Lj090001867p

M i t o x a n t r o n e , E t o p o s i d e , a n d C y c l o s p o r i n e T h e r a p y i n
P e d i a t r i c P a t i e n t s W i t h R e c u r r e n t o r R e f r a c t o r y A c u t e
M y e l o i d L e u k e m i a
By Gary V. Dahl, Norman J. Lacayo, Nathalie Brophy, Kyriaki Dunussi-Joannopoulos, Howard J. Weinstein, Myron Chang, Purpose: To determine the remission rate and toxic-
remission, and 9% of patients (n ؍ 6) died of infection.
ity of mitoxantrone, etoposide, and cyclosporine (MEC)
Exposure to CSA levels of greater than 2,400 ng/mL
therapy, multidrug resistance-1 (MDR1) status, and
was achieved in 95% of patients (n ؍ 56 of 59). Toxic-
steady-state cyclosporine (CSA) levels in children with
ities included infection, cardiotoxicity, myelosuppres-
relapsed and/or refractory acute myeloid leukemia.
sion, stomatitis, and reversible increases in serum cre-
Patients and Methods: MEC therapy consisted of mi-
atinine and bilirubin. In most who had relapsed while
toxantrone 6 mg/m2/d for 5 days, etoposide 60 mg/
receiving therapy or whose induction therapy had
m2/d for 5 days, and CSA 10 mg/kg for 2 hours followed
failed, response was not significantly different for
by 30 mg/kg/d as a continuous infusion for 98 hours.
MDR1-positive and MDR1-negative patients.
Because of pharmacokinetic interactions, drug doses
Conclusion: Serum levels of CSA capable of revers-
were decreased to 60% of those found to be effective
ing multidrug resistance are achievable in children with
without coadministration of CSA. MDR1 expression was
acceptable toxicity. The CR rate of 35% achieved in this
evaluated by reverse transcriptase polymerase chain re-
study is comparable to previously reported results us-
action, flow cytometry, and the ability of CSA at 2.5
ing standard doses of mitoxantrone and etoposide. The
␮mol/L to increase intracellular accumulation of 3H-
use of CSA may have improved the response rate for
daunomycin in blasts from bone marrow specimens.
the MDR1-positive patients so that it was not different
Results: The remission rate was 35% (n ؍ 23 of 66).
from that for the MDR1-negative patients.
Overall, 35% of patients (n ؍ 23) achieved complete
J Clin Oncol 18:1867-1875. 2000 by American
remission (CR), 12% of patients (n ؍ 8) achieved partial
Society of Clinical Oncology.
WITH CURRENT chemotherapy regimens, approxi- toxantrone, respectively), the vinca alkaloids (vincristine mately 75% to 85% of newly diagnosed children and vinblastine), and the epipodophyllotoxins (etoposide with acute myeloid leukemia (AML) will achieve complete remission (CR).1 Further consolidation/maintenance che- Several noncytotoxic drugs, such as verapamil and phe- motherapy results in 5-year event-free survival (EFS) of nothiazines, have been shown to modulate MDR, at least in 35% to 60%.2-6 The major reason for treatment failure is part by competitive inhibition of P-gp–mediated drug ef- leukemic relapse, which is most likely to occur within the flux.11 Verapamil and phenothiazines modulate MDR and first 2 years after diagnosis. Only 30% to 50% of patients P-gp at drug concentrations that produce unacceptable who relapse will achieve a second remission with chemo- clinical toxicities (heart block and depression of the CNS).
therapy, and these remissions are not durable. The likeli- Cyclosporine (CSA) has also been shown to be a potent hood of achieving a second remission is directly propor- inhibitor of P-gp drug efflux in vitro. CSA concentrations of tional to the length of the first remission.7 The lack ofresponse to chemotherapy suggests the development ofacquired drug resistance.
From the Divisions of Pediatric Oncology, Medical Oncology, and Drug resistance may be present at the time of diagnosis or Clinical Pharmacology, Stanford University School of Medicine, Palo it may arise by somatic mutations during tumor growth.
Alto, and Children’s Cancer Group, Arcadia, CA, and Pediatric Numerous reports implicate the increased expression of the multidrug resistance-1 (MDR1)– encoded P-glycoprotein Submitted August 4, 1999; accepted January 5, 2000.
Supported by the Career Development Award from the American (P-gp) multidrug transporter as one cause of both intrinsic Society of Clinical Oncology, Alexandria, VA, and by grants no. R01 and acquired drug resistance.8-10 P-gp is a transmembrane CA 52168 and M01 RR 00070 from the National Institutes of Health, ATP-dependent efflux pump that has broad substrate spec- Bethesda, MD (to General Clinical Research Center, Stanford Univer- ificity. Increased amounts of P-gp confer multidrug resis- tance (MDR) in cells by reducing intracellular accumulation Address reprint requests to Gary V. Dahl, MD, Department of Pediatrics, Division of Hematology, Oncology, 300 Pasteur Dr, Room of a variety of cytotoxic drugs. Several key drugs used in the G313, Stanford, CA 94305-5208; email [email protected]. therapy of AML are transported by P-gp. These include 2000 by American Society of Clinical Oncology. anthracyclines and anthracenediones (daunorubicin and mi- Journal of Clinical Oncology, Vol 18, No 9 (May), 2000: pp 1867-1875 1,000 to 2,000 ng/mL, which reverse P-gp in vitro, are Table 1. Patient Characteristics
readily achievable with short-term intravenous administra- tion in adults patients with acceptable toxicities.12 Several studies in patients with AML have shown that P-gp overexpression is relatively frequent, especially in refractory and relapsed leukemia.8-10 In several adult AML studies, CR rates were significantly lower in AML patients with overexpression of MDR1/P-gp, but the overall prog- nostic significance of P-gp expression on outcome remains controversial. Interpretation of the clinical data has been hampered by lack of well-defined criteria for detecting a level of P-gp expression considered positive and inadequate standardization of drug efflux assays.13 Two recent reports of phase II trials using cyclosporines as MDR modulators in adult AML patients have shown improved remission rates in We present the results of the largest pediatric study of an MDR reversal agent, and the first in childhood AML. The goals of this phase II study were (a) to determine the remission rate and toxicity of mitoxantrone, etoposide, and CSA (MEC) in children with refractory or relapsed AML,(b) to document steady-state CSA levels of greater than2,400 ng/mL, (c) to determine MDR1 mRNA and P-gpexpression in leukemic blasts, and (d) to perform functional The diagnosis of AML was made on bone marrow smears routinely stained and evaluated according to the revised French-American- A phase III AML study—Pediatric Oncology Group British (FAB) criteria.17 All patients had normal cardiac, renal, liver,and pulmonary function, and all but two had received less than 500 (POG) 9421—with etoposide and mitoxantrone with/with-
mg/m2 of prior daunorubicin therapy. The median prior daunorubicin out CSA during the consolidation phase has just been dose was 265 mg/m2 (range, 72 to 500 mg/m2), and four patients had received additional cardiotoxic therapy (amsacrine in three and idaru-bicin in one). This study was approved by the institutional review boards of each participating POG institution, and the patient and/orparent or guardian was required to sign an approved consent form.
Between April 1992 and June 1994, a total of 68 pediatric patients with AML were entered onto the MEC protocol, a POG phase II multicenter trial (POG 9222). These children had received unsucessful The National Cancer Institute criteria for CR and partial remission induction therapy with initial therapy or had relapsed during therapy or (PR) were used. A CR was defined as the presence of a morphologi- after one or more CRs. Forty-nine of the 66 assessable patients were cally normal bone marrow, a granulocyte count of at least 1,500/␮L, initially treated on or as per POG 8821.16 In this study, patients and a platelet count of 100,000/␮L. A PR was defined as 5.1% to received induction therapy with daunorubicin, cytarabine (Ara-C), and 25.0% of bone marrow blast cells with recovering counts and no thioguanine, followed by a second induction with high-dose Ara-C circulating blasts. Relapse was defined as more than 25% blasts.17 (total anthracyclines, 135 mg/m2). Patients who entered remission wereeligible for randomization. After randomization, an additional course of etoposide and azacytidine with intrathecal Ara-C was given, followedby either intensive chemotherapy (cumulative anthracyclines, 360 Echocardiography or planar radionuclide ventriculography (multi- mg/m2), autologous transplantation, or allogeneic transplantation. Eight gated angiogram scan) was required of all patients before the start of patients were treated according to POG pilot study 9194 (three cycles the study and before each course of therapy. If the shortening fraction with daunorubicin, high-dose Ara-C, and thioguanine; etoposide and was greater than 27% or ejection fraction was greater than 50%,therapy could be initiated. If clinical heart failure developed or the high-dose Ara-C; and high-dose Ara-C and daunorubicin [cumulative shortening fraction dropped below 25%, the second cycle of therapy anthracyclines, 225 mg/m2]). The remaining patients were treated on was held until cardiac function improved.
various anthracycline-containing induction regimens. Relapse or induc-tion failure was documented by bone marrow aspirate and biopsy in all patients. The characteristics of this patient population are listed inTable 1. In 68% of the assessable patients, either induction therapy MEC therapy consisted of mitoxantrone 6 mg/m2/d for 5 days and failed or the patient relapsed while on therapy; the remainder had first etoposide 60 mg/m2/d for 5 days in combination with CSA 10 mg/kg remission duration of less than 1 year.
for 2 hours followed by 30 mg/kg/d continuous infusion for 98 hours (total of 100 hours). The CSA dose was adjusted at hours 14, 26, 38, 50, 75%; eight samples showed 100% positive blasts). Similar results were and 74 in an attempt to maintain a steady-state serum CSA level above obtained when D values were calculated, with most samples falling in 2.5 ␮mol/L (range, 3,000 to 5,000 ng/mL).
the low expression range (D values Ͼ 0.1 and Ͻ 0.2).20 Because the A bone marrow aspirate and biopsy samples were obtained 10 to 14 number of positive leukemia samples was relatively low (a total of 14 days from the start of the 5-day MEC course. If the aspirate had more patient samples), samples were not broken down into subgroups of than 25% blasts, patients were removed from study regardless of the cellularity. If patients had fewer than 25% blasts and cellularity of less RT-PCR analysis of MDR1 expression. than 25%, they were immediately given a second 5-day MEC course.
was determined in bone marrow samples by RT-PCR using MDR1 and If fewer than 5% blasts were noted and the biopsy was hypocellular beta-2-microglobulin specific oligonucleotide primers as described by (5% to 20%), a second course was delayed until the counts recovered.
