Relationship between major vault protein/lung resistance protein, multidrug resistance-associated protein, P-glycoprotein expression, and drug resistance in childhood leukemia

Cellular drug resistance is related to a poor prognosis in childhood leukemia, but little is known about the underlying mechanisms. We studied the expression of P-glycoprotein (P-gp), multidrug resistance (MDR)-associated protein (MRP), and major vault protein/lung resistance protein (LRP) in 141 children with acute lymphoblastic leukemia (ALL) and 27 with acute myeloid leukemia (AML) by flow cytometry. The expression was compared between different types of leukemia and was studied in relation with clinical risk indicators and in vitro cytotoxicity of the MDR-related drugs daunorubicin (DNR), vincristine (VCR), and etoposide (VP16) and the non-MDR-related drugs prednisolone (PRD) and L-asparaginase (ASP). In ALL, P-gp, MRP, and LRP expression did not differ between 112 initial and 29 unrelated relapse samples nor between paired initial and relapse samples from 9 patients. In multiple relapse samples, LRP expression was 1.6-fold higher compared with both initial (P = .026) and first relapse samples (P = .050), which was not observed for P-gp and MRP. LRP expression was weakly but significantly related to in vitro resistance to DNR (Spearman's rank correlation coefficient 0.25, P = .016) but not to VCR, VP16, PRD, and ASP. No significant correlations were found between P-gp or MRP expression and in vitro drug resistance. Samples with a marked expression of two or three resistance proteins did not show increased resistance to the tested drugs compared with the remaining samples. The expression of P-gp, MRP, and LRP was not higher in initial ALL patients with prognostically unfavorable immunophenotype, white blood cell count, or age. The expression of P-gp and MRP in 20 initial AML samples did not differ or was even lower compared with 112 initial ALL samples. However, LRP expression was twofold higher in the AML samples (P < .001), which are more resistant to a variety of drugs compared with ALL samples. In conclusion, P-gp and MRP are unlikely to be involved in drug resistance in childhood leukemia. LRP might contribute to drug resistance but only in specific subsets of children with leukemia.