EurJ Cancer, I’ol. 29.4, No. I, pp. 57-60. l’nnred 1.1Gre,u Hnrum 1993 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 0964~1947~93%5.00 i 0.w p 1992 Perpmm Press Ltd Treatment of Melphalan-resistant Multiple Myeloma with Vincristine, BCNU, Doxorubicin, and High-dose Dexamethasone (VBAD) Joan Bladk, Jestis San Miguel, Miguel Angel Sanz-Sanz, Antonio Alcal& Jose M. Hernhndez, Miguel Martinez, Javier Garcia-Conde, Jestis Moro, Fernando Ortega, Montserrat Fontanillas, Ciril Rozman and Jordi Estap6 A total of 65 patients (35 male/30 female) with multiple myeloma primarily (33) or secondarily (32) resistant to melphalan and prednisone were treated with vincristine, carmustine (BCNU), doxorubicin, and high-dose dexamethasone (VBAD) at 4-week intervals. Among 60 evaluable patients the overall response was 36.6% (21.6% objective response plus 15% improvements). The response rate was significantly higher in primarily resistant patients than in those becoming resistant after a prior response (48.4 vs. 24.1%, P < 0.05). The median duration of response was 17.5 months. When survival of responders and non-responders were compared by the conventional method, a highly significant difference was observed (P < 0.001). However, using the Mantel and Byar procedure and the landmark method, only a trend for longer survival in the responders was registered. These results indicate that although VBAD is effective in at least one third of patients with advanced multiple myeloma resistant to melphalan, its impact on survival is limited. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHG EurJ Cancer, Vol. 29A, No. 1, pp. 57-60,1993. were entered into the study. The diagnosis of MM was estabINTRODUCTION zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA multiple myeloma (MM) who either fail to respond or become refractory to initial alkylating treatment usually have a low response rate to subsequent chemotherapy and a short survival [l-3]. Based on the observations that carmustine (BCNU) and doxorubicin seem to have a synergistic effect in myeloma [4], and that the addition of vincristine and prednisone also appear to have a positive effect on response rate and/or survival [S, 61, the Southwest Oncology Study Group, in a large cooperative study, employed a combination of vincristine, BCNU, doxorubicin and prednisone (VBAP) in refractory myeloma with encouraging results [7]. In this regard, we obtained similar results in a series of 33 patients from a single institution with myeloma resistant to alkylating agents [8]. On the other hand, high-dose glucocorticoids, particularly dexamethasone, seem to have higher antitumour activity in refractory myeloma than prednisone at standard doses [9, lo]. Taking into account these facts, we designed a regimen combining vincristine, BCNU, doxorubicin and high-dose dexamethasone (VBAD). We report the results obtained in 65 patients with melphalanresistant MM treated with such a regimen by the Spanish Cooperative Group PETHEMA (Program for the Study and Therapy of Haematological Malignancies, Spanish Society of Haematology). PATIENTS WITH PATIENTS AND METHODS Patients’ characteristics From January 1985 to March 1991, 65 patients with MM refractory to melphalan from 16 institutions from PETHEMA Correspondence to J. BladC at the Postgraduate School of Haematology, Hospital Clinic i Provincial, Villarroel 170,08036 Barcelona, Spain. The authors are members of PETHEMA, Spanish Cooperative Group for Hematological Malignancies Treatment, Spanish Society of Haematology at the Hospital Clinic, Barcelona, Spain. Revised 11 May 1992; accepted 20 May 1992. lished according to the criteria of the Chronic Leukaemia Myeloma Task Force (111. Previous treatment consisted of intermittent courses of oral melphalan 9 mg/mZ and prednisone 60 mgim’ orally or intramuscularly for 4 days, repeated at 4week intervals [ 121. Resistance to initial treatment was defined as any of the following: (a) progressive disease after at least 4 courses; (b) lack of response after 8 courses of adequate therapy (primary resistance) or (c) progression of myeloma during therapy following an initial response (secondary resistance). 