Background : The complete remission rate of the first induction chemotherapy is 60~80% in acute leukemia. Most patients who achieve complete remission will ultimately relapse within the first year and are much less responsive to the therapy with a bri...
Background : The complete remission rate of the first induction chemotherapy is 60~80% in acute leukemia. Most patients who achieve complete remission will ultimately relapse within the first year and are much less responsive to the therapy with a brief second remission in a minority. Acquisition of drug resistance in cellular level is the major cause of treatment failure. Therefore, the identification of the mechanisms of drug resistance is an important goal of current treatment. One mechanism that has been extensively studied in vitro is a multidrug resistance (MDR) mediated by P-glycoprotein which is the product of the MDR1 gene, but there is a little information on it`s clinical application. Methods : In order to find out the suitable method for detecting MDR, and furthermore, to see if the detection of MDR can be used in predicting treatment results, here, the existence of MDR1 amplification/overexpression in DNA, RNA and P-glycoprotein level was studied. Results : P-glycoprotein overexpression by immunohis-tochemical staining method was more sensitive (55.4%) than KNA amplification (0.0%) and mRNA overexpression (7.7%) in detecting MDR in acute leukemic patients. The incidence of P-glycoprotein overexpression was higher in relapsed (78.6%) and in refractory patients (88.9%) than in initial state (33.3%). There was a tendency of low complete remission rate in patients with primary P-glycoprotein overexpression. The incidence of acquired P-glycoprotein overexpression was 71.4%. When comparing the correlation between the P-glycoprotein overexpressin and the clinical results, the predictability for th treatment failure in P-glycoprotein overexpressing patients was acceptable (84%), but the predictability for the remission in P-glycoprotein non-expressing patients was low (54%). These predictability were similar to the predictability for drug resistance by comparing the correlation between the P-glycoprotein overexpression and in vitro cytotoxic assay results; the predictability for drug resistance to adriamycin in P-glycoprotein overexpressing patients was acceptable (75%), but the predictability of drug sensitivity in P-glycoprotein non-expressing patients was low (57%). Combination of verapamil and cytotoxic drug reversed MDR in 33.3% for adriamycin and 16.7% for vincristine in P-glycoprotein overexpressing patents. Conclusions : These results suggest the P-glycoprotein by immunohistochemical staining method could be a useful method for predicting treatment results. However, the low predictability for the remissionn in P-glycoprotein non-expressing patients, suggested by some other drug resistance mechanisms, can be overcome by combination with in vitro cytotoxic assay. This fact might contribute to the better treatment results in acute leukemia by optimization of drug selection.