ALBERT

Description: Background. Lenalidomide plus Dexamethasone is approved at first relapse and beyond in Europe, and has transformed the prognosis of Myeloma in the relapse setting. Lenalidomide plus Dexamethasone is approved until progression, that could last for years, the median PFS in phase 3 studies being at 17 months at first relapse, but many patients eventually reach 5 to 7 years these days. Dexamethasone was showed to enhance lenalidomide-antitumor efficacy and to prolong the progression-free survival. However, long term exposure to dexamethasone is also known to be associated to an array of adverse events. Finally, IMiDs are known to act through immunomodulation a class-based mechanism. It is possible that lenalidomide might show efficacy on the long run without need to dexamethasone use, at least for some patients with myeloma. We sought to study the impact of dexamethasone discontinuation beyond six months and one year, and compare this analysis to patients treated on lenalidomide plus dexamethasone. Method. We have recruited 200 relapse refractory myeloma patients for this study from various IFM centers. The patients were to be older than 18 years old and treated with lenalidomide plus dexamethasone. We sought to study the impact of the various ways to use dexamethasone in the real life, and therefore dexamethasone was given according to physician decision. We identified groups according to dexamethasone given high dose (4 days 160mg total in a raw), given once a week at 40mg (considered standard dose), given at lower dose (considered low dose) and a group that had dexamethasone discontinued. Patients were not allowed to have other type of combination but lenalidomide plus dexamethasone. Result. A total of 200 patients were analyzed, median age of 57 years old (range 25-76). 17,5% patients had renal dysfunction at diagnosis. ISS was 2 for 20% and 3 for 20%. Approximately 10% had either del17p or t(4;14). 7% of patients had previous history of venous thrombosis before the treatment. Response rate, survival, including TTP, PFS, EFS and overall survival will be presented at ASH with updated follow-up. Conclusion. This study aims to investigate the importance of long run and exposure to Dexamethasone in the Lenalidomide-Dexamethasone regimen. We also wished to assess the optimal dose of dexamethasone that could be given to patients with prolonged exposure to lenalidomide plus dexamethasone. Disclosures Arnulf: Celgene: Membership on an entity's Board of Directors or advisory committees; Janssen: Membership on an entity's Board of Directors or advisory committees. MACRO:millenium: Membership on an entity's Board of Directors or advisory committees; jansen: Membership on an entity's Board of Directors or advisory committees; celgene: Membership on an entity's Board of Directors or advisory committees. Facon:Celgene: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Millenium: Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees; Onyx: Membership on an entity's Board of Directors or advisory committees; Janssen: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Pierre Fabre: Membership on an entity's Board of Directors or advisory committees. Leleu:LeoPharma: Honoraria; Pierre Fabre: Honoraria; BMS: Honoraria; Novartis: Honoraria; TEVA: Honoraria; Amgen: Honoraria; Takeda: Honoraria; Celgene: Honoraria; Janssen: Honoraria; Chugai: Honoraria.