33 patients showed primary resistance, whereas the remaining 32 became resistant after having achieved a response with melphalan prednisone (MP) therapy. Patients with symptomatic cardiopathy, severe arrhythmia or gastroduodenal ulcus were excluded. Table 1 summarises the patients’ characteristics imediately before the start of VBAD therapy. Treatment schedule A cycle of treatment consisted of vincristine 1 mg intravenously, BCNU and doxorubicin 30 mg/m2 each intravenously on day 1, and dexamethasone 25 mg/m2 on days l-4,9-12, and 17-20 in the first course and on days l-4 in subsequent courses. Cycles were repeated at 4-week intervals. Responding patients continued to receive treatment until disease progression was observed. When an accumulative dose of doxorubicin of 540 mg/m2 was reached, this drug was discontinued and therapy was restricted to vincristine, BCNU and dexamethasone. The mean number of VBAD or VBD cycles per patient was nine (range: l-30). All patients received antacid treatment with cimetidine or ranitidine and antibiotic prophylaxis with trimethroprim-sulphametoxazole at least for the first 5 weeks of therapy, when higher doses of dexarnethasone were administered. J. Blade et zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIH al. 58 Table 1. Patients characteristics immediately before Table 2. Response to VBAD VBAD therapy No. of patients Males/females Age [mean (SD)] years Myeloma protein type IgG IgA Light chain only Non-secretory Clinical stage IA IIA IIIA IIIB LDH > 500 U/l 65 35130 63.9 (7.04) 37 18 9 3 13 40 9 20% (lO/SO) B,-microglobulin > 6 mg/l Response to MP Good (with secondary resistance) Primary resistant Duration of MP chemotherapy [Mean (SD)] months No. of MP courses [mean (SD)] according to the prior response to M P (60 patients) 29% (9/31) 32 33 16.2 (10.9) 11.8 (7.04) LDH =Lactate dehydrogenase; MP = melphalan’prednisone. Criteria of response An objective response was defined as a reduction of 50% or more of the M-component, improvement in performance status by at least two grades, and a decrease of more than 50% in measured cross-sectional area of plasmocytomas [ 111. Furthermore, the size and number of lytic bone lesions must not have increased and also there must have been correction of hypercalcaemia (< 11.5 mgldl), anaemia (> 9 gldl) and hypoalbuminaemia (> 3 gldl). In those patients fulfilling the above criteria but with a reduction in M-component of less than 50%, a partial response (or improvement) was considered. When the criteria for objective or partial response were not found, the case was considered a failure. Finally, patients who died within 2 months from start of therapy were considered as early deaths. Response to VBAD Prior response to MP Objective Partial Failure 8 7 16 5 2 22 Primarily resistant (n = 31) Secondarily resistant (n = 29) P = 0.045 38 patients showed no response. Table 2 shows the correlation between response to initial MP treatment and response to VBAD. As it can be observed, the response rate was 24.1% (7/29) in patients with secondary resistance to MP and 48.4% (15/31) among patients who were primarily resistant (P = 0.045). The median duration of response was 17.5 months (Fig. 1). zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIH Toxicity The treatment was generally administered on an outpatient basis and was clinically well tolerated. Side-effects of VBAD were graded according to the WHO scale [ 171. Although grade 2 or more myelotoxicity was recorded in 26 patients (40%), severe granulocytopenia (< 5 zyxwvutsrqponmlkjihgfedcbaZYXWVUT x 109/1) and thrombocytopenia (< 25 x 109/1)were only observed in 4 and 1 patients, respectively. There were nine episodes of severe infection requiring hospitalisation during VBAD therapy. 5 patients developed heart failure after receiving a total doxorubicin dose ranging from 120 to 450 mg/m2. In all of them doxorubicin was discontinued. 4 patients had grade 2 peripheral neuropathy attributed to vincristine. In one patient who developed an steroid psycosis, dexamethasone was discontinued. No treatment-related deaths were observed. Survival The overall median survival of the series was 13 months, with 47 patients having died at the time of this analysis. The median survival for the reponders was 30 months vs. 9.8 months for Statistical methods The Fisher’s exact test was used to assess the statistical significance of comparison of response rates. Survival times were estimated from the start of the VBAD therapy to the date of death or last follow-up. Survival curves were plotted according to the method of Kaplan and Meier [ 131 and statistically compared by means of the log-rank test [ 141. Finally, the influence of response on survival was analysed by the Mantel and Byar procedure and the landmark method [ 15, 161. RESULTS Among the 65 patients entered into the study, 60 were evaluable for response. 