BONE MASS AND BODY COMPOSITION AFTER CESSATION OF THERAPY FOR CHILDHOOD CANCER Karsten NYSOM 1 , Christian MØLGAARD 3 , Kirsten HOLM 2 , Henrik HERTZ 1 and Kim FLEISCHER MICHAELSEN 3 1 Paediatric Haematology and Oncology Section, Juliane Marie Centre, Rigshospitalet, Copenhagen, Denmark 2 Growth and Reproduction Section, Juliane Marie Centre, Rigshospitalet, Copenhagen, Denmark 3 Research Department of Human Nutrition and Centre for Advanced Food Studies, Royal Veterinary and Agricultural University, Frederiksberg, Denmark O ur aim was to review current information on body composition and bone mass after cessation of therapy for childhood cancer and to present preliminary data on body composition and bone mass in a group of Danish survivors of childhood leukaemia or lymphoma. Elevated body-mass in- dex (weight/height 2 ; BMI) is frequent after treatment for childhood acute lymphoblastic leukaemia. BMI increases dur- ing therapy or within the first year after therapy and remains abnormal thereafter. T reatment with corticosteroids, abnor- mal growth-hormone secretion after treatment with cranial irradiation (CI) or corticosteroids, younger age at diagnosis, or female gender were risk factors for elevated BMI in earlier studies. W e evaluated 185 survivors of childhood leukaemia or lymphoma by dual-energy X -ray absorptiometry scanning. W e found elevated whole-body relative fat mass, which was associated with CI. O ther studiesfound reduced bone massin the radius, the lumbar spine and the whole body after treatment for childhood cancer. Growth-hormone deficiency that is not adequately corrected, CI, reduced height or reduced weight were risk factors for reduced bone mass. In our 185 participants, the whole-body bone mass was also reduced significantly compared with reference values. CI and older age at follow-up were risk factors for reduced bone mass. W e conclude that the elevated relative fat mass and reduced bone massseen after treatment for childhood leukae- mia or lymphoma is associated mainly with CI. Int. J. Cancer Supplement 11: 40–43, 1998.  1998 Wiley-Liss, Inc. With improved survival after childhood cancer, it has become more important to limit delayed side-effects in the survivors. Late effects after treatment for childhood cancer have been described for many organ systems. The frequency and severity of late effects depends primarily on the organ in question, on the original disease and on the methods of treatment. Impaired growth and pubertal development have long been recognized as frequent complications after treatment for childhood cancer. Less attention has been paid to related outcomes, such as elevated relative fat mass and reduced bone mass. The aim of the present paper is to review current information on body composition and bone mass after cessation of therapy for childhood cancer and to present preliminary data on body composition and bone mass in a group of Danish survivors of childhood leukaemia or lymphoma. EARLIER STUDIES Body composition Elevated body-mass index (weight/height 2 , BMI) (Zee and Chen, 1986; Schell et al., 1992; Odame et al., 1994; Didi et al., 1995; Van Dongen-Melman et al., 1995), weight for height (Sainsbury et al., 1985; Groot-Loonen et al., 1996), or skin-fold thickness (Dacou Voutetakis et al., 1993) are frequent after treatment for childhood acute lymphoblastic leukaemia (ALL). BMI increases during treatment of the malignancy (Van Dongen- Melman et al., 1995) or during the first year after cessation of therapy (Zee and Chen, 1986) and remains abnormal during the following years. Several risk factors for elevated BMI after childhood ALL have been suggested, including treatment with corticosteroids (Didi et al., 1995; Van Dongen-Melman et al., 1995), in particular dexamethasone (Van Dongen-Melman et al., 1995), abnormal growth-hormone secretion caused by cranial irradiation (CI) or dexamethasone (Zee and Chen, 1986; Groot- Loonen et al., 1996), and younger age at diagnosis (Sainsbury et al., 1985; Didi et al., 1995). In some studies, elevated BMI was more frequent (Odame et al., 1994) or severe (Didi et al., 1995) in females than in males, but other studies could not confirm this finding (Dacou Voutetakis et al., 1993; Van Dongen-Melman et al., 1995; Groot-Loonen et al., 1996). Significantly elevated whole-body relative fat mass measured by dual-energy X-ray-absorptiometry (DXA) scanning has been found in girls, but not in boys, an average of 7 years after treatment for ALL (Warner et al., 1997). All patients treated for ALL had received CI. Elevated relative fat mass was associated with reduced energy expenditure during exercise testing in boys and in girls treated for ALL. The energy expenditure of patients treated for other malignancies without CI was similar to that of controls. The authors suggested that the reduced energy expenditure, and thus the elevated relative fat mass, after treatment for childhood ALL may be caused by reduced heart-stroke volume after anthracycline therapy. Most ALL patients had received cumulative anthracycline doses ranging from 90 to 270 mg/m 2 . In patients treated for malignancies other than ALL, however, relative fat mass and energy expenditure were normal, even though 9 of 21 patients had received a median anthracycline dose of 300 mg/m 2 . Bone mass Bone-mineral status can be evaluated by different methods, including single photon absorptiometry, dual photon absorptiom- etry, DXA, quantitative computed tomography (CT), and ultraso- nography. Results can be expressed as crude bone-mineral mass (e.g., bone-mineral content of the lumbar spine in g), mineral mass adjusted for the projected bone area (e.g., bone-mineral areal density of the lumbar spine in g/cm 2 ), or mineral mass adjusted for the bone volume (e.g., bone-mineral volumetric density of the lumbar spine in g/cm 3 ). The bone-mineral volumetric density, which can be measured only by CT scanning, is independent of bone size, whereas the bone-mineral content and the bone-mineral areal density both increase with increasing bone size. The bone- mineral areal density increases with increasing bone size, since it is corrected for bone size in only 2 dimensions. Increasing bone thickness, in parallel with increasing projected bone area, results therefore in increasing bone-mineral areal density even when the bone-mineral volumetric density remains unaltered. Bone-mineral areal density is used widely in osteoporosis research, since in adult women reduced bone-mineral areal density is strongly associated with increased risk of osteoporotic fractures (Kanis et al., 1994). Bone-mineral areal density is, however, not as suitable for analyzing bone mineral in growing subjects, mainly Grant sponsors: Torkil Steenbeck Foundation; Ville Heise Foundation; Rosalie Petersen Foundation; FØTEK (Food Technology Research and Development Programme; Danish Dairy Research Foundation. Correspondence to: Paediatric Haematology and Oncology, Section 4074, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen Ø, Denmark. Fax: (45) 35-456–543. E-mail: nysom@inet.uni2.dk Int. J. Cancer: Supplement 11, 40–43 (1998)  1998 Wiley-Liss, Inc. Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer