Clinical Endocrinology (2015) 83, 508–517
doi: 10.1111/cen.12773
ORIGINAL ARTICLE
Growth hormone deficiency after childhood bone marrow transplantation with total body irradiation: interaction with adiposity and age N.L. Davis*, C.E. Stewart†, A.D. Moss‡, W.W.W. Woltersdorf*, L.P. Hunt§, R.A. Elson*, J.M. Cornish¶, M.C.G. Stevens§ and E.C. Crowne* *Department of Paediatric Endocrinology and Diabetes, University Hospitals Bristol NHS Foundation Trust, Bristol, †Department of Stem Cells, Ageing and Molecular Physiology Unit, Research Institute for Sport & Exercise Sciences, Liverpool John Moores University, Liverpool, ‡Department of Exercise and Sport Science, Manchester Metropolitan University, Crewe, Cheshire, §School of Clinical Sciences, University of Bristol, and ¶Department of Paediatric Oncology, University Hospitals Bristol NHS Foundation Trust, Bristol, UK
Summary Objective Bone marrow transplantation with total body irradiation (BMT/TBI) has adverse effects on growth, growth hormone status and adiposity. We investigated the GH–IGF-I axis in relation to adiposity. Design Cross-sectional case control study. Patients BMT/TBI survivors (n = 22) and short stature control participants (n = 19), all GH-na€ıve or off GH treatment >3 months. Measurements Auxology, DEXA scans and GH–IGF-I axis investigation: (i) 12-h overnight GH profiles; (ii) insulin tolerance test (ITT); and (iii) IGF-I generation test. Analysis: autodeconvolution of GH profile data and comparison of quantitative parameters using ANOVA. Results Eighty-two percent of BMT/TBI survivors had growth hormone deficiency (GHD) using ITT. GH profile area-underthe-curve (GH-AUC) was reduced in BMT/TBI survivors vs short stature control participants [geometric mean (range) 209 (21–825) vs 428 (64–1400) mcg/l/12 h, respectively, P = 0007]. GHD was more marked in those who had additional cranial irradiation (CRT) [ITT peak 14 (02–30) vs TBI only 41 (11– 148) mcg/l, P = 0036]. GHD was more marked at the end of growth in BMT/TBI survivors vs short stature control participants (GH-AUC 551 (64–2474) vs 1369 (192–4197) mcg/l/12 h, respectively, P = 0011) and more prevalent (9/11 vs 1/9, respectively, P = 0005). GH profile data were consistent with ITT results in 80% of participants. IGF-I generation tests were normal. BMT/TBI survivors still demonstrated lower GH levels after
Correspondence: Dr Elizabeth Crowne, Department of Paediatric Endocrinology and Diabetes, Level 6 UH Bristol Education Centre, Upper Maudlin Street, Bristol BS2 8AE, UK. Tel.: +44 117 3420165; Fax: +44 117 3420186; E-mail: [email protected]
508
adjustment for adiposity (fat-adjusted mean difference for GHAUC 909 mcg/l/12 h, P = 0025). Conclusions GHD was more prevalent in BMT/TBI survivors than expected for the CRT dose in TBI, worsened with time and persisted into adulthood. GHD could not be explained by adiposity. There was no evidence of GH neurosecretory dysfunction or resistance after BMT/TBI. (Received 22 September 2014; returned for revision 24 October 2014; finally revised 7 March 2015; accepted 12 March 2015)
Introduction Poor growth, found consistently after childhood bone marrow transplantation with total body irradiation (BMT/TBI), is multifactorial.1 GHD is a well-established consequence of high-dose CRT2,3 and is dose, fractionation and time since treatment dependent. Its reported prevalence varies following lower dose regimens (12-24 Gy),4 including TBI.5–7 The biologically effective dose (BED) of irradiation to the pituitary should relate to hypothalamic and pituitary late effects in terms of GHD,8,9 as it represents the effects of ionizing radiation delivered and takes into account the type of radiation and the nature of the tissue being irradiated. Unsurprisingly, both GH secretion and growth are more impaired after additional CRT given for central nervous system (CNS) disease.10,11 However, other factors must contribute to GHD found after TBI alone at doses of 10–14 Gy. In vitro studies have also demonstrated additional attenuation of GH levels after chemotherapy, and it has been suggested that alkylating agents and antimetabolites may also be important additional factors affecting GH secretion.9,12 In addition, it is likely that chemotherapy for ALL also has a direct effect on the growth plates.12,13 Other factors known to affect GH secretion, including age,14 gender,15 pubertal status16 and adiposity,17 may contribute to this variation in GH status. © 2015 John Wiley & Sons Ltd
