Document YrjwgdO03N6X7EkkLrBG98Q80

Chemoimmunotherapy for Multiple Myeloma RAYMOND ALEXANIAN, MD,' SYDNEY SALMON, MD,t JORDAN GUTTERMAN, MD,' DENNIS DIXON, PHD,* JOHN BONNET, MD,$ AND ARTHUR HAUT, MDS The effect of chemoimmunotherapy consolidation treatment using alternating courses of alkylating agents and BCG was studied in 105 responding patients with multiple myeloma. The survival time of these patients was similar to that of responding patients treated on previous maintenance programs, indicating no apparent value from BCG for myeloma. The duration of unmaintained remission was langest in those with low numbers of residual plasma cells, and relapse usually developed within one year in patients with persistent serum myeloma peaks. Disease recontrol was achieved in 50%of relapsing patients whose median survival from retreatment was 20 months. Patient follow-up without chemotherapy until relapse was justified mainly for those responding patients with disappearance of myeloma proteins. Cancer 47:1923-1929. 1981. TH E M E D I A N S U R V I V A L time for patients with multiple myeloma who have responded to chemotherapy is about two years longer than that of unresponsive patients, a prolongation in life span that has been accounted for by the remission duration.' Survival was not improved in patients who received melphalanprednisone courses indefinitely in comparison with patients who had been followed without therapy until relapse.' These similar results were attributed to the high frequency of second remissions in patients relapsing on no treatment in that study. Repeated azathioprine-prednisone courses given on the possibility that an increased percentage of dividing cells may be destroyed by an antimetabolite also failed to confer any survival benefit.g Between 1975 and 1978, the Southwest Oncology Group (SWOG) evaluated the utility of a chemoimmunotherapy program that combined a melphalanprednisone combination with Bacillus Calmette-Guerin (BCG) vaccine in responding patients. BCG consolida- For the Southwest Oncology Group. From the X:Universityof Texas M. D. Anderson Hospital and Tumor Institute, Houston, Texas, the tUniversity of Arizona, Tucson, Arizona, the $Scott and White Clinic, Temple. Texas. and the %University of Arkansas Medical Center, Little Rock, Arkansas. Supported by Grant nos. CA-03195, CA-03389, CA-03392, CA04915 , C A-04919. CA-04920, CA- 10187. CA-12014, CA-I 2213, CA-12644, CA-13238. CA-16943. and CA-19980 from the National Cancer Institute. Address for reprints: Raymond Alexanian, MD, The University of Texas M . D. Anderson Hospital, Houston. T X 77030. The authors thank Kay Delasalle and Robert Dreicer for many portions of this analysis. Accepted for publication April 28, 1980. tion, in combination with chemotherapy, had been associated with survival prolongation in melanoma!' and acute leukemia"' but had not been evaluated in myeloma. This report indicates that no survival benefits were observed by adding BCG to chemotherapy for consolidation therapy in patients with myeloma. After the completion of treatment, the duration of unmaintained remission was longest in those patients with very low numbers of plasma cells. Retreatment with a melphalan-prednisone combination during relapse was followed by a median survival time of 20 months despite the occurrence of second remissions in only one half of the patients. Methods This report presents an analysis of 289 consecutive patients with multiple myeloma whose treatment began at 12 SWOG institutions between October 1974 and February 1977. Only patients previously untreated for myeloma were included in the study. The diagnosis was based on criteria defined by the Chronic LeukemiaMultiple Myeloma task force that included bone marrow plasmacytosis and a myeloma globulin peak on serum or urine electrophoresis in symptomatic patients'; seven patients without detectable peaks (2%) on immunoelectrophoresis were also included when marked bone marrow plasmacytosis, lytic bone lesions, and marked depression of normal serum immunoglobulins were present. The pretreatment tumor mass was rated as high, intermediate, or low in accordance with a clinical staging system described previously.4.n 0008-543X/81/0415/1923 $0.90 b American Cancer Society 1923 1924 CANCEARpril 15 1981 Vol. 41 T ~ B L1E. Response and Survival from Different Drug Combinations in Multiple Myeloma No. treated Early death Inadequate report No. evaluable No. responsive Response rates 9% Evaluable % All Median survival (months) Courses every 3 weeks C+A+P Cyclophosphamide, 100 mg/m2, p.o./d x 4 Adriamycin, 25 mdm'. i.v., day 1 Prednisone. 