MDR-1 Expression and Response to Vincristine, Doxorubicin, and Dexamethasone Chemotherapy in Multiple Myeloma Refractory to Alkylating Agents By J.J. Cornelissen, P. Sonneveld, M. Schoester, H.G.P. Raaijmakers, H.K. Nieuwenhuis, A.W. Dekker, and H.M. Lokhorst Purpose: To assess whether the presence of enhanced multiple rug resistance (MDR)- 1 gene expression in multiple myeloma (MM) patients predicts survival, as well as response to vincristine, doxorubicin, and dexamethasane (VAD) chemotherapy. Patients and Methods: Sixty-three MM patients refractory to alkylating therapy were studied. The presence of the MDR-1 gene product, a 170-kd glycoprotein (P-170), was analyzed in bone marrow plasma cells by means of the alkaline phosphatase (APAAP) technique using the P-170-specific monoclonal antibody (MoAb) C219. The prognostic value of MDR-1 gene expression, examined before VAD treatment, was compared with other established prognostic factors including p 2 -micro-

globulin, albumin, lactate dehydrogenase (LDH), and the plasma cell labeling index. Results: Fifty-nine percent of all samples were P- 170-

M

OST PATIENTS with multiple myeloma (MM) are initially treated with chemotherapy based on a combination of alkylating agents, such as melphalan and prednisone. Patients who become refractory may subsequently be treated with the combination of a continuous infusion of vincristine and doxorubicin plus dexamethasone (VAD), which induces response rates in 50% to 75% of melphalan-resistant patients.' However, response duration following VAD is generally short, and most patients relapse within 1 year. The failure of MM patients to respond to VAD chemotherapy may be associated with expression of the multidrug resistance phenotype.2 Multidrug resistance (MDR) is caused by increased expression of a 170-kd membrane glycoprotein (P-170) encoded by the MDR-1 gene.3 P-170 acts as an drug efflux pump that prevents the intracellular accumulation of various nonrelated drugs, such as doxorubicin and vincristine. Resistance to VAD has been reported to correlate with the extent of P-170 expression.4 Clinical resistance may be modulated by the use of noncytotoxic agents, such as verapamil or cyclosporine, which inhibit the function of P-170.25,6 However, it is still unknown whether P-170 expression is of prognostic value in MM patients who are to be treated with VAD chemotherapy. We examined whether the presence of MDR-1 gene expression in 63 patients refractory to first-line treatment with alkylating agents would predict their response to VAD, as well as their survival.

positive. No association could be demonstrated between response to VAD and MDR-1 gene expression (X2 P = .359), in contrast to high serum 0 2 -microglobulin levels, which were positively correlated with response (P = .006). P-170-positive and -negative patients showed a median survival duration of 23 and 22 months, respectively, a difference that was not statistically significant (P = .9). 32 -microglobulin, LDH, albumin, and the plasma cell labeling index were all significantly correlated with survival. Conclusion: These results indicate that other mechanisms of resistance must be involved in MM apart from MDR. The role of MDR status at this stage of disease may be biased by the major contribution of dexamethasone to induction of response by VAD in MM patients. J Clin Oncol 12:115-119. O 1994 by American Society of Clinical Oncology.

PATIENTS AND METHODS The study was performed on available frozen cytocentrifuge slides prepared from Ficoll-Hypaque-purified bone marrow aspirates obtained from MM patients before VAD treatment.

Patients Sixty-three patients presenting between 1986 and 1991 were studied. Patients were treated with the VAD regimen at the departments of Haematology of the University Hospital Utrecht or the University Hospital Rotterdam Dijkzigt. The median age was 58 years (range, 29 to 79). Fifty-one patients had stage Ila MM, nine stage IlIb, and three stage II (Table 1).7 Performance status was determined

according to the criteria of the Eastern Cooperative Oncology Group (ECOG).8 Serum 62-microglobulin level was determined by means

of a competitive enzyme immunoassay (Phadezym; Pharmacia, Uppsala, Sweden). Serum levels of lactate dehydrogenase (LDH) and albumin were measured according to standard methods. The kinetic

state of the plasma cells was measured as described previously, by determining the plasma cell labeling index.' All 63 patients received VAD following first-line treatment, including 57 who were refractory to alkylating agents and six who were refractory to both alkylating agents and doxorubicin. Twenty-

