Promising new drugs are being evaluated for treatment of multiple myeloma (MM), but their impact should be measured against the expected outcome in patients failing current therapies. However, the natural history of relapsed disease in the current era remains unclear. We studied 286 patients with relapsed MM, who were refractory to bortezomib and were relapsed, refractory, or ineligible, to an IMiD (Immunomodulatory Drug), with measurable disease and ECOG PS of 0, 1 or 2. The date patients satisfied the entry criteria was defined as time zero (T0). The median age at diagnosis was 58 years and time from diagnosis to T0 was 3.3 years. Following T0, 213 (74%) patients had a treatment recorded with one or more regimens (median=1; range 0-8). The first regimen contained bortezomib in 55 (26%) patients and an IMiD in 70 (33%). A minor response or better was seen to at least one therapy after T0 in 94 patients (51%) including >=partial response in 69 (38%). The median overall survival and event free survival from T0 were 9 and 5 months respectively. This study confirms the poor outcome once patients become refractory to current treatments. The results provide context for interpreting ongoing trials of new drugs.

4477 Tyrosine kinases Inhibitors (TKI) of first and second generation are presently the choice of treatment for chronic myeloid leukemia (CML) and adherence to the treatment is an important factor to obtain good therapeutic results. We used the World Health Organization adherence definition that enrolls a global person's behavior including how the medication is used, diet, and/or lifestyle change. Objective: To evaluate the adherence to TKI in patients with CML and to identify factors that can affect adherence to TKI treatment. Methods: A prospective and observational analysis was performed with 122 patients taking TKI, in a public institution. The median of observation time was 169 days (131-240). We used the Hasford scale to risk assessment of diagnosis. The adherence was calculated using the mean medication possession ratio (MPR), calculated as total days’ dose of TKI divided by the number of the days in the observation time. Statistical analysis began with descriptive analysis and after that we applied Pearson or Spearman's correlation and t-Test, what it is adequate, considering significant p-value £ 0.05. Results: A total of 122 patients were evaluated: 92 (75.5%) were taking Imatinib, 16 (13%) Nilotinib, 09 (7.5%) Dasatinib and 5 (4%) Bosutinib. The mean age was 48 (20-82) years. The majority of patients were male (61%) and the median of TKI treatment time was 45 months (7-114). There were 103 (84.5%) patients in first line treatment, 17 (14%) in second line and 2 (1.5%) in third line. Considering all patients, median and mean of adherence was 96% and 90% whereas 23% of patients had 100% of MPR. There was a significant difference between adherence distribution and cytogenetic results, best adherent patients had most complete and major cytogenetic response (p= 0.01) in CML Hasford lower risk group. In all patients, the MPR decreased with longer use of TKI treatment (p= 0.02) and with longer time of diagnosis (P= 0.002). The adherence was superior in patients participating in clinical trials (p= 0.01) and using second generation TKI (p= 0.001). Conclusion: The best adherence was correlated to a better cytogenetic response in Hasford lower risk group. Longer time of treatment was related to a poorer adherence, suggesting a continuous approach by health team can make a difference. Disclosures: No relevant conflicts of interest to declare.

2755 Background: In ENESTnd, nilotinib demonstrated superior efficacy vs imatinib in newly diagnosed patients (pts) with CML-CP, including a significantly reduced rate of progression to AP/BC on treatment. Here, we examined the occurrence of emergent mutations on treatment and their impact on response. Data on the incidence of mutations and impact on efficacy with a minimum follow-up of 36-months (mo) for all pts will be presented. Methods: Pts with CML-CP were randomized to receive nilotinib 300 mg BID (n = 282), nilotinib 400 mg BID (n = 281), or imatinib 400 mg QD (n = 283). Mutation testing was performed by direct sequencing of the kinase domain (amino acids 230 to 490; sensitivity, 10%-20%) in a central lab at: baseline, 5-fold increase in BCR-ABL levels, lack of MMR at 12 mo, loss of MMR, or treatment discontinuation. Results: With a minimum follow-up of 24 mo, twice as many pts had emergent mutations on imatinib (n = 20) vs nilotinib (n = 10, nilotinib 300 mg BID; n = 8, nilotinib 400 mg BID), with the majority of mutations emerging in pts with high and intermediate Sokal scores (Table). Of pts with mutations emerging on imatinib, the majority (65%) had nilotinib-sensitive, imatinib-resistant mutations; whereas nilotinib was effective in preventing the emergence of clones with nilotinib-sensitive mutations. The incidence of T315I mutations was similar for the nilotinib (n = 3, nilotinib 300 mg BID; n = 2, nilotinib 400 mg BID) and imatinib (n = 3) arms and most of these T315I mutations (6/8) were detected within the first 12 mo of therapy. All but 1 pt with the T315I mutation had a high Sokal risk; the other pt had an intermediate Sokal risk. Overall, across the 3 treatment arms, the incidence of any mutation was 14% in pts who had BCR-ABLIS > 10% at 6 mo vs 4% in pts with BCR-ABLIS ≤ 10% at 6 mo. The majority of pts with emerging mutations had suboptimal response (SoR) or treatment failure (TF) on treatment; all pts with the T315I mutation had SoR or TF. Of the pts with mutations, 1/10 pts on nilotinib 300 mg BID, 2/8 pts on nilotinib 400 mg BID, and 7/20 pts on imatinib, progressed to AP/BC on treatment. BCR-ABL mutations did not account for