Acute lymphoblastic leukemia, or ALL, accounts for about 28 percent of cancers in children under fourteen and 13 percent in adolescents aged fifteen to nineteen, with an incidence of 58.7 cases per million children per year. Thanks to modern chemotherapy protocols and targeted treatments, survival rates now approach 90 percent. That success has shifted clinical attention toward the early and late adverse effects of therapy, and bone disorders rank among the most consequential. Fractures occur in 16 to 36 percent of patients around ALL onset or during treatment, mostly within the first two years and peaking in the spine at roughly twelve months. Long-term survivors face reduced bone mineral density and a six-fold increased risk of fragility fractures compared with healthy peers, with culprits including glucocorticoids, methotrexate, reduced weight-bearing activity, malnutrition, and endocrine disturbances.

The Italian team enrolled 49 patients aged three to eighteen who were treated under the AIEOP-BFM ALL 2009 protocol between March 2012 and April 2016. Participants underwent dual-energy X-ray absorptiometry, or DXA, scans and extensive blood and urine testing at seven timepoints: at diagnosis, every six months thereafter, and until one year after the end of the 24-month treatment program. The researchers measured both lumbar spine bone mineral apparent density, a volume-corrected metric for the vertebrae, and total body less than head bone mineral density, chosen because the growing skull contributes unpredictably to total-body readings in children. All scans were performed on a single densitometer and reviewed by one expert to eliminate inter-observer variability, and body composition indexes were expressed as standard deviation scores against age- and sex-specific pediatric reference ranges.

The baseline results were striking. At diagnosis, before any chemotherapy had begun, 71.7 percent of patients had lumbar spine Z-scores at or below -2.0, with a median of -2.8, indicating severely reduced vertebral bone density. Yet total body bone density was essentially normal, exceeding -2.0 in 97.8 percent of patients. The researchers attribute this dissociation to the biology of the disease itself: leukemic blasts infiltrate trabecular bone, the spongy, metabolically active tissue that predominates in the lumbar vertebrae and serves as a site of hematopoiesis, while cortical bone throughout the rest of the skeleton remains largely spared. Tumor-released cytokines may further erode trabecular integrity. Because the study’s strict enrollment criteria excluded children with prior steroid exposure, immobilization, or conditions affecting bone health, the authors argue that the malignant cells themselves are the most plausible explanation for the spinal damage present at day one.

The picture reversed once treatment began. Total body bone density plummeted within the first six months, reaching its nadir twelve to eighteen months after diagnosis, a timing that coincides with the most intensive phase of chemotherapy, when patients receive high-dose corticosteroids known to trigger apoptosis in both osteoblasts and osteoclasts, along with methotrexate and prolonged hospitalization that drastically reduces weight-bearing activity. Linear mixed-effects models confirmed the pattern statistically: at twelve months, total body Z-scores were on average 1.66 standard deviations below baseline, the largest drop recorded. In contrast, lumbar spine density stayed flat during treatment, suggesting that therapy added little further harm to bone already devastated by the leukemia. Recovery began after the twenty-fourth month, but it was incomplete on both fronts.

Biochemical markers told a parallel story of bone in turmoil. Vitamin D levels below 20 nanograms per milliliter were found in 40.8 percent of patients at diagnosis, and 18.4 percent showed elevated parathyroid hormone. During the first six months of therapy, both bone formation markers, including bone-specific alkaline phosphatase and osteocalcin, and the resorption marker beta-CTX rose significantly, signaling heightened bone turnover accompanied by increased serum and urinary calcium and compensatory parathyroid activation. The researchers also tracked the RANKL-to-OPG ratio, a molecular dial that governs osteoclast activity, and observed it fall between months twelve and eighteen, which they attribute to the withdrawal of steroids and the start of methotrexate-based maintenance therapy, which suppresses RANKL secretion. Notably, alkaline phosphatase correlated negatively with spine density as late as one year after treatment ended.

Body composition shifted in a worrying direction simultaneously. Lean mass percentage declined steadily from the first six months onward, dropping from a median of 74.7 percent at diagnosis to 68.7 percent at the three-year mark, while fat mass rose progressively, a pattern consistent with the sarcopenic obesity reported in long-term ALL survivors, which one prior study found in 43 percent of patients a decade or more after diagnosis. The most clinically meaningful result emerged from multivariable analysis: improvements in the appendicular skeletal muscle mass index over time were significantly associated with recovery of total body bone density between months six and thirty-six (p = 0.008), while fat mass changes showed no such link. Age also mattered, with younger children showing worse spine density at diagnosis but milder, more recoverable losses in total body bone, possibly reflecting a greater bone turnover reserve.

The study’s authors suggest a practical implication: protecting and rebuilding lean muscle mass through timely, weight-bearing physical activity may help shield young survivors from cortical bone loss, though they caution that dedicated interventional studies are needed to confirm the causal pathway. Mechanistically, the connection is plausible, because muscle contraction loads the skeleton and lean mass is one of the strongest determinants of bone accrual during growth. The researchers also note limitations, including the modest sample size, the absence of systematic spinal imaging to detect asymptomatic vertebral compression fractures, unavoidable missed scans among the sickest patients, and the wide age range, which makes pubertal development difficult to disentangle.

Nevertheless, the prospective design, more than 250 DXA scans, and rigorous standardization lend unusual weight to the findings. No symptomatic fractures occurred during the three-year observation window, but when 43 surviving patients were contacted roughly 11.6 years after diagnosis, seven reported at least one fracture afterward, several involving the forearm, a classic site of fragility injury. With survival from childhood ALL now the norm rather than the exception, the study argues that bone health deserves a permanent place in survivorship care. Monitoring both spine and total body density, supporting muscle mass through nutrition and exercise, and correcting vitamin D deficiency may together determine whether today’s cured children face tomorrow’s fractures as adults.

Subject of Research: Longitudinal changes in bone mineral density, bone turnover, and body composition in children treated for acute lymphoblastic leukemia

Article Title: Longitudinal changes in bone mineral density, bone turnover markers, and body composition in children and adolescents treated for acute lymphoblastic leukemia

Article References: Cattoni, A., Molinari, S., Capitoli, G., Vai, S., Carrara, S., Galliera, E., Massaccesi, L., Nicolosi, M. L., Radaelli, S., Sarra, M., Tondelli, D., Corso, R., Speziali, G., Biondi, A., Balduzzi, A., Corbetta, S., & Sala, A. (2026). Longitudinal changes in bone mineral density, bone turnover markers, and body composition in children and adolescents treated for acute lymphoblastic leukemia. Frontiers in Endocrinology, 17, Article 1901073. https://doi.org/10.3389/fendo.2026.1901073

Image Credits: AI Generated

DOI: 10.3389/fendo.2026.1901073

Keywords: acute lymphoblastic leukemia, bone mineral density, pediatric oncology, DXA, body composition, lean mass, bone turnover markers, osteoporosis, glucocorticoids, childhood cancer survivors, sarcopenic obesity, parathyroid hormone