Abstract #664
A New International Staging System for Multiple Myeloma From the International Myeloma Working Group
From: Greipp PR, San Miguel JF, Durie BG, Loiseau HA, Fonseca R, Jacobson JL, Rasmussen E, Crowley JJ. A New International Staging System for Multiple Myeloma From the International Myeloma Working Group. Blood 2003;102:190a, abstract 664. Copyright American Society of Hematology, used with permission. P. R. Greipp, J. F. San Miguel, B. G. Durie, H. A. Loiseau, R. Fonseca, J. L. Jacobson, E. Rasmussen, J. J. Crowley Hematology, Mayo Clinic, Rochester, Minnesota; Hematology, University of Salamanca, Salamanca, San Vicente, Spain; Cedars Sinai Comprehensive Cancer Center, Los Angeles, California; University of Nantes, Nantes, France; Cancer Research and Biostatistics, Seattle, Washington Introduction: In 1975, Durie and Salmon (DS) developed a staging system that has remained, for over 25 years, the gold standard for stratification of MM patients. However the system does not contain B2M, widely recognized as the most useful prognostic factor. Acceptance of staging systems utilizing B2M has been impeded by lack of agreement on the merits of other components. Similar to what has been done in other malignancies, an international cooperative effort gathered data on 11,171 cases from 17 institutions in Asia, Africa, Europe, and North America; 8,270 received standard treatment, and 2,901 received high dose therapy as initial planned treatment. Results: Median age was 60 years; 61% were IgG, 24% IgA, 10% light chain only, 3% IgD, and 2% biclonal or other. Median serum M-protein level was 3.9 g/dL, hemoglobin 10.5 g/dL, platelets 221 x 103/mcL, creatinine 1.1 mg/dL, B2M 3.8 mcg/mL, albumin 3.6 g/dL, and bone marrow plasma cell percent (PC) 40.0%. 42% had 3 or more bone lesions, 24.4% had pathologic fractures, and 33.2% had compression fractures. Univariate survival analysis. Cutoffs were chosen by CART analysis. Overall survival of the entire group was 44 months. Significant variables predicting survival in order of hazard ratio (HR) were: B2M ≥3.5 (1.78), creatinine≥2 (1.76), platelets <130,000 (1.71), age ≥65 (1.63), LDH above normal (1.52), hemoglobin <10 (1.50), albumin <3.5 (1.34), BMPC ≥33% (1.34), and CRP≥8 (1.30). Multivariate survival analysis. Regression tree branching analysis was performed on univariate factors available on all patients. Three branching models showed that B2M and albumin were independent prognostic factors for survival. The new International Staging System (ISS) was as follows: Stage I = B2M <3.5 and albumin ≥3.5; Stage II = B2M <3.5 and albumin <3.5, or B2M ≥3.5 to <5.5; Stage III = B2M ≥5.5. Of 3278 Stage II patients, 1057 (32%) were classified Stage II because of an albumin <3.5. The same survival differences were observed in the training set and a validation set comprised of 1/2 the patients. Moreover the ISS effectively stratified survival for standard or transplant therapy, for different geographic regions, individual institutions, and cooperative groups. Comparing the ISS with the Durie and Salmon (DS) classification, the former provides more powerful discrimination between staging groups and more even distribution of patients among stages. Creatinine ≥2 significantly worsened survival in a small group of 255 Stage I and II patients: Stage I, 35 vs 62 months; Stage II, 32 vs 45 months. Within stage III, a small subset of 755 patients with low platelets (<130,000/ μL) or above normal serum LDH level had a median survival of only 20 months. Conclusion: The new ISS classification for MM improves the stratification of the previous DS system and can be widely used since it is based on simple variables (B2M and albumin). Although not as robust a part of the statistical model, subclassification of stages I and II according to creatinine, and of stage III patients as high risk according to low platelets or high LDH levels can improve the discrimination and clinical usefulness of the ISS. A second staging system, based on variables such as cytogenetics will be developed for reference MM institutions. Abstract #662