Noonan et al.24 Total RNA (100 ng) was used to make cDNA followed Patients who recovered with greater than 25% blasts before the second by amplification in a 4800 Thermocyler (Perkin Elmer, Norwalk, CT) course were removed from the study. If there was residual leukemia set for a 30-second denaturation at 94°C, 1 minute of primer annealing after two courses, the patient was taken off the study. Patients who at 55°C, and 2 minutes of extension/synthesis at 72°C. Primers used for achieved a CR with one or two courses could receive two or one amplification of MDR1 RNA were from nucleotides 2596 to 2615 for subsequent course(s), respectively, for consolidation. No more than the sense direction and nucleotides 2733 to 2752 for the antisense three courses of MEC were given to any patient.
direction, yielding a 167-nucleotide product. Primers used for ampli-fication of ␤2-microglobulin were from nucleotides 1544 to 1563 for the sense and from nucleotides 2253 to 2263 for the antisense direction.
This combination yielded a 120-nucleotide product. PCR products were Flow cytometric analysis of MDR1 expression. electrophoretically separated in a 3% agarose gel and the specific PCR was determined by both flow cytometric analysis and reverse transcrip- amplicons were excised after ethidium bromide staining. The amount tase polymerase chain reaction (RT-PCR). Analysis of MDR1 P-gp cell of radioactivity incorporated into each amplicon was then determined surface expression was performed on a FACSCAN flow cytometer by liquid scintillation counting. Analyses were done after 25 and 35 using Lysis II software (Becton Dickinson, Thousand Oaks, CA) and cycles of amplification. A sample was considered positive for MDR the 4E3 monoclonal antibody (MoAb) directed to an external epitope of expression if a band was visibly present after ethidium bromide P-gp.18,19 Similar results were obtained using MRK16, another external staining and/or the MDR1 to ␤2-microglobulin ratio of incorporated epitope anti–P-gp MoAb. Although there were subtle differences, the radioactive counts was greater than 0.7 (based on comparison with the results reported here are only for the 4E3 antibody because expression drug-sensitive [P-gp–negative] CEM cell line, which had a ratio of less of surface P-gp was observed to be similar and did not change the than 0.7 under the RT-PCR conditions used). The range of values number of patients who were positive.
varied from 0.8 to 1.8, with an average of 1.2 and a median of 1.1.
Fresh bone marrow samples were layered over Ficoll-Hypaque cushions and centrifuged at 600 ϫ g for 15 minutes to separate Cytotoxic Drug Accumulation Studies mononuclear cells, which were then washed in cold phosphate-buffered Although the functional analysis of MDR1/P-gp has evolved over the saline containing 2% fetal bovine serum. These cells were then used for past several years with the discovery of a variety of dyes effluxed by staining with 4E3, other hematopoietic markers (including CD34, the MDR transporter, when this clinical study was begun these methods CD13, CD15, CD33, CD14, CD2, CD19, and CD10), or isotype had not been established.19,25-27 The determination of intracellular drug controls. 4E3 or MRK16 was detected using secondary phycoerythrin- accumulation was used to measure the effect of verapamil and/or CSA conjugated goat-antimouse immunoglobulin (Fab)2 fragment (Tago, on reversing the level of cytotoxic drug accumulation, even though, as Burlingame, CA) at a 1/30 dilution. All other MoAbs were directly many studies have now shown, the correlation with P-gp and the conjugated to fluorescein (Becton Dickinson or AMAC, Inc, Miami, reversal of drug uptake or accumulation with these inhibitors expres- FL). 4E3 staining of gated leukemic blasts was compared with isotype sion is not perfect, in part due to the non-MDR1 transport sys- control as well as control cells expressing different levels of MDR1 P-gp (CEM, CEM/VBL-100, and CEM/VBL-300).18 The intracellular accumulation of cytotoxic agents was determined in Quantitation was accomplished initially by comparing mean fluores- the absence and presence of P-gp inhibitors verapamil and CSA, as cence intensities or using the Kolmogorov-Smirnov statistic, referred to previously described.18,27-30 3H-Daunomycin (specific activity, ϳ1 to 5 as D, which measures the difference between the means of two Ci/mmol; Dupont, Wilmington, DE) and 3H-vinblastine (specific ac- distributions.13,20-22 The ratio of the mean fluorescence intensity for tivity, ϳ5 to 25 Ci/mmol; Amersham International, Les Uris, France) 4E3 expression compared with the value using an isotype control were used in these studies. Isolated bone marrow mononuclear cells antibody was then calculated. This ratio never exceeded 1.2 for the were resuspended at 1 ϫ 106 cells/mL of RPMI medium containing drug-sensitive (P-gp–negative) cell line, CEM. AML blasts were 10% fetal bovine serum. Aliquots of 0.25 mL were then dispensed into therefore