From the Department of Haematology, University Hospital Utrecht, Utrecht; and University Hospital Rotterdam Dijkzigt, Rotterdam, the Netherlands. Submitted March 16, 1993; accepted September 16, 1993. Address reprint requests to J.J. Cornelissen, MD, Dr Daniel den Hoed Clinic,Departmentof Hematology, Groene Hilledijk301, 3075 EA Rotterdam, the Netherlands. © 1994 by American Society of Clinical Oncology. 0732-183X/94/1201-0019$3.00/0

Journal of Clinical Oncology, Vol 12, No 1 (January), 1994: pp 115-119

Downloaded from ascopubs.org by 37.44.207.161 on January 18, 2017 from 037.044.207.161 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

115

CORNELISSEN ET AL

116 Table 1. Patient Characteristics 63

No. of patients Age, years Median Range Sex (male/female) Immunoglobulin G Immunoglobulin A Immunoglobulin D Light-chain disease Stage Illa

58 29-77 36/27 38 15 1 9

II Performance status (ECOG) 0 1 2 3 4 Previous response Primarily refractory Relapsed Resistance To alkylating agents To anthracyclines and alkylating agents

51 9 3 23 33 5 2 0 22 41 57 6

two patients had never previously responded to chemotherapy and 41 were in relapse after a previous response (Table 1). The VAD regimen consisted of vincristine (0.4 mg/d) and doxorubicin (9 mg/ m2 ), both administered by continuous infusion for 4 days, dexamethasone administered orally at a dose of 40 mg/d on days 1 to 4, 9 to 12, and 17 to 20. Cycles were repeated every 4 weeks. The second and third dexamethasone pulse was omitted every second VAD session.

Response to VAD treatment was evaluated by monthly serial measurement of serum and urine myeloma proteins. Marrow aspirates were performed every 3 months. A complete response was defined as a minimal 75% reduction of serum myeloma protein or the disappearance of Bence-Jones proteins in the case of light-chain disease. A partial response was defined as a 50% to 75% reduction in myeloma proteins in serum. VAD treatment was continued until no further tumor regression could be observed, but with a minimum of six courses in partial and complete responders. Kaplan-Meier survival curves were calculated from the start of treatment and Wil0 coxon-Gehan statistics were used to compare survival curves.' Statistical analysis to compare response rates in subgroups of patients was performed with the X' test. MDR-1 expression was analyzed on bone marrow cytocentrifuge slides using the P-170-specific monoclonal antibody (MoAb) C219 (Centocor, Malvern, PA). Bone marrow cells were separated by Ficoll-Hypaque and washed twice in phosphate-buffered saline. Cytocentrifuge slides were prepared and air-dried for 18 hours. Afterwards, the slides were fixed in methanol/acetone (1/1), soaked in Tris-buffered saline for 15 minutes, and incubated with 10% heatinactivated rabbit serum and 1% goat serum for 30 minutes. Undiluted C219 MoAb or idiotype-matched control (immunoglobulin G2a [IgG2a], Sigma, St Louis, MO) were added and incubated overnight at 4'C. Antibody to mouse immunoglobulin was added to

each slide. After washing, the cells were incubated with alkaline phosphatase substrate (APAAP) for 60 minutes at 37 0 C, then washed three times. Only cells with membrane staining were scored positive. As described previously," a sample was considered positive if at least 30% of the plasma cells showed membrane staining and if the isotype-matched control staining was indeed negative. All samples were independently scored by two investigators. To exclude MDR3 encoded P-glycoprotein staining by C219, the staining reaction was also performed using the MDR-l-specific MoAbs JSB1 (a gift Dr R. Schepers, Amsterdam, the Netherlands) in 23 randomly chosen patients and C494 (Centocor) in 21 patients. The patient samples were compared with simultaneously stained control Chinese hamster ovary cell lines (a drug-sensitive MDR-1-negative line; American Tissue Type Culture, Rockville, MD) and a drug-resistant MDR-1positive line derived from this line resistant to 100 pg/L of colchicine.