Gene Expression Profiling Can Be Used To Predict EFS in Myeloma Patients Treated with High Dose Therapy and Tandem Stem Cell Transplant From: Shaughnessy J, Rasmussen E, Zhan F, Barlow W, Jacobson J, Crowley J, Barlogie B. Gene Expression Profiling Can Be Used To Predict EFS in Myeloma Patients Treated With High Dose Therapy and Tandem Stem Cell Transplant. Blood 2003;102:190a, abstract 662. Copyright American Society of Hematology, used with permission. J. Shaughnessy, E. Rasmussen, F. Zhan, W. Barlow, J. Jacobson, J. Crowley, B. Barlogie Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, Little Rock, Arkansas; Cancer Research and Biostatistics, Seattle, Washington One of the hopes for global gene expression profiling is that it will account for the tremendous variability in survival of multiple myeloma, which can range from several months to > 10 years after diagnosis. The presence of cytogenetic abnormalities (CA) and/or interphase FISH chromosome 13 deletion (FISH13) has emerged as the most powerful variables predicting for early failure of tandem auto-transplants in MM, but even these variables only account for ~15% of this outcome variability. Gene expression profiling using Affymetrix U95Av2 microarrays was performed on cRNA from CD138-enriched plasma cells from 212 newly diagnosed multiple myeloma patients registered to the Total Therapy II protocol. The median follow-up time was 20 months. There were 34 events representing either disease specific death or progression/relapse. Cox regression was used to model disease specific event-free survival (EFS) using both standard prognostic values and Affymetrix signal. Gene expression values considered were based on significance of univariate association with EFS based on Cox regression models. The median Signal call was used as a cut point prior to modeling inclusion. The 100 genes most significantly associated with EFS based on the score test were potential variables in multivariate modeling. Cox regression and stepwise selection with a 0.05 entry criteria and 0.005 significance criteria was used to model EFS using gene expression variables after adjusting for the standard prognostic variables model. A generalized estimate of R2 was obtained using the approach suggested by Cox and Snell (1989). Using standard prognostic variables only, the model which best fit disease specific event-free survival was FISH13 and CA. Adjusting for CA, patients with FISH13 had an increased risk of event (HR=2.4, 95% CI=1.1-5.4, p=0.0329). Adjusting for FISH13, patients with CA had an increased risk of event (HR=3.9, 95% CI=1.8-8.5, p=0.0007). Adjusting for CA, FISH13 and other genes in the model, three genes were significant at the 0.005 level: 98-026.1, 98-026.2, and 98-026.3. After adjustment for other model variables, patients with high 98-026.1 expression had increased risk of event (HR=25, 95% CI=3.3-195.9, p=0.0019), patients with high 98-026.2 had decreased risk of event (HR=0.06, 95% CI=0.008- 0.42,p=0.005), and patients with high 98-026.3 had decreased risk of event (HR=0.15, 95% CI=0.04-0.54, p=0.0036). After adjusting for these 3 genes and each other, FISH13 patients had an increased risk of event (HR=2.8, 95% CI=1.2-6.5, p=0.0157) but CA did not have a significant increase in risk (HR=1.09, 95% CI=0.47-2.5, p=0.8442). R2 for the FISH13 and CA model was 9.7%. With the addition of 98-026.1, 98-026.2, and 98-026.3 the estimate for R2 was 38.5%, an increase of 28.8%. This is an example of improved modeling of disease specific event-free survival in myeloma using microarray technology. Importantly, the genes identified may not only represent useful biomarkers, but may have a pathogenetic role in disease initiation or progression and therefore may be targeted with novel therapeutics. A discussion of the biological relevance of these genes will be presented. Abstract #509
Oral Melphalan(Drug information on melphalan), Prednisone(Drug information on prednisone) and Thalidomide(Drug information on thalidomide) for Newly Diagnosed Myeloma From: Palumbo A, Bertola A, Musto P, Nunzi M, De Stefano V, Callea V, Rotoli B, Petti MC, Caravita T, Lauta VM, Patti C, Bringhen S, Cavallo F, Falco P, Carella AM, Liberati AM, Boccadoro M. Oral Melphalan, Prednisone and Thalidomide for Newly Diagnosed Myeloma. Blood 2003;102:148a, abstract 509. Copyright American Society of Hematology, used with permission. A. Palumbo, A. Bertola, P. Musto, M. Nunzi, V. De Stefano, V. Callea, B. Rotoli, M. C. Petti, T. Caravita, V. M. Lauta, C. Patti, S. Bringhen, F. Cavallo, P. Falco, A. M. Carella, A. M. Liberati, M. Boccadoro Divisione Universitaria di Ematologia, Azienda Ospedaliera San Giovanni Battista, Torino, Italy; Italian Multiple Myeloma Study Group This work was supported by AIRC, AIL and MIUR Purpose: Thalidomide (THAL) is an useful drug for the treatment of refractory myeloma. Few data are available on the activity of THAL for newly diagnosed