ALBERT

Description: Biallelic variants in the ERCC excision repair 6 like 2 gene (ERCC6L2) are known to cause bone marrow failure (BMF) due to defects in DNA repair and mitochondrial function. Here, we report on eight cases of BMF from five families harboring biallelic variants in ERCC6L2, two of whom present with myelodysplasia. We confirm that ERCC6L2 patients’ lymphoblastoid cell lines (LCLs) are hypersensitive to DNA-damaging agents that specifically activate the transcription coupled nucleotide excision repair (TCNER) pathway. Interestingly, patients’ LCLs are also hypersensitive to transcription inhibitors that interfere with RNA polymerase II (RNA Pol II) and display an abnormal delay in transcription recovery. Using affinity-based mass spectrometry we found that ERCC6L2 interacts with DNA-dependent protein kinase (DNA-PK), a regulatory component of the RNA Pol II transcription complex. Chromatin immunoprecipitation PCR studies revealed ERCC6L2 occupancy on gene bodies along with RNA Pol II and DNA-PK. Patients’ LCLs fail to terminate transcript elongation accurately upon DNA damage and display a significant increase in nuclear DNA–RNA hybrids (R loops). Collectively, we conclude that ERCC6L2 is involved in regulating RNA Pol II-mediated transcription via its interaction with DNA-PK to resolve R loops and minimize transcription-associated genome instability. The inherited BMF syndrome caused by biallelic variants in ERCC6L2 can be considered as a primary transcription deficiency rather than a DNA repair defect.

Description: Iron-mediated carcinogenesis is thought to occur through the generation of oxygen radicals. Iron chelators are used in attempts to prevent the long term consequences of iron overload. In particular, 1,2-dimethyl-3-hydroxypyrid-4-one (L1), has shown promise as an effective chelator. Using an established hepatocellular model of iron overload, we studied the generation of iron-catalyzed oxidative DNA damage and the influence of iron chelators, including L1, on such damage. Iron loading of HepG2 cells was found to greatly exacerbate hydrogen peroxide–mediated DNA damage. Desferrithiocin was protective against iron/hydrogen peroxide–induced DNA damage; deferoxamine had no effect. In contrast, L1 exposure markedly potentiated hydrogen peroxide–mediated oxidative DNA damage in iron-loaded liver cells. However, when exposure to L1 was maintained during incubation with hydrogen peroxide, L1 exerted a protective effect. We interpret this as indicating that L1's potential toxicity is highly dependent on the L1:iron ratio. In vitro studies examining iron-mediated ascorbate oxidation in the presence of L1 showed that an L1:iron ratio must be at least 3 to 1 for L1 to inhibit the generation of free radicals; at lower concentrations of L1 increased oxygen radical generation occurs. In the clinical setting, such potentiation of iron-catalyzed oxidative DNA damage at low L1:iron ratios may lead to long-term toxicities that might preclude administration of L1 as an iron chelator. Whether this implication in fact extends to the in vivo situation will have to be verified in animal studies.

Description: Introduction The TP53 gene encodes the tumour suppressor and cell cycle regulatory protein and is found to be mutated in a variety of carcinomas. Mutation in TP53 gene is associated with resistance to conventional therapy, disease progression and overall poor prognosis in solid tumours and haematological malignancies including Myelodysplastic Syndromes (MDS). TP53 mutated sub-clones in MDS have been demonstrated by deep sequencing technology in prior studies. Strong nuclear staining of p53 protein by immunohistochemistry has been used as a surrogate marker for TP53 gene mutation in haematological and other malignancies. Methods We analysed sequential marrow samples for p53 expression on 35 patients with MDS from a single institution pre and post Azacitidine therapy. Formalin fixed, paraffin embedded marrow biopsies were stained with DO-7 mouse p53 monoclonal antibody. 1000 haematopoietic cells were examined under the high power and p53 expression was determined as per Modified Quick Score. Results Median age of the patients was 70 and WHO subgroups were identified as follows: 7 RCMD, 1 5q-syndrome, 1 MDS/MPN, 8 CMML, 6 RAEB-1, 6 RAEB-2 and 6 t-MDS. Cytogenetic risk as per IPSS-R/CPSS showed 17 (50%) lower risk, 4(12%) intermediate risk and 13(38%) higher risk groups. Patients received a median 13 cycles of Azacitidine. Marrows were assessed prior to treatment and after 3-6 cycles. Median overall survival of the study group was 20 months and transformation to AML occurred in 13 patients (37%). At diagnosis, 27 patients (77%) were p53 negative and 8 patients (23%) were p53 positive. At reassessment, 24 patients (69%) remained p53 negative while 6 patients (17%) remained p53 positive. Two patients (6%) became p53 negative and 3 (8%) became p53 positive following Azacitidine treatment. Median overall survival of patients who remained p53 negative during Azacitidine treatment was 28 months compared to 11 months in patients who remained p53 positive, p=0.005. Similarly, median overall survival of patients who remained or became p53 negative was 28 months compared to 18 months for those who remained or became p53 positive during Azacitidine therapy, p=0.012. p53 expression at diagnosis or changes in p53 expression during Azacitidine treatment did not correlate with transformation to Acute Myeloid Leukaemia (AML) or time to progression to AML. Among the p53 positive group, patients who had more than 10% p53 expression had lower overall survival compared to those who were 10% at DiagnosisPositive p53

