postinduction therapy (Table 1).20-22 The bone mineral density (expressed as the mean of the L1-L2 trabecular bone) of all participants was measured at least 4 years after completing ALL therapy by quantitative computed tomography (QCT) with a Siemens Somatom-Plus spiral CT scanner (Siemens, Iselin, NY) and with Mindwaves QCT Calibration Phantoms and software (Mindwaves Software, South San Francisco, CA), as previously described.1 A standardized bone mineral density z score (expressed as standard deviations greater than or less than the mean) was calculated for each patient, which represented the difference between the calculated bone density in a patient and the average bone density in age- and sex-matched controls.1 Data for normal children were included in the QCT-based software.
Patients were grouped according to five clinical factors that were previously shown to be important to bone mineral density: age group, sex, ethnicity, body mass index (BMI), and antileukemic treatment.5-8,23-26 Patients were grouped by age as follows: age group 1 (n = 119) included patients younger than 14 years; group 2 (n = 107) included patients aged 14 to 18 years; and group 3 (n = 83) included adult patients older than 18 years. Ethnicity was categorized as white (n = 266), black (n = 37), or other (n = 6).
BMI groups were based on age and body mass index (BMI) at the time the z score was determined.27 BMI group 1 (n = 148) included normal-weight patients (weight within 10th to 85th percentile if < 20 years old, or BMI 18.5 to 25 if Three protocol groups were formed according to similarities in corticosteroid and antimetabolite therapy (Table 1). Patients enrolled in Total XI and XIII were given risk-directed therapy and were grouped accordingly. Protocol group 1 (n = 196) included patients who had received monthly pulses of corticosteroids on a backbone of pairs of multiple antileukemic drugs that were less inclusive of methotrexate/mercaptopurine (arms 2 and 3 of Total XI and the high-risk arms of Total XIII).20,21 Group 2 (n = 48) included patients that received monthly pulses of corticosteroids on a backbone of intensive methotrexate/mercaptopurine therapy (arm 1 of Total XI and the low-risk arms of Total XIII).20,21 Group 3 (n = 65) included patients enrolled solely on Total XII and was unique, because postinduction therapy included no corticosteroids and heavily emphasized antimetabolites.22
Genotyping
Statistical Analyses A classification and regression tree (CART) analysis was employed to determine the interaction of clinical factors and CRHR1 SNPs on bone mineral density z score.28 The underweight group was excluded from all CART analyses because of the small number of patients (n = 8). Protocol group was the most significant determinant of z score and was selected as the first node of the CART analysis. Multiple-factor linear models were applied by using the CRHR1 SNPs and the remaining clinical factors to determine subsequent nodes. Because the rs1876828 SNP has been linked to corticosteroid response in other studies,15 we performed an additional CART analysis that allowed only the rs1876828 SNP to enter the CART. Five patients in protocol group 1 were excluded because of missing rs1876828 SNP genotypes. An identical CART analysis was performed in the largest racial group (white patients) to minimize the possibility that the results were confounded by ethnicity.
Clinical Factor and Haplotype Analysis The HAPLOTYPE procedure in SAS/Genetics (SAS Institute, Cary, NC) that used an expectation-maximization (EM) algorithm was applied to iteratively furnish the maximum-likelihood estimates of population haplotype frequencies, with the assumption of Hardy-Weinberg equilibrium.29 For each patient, the most likely pair of haplotypes was used in the analyses. A haplotype substitution model30,31 was employed to test whether haplotypes had a significant association with QCT bone mineral density z score. To determine significant associations with z score, estimates of the effect of individual haplotypes as a deviation from the effect of the most frequent haplotype (Hap1) were evaluated for significance. The CART analysis was applied to the estimated haplotypes by using general linear models, similar to the analysis for individual SNPs. In each model, the z score was taken as the response variable, and significant clinical factors from the univariate analysis and one of the five major haplotypes were taken as explanatory variables. An identical haplotype analysis was performed in the white subcohort to ensure that the haplotype structure and frequencies identified were not confounded by ethnicity.