60 mg/m', p.o./d x 4 V+C+A+P Vincristine, 1.0 mg, i.v.. day 1 Cyclophosphamide. 100 mg/m2, p.o./d X 4 Adriamycin, 25 mg/m', i.v., day 1 Prednisone. 60 mg/m2, p.o./d x 4 V+M+C+P Vincristine, 1.0 mg i.v., day I Melphalan. 5 mg/m', p.o./d x 4 Cyclophosphamide, 100 mg/m2, p.o./d x 4 Prednisone, 60 mpim', p.o./d x 4 V+B+A+P Vincristine. 1.0 mg, i.v., day I BCNU, 25 mg/m2. i.v., day 1 Adriamycin, 25 mg/m'. i.v., day 1 Prednisone, 60 mg/m2, p.o./d x 4 57 87 94 51 2 3 1 3 I 1 6 0 54 26 48 46 32 83 56 67 64 32 87 52 60 55 30 48 31 65 61 32 Tretitmrtit Regitnetis Retri issioti Mcr inteti ci 11 C P All patients were assigned at random to a combina- Between April 1975 and January 1978, 105 eligible tion of either cyclophosphamide-Adriamycin-pred- responding patients who had received at least six nisone (CAP), vincristine-cyclophosphamide-Adria- months of drug treatment on any of the four induction mycin-prednisone (VCAP), or vincristine-melphalan- regimens described above were assigned to a chemo- cyclophosphamide-prednisone (VMCP) in the dose immunotherapy program consisting of alternating regimens described in Table 1. After a lower response cycles of VMCP and BCG for 12 more months (Table rate to the CAP combination was found, a vin- 2). Three patients were ineligible for this chemoim- cnstine-BCNU (Bis-chlorethy1nitrosourea)-Adriamycin- munotherapy program since they were randomized to prednisone regimen (VBAP) was substituted in the receive treatment later than five months following their randomization scheme. All patients received doses eligibility. Of the 105 eligible responding patients who adjusted to produce moderate degrees of bone marrow were registered, 19 had been induced into remission toxicity ( i . e . , a nadir in granulocytes between 1000 with CAP, 32 with VCAP, 34 with VMCP, and 20 and 2500/mm3 or in platelets between 50,000 and with VBAP. Because the survival times for responders 100,OOO/mms) between courses that were repeated in the different groups were similar, all eligible patients at three-week intervals. Clinical response was based registered on VMCP-BCG maintenance were com- on criteria previously described and based on a 75% bined in the analysis. The consolidation treatment reduction of myeloma protein production rate." consisted of one VMCP course given every three Randomize for 6 months of treatment TABLE2. Schema of Treatment ~ ~ ~___ Still in remission after Responding patients 12 months of BCG Relapse VMCP VMCP + BCG for 12 months-No treatment until relapse ___j Resume VMCP A = Adriamycin; B = BCNU; C = Cytoxan; M = melphalan; P = prednisone: V = vincristine. No. 8 CHEMOIMMUNOTHEFRORAPMYULTIPLME YELOMA. Alexaiiian et (11. 1925 to four weeks in doses previously described (Table 1); 2 x loxattenuated Pasteur-strain BCG organisms were given by scarification 8 and 15 days after the start of each VMCP course. Treatment continued for 12 months, after which no chemotherapy was given to those still in remission until there was evidence of relapse (Table 2). When relapse developed, the VMCP combination was reinstituted and continued until the patient relapsed or died. Relapse was defined by the earliest of either an increase in myeloma protein production to more than 25% of the pretreatment value, a doubling of myeloma protein production, the reappearance of myeloma proteins that had disappeared, or unequivocally progressive lytic bone lesions. The frequency of second reductions in tumor mass was defined in 20 patients with evaluable data at the M. D. Anderson Hospital; results were compared with those in 13 patients relapsing after unmaintained remission periods that followed only one year of initial chemotherapy in a previous study.2 The survival of responding patients who received BCG was compared with those maintained on monthly courses of either a melphalan-cyclophosphamideBCNU-prednisone combination (MCBP) or an azathioprine-prednisone combination (IP) with MCBP reinforcement in a preceding SWOG study.3All eligible and registered patients were included in the survival calculations that used the life table method from the start of both the initial and the maintenance chemotherapy until death." Delrryetl Hypersetisitivity Studies Delayed hypersensitivity was evaluated from the magnitude of skin induration to test antigens in 42 previously untreated patients with multiple myeloma older than 40 who were studied at the M. D. Anderson Hospital. Patients were selected when an outpatient evaluation over several days was possible, when no corticosteroid or radiotherapy was necessary, and when the same technician was available to quantify all results. Five patients were included who had received local radiotherapy more than four months previously. The antigens used were dermatophytin, streptococcal varidase, candida, mumps, PPD, and KLH using preparations, doses, and techniques of injection described previously.'" The maximum induration as an average of two right angle measurements was recorded in millimeters 24 and 48 hours following the intracutaneous injection of test substances; the maximum response was compared with those from a control group of healthy volunteers over the age of 40 studied TABLE3. Comparability of Patient Groups on Different Maintenance Treatments AzathioVMCP-BCG prine* MCBP* No. patients Age, yr. (median) Sex (amale) Race (5% black) Pretreatment abnormalities (% of total) Hemoglobin ( ~ 8 . g5fl00 ml) Serum calcium (>11.5 mgil00 mi) BUN (>40 mg/100 ml) Tumor mass grade 7% High %' Intermediate 5% Low 105 61 64 18 11 15 8 37 40 23 33 33 61 58 58 58 18 21 12 21 18 9 39 30 43 46 21 24 30 * Includes all patients maintained on these regimens in a previous study.2 at the same institution."j Multiple assessments of the skin test reaction were conducted in 32 myeloma patients among whom were 20 responsive and 12 unresponsive patients. No chemotherapy had been given for at least four weeks before any follow-up skin test. Results Remission Maintetiarzce The response rates for patients who received one of three initial treatment combinations that included vincristine ranged from 60-67931 in comparison with 48% for CAP treatment (Table 1). Further details on the effects on these treatments have been described previously.:' The survival time was evaluated in 105 responding patients who received a VMCP-BCG chemoimmunotherapy combination. Results were compared with those of 66 responding patients from the same institutions maintained until relapse on either an azathioprine-prednisone or an MCBP combination after the first six months.3 Table 3 confirms the similarity of the three patient groups in those major factors known to affect prognosis. Bacillus Calmette-Guerin produced mild to moderate local skin reactions at the scarification site in all patients: treatment was terminated in three patients because of generalized reactions such as diffuse papular eruption (two patients) or recurrence of active tuberculosis (one patient). As indicated in Figure 1, the median survival for patients receiving BCG was 38 months, a duration about five months longer than that from the other treatments. 1926 CANCERApril 15 1981 V O l . 47 VMCP + /BCG 30 - I I I I \I 10 20 30 40 50 Survival from Treatment (Mod FIG.1 . Survival time from start of remission maintenance treatment for 105 responding patients who received a VMCP-BCG chemoimmunotherapy regimen for 12 months. Results were compared with those in 66 responding patients maintained previously until relapse on either azathioprine-prednisone or an MCBP combination. This superiority was not significant ( P > 0.2), and the longevity from all programs was also similar to that for 64 responding patients followed without treatment or continued on melphalan-prednisone in a previous SWOG study.2 Unmuintuined Remissions Of the 105 patients who received the VMCP-BCG program, 30 relapsed or died during the 12-month TABL4E. Correlation of Unmaintained Remission Time with Myeloma Cell Mass No. patients Unmaintained remission duration (median mo.) All evaluable patients* Pretreatment tumor mass High and intermediate All patients Serum peak persists Serum peak disappears Bence Jones protein dis- appears Low All patients Serum peak persists Serum peak disappears Bence Jones protein dis- appears 69 56 33 12 11 13 2 6 5 9 8 6 11 10 14 6 21 12 ~~~~~ * Six additional patients never had a myeloma protein, and their unmaintained remission times could not be measured. period of remission maintenance, and patients with a persistent serum peak constituted 24 of these patients. Thirty-four of the 40 patients with disappearance of myeloma proteins on cellulose acetate electrophoresis completed the 12-month chemoimmunotherapy program, including 16 with disappearance of only Bence Jones protein as evaluated by electrophoresis of urine concentrates (Table 4). Follow-up data were available to evaluate the duration of unmaintained remission in 69 patients, of whom 14 were still alive in remission, four died in remission of unrelated disease, and 51 relapsed from myeloma. Six patients without an abnormal protein at diagnosis were excluded because their unmaintained remission time could not be measured precisely. The median duration of unmaintained remission for all patients was nine months, and longer periods of disease stability were found mainly in patients with disappearance of myeloma protein (Table 4). Withholding chemotherapy in patients with a persistent peak was usually followed by rising levels of the monoclonal component within 12 months. Reinduction Therapy When chemotherapy was resumed in 20 relapsing patients with rising myeloma proteins treated at the M. D. Anderson Hospital, a greater than 50% reduction of myeloma cell mass occurred in ten. Thus, second partial remissions developed in 50% of the relapsing patients in this study, in comparison with 77% of those relapsing in a previous study,' but this difference was not significant ( P > 0.2). Patients who relapsed after six months of unmaintained remission had a higher frequency of tumor recontrol (6/8) than patients who relapsed within the first six months of no treatment follow-up (4/12). The frequency of disease recontrol was similar in all patient groups in Table 4, regardless of the pretreatment tumor mass or the myeloma protein status. The median survival from the resumption of chemotherapy was 20 months, a duration similar to that of 15 months for patients retreated after relapsing with only one year of initial chemotherapy.z Figure 2 demonstrates the prolonged tumor mass stability and survival time in two patients rated as unresponsive to retreatment with VMCP. Skin Tests The median diameter of induration from each of the antigens evaluated in 42 patients before chemotherapy was virtually identical to those found in normal subjects of similar age (Table 5 ) . No correlation was noted between the intensity of the reaction to any of the antigens tested and the pretreatment tumor mass, the likelihood of response, or the survival time. The effects of chemotherapy were evaluated on the skin test re- No. 8 CHEMOIMMUNOTHERAPY FOR MULTIPLEMYELOMA . Alexaniun et a / . 1927 action in 32 patients with follow-up studies. No consistent correlations were apparent between the followup skin test reactions and either response or survival. When it developed, myeloma protein relapse was not associated with any consistent changes in skin test reactions. Responsive patients who received BCG usually improved their response to PPD; 12 of 15 patients so treated tripled their mean diameter of induration to more than 15 mm, a change that never occurred in any unresponsive patient not given BCG. Discussion Recent SWOG studies have shown that 60% of patients with multiple myeloma achieved remissions as defined by a 75% or greater reduction in myeloma protein production rate.3 In these trials, the response rate was about 15% higher and the median survival time about five months longer when vincristine was included in the drug combination, and the treatments were repeated at three-week intervals. This report indicates that a similar high response rate resulted from a combination of vincristine-BCNU-Adriamycinprednisone (VBAP), a regimen that also produced 50% reductions in tumor load in about 30% of relapsing patients .5 A large number of responding patients received a 12-month consolidation treatment using a VMCPBCG combination to enhance patient immunity. The VMCP drug combination was chosen because this treatment had produced the highest response rate in a previous SWOG study.3 Chemoimmunotherapy using BCG has been associated with the prolongation of survival and remission in patients with malignant melanoma,y acute leukemia," and advanced breast cancer.6 Cyclophosphamide may depress B-cell proliferation, and its administration with BCG vaccine may produce increased numbers of T cells with antitumor activity.15*18Consecutive responding patients were given the VMCP-BCG combination, and their survival was compared with almost identical patient groups in an immediately preceding study.3 The survival of patients receiving the VMCP-BCG program was not significantly superior to that of responders Resume -Ut 1 No Rx VMCP + BCG 20 40 I+BCG, I 20 40 Months of Treatment FIG.2. rumor mass curves in two typical patients relapsing from unmaintained remissions. While the resumption of VMCP did not