Frequency and clinical significance of the expression of the multidrug resistance proteins MDR1/P-glycoprotein, MRP1, and LRP in acute myeloid leukemia. A Southwest Oncology Group Study
Cp. Leith et al., Frequency and clinical significance of the expression of the multidrug resistance proteins MDR1/P-glycoprotein, MRP1, and LRP in acute myeloid leukemia. A Southwest Oncology Group Study, BLOOD, 94(3), 1999, pp. 1086-1099
Therapeutic resistance is a major obstacle in the treatment of acute myeloi
d leukemia (AML). Such resistance has been associated with rapid drug efflu
x mediated by the multidrug resistance gene 1 (MDR1; encoding P-glycoprotei
n) and more recently with expression of other novel proteins conferring mul
tidrug resistance such as MRP1 (multidrug resistance-associated protein 1)
and LRP (lung resistance protein). To determine the frequency and clinical
significance of MDR1, MRP1, and LRP in younger AML patients, we developed m
ultiparameter flow cytometric assays to quantify expression of these protei
ns in pretreatment leukemic blasts from 352 newly diagnosed AML patients (m
edian age, 44 years) registered to a single clinical trial (SWOG 8600). Pro
tein expression was further correlated with functional efflux by leukemic b
lasts [assessed using two substrates: Di(OC)(2) and Rhodamine 123] and with
the ability of MDR-reversing agents to inhibit efflux in vitro. MDR1/P-gly
coprotein expression, which was highly correlated with cyclosporine-inhibit
ed efflux, was noted in only 35% of these younger AML patients, distinctly
lower than the frequency of 71% we previously reported in AML in the elderl
y (Blood 89:3323, 1997). Interestingly, MDR1 expression and functional drug
efflux increased with patient age, from a frequency of only 17% in patient
s less than 35 years old to 39% in patients aged 50 years (P =.010). In con
trast, MRP1 was expressed in only 10% of cases and decreased with patient a
ge (P =.024). LRP was detected in 43% of cases and increased significantly
with increasing white blood cell counts (P =.0015). LRP was also marginally
associated with favorable cytogenetics (P =.012) and French-American-Briti
sh (FAB) AML FAB subtypes (P =.013), being particularly frequent in M4/M5 c
ases. Only MDR1/P-glycoprotein expression and cyclosporine inhibited efflux
were significantly associated with complete remission (CR) rate (P-MDR1 =
012; P-efflux =.039) and resistant disease (RD; P-MDR1 =.0007; P-efflux =.0
092). No such correlations were observed for MRP1 (P-CR = .93; P-RD =.55) o
r LRP (P-CR = 50; P-RD =.53). None of these parameters were associated with
overall or relapse-free survival. Unexpectedly, a distinct and nonoverlapp
ing phenotype was detected in 18% of these cases: cyclosporine-resistant ef
flux not associated with MDR1, MRP1, or LRP expression, implying the existe
nce of other as yet undefined efflux mechanisms in AML. In summary, MDR1 is
less frequent in younger AML patients, which may in part explain their bet
ter response to therapy. Neither MRP1 nor LRP are significant predictors of
outcome in this patient group. Thus, inclusion of MDR1 modulators alone ma
y benefit younger AML patients with MDR1(+) disease. (C) 1999 by The Americ
an Society of Hematology.