Description: Background. Bortezomib-melphalan-prednisone (VMP) and melphalan-prednisone-Thalidomide (MPT) are the 2 standards of care upfront in Multiple Myeloma (MM) ineligible for transplantation, along with Bendamustine-prednisone in a very limited indication in Europe. These regimens are based on an alkylator plarform, to which cyclophosphamide might replace melphalan in some countries. For patients initially exposed to Thalidomide several options are offered at first relapse with bortezomib or lenalidomide-based therapy and vive versa at subsequent relapses. For patients initially exposed to bortezomib, either they are retreated with a bortezomib-based regimen or receive a lenalidomide-based therapy, but these patients often have never been exposed to thalidomide throughout their myeloma disease history. We hypothesized that patients that will receive the 3 agents, thalidomide first followed by bortezomib and lenalidomide at subsequent relapses, will have a prolonged survival compared to patients that had bortezomib-based first followed by lenalidomide at subsequent relapses but never been exposed to thalidomide upfront. We sought to understand the prognostic impact of receiving versus being spared from Thalidomide in elderly MM newly diagnosed. Method. A total of 76 patients were recruited, 37% had receive thalidomide and 63% never been exposed to thalidomide. Patients were required to be aged ≥65 years, NDMM treated with either thalidomide upfront or never been exposed to thalidomide upfront or later in the myeloma disease course. Response rate was determined according to IMWG. All survival endpoints were evaluated using Kaplan-Meier estimates and compared with the log-rank test. Results. Overall, the median age was 73 years (range, 65 - 85), with 46% aged 〉75. The m:f ratio was 1.2, 49% of the patients were ISS 3, the median b2m was 4.5mg/L, 33% had an ECOG score ≥ 2, 47% renal insufficiency, 11% had elevated LDH, 8% presence of plasmacytoma, and 11% had adverse FISH (del17p, t(4;14) and or t(14;16)). There was no difference between the 2 studied groups, according to exposure or not to thalidomide. In the thalidomide group, all patients had MPT initially for a median of 8 cycles (range 3 – 12), at a median dose of thalidomide of 100mg/day (50-200), 11% dose reduction, an ORR of 79%, a median PFS of 30 months (CI95% 27;32). In the bortezomib group upfront, patients received Vd, VCd or VMP upfront. The median dose administered of bortezomib was 1.3mg/m², for a median of 5 cycles (2-9). The ORR was 67%, a median PFS of 17 months (CI95% 13;20) with 44% at 2-years PFS. With a median follow-up of 5 years, 93% had relapse, 47% have died. We then sought to compare the OS according to whether the patients were exposed to thalidomide. Interestingly, the median OS of the thalidomide group was 4 years (CI95% 3;5) versus 5 years for the group with no exposition to thalidomide (3.5;6), p=ns. The estimated 6-years OS was 32% and 44% for the 2 groups, respectively. Conclusion. The sequence of bortezomib-based regimen upfront followed by lenalidomide with no exposure to thalidomide in transplant ineligible patients appeared to be slightly superior to the sequence including-based regimen upfront followed by bortezomib and lenalidomide at subsequent relapses. This data needs to be confirmed in a larger study, but it seems that thalidomide could be spared for elderly NDMM that receive bortezomib-based and lenalidomide-based regimens with possibly an improvement of OS in this latter group with a prolonged follow up. Disclosures Gay: Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria; Sanofi: Membership on an entity's Board of Directors or advisory committees. Zweegman:Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Millennium: Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen-Cilag: Membership on an entity's Board of Directors or advisory committees, Research Funding.

Description: Background. Bortezomib has improved overall survival (OS) in light chain (AL) amyloidosis; however, data on its activity in severe cardiac AL are sparse. Furthermore, the impact of the safety profile of Bortezomib on overall survival in severe cardiac AL amyloidosis remains unknown given the fragile population. We sought to outline the activity and safety profile of Bortezomib in severe cardiac AL amyloidosis. Methods. Twenty-seven patients diagnosed with AL amyloidosis and treated with Bortezomib were included, mean age was 63 years (36-85), with a sex ratio of 18/9. Eighteen patients had cardiac involvement, among which all had Mayo-Clinic stage III staging but 3, and 9 had kidney involvement only. Seventy percent of patients received Bortezomib as a 1st-line therapy, once (19%) or twice weekly (81%), given IV at the starting dose of 1.3g/m2 in combination to Dexamethasone. Thirteen (48%) patients also received an alkylating agent. Results. Overall hematological response rate was 75% in patients who received at least 1 cycle of Bortezomib, and 83% and 62.5% in patients with and without cardiac involvement, respectively. Complete response was obtained in 45%, and 42% and 50% in the 2 groups, respectively. 