Description: Abstract 1791 Six courses of FCR is considered standard therapy for fit, non-(17p) deleted patients with Chronic Lymphocytic Leukaemia (CLL), however the optimal dose of Rituximab in combined chemotherapy for CLL or the number of FCR courses required has never been formally assessed. Minimal residual disease (MRD) eradication (assessed by 6 colour flow-cytometry(FC) is associated with an increased disease free interval and is a logical endpoint in determining length/efficacy of treatment. Serial MRD testing following therapy provides an objective test of disease re-emergence. A multi-centre prospective phase II study in treatment-naïve CLL patients with modified FCR (Rituximab 375mg/m2) using MRD to determine treatment length/efficacy and identify disease emergence recruited from 2009–2012. Standard pre-treatment assessment plus C-T scan, FISH and Ig somatic hypermutation (SHM) analyses were performed. Patient in radiological and MRD-ve remission after 4 courses of FCR stopped therapy and the remainder received 6 courses. All were followed by 6 monthly MRD assessment. Fifty-two patients were included (35M/17F), mean age 52 years (range 37–72), mean WCC 51 × 109/L (range 8–386), elevated LDH 23 of 45 (51%), lymphadenopathy 43 (83%) and hepatosplenomegaly in 37 (71%). Abnormal FISH results were, del(11q) in 15, +12 in 5, del(13q) in 18; SHM 15(29%) mutated and 37(71%) unmutated or with V3–21 gene usage. Post-treatment MRD is available in 43 patients; 36 (70%) were MRD –ve including 18 (42%) after 4 courses. Nine patients did not complete treatment (toxicity −7, progression-1, non-compliance-1). With a mean follow-up of 20.7 months (range 4.5–38), 6 patients have reverted to MRD +ve at a mean of 13 months (range 10–19) from therapy; only the patient with primary treatment failure has required further chemotherapy. Myelotoxoicty resulted in 23 NCI grade 3 episodes and 7 treatment delays of a mean duration of 25 days (range15–32). One further grade 3 toxicity of pneumonia was identified. Modified FCR was effective in this patient cohort with high risk features (38% unfavourable FISH, 71% unfavourable SHM), with 70% patients achieving MRD-ve status on completion of therapy. 42% of patients became MRD-ve after 4 courses of FCR, suggesting that some patients may not require 6 courses of therapy. Myelotoxicity remains an issue with 7 patients not completing therapy. Longer follow-up will clarify whether shortened therapy will have an impact on MRD+ve reversion, time to re-treatment and survival. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 3171 Autoimmune haemolytic anaemia (AIHA) is characterised by severe haemolysis due to development of auto antibodies directed against patient's own red cells. Rituximab is a chimeric monoclonal antibody that specifically depletes B cells by targeting CD20 on both immature and mature B lymphocytes We evaluated the use of Rituximab in patients with AIHA who are refractory to conventional Immunosupression in 5 centres in ireland. Methods: 24 cases of AIHA from five centres in the Republic of Ireland were identified who were treated with Rituximab at the standard dose of 375mg/m2 weekly for four weeks. All patients had received prior therapy with steroids and 50% of patients received two or more courses of Immunosupression. One patient had splenectomy prior to Rituximab. Five patients had an underlying lymphoproliferative disorder and two had a myeloproliferative neoplasm. Response was assessed at 1,3,6,12,36, 48 and 60 months. Medium duration of follow up was 34 months. Response was defined as normalisation of Haemoglobin concentration and haemolytic parameters reticulocyte count, bilirubin and lactate dehydrogenase (LDH). Overall response to Rituximab was 83.3% at 28 days post treatment. Four patients responded to Rituximab in combination with other Immunosupression such as Azathioprine or low dose steroids. Medium response duration was 11.5 months. Sustained response at 1 year, 2 years and 4 years post treatment were 60%, 55% and 30% respectively. Rituximab was well tolerated without significant complications in the majority of patients. One patient developed neutropenic sepsis. One patient received second course of Rituximab after relapse within a year and had sustained response of over two years after the second course. Conclusion: Rituximab is a safe and effective treatment for patients with AIHA who are resistant to steroids.The majority of patients will initially respond but a significant number will gradually relapse over time. Re-treatment with rituximab maybe considered on relapse. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction Myelodysplastic Syndrome (MDS) is classically a disease of older people, with median age at presentation of 70-75 years. The incidence of MDS is estimated at 5-13/100,000/year, but rises to 〉20/100,000/year in older populations. An increase in diagnosis over the last decades is in part due to improved recognition of MDS, but likely also to an increase in the ageing population. There is very little data on the clinical course, management and outcomes for very old patients (≥85 years of age) with MDS. Patients and Methods: This was a retrospective, multicentre analysis of 84 patients with MDS or Chronic Myelomonocytic Leukemia (CMML) aged ≥85 years at diagnosis from 6 centres in Ireland. Results: We identified 84 patients aged ≥85 years at time of diagnosis of MDS (n= 70) or CMML (n=14), including 47 men (56%) and 37 women (44%). Median age at diagnosis was 87 years (range 85-98). Most patients (93%) were anemic at presentation, including 45/47 men (96%) and 33/37 women (89%). Median hemoglobin (Hb) was 9.5 g/dl (range 5.9 -13.8). Median neutrophil count was 2.4 x109/L (range 0-72). Forty-four patients had thrombocytopenia (median platelet count 144 x 109/L (range 18-624)). Data regarding co-morbidities were available for 75 patients: 69% had hypertension, 36% ischemic heart disease, 39% atrial fibrillation, 31% heart failure, 19% diabetes and 39% renal dysfunction. Ferritin was elevated in 18 (32%) of 57 patients tested. 2006 WHO subgroups were reported for 81 patients: RCMD (32; 40%), CMML (14; 17%), RA (10; 12%), RAEB-1 (10; 12%), RAEB-2 (7; 9%), RARS (2; 3%), t-MDS (2; 3%), Hypoplastic MDS (1; 1%) and 5q- Syndrome (1; 1%). Cytogenetic analysis was performed in 49 patients (58%); results were available for 39 (46%). No patient had molecular studies for MDS-associated mutations or p53 deletions/mutations. Karyotype was normal in 23 patients (59% of those with results available), deletion Y in 5 (13%), Trisomy 8 in 5 (13%), complex in 3 (7.7%), 5q- in 2 (5.2%), and monosomy 7 in 1 (2.5%). Risk stratification by IPSS-R was available for only 37/84 patients, primarily due to lack of cytogenetic testing. Data were available regarding treatment strategies for 81 patients. Thirty-five (43%) received supportive care only. Forty-five patients (57%) were transfused; 29 (34%) became transfusion-dependent during the course of their disease. Of these, only 14 (48%) received erythropoietin (EPO). Of 50 patients with significant anemia likely to cause symptoms (Hb 〈 10g/dl), only 21 (42%) received EPO. Five patients (6%) received azacitidine (1-18 cycles; median 5), 7 (8%) received G-CSF; none received lenalidomide or iron chelation. Median survival for all patients was 17 months (range 0-147), 16 months for men (range 0-70), and 27 months for women (range 1-147). In 35 patients who had IPSS-R data available, median survival was 49 months for Very Good, 30 months for Good and 13 months for Intermediate category patients. For 4 patients in the Poor and Very Poor categories median survival was 1, 5, 7 and 28 months. Median survival for patients with RA was 28 months (n=10), RCMD 25 months (n=29), CMML 13 months (n= 14), RAEB-1 10 months (n=10) and RAEB-2 19 months (n=7). Six patients (including 3 with RAEB-1, 1 with CMML and 1 with t-MDS) developed Acute Myeloid Leukaemia (7%) at a median of 4.5 months from diagnosis. Median survival for these patients was 9.5 months. Of 84 patients, 60 have died. The main causes of death included marrow failure, sepsis, cardiac events, other malignancies and gastrointestinal bleeding. Conclusions Anemia is the commonest presenting feature of MDS in the very old, and may be the sole cytopenia. Unexplained anemia in the very old should trigger suspicion of underlying MDS, especially if associated with a high MCV. In many patients over 85 years cytogenetic analysis is not performed, precluding accurate prognostic evaluation. MDS in these very old patients is not often actively managed with pharmacological intervention or chemotherapy. Up to 50% of transfusion-dependent patients do not receive erythropoeitin. Azacitidine and lenalidomide are infrequently used. Co-morbidities (especially cardiac and renal disorders) are very common. Survival can be prolonged, especially in patients with low-risk disease. With an ageing population, management of very elderly patients with MDS is becoming more challenging and a more proactive approach should be considered. Figure. Figure. Disclosures Quinn: Janssen: Honoraria.