Clinical Factors and SNP Effects on Bone Mineral Density z score In the univariate analysis, sex, BMI group, and protocol group all were significantly associated with QCT bone mineral density z score (P < .05), and ethnicity was marginally significantly associated with z score (P = .08; Table 2; Fig 1). Males had lower mean z scores than females, black patients had higher z scores than white patients and those with other ethnicities, and overweight patients had higher z scores than normal-weight and underweight patients. No SNP genotypes were significantly associated with z score in univariate analyses (Appendix Table A1). Age group was not associated with z score (Table 2).
Multiple-factor linear models, including the four significant clinical factors, showed that treatment protocol group was the strongest factor to predict z score (P < .0001), and those patients who received more intense antimetabolite therapy in addition to glucocorticoids (protocol group 2) had the lowest z scores (Fig 1). Because the protocol group was the most significant factor for all models in the multiple-factor analysis, it was taken as the first node in the CARTs. As the rs1876828 SNP was shown previously to be associated with asthmatic response to glucocorticoids,15 we performed a CART analysis with only this SNP. In protocol group 1 (low antimetabolites, high glucocorticoids), sex was the most significant factor that affected the z score (P = .01; Fig 2). The GG genotype was associated with a lower bone density in males (P = .02) and a higher bone density in normal-weight females (P = .09; Figs 2 and 3). In protocol group 3 (low glucocorticoids), the BMI group was the most significant factor for bone density (P = .008), which was higher in overweight patients (Fig 2). There were no significant predictors of bone density in protocol group 2, which is the group that received intensive steroids and intensive antimetabolites. The CART analysis on z score and SNP rs1876828 in white patients showed the same structure as in the whole cohort (Appendix Fig A1, online only).
To test for an interaction effect of SNP rs1876828 and sex, we performed a direct statistical test by using a multiple-factor linear model that was adjusted for protocol group, BMI group (excluding eight underweight patients), sex, rs1876828 genotype (AG/AA versus GG) (excluding five patients with missing genotype data), and the two-way interactions between the rs1876828 binary group with the protocol group, BMI group and sex, respectively. We found that the protocol group (P < .0001), the BMI group (P = .0003), sex (P = .005), and the interaction between sex and SNP rs1876828 (P = .02) all were significantly associated with z score. In males (n = 93), the mean z score for the rs1876828 GG genotype was –0.82 (standard error [SE], 0.12) versus 0.11 (SE, 0.13) in females (n = 92). In the rs1876828 AG/AA binary group, the mean z score was –0.42 (SE, 0.16) in males (n = 57) and –0.50 (SE, 0.18) in females (n = 54). To further validate the significant interaction effect of SNP 1876828 and sex on z score, we used the Akaike Information Criterion and cross-validation to prove that this significant finding was not caused by overfitting of the data (Fig A5). When all nine SNPs were allowed to compete in the CART along with clinical risk factors among all patients (Appendix Fig A2, online only) and among white patients only (Appendix Fig A3, online only), we observed associations with CRHR1 genotypes and z score, but these were confounded by the high linkage disequilibrium among CRHR1 SNPs. Therefore, we analyzed haplotypes for CRHR1.
Clinical Factor and Haplotype Effect on Bone Mineral Density z score A haplotype substitution model was used to estimate associations of haplotypes with the QCT bone mineral density z score. In these analyses, the most frequent haplotype (Hap1) was set to have zero effect (baseline). Haplotypes were significantly associated with z score in multivariate analyses that accounted for clinical risk factors. By CART analysis, within protocol group 1 (low antimetabolites, high glucocorticoids), males again had lower bone density than females (P = .008; Fig 4). Hap3, CCAGCGAGA, was associated with a significantly higher bone density in males (P = .03) but tended to have the opposite association in normal-weight females (P = .07; Figs 3 and 4). An alternative haplotype (Hap5, TCAGCAGAG) was also a significant predictor for z score in the normal-weight females (P = .018). The sex-specific opposing effects of Hap3 in the entire cohort are shown by the negative correlation between the number of copies (0, 1, or 2) of Hap3 and z score in females and the positive correlation to z score in males (Fig 3). The haplotype data (Fig 4) are in agreement with the individual SNP data, because the rs1876828 A allele correlates to a higher bone density in males and a lower bone density in females (Figs 2 and 3), and this SNP is an important defining tag SNP for Hap3.