patients. No data are available on the association of THAL with the standard oral Melphalan and Prednisone (MP).To address this issue, we evaluated the potential additive and synergistic effect of the combination Melphalan, Prednisone and Thalidomide (MP-THAL). Materials and methods: Between June 2002 and June 2003, 56 patients with newly diagnosed symptomatic multiple myeloma were assigned to a treatment with 6 monthly courses of MP (melphalan 4 mg/sqm and prednisone 40 mg/sqm for 7 days every month) plus Thalidomide that was administered at 100 mg/day continuously until any sign of disease progression or relapse. The dose of THAL was reduced to 50% when grade II WHO toxicity occurred, and suspended for any grade III. At present, 31 patients (median age 72, range 61-82) have completed the 6 assigned MP courses and were evaluated. Results: After a minimum of 3 months of treatment, no one was in complete remission (according to the EBMT/IBMTR criteria), 11 patients (36%) showed a myeloma protein reduction of 75- 99%, 14 patients (45%) a response of 50-74%, and 6 (19%) a response of <50% or disease progression. After a minimum of 6 months of treatment, CR was observed in 7 patients (22%), myeloma protein reduction of 75-99% was detected in 13 patients (42%), response of 50-74% in 8 patients (26%) and response <50% or disease progression in 3 patients (10%). Response was followed by significant improvement of performance status, skeletal pain, anemia and transfusion requirement. The major acute adverse event was deep-vein thrombosis (6 patients) and pulmonary thromboembolism (1 patient). Neurological toxicity was recorded in 10 patients (5 constipations, 2 tremors, 2 tingling and numbness, 1 dizziness) . Thalidomide increased the hematological toxicity induced by MP: grade III-IV WHO neutropenia was observed in 6 patients. Infections were frequent (4 cases of pneumonia, 1 Herpes Zoster infection and 1 fever of unknown origin). Four patients required hospitalization and intravenous antibiotic therapy. One patient died for septicemia after a severe pneumonia. Ten patients discontinuated THAL (7 because of thrombo-embolic events, 2 due to constipation and 1 due to infectious disease), 50% dose- reduction was required in 4 patients (grade II neurological toxicity in 2 cases, 1 cutaneous rush and 1 infection). Conclusion: MP-THAL is a feasible and promising approach for newly diagnosed patients. Response rate is significantly higher than those previously reported with any other conventional poly-chemotherapy regimen. A 20% CR rate is quite similar to those observed after transplant. The occurrence of a significant incidence of sepsis and deep-vein thrombosis suggests the need for antibiotic and anticoagulant prophylaxis. An update of these data will be presented. Abstract #135
Comparable Survival in Multiple Myeloma With High Dose Therapy Employing MEL 140 mg/m² + TBI 12 Gy Autotransplants vs Standard Dose Therapy with VBMCP and No Benefit From Interferon Maintenance: Results of Intergroup Trial S9321 From: Barlogie B, Kyle R, Anderson K, Greipp P, Lazarus H, Jacobson J, Rasmussen E, Cromer J, Crowley J. Comparable Survival in Multiple Myeloma With High Dose Therapy Employing MEL 140 mg/m² + TBI 12 Gy Autotransplants vs Standard Dose Therapy with VBMCP and No Benefit From Interferon Maintenance: Results of Intergroup Trial S9321. Blood 2003;102:42a, abstract 135. Copyright American Society of Hematology, used with permission. B. Barlogie, R. Kyle, K. Anderson, P. Greipp, H. Lazarus, J. Jacobson, E. Rasmussen, J. Cromer, J. Crowley Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, Little Rock, Arkansas; Mayo Clinic, Rochester, Minnesota; Dana-Farber Cancer Institute, Boston, Massachusetts; Ireland Cancer Center, University Hospital of Cleveland, Cleveland, Ohio; Cancer Research and Biostatistics, Seattle, Washington; on behalf of SWOG, CALGB, ECOG ECOG, CALGB and SWOG enrolled 899 patients with newly diagnosed MM to receive VAD induction x 4 cycles (n=805), followed by randomization to a single PBSC-supported HDT (n=258) vs VBMCP (n=252) using CTX 4.5 g/m² + G-CSF for PBSC mobilization in all patients; 39 patients with HLA compatible sibling donors received allo-Tx with MEL 140 + TBI 12 Gy. Responders to VBMCP or HDT were randomized to ±IFN (n=120/119). Response rates were similar with HDT/VBMCP: CR 17% /15%, PR 93%/91%. Median PFS and OS from VAD were 22 and 46 mos. PFS was superior after HDT (25 vs 21 mos, p=.05); OS was not (58 vs 53 mos, p=.8) (Fig 1). PFS/OS