Growth hormone abnormalities after BMT/TBI 509 Investigating GH status in BMT/TBI survivors requires careful consideration. Circulating IGF-I is unhelpful in the diagnosis of GHD in this group.18 ‘Neurosecretory dysfunction’,19 the term coined to encompass discrepancies between GH responses in provocation tests vs endogenous GH production, has also been suggested. GH treatment does improve growth in BMT/TBI survivors,20 but final height is still significantly compromised, presumably because of the other factors adversely affecting growth. Indeed, the long-term BMT/TBI survivors’ phenotype resembles the GHD phenotype, with short stature, poor pubertal growth, fatigue and reduced QOL,21 raising the possibility of a degree of GH resistance. TBI irradiates all tissues, including growth plates and gonads, potentially impacting on both the skeletal response to GH and the pubertal growth spurt. Final height after BMT/ TBI is more impaired in males and in those irradiated at a younger age,5,6 and pubertal growth after TBI is particularly impaired1 principally reflecting poor spinal growth,22 a direct effect of spinal irradiation.23 Central adiposity and an increased risk of diabetes and cardiovascular disease18,24 are increasingly recognized in survivors of childhood cancer with and without GHD.25 It is not clear whether GH treatment improves these outcomes, or whether GHD after BMT/TBI persists into adulthood necessitating longterm GH treatment.22,26 In this study, we have investigated the GH–IGF-I axis in detail in BMT/TBI survivors in comparison with a control group with short stature, using endogenous overnight GH profiles, ITTs and IGF-I generation tests.
Methods Clinical details of study participants: BMT/TBI survivors Fig. 1 describes the recruitment of participants into the study (undertaken 2006–2008). Survivors of childhood BMT/TBI for haematological malignancy previously treated at a single centre between October 1988 and 2004, now aged 6–25 years and still resident in the region, were identified from the BMT database (n = 94) in 2006–2008. Of these, 32 requiring GH axis assessment were contacted, and the remainder were either already established on GH (n = 29) or did not have a clinical indication for GH testing or were not followed up in our centre (n = 33). Clinical indications for GH testing included poor height velocity, end of growth testing in those previously treated with GH, and symptoms of GHD in postpubertal survivors not previously tested. Twenty-two of 32 (12 male; aged 6–245 years and 88(14–192) years post-BMT/TBI) agreed to participate. All participants were euthyroid at the time of study, and five were on thyroxine replacement for primary hypothyroidism. All of the pubertal and postpubertal female BMT/TBI survivors were taking oestrogen replacement therapy (n = 7) and four of eight pubertal and postpubertal male BMT/TBI survivors were on testosterone replacement therapy. Survivors had ALL (n = 16), AML (n = 4) and CML (n = 2): 11 matched-related donor transplants, 10 matched-unrelated and one unrelated stem cell © 2015 John Wiley & Sons Ltd Clinical Endocrinology (2015), 83, 508–517
transplant. All received conditioning with cyclophosphamide (60 mg/kg) and Campath (1 mg/kg, maximum dose 50 mg) and TBI [144 Gy (eight fractions), n = 20; 10 Gy (single fraction), n = 2]. Four had a CNS boost of 6 Gy (four fractions), whilst two also received CNS prophylactic radiotherapy at doses of 12 Gy and 18 Gy (10 fractions), respectively. Ten survivors had developed graft-vs-host disease and received oral prednisolone treatment, whilst three received topical steroid treatment (see Fig. 1; Table 1). Short stature control participants Nineteen short stature control participants (13 male, aged 78–201 years) agreed to participate from 28 contacted who were due GH status assessment in the paediatric endocrinology clinic in 2006–2008. All were euthyroid. For ethical reasons, normal children were not recruited for ITTs. Participants all required GH testing for clinical reasons. The short stature control postpubertal participants had previously been diagnosed with isolated GHD in childhood (IGHD) and had therefore received GH treatment until the end of growth and now required repeat GH testing. None had taken glucocorticoid medication in the previous year or had other conditions likely to affect growth or pituitary function. Both BMT/TBI and short stature control groups included (i) GH-na€ıve participants having initial investigation of GH status and (ii) postpubertal participants having end of growth retesting after completion of GH treatment. Final height was defined as Tanner stage 4–5 on examination with a height velocity of