all cases of progression to AP/BC, loss of CCyR, and loss of MMR on treatment (Table). Of the pts who achieved an MMR on treatment, 0/203 (0%), 2/192 (1%) and 3/131 (2%) had a mutation and lost MMR with nilotinib 300 mg BID, nilotinib 400 mg BID or imatinib, respectively. Conclusions: Nilotinib may be more effective in preventing the development of emerging mutations vs imatinib. More pts with new mutations progressed to AP/BC on imatinib than on nilotinib. These data suggest that deeper molecular responses with nilotinib protect from the development of emerging mutations and progression to AP/BC vs imatinib. Disclosures: Hughes: Bristol Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Ariad: Honoraria, Membership on an entity's Board of Directors or advisory committees. Kim:Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Bristol Myers Squibb: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pfizer: Membership on an entity's Board of Directors or advisory committees, Research Funding. Kurokawa:Novartis Pharmaceutical: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Kalaycio:Novartis Pharmaceutical: Honoraria, Research Funding, Speakers Bureau. Saglio:Bristol Myers Squipp: Consultancy, Speakers Bureau; Novartis Pharmaceutical: Consultancy, Speakers Bureau; Pfizer: Consultancy. Larson:Novartis Pharmaceuticals: Consultancy, Honoraria, Research Funding. Kantarjian:Pfizer: Research Funding; Novartis: Research Funding; Novartis: Consultancy; BMS: Research Funding. Hoenekopp:Novartis Pharmaceutical: Employment, Equity Ownership. Shou:Novartis: Employment. Yu:Novartis: Employment, Equity Ownership. Blakesley:Novartis Pharmaceutical: Employment. Rosti:Novartis: Consultancy, Honoraria, Research Funding; Bristol Myers Squibb: Consultancy, Honoraria; Roche: Honoraria. Hochhaus:Bristol Myers Squibb: Consultancy, Honoraria, Research Funding; Novartis Pharmaceutical: Consultancy, Honoraria, Research Funding; Ariad: Consultancy, Honoraria, Research Funding; Merck: Consultancy, Honoraria, Research Funding.

Background: Paroxysmal nocturnal hemoglobinuria (PNH) is a rare acquired disorder of hematopoietic stem cells. The disease is diagnosed with hemolytic anemia, marrow failure or episodes of venous thrombosis. Stem cell transplantation (SCT) is the only curative treatment. Until now, there are few reports of SCT in PNH and all but one includes small numbers of patients (Saso R et al, British Journal of Haematology, 1999). Aim: To evaluate the outcome and survival risk factors of PNH patients after SCT according to clinical status of PNH at time of transplantation, and to perform a comparative analysis with a matched cohort of PNH patients without transplantation. Methods: We analyzed the outcome of PNH patients who underwent SCT in Europe through the European Blood and Marrow Transplant (EBMT) Group Promise registry. An additional questionnaire was sent to all centers on PNH specific history and classification. One hundred and six hematological centers were contacted through the EBMT group. Non-transplanted PNH patients were previously reported to the French Society of Hematology registry (Peffault de Latour R et al, Blood 2008). Both populations will be matched according to the date of PNH complications known to be related to bad outcome (thrombosis, aplastic anemia and/or malignant disease). Results: Between September 1978 and April 2007, 268 PNH patients were reported to the EBMT registry. Among those patients, 141 were updated (74 male, 52%) for the present study from 42 hematological centers. The 3 main indications for SCT were aplastic anemia (n=86, 61%), severe recurrent hemolytic crisis (n=43, 30% and thrombosis (n=31, 30%). Concerning disease subcategories recently described (Parker C et al, Blood 2005) 75 patients were transplanted because of PNH in the setting of another specified bone marrow disorder (54%) and 62 patients because of classical PNH (hemolysis without marrow failure, 45%). At time of SCT, the median age of the population was 30 years (IQR 23 to 36). The median interval from diagnosis of PNH to the time of SCT was 21 months (IQR 7 to 69). Eighty-nine patients were transplanted from HLA-identical siblings (64%). The source of stem cells was mostly bone marrow (n=95, 68%). The 2 main conditioning regimens consisted in Cyclophosphamide with Antihymoglobulin or Busulfan. The graft versus host disease (GvHD) prophylaxis associate ciclosporine with or without methotrexate in 102 patients (72%). During evolution, 9 patients did not engraft (6%). Acute GvHD of grade II or more developed in 35 patients (Grade III–IV, n=12). Chronic GvHD developed in 45 (limited, n=26; extensive, n=19). With a median follow-up (±SE) of 62.5 months (±6.2), 39 patients died. The 5-year survival rate was 70% (±4.2%). The main causes of death were infections (n=19), GvHD (n=9) and Hemorrhage (n=4). None of the variables examined for associations with transplant outcome were statistically significant predictors of survival including: age, interval from diagnosis to transplant, year of transplant, stem cell source, donor type, HLA-matching, indications for SCT and PNH subcategories. The non transplanted PNH population includes 401 patients, diagnosed in France from 1950 to 2005. The median follow-up (±SE) is 123.2 months (±9.8). The 10-year survival rate was 67.8% (±4.6%). Conclusions: This study confirms on a large cohort of PNH patients that SCT using related or unrelated donor is a valuable curative option for PNH complications. Nevertheless, no survival associated risk factor was identified. Due to the reduced number of PNH SCT patients, the on-going matched-control analysis with non transplanted PNH patients is the only way to access the potential benefit of SCT in case of PNH complications.

