British Journal of Haematology, 2002, 116, 519–527
P-glycoprotein, lung resistance-related protein and multidrug resistance-associated protein in de novo adult acute lymphoblastic leukaemia Daniela Damiani, 1 Angela Michelutti, 1 Mariagrazia Michieli, 2 Paola Masolini, 1 Raffaella Stocchi, 1 Antonella Geromin, 1 Anna Ermacora, 1 Domenico Russo, 1 Renato Fanin 1 and Michele Baccarani 3 1Division of Haematology, Department of Medical and Morphological Research, University Hospital, Udine, 2Medical Oncology B, Centro di Riferimento Oncologico, National Cancer Institute, IRCCS, Aviano, and 3Institute of Haematology and Medical Oncology ÔL and A. Sera`gnoliÕ, University of Bologna, Bologna, Italy Received 31 July 2001; accepted for publication 3 October 2001
Summary. P-glycoprotein (P-gp), lung resistance-related protein (LRP) and multidrug resistance-associated protein (MRP) expression, and blast cell intracellular daunorubicin accumulation (IDA) were evaluated in 95 previously untreated cases of adult acute lymphoblastic leukaemia (ALL) using flow cytometry. Forty-five out of 95 (47%) patients were P-gp positive (+), 12/66 (18%) were LRP+ and 11/66 (17%) were MRP+. Eighteen out of 66 (28%) patients showed a simultaneous multidrug resistance (MDR)-related protein expression higher than controls for more than one protein, while 24/66 (36%) cases did not overexpress any protein. Twenty-one out of 24 (87%) cases overexpressing at least one MDR-related protein had a defect in accumulating daunorubicin into their blast cells, while only 4/24 (16%) cases who did not overexpress any protein had similar features. The complete remission rates were similar in MDR-
Intensive treatment with anticancer agents and welldesigned chemotherapeutic strategies significantly improve outcome in childhood acute lymphoblastic leukaemia (ALL), and long-term disease free survival is obtained in 70% of cases. In contrast, adults with ALL have lower complete remission (CR) rates and substantially poorer rates of longterm survival despite the application of therapeutic strategies that are successfully used in children (Copelan & McGuire, 1995). To explain the differing prognoses, and the poorer outcome in older patients, many factors must be taken into consideration: differences in disease biology, including a
Correspondence: Dr Daniela Damiani, Division of Haematology, University Hospital, P.le S. Maria della Misericordia, 33100 Udine, Italy. E-mail:
[email protected] Ó 2002 Blackwell Science Ltd
positive and -negative (–) patients but relapses within 6 months were more frequent in P-gp+ cases, and therefore the disease-free survival duration was shorter in P-gp+ than in P-gp– patients (P ¼ 0Æ01). The number of MRP+ and/or LRP+ cases was too small to be able to draw any conclusion on their role in affecting or predicting therapy outcome. In conclusion, P-gp overexpression associated with a defect in daunorubicin accumulation is a frequent feature in adult ALL at onset and seems to be related to poorer therapy outcome and, consequently, a shorter disease-free survival. LRP and MRP overexpression seems to be a rare event and no conclusion can be drawn on its prognostic role. Keywords: P-glycoprotein, multidrug resistance-associated protein, lung resistance-related protein, adult acute lymphoblastic leukaemia.
higher incidence of chromosome translocations associated with poor prognosis, in particular the Philadelphia chromosome (Copelan & McGuire, 1995), differences in drug metabolism, especially regarding methotrexate (Goker et al, 1993; Copelan & McGuire, 1995), a greater haematological and extrahaematological toxicity in older patients (Hoelzer, 1993). More recently, another potential factor causing poor adult response to chemotherapy has been related to the presence of the multidrug resistance (MDR) phenomenon (Endicott & Ling, 1989; Goasguen et al, 1993; List, 1993). Classic MDR is due to the expression of P-glycoprotein (P-gp), a protein encoded by the mdr-1 gene, which belongs to the ATP-binding cassette of the transporter gene superfamily. This gene maps on the long arm of chromosome 7 and its product, P-gp, acts as an energy–dependent transmembrane pump that actively effluxes drugs and dyes
519
520
D. Damiani et al
(Tsuruo, 1988; Chin et al, 1989; Weinstein et al, 1990). P-gp is physiologically expressed at different levels in many normal tissues including the biliary lining of the liver, renal cells, colon, brain and dermal endothelial cells (Thiebaut et al, 1987; Sugawara et al, 1988; Pileri et al, 1991; Chaudhary et al, 1992; Drach et al, 1992). This disposition suggests that the physiological role of P-gp is to contribute to the detoxification and excretion mechanisms. Transfection experiments showed that MDR overexpression can cause the MDR phenomenon, which involves a broad spectrum of drugs and natural products (Tsuruo, 1988). Substrates of P-gp are anthracyclines, vinca alkaloids, epipodophyllotoxins and taxanes. The MDR phenomenon has been observed in vitro, but P-gp is overexpressed at different levels in acute and chronic leukaemias, in lymphomas and multiple myelomas (Riccardi et al, 1991; Solary et al, 1991; Campos et al, 1992; Michieli et al, 1992; Goasguen et al, 1993; Michelutti et al, 1994; Michieli et al, 1997, 2000). Many authors have reported that acute leukaemia is characterized by a positive correlation between P-gp expression and lower remission rates and a higher frequency of relapse (Marie et al, 1991; Pirker et al, 1991; Campos et al, 1992; Michieli et al, 1992; Goasguen et al, 1993; Ino et al, 1994; Wood et al, 1994; Zo¨chbauer et al, 1994). Clinical resistance to chemotherapies is not always accomplished by a P-gp overexpression and can be due to the presence of other atypical MDR mechanisms, for example, the overexpression of the lung resistance-related protein (LRP) or the multidrug-associated protein (MRP). MRP also belongs to the ATP-binding cassette of drug transporter proteins; in humans it is encoded by the mrp-1 gene that maps on the short arm of chromosome 16 and P-gp is expressed at different levels in many normal tissues (Cole et al, 1992). Transfection experiments showed that MRP overexpression can cause the development of MDR which involves more or less the same classes of cytotoxic drugs affected by P-gp (Grant et al, 1994). In acute leukaemia, MRP is expressed with lower frequency than P-gp and its role