44% patients with cardiac involvement had an organ response. An hematological toxicity occurred in 26% of patients, similarly in the 2 groups, consisting mainly of thrombocytopenia with no need for treatment modification. Non-hematological toxicity (grade ≥2) rate was 62% in patients with cardiac involvement and 38% in patients with kidney involvement (p=ns), consisting mostly of fatigue, peripheral neuropathy, infection and gastro-intestinal adverse effects, and leading to 25% of dose reduction, and 33% of Bortezomib interruption before cycle 4, similarly in both groups. The median follow-up was 41 months from start of Bortezomib. Seven patients died during the first cycle of treatment, all of them but one had severe stage III cardiac involvement with LVEF

Description: BACKGROUND: It has been recently shown that nivolumab, a programmed death 1 (PD-1) blocking antibody, had substantial therapeutic activity and an acceptable safety profile in patients with relapsed or refractory Hodgkin's lymphoma (HL). To our knowledge, this drug has never been tested after allogeneic stem cell transplantation (allo-SCT) mainly because PD-1-blocking strategy may substantially increase the risk of Graft Versus Host Disease (GVHD). Nevertheless, some studies suggested that Reed-Sternberg cells exploit the PD-1 pathway to evade immune detection, possibly explaining why some patients resist to graft-versus-lymphoma effect. PATIENTS AND METHODS: We retrospectively assessed the efficacy and toxicity of nivolumab as a single agent in 12 HL patients relapsing after allo-SCT. All patients had progressive disease at the nivolumab initiation. The dose was 3 mg/kg of body weight every 2 weeks. The median of previous systemic therapies was 9 (range 7-11), including allo-SCT. Patients were required to be off immune suppression for more than 4 weeks prior to nivolumab initiation, with no history of grades III-IV acute or extensive chronic GVHD. Toxicities were graded according to the National Cancer Institute Common Toxicity Criteria for AEs (version 4.0). Efficacy was assessed by physical examination before each infusion and per IWG guidelines (Cheson 2007) after four injections of nivolumab. RESULTS: Eight out of 12 patients were evaluable at the time of submission. Median age at time of transplant was 31 (range 28-42) and median time from allo-SCT to study treatment was 11 months (range 3 - 122). Eight patients had related donors and 4 patients had unrelated donors including one haploidentical transplantation. Best response before allo-SCT was RP in 8 patients and RC in 4 patients. Allo-SCT improved response in only 2 patients. Six patients had received therapy for HL progression/relapse post allo-SCT, including donor lymphocyte infusions (n=4). Median number of injection of nivolumab was 4 (range 1-10). Acute GVHD occurred in 2 patients after 1 injection (grade III skin acute GVHD) and 2 injections (grade IV skin acute GVHD) of nivolumab. In one patient, GVHD responded quickly to corticosteroids therapy. The patient who received haplo-identical allo-SCT presented with corticosteroid-refractory grade IV skin GVHD reversed after several extracorporeal photopheresis sessions. Both patients had prior history of grade II acute GVHD in the three months before nivolumab administration and treatment was stopped. Nivolumab did not modify chimerism. The only serious hematological adverse events were grade 4 neutropenia (1 patient) and grade 3 thrombocytopenia (1 patient). No other non-hematological adverse event was observed except a grade 2 cerebellar ataxia. With a median follow-up of 60 days post nivolumab initiation, 1 patient discontinued due to progressive disease, 2 patients due to acute GVHD and 9 patients remain on nivolumab. In our preliminary analysis of efficacy, 7 out of the 8 (87.5%) patients evaluable for response had clinical benefit, with 4 achieving partial response and 3 in complete response according to Cheson 2007 criteria. Of note, one of the patient developing acute GVHD post nivolumab achieved partial response. No data for the other patient presenting with acute GVHD after treatment was available at the time of analysis. CONCLUSIONS: Our preliminary results suggest that nivolumab is effective with manageable toxicity in patients with relapsed or refractory Hodgkin's lymphoma after allo-SCT. These encouraging results emphasize the need to delineate indications and perform prospective protocols. Disclosures Morschhauser: Genentech Inc./Roche: Other: Advisory boards.

Description: Key Points PD-1 blockade with nivolumab provides durable disease control after allo-HCT. PD-1 blockade with nivolumab after allo-HCT is associated with 30% acute GVHD.