considered positive if they had a shift of Ն 1.3 times a mean 75-mm round-bottom plastic tubes and MDR reversal agents were fluorescent intensity for P-gp compared with the isotype control. In added to a final concentration of 2.5 ␮mol/L. After a 30-minute addition, Ն 10% of marrow mononuclear cells in the leukemic blast incubation with gentle agitation at 37°C, tritiated cytotoxic drug was population had to be positive for a sample to be considered positive for added to a final concentration of 8 nmol/L in the presence of the same P-gp expression.23 The observed ratios of the mean fluorescence but unlabeled drug at a final concentration of 10 ␮mol/L. After a 2-hour intensity for 4E3 staining compared with the isotype control for all incubation, the entire contents of each tube were overlaid on a 200-␮L patient samples considered to be positive by these criteria ranged from cushion of Dow 550 silicone oil (Dow Corning, Midland, MI) and 1.3 to 5, with a mean of 2.2 and a median value of 1.65. The values for mineral oil (4:1) in a 1.5-mL Eppendorf tube and centrifuged at percentage of positive cells in the leukemic blast populations from 10,000 ϫ g at 4°C for 1 minute to separate the cells from drug- different patients ranged from 10% to 100%, with a mean of 70% and containing medium. The medium and oil mixture was removed by a median of 100% (values included 10%, 11%, 13%, 20%, 53%, and aspiration, and the cellular pellets were solubilized in 1 mL of 1 M NaOH at 60°C overnight before the amount of radioactivity accumu- therapy; five achieved CR after the second course. CR rates lated in the cells was determined by liquid scintillation counting.
by patient characteristics are listed in Table 1. The remission The “fold increase” of radiolabeled drug accumulation was calcu- rates were 62%, 15%, and 33 for patients with off-therapy lated as follows: The ratio of the drug accumulation as described earlierwithout a reversal agent compared to with a reversal agent was relapse, on-therapy relapse, and primary refractory AML or calculated. Along with each patient sample, drug accumulation for the secondary AML%, respectively. The overall remission rate drug-sensitive cell line, CEM, and the drug-resistant derivative, CEM/ was 35% (23 of 66 patients), with a 95% confidence interval of VBL 300, was simultaneously determined. The values for each pa- 23% to 46%. The median EFS and median survival times were tient’s drug accumulation ratio and for CEM cells were normalized by 2.5 months and 4.5 months, respectively. Fifty-seven (86%) of dividing by the ratio for CEM/VBL 300 cells. Last, a fold increase forthe patient sample compared with CEM cells was determined. Values 66 patients died during the first year. When patients were reported are the average of at least three determinations. A fold increase stratified based on either the status of MDR1 or status of prior value of greater than 1.2 was considered positive, as this was the upper relapse, the number within each stratum was too small to level observed using the drug-sensitive cell line, CEM.18,30 The range of values for patient samples greater than 1.2 varied from 1.23 to 6.59, Marrow samples were received for determination of MDR1 status in 59 (90%) of 66 patients. Table 2 lists the disease status and response of patients with AML accordingto their MDR1 status as defined in the Materials and The significance of observed differences in proportions was tested using the ␹2 statistic and, when appropriate for small sample size, Methods. Fifty-five patients were tested by flow cytometry Fisher’s exact test. The Mantel-Haenszel test31 was used to compare for the presence of 4E3 staining; 14 (25%) were positive the CR rate in MDR1-negative patients with that in MDR1-positive and 41 (75%) were negative. Of these patients, CR was patients with adjustment for imbalance of prior relapse status. The test achieved by four (29%) of the 4E3-positive and by 14 is based on the difference between the observed number of CR in (34%) of 4E3-negative patients. Fifty patients were tested MDR1-negative patients and the expected number of CRs in the samegroup under the hypothesis that the remission rate in MDR1-negative for MDR1 status by RT-PCR. Ten (20%) were positive and patients is the same as that in MDR1-positive patients.
40 (80%) were negative. Of these patients, CR was achieved Kendall’s tau-b32 was used to evaluate the correlation between the in four (40%) of the PCR-positive patients and 13 (33%) of various measurements associated with the MDR1 status. The estimator the PCR-negative patients. Fifty-two patients were tested of Kendall’s tau-b is based on the fraction of concordant versus for MDR1 reversal by CSA in vitro. Twenty (38%) exhib- discordant pairs of observations on two variables. A value of theestimator close to Ϫ1, ϩ1, or 0 indicates strongly negative, strongly ited reversal and 32 (62%) did not. Of these patients, CR positive, or no correlation, respectively, between the variables.