in patients’ clinical outcome is still a question of debate; this can be due to the low number of cases studied and the frequent co-expression of the other proteins involved in MDR, which can contribute to clinical response. Studies by Schneider et al (1995) and Hart et al (1993) seem to underline an increase in MRP expression at relapse in ALL and acute non-lymphoblastic leukaemia (ANLL). Lung resistance-related protein (LRP) is the human major vault protein. It is coded by the lrp gene, which maps on the short arm of chromosome 16 (Scheffer et al, 1995). High LRP expression has been found in several normal tissues, especially in those chronically exposed to xenobiotics or toxic agents, such as bronchial epithelium or in tissues with secretory or excretory functions (Scheper et al, 1993). Relationships between LRP overexpression and development of MDR are still unclear. Transfection of lrp cDNA in tumour cell lines does not display an MDR phenotype (Scheper et al, 1993). Studies on cellular pharmacokinetics of anthracyclines [doxorubicin and methotrexate (MTX)] in cell lines with LRP overexpression seem to confer to LRP a role in cellular transport mechanisms
(Kuiper et al, 1990). LRP seems to be overexpressed in acute secondary leukaemia and in de novo leukaemia, frequently associated with P-gp (List et al, 1993; Michieli et al, 1997, 1999; Damiani et al, 1998). Only a few data on the MDR phenomenon in ALL have been reported in an attempt to demonstrate its correlation with treatment response or patient follow-up (Goasguen et al, 1993; Beck et al, 1995; Wattel et al, 1995; Goasguen et al, 1996; den Boer et al, 1998; Dhooge et al, 1999). Studies on adult ALL series are scarce, in fact most of these studies combine patients at diagnosis and at relapse or children with adults and often examine only one of the mechanisms involved in MDR, so the role of multidrug resistance in adult and childhood ALL is still unclear. In this study, we examined the expression and function of three proteins involved in the MDR phenomenon, P-glycoprotein, multidrug resistance-associated protein and lung resistance -related protein, in 95 cases of adult de novo ALL. PATIENTS AND METHODS Patients. We studied 95 consecutive patients with previously untreated ALL who were admitted to the Division of Haematology, University of Udine, between January 1991 and February 2001. The diagnosis was based on French– American–British (FAB) guidelines (Bennet et al, 1985) and was confirmed by the immunophenotype of blast cells which was determined by using a panel of monoclonal antibodies (mAbs): anti CD13, 33, 34, 10, 19, 22, DR, SmIg, 1a, 7, 3, 4, 8. (BD). CD34 expression was evaluated in 92 patients. Cytogenetic analysis was performed in 73 cases: 26/73 (36%) were normal diploid and 47/73 (64%) were abnormal. Recent studies have shown a correlation between chromosomal abnormalities and prognosis, describing three types of abnormalities: favourable, unfavourable and unrelated to differences in treatment outcome (undefined) (Secker-Walker et al, 1997; Ferrando & Look, 2000; Charrin, 1996). We detected 33 unfavourable abnormalities: 25 cases of t(9;22), one of t(4;11), one of t(1;19), one of t(8;14), one case of deletion involving chromosome 9p, four cases of hypodiploidy (with less than 46 chromosomes); also detected were 10 favourable abnormalities: one of t(10;14), one case of deletion involving chromosome 12p, five cases of low hyperdiploidy (47– 49 chromosomes), three cases of high hyperdiploidy (50–60 chromosomes), and four undefined abnormalities: one case of t(2;14), one case of 5q–, one case of t(10;12), one case of t(1;4). Patients’ characteristics are summarized in Table I. All the patients were treated according to the Institution’s ongoing first-line induction and consolidation regimens. The first regimen included an induction course of vincristine (1Æ5 mg/m2 weekly for the first 4 weeks), prednisone (50 mg/m2) and daunorubicin (40 mg/m2/d, d 1–3) (38 patients) or idarubicin (8 mg/m2/d, d 1, 3, 5) (57 patients), and a post-induction course of high dose cytarabine (Ara-C, 2000 mg/m2 i.v. twice daily, d 1–3) and idarubicin (12 mg/m2/d i.v., d 4–6). Central nervous system (CNS) prophylaxis with intrathecal (IT) Ara-C and MTX was performed weekly in all the patients.
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
MDR-related Protein Expression in De Novo Adult ALL Table I. Biological and clinical characteristics of the 95 patients with de novo acute lymphoblastic leukaemia who entered the study.
Patient number Age Mean ± SD Median (range) Sex Female Male Immunophenotype B-cell lineage Pre-B B T-cell lineage Pre-T T FAB classification L1 L1/L2 L2 L3 Karyotype (73 patients) Normal diploid Abnormal Favourable Unfavourable Undefined CD34 expression (92 patients) Positive Negative WBC count (109/l) Mean ± SD Median (range) LDH > 460 UI/l (88 patients)
95 39Æ5 ± 16Æ61 37Æ5 (13–80) 47/95 (49%) 48/95 (51%) 75/95 62/95(65%) 13/95(14%) 20/95 15/95(16%) 5/95(5%) 9/95 (10%) 46/95 (48%) 30/95 (32%) 10/95 (10%) 26/73 (36%) 47/73 (64%) 10/73 (14%) 33/73 (45%) 4/73 (5%) 62/92 (67%) 30/92 (33%) 60Æ7 ± 130 13Æ5 (0Æ8–850) 73/88 (83%)
Patients younger than 55 years of age underwent a post-remission therapy with three courses of vincristine (1Æ5 mg/m2, d 1), MTX (1000 mg/m2 c.i. d 1), Ara-C (100 mg push plus 400 mg/m2 c.i. d 1) and desametasone 10 mg/m2/d, d 1–5), and two courses of teniposide (VM-26, 165 mg/m2/d on d 1, 5, 9, 13) and Ara-C (300 mg/m2/d on d 2, 5, 9, 13), and a ÔmaintenanceÕ therapy with MTX and 6-mercaptopurine in combination with pulses of vincristine and prednisone. Patients older than 55 years received reduced drug doses. Seventeen out of 95 (18%) patients underwent an allogeneic bone marrow transplant (BMT) and 23/95 (24%) underwent autologous bone marrow transplantation (ABMT) within 6 months of the diagnosis. Response to treatment was evaluated after the induction course and classified as follows: death during induction (DDI, death during or after the first course of therapy before haematological reconstitution), primary resistance (PR, cellular marrow with > 5% of blast cells or evidence of leukaemia in other sites, after at least two courses of chemotherapy), complete remission (CR, cellular marrow with < 5% of blast cells, a neutrophil count higher than 1Æ5 · 109/l, platelet count higher than 100 · 109/l and no
521