Description: Background: An increased incidence of second primary malignancies (SPM) is described in chronic lymphocytic leukemia (CLL). Current hypothesis to explain this phenomenon mainly relies on immunosuppression conferred by both CLL itself and CLL treatment. The imputability of fludarabine has been long recognized but the risk associated with monoclonal antibodies, especially rituximab recently shown to improve overall survival in CLL when combined with fludarabine and cyclophosphamide (FC), remains unknown. Materials and methods: We conducted a retrospective study of the incidence of secondary cancers in 1255 CLL patients diagnosed since 1980 in the University Hospital of Lille to better characterize the possible risk of SPM associated with rituximab Results: Of 1255 patients, 651 (52%) received therapy including rituximab (38%), FC (26.7%), F alone (22.4%), chlorambucil (27.5%), alemtuzumab (15.5%) and bendamustine (9.3%). Rituximab was given either in combination with FC (27.5%), other chemotherapy (2.6%) or as a single-agent (3.5%). There was no significant difference in terms of age (58 versus 62 years), gender, Binet stage, cytogenetic abnormalities and number of regimen received between patients treated with or without rituximab. Median follow-up was 6 years for all patients (range 2-23), 4.8 years (range, 2-8) since initiation of therapy for patients treated without rituximab and 3.6 years (range, 0.2-11) for patients who received rituximab. Median overall survival (OS) was 18 years for patients treated with R-chemotherapy versus 11 years for patients treated with chemotherapy alone (p

Description: Abstract 2964 Background. Lenalidomide is an oral IMiD®, immunomodulatory compound, approved for use in combination with dexamethasone (Len/Dex) in patients with RRMM who have received one prior therapy. Len/Dex is indicated until evidence of disease progression at the best-tolerated dose of both Len and Dex (Dimopoulos et al. Leukemia 2011). However, the tolerability profile of long term exposure to Len/Dex is not well described, and evidence that long term exposure to Len/Dex would improve on the response rate and survival has yet to be determined. We sought to determine the efficacy and safety profile of long term exposure to Len/Dex in RRMM pts in a multicentre study. Method. We retrospectively reviewed the medical records of 50 RRMM pts treated with Len/Dex and remaining on Len for ≥2 years with a special focus on pts receiving Len for ≥3 years. All pts included had complete follow up records. Results. The median (range min-max) age was 58 years (39–79) with 30% (n=15) 〉 65 years (elderly MM), the sex ratio M/F was 1.2, 49% (n=24) ISS 2 and 3, 12% (n=6) severe renal insufficiency (CrCl 〈 30mL/min), and 8% (n=4) adverse FISH [del17p and/or t(4;14)]. Overall, 25 pts (50%) had Len/Dex at first relapse, 19 pts (38%) at second relapse and 6 (9%) pts in subsequent relapses. Len/Dex was given at first relapse in 10 (66%) elderly patients. The median time from diagnosis to starting Len/Dex was 4.5 years (1–16) for overall cohort and 3 years (1–8) for elderly patients (p=0.05). 28 pts (56%) received Len/Dex for ≥3 years. The median duration on Len/Dex was 3 (2–7) years for the overall cohort, and was 4 (3;7) years for patients exposed to Len ≥3 years. Treatment duration was similar across age categories and across number of previous relapses. With a median follow up of 4 years, 19 patients had stopped Len/Dex. The response rate (ORR, ≥PR) was 96% (n=48), including 37 (74%) patients with ≥VGPR, similar across age categories. Interestingly, the ORR and ≥VGPR were similar irrespective of whether patients have stopped Len/Dex in our study. The ORR was also similar across number of previous relapses, but the ≥VGPR rate was lower in patients at third relapse and beyond, (50%; p=ns). The ORR and ≥VGPR rate was 93% and 77% in patients exposed to Len ≥3 years, similar to the whole cohort. The median time to first response and best response were 2 (1–5) months and 4.5 (2–9) months, respectively. Overall, 9 (18%) patients stopped treatment due to toxicity, 9 (18%) progression of MM, and 1 (0.5%) patient decision. With a median follow up of 4 years, the median (95%CI) TTP was not reached, the estimated 4-yr TTP was 51.5%. There was no imbalance in the incidence of toxicity based on age, number of previous relapses, and patients exposed to Len ≥3 years did not discontinue more often due to toxicity, 14% versus 19% for those receiving Len 〈 3 years. The hematological safety profile was similar across age categories, number of previous relapses, and patients exposed to Len ≥ 3 years; overall, 8 (16%) patients experienced grade 3–4 neutropenia, 6% thrombopenia, and 6% anemia. Ten (20%) patients experienced a thromboembolic event (VTE), all of them of venous type. Two patients had previous history of VTE, but none of them experienced VTE on Len/Dex, likely related to adequate VTE prophylaxis. The median time to first occurrence was 5 (1;28) months, although 4/10 occurred in patients with ≥3 years on Len. All VTE occurred while on VTE prophylaxis except for 1 patient, 5 on aspirin, 2 on prophylactic doses of LMWH, and 2 on VKA (target INR 2–3). The incidence rate of second primary malignancy (SPM) was 3 (6%) (larynx, lung, and MDS). The SPMs occurred at a median time of 4 years from start of Len, while Len was already stopped in 2/3 patients, the latter stopped len at time SPM was diagnosed. Interestingly, none of the patients with more than 3 years exposure on Len had SPM. Conclusions. The current study provides estimates of responses, TTP and safety in a series of MM pts with long-term exposure to Len-based regimen at relapse. 62% of patients remained on Len beyond 3 years reflecting the efficacy and good safety profile of Len in relapsed MM, irrespective of age and number of prior therapies. Furthermore, no excess of long term side effects, including SPM, was observed with a prolonged long follow-up in this study. Disclosures: Robinson: Celgene: Employment. Miljkovic:Celgene: Employment. Morel:Celgene: Employment. Boccacio:Celgene: Employment. Facon:millenium: Membership on an entity's Board of Directors or advisory committees; janssen: Membership on an entity's Board of Directors or advisory committees; celgene: Membership on an entity's Board of Directors or advisory committees; onyx: Membership on an entity's Board of Directors or advisory committees. Hulin:celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; janssen: Membership on an entity's Board of Directors or advisory committees. Leleu:Onyx: Honoraria, Speakers Bureau; Sanofi: Honoraria; Amgen: Honoraria, Research Funding; Novartis: Honoraria, Research Funding; Janssen: Honoraria, Research Funding, Speakers Bureau; Celgene: Honoraria, Research Funding, Speakers Bureau; LeoPharma: Honoraria, Speakers Bureau.

Description: Abstract 3970 Background: Measurement of serum M-spike is used to assess response to therapy and treatment-free survival in IgA myeloma. However, its resolution on SPEP and the presence of IgA-dymers can make accurate measurement difficult. The more the M-spike decreases on SPEP, the more imprecise the M-spike measurement is using SPEP. IgA M-spike often migrates in the betaglobines that renders the M-spike measurement often complicated to analyze. Furthermore, quantification of the clonal IgA chain by nephelometry (IgAneph), which inherently includes monoclonal and polyclonal immunoglobulins, does not accurately reflect the tumour burden. Currently, there is a need in myeloma with IgA isotype (approximately 30–40% of patients) to identify new markers that better reflect the disease burden and the response to treatment, and correlate to patients' outcome. Hevylite® measures IgAkappa and IgAlambda and might provide precise quantitative measurement of the IgA M-spike. We sought to determine whether Hevylite® can be used as a reliable marker for diagnosis and response to therapy in IgA myeloma as compared to the M-spike measurement on SPEP and nephelometry. Methods: We conducted a retrospective analysis on 113, smoldering or symptomatic, IgA myeloma patients at diagnosis referred to our department from 1997 to 2011. All serum samples were collected prior to treatment or at relapse (for sequential data) and were kept frozen since collection. Hevylite® measurements were made at The Binding Site Ltd, Birmingham, UK. A normal range was produced from normal (blood donor) sera (n=138), and were for IgA kappa 0.48–2.82g/L, IgA lambda 0.36–1.98g/L and IgA K/L ratio 0.80–2.04. For ease of comparison we have studied the IgA hevylite ratios expressed as IgA K/L ratio. Results: The median age at diagnosis was 65 years (range: 33–93) and the M/F sex ratio was 0,91. Forty-eight percent of patients had an ISS greater than 2 (n=86). On nephelometry the median IgA level was 22g/L (range min-max, 0–100). Fifty patients had M-spikes migrating among the betablobulines, 27 among the gammaglobulines, 10 migrated in both and 23 unspecified. Fourteen patients had oligosecretory disease (M-component 〈 10g/L). Forty-two patients had an IgA kappa clonal chain, 70 others an IgA lambda and for one patient the data was not available. Among IgA kappa patients the mean HCLratio was 616.6 (median 117.7 [0.021–4323.7]) whereas it was 0.61 (0.21 [0.004–0.455]) among the IgA lambda patients. Across the entire population, 58 patients were identified by SPEP and HCL ratios. Among the 55 patients whose M-component was not quantified on SPEP, HCL ratios were abnormal in 53/55 (96%) cases. In the subgroup of patients whose M-protein migrates in the beta-region 29 out of 50 are identified by HCL ratios and SPEP. Another 19 patients (38%) had abnormal HCL ratios while unquantified on SPEP. The same was seen in patients whose M-protein migrates among the gammaglobulins with 16/27 and 11/27 (40%), respectively. In the last subgroup of patients with beta and gamma migrating M-proteins, all (10/10) were identified by HLC and SPEP. More interestingly, among the oligosecretory MM patients identified by an M-component 〈 10g/L, all (14 patients) had an abnormal HCL ratio. In our series of IgA myeloma, 51% of patients were accurately quantified on SPEP. When using HCLratios, an extra 47% of patients became measurable, with an abnormal HLC ratios allowing IgA myeloma in up to 98% of cases. When considering IgA measurement using nephelometry, 102 patients had both high IgA levels (90%) and abnormal HCL ratios. Interestingly, 7 patients had normal IgA levels with IgAneph and abnormal HCL ratio offering a useful diagnostic tool for 96% of patients as compared to IgAneph. In our series, IgA myeloma was neither associated with a poor outcome (median [range] OS: 119 months [0,5–604]) nor a poor response to therapy (median [range] TTP: 16 months (1–92). HCL involved chain ROC analysis identified a 37g/L cut-off as prognostic in IgA myeloma(p=0.039). Conclusion: Hevylite® is a new and reliable marker for diagnostic and monitoring of IgA myeloma. It enables to quantify accurately up to 98% of IgA-MM patients. These preliminary data need confirmation in further prospective trials in order to monitor further the impact of this marker in IgA myeloma patients before it becomes the gold standard to monitor the IgA M-protein in years to come. Disclosures: Boyle: Chugai: Consultancy, Honoraria. Combat:The Binding Site: Employment. Pietrantuono:The Binding Site: Employment. Facon:onyx: Membership on an entity's Board of Directors or advisory committees; celgene: Membership on an entity's Board of Directors or advisory committees; janssen: Membership on an entity's Board of Directors or advisory committees; millenium: Membership on an entity's Board of Directors or advisory committees. Harding:The Binging Site: Employment. Leleu:Celgene: Honoraria, Research Funding, Speakers Bureau; Janssen: Honoraria, Research Funding, Speakers Bureau; Novartis: Honoraria, Research Funding; Amgen: Honoraria, Research Funding; Sanofi: Honoraria; Onyx: Honoraria, Speakers Bureau; LeoPharma: Honoraria, Speakers Bureau.