achieve a 50% reduction in cell mass, and the patients were considered unresponsive, prolonged tumor mass stability continued for more than two years. maintained on previous SWOG treatments or on no treatment. Thus, these results did not support a useful role for BCG immunotherapy in patients with myeloma, although further evaluations may reveal a small percentage of patients with very long survival. These negative results may have been because of the preserved cell-mediated immunity in patients with multiple myeloma, so that any further enhancement with BCG, as occurred in patients with acute leukemia or melan ~ m a , ~w. ~aOs less likely to develop. On the other hand, several groups have reported a survival benefit for BCG in patients with non-Hodgkin's lymphoma,11.L3 another B-cell neoplasm derived from cells similar to plasma cells in the differentiation pathway.I7 For patients with nodular lymphoma, the benefit from BCG was observed only when this immunostimulant was added during induction and not during maintenance.lo This suggested that a similar strategy may be useful in myeloma patients. TABLE5 . Pretreatment Skin Test Reactions to Various Antigens (Median Values for Mean Diameter of Induration) No. patients Dermatophytin Varidase Candida Mumps PPD KLH Normal volunteers* (40-80 years old) Multiple myeloma All patients High tumor mass Responsive Three year survival * Reference 16. t Data not available. 131 42 12 21 21$ 8 7 16 18 I 10 10 18 20 5 9 10 21 24 5 1 1 10 18 19 7 13 12 18 20 7 $ lncludes 17 responsive and four unresponsive patients. t 5 4 5 6 1928 CANCERApril 15 1981 Vol. 47 In comparison with normal subjects of comparable frequency of recontrol may be intermediate between age, myeloma patients showed no impairment of de- these figures (namely about 60%), one wonders layed hypersensitivity as measured by their reaction whether the longer duration of initial therapy provided to a battery of skin tests. These findings emphasized more effective control of the sensitive tumor subclone the importance of making comparisons with an older and the more likely emergence of subclones resistant age control population. In this first study reported of to the drugs given.I2 In fact, the persistence of any serial measurement of skin tests in myeloma, no cor- myeloma protein at the end of the remission main- relations were noted between the intensity of the re- tenance phase must have reflected the presence of action to recall antigens and the likelihood of aresponse resistant cells even before the unmaintained remission to treatment or the survival time. This contrasts with phase began. Thus, resistance to individual drugs similar studies in acute leukemia, breast cancer, and may develop early after the start of treatment but melanoma in which the reaction to some skin tests may be expressed slowly in accordance with the growth appeared to predict the response to chemotherapy.'~~9~1o"f resistant cells. In that case, the value of any specific Pretreatment and serial skin testing with the antigens maintenance treatment has less relevance than the pre- available may have provided an insensitive technique vention of tumor resistance to the drugs now available. for assessing immunity in patients with multiple Treatment regimens that include all effective drugs myeloma. from the start, such as in current SWOG alternating The duration of unmaintained remission, as defined and sequential drug combinations, may clarify this by the duration of low and constant myeloma protein question. values, was evaluated in 69 patients who had com- The increasing tumor resistance with time is con- pleted the chemoimmunotherapy program with at least sistent with previous analyses indicating that resistant 18 months of total therapy. The unmaintained remis- cells constitute most of the myeloma cell population sion time of nine months for all patients was similar after two years of treatment.'