evidence of leukaemia in other sites), Early relapse (ER, relapse within 6 months from CR). The sum of DDI, PR and ER accounted for the total early failure of treatment. Leukaemic blast cells. Peripheral blood and marrow samples anticoagulated with heparin were collected during diagnostic procedures after receiving the patients’ informed consent. Mononuclear cells were separated by sedimentation on Ficoll, washed twice in phosphate-buffered saline (PBS) and checked for viability using the trypan blue exclusion test, and suspended in medium or PBS as required. In all samples over 80% of cells were blasts. Moreover, to avoid possible contamination with residual normal leucocytes, analysis for MDR phenotyping and functional studies was performed after gating blast cells by using scatter parameters. P-gp, LRP and MRP expression. P-gp, LRP and MRP expression was evaluated by flow cytometry using the MRK16 (anti-P-gp), LRP-56 (anti-LRP) and MRPm6 (anti-MRP) mAbs (Kamiya, Seattle, USA) as previously described (Michieli et al, 1997). To study P-gp, 1 · 106 blast cells were fixed in PLP (periodate lysine paraformaldehyde) for 15 min, washed twice in PBS, and incubated for 30 min in 50 ll of a PBS-saponine solution (0Æ02%) containing MRK16 (2 lg/ml). After two washes in PBS, cells were incubated with 2Æ5 ll of fluorescein isothiocyanate (FITC) goat antimouse (Dako, Denmark). To study LRP and MRP, 1 · 106 blast cells were fixed and permeabilized in the Becton Dickinson (BD) lysing solution and, after two washes, incubated with 2Æ0 ll of MRPm6 or 10 ll of LRP-56 for 1 h at 4°C in a PBS-saponine solution, as recommended by the manufacturer. Staining was revealed using 2Æ5 ll of an FITC goat anti-mouse (Dako). Control samples were carried out simultaneously by replacing the primary antibodies with their respective isotypic control (Dako). Data acquisition and analysis were made using a FACScan and the Lysis II software programme (BD). Results were expressed using the mean fluorescence index (MFI ¼ the ratio between the mean fluorescence intensity of cells incubated with MRK-16, LRP56 or MRPm6 and the mean fluorescence intensity of respective controls) rather than by the percentage of stained cells. Leukaemic cells were defined as positive (overexpressing) when their MFI exceeded the highest MFI of either the negative cell lines, normal marrow or blood cells, so a case had P-gp overexpression (P-gp+) when the MFI was higher than 6 for MRK-16, an LRP overexpression (LRP+) when the MFI was higher than 5 for LRP-56, and an MRP overexpression (MRP+) when the MFI was higher than 3 for MRPm6. All the other cases were defined as negative (–) or non-overexpressing for one or more proteins (Michieli et al, 1997). Intracellular daunorubicin accumulation. The amount of daunorubicin accumulated by blast cells was evaluated by flow cytometry as previously described (Damiani et al, 1995). Briefly, mononuclear cells separated on Ficoll-Hypaque were suspended at 2Æ5 · 106 cells/ml in Roswell Park Memorial Institute 1640 medium (RPMI-1640; Biochrom KG, Seromed, Berlin, Germany), and incubated at 37°C with 5% CO2 for 2 h, with 1000 ng/ml of DNR (Pharmacia,
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
522
D. Damiani et al
Milan, Italy). After two washes in cold PBS, cells were kept on ice and immediately analysed for their FL2 and forward scatter (FSC), using a FACScan equipped with an argon laser tuned at 488 nm and Lysis II software (BD). Results were expressed using the normalized mean fluorescence index (NMFI), calculated according to Luk & Tannock, 1989). The normal peripheral and bone marrow cells of healthy donors had a mean NMFI of 345 ± 61, so samples were considered to have a reduced uptake of daunorubicin when their NMFI was < 280 (Michieli et al, 1997). Statistical analysis. Yates correct chi-square test, Fischer’s exact test and the Mann–Whitney U-test were used to compare the differences between two groups. The univariate logistic regression test was used to identify the variables affecting therapy outcome. All the parameters of significant value in univariate analysis (P < 0Æ10) were entered in the multivariate analysis. Then, a backward procedure was adopted to remove the least significant factors until only variables with significant value were retained. The linear regression test was used to evaluate the relationship between P-gp expression and clinical or laboratory parameters. Survival was calculated according to Kaplan & Meier (1958). The log rank test for heterogeneity (Peto et al, 1977) and Cox’s proportional hazard model for covariate analysis of censored data on survival (Cox, 1970) were applied. RESULTS P-gp, LRP, MRP expression We classified as P-gp, LRP or MRP overexpressing (positive, +) those cases in which the MFI for anti-P-gp (MRK-16), antiLRP (LRP-56) or anti-MRP (MRPm6) monoclonal antibodies was higher than negative controls. All other cases were defined negative or non-overexpressing (–). All patients were studied for P-gp expression and the number of cases overexpressing MRK-16 was 45/95 (47%) (Table II). Sixtysix out of the 95 patients enrolled in this study were studied for MRP and LRP; 11/66 (17%) cases overexpressed MRP and 12/66 (18%) overexpressed LRP. In 18 cases (28%) there was a co-expression of at least two proteins: in 7/66 cases (11%) the P-gp overexpression was associated with MRP overexpression, in 8/66 (12%) cases with LRP overexpression, in 1/66 (2%) cases MRP and LRP were overexpressed simultaneously, and 2/66 (3%) cases overexpressed the three proteins simultaneously. Twenty-four out of 66 (36%) patients did not overexpress any protein (P-gp–/MRP–/LRP–). P-gp overexpression was associated with an advanced age (P ¼ 0Æ037) and a higher white blood cell count (P ¼ 0Æ018). On LRP expression, a significant correlation was found with the immunophenotype: in fact, this protein was overexpressed in 6/14 (43%) T-cell lineage ALL and in only 6/52 (12%) B-cell lineage ALL (P ¼ 0Æ02). No relationship was found between P-gp, LRP or MRP expression and the other clinical and biological characteristics. The clinical and biological characteristics of P-gp-positive and -negative cases are summarized in Table II. The clinical and biological characteristics of cases with the overexpres-
Table II. Clinical and biological characteristics of 95 de novo acute lymphoblastic leukaemia patients according to P-glycoprotein (P-gp) expression.
