Pharmacokinetic, pharmacodynamic, and pharmacogenetic determinants of osteonecrosis in children with acute lymphoblastic leukemia

Abstract

Osteonecrosis is a severe glucocorticoid-induced complication of acute lymphoblastic leukemia treatment. We prospectively screened children (n = 364) with magnetic resonance imaging of hips and knees, regardless of symptoms; the cumulative incidence of any (grade 1-4) versus symptomatic (grade 2-4) osteonecrosis was 71.8% versus 17.6%, respectively. We investigated whether age, race, sex, acute lymphoblastic leukemia treatment arm, body mass, serum lipids, albumin and cortisol levels, dexamethasone pharmacokinetics, and genome-wide germline genetic polymorphisms were associated with symptomatic osteonecrosis. Age more than 10 years (odds ratio, = 4.85; 95% confidence interval, 2.5-9.2; P = .00001) and more intensive treatment (odds ratio = 2.5; 95% confidence interval, 1.2-4.9; P = .011) were risk factors and included as covariates in all analyses. Lower albumin (P = .05) and elevated cholesterol (P = .02) associated with symptomatic osteonecrosis, and severe (grade 3 or 4) osteonecrosis was linked to poor dexamethasone clearance (P = .0005). Adjusting for clinical features, polymorphisms of ACP1 (eg, rs12714403, P = 1.9 × 10(-6), odds ratio = 5.6; 95% confidence interval, 2.7-11.3), which regulates lipid levels and osteoblast differentiation, were associated with risk of osteonecrosis as well as with lower albumin and higher cholesterol. Overall, older age, lower albumin, higher lipid levels, and dexamethasone exposure were associated with osteonecrosis and may be linked by inherited genomic variation.

Figure 1

Therapy, testing, and osteonecrosis on Total XV study. (A) Schema of sample collection, asparaginase and glucocorticoid dosing, and MRI screening. Blood samples were collected on consolidation day 15 and week 2, week 7, week 8, and week 12 of continuation phase of therapy. Prednisone was administered at 40 mg/m² per day during remission induction. Dexamethasone was given at 12 mg/m² per day (SR/HR arm) and 8 mg/m² per day (LR) during continuation weeks 1 to 6 and 10 to 16; and at 8 mg/m² per day for both treatment arms at weeks 7, 9, 17, and 19. Asparaginase was administered at 10 000 U/m² per dose intramuscularly 3 times weekly for 6 to 9 doses during remission induction; those in SR/HR arms received 25 000 U/m² per week from weeks 1 to 19; those in the LR arm received 10 000 U/m² 3 times weekly for 9 doses at weeks 7-9 and 17-19. MRI screens were scheduled to be performed at weeks 9-10 and 19-20. (B) Cumulative incidence of asymptomatic and symptomatic osteonecrosis. The top curve represents the first incidence of asymptomatic osteonecrosis in those whose worst grade was grade 1; and lower curve, the first incidence of symptomatic osteonecrosis in those who eventually developed grade 2 to 4 osteonecrosis.

Figure 2

Age and treatment arm were associated with symptomatic osteonecrosis. Percentage of patients who developed grade 2 to 4 osteonecrosis by (A) age, (B) treatment arms, (C) race, and (D) sex.

Figure 3

Lower serum albumin and higher cholesterol were associated with symptomatic osteonecrosis (ON). After adjustment for age and treatment arm, (A) serum albumin levels (g/dL) were lower and (B) cholesterol levels (mg/dL) were higher at week 8 in patients who did versus did not develop grade 2 to 4 osteonecrosis.

Figure 4

Dexamethasone AUC was associated with severe osteonecrosis. (A) Dexamethasone AUC (nM × h) at week 8 in patients with grade 0 or 1, grade 2, and grade 3 or 4 osteonecrosis (ON). Statistical significance was determined after adjustment for age and treatment arm. (B) Representative plasma concentration versus time curves in patients with low, medium, and higher dexamethasone exposure.

Figure 5

SNPs in ACP1 were associated with symptomatic osteonecrosis, lower albumin, and higher cholesterol levels. After adjustment for age and treatment arm, association of SNP genotypes in ACP1 (rs12714403) to (A) grade 2 to 4 osteonecrosis (ON), (B) albumin levels, and (C) cholesterol levels at week 8.

Figure 6

CART analysis. Age, sex, race, treatment arm, body mass index, dexamethasone AUC, cholesterol, albumin, and 20 SNPs described in supplemental Table 8 were allowed to compete in the CART analysis. Black and white colors in the pie chart represent the percentage of patients with grade 0 to 1 and grade 2 to 4 osteonecrosis (ON), respectively. Each number indicates the number of patients at the branch point.