In the white subcohort (n = 262), seven haplotypes (codes 1, 3, 4, 5, 6, 10, and 12) were estimated (Appendix Table A3). The CART analysis on z score and haplotypes in the white subcohort showed the same structure as in the whole cohort (Appendix Fig A4, online only).
One of the more common sequelae associated with long-term corticosteroid therapy is diminished bone mineral density. In the treatment of ALL, postinduction therapy has often included corticosteroids and antimetabolites, and it has been shown that low bone density is a common complication after completion of therapy.1-3 Factors such as age/puberty, sex, ethnicity, and BMI all can affect bone mineral density,32-37 although genetic predisposition to low bone density has not been assessed previously in patients with ALL. We studied whether CRHR1 variations predicted bone density in the context of other clinical characteristics important for bone density. We found that sex, ethnicity, treatment protocol, and BMI were significantly associated with bone density, and treatment protocol was the strongest predictor, followed by sex and BMI. The rs1876828 SNP variant A allele was predictive of a higher bone density in males but a lower bone density in normal-weight females. The haplotype analysis showed that Hap3 had a similar sex-specific association, with Hap3 copy number correlated to a higher bone density in males but a lower bone density in females, which is consistent with the individual SNP data. Interestingly, among patients who received high cumulative doses of antimetabolites in addition to intense glucocorticoid therapy, there were no significant clinical factors or CRHR1 SNPs that were predictive of bone density, possibly because of the smaller number of patients. It is also possible that the high exposure to antimetabolites could have obscured genetic effects on glucocorticoid response. Our results are consistent with a possible contribution of antimetabolites to steroid-induced bone loss.5-9,12,13 CRHR1 is the primary receptor in the anterior pituitary gland that binds corticotropin hormone, which results in the release of corticotropin and, ultimately, of glucocorticoids (ie, cortisol) from the adrenal glands.38,39 We hypothesize that CRHR1 gene variants can result in a varying release of corticotropin from the anterior pituitary and in altered levels of circulating endogenous glucocorticoids, hence causing a variable susceptibility to exogenous corticosteroid therapy. Our data are consistent with prior studies of CRHR1 for a completely different glucocorticoid-induced phenotype: the rs1876828 SNP was predictive of significantly improved lung function in patients with asthma in a study that included a high percent of females (but not in a study with a high percent of males).15 The effect of sex on bone mineral density has been studied extensively.26,32,33,40,41 Although there are conflicting results,26,42 several prior studies indicate that bone mineral density differs in males and females,1,2,40,43 and our data are consistent with prior reports of lower bone density in males than females after treatment for childhood ALL.1,2 It has been postulated that sex steroids may be responsible for sex differences in disease susceptibility and drug response.44,45 The sex-specific inverse correlation of the rs1876828 variant A allele with bone density (increased toxicity in females, but decreased toxicity in males) could be caused by the effects of sex steroids on the hypothalamic pituitary adrenal axis. Estrogens play a stimulatory role in corticotropin release, as reflected by higher basal and stress-induced levels of endogenous glucocorticoids in females compared with males.46 Because females inherently have higher levels of corticotropin, they may be less likely to succumb to consequences associated with a defective CRHR1 protein (ie, bone mineral decrements caused by increased response to exogenous corticosteroids). Estrogens are more important to bone accumulation than are androgens,47,48 and they may be a factor contributing to the sex-specific differences observed in the current study. The notion that autosomal germline polymorphisms might impact one sex more than the other has precedence. For example, the MDM2 polymorphism effect on cancer risk in females but not males,44 the XbaI GYS1 polymorphism association with cardiovascular mortality in males but not females,49 or the SCARB1 polymorphism association with higher HDL-C levels in females but not males.50 However, it has been questioned whether many of these studies have internal and external validity to claim sex-related differences.51 In our study, we performed a direct test and showed that the interaction between SNP rs1876828 and sex was indeed significant; a previous study showed that the rs1876828 SNP was suggestive of a sex difference in patients with asthma who were treated with inhaled corticosteroids,15 which supports the internal and external validity of our results. Although it is clear that sex and body size can exert effects on bone density, their impacts may be further modified by environmental exposures (in addition to genetic factors).34-37,52-55 Genetic variation in CRHR1 has now been associated with bone mineral density changes. Our findings indicate that the importance of CRHR1 polymorphisms in other settings will depend not only on the intensity of steroids and antimetabolites but also on the sex composition of the study group.