P
MDR phenotype
P-gp positive
P-gp negative
Patient number Age Mean ± SD Median (range) Sex (female/male) Immunophenotype B-cell lineage T-cell lineage WBC count (109/l) Mean ± SD Median (range) CD34 expression Positive Negative LDH > 460 UI/l IDA < 280 NMFI > 280 NMFI Karyotype Favourable Unfavourable Undefined
45/95 (47%)
50/95 (53%)
41Æ4 ± 15Æ9 39Æ5 (13–80) 21/24
37 ± 17Æ2 35 (14–80) 26/24
0Æ037
32/45 (71%) 13/45 (29%)
43/50 (86%) 7/50 (14%)
0Æ35
77Æ7 ± 161Æ9 32Æ3 (0Æ8–850)
45Æ1 ± 90Æ6 11Æ7 (1Æ2–496)
0Æ018
26/44 (59%) 18/44 (41%) 34/42 (81%)
36/48 (75%) 12/48 (25%) 39/46 (85%)
0Æ46
44/45 (98%) 1/45 (2%)
5/50 (10%) 45/50 (90%)
0Æ0001
16/32 (50%) 15/32 (47%) 1/32 (3%)
20/41 (49%) 18/41 (44%) 3/41 (7%)
0Æ8
0Æ32
We classified as P-gp positive those cases with an MFI (mean fluorescence index) for MRK-16 higher than 6 (see Patients and methods).
sion of only one protein, of cases co-overexpressing at least two proteins and of cases lacking any MDR-related protein overexpression are summarized in Table III. Intracellular daunorubicin accumulation The ability of blast cells to accumulate ex vivo daunorubicin was chosen as a marker of MDR-related protein activity. Forty-nine out of 95 (51%) patients had a defect in accumulating ex vivo daunorubicin as shown using an intracellular daunorubicin accumulation (IDA, i.e. the DNR NMFI) lower than 280 (see Patients and methods). Forty-four out of 45 (98%) of the P-gp-positive cases as well as 17/18 (94%) cases that simultaneously overexpressed at least two MDR proteins had an impaired ability in accumulating DNR (median DNR NMFI 180, range 78–303 in P-gp+, and median 187, range 78–270 in co-overexpressing cases). In contrast, only 4/24 (16%) cases with no overexpression of any MDR protein (P-gp–/MRP–/LRP–) had similar features (median DNR NMFI 311; range 230–543). Cases with one or more MDR-related protein overexpression had a statistically significant reduction of IDA compared with cases without any protein overexpression (P ¼ 0Æ0001). Results are reported in Tables II and III. Cytogenetics Cytogenetic analysis was obtained in 73/95 cases. According to recent literature (Charrin et al, 1996; Secker-Walker
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
MDR-related Protein Expression in De Novo Adult ALL
523
Table III. Clinical and biological characteristics of 66 de novo acute lymphoblastic leukaemia patients according to P-glycoprotein (P-gp), multidrug resistance-associated protein (MRP) and lung resistancerelated protein (LRP) expression.
MDR phenotype Patient number Age Mean ± SD Median (range) Sex (female/male) Immunophenotype B-cell lineage T-cell lineage WBC count (109/l) Mean ± SD Median (range) CD34 expression Positive Negative LDH > 460 UI/l IDA < 280 NMFI > 280 NMFI Karyotype Favourable Unfavourable Undefined
Negative cases (P-gp–/MRP–/LRP–)
Overexpressing cases (only one protein)
Co-overexpressing cases (two or more proteins)
24/66 (36%)
24/66 (36%)
18/66 (28%)
40Æ5 ± 18Æ7 36Æ5 (14–80) 15/9
43Æ2 ± 14Æ4 40 (25–72) 13/11
39Æ3 ± 16Æ7 36 (15–80) 9/9
22/24 (92%) 2/24 (8%)
21/24 (87%) 3/24 (13%)
52Æ5 ± 107Æ2 12Æ1 (1Æ2–496)
48Æ4 ± 64Æ2 18Æ2 (0Æ8–219)
32Æ9 ± 33Æ3 15Æ4 (2–117)
18/24 (75%) 6/24 (25%) 18/22 (81%)
17/24 (71%) 7/24 (29%) 21/23 (91%)
10/18 (56%) 8/18 (44%) 11/18 (61%)
4/24 (16%) 20/24 (84%)
21/24 (87%)*** 3/24 (13%)
17/18 (94%)** 1/18 (6%)
8/20 (40%) 10/20 (50%) 2/20 (10%)
7/17 (41%) 9/17 (53%) 1/17 (6%)
9/18 (50%)* 9/18 (50%)
9/15 (60%) 5/15 (33%) 1/15 (7%)
We classified as negative those cases without overexpression for all three proteins (P-gp–/MRP–/LRP–) as overexpressing those with the overexpression of only one protein (P-gp, MRP or LRP) and as co-overexpressing those cases positive for at least two proteins. (*P ¼ 0Æ004; **P ¼ 0Æ0001; ***P ¼ 0Æ0001)
et al, 1997; Ferrando & Look, 2000), the karyotype was defined as favourable in 36 cases (26 normal diploid cases plus 10 favourable abnormal cases) and unfavourable in 33 cases. Four cases had an undefined karyotype, i.e. cytogenetic abnormalities unrelated to differences in prognosis. No relationship could be detected between MDR phenotype and favourable or unfavourable karyotype. In fact, the karyotype was favourable in 16/32 (50%) P-gp+, 20/41 (49%) P-gp– and in 8/20 (40%) patients with no MDR-related protein overexpression. An unfavourable karyotype was observed in 15/32 P-gp+ cases (47%) as well as in 18/41 P-gp– (44%) and 10/20 (50%) cases lacking any protein overexpression. In co-overexpressing cases, an unfavourable karyotype was observed in 5/15 cases (33%). Tables II and III summarize the data. The presence of the Ph chromosome had no relationship with MDR phenotypes in 10/35 Ph+ cases (28%) overexpressing at least one MDRrelated protein versus 9/20 (45%) cases showing no MDR protein overexpression at all (P-gp–/LRP–/MRP–). Response to treatment Response to treatment and P-gp, LRP and MRP expression. Seventy-six out of 95 (80%) patients obtained a CR. Deaths during induction (DDI) and primary resistance (PR) rates were not statistically different between P-gp-positive
and -negative cases (P ¼ 0Æ27 and 0Æ24 respectively). CR rates were similar in P-gp-positive (33/43, 76%) and -negative (43/50, 86%) patients (P ¼ 0Æ81). Nevertheless, relapses within 6 months were more frequent in P-gppositive (15/33, 45%) than in P-gp-negative (7/43, 16%) cases (P ¼ 0Æ03). Accordingly, disease-free survival (DFS) duration was shorter in P-gp-positive (6 months) than in P-gp-negative cases (22 months) (P ¼ 0Æ007, Fig 1). Overall survival (OS) was calculated from the date of diagnosis to the date of death or last follow-up (range from 1 to 136 months), irrespective of the treatment given after induction and consolidation therapy. Survival of P-gpnegative patients (median 14Æ5, range 1–136) was not longer than that of P-gp-positive patients (median 12, range 1–108) (P ¼ 0Æ38). The response to treatment in the 24 P-gp, MRP and LRP non-overexpressing patients was the following: 4/24 cases (16%) DDI, 18/24 cases (76%) CR, 2/24 (8%) patients did not respond to therapy. Four out of 18 (22%) cases relapsed within 6 months. As regards patients with more than one MDR-related protein, 4/5 (80%) P-gp and MRP co-overexpressing patients obtained a CR and three of them relapsed in < 6 months; 7/8 (88%) P-gp and LRP co-overexpressing patients obtained a CR but three had an early relapse. Two out of two (100%) P-gp, MRP and LRP co-overexpressing
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
524
D. Damiani et al
Fig 1. Kaplan–Meier estimate of disease-free survival (DFS) duration of P-gp-positive (45 cases) and P-gp-negative (50 cases) patients. Median DFS was 6 and 22 months respectively (P ¼ 0Æ007, log rank test).
