Children treated for acute lymphoblastic leukemia may develop reduced bone mineral density during treatment, but there is little information on the mechanisms involved. In a prospective, longitudinal study on 15 children with ALL, we undertook serial measurements of markers of bone and collagen turnover, insulinlike growth factor (IGF)-I and its binding proteins (IGFBPs)-3 and -2 during the second year of continuing chemotherapy. In eight patients we also measured lower leg length by knemometry. Height SD scores, lower leg length velocity, IGF-I, and markers of bone collagen turnover did not differ significantly from healthy children. However, bone alkaline phosphatase, a marker of the differentiated osteoblast, was lower (mean SD score, Ϫ0.64; p Ͻ 0.0001), whereas procollagen type III N-terminal propeptide (P3NP, a marker of soft tissue collagen turnover; mean SD score, ϩ0.93, p Ͻ 0.05), IGFBP-3 (mean SD score, ϩ0.76; p Ͻ 0.01), and IGFBP-2 (mean SD score, ϩ1.24, p ϭ 0.01) were all higher than in healthy children. IGFBP-3 decreased during episodes of afebrile neutropenia (p Ͻ 0.05). Within 3 mo after completion of treatment, bone ALP increased in all eight patients, but collagen markers showed little change. IGFBP-2 returned to normal posttreatment, but P3NP and IG-FBP-3 remained significantly elevated compared with healthy children (mean SD scores, ϩ1.51 and ϩ1.36, respectively; p Ͻ 0.01). We conclude that continuing chemotherapy was associated with normal growth and bone collagen turnover but enhanced soft tissue collagen turnover. Bone bone alkaline phosphatase was low throughout treatment, which suggests impaired osteoblast differentiation resulting from a direct effect of chemotherapy on bone. Although the effect was reversible, the long-term implications for bone health in survivors remain uncertain. There have been many reports of poor growth in children with acute lymphoblastic leukemia (ALL), especially during periods of intensive chemotherapy, but growth may return to normal during less-intensive periods and may show evidence of further catch-up after completion of chemotherapy (1-13). Of greater concern are a number of retrospective crosssectional studies on survivors of ALL that describe reduced bone mineral density (BMD) at various sites (14 -16). This may be associated with increased fracture risk, reduced peak bone mass, and a theoretical risk of osteoporosis in adult life. However, these previous studies reflect the outcome of earlier treatment protocols, most of which included cranial irradiation and are now obsolete. Current protocols generally use more intensive chemotherapy regimens but avoid cranial irradiation in all but a few high-risk cases. Recently, a number of prospective longitudinal studies have reported that the prevalence of radiologic osteopenia and fracture in children with ALL increases and that BMD decreases during chemotherapy, in the absence of cranial irradiation (17)(18)(19). However, the mechanisms responsible for these observations remain to be elucidated.There are a num...