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a "U" are those for which no compensation was received; those relationships marked with a "C" were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment or Leadership Position: None Consultant or Advisory Role: None Stock Ownership: None Honoraria: None Research Funding: Cheng Cheng, National Institutes of Health; Mary V. Relling, National Institutes of Health Expert Testimony: None Other Remuneration: None
Conception and design: Cheng Cheng, Kelan G. Tantisira, Mary V. Relling Administrative support: Mary V. Relling Provision of study materials or patients: Sue C. Kaste, Ching-Hon Pui Collection and assembly of data: Terreia S. Jones, Sue C. Kaste, Wei Liu, Cheng Cheng, Kelan G. Tantisira, Mary V. Relling Data analysis and interpretation: Terreia S. Jones, Wei Liu, Cheng Cheng, Wenjian Yang, Mary V. Relling Manuscript writing: Terreia S. Jones, Sue C. Kaste, Wei Liu, Cheng Cheng, Ching-Hon Pui, Mary V. Relling Final approval of manuscript: Terreia S. Jones, Sue C. Kaste, Cheng Cheng, Kelan G. Tantisira, Ching-Hon Pui, Mary V. Relling
Patients and statistical analysis. Three hundred nine patients who were treated previously on a front-line protocol for acute lymphoblastic leukemia (ALL; Total XI-XIII) between 1984 and 1997 at St Jude Children's Research Hospital, who were in remission for at least 4 years, and who had subsequently undergone a standardized bone density study as a part of one of two St Jude bone studies (Bone I or II) were included in this study. Exclusion criteria included disease recurrence, development of a secondary cancer, pregnancy or lactation, history of spinal irradiation, allogeneic bone marrow transplantation, or absence of informed consent (Kaste SC, Jones-Wallace D, Rose SR, et al: Leukemia 15:728-734, 2001). All patients received similar remission induction therapy that contained prednisone but had primary differences in the postinduction therapy, and all patients were grouped according to five clinical factors (age group, sex, ethnicity, body mass index [BMI], and antileukemic treatment).
Clinical factor and single nucleotide polymorphism analysis.
Clinical factor and haplotype analysis.
The Y value represents the z score; µ represents the overall mean of the population; Sex, Ethnicity, Protocol_Grp, and BMI_Grp represent the effects of sex (female, male), ethnicity (white, black, other), protocol group (1, 2, or 3), and BMI group (normal weight, underweight, and overweight), respectively; Hapk is 0, 1, or 2 copies of haplotype k present in the individual; Hap1 represents the most frequent of M marker haplotypes; and the remaining haplotypes are denoted Hap2, Hapk, and HapM; βk is the regression coefficient that corresponds to the effect of haplotype k as a deviation from the effect of the most frequent haplotype (Hap1), which is set to zero (baseline) to make the model have full rank; and The CART analysis was applied to the estimated haplotypes by using general linear models. The significant clinical factors from the univariate analysis (protocol group, sex, ethnicity, and BMI group) and the number of copies (0, 1, or 2) of each of the five major haplotypes (Hap1 to Hap5) were taken as explanatory variables, and the z score was taken as the response variable in each model. Because Hap6 was a combination of rare haplotypes, it would have been difficult to determine if any particular rare haplotype was associated with the z score in each subcohort; thus, we did not consider Hap6 in the CART analysis.
Results. The AIC decreases as the interaction effects are added, and a large decrement is realized by adding the SNP-sex interaction effect. This indicates that the significance of the full model and the SNP-sex interaction effect are unlikely caused by over-fitting, because AIC measures the goodness of fit with a penalty for model complexity. We next assessed how much the full model–predicted z score could be affected by small perturbations to the data set by using leave-one-out cross validation. First, the full model–predicted z score was calculated for each patient on the basis of the whole data set. Then, in the ith (i = 1,2,... ,n) cross validation round, the full model was fitted without the data of the ith patient, and the z score of the ith patient was predicted by using this model. We obtained n pairs of predicted values: one pair for each patient; in each pair, one value is the full model–predicted z score that is based on the whole (original) data set, and the other value is the full model–predicted z score that is based on the data without that patient (perturbed data). Appendix Figure A5 shows an excellent agreement between the values in each pair (sample correlation > 0.99; P < .0001). Thus, prediction by the full model is not substantially affected by small perturbations to the data set, which further indicates that it is unlikely that the model has overfit the data. Additionally, the P value of the SNP-sex interaction effect ranged from .0096 to .0294 in the 296 cross-validation rounds, which indicates that there is likely a true interaction between SNP genotype and sex.