Description: Fludarabine, cyclophosphamide and rituximab (FCR) therapy results in a complete remission (CR) rate of 72% and a median progression free survival (PFS) of 80 months in non-del (17P) CLL1. Achieving an MRD negative (MRD-ve) CR after completing therapy is an early surrogate marker for overall survival (OS) and PFS2. Specific genetic CLL subtypes determined by fluorescent in-situ hybridisation (FISH) analysis, immunoglobulin mutation IgVH, NOTCH1 and SF3B1 status determine response to chemotherapy3,4. We completed a multi-centre prospective study between 2008 and 2012, with a median follow up of 62.6 months using MRD status to determine length of therapy. Patients who achieved an MRD-veCR after 4 courses of FCR received no further therapy and the remaining patients completed 6 cycles of FCR. MRD status was tracked 6 monthly in patients who became MRD-veuntil MRD was detected. The genetic subtype was also analysed but did not influence treatment. Fifty-two patients {35M;17F, median age 61years (range 37-73)} were enrolled. Forty-six patients completed the MRD assessment after 4 cycles. Eleven patients discontinued assigned FCR therapy for the following reasons: prolonged cytopenia (4); non-compliance (1); autoimmune haemolytic anaemia (2); renal impairment (1); pleural effusion (1); not recorded (2). Eighteen (34.6%) patients achieved an MRD-veCR after 4 cycles and a further11 after 6cycles resulting in 29/52 (55.8%) MRD-veCRs in total. The median PFS was 72.3 months (95% Confidence Interval 61.3-84.1 months) and the median OS has not been reached. Patients who attained an MRD-veversusMRD+vestatus had a prolonged PFS (81.1 vs 46.2 months, p

Description: Background: The presence of significant marrow fibrosis has previously been shown to be a poor prognostic factor in patients with myelodysplastic syndrome (MDS). Associations between fibrosis and higher transfusion requirements, multilineage dysplasia, and an increased rate of leukaemic transformation have been demonstrated. Currently, the presence of fibrosis is not included in standard risk scores for MDS such as the Revised International Prognostic Scoring System (IPSS-R) nor is fibrosis included in the current WHO classification for MDS. It is also not certain whether the presence of fibrosis should alter current treatment algorithms for patients. More information is needed to further assess both the prognostic and therapeutic implications of the presence of significant fibrosis in this patient population. Methods: We conducted a retrospective study utilizing a database of 247 patients with diagnosed MDS in a single center from 2000-2014. Bone marrow trephine samples for 200 of these patients were assessed using the European consensus on grading bone marrow fibrosis. Data was collected on: age, gender, WHO classification, marrow blast percentage, cytogenetics, progression to AML, treatment with azacitadine, and overall survival. These characteristics were compared between patients without significant fibrosis (grade 0/1) and those with significant fibrosis (grade 2/3). Our aim was to identify potential significant associations with MDS fibrosis and to assess whether treatment with azacitadine influenced these parameters or overall survival between the two groups. Results: Of 200 patients, 38 were found to have significant fibrosis (19%) versus 162 without significant fibrosis (81%). There was no significant difference in age or gender between the two groups. The commonest WHO category in both groups was RCMD (31.58% v 48.77%) in the fibrosis and non-fibrosis groups respectively. IPSS-R score was determined for 152 patients where data was available. There was no significant difference observed in IPSS-R Score between the two groups. Cytogenetic data was available on 175 patients. The commonest cytogenetic result in both groups was normal karyotype (55.3% v 52.5%). In the fibrosis group this was followed by complex cytogenetics (〉 2 abnormalities) (23.68%) and trisomy 8 (7.89%). The presence of a cytogenetic abnormality was not significantly different in those with or without fibrosis (p=0.69). A significant difference was found between patients' marrow blast percentage at diagnosis (average 5.4% blasts in fibrotic patients versus 3.6% in non-fibrotic patients (p=0.04)). There were no differences in diagnostic haemoglobin level, neutrophil count, and platelet count. Acute Myeloid Leukaemia (AML) developed in 31 patients. The presence of fibrosis was associated with an increased rate of AML transformation with 27% compared with 13.5% in patients without fibrosis (p=0.05). Median overall survival was decreased in the fibrosis group compared to the non-fibrosis group (29 months versus 42 months, p=0.02). A total of 36 patients (25 of whom progressed to AML) received azacitadine treatment (9 (24%) patients with fibrosis and 27 (17%) patients without fibrosis). Patients with fibrosis had a longer median survival than those without (29 months and 19 months respectively) but this difference was not statistically significant (p=0.48). Conclusion: Marrow fibrosis adversely affects overall survival in patients with MDS. Patients with fibrosis are more likely to present with higher marrow blast counts and to progress to AML. Patients with fibrosis who received azacitadine appeared to have an overall higher survival than those without fibrosis. However, the numbers of patients who received azacitadine were small. This study confirms the adverse prognostic influence of marrow fibrosis in patients with MDS, but the presence of fibrosis may not adversely affect the responsiveness to azacitadine therapy. Table IPSS-R Score - Fibrosis versus Non-Fibrosis Table. IPSS-R Score - Fibrosis versus Non-Fibrosis Figure Overall Survival (Non-Fibrosis versus Fibrosis) Figure. Overall Survival (Non-Fibrosis versus Fibrosis) Disclosures Desmond: Novartis: Honoraria.