was achieved in nine (45%) with positive reversal by CSAand in nine (28%) with negative reversal by CSA in blast cells. In samples from 48 patients, results of all threeMDR1/P-gp detection methods outlined earlier were avail- able. When MDR1/P-gp positivity was defined as either Sixty-eight patients (age range, 1.4 months to 20 years) 4E3ϩ, PCRϩ, ϩCSA reversal, or any combination thereof, were entered onto the study and 66 were assessable for there were 27 (56%) positive and 21 (44%) negative response and toxicity (Table 1). No patient was considered samples. Of these patients, CR was achieved in 10 (37%) of ineligible for entry onto the study, but two patients were MDR1/P-gp–positive patients and in seven (33%) of MDR1/ administered nonprotocol cytotoxic therapy and therefore were considered nonassessable. The distribution of FAB Only five patients were positive for MDR1 expression by subtypes was consistent with previously published experi- all three methods, and 21 were negative for MDR1 by all ence.33 There were two cases of treatment-related secondary three methods. The response (CR ϩ PR) rates were 60% AML (tAML). The “other” category includes one patient (three of five patients) and 38% (eight of 21 patients) for with refractory anemia with excess blasts developing after a these two groups of patients. The difference was insignifi- 3-month remission of AML and seven AML cases not given cant (two-sided P ϭ .62 by Fisher’s exact test).
The Mantel-Haenszel test shows that after adjustment for prior relapse status the CR rate in MDR1/P-gp–negative patients was not higher than that in MDR1/P-gp–positive Overall, 23 patients (35%) achieved CR, eight patients patients when MDR1/P-gp positivity is defined as ϩ4E3, (12%) achieved PR, and six patients (9%) died of infection ϩRT-PCR, ϩCSA reversal, or the presence of at least one during marrow hypoplasia. Among the complete respond- of the above (one-sided P values were .30, .55, .86, and .39, ers, median time to CR was 39 days (range, 14 to 62 days).
respectively). In addition, the variables of age, sex, and Eighteen (27%) achieved CR after a single course of FAB group were not significantly correlated with CR in Table 2. CR Outcome Versus MDR1 Status
*Defined as Ͼ 10% positive cells.
†Defined as a PCR value Ͼ 0.7 relative to 1.0 for positive control.
‡Defined as a CSA value Ͼ 1.2ϫ, which is above that of P-gp-negative cells.
§A combination of antibody and PCR-positive samples, or reversible to CSA.
univariate analysis. The two-sample binomial test shows of the hands and lips was frequently reported and resolved that the remission rate (13 of 21 patients, 62%) in patients with completion of the CSA infusion. All patients experi- with off-therapy relapse was higher than the CR rate (10 of enced profound myelosuppression, with an absolute neutro- 45 patients, 22%) in patients with on-therapy relapse or phil count of zero and platelet counts of less than 20,000/ induction failure (one-sided P ϭ .0006).
mm3. In the responding patients, the median time togranulocyte and platelet recovery was 35 days (range, 21 to Correlation Among 4E3 Staining, RT-PCR, and Cytotoxic 55 days) and 36 days (range, 22 to 69 days), respectively.
Drug Accumulation With or Without CSA The major toxicity was infection, with 28 episodes; among The Kendall’s tau-b correlation coefficients among the these, 25 patients developed documented bacterial or fungal 4E3 staining, RT-PCR, and CSA analyses were not signif- infections. As determined using National Cancer Institute icantly different from zero, except for the correlation toxicity criteria, grade 4 mucositis developed in eight coefficient (r ϭ .49) between 4E3 staining and RT-PCR children. Significant increases in serum creatinine occurred (P ϭ .0005) and the correlation coefficient (r ϭ .34) in three patients; one was grade 4 in a 19-month-old boy and between 4E3 staining and CSA modulation (P ϭ .0023).
was associated with severe mucositis, candidiasis, cardiacfailure, and early death. The other two were related to poor state of hydration before the start of the CSA infusion, and Whole blood or serum CSA levels were performed by both resolved with adequate parenteral fluid administration.
clinical laboratories at each participating institution. No Cardiotoxicity, manifested by decreased ejection fraction standard measurement method was suggested. For the 59 or shortening fraction, was noted in 15 (23%) of the patients patients with available data, the median serum CSA level after induction therapy. In 12 (80%) of these patients, after a 2- hour loading dose was 3,449 ng/mL and the mean cardiotoxic events were transient, were due to multifactorial was 2,820 ng/mL (range, 1,050 to 16,599 ng/mL). For the clinical problems, and resolved without specific therapy.
5-day continuous infusion, the median level was 3,187 Three children developed congestive heart failure that ng/mL and the mean was 3,446 ng/mL (range, 1,397 to 6,534 ng/mL). The median peak CSA level during the 5-dayinfusion was 4,911 ng/mL (range, 2,160 to 22,296 ng/mL).
Although steady-state concentrations revealed significant Table 3. Grades 3 to 4 Toxicity Among Assessable Patients
interpatient variability, prolonged exposure to levels of greater than 2,400 ng/mL was achieved in 56 of 59 patients.
Table 3 lists toxicities related to the induction regimen.
