Changes in Bone Metabolic Parameters Under Imatinib Treatment in Children with Chronic Myeloid Leukemia (CML)

Objectives: Tyrosine kinase inhibitors (TKIs) demonstrate outstanding efficacy for treatment of CML, however, also exert well-known off-target effects. Among these, inhibition of osteoclast's and osteoblast's growth and differentiation [Vandyke K et al Blood 2010] results in longitudinal growth impairment in children and adolescents [Shima H et al J Pediatr 2011; Millot F et al Eur J Cancer 2014; Tauer JT et al Blood 2014, ASH abstract 522] under longtime imatinib treatment. In children enrolled into trial CML-paed II on ongoing imatinib exposure we investigated blood levels of calcium, phosphate, 25-vitamin D3 [25-vitD3], 1.25-vitamin D3 [1.25-vitD3], parathyroid hormone [PTH], and metabolic markers of bone resorption (osteoclasts: C-terminal cross-linking telopeptide of collagen type I [CTX-I]) and bone formation (osteoblasts: bone alkaline phosphatase [BAP], osteocalcin [OC], procollagen type 1 N propeptide [PINP]) in order to get a deeper insight in alterations associated with bone remodeling in a longitudinal timely fashion. 

Methods: 119 patients [pts] (70 male/49 female, median age 12 years, range 1-18 yrs) enrolled into the trial received 260-340 mg imatinib/sqm daily within one week after diagnosis (Dx) when CML had been confirmed by either cytogenetics or PCR. 20 to 30 pts out of this cohort could successfully be monitored repeatedly over a median period of 3 years for all parameters planned to be analyzed by collecting blood and urine three-monthly under appropriate circumstances (immediate processing, freezing, light protection, dry ice shipment to study center). Assays were performed in a central laboratory as described previously [Jaeger BA et al Med Sci Monit 2012]. Age normalized reference values were used as standard deviation scores (SDS) for statistical analysis by REML fit linear mixed effect-model with software R version 3.1.1. [https://www.r-project.org/].

Results: At Dx mean OC levels were found elevated to the upper normal range (+1.5 SDS) but increased further on up to +5 SDS within the next 3 mo. Also mean BAP-SDS and PINP-SDS increased during follow-up compared to measurements at diagnosis but remained within the normal range. At mo 3 increased CTX-I (+3 SDS) indicated enhanced bone resorption followed by elevated measurements in the upper normal range (+1.5 SDS) over the next 33 mo. 25-vitD3 levels were found stable in the very low but still normal -2 SDS range in the summer time (May - Oct.) but below that range (-2.5 SDS) in the winter time (Nov.-April). 1.25-vitD3 measurements showed a course with a sharp and steep increase from -2 SDS at Dx to + 2 SDS at mo 3 of treatment falling to normal levels (0 SDS) at mo 12 and thereafter. Serum calcium and phosphate remained in the normal range throughout the whole treatment period of 33 mo analyzed while all pts showed low urinary calcium (-6 SDS) at diagnosis and at mo 3 followed by levels in the lowest normal range (-2 SDS) from mo 9 onwards. Urinary phosphate levels also gradually increased from -6 SDS at Dx to normal values at mo 12. PTH ranged within +1SDS to +2 SDS during the total period of monitoring.        

Conclusion: Children with CML under imatinib treatment exhibit moderate secondary hyperparathyroidism in conjunction with pathologically low 25-vitD3 levels. OC as marker of bone formation is highly elevated possibly due to inhibition of osteoclasts' activity by imatinib via blockade of c-FMS. Bone resorption is also increased, however, to a lesser extent (+ 1.5 SDS) as demonstrated by elevated CTX-I in the upper normal range (+2 SDS). This dysbalanced osseous activity affects bone remodelling and in not-outgrown pts causes longitudinal growth impairment. Most importantly, a typical pattern of changes in most osseous metabolic markers is observed showing an initial increase at mo 3 (phase 1) followed by stable, but not normalized values (phase 2) from mo 9 onwards [see Fig. 1]. This is in line with the clinical observation that skeletal pain may be severe under TKI treatment but usually disappears by the end of the first year. Also in rodent models a typical biphasic spatio-temporal shifting of bone formation and resorption at the epiphyseal line could be demonstrated [Tauer JT et al PLoS One 2015]. These osseous changes mandate close monitoring in pediatric patients under long-term TKI exposure.