Keywords

Full Text

References

Figures

Tables

Drug resistance factors in acute myeloid leukemia: a comparative analysis

M Filipits¹, T Stranzl¹, G Pohl¹, H Heinzl³, U Jäger², K Geissler², C Fonatsch⁴, OA Haas⁵, K Lechner² & R Pirker¹

¹Division of Oncology, Department of Internal Medicine I, University of Vienna Medical School, Austria

²Division of Hematology, Department of Internal Medicine I, University of Vienna Medical School, Austria

³Department of Medical Computer Sciences, University of Vienna Medical School, Austria

⁴Department of Medical Biology, University of Vienna Medical School, Austria

⁵St Anna Kinderspital, Vienna, Austria

Correspondence to: R Pirker, Department of Internal Medicine I, Division of Oncology, Währinger Gürtel 18, A-1090 Vienna, Austria; Fax: (43) (1) 40400-4461

To compare the clinical relevance of drug resistance factors in de novo acute myeloid leukemia (AML), we determined their relationship to both response to induction chemotherapy and survival of the patients in univariate as well as multivariate analyses. The drug resistance factors immunocytochemically studied in 111 patients at the time of diagnosis included the lung resistance protein (LRP), P-glycoprotein (P-gp), multidrug resistance protein (MRP1) and bcl-2. In the univariate analyses, age (P = 0.005), karyotype (P = 0.03), LRP (P = 0.003), P-gp (P = 0.02) and bcl-2 (P = 0.03) predicted for response to induction chemotherapy, whereas MRP1 had no predictive value. Age (P = 0.05), karyotype (P = 0.05) and LRP (P = 0.03) retained their predictive value in the multivariate logistic regression analyses. With regard to overall survival, age (P = 0.008), karyotype (P = 0.006), LRP (P = 0.001) and P-gp (P = 0.01) were of prognostic value in the univariate Cox regression analyses but only age (P = 0.01), karyotype (P = 0.02) and LRP (P = 0.01) retained their prognostic significance in the multivariate analyses. A risk score based on the number of independent prognostic factors allowed division of patients into four groups with different outcome. In these groups, the complete remission rates were 93%, 75%, 47% and 33%, respectively, and median overall survival was 2.4, 1.2, 0.6 and 0.2 years, respectively. Thus, several drug resistance factors did predict outcome in the univariate analyses but LRP was the only drug resistance factor with independent predictive and prognostic significance. The proposed risk score might be useful for risk-adapted treatment in the future. Leukemia (2000) 14, 68–76.

Acute myeloid leukemia (AML) lends itself as a model disease for the evaluation of the clinically relevant mechanisms of drug resistance. Several mechanisms of multidrug resistance have previously been studied for their clinical relevance in AML (see Refs 1–3 for review). MDR1/P-glycoprotein expression was associated with poor response to induction chemotherapy and shorter survival,^4,5,6,7,8 whereas expression of the multidrug resistance protein (MRP1) had no impact on clinical outcome.^9,10 Expression of the lung resistance protein (LRP) was recently shown to correlate with lower response rates of induction chemotherapy and shorter survival by List et al¹¹ and our group.¹² Expression of the proto-oncogene bcl-2, which is involved in the regulation of apoptosis,¹³ indicated poor clinical outcome.¹⁴

To compare the clinical relevance of these mechanisms of drug resistance on a representative patient population, we studied four drug resistance factors (LRP, P-gp, MRP1, bcl-2) in addition to age and karyotype for their predictive value with regard to outcome of induction chemotherapy and their prognostic value with regard to survival in both univariate and multivariate analyses in 111 patients with de novo AML. Here we report the results of these studies.

Patients

One hundred and eleven patients (49 female, 62 male) with de novo AML were included in this study after obtaining informed consent. Eighty of these patients had been included in a previous study on the clinical relevance of LRP.¹² Patients were treated between January 1990 and February 1998. The characteristics of the patients are summarized in Table 1.

Patients received one to two cycles of standard induction chemotherapy. Chemotherapy consisted of daunorubicin 45 mg/m² daily on days 1–3 and cytarabine 200 mg/m² daily on days 1–7 (DA protocol) in 25 patients and additional etoposide 100 mg/m² daily on days 1–5 (DAE protocol) in 71 patients. Four patients were treated with idarubicin plus cytarabine (IA protocol). Eight patients with FAB subtype M3 received all-trans retinoic acid (ATRA) followed by chemotherapy. Three patients received intermediate-dose cytarabine followed by DAE as second induction chemotherapy cycle. Response to induction chemotherapy was assessed according to standard criteria.

Seventy-five of 78 patients in complete remission received consolidation therapy with conventional-dose cytarabine (n = 29) or high-dose cytarabine (n = 46). Nineteen patients underwent bone marrow transplantation.