doi: 10.1111/cen.12773
ORIGINAL ARTICLE
Growth hormone deficiency after childhood bone marrow transplantation with total body irradiation: interaction with adiposity and age N.L. Davis*, C.E. Stewart†, A.D. Moss‡, W.W.W. Woltersdorf*, L.P. Hunt§, R.A. Elson*, J.M. Cornish¶, M.C.G. Stevens§ and E.C. Crowne* *Department of Paediatric Endocrinology and Diabetes, University Hospitals Bristol NHS Foundation Trust, Bristol, †Department of Stem Cells, Ageing and Molecular Physiology Unit, Research Institute for Sport & Exercise Sciences, Liverpool John Moores University, Liverpool, ‡Department of Exercise and Sport Science, Manchester Metropolitan University, Crewe, Cheshire, §School of Clinical Sciences, University of Bristol, and ¶Department of Paediatric Oncology, University Hospitals Bristol NHS Foundation Trust, Bristol, UK
Summary Objective Bone marrow transplantation with total body irradiation (BMT/TBI) has adverse effects on growth, growth hormone status and adiposity. We investigated the GH–IGF-I axis in relation to adiposity. Design Cross-sectional case control study. Patients BMT/TBI survivors (n = 22) and short stature control participants (n = 19), all GH-na€ıve or off GH treatment >3 months. Measurements Auxology, DEXA scans and GH–IGF-I axis investigation: (i) 12-h overnight GH profiles; (ii) insulin tolerance test (ITT); and (iii) IGF-I generation test. Analysis: autodeconvolution of GH profile data and comparison of quantitative parameters using ANOVA. Results Eighty-two percent of BMT/TBI survivors had growth hormone deficiency (GHD) using ITT. GH profile area-underthe-curve (GH-AUC) was reduced in BMT/TBI survivors vs short stature control participants [geometric mean (range) 209 (21–825) vs 428 (64–1400) mcg/l/12 h, respectively, P = 0007]. GHD was more marked in those who had additional cranial irradiation (CRT) [ITT peak 14 (02–30) vs TBI only 41 (11– 148) mcg/l, P = 0036]. GHD was more marked at the end of growth in BMT/TBI survivors vs short stature control participants (GH-AUC 551 (64–2474) vs 1369 (192–4197) mcg/l/12 h, respectively, P = 0011) and more prevalent (9/11 vs 1/9, respectively, P = 0005). GH profile data were consistent with ITT results in 80% of participants. IGF-I generation tests were normal. BMT/TBI survivors still demonstrated lower GH levels after
Correspondence: Dr Elizabeth Crowne, Department of Paediatric Endocrinology and Diabetes, Level 6 UH Bristol Education Centre, Upper Maudlin Street, Bristol BS2 8AE, UK. Tel.: +44 117 3420165; Fax: +44 117 3420186; E-mail: [email protected]
508
adjustment for adiposity (fat-adjusted mean difference for GHAUC 909 mcg/l/12 h, P = 0025). Conclusions GHD was more prevalent in BMT/TBI survivors than expected for the CRT dose in TBI, worsened with time and persisted into adulthood. GHD could not be explained by adiposity. There was no evidence of GH neurosecretory dysfunction or resistance after BMT/TBI. (Received 22 September 2014; returned for revision 24 October 2014; finally revised 7 March 2015; accepted 12 March 2015)
Introduction Poor growth, found consistently after childhood bone marrow transplantation with total body irradiation (BMT/TBI), is multifactorial.1 GHD is a well-established consequence of high-dose CRT2,3 and is dose, fractionation and time since treatment dependent. Its reported prevalence varies following lower dose regimens (12-24 Gy),4 including TBI.5–7 The biologically effective dose (BED) of irradiation to the pituitary should relate to hypothalamic and pituitary late effects in terms of GHD,8,9 as it represents the effects of ionizing radiation delivered and takes into account the type of radiation and the nature of the tissue being irradiated. Unsurprisingly, both GH secretion and growth are more impaired after additional CRT given for central nervous system (CNS) disease.10,11 However, other factors must contribute to GHD found after TBI alone at doses of 10–14 Gy. In vitro studies have also demonstrated additional attenuation of GH levels after chemotherapy, and it has been suggested that alkylating agents and antimetabolites may also be important additional factors affecting GH secretion.9,12 In addition, it is likely that chemotherapy for ALL also has a direct effect on the growth plates.12,13 Other factors known to affect GH secretion, including age,14 gender,15 pubertal status16 and adiposity,17 may contribute to this variation in GH status. © 2015 John Wiley & Sons Ltd