Nausea and vomiting associated with headache and tingling For the 55 patients with available data, the median total level of MDR1 transcripts by RT-PCR or RNAse protection bilirubin level during CSA infusion was 2.2 mg/100 mL assays. Later, protein detection by immunoblotting and immu- (range, 0.6 to 9.7 mg/100 mL). The mostly direct hyperbi- nohistochemistry was introduced. More recently, functional lirubinemia was rapidly reversible once the CSA infusion assays have been developed. Despite these advances, it is was discontinued and was not accompanied by any other difficult to assess the role that P-gp plays in clinical drug evidence of additional hepatic dysfunction. There was no resistance due to tumor cell heterogeneity, sensitivity, and relationship between the level of bilirubin, response rate, specificity of methods of detection. The best approach, as CSA level achieved, or MDR1/P-gp status.
recommended by the international workshop on MDR1 detec- tion methods, is to use tissue-specific controls, antibody con-trols, standardized MDR1 cell lines to calibrate detection The major objective of this protocol was to determine methods, and two or more vendor-standardized anti–P-gp whether the MDR1/P-gp reversal agent CSA could be safely antibodies, and to report immunostaining data as staining tolerated and produce significant response rates in pediatric intensity and the percentage of positive cells.11 Although this patients when given in high doses with an active reinduction study was completed before these consensus recommenda- regimen of mitoxantrone and etoposide. In this study, tions, many of these guidelines were followed. The RT-PCR high-dose CSA was administered continuously in a group- data were complemented with immunostaining and drug accu- wide setting and found to be generally safe with acceptable mulation studies with and without modulators. Standardized toxicity. Grades 3 to 4 cardiac toxicity, likely related to cell lines with low expression of MDR1/P-gp were used as a chemotherapy, occurred in three patients (4.5%), who de- veloped congestive heart failure and required treatment with Flow cytometric assessment of P-gp had an adequate correla- digitalis. Although hyperbilirubinemia was common and tion with both RT-PCR and CSA enhancement of drug accumu- directly related to CSA administration, it quickly normal- lation. Although adequate, the level of correlation highlights the ized when the infusion was discontinued.
need for protein and functional assays as necessary tools in In addition, the combination of CSA, mitoxantrone, and evaluating the MDR1 status in clinical samples and in validating etoposide was determined to be an active induction regimen the analysis of MDR1 modulation trials.
for children with relapsed and refractory AML, even though Limited data are available on the expression of MDR1/ the doses of mitoxantrone and etoposide given during P-gp in marrow or blood samples from pediatric patients induction were decreased to 60% of those found to be with AML. This series includes the largest sample of effective without coadministration of CSA. Dose modifica- relapsed or refractory pediatric AML samples analyzed for tions were necessary because high-dose CSA produces MDR1/P-gp expression and function. There is one pub- significant increases in etoposide and mitoxantrone sys- lished report of MDR1/P-gp expression in pediatric patients temic exposure and resultant leukopenia and mucositis. Our with de novo AML.36 In this study, samples from 130 group has performed pharmacokinetic analysis of patientsreceiving etoposide and mitoxantrone with and without patients were evaluated, and 30% of infants and 8% of CSA in an ongoing phase III trial.34 CSA produced signif- children older than 1 year expressed MDR1/P-gp at diagno- icant increases in the area under the curve for etoposide and sis. This study was limited to the use of the MRK-16 mitoxantrone, thus confirming the need for the dose adjust- antibody with a fluorescein isothiocyanate– conjugated sec- ments used in this study and in our ongoing phase III trial.
ond-step reagent. MDR1/P-gp positivity did not have an The CR rate of 35% achieved in this study is comparable to effect on induction failure, incidence of relapse, EFS, or other results reported with mitoxantrone/etoposide.35 As in overall survival in this group of patients. The method of other studies, the likelihood of achieving a CR was higher in analysis was substandard because it relied on a single patients who relapsed off therapy, compared with those who technique to study MDR1/P-gp status. If the data from this relapsed on therapy or did not go into remission (62% v Children’s Cancer Group study are corroborated, then our 15% and 33%, respectively). Approximately 50% of pa- group of relapsed and/or refractory pediatric AML patients tients achieved a CR or PR with this MEC regimen, which shows an increased expression of MDR1/P-gp, similar to allowed most to go on to bone marrow transplant.
what is seen when comparing de novo and relapsed AML in A second aim of this study was to determine the P-gp status of blasts in refractory childhood AML by several methods, Although the response rate for MDR1/P-gp-negative including MDR1 mRNA, P-gp expression, and drug efflux patients is not significantly higher than that for MDR1/P- studies. The methodology used to assess P-gp expression has gp–positive patients, it is interesting to note that the patients evolved over the last several years. Early studies reported the relapsing off therapy before entry onto this regimen were more likely to achieve remission and more likely to be and hyperbilirubinemia were found to be major treatment- MDR1/P-gp positive. MDR1/P-gp status as a predictor of related toxicities in this population. Remissions were seen response to induction may be important for patients for in heavily pretreated patients and in children who had failed whom induction therapy fails or for patients who relapse on to achieve an initial remission. The remission rate in therapy, but it may have little prognostic importance in MDR1/P-gp–negative patients was not significantly greater those with late relapses. It is interesting to note that the than that in MDR1/P-gp–positive patients.