Immunocytochemistry

Immunocytochemistry was performed as described.^9,12 Briefly, cells were fixed in cold acetone and incubated for 2 h with the primary monoclonal antibodies described below. Antibody binding was detected by the avidin–biotin– peroxidase method. Staining of at least 200 leukemic cells was evaluated independently by three investigators who had no previous knowledge of clinical data of the patients.

Because cut-off levels ranging from 5%^7,15,16 to 20%^{6,17,18,19,20} have been described for P-gp and from 5%²¹ to 20%⁹ for MRP1, we decided to divide P-gp and MRP1 expression into three categories (5%, 6–20%, >20% staining cells).

LRP:

The LRP-56 antibody was used (Alexis, Läufelfingen, Switzerland).¹² Dependent on the percentage of staining cells, LRP expression was divided into negative (<5% staining cells) and positive (5%) expression. The small-cell lung cancer cell line SW1573 and its drug-resistant variant SW1573/2R120 (both cell lines provided by Dr RJ Scheper) were used as negative and positive controls, respectively.²²

P-gp:

The C219 (Alexis) antibody was used.^9,12 Dependent on the percentage of staining cells, P-gp expression was divided into low (5%), intermediate (6–20%) and high (>20%) expression. Drug-sensitive KB-3-1 and multidrug-resistant KB-8-5 cells (provided by Drs I Pastan and MM Gottesman, National Cancer Institute, Bethesda, MD, USA) were used as negative and positive controls, respectively.⁴

MRP1:

A combination of QCRL-1/QCRL-3 antibodies was used.⁹ Dependent on the percentage of staining cells, MRP1 expression was divided into low (5%), intermediate (6–20%) and high (>20%) expression. C1 cells were used as negative controls and T5 cells as positive controls (provided by Drs SPC Cole and RG Deeley, Queen's University, Kingston, Canada).²³

bcl-2:

The bcl-2 antibody was used (Neomarkers, Fremont, CA, USA). Dependent on the percentage of staining cells, bcl-2 expression was divided into negative (20%) and positive (>20%).

To ensure specificity of staining and to overcome inter-assay variability, several controls were performed. These controls included negative and positive control cell lines. In addition, experiments without monoclonal antibodies were performed as negative controls in all cases, and controls with irrelevant isotype-specific antibodies were done in some cases. There were no differences in staining between the irrelevant isotype-specific antibodies and the negative controls without any primary antibody. Finally, we analyzed selected samples (at least one sample per expression category for C219, QCRL-1/QCRL-3, LRP-56 and bcl2-) at least three times in order to confirm the reproducibility of our immunocytochemical assay. In these cases, the repetitions gave consistent results with identical categories of expression for all drug resistance factors (data not shown).

Karyotype analysis

Good prognosis karyotype included inv(16), t(8;21) or t(15;17). Poor prognosis included del5q or -5, del7q or -7, +8, inv(3) or t(3;3), and complex aberrations. Cytogenetic abnormalities others than the ones described above or normal karyotype were regarded as indicators of intermediate prognosis.

Statistical analysis

Associations between clinical parameters and drug resistance factors were assessed by chi-squared tests. Survival probabilities were calculated with the product limit method according to Kaplan–Meier.²⁴ Overall survival time was defined as the period between the time of diagnosis and the time of death. Disease-free survival time was defined as the period between the time when complete remission was first demonstrated and the time of relapse or death. Survival times of patients still alive or patients who underwent bone marrow transplantation were censored with the last follow-up date or transplantation date, respectively. Logistic regression models and Cox proportional hazards regression models²⁵ were used to assess the independent effects of covariables on complete remission and survival. All P values are results of two-sided tests. The SAS statistical software system (SAS Institute, Cary, NC, USA, 1990) was used for calculations.

Expression of drug resistance factors

One hundred and eleven patients with de novo AML were studied and their characteristics are summarized in Table 1. Drug resistance factors were determined in leukemic cells at the time of diagnosis (Table 2). LRP expression was negative in 66% and positive in 34% of the patients. Low, intermediate and high expression of P-gp was seen in 59%, 29% and 12% of the patients, respectively, and of MRP1 in 30%, 47% and 23%. Expression of bcl-2 was negative in 46% and positive in 54% of the patients.

Next, the drug resistance factors were assessed for potential associations with either each other or clinical parameters. LRP positivity was associated with higher white blood cell count (P = 0.0003), higher lactate dehydrogenase serum levels (P = 0.03), and increased expression of P-gp, MRP1 and bcl-2 (Table 3 and data not shown). P-gp correlated with expression of MRP1 and bcl-2 (Table 3), and was less frequent in both certain FAB subtypes (M3, M4Eo) (P = 0.03) and good karyotype (P = 0.03) (data not shown). P-gp expression increased with age (P = 0.1) and white blood cell count (P = 0.1) (data not shown).