We thank our protocol coinvestigators, the clinical and research staff, and the patients and their families for their participation.
Supported by Grants No. NCI CA 51001, CA 78224, CA21765 from the National Institutes of Health; by the NIH/NIGMS Pharmacogenetics Research Network and Database Grants No. U01 GM61393, U01 HL65899, U01GM61374 (http://pharmgkb.org), and PS207386 from the National Institutes of Health; by the American Cancer Society; by the Center of Excellence grant from the state of Tennessee; and by American Lebanese Syrian Associated Charities. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Kaste SC, Jones-Wallace D, Rose SR, et al: Bone mineral decrements in survivors of childhood acute lymphoblastic leukemia: Frequency of occurrence and risk factors for their development. Leukemia 15:728-734, 2001[CrossRef][Medline] 2. Kaste SC, Rai SN, Fleming K, et al: Changes in bone mineral density in survivors of childhood acute lymphoblastic leukemia. Pediatr Blood Cancer 46:77-87, 2006[CrossRef][Medline] 3. Brennan BM, Rahim A, Adams JA, et al: Reduced bone mineral density in young adults following cure of acute lymphoblastic leukaemia in childhood. Br J Cancer 79:1859-1863, 1999[CrossRef][Medline] 4. Halton JM, Atkinson SA, Fraher L, et al: Altered mineral metabolism and bone mass in children during treatment for acute lymphoblastic leukemia. J Bone Miner Res 11:1774-1783, 1996[Medline] 5. Stanisavljevic S, Babcock AL: Fractures in children treated with methotrexate for leukemia. Clin Orthop Relat Res 125:139-144, 1977[Medline] 6. Ragab AH, Frech RS, Vietti TJ: Osteoporotic fractures secondary to methotrexate therapy of acute leukemia in remission. Cancer 25:580-585, 1970[CrossRef][Medline] 7. Nesbit M, Krivit W, Heyn R, et al: Acute and chronic effects of methotrexate on hepatic, pulmonary, and skeletal systems. Cancer 37:1048-1057, 1976[CrossRef][Medline] 8. Shane E, Rivas MC, Silverberg SJ, et al: Osteoporosis after cardiac transplantation. Am J Med 94:257-264, 1993[CrossRef][Medline] 9. Joseph JC: Corticosteroid-induced osteoporosis. Am J Hosp Pharm 51:188-197, 1994[Abstract] 10. Lukert BP, Raisz LG: Glucocorticoid-induced osteoporosis: Pathogenesis and management. Ann Intern Med 112:352-364, 1990 11. Bockman RS, Weinerman SA: Steroid-induced osteoporosis. Orthop Clin North Am 21:97-107, 1990[Medline] 12. Sala A, Barr RD: Osteopenia and cancer in children and adolescents: The fragility of success. Cancer 109:1420-1431, 2007[Medline] 13. Nevinny HB, Krant MJ, Moore EW: Metabolic studies of the effects of methotrexate. Metabolism 14:135-140, 1965[CrossRef][Medline] 14. Thomsen OO, Cortot A, Jewell D, et al: A comparison of budesonide and mesalamine for active Crohn's disease. International Budesonide-Mesalamine Study Group. N Engl J Med 339:370-374, 1998 15. Tantisira KG, Lake S, Silverman ES, et al: Corticosteroid pharmacogenetics: Association of sequence variants in CRHR-1 with improved lung function in asthmatics treated with inhaled corticosteroids. Hum Mol Genet 13:1353-1359, 2004 16. Hiraoka M, Tsukahara H, Matsubara K, et al: A randomized study of two long-course prednisolone regimens for nephrotic syndrome in children. Am J Kidney Dis 41:1155-1162, 2003[CrossRef][Medline] 17. Malmstrom K, Rodriguez-Gomez G, Guerra J, et al: Oral montelukast, inhaled beclomethasone, and placebo for chronic asthma. A randomized, controlled trial. Montelukast/Beclomethasone Study Group. Ann Intern Med 130:487-495, 1999 18. Szefler SJ, Martin RJ, King TS, et al: Significant variability in response to inhaled corticosteroids for persistent asthma. J Allergy Clin Immunol 109:410-418, 