Growth hormone abnormalities after BMT/TBI 509 Investigating GH status in BMT/TBI survivors requires careful consideration. Circulating IGF-I is unhelpful in the diagnosis of GHD in this group.18 ‘Neurosecretory dysfunction’,19 the term coined to encompass discrepancies between GH responses in provocation tests vs endogenous GH production, has also been suggested. GH treatment does improve growth in BMT/TBI survivors,20 but final height is still significantly compromised, presumably because of the other factors adversely affecting growth. Indeed, the long-term BMT/TBI survivors’ phenotype resembles the GHD phenotype, with short stature, poor pubertal growth, fatigue and reduced QOL,21 raising the possibility of a degree of GH resistance. TBI irradiates all tissues, including growth plates and gonads, potentially impacting on both the skeletal response to GH and the pubertal growth spurt. Final height after BMT/ TBI is more impaired in males and in those irradiated at a younger age,5,6 and pubertal growth after TBI is particularly impaired1 principally reflecting poor spinal growth,22 a direct effect of spinal irradiation.23 Central adiposity and an increased risk of diabetes and cardiovascular disease18,24 are increasingly recognized in survivors of childhood cancer with and without GHD.25 It is not clear whether GH treatment improves these outcomes, or whether GHD after BMT/TBI persists into adulthood necessitating longterm GH treatment.22,26 In this study, we have investigated the GH–IGF-I axis in detail in BMT/TBI survivors in comparison with a control group with short stature, using endogenous overnight GH profiles, ITTs and IGF-I generation tests.
Methods Clinical details of study participants: BMT/TBI survivors Fig. 1 describes the recruitment of participants into the study (undertaken 2006–2008). Survivors of childhood BMT/TBI for haematological malignancy previously treated at a single centre between October 1988 and 2004, now aged 6–25 years and still resident in the region, were identified from the BMT database (n = 94) in 2006–2008. Of these, 32 requiring GH axis assessment were contacted, and the remainder were either already established on GH (n = 29) or did not have a clinical indication for GH testing or were not followed up in our centre (n = 33). Clinical indications for GH testing included poor height velocity, end of growth testing in those previously treated with GH, and symptoms of GHD in postpubertal survivors not previously tested. Twenty-two of 32 (12 male; aged 6–245 years and 88(14–192) years post-BMT/TBI) agreed to participate. All participants were euthyroid at the time of study, and five were on thyroxine replacement for primary hypothyroidism. All of the pubertal and postpubertal female BMT/TBI survivors were taking oestrogen replacement therapy (n = 7) and four of eight pubertal and postpubertal male BMT/TBI survivors were on testosterone replacement therapy. Survivors had ALL (n = 16), AML (n = 4) and CML (n = 2): 11 matched-related donor transplants, 10 matched-unrelated and one unrelated stem cell © 2015 John Wiley & Sons Ltd Clinical Endocrinology (2015), 83, 508–517
transplant. All received conditioning with cyclophosphamide (60 mg/kg) and Campath (1 mg/kg, maximum dose 50 mg) and TBI [144 Gy (eight fractions), n = 20; 10 Gy (single fraction), n = 2]. Four had a CNS boost of 6 Gy (four fractions), whilst two also received CNS prophylactic radiotherapy at doses of 12 Gy and 18 Gy (10 fractions), respectively. Ten survivors had developed graft-vs-host disease and received oral prednisolone treatment, whilst three received topical steroid treatment (see Fig. 1; Table 1). Short stature control participants Nineteen short stature control participants (13 male, aged 78–201 years) agreed to participate from 28 contacted who were due GH status assessment in the paediatric endocrinology clinic in 2006–2008. All were euthyroid. For ethical reasons, normal children were not recruited for ITTs. Participants all required GH testing for clinical reasons. The short stature control postpubertal participants had previously been diagnosed with isolated GHD in childhood (IGHD) and had therefore received GH treatment until the end of growth and now required repeat GH testing. None had taken glucocorticoid medication in the previous year or had other conditions likely to affect growth or pituitary function. Both BMT/TBI and short stature control groups included (i) GH-na€ıve participants having initial investigation of GH status and (ii) postpubertal participants having end of growth retesting after completion of GH treatment. Final height was defined as Tanner stage 4–5 on examination with a height velocity of