Drug resistance factors and outcome of induction chemotherapy

The complete remission (CR) rate of induction chemotherapy was 70% for the total study population (Table 2). Resistant disease and early death occurred in 12% and 13% of the patients, respectively, and 5% of the patients were classified as not evaluable for response because they received only one treatment cycle which did not result in CR. The CR rates were 83% for patients <60 years, 59% for patients 60 years, 76% for patients with good or intermediate karyotype, and 52% for those with poor karyotype (Table 2). White blood cell count and lactate dehydrogenase serum levels did not affect response to chemotherapy (data not shown).

Expression of LRP, P-gp and bcl-2 but not MRP1 were associated with poor response to induction chemotherapy (Table 2). Patients with negative LRP had a CR rate of 79% but those with positive LRP of only 53% (P = 0.003). For patients with low, intermediate and high P-gp expression, the CR rates were 77%, 68% and 38% (P = 0.02), respectively. For patients with low, intermediate and high MRP1 expression, the CR rates were 76%, 71% and 61% (P = NS). The CR rate was 80% for patients with negative bcl-2, and 62% for those with positive bcl-2 (P = 0.03). Similar results on the association between drug resistance factors and complete remission were seen when patients with FAB M3 were excluded from the analysis (data not shown).

Next, we performed logistic regression analyses on 101 patients for whom complete data were available. In the univariate analyses (Table 4), the odds ratios for complete remission were 0.3 for age 60 years (P = 0.005), 0.4 for poor karyotype (P = 0.03), 0.3 for positive LRP (P = 0.003), 0.6 for intermediate and 0.2 for high P-gp expression (P = 0.02), 0.8 for intermediate and 0.5 for high MRP1 expression (P = NS), and 0.4 for positive bcl-2 (P = 0.03).

The multivariate analyses included age, karyotype and drug resistance factors (Table 4). The odds ratios for complete remission were 0.4 for age 60 years (P = 0.05), 0.3 for poor karyotype (P = 0.05), 0.3 for positive LRP (P = 0.03), 1.1 for intermediate and 0.4 for high P-gp expression (P = NS), 0.8 for intermediate and 0.5 for high MRP1 expresssion (P = NS), and 0.6 for positive bcl-2 (P = NS).

Drug resistance factors and survival

The associations between drug resistance factors or clinical parameters and overall survival are shown in Figure 1. Survival was better for patients <60 years than for those with age 60, and for patients with good or intermediate karyotype than for those with poor karyotype (Figure 1), but was independent of white blood cell count and lactate dehydrogenase serum levels (data not shown). Expression of LRP and of P-gp were associated with shorter overall survival. Median overall survival was 1.4 years for LRP-negative patients but only 0.7 years for LRP-positive patients (P = 0.001). Median overall survival was 1.4, 0.8 and 0.4 years for patients with low, intermediate or high P-gp expression (P = 0.01), respectively. Within the group of patients with good prognosis (patients <60 years with good or intermediate karyotype), survival was shorter in the presence than in the absence of LRP expression (median overall survival 0.9 vs 2.2 years, P = 0.02) (Figure 2). In the patients receiving consolidation chemotherapy, overall survival was longer in patients receiving high-dose cytarabine consolidation than in those receiving normal-dose cytarabine consolidation (data not shown). In patients receiving consolidation chemotherapy with high-dose cytarabine (n = 46), overall survival was shorter in LRP-positive patients than in LRP-negative patients (P = 0.1) (data not shown). Positive bcl-2 showed a trend toward shorter survival (median overall survival 0.7 vs 1.7 years, P = 0.06), whereas MRP1 expression had no impact on overall survival.

In the univariate Cox regression analyses (Table 5), the relative risk for death were 2.0 for age 60 years (P = 0.008), 2.4 for poor karyotype (P = 0.006), 2.5 for positive LRP (P = 0.001), 1.8 for intermediate and 2.8 for high P-gp expression (P = 0.01), 1.9 for intermediate and 1.8 for high MRP1 expression (P = NS), and 1.7 for positive bcl-2 (P = 0.06). In the multivariate Cox regresssion analyses (Table 5), which included age, karyotype and the drug resistance factors, the relative risks for death were 2.1 for age 60 years (P = 0.01), 2.3 for poor karyotype (P = 0.02) and 2.3 for positive LRP (P = 0.01). P-gp, MRP1 and bcl-2 were not of independent prognostic significance (Table 5).

In the univariate Cox regression analyses of disease-free survival, P-gp (P = 0.02) and conventional-dose cytarabine (as compared to high-dose) consolidation therapy (P = 0.006) were associated with shorter survival of the patients, and positive LRP (P = 0.08) showed a trend towards shorter survival (Table 6). No associations between age, karyotype, MRP1 or bcl-2 and disease-free survival were observed (Table 6). In the multivariate analyses, P-gp (P = 0.02) and conventional-dose cytarabine (P = 0.01) retained their prognostic significance, and LRP (P = 0.08) showed a trend toward independent prognostic value, whereas the other factors had no independent prognostic value (Table 6).