4561 Introduction: In recent decades, several advances were made in the management of patients (pts) who underwent allogeneic hematopoietic stem cell transplantation (HSCT). Despite that, the transplant related mortality (TRM) is not negligible. Tools that could predict TRM, based on pre HSCT characteristics and co-morbidities were always sought. Some of these tools are the modified hematopoietic cell transplantation co-morbidity index (HCT-CI). Objectives: To apply the modified HCT-CI and correlate it with overall survival (OS) and non-relapse mortality (NRM). Pts and Methods: This analysis include patients who received allogeneic HSCT, for malignant and non malignant hematological diseases, after high or low dose conditioning regimens, from related and unrelated donor, at the HSCT unit of the University Hospital of Campinas between 1993 to 2010. The conditioning regimens and graft versus host disease (GVHD) prophylaxis were selected according to ongoing protocols at the University Hospital. Informed consent for transplant was obtained from all patients. We defined low dose conditioning regimen as follow: busulfan dose < 9 mg/kg, melphalan dose < 150mg/m2, and total body irradiation dose of 2Gy. All patients received cyclosporine-A (CsA) with methotrexate or CsA with mycophenolate mofetil as GVHD prophylaxis. Screening for cytomegalovirus with antigenemia and preemptive treatment with ganciclovir were made for all patients. Prophylaxis for Candida infections was made with fluconazole; for bacterial infections ciprofloxacin was used. Prophylaxis for Pneumocystis jirovecii was made with sulfamethoxazole and trimethoprim. For viral agents as herpes simplex virus and varicella zoster virus, acyclovir was used. The diagnosis and clinical grading of acute and chronic GVHD were performed using standard criteria. Information on co morbidities was extracted, retrospectively, from medical charts. Each co-morbidity was assigned an integer weight, based on derived scores from Sorror et al. The HCT-CI score was the sum of this integer weights, and the patients were then assigned to one of the risk groups: 0, 1, 2 and 3 or more. Kaplan-Meier was applied for estimation of overall survival. NRM was estimated by cumulative incidence considering primary relapse disease as competing risk. Results: 457 pts were analyzed at the beginning. However, 331 (72.4%) were evaluable, 126 (27.6%) were excluded when at least one parameter was missing. 208 (63%) were male with a median age of 36.5 (5–65) years, 249 (75%) pts received high dose (HD) conditioning and 82 (25%) low dose (LD). Distribution of diseases in the HD was: 110 (44%) acute leukemias; 109 (43%) CML and 30 (13%) others; LD was: 42 (51%) severe aplastic anemia; 14 (17%) lymphomas; 12 (15%) myeloma and 14 (17%) others. HCT-CI in HD was: 127 (51%) score 0; 39 (16%) score 1; 19 (7%) 2 and 64 (26%) score ≥ 3; in LD 33 (40%) score 0; 14 (17%) score 1; 8 (10%) score 2 and 27 (33%) score ≥ 3. The median follow up in HD was 22 months (0–201) and 9 (0–203) in LD. The adjusted OS at 10 year for HD conditioning according to the HCT-CI was 55% for score 0, 34% for score 1–2 & 34% for score ≥ 3 and for LD conditioning was 58% for score 0, 53% for score 1–2 & 15% for score ≥ 3 (P< 0.0001). The NRM rate in the HD cohort was 32% in the pts with a score 0, 42% for score 1–2 and 53% in the score ≥ 3. In the LD cohort, the NRM was 23% in the pts with a score of 0; 35% for score 1–2 and 54% in the score ≥ 3 (P= 0.002). The adjusted OS for conditioning type in patients with cardiac valve disease was associated to a lower OS (29% and 33% P= 0.002) whereas mild hepatic involvement was just significant in the HD cohort. Conclusion: The OS and NRM were worse for score ≥ 3 either in HD and LD groups. The heart valve disease emerged as a worse outcome in both conditioning type and mild hepatic involvement only in HD group. The HCT-CI showed to be a valid tool to predict outcome after allogeneic HSCT. Disclosures: No relevant conflicts of interest to declare.