prognostic importance of MDR1/P-gp expression and func- The correlation among 4E3 staining, RT-PCR, and CSA tion was not seen in this relatively large group of patients inhibition of drug efflux was adequate. Our methods of treated with an MDR1/P-gp modulator during induction. It analysis using several diagnostic tools to determine the may be that the modulator improved the response rate for MDR1 status illustrate the pitfall of relying on a single the MDR1/P-gp–positive patients so that it was not different method to diagnose the MDR1/P-gp status. Future studies from that of the MDR1/P-gp–negative patients. However, will require the use of new flow cytometric analysis re- the contribution of dose-intensity to this outcome needs to agents and strict guidelines regarding clinical sample col- lection, preservation, and amount of sample to perform It is impossible to determine the prognostic importance of MDR1/P-gp status and response to this therapy because of In the laboratory setting it has been shown that the use of variations in treatment history and drug exposure before MDR1/P-gp modulators during cytotoxic therapy can abol- admission to this protocol. The induction failure and on- and ish the selection of MDR1-mediated drug resistance in off-therapy relapse groups were not balanced for MDR1/ cancer cell lines.37 Thus a possible benefit of introducing P-gp expression. The second largest subgroup of patients MDR1/P-gp modulators in the therapy of patients with de consisted of those who relapsed off therapy. These patients novo AML may be a decrease in acquired drug resistance were more likely to be MDR1/P-gp–positive and historically based on MDR1/P-gp overexpression. To follow up this have a higher chance of achieving remission. To determine the study and to study further the incidence of P-gp positivity prognostic importance of MDR1/P-gp status and reinduction and the effect of P-gp reversal with CSA on outcome in de rate, a randomized trial, in which the target plasma concentra- novo childhood AML, a randomized phase III study (POG tion of the chemotherapeutic agents is matched in the group 9421) was recently completed at POG institutions. When given the reversal agent and the control, is needed.
they become available, results from this study may help to In summary, the results of this study indicate that levels determine the prognostic significance of MDR1/P-gp ex- of CSA capable of reversing MDR1/P-gp in vitro are pression and the effect of modulation by CSA in children achievable in children with acceptable toxicity. Mucositis San Antonio Military Pediatric Center & Blood Disorders Center 1. Ebb DH, Weinstein HJ: Diagnosis and treatment of childhood 17. Cheson BD, Cassileth PA, Head DR, et al: Report of the acute myelogenous leukemia. Pediatr Clin North Am 44:847-862, 1997 National Cancer Institute-sponsored workshop on definitions of diag- 2. Creutzig U, Ritter J, Schellong G: Identification of two risk nosis and response in acute myeloid leukemia. J Clin Oncol 8:813-819, groups in childhood acute myelogenous leukemia after therapy inten- sification in study AML-BFM-83 as compared with study AML-BFM- 18. Arceci RJ, Stieglitz K, Bras J, et al: Monoclonal antibody to an external epitope of the human MDR 1 P-glycoprotein. Cancer Res 3. Ravindranath Y, Steuber CP, Krischer J, et al: High-dose cytar- abine for intensification of early therapy of childhood acute myeloid 19. Leith CP, Chen IM, Kopecky KJ, et al: Correlation of multidrug leukemia: A Pediatric Oncology Group study. J Clin Oncol 9:572-580, resistance (MDR 1) protein expression with functional dye/drug efflux in acute myeloid leukemia by multiparameter flow cytometry: Identi- 4. Wells RJ, Woods W, Lamphin B, et al: Impact of high-dose fication of discordant MDRϪ/effluxϩ and MDRϩ/effluxϪ cases.
cytarabine and asparaginase intensification on childhood acute myeloid leukemia: A report from the Children’s Cancer Group. J Clin Oncol 20. Leith CP, Kopecky KJ, Godwin J, et al: Acute myeloid leukemia in the elderly: Assessment of multidrug resistance (MDR 1) and 5. Hann IM, Stevens RF, Goldstone AH, et al: Randomized com- cytogenetics distinguishes biologic subgroups with remarkably distinct parison of DAT versus ADE as induction chemotherapy in children and responses to standard chemotherapy: A Southwest Oncology Group younger adults with acute myeloid leukemia: Results of the Medical Research Council’s 10th AML trial (MRC AML10). Adult and Child- 21. Ivy SP, Olshefski RS, Taylor BJ, et al: Correlation of P- hood Leukaemia Working Parties of the Medical Research Council.