Risk score

Next we used the three independent predictive and prognostic factors (age 60 years, poor karyotype, LRP expression) in order to define a risk score. Dependent on the number of independent prognostic factors present in a patient, patients were divided into four groups with risk scores from 0 to 3. The outcome was different between these groups (Table 7, Figure 3). Patients with a risk score of 0, 1, 2 and 3 had complete remission rates of 93%, 75%, 47% and 33% (P < 0.001), respectively (Table 7), and a median overall survival of 2.4, 1.2, 0.6 and 0.2 years (P = 0.0001), respectively (Table 7, Figure 3).

In the present study, age, karyotype and LRP were found to be independent predictive factors with regard to outcome of induction chemotherapy and independent prognostic factors with regard to overall survival in de novo AML. A risk score based on these three factors defined four groups of patients with different outcomes (Table 7, Figure 3). P-gp was of predictive and prognostic value in the univariate analyses but not in the multivariate analyses. Bcl-2 only predicted outcome of induction chemotherapy in the univariate analysis. MRP1 had no predictive or prognostic value.

The present univariate results on LRP confirm our initial results in 82 patients,¹² and are consistent with published studies.^11,20,26 List et al¹¹ reported a significant association between LRP overexpression and both an inferior response to induction chemotherapy and a trend toward shorter survival on, in comparison to our study, a more heterogeneous study population that included patients with de novo, secondary or relapsed AML. In the study by Hart et al²⁶ LRP expression predicted response of induction chemotherapy. Borg et al²⁰ demonstrated that LRP is an unfavorable predictive and prognostic factor in patients with de novo or secondary AML. However, our findings are in contrast to other studies including the one of the Southwest Oncology Group in younger AML patients.^10,27,28 These discrepancies might be due to differences in patient populations both with regard to age and number of patients, treatment protocols, detection methods and other factors.³

Our univariate findings on the predictive and prognostic significance of P-gp are consistent with previous reports.^4,5,6,7,8 The lack of an association between MRP1 expression and clinical outcome confirms our previous report on a smaller patient population, and is consistent with data from other groups.^9,10,20 In a recent study by Marie's group, outcome of induction chemotherapy correlated with MRP function but not with MRP expression.²⁸ Shorter survival in the presence of bcl-2 expression (Figure 1) has recently also been shown by others.¹⁴

The multivariate analyses revealed the independent predictive and prognostic values of age, karyotype and LRP. Whereas the predictive and prognostic values of age and karyotype are well established,^29,30 our results characterize LRP as an independent predictive and prognostic factor in de novo AML. Data from multivariate analyses on LRP are currently available also from two other groups.^11,20 List et al¹¹ found that LRP but not P-gp had independent predictive significance with regard to response to induction chemotherapy. Borg et al²⁰ performed a multivariate analysis that included age, white blood cell count, CD34, bcl-2, LRP, MRP, P-gp expression as well as rhodamine 123 efflux, FAB subtype, and cytogenetic subgroups. Age, LRP and P-gp function independently predicted for poor response to induction chemotherapy, whereas LRP, P-gp function and cytogenetics were independent prognostic factors with regard to overall survival and leukemia-free survival.²⁰

The finding that P-gp lost its prognostic value with regard to overall survival in the present multivariate analyses is consistent with data obtained by other groups.^11,30 However, we would like to stress that the lack of an independent prognostic significance of P-gp does not mean that P-gp expression is without clinical relevance in AML. It is unlikely that the lack of prognostic significance of P-gp in our study is due to the selection of the C219 antibody because we clearly demonstrated the predictive and prognostic value of P-gp in the univariate analyses. We have selected the C219 data in this study because, firstly, data from previous studies on smaller patient populations were available and, secondly, we previously did not observe significant differences between C219 and MRK16.³¹