Description: Background Light chain amyloidosis (AL) is characterized by a median overall survival (OS) from diagnosis of approximately 3 years, but with clinically overt cardiac involvement this is reduced to 1 year. Bortezomib (B) has shown great promise in the treatment of AL, especially of cardiac involvement. However, efficacy might be hampered by severe safety issues with use of B, primarily of neuropathy type. We sought to study the prognostic impact of the safety profile of B in AL with attention to cardiac involvement. Method This study has included 27 patients with AL, of these 18 had cardiac and 9 kidney but no cardiac involvement. AL was diagnosed as outline in international consensus criteria, as to the hematologic and organ responses. B was given IV twice weekly at the starting dose of 1.3 mg/m², in combination to weekly cyclophosphamide in 13 patients (52%). Results The median age was 63, sex ratio was 18/9, all cardiac AL had Mayo Cardiac Stage III but 3, and none in patients with no cardiac AL. 70% were at diagnostic. Seven patients died during the first month, all of them but one had cardiac AL. We then looked at the safety profile of B in the studied population, and found that 25% and 75% experienced some degree of hematotoxicity in cardiac and in no cardiac AL (p=NS). The non hematology toxicity rate was 62% and 38%, respectively (p=NS). 26% of patients needed dose reduction of B and 33% dose interruption of B in the study before cycle 4, all related to non hematological toxicity of neuropathy, fatigue and GI AEs; and was similar in cardiac as compared to no cardiac AL. The overall hematologic response rate (ORR) was 56% and at least VGPR (with 〉90% decrease in difference between involved/uninvolved light chain) 41%. The responses were 56% and 39% in cardiac AL, similar to patients with no cardiac involvement (56% and 44%), respectively. The median duration of response was 13 months overall, 10 and 20 months in cardiac and in no cardiac AL, respectively (p=NS). Non hematological toxicity did not impact the response rate or the duration of response. With a median follow-up of 41 months from start of B, 70% relapsed and 59% died in the study as a whole, and 67% and 67% in the cardiac group, respectively. The median time to progression was 20 (95CI 4;35.5) months as a whole, and 13 (9;17) months and 20 (0;43) months in cardiac and in no cardiac AL (p=NS). The median OS was not reached in the cohort as a whole and in patients with no cardiac involvement, but was 5 months in cardiac AL (p=NS); the estimated 4-year OS was 55%, 62% and 50%, respectively. Interestingly, all cardiac AL that survived beyond 6 months remain alive at F-up date. Several variables negatively impacted survival in univariate analysis in the group with cardiac AL, including decreased LVEF (p=0.025), NYHA greater than 2 (p=0.007),Mayo Cardiac Stage III (p=0.028), no hematological (p=0.002) and no organ responses (p=0.05), occurrence of non hematological toxicity (p=0.002) with the consequence of dose reduction of B (p=0.09) and dose interruption of B (p=0.04). In multivariate analysis, independent variables that impacted survival were hematological response (OR = 5.1, 95%CI = 1.5-18; p = 0.011) and non hematological toxicity (OR = 3.6, 95%CI = 0.8-14; p = 0.05). Conclusion Bortezomib is a very rapid and effective therapy for AL particularly of cardiac involvement. However, patients may develop severe side effects with Bortezomib that preclude efficacy of Bortezomib given IV, and consequently impact negatively survival. This data favours use of sub cutaneous Bortezomib in AL although not validated in this indication yet. Disclosures: Leleu: Janssen: Consultancy, Honoraria; Celgene: Consultancy, Honoraria; Onyx: Consultancy, Honoraria; Leopharma: Consultancy, Honoraria; Millennium : Honoraria; Amgen: Honoraria; Novartis: Honoraria.