from ±IFN were similar at 21/58 mos vs 19/79 mos. 52% of 161 VBMCP failures received salvage transplants (STx), effecting PR in 59% with a post-relapse survival of 30 mos vs 23 mos without STx (p=.05). PFS and OS were adversely affected by B2M > 4 mg/L and PCLI > 0.4%, regardless of treatment arm. Excessive TRM of 39% called for early closure of alloTx arm after 39 pts, however with similar 6 yr PFS/OS rates. The equivalent outcome after VBMCP vs HDT on S9321, interpreted in the context of other randomized clinical trials, may relate to (1) frequent STx in VBMCP failures resulting in significant survival extension in comparison to no STx; (2) superior performance of VBMCP + CTX 4.5 g/m² in comparison with prior SWOG standard regimens (52 mos vs 37 mos; p=.005) in comparison to matched historical controls (Fig 2); and (3) failure to achieve a substantial increase in CR with HDT vs VBMCP (MEL 200 superior to MEL 140 + TBI, reported in IFM 95). We conclude that, while S9321 results overall are significantly improved over prior standard SWOG regimens, the strategic purpose of HDT, effecting sufficiently steep tumor cytoreduction resulting in CR rates in excess of 40%, was not achieved. SWOG 0204 evaluates, in a phase II design, feasibility and efficacy of MEL 200-based tandem autotransplant in the context of THAL + DEX for induction and maintenance. Abstract #137
High-Dose Therapy Autotransplantation/ Intensification vs Continued Conventional Chemotherapy in Multiple Myeloma Patients Responding to Initial Chemotherapy. Definitive Results from PETHEMA after a Median Follow-Up of 66 Months From: Bladé J, Sureda A, Ribera JM, Diaz-Mediavilla J, Garcia-Larana J, Palomera L, Fernández-Calvo J, Marti JM, Giraldo P, Carbonell F, Callis M, Trujillo J, Gardella S, Moro MJ, Barez A, Soler JA, Font LI, Fontanillas M, San Miguel JF. High-Dose Therapy Autotransplantation/Intensification vs Continued Conventional Chemotherapy in Multiple Myeloma Patients Responding to Initial Chemotherapy. Definitive Results from PETHEMA after a Median Follow-Up of 66 Months. Blood 2003;102:42a, abstract 137. Copyright American Society of Hematology, used with permission. J. Bladé, A. Sureda, J. M. Ribera, J. Diaz-Mediavilla, J. Garcia-Larana, L. Palomera, J. Fernández-Calvo, J. M. Marti, P. Giraldo, F. Carbonell, M. Callis, J. Trujillo, S. Gardella, M. J. Moro, A. Barez, J. A. Soler, L. I. Font, M. Fontanillas, J. F. San Miguel, on behalf of Pethema Hospital Clinic, Barcelona, Spain High-dose therapy/stem cell rescue (HDT/SCT) is offered to most younger patients with multiple myeloma (MM) as part of the initial therapy. However, the role of HDT/SCT in the management of patients with MM who have responded to the initial chemotherapy still remains controversial. The aim of this randomized trial was to compare HDT/SCT versus continued conventional chemotherapy in MM patients responding to the initial treatment. From May 1994 to October 1999, 216 patients (122M/94F, median age 56 yrs, stage II or III, ECOG < 3) were registered. The initial chemotherapy consisted of 4 courses of alternating BVMCP/ VBAD. Responding patients received either 8 additional courses of BVMCP/VBAD (arm A) or intensification with HDT/SCT - melphalan 140 mg/m²/TBI 12 Gys or melphalan 200 mg/m2 (arm B). Maintenance treatment consisted of alphainterferon and dexamethasone(Drug information on dexamethasone) in both arms. Onehundred and eighty five patients responded to initial chemotherapy (CR: 15%, PR: 68%, and MR: 17%). Twenty-one of these responding patients were not randomized due to different reasons. Among the 164 randomized patients 83 were allocated to continued chemotherapy and 81 to HDT/SCT. The degree of initial response as well as prognostic features did not differ in both groups. The results were updated as of April 30, 2003 after a median follow-up of 66 months from the initiation of treatment and analyzed on an intention- to-treat basis. CR (negative electrophoresis) was significantly higher in the HDT/SCT arm (30 vs. 11%, p=0.002). However, PFS was not significantly different between HDT/SCT and conventional chemotherapy (median, 42 vs. 33 mos; p=NS) and the OS was similar in both groups (median, 65 vs. 67 mos.; p=NS). The median survival after relapse was identical (16 mos.) in both arms. These results show that HDT/SCT intensification when given to myeloma patients who have responded to the initial chemotherapy, significantly increases complete remission rate but has no significant impact on PFS and OS.