Several other multivariate analyses of drug resistance factors in AML have been published.^{19,32,33,34,35} The parameters entered into these analyses varied but did not include LRP. We previously characterized age, MDR1 RNA expression and FAB subtype M4Eo as independent predictive factors, and age and MDR1 RNA expression as independent prognostic factors.³² These previous data are not in contrast to the present data for the following reasons: firstly, the previous multivariate analyses did not include LRP and, secondly, P-gp-positive AML was also associated with lower CR rates and shorter overall survival as well as disease-free survival as compared to P-gp-negative AML in the present study, but the differences were not statistically significant in the case of CR and overall survival. Nüssler et al³³ demonstrated the prognostic significance of age, P-gp expression and t(15;17) karyotype for patients treated with protocols containing daunorubucin and vincristine, but only of age and karyotype for patients treated with intermediate-dose cytarabine/amsacrine. In our population, however, a shorter overall survival for LRP-positive patients as compared to LRP-negative patients was seen also among patients receiving high-dose cytarabine consolidation chemotherapy, but the difference did not reach the level of statistical significance, probably due to the low number of patients. Van den Heuvel-Eibrink et al¹⁹ showed that age, karyotype, P-gp and CD34 were independent predictive and prognostic factors. Hunault et al³⁴ found that MDR1 overexpression predicted resistance to induction chemotherapy. In the Southwest Oncology Group study of elderly AML patients,³⁰ MDR1 expression independently predicted for complete remission, but did not indicate prognosis with regard to both overall and disease-free survival. In the study by Lohri et al³⁵ bcl-2 RNA expression and topoisomerase II RNA expression independently predicted overall survival and disease-free survival, whereas MDR1 RNA and MRP1 RNA had no impact.

Several explanations for the discrepancies between the various multivariate analyses exist. Firstly, the numbers and the types of factors entered into the analyses varied between the studies and probably affected the results. Secondly, patient populations differed with regard to age, disease category and sample size. In particular, the number of patients can affect the statistical power of a study. Thirdly, differences in detection levels, eg RNA vs protein vs function, and/or cut-off levels might have played a role. Finally, different treatment protocols might have affected the association between drug resistance factors and clinical outcome.³³

With regard to disease-free survival, P-gp and consolidation chemotherapy with high-dose cytarabine were of independent prognostic significance. LRP showed a trend toward shorter disease-free survival. The findings on the impact of consolidation therapy are consistent with recent studies in which high-dose cytarabine consolidation chemotherapy resulted in longer disease-free^36,37 and overall survival³⁷ as compared to conventional-dose cytarabine consolidation chemotherapy.

The clinical implications of our present findings are several-fold. Firstly, LRP and/or the proposed risk score might serve as prognostic factors and, in particular, might allow risk-adapted treatment in the future. For example, patients with LRP expression but otherwise good prognosis (age <60 years with good or intermediate karyotype) might be selected for more aggressive treatments, because they have worse outcome than corresponding patients without LRP expression. Secondly, the data support the multifactorial nature of drug resistance in AML. LRP was previously shown to be associated with resistance to doxorubicin, vincristine, carboplatin, cisplatin and melphalan.³⁸ In AML cells, LRP expression correlated with intracellular daunorubucin accumulation in one study²⁷ but not in another.³⁹ Thus blockade of a single mechanism, such as the one of MDR1/P-gp,^40,41,42 might be insufficient for improved clinical outcome.

We observed several correlations between the drug resistance factors with each other (Table 3). Associations between LRP and P-gp^11,12 or MRP1 expression,^20,26,28 between MDR1 RNA and MRP1 RNA,^26,34,35 and between P-gp and bcl-2¹⁷ have also been described before. The mechanisms of these co-expressions of drug resistance factors remain to be determined. Whereas LRP was not associated with karyotype or FAB subtype, P-gp was less frequently expressed in both patients with good karyotype and certain FAB subtypes (M3, M4Eo) which is consistent with data in the literature.^43,44

The optimal method for the determination of the expression of drug resistance factors in clinical samples remains a matter of debate. Immunohisto(cyto-)chemistry is a well established method in clinical oncology and also a reproducible technique for the assessment of drug resistance factors. Several studies demonstrated the reproducibility of immunocytochemistry with regard to P-gp^45,46 and LRP.⁴⁷ Immunocytochemical results also correlated with those obtained by flow cytometry with regard to P-gp,^18,20 LRP⁴⁷ and bcl-2.¹⁴ In order to resolve the issue of the optimal detection method, future comparative studies on large patient populations are required.

In summary, our data indicate that LRP is a drug resistance factor with independent predictive and prognostic significance in de novo AML. LRP or the proposed risk score might be useful for the selection of patients for risk-adapted treatment in the future. Our results also support the multifactorial nature of drug resistance which will have to be considered for strategies to reverse drug resistance in this disease.

Acknowledgements

This study was supported by the 'Fonds zur Förderung der wissenschaftlichen Forschung' (project number P12264-MED).