2002[CrossRef][Medline] 19. Tornatore KM, Reed K, Walshe JJ, et al: Cortisol pharmacodynamic response to long-term methylprednisolone in renal transplant recipients. Pharmacotherapy 14:111-118, 1994[Medline] 20. Pui CH, Mahmoud HH, Rivera GK, et al: Early intensification of intrathecal chemotherapy virtually eliminates central nervous system relapse in children with acute lymphoblastic leukemia. Blood 92:411-415, 1998 21. Pui CH, Sandlund JT, Pei D, et al: Improved outcome for children with acute lymphoblastic leukemia: Results of Total Therapy Study XIIIB at St Jude Children's Research Hospital. Blood 104:2690-2696, 2004 22. Evans WE, Relling MV, Rodman JH, et al: Conventional compared with individualized chemotherapy for childhood acute lymphoblastic leukemia. N Engl J Med 338:499-505, 1998 23. Nussdorfer GG, Mazzocchi G, Robba C, et al: Effects of ACTH and dexamethasone on the zona glomerulosa of the rat adrenal cortex: An ultrastructural stereologic study. Acta Endocrinol (Copenh) 85:608-614, 1977 24. Kalkwarf HJ, Zemel BS, Gilsanz V, et al: The Bone Mineral Density in Childhood Study (BMDCS): Bone Mineral Content and Density According to Age, Sex and Race. J Clin Endocrinol Metab 92:2087-2099, 2007 25. Rosen CJ, Bouxsein ML: Mechanisms of disease: Is osteoporosis the obesity of bone? Nat Clin Pract Rheumatol 2:35-43, 2006[CrossRef][Medline] 26. Gilsanz V, Gibbens DT, Roe TF, et al: Vertebral bone density in children: Effect of puberty. Radiology 166:847-850, 1988 27. Hijiya N, Panetta JC, Zhou Y, et al: Body mass index does not influence pharmacokinetics or outcome of treatment in children with acute lymphoblastic leukemia. Blood 108:3997-4002, 2006 28. Breiman L, Friedman JH, Olshen RA, et al: Classification and Regression Trees. Belmont, CA, Wadsworth International Group, 1984, pp 27-43 29. Excoffier L, Slatkin M: Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol 12:921-927, 1995[Abstract] 30. Sweeney C, Curtin K, Murtaugh MA, et al: Haplotype analysis of common vitamin D receptor variants and colon and rectal cancers. Cancer Epidemiol Biomarkers Prev 15:744-749, 2006 31. Khatib H, Heifetz E, Dekkers JC: Association of the protease inhibitor gene with production traits in Holstein dairy cattle. J Dairy Sci 88:1208-1213, 2005 32. Jones G, Dwyer T: Bone mass in prepubertal children: Gender differences and the role of physical activity and sunlight exposure. J Clin Endocrinol Metab 83:4274-4279, 1998 33. El-Hajj Fuleihan G, Nabulsi M, Tamim H, et al: Effect of vitamin D replacement on musculoskeletal parameters in school children: A randomized controlled trial. J Clin Endocrinol Metab 91:405-412, 2006 34. McVeigh JA, Norris SA, Cameron N, et al: Associations between physical activity and bone mass in black and white South African children at age 9 yr. J Appl Physiol 97:1006-1012, 2004 35. Bass S, Pearce G, Bradney M, et al: Exercise before puberty may confer residual benefits in bone density in adulthood: Studies in active prepubertal and retired female gymnasts. J Bone Miner Res 13:500-507, 1998[CrossRef][Medline] 36. McCulloch RG, Bailey DA, Houston CS, et al: Effects of physical activity, dietary calcium intake and selected lifestyle factors on bone density in young females. CMAJ 142:221-227, 1990[Abstract] 37. Sentipal JM, Wardlaw GM, Mahan J, et al: Influence of calcium intake and growth indexes on vertebral bone mineral density in young females. Am J Clin Nutr 54:425-428, 1991 38. Aguilera G, Nikodemova M, Wynn PC, et al: Corticotropin releasing hormone receptors: Two decades later. Peptides 25:319-329, 2004[CrossRef][Medline] 39. Dautzenberg FM, Hauger RL: The CRF peptide family and their