5 patients were considered non-evaluable for the following reasons: major protocol violation (3 cases), early death (1 patient) and lost to follow-up (1 patient). 0 Response to treatment lb 2b 3b onths (“1 40 50 60 The overall favourable response rate was 36.6% (22 out of 60 Fig. 1. Response durationb 22 responding patients. patients; 13 objective responses and 9 improvements), whereas zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Treatment of Melphalan-resistant Multiple Myeloma with VBAD 59 from high-dose therapy with haematopoietic stem cell grafting or haematopoietic growth factors [19, 201, this approaches can be applied to a minority of patients only. Second line treatments 1.55 m for refractory multiple myeloma produce disappointing results 7 P <O.OOl 2 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 0.6. i, due to both the low response rate to salvage therapy and the ;i zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA --i short duration of clinical response [l-3]. With the combination of vincristine, BCNU, doxorubicin and prednisone (VBAP), response rates of about 25%, as well as survival prolongation for responding patients, have been reported [7, 81. In addition, high-dose corticosteroids, particularly dexamethasone, have clearly shown antitumour activity in refractory myeloma. Thus, Alexanian er al. [lo] reported a response rate of 27 and 21% in primary refractory and relapsing patients, respectively, with Fig. 2. Survival of responders (thick line) and non-responders (thin dexamethasone alone. Furthermore, according to Buzaid and line) to VBAD therapy (P < 0.001). zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Durie [2], the expected response to high-dose glucocorticoids in resistant myeloma is about 25%. With this background, we employed the VBAP regimen with high-dose dexamethasone instead of prednisone. However, taking into account both the considerable toxicity of high-dose glucocorticoids in pretreated myeloma patients and that response to dexamethasone-containing regimens usually occurs rapidly [21-233, dexamethasone was given for 4-day courses, beginning on days 1, 9 and 17 in the first cycle and only for days 1 to 4 in subsequent cycles in our study, in order to prevent severe steroid toxicity. As VBAD regimen constituted the rescue therapy of the MP arm in a cooperative randomised study of the PETHEMA group IO 20 30 40 io ii0 0 [12], all of our patients had been homogeneously treated and more than one third of them responded to VBAD. It is of note Months that the response rate was significantly higher in primary than Fig. 3. Survival of the overall series (thick line) as compared with in secondary resistant patients (48.4 vs. 24.2%, P < 0.05). This that of non-responding patients (thin line) by the Mantel and Byar fact may be due to the emergence of chemotherapy-resistant test (x2: 2.85, P = 0.091). clones in patients heavily pretreated with melphalan. An alternative explanation would be the usefulness of dexamethasone in non-responders (Fig. 2). This difference was highly significant patients unresponsive to initial therapy [2, IO]. In fact, in when analysed by means of the standard method, considering previous studies (7, 81 the response rate of primary resistant the response as an initial variable (P < 0.001). However, when myeloma to VBAP was only 25 and 7%. The median duration the Mantel and Byar procedure [ 151 (Fig. 3) and the landmark of response to salvage regimens, including VAD and high-dose method [16] (Fig. 4) were applied, only a trend in favour of melphalan, usually ranges between 6 and 9 months in refractory responders was observed. Disease progression and infection MM [3, 10, 21-25J. In the present series, the median duration were the main causes of death. of response is longer than in the above reports. This might be explained in part by the fact that our patients had received only DISCUSSION MI’ as previous therapy. The combination of melphalan and prednisone constitutes As in the previous series with VBAP [7, 81, VBAD was well the standard