²  Malayeri R, Filipits M, Suchomel RW, Zöchbauer S, Lechner K, Pirker R Multidrug resistance in leukemias and its reversal Leuk Lymphoma 1996 23: 451–458 MEDLINE

³  Leith C Multidrug resistance in leukemia Curr Opin Hematol 1998 5: 287–291 MEDLINE

⁴  Pirker R, Wallner J, Geissler K, Linkesch W, Haas OA, Bettelheim P, Hopfner M, Scherrer R, Valent P, Havelec L, Ludwig H, Lechner K MDR1 gene expression and treatment outcome in acute myeloid leukemia J Natl Cancer Inst 1991 83: 708–712 MEDLINE

⁵  Marie J-P, Zittoun R, Sikic BI Multidrug resistance (mdr1) gene expression in adult acute leukemias: correlations with treatment outcome and in vitro drug sensitivity Blood 1991 78: 586–592 MEDLINE

⁶  Campos L, Guyotat D, Archimbaud E, Calmard-Oriol P, Tsuruo T, Troncy J, Treille D, Fiere D Clinical significance of multidrug resistance P-glycoprotein expression on acute nonlymphoblastic leukemia cells at diagnosis Blood 1992 79: 473–476 MEDLINE

⁷  Zöchbauer S, Gsur A, Brunner R, Kyrle PA, Lechner K, Pirker R P-glycoprotein expression as unfavorable prognostic factor in acute myeloid leukemia Leukemia 1994 8: 974–977 MEDLINE

⁸  Wood P, Burgess R, MacGregor A, Yin JAL P-glycoprotein expression on acute myeloid leukaemia blast cells at diagnosis predicts response to chemotherapy and survival Br J Haematol 1994 87: 509–514 MEDLINE

⁹  Filipits M, Suchomel RW, Zöchbauer S, Brunner R, Lechner K, Pirker R Multidrug resistance-associated protein (MRP) in acute myeloid leukemia: no impact on treatment outcome Clin Cancer Res 1997 3: 1419–1425 MEDLINE

¹⁰  Leith CP, Kopecky KJ, Chen I-M, Slovak ML, Head DR, Weick J, Appelbaum FR, Willman CL Frequency and clinical significance of expression of the multidrug resistance proteins, MDR1, MRP1 and LRP in acute myeloid leukemia patients less than 65 yrs old. A Southwest Oncology Group study Blood 1997 90: 389a (Abstr.)

¹¹  List AF, Spier CS, Grogan TM, Johnson C, Roe DJ, Greer JP, Wolff SN, Broxterman HJ, Scheffer GL, Scheper RJ, Dalton WS Overexpression of the major vault transporter protein lung-resistance protein predicts treatment outcome in acute myeloid leukemia Blood 1996 87: 2464–2469 MEDLINE

¹²  Filipits M, Pohl G, Stranzl T, Suchomel RW, Scheper RJ, Jäger U, Geissler K, Lechner K, Pirker S Expression of the lung resistance protein predicts poor outcome in de novo acute myeloid leukemia Blood 1998 91: 1508–1513 MEDLINE

¹³  Kroemer G The proto-oncogene Bcl-2 and its role in regulating apoptosis Nature Med 1997 3: 614–620 MEDLINE

¹⁴  Campos L, Rouault J-P, Sabido O, Oriol P, Roubi N, Vasselon C, Archimbaud E, Magaud J-P, Guyotat D High expression of bcl-2 protein in acute myeloid leukemia cells is associated with poor response to chemotherapy Blood 1993 81: 3091–3096 MEDLINE

¹⁵  List AF, Spier CM, Cline A, Doll DC, Garewal H, Morgan R, Sandberg AA Expression of the multidrug resistance gene product (P-glycoprotein) in myelodysplasia is associated with a stem cell phenotype Br J Haematol 1991 78: 28–34 MEDLINE

¹⁶  Zhou DC, Marie JP, Suberville AM, Zittoun R Relevance of mdr1 gene expression in acute myeloid leukemia and comparison of different diagnostic methods Leukemia 1992 6: 879–885 MEDLINE

¹⁷  Campos L, Oriol P, Sabido O, Guyotat D Simultaneous expression of P-glycoprotein and BCL-2 in acute myeloid leukemia blast cells Leuk Lymphoma 1997 27: 119–125 MEDLINE

¹⁸  Del Poeta G, Venditti A, Aronica G, Stasi R, Cox MC, Buccisano F, Bruno A, Tamburini A, Suppo G, Simone MD, Epiceno AM, Del Moro B, Masi M, Papa G, Amadori S P-glycoprotein expression in de novo acute myeloid leukemia Leuk Lymphoma 1997 27: 257–274 MEDLINE

¹⁹  van den Heuvel-Eibrink MM, van der Holt B, te Boekhorst PAW, Pieters R, Schoester M, Löwenberg B, Sonneveld P MDR1 expression is an independent prognostic factor for response and survival in de novo acute myeloid leukaemia Br J Haematol 1997 99: 76–83 MEDLINE

²⁰  Borg AG, Burgess R, Green LM, Scheper RJ, Liu Yin JA Overexpression of lung-resistance protein and increased P-glycoprotein function in acute myeloid leukaemia cells predict a poor response to chemotherapy and reduced patient survival Br J Haematol 1998 103: 1083–1091 MEDLINE