patients obtained a CR with one patient relapsing within 6 months. Because of the small number of co-overexpressing cases, no statistical analysis could be performed. Response to treatment and other prognostic factors. CR rates and DFS were compared with several clinical or biological parameters, including age, immunophenotype, MDR features (P-gp, LRP MRP expression and ex vivo blast cell IDA), karyotype, CD34 expression, white blood cell (WBC) count, lactate dehydrogenase (LDH) levels, treatment protocol with daunorubicin or idarubicin. In univariate as well as multivariate analysis, lower CR rates were significantly associated only with an advanced age (P ¼ 0Æ016). Differently, DFS duration was significantly associated with P-gp expression (P ¼ 0Æ007) and blast cell IDA (P ¼ 0Æ0026), both in univariate and multivariate analysis. At the time of diagnosis no difference was observed in clinical and laboratory characteristics, MDR phenotype or treatment outcome between patients who received an induction course containing daunorubicin and those who received an induction course containing idarubicin. DISCUSSION In this study we evaluated the expression and function of three MDR-related proteins, P-gp, LRP and MRP, in the leukaemic blast cells of 95 adult ALL patients at the onset of the disease, and we related this data to their biological and clinical features and their treatment outcome. An ex vivo evaluation of the ability of blast cells to accumulate daunorubicin was used as a marker of protein activity. Protein expression was evaluated by flow cytometry using three specific monoclonal antibodies, and the results were expressed as mean fluorescence index (MFI ¼ the ratio between the mean fluorescence intensity of cells incubated with MRK-16, LRP-56 or MRPm6, and the mean fluorescence intensity of respective controls) according to our previous studies on AML (Michieli et al, 1997, 1999).
Leukaemic cells were defined as overexpressing (positive) when their MFI exceeded the highest MFI of negative (nonMDR) cell lines and that of normal bone marrow or peripheral blood mononuclear cells (Michieli et al, 1997). Our data clearly showed that P-gp overexpression in AML was associated with a failure in achieving or maintaining a CR as stated in the literature (Marie et al, 1991; Pirker et al, 1991; Campos et al, 1992; Michieli et al, 1992; Goasguen et al, 1993; Ino et al, 1994; Wood et al, 1994; Zo¨chbauer et al, 1994). The clinical significance of a P-gp overexpression has been poorly studied in ALL. In childhood ALL conflicting results preclude any conclusion, mainly owing to the different probes and methods used in defining a P-gp overexpression (Goasguen et al, 1993, 1996; Beck et al, 1995; den Boer et al, 1998; Dhooge et al, 1999). Data on adult ALL is still scarce (Kuwazuru et al, 1990; Wattel et al, 1995) and the role of MDR is still debatable. In this study involving a series of 95 consecutive adult cases evaluated at onset, P-gp overexpression was found in 47% of cases. P-gp overexpression was significantly associated only with an advanced age and a high WBC count, while no relationship was found with sex, immunophenotype, CD34 expression and karyotype including the presence of the Ph chromosome. Although we used a different technique, the percentage of positive cases in our study is in agreement with the series of patients reported in the literature, regardless of their adult or paediatric age, in which the amount of P-gp+ cases varied between 14% and 59% (Goasguen et al, 1993, 1996; Wattel et al, 1995; Dhooge et al, 1999). Goasguen et al (1993) reported a slightly higher incidence of P-gp expression in B-ALL (with membrane Ig) and in patients with an abnormal karyotype, although differences were not significant. Our data agrees only in part with that reported in the literature, because the majority of our B-mature ALL (8/13 cases) showed P-gp overexpression, while no association was observed with cytogenetics. We found a remarkable lower frequency of LRP overexpression in adult ALL (12/66, 18%) with respect to an analogous series of adult ANLL at onset (44/96, 46%) that we had previously studied using the same methods (Michieli et al, 1999). LRP positivity was more frequent in the T-cell lineage. No relationship was found between LRP and other clinical or biological features. Goasguen et al (1996) found a higher frequency of LRP positivity (35%) in a mixed paediatric and adult patient group, while Volm et al (1997) described 47% of LRP+ cases in a series of childhood ALL. den Boer et al (1998) reported that, in childhood leukaemias, cases overexpressing LRP showed stain intensity (MFI) significantly lower in ALL compared with ANLL. No data is available in the literature about MRP expression using flow cytometry in a series of adult ALL. In children, den Boer et al (1998) found that MRP expression did not differ between onset and relapse. Similar to the data observed in ANLL, MRP was rarely overexpressed in our series of adult ALL (17%) and no correlation was found between MRP positivity and clinical or biological parameters (Michieli et al, 1999). We used the intracellular DNR accumulation, expressed as NMFI (see Patients and methods), to highlight the
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