treatment for most patients with MM. However, tolerated. There were no treatment-related deaths, the most about 40% of patients do not respond to MP and, moreover, all frequent toxicity being moderate myelosuppresssion due to patients initially responding eventually become resistant. The BCNU and doxorubicin. In contrast with other studies in which use of other alkylating agents is precluded by cross-resistance dexamethasone was employed at higher doses [21-231, in the [18]. On the other hand, although selected patients can benefit present series no excessive dexamethasone toxicity was registered. Indeed, some studies have reported a decrease in severe dexamethasone-related toxicity with reductions in the initial doses [21, 231. The method more frequently used to analyse the impact of response on survival consists of separating patients into two groups according to whether or not they have achieved a response [ 131, the survival curves being compared by different statistical methods, such as the log-rank test [14]. In our series, this comparison showed a highly significant difference between responders and non-responders. However, this method is mis60 20 30 40 50 0 10 leading, since it introduces a bias in favour of responders (i.e. the time from the start of treatment necessary to detect the Months response, that in myeloma usually varies over a wide range). In Fig. 4. Survival of responders vs. non-responders by the landmark addition, responders may survive longer than non-responders method with landmark at 8 months after starting VBAD (x2: 3.57, P = 0.057). for other reasons than the treatment itself (i.e. prognostic Observed 1 I(221 33(381 Expected 22.7 21.3 O/E 0.48 60 J. Blade zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGF et al. 15. Mantel N, Byar DP. Evaluation of response time data involving factors), [16, 26, 271. To reliably assess the impact of treatment transient states: an illustration using heart transplant data. 3 Am results on survival, the Mantel and Byar procedure and the StarAssoc19?4,69,81-86. landmark method are currently recommended [ 15, 161. Using 16. Anderson JR, Cain KC, Gelber RD. Analysis of survival by tumor these methods, no significant differences between responders response.3 Clin O ncoll983,1,7lO - 719. 17. World Health Organization. Handbook for Reporting Resulrs of and non-responders were observed in our series, although both Cancer Treatment. WHO Offset Publication No. 48. Geneva, WHO, methods showed a trend in favour of responders. In this sense, 1979. a lack of correlation between response and survival has recently 18. Blade J, Fellu E, Rozman C, Estape J, Milla A, Montserrat E. been reported in previously untreated myeloma patients [28, Cross-resistance to alkylating agents in multiple myeloma. Cancer 291. Because of the poor correlation between response and 1983,52,786789. 19. Barlogie B, Jaganath S, Dixon DO, et al. High-dose melphalan and survival in myeloma, it has been suggested that the magnitude granulocyte-macrophage colony-stimulating factor for refractory of response, as currently defined, may not adequately assess the multiple myeloma. Blood 1990,76,677- 680. efficacy of current therapy [28-301. 20. Gahrton G, Tura S, Ljungman P, et al. Allogeneic bone marrow In conclusion, VBAD is a well-tolerated and effective salvage transplantation in multiple myeloma. N Eng13 Med 1991, 325, 1267-1273. regimen for at least one-third of patients with refractory mye21. Barlogie B, Smith L, Alexanian R. Effective treatment of advanced loma, particularly those primarily resistant to melphalan. This zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA multiple myeloma refractory to alkylating agents. N Engl3 M ed regimen can be useful in alkylating-resistant patients who are 1984,310,1353-1356. not candidates for experimental treatments, such as high-dose 22. Monconduit M, Loet le X, Bernard JF, Michaux JL. Combination therapy followed by bone marrow stem cell grafting and/or chemotherapy with vincristine, doxorubicin, dexamethasone for refractory or relapsing multiple myeloma. Br3 Haemarol 1986,63, haematopoietic growth factors [19, 201. 599- 601. 23. Friedenberg WR, Kyle RA, Knospe WH, Bennett JM, Tsiatis AA, Oken MM. High-dose dexamathasone for refractory or relapsing multiple myeloma. Am3 Hemaroll991,36, 171-175. 