²¹  Nooter K, Westerman AM, Flens MJ, Zaman GJR, Scheper RJ, van Wingerden KE, Burger H, Oostrum R, Boersma T, Sonneveld P, Gratama JW, Kok T, Eggermont AMM, Bosman FT, Stoter G Expression of the multidrug resistance-associated protein (MRP) gene in human cancers Clin Cancer Res 1995 1: 1301–1310 MEDLINE

²²  Scheffer GL, Wijngaard PLJ, Flens MJ, Izquierdo MA, Slovak ML, Pinedo HM, Meijer CJLM, Clevers HC, Scheper RJ The drug resistance-related protein LRP is the human major vault protein Nature Med 1995 1: 578–582 MEDLINE

²³  Grant CE, Valdimarsson G, Hipfner DR, Almquist KC, Cole SPC, Deeley RG Overexpression of multidrug resistance-associated protein (MRP) increases resistance to natural product drugs Cancer Res 1994 54: 357–361 MEDLINE

²⁴  Kaplan EL, Meier P Nonparametric estimation from incomplete observations J Am Stat Assoc 1958 53: 457–481

²⁵  Cox DR Regression models and life tables J R Stat Soc 1972 34: 187–220

²⁶  Hart SM, Ganeshaguru K, Scheper RJ, Prentice HG, Hoffbrand AV, Mehta AB Expression of the human major vault protein LRP in acute myeloid leukemia Exp Hematol 1997 25: 1227–1232 MEDLINE

²⁷  Michieli M, Damiani D, Ermacora A, Masolini P, Raspadori D, Visani G, Scheper RJ, Baccarani M P-glycoprotein, lung-resistance-related protein and multidrug resistance associated protein in de novo acute non-lymphocytic leukaemias: biological and clinical implications Br J Haematol 1999 104: 328–335 MEDLINE

²⁸  Legrand O, Simonin G, Perrot J-Y, Zittoun R, Marie J-P Pgp and MRP activities using calcein-AM are prognostic factors in adult acute myeloid leukemia patients Blood 1998 91: 4480–4488 MEDLINE

²⁹  Bloomfield CD, Shuma C, Regal L, Philip PP, Hossfeld DK, Hagemeijer AM, Garson OM, Peterson BA, Sakurai M, Alimena G, Berger R, Rowley JD, Ruutu T, Mitelman F, Dewald GW, Swansbury J Long-term survival of patients with acute myeloid leukemia. A third follow-up of the Fourth International Workshop on Chromosomes in Leukemia Cancer 1997 80: 2191–2198 MEDLINE

³⁰  Leith CP, Kopecky KJ, Godwin J, McConnell T, Slovak ML, Chen I-M, Head DR, Appelbaum FR, Willman CL Acute myeloid leukemia in the elderly: assessment of multidrug resistance (MDR1) and cytogenetics distinguishes biologic subgroups with remarkably distinct responses to standard chemotherapy. A Southwest Oncology Group study Blood 1997 89: 3323–3329 MEDLINE

³¹  Filipits M, Suchomel RW, Lechner K, Pirker R Immunocytochemical detection of the multidrug resistance-associated protein and P-glycoprotein in acute myeloid leukemia: impact of antibodies, sample source and disease status Leukemia 1997 11: 1073–1077 MEDLINE

³²  Pirker R, Wallner J, Götzl M, Gsur A, Geissler K, Havelec L, Knapp W, Haas O, Linkesch W, Lechner K MDR1 RNA expression is an independent prognostic factor in acute myeloid leukemia Blood 1992 80: 557–558 MEDLINE

³³  Nüssler V, Pelka-Fleischer R, Zwierzina H, Nerl C, Becker B, Gieseler F, Diem H, Ledderose G, Gullis E, Sauer H, Wilmanns W P-glycoprotein expression in patients with acute leukemia – clinical relevance Leukemia 1996 10 (Suppl. 3): S23–S31 MEDLINE

³⁴  Hunault M, Zhou D, Delmer A, Ramond S, Viguié F, Cadiou M, Perrot J-Y, Levy V, Rio B, Cymbalista F, Zittoun R, Marie J-P Multidrug resistance gene expression in acute myeloid leukemia: major prognosis significance for in vivo drug resistance to induction treatment Ann Hematol 1997 74: 65–71 MEDLINE

³⁵  Lohri A, van Hille B, Bacchi M, Fopp M, Joncourt F, Reuter J, Cerny T, Fey MF, Herrmann R Five putative drug resistance parameters (MDR1/P-glycoprotein, MDR-associated protein, glutathione-S-transferase, bcl-2 and topoisomerase II) in 57 newly diagnosed acute myeloid leukaemias Eur J Haematol 1997 59: 206–215 MEDLINE