MDR-related Protein Expression in De Novo Adult ALL association between an MDR protein overexpression and protein activity leading to an impaired ability of leukaemic cells to accumulate anthracyclines. As with MFI, cells with an NMFI lower than normal peripheral or marrow leucocytes are believed to have a reduced IDA (Michieli et al, 1997). IDA was significantly lower in P-gp-positive cases with respect to their negative counterparts: 38/42 (90%) cases overexpressing at least one protein showed a defect in accumulating DNR, while only 4/24 (16%) cases lacking any protein overexpression (P-gp–/LRP–/MRP–) had similar features. An explanation for these four cases could be the overexpression of another drug resistance-related protein, different from P-gp, MRP and LRP, that we have not investigated, such as the BCR protein (breast cancer resistance protein) (Ross et al, 2000). An important consideration is that, in our series, all the proteins were associated with a low IDA so that a simultaneous evaluation of all MDR proteins was necessary for a correct screening of the blast cells ability to accumulate drugs. This is true in particular for MRP- and LRP-negative cases who were frequently found to be P-gp-overexpressing cases with a low IDA. In our study, only P-gp positivity and a low IDA were significantly associated with shorter CR duration. This is in agreement with the results of Dhooge et al (1999) who found that P-gp is an independent prognostic factor in paediatric ALL and determines a shorter event-free survival. Goasguen et al (1996) found that the median survival for LRP– cases was statistically higher than for LRP+ cases, and in their series of adult and paediatric patients the co-expression of P-gp and LRP was related to a shorter survival. Volm et al (1997) found significantly longer relapse-free intervals in children without LRP expression. In our cases we did not find any significant difference in response to treatment or CR duration between LRP+ or MRP+ cases and their negative counterparts, either in unior multivariate analysis. Nevertheless, the number of LRPor MRP-positive cases was remarkably small. Moreover, a high percentage of MRP– or LRP– cases had P-gp overexpression. Thus, in our opinion the real role played by LRP and MRP is still unclear. In conclusion, in agreement with adult AML, this data suggests that in adult ALL, P-gp overexpression as well as a low IDA is a frequent feature, especially in cases with advanced age or with high WBC counts. These MDR characteristics could be the biological reason for the poorer outcome in adult ALL than in childhood ALL. In contrast, an MRP and LRP overexpression is less frequent and often accompanies P-gp overexpression. The role of a simultaneous P-gp, LRP or MRP overexpression is still unclear. The clinical outcome in our series of patients showed that advanced age is the main factor negatively associated with CR rates. In fact, a P-gp positivity and a low IDA were not correlated with the achievement of CR. Nevertheless, the multivariate analysis highlighted the fact that even P-gp expression and a low IDA were the only parameters affecting CR duration (DFS duration). No correlation was found between overall survival and MDR-related protein expression or other clinical and biological parameters. This discrepancy could be explained by the differences in the
525
consolidation, maintenance and intensification therapy offered to patients, which often included an autologous or allogeneic bone marrow transplantation. Larger studies on MDR in ALL, including a multiparametric evaluation of MDR proteins, are needed to confirm our findings and to clarify whether evaluation may be a useful ALL prognostic factor that could be taken in account when planning maintenance therapy. ACKNOWLEDGMENTS This work was supported by CNR, Comitato Nazionale per le Scienze Biologiche e Mediche, Contratto no. 98.00407.CT04 e 99.02375.CT04 and by MURST, Programmi di Ricerca Scientifica di Rilevante Interesse Nazionale, 1998. REFERENCES Beck, J., Handgretinger, R., Dopfer, R., Klingebiel, T., Niethammer, D. & Gekeler, V. (1995) Expression of mdr1, mrp, topoisomerase IIa/b, and cyclin A in primary or relapsed states of acute lymphoblastic leukemias. British Journal of Haematology, 89, 356–363. Bennet, J.M., Catovsky, D., Daniel, M.T., Flandrin, G., Galton, D., Gralnick, H. & Sultan, C. (1985) Proposed revised criteria for the classification of acute leukemia cells. Blood, 79, 3267–3273. den Boer, M.L., Pieters, R., Kazemier, K.M., Rottier, M.M.A., Zwaan, C.M., Kaspers, G.J.L., Janka-Schaub, G., Henze, G., Creutzig, U., Scheper, R.J. & Veerman, A.J.P. (1998) Relationship between major vault protein/lung resistance protein, multidrug resistance-associated protein, P-glycoprotein expression, and drug resistance in childhood leukemia. Blood, 91, 2092–2098. Campos, L., Guyotat, D., Archimbaud, E., Almard-Oriol, P., Tsuruo, T., Troncy, J., Treille, D. & Fiere, D. (1992) Clinical significance of multidrug resistance P-glycoprotein expression on acute non lymphoblastic leukemia cells at diagnosis. Blood, 79, 473–476. Charrin, C. for the Groupe Franc¸ais de Cytogenetique Hematologique (1996) Cytogenetic abnormalities in adult acute lymphoblastic leukemia: correlations with hematologic findings and outcome. A collaborative study of the Group Franc¸ais de Cytogenetique Hematologique. Blood, 87, 3135–3142. Chaudhary, P.M., Mechetner, E.B. & Roninson, I.B. (1992) Expression and activity of the multidrug resistance P-glycoprotein in human peripheral blood lymphocytes. Blood, 80, 2735–2739. Chin, J.E., Soffir, R., Noonan, K.E., Choi, K. & Roninson, I.B. (1989) Structure and expression of the human MDR (P-glycoprotein) gene family. Molecular and Cellular Biology, 9, 3808–3820. Cole, S.P.C., Bhardwaj, G., Gerlach, J.H., Mackie, J.E., Grant, C.E., Almquist, K.C., Stewart, A.J., Kurz, E.U., Duncan, A.M.V. & Deeley, R.G. (1992) Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science, 258, 1650–1656. Copelan, E.A. & McGuire, E.A. (1995) The biology of acute lymphoblastic leukemia in adults. Blood, 85, 1151–1168. Cox, D.R. (1970) The Analysis of Binary Data. Chapman & Hall, London. Damiani, D., Michieli, M., Michelutti, A., Fanin, R., Russo, D. & Baccarani, M. (1995) P170 glycoprotein expression and impaired anthracycline retention in chronic myeloid leukemia. Leukemia and Lymphoma, 17, 289–294. Damiani, D., Michieli, M., Ermacora, A., Candoni, A., Raspadori, D., Geromin, A., Stocchi, R., Grimaz, S., Masolini, P., Michelutti, A., Scheper, R.J. & Baccarani, M. (1998) P-glycoprotein (P-gp) and