RESULTS The overall response rate to VAD was 51% (26 of 63), including seven complete responses and 25 partial responses (Table 2). Response duration was short, with a median disease-free survival duration of 10 months and

an overall median survival duration of 21 months. In the group of patients (n = 22) without a previous response, 10 patients responded and two toxic deaths were observed. Twenty-two patients (54%) responded in the group of patients (n = 41) who had relapsed after a response to first-line treatment. The median survival duration in primary refractory patients was 22 months, and in relapsing patients, 23 months. With a cutoff point of 30% positive plasma cells, 37 of 63 samples studied were P-glycoprotein-positive (59%).

Forty-four samples were also stained with a positive control MoAb (either JSB 1 or C494), and these staining reactions yielded the same results. In the group of patients who had not responded to previous therapy (n = 22), 14 samples were MDR- -positive. In the group of patients (n = 41) who relapsed after a previous response, 23 samples were MDR-1-positive. No association could be demonstrated between response to VAD treatment and enhanced MDR-1 expression: 17 responses in the group of MDR-l-positive patients, as opposed to 15 responses in the group of MDR-1 -negative patients (x 2 P = .359) (Table 2). The subgroups of MDR-I -positive and MDR-

Table 2. Response of VAD Chemotherapy and MDR Status No. of Patients Complete and Partial Response

Stable Disease and Progressive Disease

Toxic Death

Positive Negative

17 15

18 11

2

37 26

Total

32

29

2

63

MDR Status

Downloaded from ascopubs.org by 37.44.207.161 on January 18, 2017 from 037.044.207.161 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

Total No.

117

PROGNOSTIC VALUE OF MDR IN MULTIPLE MYELOMA '

'

12

24

N

I

0

72

54 Months

1.00

075

0.50

025

12

24

N

a

Wo

72

84

Months

Fig 1. (A) KapIlan-Meier survival curves and (B)relapse-free survival curves for patients with refractory MM treated with VAD chemotk ernv Diffrran f nntint witk nhnnh... kh r.urve. s •e•teen MDR-1 expression and MDR-negative patients are not statistically significant.

patient sample are presented in Fig 2 together with corresponding survival, as well as response to VAD for each individual patient. No correlation could be demonstrated between survival and degree of P-170 expression. Furthermore, Kaplan-Meier survival curves for MDR-1 -positive and MDR-1-negative patients using different cutoff levels (10%, 20%, 30%, and 40%) showed similar results and did not yield significant differences between such subgroups of patients (results not shown). In addition, the significance of recognized prognostic factors was tested by univariate analysis using previously defined cutoff values 9' 12 -14 (Table 3). /2-microglobulin, LDH, albumin, and the plasma cell labeling index all proved to be of significant prognostic value. Using these predefined cutoff values, subgroups of patients with a poor median survival (12 to 17 months) could be identi-

0 o

100-

1-negative patients were well matched with respect to other established prognostic factors, including $ 2-microglobulin, LDH, and the plasma cell labeling index. However, low serum albumin level was not equally distributed, with two patients in the MDR-1-positive group (two of 37, 5%) and three in the group of MDR-I -negative patients (three of 26, 12%). However, the difference of distribution was not statistically significant (X2 P = .367) and survival analysis without these five patients did not result in a difference of Kaplan-Meier survival curves for MDR-1-positive and MDR-1-negative patients. The prognostic variables 62-microglobulin, LDH, albumin, and the plasma cell labeling index were analyzed separately for a possible correlation with response to VAD. Of these, Q2 -microglobulin (cutoff value, 6 pg/mL) positively correlated with response to VAD (P = .006). Subsequently, the prognostic value of enhanced MDR1 expression with respect to survival and disease-free