³⁶  Mayer RJ, Davis RB, Schiffer CA, Berg DT, Powell BL, Schulman P, Omura GA, Moore JO, McIntyre OR, Frei III E for the Cancer and Leukemia Group B. Intensive postremission chemotherapy in adults with acute myeloid leukemia New Engl J Med 1994 331: 896–903 MEDLINE

³⁷  Fopp M, Fey MF, Bacchi M, Cavalli F, Gmuer J, Jacky E, Schmid L, Tichelli A, Tobler A, Tschopp L, von Fliedner V, Gratwohl A for the Leukaemia Project Group of the Swiss Group for Clinical Cancer Research (SAKK) Post-remission therapy of adult acute myeloid leukaemia: one cycle of high-dose versus standard-dose cytarabine Ann Oncol 1997 8: 251–257 MEDLINE

³⁸  Izquierdo MA, Shoemaker RH, Flens MJ, Scheffer GL, Wu L, Prather TR, Scheper RJ Overlapping phenotypes of multidrug resistance among panels of human cancer-cell lines Int J Cancer 1996 65: 230–237 MEDLINE

³⁹  Broxterman HJ, Sonneveld P, Pieters R, Lankelma J, Eekman CA, Loonen AH, Schoester M, Ossenkoppele GJ, Löwenberg B, Pinedo HM, Schuurhuis GJ Do P-glycoprotein and major vault protein (MVP/LRP) expression correlate with in vitro daunorubucin resistance in acute myeloid leukemia? Leukemia 1999 13: 258–265 MEDLINE

⁴⁰  Pirker R, Keilhauer G, Raschack M, Lechner C, Ludwig H Reversal of multi-drug resistance in human KB cell lines by structural analogs of verapamil Int J Cancer 1990 45: 916–919 MEDLINE

⁴¹  Lum BL, Fisher GA, Brophy NA, Yahanda AM, Adler KM, Kaubisch S, Halsey J, Sikic BI Clinical trials of modulation of multidrug resistance Cancer 1993 72: 3502–3514 MEDLINE

⁴²  Advani R, Saba HI, Tallman MS, Rowe JM, Wiernik PH, Ramek J, Dugan K, Lum B, Villena J, Davis E, Paietta E, Litchman M, Sikic BI, Greenberg PL Treatment of refractory and relapsed acute myelogenous leukemia with combination chemotherapy plus the multidrug resistance modulator PSC 833 (Valspodar) Blood 1999 93: 787–795 MEDLINE

⁴³  Zöchbauer S, Haas OA, Schwarzinger I, Lechner K, Pirker R Multidrug resistance in acute myeloid leukaemia with inversion in chromosome 16 or FAB M4Eo subtype Lancet 1994 344: 894 MEDLINE

⁴⁴  Drach D, Zhao S, Drach J, Andreeff M Low incidence of MDR1 expression in acute promyelocytic leukaemia Br J Haematol 1995 90: 369–374 MEDLINE

⁴⁵  Gala JL, McLachlan JM, Bell DR, Michaux JL, Ma DDF Specificity and sensitivity of immunocytochemistry for detecting P-glycoprotein in haematological malignancies J Clin Pathol 1994 47: 619–624 MEDLINE

⁴⁶  Beck WT, Grogan TM, Willman CL, Cordon-Cardo C, Parham DM, Kuttesch JF, Andreeff M, Bates SE, Berard CW, Boyett JM, Brophy NA, Broxterman HJ, Chan HSL, Dalton WS, Dietel M, Fojo AT, Gascoyne RD, Head D, Houghton PJ, Srivastava DK, Lehnert M, Leith CP, Paietta E, Pavelic ZP, Rimsza L, Roninson IB, Sikic BI, Twentyman PR, Warnke R, Weinstein R Methods to detect P-glycoprotein-associated multidrug resistance in patients' tumors: consensus recommendations Cancer Res 1996 56: 3010–3020 MEDLINE

⁴⁷  Legrand O, Simonin G, Zittoun R, Marie J-P Lung resistance protein (LRP) gene expression in adult acute myeloid leukemia: a critical evaluation by three techniques Leukemia 1998 12: 1367–1374 MEDLINE

Figure 2  LRP and overall survival in good risk patients. Kaplan–Meier survival estimates stratified by LRP expression are shown for patients of age <60 years with good or intermediate karyotype.

Figure 3  Risk score and overall survival. Dependent on the number of independent prognostic factors (age 60 years, poor karyotype, LRP expression) a risk score ranging from 0 to 3 was developed. Overall survival curves stratified according to the risk score are shown.

Table 2 Age, karyotype and drug resistance factors: impact on complete remission of induction chemotherapy

Table 3 Correlations between drug resistance factors

Table 4 Logistic regression analysis of complete remission

Table 5 Cox regression analysis of overall survival

Table 6 Cox regression analysis of disease-free survival

Table 7 Risk score

^© Macmillan Publishers Ltd 2000