glycoprotein expression and function in childhood acute leukemia: A Children’s Cancer Group study. Blood 88:309-318, 1996 6. Woods WG, Kobrinsky N, Buckley JD, et al: Timed-sequential 22. Pearson L, Leith CP, Duncan MH, et al: Multidrug resistance-1 induction therapy improves postremission outcome in acute myeloid (MDR 1) expression and functional dye/drug efflux is highly correlated leukemia: A report from the Children’s Cancer Group. Blood 87:4979- with the t(8;21) chromosomal translocation in pediatric acute myeloid 7. Stahnke K, Boos J, Bender-Gotze C, et al: Duration of first 23. Ino T, Miyazaki H, Isogai M, et al: Expression of P-glycoprotein remission predicts remission rates and long-term survival in children in de novo acute myelogenous leukemia at initial diagnosis: Results of with relapsed acute myelogenous leukemia. Leukemia 12:1534-1538, molecular and functional assays, and correlation with treatment out- 8. Marie JP, Zittoun R, Sikic BI: Multidrug resistance (MDR1) gene 24. Noonan KE, Beck C, Holzmayer TA, et al: Quantitative analysis expression in adult acute leukemia: Correlation with treatment outcome of MDR1 (multidrug resistance) gene in human tumors by polymerase and in vitro drug sensitivity. Blood 78:586-592, 1991 chain reaction. Proc Natl Acad Sci U S A 87:7160-7164, 1990 9. Pirker R, Wallner J, Geissler K, et al: MDR 1 gene expression and 25. Mechetner EB, Roninson IB: Efficient inhibition of P-glycop- treatment outcome in acute myeloid leukemia. J Natl Cancer Inst rotein mediated multidrug resistance with a monoclonal antibody. Proc 10. Willman CL: Immunophenotyping and cytogenetics in older 26. Broxterman HJ, Sonneveld P, Feller N, et al: Quality control of adults with acute myeloid leukemia: Significance of expression of the multidrug resistance assays in adult acute leukemia: Correlation be- multidrug resistance gene-1 (MDR 1). Leukemia 1:S33-S35, 1996 tween assays for P-glycoprotein expression and activity. Blood 87: 11. Litman T, Zeuthern T, Skovsgaard T, et al: Competitive, 27. Broxterman HJ, Lankelma J, Pinedo HM, et al: Theoretical and non-competitive and cooperative interactions between substrates of practical considerations for the measurement of P-glycoprotein func- P-glycoprotein as measured by its ATPase activity. Biochim Biophys tion in acute myeloid leukemia. Leukemia 11:1110-1118, 1997 28. Cornwell MM, Pastan I, Gottesman MM: Certain calcium 12. Fisher GA, Sikic BI: Clinical studies with modulators of channel blockers bind specifically to multidrug-resistant human KB multidrug resistance. Hematol Oncol Clin North Am 9:363-382, 1995 carcinoma membrane vesicles and inhibit drug binding to P-glycopro- 13. Beck WT, Grogan TM, Willman CL, et al: Methods to detect P-glycoprotein-associated multidrug resistance in patients’ tumors: 29. Hammond JR, Johnstone RM, Gros P: Enhanced efflux of Consensus recommendations. Cancer Res 56:3010-3020, 1996 [3H]vinblastine from Chinese hamster ovary cells transfected with a 14. List AF, Kopecky KJ, Willman CL, et al: Benefit of cyclospor- full-length complementary DNA clone for the MDR1 gene. Cancer Res ine (CsA) modulation of anthracycline resistance in high-risk AML: A Southwest Oncology Group (SWOG) study. Blood 92:1998 (suppl 1, 30. Arceci RJ, Stieglitz K, Bierer BE: Immunosuppressants FK506 and rapamycin function as reversal agents of the multidrug resistance 15. Advani R, Saba HI, Tallman MS, et al: Treatment of refractory and relapsed acute myelogenous leukemia with combination chemo- 31. Mantel N, Haenszel S: Statistical aspects of the analysis of data therapy plus the multidrug resistance modulator PSC 833 (Valspodar).
from retrospective studies of disease. J Natl Cancer Inst 22:719-748, 16. Ravindranath Y, Yeager AM, Chang MN, et al: Autologous 32. Lehmann EL: Statistical methods based on ranks, in Nonpara- bone marrow transplantation versus intensive consolidation chemother- metrics. San Francisco, CA, Holden-Day, 1975, p 316 apy for acute myeloid leukemia in childhood: Pediatric Oncology 33. Barnard DR, Kalousek DK, Wiersma SR, et al: Morphologic, immunologic, and cytogenetic classification of acute myeloid leukemia and myelodysplastic syndrome in childhood: A report from the Chil- 36. Sievers EL, Smith FO, Woods WG, et al: Cell surface expres- dren’s Cancer Group. Leukemia 10:5-12, 1996 sion of the multidrug resistance P-glycoprotein (P-170) as detected by 34. Lacayo NL, Lum B, Johnson S, et al: Pharmacokinetics of etoposide monoclonal antibody MRK-16 in pediatric acute myeloid leukemia and mitoxantrone in a controlled trial of cyclosporine (CSA) as an MDR fails to define a poor prognostic group: A report from the Children’s modulator in acute myeloid leukemia (AML) from the Pediatric Oncology Group (POG). Proc Am Soc Clin Oncol 16:514, 1997 (abstr 1850) 37. Beketic-Oreskovic L, Duran GE, Chen G, et al: Decreased 35. Ho AD, Lipp T, Ehninger G, et al: Combination of mitoxantrone mutation rate for doxorubicin resistance and suppression of MDR1 and etoposide in refractory acute myelogenous leukemia—an active gene activation by PSC 833. J Natl Cancer Inst 87:1593-1602, and well-tolerated regimen. J Clin Oncol 6:213-217, 1988

Source: http://www.med.stanford.edu/labs/branimir_sikic/PDF/10%20Dahl%20JCO%202000.pdf