"

survival was examined. Kaplan-Meler survival and disease-free survival curves for MDR-1 -positive and MDR1-negative patients are presented in Fig 1A and B. MDR1-positive patients showed a median survival duration of 23 months, whereas the median survival duration for MDR-1-negative patients was 22 months. The survival difference was not statistically significant (P = .9). The durations of response were 10 and 6 months, respectively, for MDR-1-positive and MDR-1-negative patients, which was also not a statistically significant difference (P = .35). It was additionally examined whether different cutoff levels for MDR-1 positivity would result in a different prognostic value of MDR-1 gene expression. The exact percentages of P-170-positive plasma cells of each

0 OS 0

0

0o

so

o

•

0

0 0 o 0

C In C S

50-

13.1

C-

0

o

0

0

3

o 0

0-

6

12

o a

0

24

S

3

4h t Months

60b

72

e4

Fig 2. Percentage P-glycoprotein expression and survival. Scatter diagram shows the percentage of P-glycoprotein-positive plasma cells of 63 patient samples and the corresponding survival and response to VAD chemotherapy for each patient. No correlation could be demonstrated. (0) Responding patients, (e) patients with progressive or stable disease.

Downloaded from ascopubs.org by 37.44.207.161 on January 18, 2017 from 037.044.207.161 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

118

CORNELISSEN ET AL Table 3. Correlation oF Prognostic Factors and Survival Variable

MDR (%) 62-microglobulin (pg/mL) LDH (U/L) Albumin (g/L) Labeling index (%)

Cutoff Value

- 30 < 30 -6 < 6 - 300 < 300 s 30 > 30 3 < 3

Median Survival (months)

No. of Patients

23 22 18 24 18 29 10 23 17 29

37 26 22 41 25 37 5 56 23 31

P

.937 .026 .040 .025 .029

fled. Of these, a serum albumin level less than 30 g/L identified a subgroup of patients with the shortest median survival duration of 10 months. DISCUSSION The prognosis for MM patients who are primarily resistant or who have become refractory to first-line treatment with alkylating agents has improved little since the introduction of combination chemotherapy.' Although the VAD regimen produces high response rates (50% to 75%) in these patients, response duration is short and most patients relapse within 1 year. ",' The limited benefit from VAD may be explained by the emergence of MDR. In vitro studies have shown a clear-cut association between enhanced P-glycoprotein expression and intracellular concentration of, for example, doxorubicin and level of drug resistance.'16 17 Enhanced MDR- 1 expression in VAD refractory patients has been demonstrated, 4 and the addition of MDR-reversing agents may restore responsiveness in a subgroup of VAD-resistant patients.2 ',5,6 These studies have suggested that MDR-1 expression has a serious negative effect on clinical response and subsequent survival in MM. In contrast, in the present study, it is demonstrated that endogenous MDR-1 expression does not lead to a reduced response to VAD. MDR-1 expression in these patients also does not predict survival, in contrast to other well-recognized prognostic factors, such as high serum /32 -microglobulin or LDH and low albumin levels, which significantly correlated with a poor survival duration, using predefined cutoff values. These paradoxical results may be explained, in part, by previous clinical studies showing that dexamethasone accounted for most of the responses to VAD in refractory, as well as untreated patients.' "9 P-170 is a membrane protein that acts as a drug efflux pump transporting a number of unrelated drugs, including doxorubicin and vincristine, which are part of the VAD regimen. However, clinical resistance to corticosteroid therapy does not seem