Description: Background Solitary plasmacytoma (SP) is characterized by a localized proliferation of monoclonal plasma cells resulting in a mass in either bone (SBP: Solitary Bone Plasmacytoma) or soft tissue (EMP: Extra-Medullary Plasmacytoma), without evidence of Multiple Myeloma (MM). The prognosis of SP is marked by a high risk of transformation to MM. SBP (as compared to EMP), older age, tumor size 〉 5 cm, and persistence of monoclonal immunoglobulin after treatment were prognostic for progression to MM. More recently, studies demonstrated that presence of focal lesions on magnetic resonance imaging (MRI) favored progression to MM in patients (pts) with plasmacytoma. The exact role of FGD-PET CT (PET) in MM remains debatable, although it clearly provides additional valuable information to assess plasmacytoma in the context of MM, compared with MRI. The prognostic role of PET in identifying progression to MM of SP has not been formally demonstrated. We aimed to determine the impact of PET pre and post therapy on the risk of transformation of SP to overt MM. Method We retrospectively reviewed the medical records of 43 pts diagnosed with one SP clinically, confirmed with histology, either EMP (10 pts) or SBP (33 pts). All pts had one SP clinically, and were treated locally with surgery and/or radiotherapy. PET and MRI were performed at diagnosis prior to (initial) and at the end of therapy. SP was diagnosed as outlined in international consensus criteria. All pts had complete follow up records pre and post therapy. Results The median age was 57.5 years with 33% pts older than 65, the sex ratio was 1.8, IgG kappa was the most frequent isotype, the maximum M-spike value was 30g/L, 48% pts had abnormal involved serum free light chain (isFLC) value and 64% had abnormal sFLC ratio (K/L). 33% had 2 hypermetabolic lesions on initial PET, and 20% had 2 focal lesions on initial MRI. Out of the SBP, 56% were localized on spine or pelvis. With a median follow-up of 50 months (mo), the median overall survival (OS) was not reached for the whole cohort, with a 6-year OS at 79.4%. The median time to MM progression (TTMM) was 71 mo (95%CI: 59;101). The TTMM was not significantly different in SBP versus (vs) EMP although the 5-year TTMM was 58% and 83%, respectively. The TTMM for the 2 hypermetabolic lesions on initial PET group was 23 mo (9;37) vs not reached otherwise (p=0.003). Conversely, MRI at diagnosis did not have any impact on TTMM in our study, although the median TTMM for the 2 focal lesions on initial MRI group was 30 mo (9;51) vs not reached otherwise. Age had no impact on TTMM, but abnormal initial K/L ratio (p=0.022) and abnormal initial isFLC (p=0.002) did impact TTMM, 36 mo (14;58) and 21 mo (0;42) vs not reached otherwise, respectively. A normalized PET at completion of treatment did not reach significance, as to normalized MRI, but the absence of normalized isFLC value also impacted TTMM, 21 mo (10;32) vs not reached otherwise (p=0.016). Using multivariate analysis, independent variables that impacted TTMM were abnormal initial isFLC (OR=10, 95%CI=1-87; p=0.008) and initial PET (OR=5, 95%CI=0-9; p=0.032). Interestingly, initial PET did not influenced OS, median 71 mo for the 2 hypermetabolic lesions on initial PET group vs not reached otherwise, respectively (p=ns). This data suggested that pts with SP that transformed into MM did not have a worse prognostic at time of MM. The median OS of pts with SP from start of MM was not reached, the 4-year OS was 66%. With a special focus on SBP, we identified the exact same prognostic factors for TTMM as to the whole cohort in univariate and multivariate analysis. The localization of SBP in the spine is usually considered of poor prognosis, but we did not find any confirmation of this observation in our study. Conclusion FDG-PET CT and involved sFLC value at diagnosis of SP are important predictors of the risk of progression to MM. This data analysis may lead to a different management of SP for patients with one or the 2 abnormal indicators, irrespective to the solitary clinical aspect of SP. One may consider to embrace treatment of MM for SP when FDG-PET CT and involved sFLC value are abnormal at diagnosis, while surgery and/or radiotherapy would remain the appropriate therapeutic procedure for SP otherwise. Disclosures: Facon: JANSSEN: Honoraria, Speakers Bureau; CELGENE: Honoraria, Speakers Bureau. Leleu:JANSSEN: Honoraria; CELGENE: Honoraria.