." One would then expect to that found in patients followed on no treatment after tumor recontrol in almost all patients relapsing on no only 12 months of initial treatment.2 Early relapse treatment after a short period of initial treatment, usually occurred after withholding treatment in all pa- and drug resistance after several years of chemother- tients with persistent serum peaks; chemotherapy apy. Perhaps, the "ideal patient" to follow without should probably be continued in such patients to treatment is one in whom the serum myeloma protein prevent disease morbidity. The disappearance of disappeared early so that follow-up without therapy myeloma protein that was compatible with a marked could commence between 6- 12 months. While these tumor reduction was usually associated with a long patients might be assured of second remissions with unmaintained remission time, and this was most ap- retreatment for relapse, the unmaintained remission parent in patients with a low pretreatment tumor time might be short because of the high number of mass in whom serum peaks had disappeared. Myeloma residual plasma cells.2Conversely, longer durations of relapse was undoubtedly developing in most patients consolidation treatment may be followed by longer before the reappearance of myeloma proteins on the unmaintained remissions, but the frequency of tumor electrophoretic strip, but only rarely did any disease recontrol would be less as noted here. With either morbidity develop before chemotherapy was resumed. approach, a good quality survival of at least one year In view of the long duration of disease stability in most from the resumption of treatment should be antici- of these patients and the lack of curative treatment, pated, since either second remissions or a prolonged it seems reasonable to withhold treatment until electro- stable mass will ensue in most patients. phoretic or radiographic evidence of relapse becomes apparent. Previous analyses had suggested that a subclone of REFERENCES cells that was still sensitive to alkylating agent-prednisone treatment usually accounted for the initial relapse in patients who had responded and were then followed without chemotherapy." Thus, while second remissions developed in about three fourths of the patients 1. Alexanian R, Bergsagel D, Migliore P. Vaughn W, Howe CD. Melphalan therapy for plasma cell myeloma. Blood 1968; 3 1: 1- 10. 2. Alexanian R, Gehan E, Haut A, Saiki J. Weick J. Unrnaintained remissions in multiple myeloma. Bloud 1978; 51:1005- 101I . 3. Alexanian R, Salmon S, Bonnet J, Gehan E, Haut A, Weick J . Combination therapy for multiple myeloma. C U ) I L . C19~77: 40: who had relapsed after 12 months of initial therapy,' recontrol occurred in only one half of the patients who received 18 months of treatment. While the true 2765-2771. 4. Alexanian R. Balcerzak S , Bonnet J, ct d.Prognostic factors in multiple myeloma. Crrticrr 1975: 36: 1192- 1201. 5. Bonnet J, Alexanian R, Salmon S. Vincristine-BCNU- No. 8 CHEMOIMMUNOlHERAPY FOR MULTIPLEMYELOMA . Alexmiutz et (I/. 1929 Adriamycin-prednisone combination for treatment of melphalanand cytoxan-resistant multiple myeloma (abstr). Atti Soc Hr/?irito/ 1976; 122. 6 . Buzdar A, Gutterman J, Blumenschein G, el ti/. Intensive postoperative chemoimmunotherapy for patients with stage I1 and 111 breast cancer. Cutrcr,r 1978: 41: 1064- 1075. 7. Chronic Leukemia-Multiple Myeloma Task Force: Proposed guidelines for protocol studies. Ctiricrr Choiiother Rep 1968; 1: 13-39. 8. Durie B, Salmon E. Clinical staging system for multiple myeloma. Cuticrr 1975: 36:842-854. 9. Gutterman J , Mavligit G, Hersh E. Adjuvant immunotherapy of recurrent malignant melanoma with systemic BCG. Luticet 1973; 1: 1208- 1212. 10. Hersh E, Gutterman J , Mavligit G. et u / . Serial studies of immunocompetence of patients undergoing chemotherapy for acute leukemia. J C/i/i Itivest 1974: 54:401-408. 11. Hoerni B . Durand M , Richard P, ef ul. Successful maintenance immunotherapy by BCG of non-Hodgkin's malignant lymphomas. Br J Huetirutd 1979; 42:507-514. 12. Hokanson J , Brown B, Thompson J , Drewinko B, Alexanian R. Tumor growth patterns in multiple myeloma. Cliticer 1977; 39: 1077- 1084. 13. Jones S, Salmon S, Byrne G, Butler J. Improved survival in nodular lymphoma with chemoimmunotherapy (abstr). Proc ASCO.1979; 20:313. 14. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Atn Slut Assoc 1958; 534A:457-482. 15. Lagrange P, Mackaness G , Miller T. Potentiation of T-cell mediated immunity by selective suppression of antibody formation with cyclophosphamide. J Exp Med 1974; 139:1529-1539. 16. Morns DL, Hersh EM, Bartholomew P, Gutterman J, Marshall M, Mavligit G. Recall antigen delayed-type hypersensitivity skin testing in melanoma and acute leukemia patients and their associates. Crincer Kes 1979: 39:219-226. 17. Salmon S, Seligmann M. B-cell neoplasia in man. Luncei 1974; 2:1230-1233. 18. Steel G, Pierce GE. Effects of cyclophosphamide on im- munity against chemically-induced syngeneic murine sarcomas. /ti/ J Ctiticer 1974; 13:572-578.