to be caused by MDR-1 expression, and in vitro studies have suggested that resistance to glucocorticoid inhibition of myeloma cell growth is mediated by postreceptor mechanisms. 20 Our results suggest that resistance to VAD may also be explained by other mechanisms of resistance, such as corticosteroid resistance. In addition, apart from MDR due to the expression of P-glycoprotein, a number of other mechanisms of MDR have been reported, which could theoretically play a role in resistance to VAD. 21 Furthermore, studies that indicate a negative effect of MDR-I on response to VAD need to be interpreted with caution. Epstein et al4 demonstrated that a majority of VAD-resistant patients expressed P-glycoprotein on their myeloma cells. However, only a minority of patient samples were analyzed before VAD therapy, leaving the question whether P-glycoprotein expression was present before VAD or acquired during VAD. The latter possibility cannot be ignored, especially in view of findings by Grogan et al, 22 who demonstrated a correlation between Pglycoprotein expression in myeloma and administration of vincristine and doxorubicin. In the present study, a high percentage of patients proved to be MDR-positive before VAD therapy, and most of these patients were pretreated with alkylating agents, which indicates that Pglycoprotein expression is not exclusively related to (prior) administration of vincristine and doxorubicin. This finding was also observed in recent studies in patients with acute myeloid leukemia and monoclonal gammopathy of undetermined significance (MGUS), in which it was demonstrated that MDR-1 expression not induced by MDR-1-related drugs is frequently observed."' 23 The clinical relevance of endogenous MDR-I expression in MM needs to be further established, as it may be possible that MM cells with endogenous P-170 expression are further selected during initial treatment. The clinical studies using chemosensitizers and chemotherapy have demonstrated that a subgroup of VAD-resistant patients may show a secondary response to VAD after the addition of either verapamil or cyclosporine.2,5,6,24 The studies using verapamil also show that other mechanisms of resistance must play a role, given that not all drugresistant patients are MDR-positive and only a minority of patients (10% to 23%) appear to benefit from the combination of VAD and verapamil. Recently, Sonneveld et al6 reported results obtained in refractory MM patients treated with the combination of VAD and cyclosporine. A high response rate of 47% (seven of 15 patients) was observed in VAD-refractory patients. The beneficial effect of cyclosporine is most likely explained by its inhibitory effect on P-170 expression on plasma cells, but cyclosporine may also affect the pharmacokinetics of cy-

Downloaded from ascopubs.org by 37.44.207.161 on January 18, 2017 from 037.044.207.161 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.

119

PROGNOSTIC VALUE OF MDR IN MULTIPLE MYELOMA 25 26 totoxic drugs, leading to increased systemic exposure. ' In addition, it was shown recently that dexamethasone is transported by P-glycoprotein27 and that cyclosporine may modulate P-170-mediated corticosteroid efflux, as was demonstrated by Van Kalken et a128 in MDR-1positive hamster lung cells in vitro. In conclusion, in contrast to the prognostic value of recognized parameters such as f 2 -microglobulin, LDH, albumin, and the plasma cell labeling index, no association could be demonstrated between enhanced

MDR-1 expression and survival in refractory MM patients treated with VAD. Furthermore, response, as well as duration of response, was also not correlated with enhanced expression of P-glycoprotein on plasma cells of these patients. These results indicate that other mechanisms of resistance must be involved. Corticosteroid resistance may be important in view of prior studies indicating the significant contributions of dexamethasone by VAD in MM patients at an early stage of the disease.1 6

REFERENCES 1. Barlogie B, Smith L, Alexanian R: Effective treatment of advanced multiple myeloma refractory to alkylating agents. N Engl J Med 310:1353-1358, 1984 2. Salmon SE, Dalton WS, Grogan TM, et al: Multi-drug resistant myeloma: Laboratory and clinical effects of verapamil as a chemosensitizer. Blood 79:44-50, 1990 3. Pastan I, Gottesmann M: Multi-drug resistance in human cancer. N Engl J Med 316:1388-1393, 1987 4. Epstein J, Xiao H, Koba B: P-glycoprotein expression in plasma-cell myeloma is associated with resistance to VAD. Blood 74:913-917, 1989 5. Trumper LH, Ho AD, Wulf G, et al: Addition of verapamil to overcome drug resistance in multiple myeloma: Preliminary clinical observation in 10 patients. J Clin Oncol 7:1578-1579, 1989 6. Sonneveld P, Durie BGM, Lokhorst HM, et al: Modulation of multi-drug resistant multiple myeloma by cyclosporin. Lancet 340:255-259, 1992 7. Durie BGM, Salmon SE: A clinical staging system for multiple myeloma. Correlation of measured myeloma cell mass with presenting clinical features, response to treatment and survival. Cancer 36:842-852, 1975 8. Zubrod CG, Scheiderman M, Frei E, et al: Cancer appraisal of methods for the study of chemotherapy of cancer in man. J Chron Dis 11:7-23, 1960 9. Lokhorst HM, Boom SE, Bast EJEG, et al: Determination of the plasma cell labeling index with bromodeoxyuridine in a double fluorescence technique. Br J Haematol 64:271-275, 1986 10. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163167, 1966 11. Sonneveld P, Durie BGM, Lokhorst HM, et al: Analysis of multidrug-resistance (MDR-1) glycoprotein and CD56 expression to seperate monoclonal gammopathy from multiple myeloma. Br J Haematol 83:63-67, 1993 12. Durie BGM, Stock-Novack D, Salmon SE, et al: Prognostic value of pretreatment serum P2-microglobulin in myeloma: A Southwest Oncology Group study. Blood 75:823-830, 1990 13. Dimopoulos MA, Barlogie B, Smith TL, et al: High serum lactate dehydrogenase level as a marker for drug resistance and short survival in multiple myeloma. Ann Intern Med 115:931-935, 1991 14. Bataille R, Boccadoro M, Klein B, et al: C-reactive protein and lz-microglobulin produce a simple and powerful myeloma staging system. Blood 80:733-737, 1992 15. Lokhorst HM, Meuwissen OJAT, Bast EJEG, et al: VAD