24. Barlogie B, Hall R, Zander A, Dicke K, Alexanian R. Highdose melphalan with autologous bone marrow transplantation for multiple myeloma. Blood 1986,67, 1298-1301. _ 25. Selby P, McElwain TJ, Nandi AC, et al. Multiple myeloma treated with high-dose intravenous melphalan. Br3 Haemarol 1987, 66, 1. Kvle RA. Greioo PR. Germ MA. Treatment of refractorv multiole myeloma and considerations for future therapy. Simin drool 1986, 13,32&333. 2. Buzaid AC, Durie, BGM. Management of refractory myeloma: a review.gClin Oncol1988,6,889-905. 3. Alexanian R, Barlogie B, Ventura G. Chemotherapy for resistant 55- 62. and relapsing multiple myeloma. Eur 3 Haematol 1989 43, 26. Rozman C, Montserrat E. Critical factors in new therapeutic (suppl. 51), 140-144. aauroaches in chronic lvmnhocvtic leukaemia (CLL). Nouv Rev Fr ~ 4. Alberts DS, Durie BGM, Salmon SE. DoxorubiciniB.C.N.U. I% marol1988,30,45~55. . chemotherapv for multiple in relapse. Lancer 1976, 1, _ mveloma _ 27. Rozman C, Montserrat E, Rozman M, Blade J, Cervantes C. 926-928. -. Analisis de la respuesta al tratamiento. Consideraciones sobre las 5. Alexanian R, Salmon S, Bonnet J, Gehan E, Ham A, Weick J. variables pronosticas evolutivas (tiempo-dependientes). M ed Clin Combination therapy for multiple myeloma. Cancer 1977, 40, (Barr) 1989,92,52- 55. 2765-2771. 28. Palmer M, Belch A, Brox L, Pollock E, Kock M. Are the current 6. Cocnwell III GG, Paiak TF, Kochwa S, er al. Vincristine and criteria for response useful in the management of multiple myeloma? prednisone prolong the survival of patients receiving intravenous J Clin Oncoll987,5, 1373-1377. or oral melphalan for multiple myeloma: Cancer and Leukemia 29. Marmont F, Levis A, Falda M, Resegotti L. Lack of correlation Group B experience.3 Clin Oncoll988,6,1481-1490. between objective response and death rate in multiple myeloma 7. Bonnet J, Alexanian R, Salmon S, et al. Vincristine, BCNU, patients treated with oral melphalan and prednisone. Ann Oncol Doxorubicin, and Prednisone (VBAP) combination in the treatment 1991,2,191-195. of relapsing or resistant multiple myeloma: a Southwest Oncology 30. Durie BGM. Is the magnitude of initial response predictive for Group-Study. Cancer TreatRt$19%2,66,1267-1271. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA survival in multiple myeloma? Ann Oncoll991,2,166. 8. Blade J, Rozman C, Montserrat E, et al. Treatment of alkylating resistant myeloma with vincristine, BCNU, Doxorubicin and Prednisone (VBAP). Eur3 Cancer Clin O ncoll986,22,1193- 1197. Acknowledgements-The following institutions participated in this study by reporting patients to the PETHEMA registry: Hospital Clinic, 9. Alexanian R. Yan BS. Bodev GP. Prednisone oulse theraov _- for Barcelona (C. Rozman, J. Balde); Hospital General Rio Carri6n, Palencia refractory myeloma. Blood 1983,62,572-577. * (M.A. Sam-Sanz); Residencia Capimn Corms, Jaen (A. Alcala, M.L. 10. Alexanian R, Barlogie B, Dixon D. High-dose glucocorticoids treatment of resistant myeloma. Ann Intern Med 1986,105,8-l 1. Escudero); Hospital Clinico Universitario, Salarnanca (J.F. San Miguel); 11. Chronic Leukemia-Myeloma Task Force. Proposed guidelines for Hospital General, Segovia (J.M. Hernandez, M. Martinez, J. Bascones); protocol studies. II. Plasma cell myeloma. Cancer Ckenzotker Rep Hospital Clinic0 Universitario, Valencia (J. Garcia-Conde, A. Pascual); 1973,4,145-158. Residencia Virgen Blanca, Leon(M.J. Moro); Residensia Nuestra S 12. Blade J,San Miguel J, Alcala A, et al. A randomized multicentric Sonsoles, Avila (F. Ortega, J.M. Hernsindez); Clinica Alianza, Barcelona study combination chemotherapy (C. Besses); Hospital Universitario, Valladolid (L. Guerras, J. Fernandcomparing alternating (VCMPNBAP) and melphalan-prednisone in multiple myeloma. ezCalvo); Residencia Virgen Vega, Salamanca (R. Jimenez-Galindo); Blur 1990,60,31%322. Residencia Seguridad Social, Soria (M.V. Faura); Hospital Verge de la Cinta. Tortosa (Ll. Font): Residencia Son Dureta. Palma de Mallorca 13. Kaplan GL. Meier P. Non-uarametric stimation from incomolete ob&vations.JAmSrarAsso~ 1958,53,457-481. (J. Besalduch); ‘Hospital’de Sagunt, Sagunt (R. Colon&as); Hospital Z(0 - EIZ/E de1 Rio Hortega (J. Tortosa). 14. Peto R. Pike MC. Conservation of the aooroximation .. in the log-rank test for survival data or tumour incidence data. This work has been supported in part by the Spanish Grant FISss Biometrics 1973,29,57%584. 9110608.