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
526
D. Damiani et al
not lung resistance-related protein (LRP) is a negative prognostic factor in secondary leukemias. Haematologica, 83, 290–297. Dhooge, C., De Moerloose, B., Laureys, G., Kint, J., Ferster, A., De Bacquer, D., Philippe, J. & Benoit, Y. (1999) P-glycoprotein is an independent prognostic factor predicting relapse in childhood acute lymphoblastic leukemia: results of a 6 year prospective study. British Journal of Haematology, 105, 676–683. Drach, D., Zhao, S., Drach, J., Mahadevia, R., Gattringer, C., Huber, H. & Andreeff, M. (1992) Subpopulation of normal peripheral blood and bone marrow cell express a functional multidrug resistant phenotype. Blood, 80, 2729–2734. Endicott, J.A. & Ling, V. (1989) The Biochemistry of P-glycoprotein mediated multidrug resistance. Annual Review of Biochemistry, 58, 137–144. Ferrando, A.A. & Look, A.T. (2000) Clinical implications of recurring chromosomal and associated molecular abnormalities in acute lymphoblastic leukemia. Seminars in Hematology, 37, 381–395. Goasguen, J.E., Dossot, J.M., Fardel, O., Le Mee, F., Le Gall, E., Leblay, R., Le Prise, P.:Y., Chaperon, J. & Fauchet, R. (1993) Expression of the multidrug resistance-associated P glycoprotein (P-170) in 59 cases of de novo acute lymphoblastic leukemia: prognostic implications. Blood, 81, 2394–2398. Goasguen, J.E., Lamy, T., Bergeron, C., Sunaram, B.L., Mordelet, E., Gorre, G., Dossot, J.M., Le Gall, E., Grosbois, B., Le Prise`, P.Y. & Fauchet, R. (1996) Multifactorial drug-resistance phenomenon in acute leukemias: impact of P170–MDR1, LRP56 protein, glutathione-transferases and metallothionein systems on clinical outcome. Leukemia and Lymphoma, 23, 567–576. Goker, E., Lin, J.T., Trippett, T., Elisseyeff, Y., Tong, W.P., Niedzwiecki, D., Tan, C., Steinherz, P., Schweitzer, B.I. & Bertino, R. (1993) Decreased polyglutamylation of methotrexate in acute lymphoblastic leukemia blasts in adults compared to children with this disease. Leukemia, 7, 1000–1004. Grant, C.E., Valdimarsson, G., Hipfner, D.R., Almquist, K.C., Cole, S.P. & Deeley, R.G. (1994) Overexpression of multidrug resistanceassociated protein (MRP) increases resistance to natural products drugs. Cancer Research, 54, 357–361. Hart, S.M., Ganeshaguru, K., Hoffbrand, A.V. & Metha, A.B. (1993) Expression of multidrug resistance-associated protein (MRP) gene in acute myeloid leukemia (AML) (Abstract). Blood, 82, 258. Hoelzer, D. (1993) Aggressive chemotherapy of ALL in elderly patients. Hematological Oncology, 11, 12. Ino, T., Miyazaki, H., Isogai, M., Nomura, T., Tsuzuki, M., Tsuruo, T., Ezaki, K. & Hirano, M. (1994) Expression of P-glycoprotein in de novo acute myelogenous leukemia at initial diagnosis: results of molecular and functional assays, and correlation with treatment outcome. Leukemia, 8, 146–151. Kaplan, E.L. & Meier, P. (1958) Non-parametric estimation from incomplete observations. Journal of American Statistical Association, 53, 457–481. Kuiper, C.M., Broxterman, H.J., Baas, F., Schuuruius, G.J., Haisman, H.J., Scheffer, G.L., Lankelma, J. & Pinedo, H.M. (1990) Drug transport variant without P-glycoprotein overexpression from a human squamous lung cancer cell line after selection with doxorubicin. Journal of Cellular Pharmacology, 1, 35–41. Kuwazuru, Y., Yoshimura, A., Hanada, S., Utsunomiya, A., Makino, T., Ishibashi, K., Kodama, M., Iwahashi, M., Arima, T. & Akiyama, S.I. (1990) Expression of the multidrug transporter, P-glycoprotein, in acute leukemia cells and correlation to clinical drug resistance. Cancer, 66, 868–873. List, A.F. (1993) Multidrug resistance: clinical relevance in acute leukemia. Oncology, 7, 23–30. List, A.F., Spier, C.S., Abbaszadegan, M., Grogan, T.M., Greer, J.P., Wolff, S.N., Scheper, R.J. & Dalton, W.S. (1993) Non P-glyco-
protein (P-gp) mediated multidrug resistance (MDR): identification of a novel drug resistance phenotype with prognostic relevance in acute myeloid leukemia (AML) (Abstract). Blood, 82, 1755. Luk, C.K. & Tannock, I.F. (1989) Flow cytometric analysis of doxorubicin accumulation in cells from human and rodent cell lines. Journal of the National Cancer Institute, 81, 55–59. Marie, J.P., Zittoun, R. & Sikic, B.I. (1991) Multidrug resistance (mdr1) gene expression in adult acute leukemias: correlation with treatment outcome and in vitro drug sensitivity. Blood, 78, 586–592. Michelutti, A., Michieli, M., Damiani, D., Melli, C., Geromin, A., Russo, D., Fanin, R. & Baccarani, M. (1994) Overexpression of MDR related p170 glycoprotein in chronic myeloid leukemia. Haematologica, 70, 200–204. Michieli, M., Damiani, D., Geromin, A., Michelutti, A., Fanin, R., Raspadori, D., Russo, D., Visani, G., Dinota, A., Pileri, S., Tsuruo, T., Grandi, M., Baccarani, M. & Tura, S. (1992) Over-expression of multidrug resistance–associated p170 glycoprotein in acute non-lymphocytic