chemotherapy for refractory multiple myeloma. Br J Haematol 71:25-30, 1989 16. Dalton WS, Grogan TM, Rybski JA, et al: Immunohistochemical detection and quantification of P-glycoprotein in multiple drugresistant human myeloma cells: Association with level of drug resistance and drug accumulation. Blood 73:747-752, 1989 17. Dalton WS, Durie BGM, Alberts DS, et al: Characterization of a new drug resistant human myeloma cell line which expresses P-glycoprotein. Cancer Res 46:5125-5130, 1986 18. Alexanian R, Dimopoulos MA, Delasalle K, et al: Primary dexamethasone treatment of multiple myeloma. Blood 80:887-890, 1992 19. Alexanian R, Barlogie B, Dixon D: High-dose glucocorticoid treatment for resistant myeloma. Ann Intern Med 105:8-17, 1986 20. Gomi M, Moriwaki K, Katagiri S, et al: Glucocorticoid effects on myeloma cells in culture: Correlation of growth inhibition with induction of glucocorticoid recepter messenger RNA. Cancer Res 50:1873-1878, 1990 21. Vendrik CPJ, Bergers JJ, De Jong WH, et al: Resistance to cytostatic drugs at the cellular level. Cancer Chemother Pharmacol 29:413-429, 1992 22. Grogan ThM, Spier CM, Salmon SE, et al: P-glycoprotein expression in human plasma cell myeloma: correlation with prior chemotherapy. Blood 81:490-495, 1993 23. Campos L, Guyotat D, Archimbaud E, et al: Clinical significance of multidrug resistance P-glycoprotein expression on acute nonlymphoblastic leukemia cells at diagnosis. Blood 79:473-476, 1992 24. Gore ME, Selby PJ, Millar B, et al: The use of verapamil to overcome drug resistance in myeloma. Proc Am Soc Clin Oncol 7:228-233, 1988 (abstr) 25. Yahanda AM, Adler KM, Fisher GA, et al: Phase I trial of etoposide with cyclosporine as a modulator of multidrug resistance. J Clin Oncol 10:1624-1634, 1992 26. Lum BL, Kaubisch S, Yahanda AM, et al: Alteration of etoposide pharmacokinetics and pharmacodynamics by cyclosporine in a phase I trial to modulate multidrug resistance. J Clin Oncol 10:16351642, 1992 27. Ueda K, Okamura N, Hirai M, et al: Human P-glycoprotein transports cortisol, aldosterone, and dexamethasone, but not progesterone. J Biol Chem 267:24248-24252, 1992 28. Van Kalken CK, Broxterman HJ, Pinedo HM, et al: Cortisol is transported by the multidrug resistance gene product P-glycoprotein. Br J Cancer 67:284-289, 1993

Downloaded from ascopubs.org by 37.44.207.161 on January 18, 2017 from 037.044.207.161 Copyright © 2017 American Society of Clinical Oncology. All rights reserved.