receptors: Yet more partners discovered. Trends Pharmacol Sci 23:71-77, 2002[CrossRef][Medline] 40. Garnett SP, Hogler W, Blades B, et al: Relation between hormones and body composition, including bone, in prepubertal children. Am J Clin Nutr 80:966-972, 2004 41. Warholm M, Guthenberg C, Mannervik B, et al: Purification of a new glutathione S-transferase (transferase µ) from human liver having high activity with benzo{ 42. Nelson DA, Simpson PM, Johnson CC, et al: The accumulation of whole body skeletal mass in third- and fourth-grade children: Effects of age, gender, ethnicity, and body composition. Bone 20:73-78, 1997[Medline] 43. Arabi A, Tamim H, Nabulsi M, et al: Sex differences in the effect of body-composition variables on bone mass in healthy children and adolescents. Am J Clin Nutr 80:1428-1435, 2004 44. Bond GL, Hirshfield KM, Kirchhoff T, et al: MDM2 SNP309 accelerates tumor formation in a gender-specific and hormone-dependent manner. Cancer Res 66:5104-5110, 2006 45. Couzin J: Medicine. Estrogen's ties to COX-2 may explain heart disease gender gap. Science 306:1277, 2004 46. Viau V, Bingham B, Davis J, et al: Gender and puberty interact on the stress-induced activation of parvocellular neurosecretory neurons and corticotropin-releasing hormone messenger ribonucleic acid expression in the rat. Endocrinology 146:137-146, 2005 47. Bilezikian JP, Morishima A, Bell J, et al: Increased bone mass as a result of estrogen therapy in a man with aromatase deficiency. N Engl J Med 339:599-603, 8-27-1998 48. Roche JK, Crum P: Local immunity and the uterine cervix: Implications for cancer-associated viruses. Cancer Immunol Immunother 33:203-209, 1991[CrossRef][Medline] 49. Fredriksson J, Anevski D, Almgren P, et al: Variation in GYS1 Interacts with Exercise and Gender to Predict Cardiovascular Mortality. PLoS ONE 2:e285, 2007[CrossRef] 50. McCarthy JJ, Lewitzky S, Reeves C, et al: Polymorphisms of the HDL receptor gene associated with HDL cholesterol levels in diabetic kindred from three populations. Hum Hered 55:163-170, 2003[Medline] 51. Patsopoulos NA, Tatsioni A, Ioannidis JP: Claims of sex differences: An empirical assessment in genetic associations. JAMA 298:880-893, 2007 52. Havill LM, Mahaney MC, Binkley L, et al: Effects of genes, sex, age, and activity on BMC, bone size, and areal and volumetric BMD. J Bone Miner Res 22:737-746, 2007[CrossRef][Medline] 53. Leonard MB, Shults J, Wilson BA, et al: Obesity during childhood and adolescence augments bone mass and bone dimensions. Am J Clin Nutr 80:514-523, 2004 54. Schmid HH, Bandi PC, Madson TH, et al: Substrate specificity in plasmalogen biosynthesis: Desaturation of 1-O-hexadecyloxyethyl-2-acylglycero-3-phosphoethanolamine in developing rat brain. Biochim Biophys Acta 488:172-178, 1977[Medline] 55. Pocock NA, Eisman JA, Hopper JL, et al: Genetic determinants of bone mass in adults: A twin study. J Clin Invest 80:706-710, 1987[Medline] Submitted September 26, 2007; accepted March 7, 2008.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
20 years old); group 2 (n = 8) included underweight patients (< 10th percentile if < 20 years old, or BMI 
18.5 if 
is the random error associated with the individual. This model was fitted by using least squares (GLM procedure; SAS Institute, Cary, NC), and the significance of associations was determined by an F test on the sum of squares explained by the combined effect of haplotypes. We evaluated for significance the estimates of the effect of individual haplotypes as a deviation from the effect of the most frequent haplotype (Hap1). 
}pyrene-4,5-oxide. Biochem Biophys Res Commun 98:512-519, 1981