leukaemia. European Journal of Haematology, 48, 87–92. Michieli, M., Damiani, D., Ermacora, A., Raspadori, D., Michelutti, A., Grimaz, S., Fanin, R., Russo, D., Lauria, F., Masolini, P. & Baccarani, M. (1997) P-glycoprotein (P-gp) and lung-resistancerelated protein (LRP) expression and function in leukaemic blast cells. British Journal of Haematology, 96, 356–365. Michieli, M., Damiani, D., Ermacora, A., Masolini, P., Raspadori, D., Visani, G., Scheper, R.J. & Baccarani, M. (1999) P-glycoprotein, lung resistance-related protein and multidrug resistance associated protein in de novo acute non-lymphocytic leukaemias: biological and clinical implications. British Journal of Haematology, 104, 328–335. Michieli, M., Damiani, D., Ermacora, A., Geromin, A., Michelutti, A., Masolini, P. & Baccarani, M. (2000) P-glycoprotein (P-gp), lung resistance-related protein (LRP) and multidrug resistanceassociated protein (MRP) expression in acute promyelocytic leukemia. British Journal of Haematology, 108, 703–709. Peto, R., Pike, M.C., Armitage, P., Breslow, N.E., Cox, D.R., Howard, S.V., Mantel, N., McPherson, K., Peto, J. & Smith, P.G. (1977) Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II analysis and examples. British Journal of Cancer, 35, 1–39. Pileri, S.A., Sabattini, E., Falini, B., Tazzari, P.L., Gherlinzoni, F., Michieli, M.G., Damiani, D., Zucchini, L., Gobbi, M., Tsuruo, T. & Baccarani, M. (1991) Immunohistochemical detection of the multidrug transport protein P170 in human normal tissues and malignant lymphomas. Histopathology, 19, 131–140. Pirker, R., Wallner, J., Geissler, K., Linkesch, W., Hass, O.A., Bettelheim, P., Hopjner, M., Scherrer, R., Valent, P., Havelec, L., Ludwig, H. & Lechner, K. (1991) MDR1 gene expression and treatment outcome in acute myeloid leukemia. Journal of the National Cancer Institute, 83, 708–712. Riccardi, A., Invernizzi, R., Ucci, G., Luoni, R., Danova, M., Giordano, M., Zambelli, M.L., Rastaldi, M.P. & Ascari, E. (1991) P170 glycoprotein in multiple myeloma and acute leukemia. Haematologica, 76, 177–180. Ross, D.D., Karp, J.E., Chen, T.T. & Doyle, L.A. (2000) Expression of breast cancer resistance protein in blast cells from patients with acute leukemia. Blood, 96, 365–368. Scheffer, G.L., Wijingaard, P.J.L., Flens, M.J., Izquierdo, M.A., Slovak, M.L., Pinedo, H.M., Meijer, C.J.L.M., Clevers, H.C. & Scheper, R.J. (1995) The drug resistance-related protein LRP is the human major vault protein. Nature Medicine, 1, 578–582. Scheper, R.J., Broxterman, H.J., Scheffer, G.L., Kaaijk, P., Dalton, W.S., van Heijningen, T.H.M., van Kalken, G.K., Slovak, M.L.,
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
MDR-related Protein Expression in De Novo Adult ALL de Vries, E.G.E., van der Valk, P., Meijer, C.J.L.M. & Pinedo, H.M. (1993) Over-expression of a Mr 110,000 vesicular protein in non-P-glycoprotein-mediated multidrug resistance. Cancer Research, 53, 1475–1479. Schneider, E., Cowan, K.H., Bader, H., Toomey, S., Schwartz, G.N., Karp, J.E., Burke, P.J. & Kaufmann, S.H. (1995) Increased expression of the multidrug resistance-associated protein gene in relapsed acute leukemia. Blood, 85, 186–193. Secker-Walker, L.M., Prentice, H.G., Durrant, J., Richards, S., Hall, E. & Harrison, G. on behalf of the MRC Leukemia Working Party (1997) Cytogenetics adds independent prognostic information in adults acute lymphoblastic leukaemia on MRC trial UKALL XA. British Journal of Haematology, 96, 601–610. Solary, E., Bidan, J.M., Calvo, F., Chauffert, B., Caillot, D., Mugneret, F., Gauville, C., Tsuruo, T., Carli, P.M. & Guy, H. (1991) P-glycoprotein expression and in vitro reversion of doxorubicin resistance by verapamil in clinical specimens from acute leukemia and lymphoma. Leukemia, 5, 592–597. Sugawara, I., Kataoka, I., Morishita, Y., Hamada, H., Tsuruo, T., Itoyama, S. & Mori, S. (1988) Tissue distribution of P-glycoprotein encoded by a multidrug resistant gene as revealed by a monoclonal antibody, MRK16. Cancer Research, 48, 1926–1929. Thiebaut, F., Tsuruo, T., Hamada, H., Gottesman, M., Pastan, I. & Willingham, M.C. (1987) Cellular localization of the multidrug resistance gene product P-glycoprotein in normal human tissues.
527
Proceeding of the National Academy of Sciences of the United States of America, 84, 7735–7738. Tsuruo, T. (1988) Mechanisms of multidrug resistance and implications for therapy. Japanese Journal of Cancer Resistance, 79, 285–296. Volm, M., Stammler, G., Zintl, F., Kooma¨gi, R. & Sauerbrey, A. (1997) Expression of lung resistance-related protein (LRP) in initial and relapsed childhood acute lymphoblastic leukemia. Anticancer Drugs, 7, 662–665. Wattel, E., Lepelley, P., Merlat, A., Sartiaux, C., Bauyters, F., Jouet, J.P. & Fenaux, P. (1995) Expression of the multidrug resistance P glycoprotein in newly diagnosed adult acute lymphoblastic leukemia: absence of correlation with response to treatment. Leukemia, 9, 1870–1874. Weinstein, R.S., Kuszak, J.R., Kluskens, L.F. & Coon, J.S. (1990) P-glycoprotein in pathology: the multidrug resistance gene family in humans. Human Pathology, 21, 34–48. Wood, P., Burgess, R., MacGregor, A. & Liu Yin, J.A. (1994) P-glycoprotein expression on acute myeloid leukaemia blast cells at diagnosis predicts response to chemotherapy and survival. British Journal of Haematology, 87, 509–513. Zo¨chbauer, S., Gsur, A., Brunner, R., Kyrle, P.A., Lechner, K. & Pirker, R. (1994) P-glycoprotein expression as unfavourable prognostic factor in acute myeloid leukemia. Leukemia, 8, 974–977.
Ó 2002 Blackwell Science Ltd, British Journal of Haematology 116: 519–527