Description: Therapy-related myelodysplastic syndromes and acute myeloid leukaemia (t-MDS/AML) occur as a late complication of cytotoxic chemotherapy and/or radiation for neoplastic and non-neoplastic disorders and are a major cause of non-relapse mortality. The prognosis of t-MDS/AML is poor because of an increased incidence of adverse cytogenetic features and relative resistance/intolerance to conventional therapies. TP53 is a tumour suppressor and cell cycle regulatory protein. Mutations in the TP53 gene are known to be associated with adverse outcomes in a variety of cancers including haematological malignancies. We examined p53 expression using immunohistochemical staining, as a surrogate marker for TP53 mutation, in marrow biopsies of 39 patients with t-MDS/AML from a single institution and correlated p53 expression with survival. Formalin fixed, paraffin embedded marrow trephine samples at diagnosis of t-MDS/AML were stained with DO-7 mouse p53 monoclonal antibody. Positive expression was defined as per Modified Quick Score. Of 39 patients, 35 had t-MDS and 4 had t-AML. 51% of patients had marrows at diagnosis that showed p53 positivity and 49% were negative. In a control group of 47 MDS patients in our centre, 22% were p53 positive and 78% were p53 negative. Of the t-MDS/AML group, age at diagnosis and gender distribution was similar in p53 positive and negative patients. Median latency (time from diagnosis of primary malignancy to diagnosis of t-MDS/AML) was similar between p53 positive and negative patients (66 vs. 52 months, p=0.51). A similar distribution was noted for the type of primary tumours (haematological vs. solid tumour) among the two groups, but all patients with more than one prior malignancy were noted to be p53 positive (N=4) whereas all patients who developed t-MDS after cytotoxic therapy for non-neoplastic disorders were p53 negative(N=3). There was no difference in p53 expression based on the type of primary therapy received (chemotherapy and/or radiotherapy) but a greater proportion (71%) of patients who received combined chemotherapy and radiation was p53 positive. A higher proportion of p53 negative patients had lower risk cytogenetics and normal karyotype whereas in p53 positive group there was lower incidence of good risk and a higher proportion to intermediate risk cytogenetics. However, equal numbers of patients with higher risk cytogenetics were found between p53 positive and negative groups. These findings were again reflected by IPSS-R risk groups (Table 1). Median overall survival was 10.5 months in p53 positive group compared to 22.5 months in p53 negative patients, p= 0.208. (Figure 1) In conclusion, our study demonstrated that a higher proportion of patients with t-MDS/AML were positive for p53 than in de novo MDS which may be related to poor outcomes in this group. p53 positive t-MDS/AML patients tend to have lower incidence of favourable karyotypes and showed poor survival compared to their counterparts who were p53 negative. Analysis of p53 expression by immunohistochemistry is a readily accessible, cost-effective method of assessment without the need for expensive gene sequencing and is a clinically useful prognostic tool. It may be particularly useful in patients with t-MDS/AML. Table 1. p53 expression in Therapy-related Myeloid Neoplasms p53 Positive (N=20, 51%) p53 Negative (N=19, 49%) Primary TumourHaematological (22) Solid Tumour (10) Autoimmune disease (3) More than one disorder (4) 45% 60% 0 100% 55% 40% 100% 0 Primary TherapyChemotherapy alone (27) Radiotherapy alone (5) Combined therapy (7) 44% 60% 71% 56% 40% 29% Cytogenetic RisksLow (very good, good) (14) Intermediate (8) High (poor, very poor) (12) 29% 75% 50% 71% 25% 50% IPSS-RLower (very low, low)(12) Intermediate (4) Higher (high, very high) (16) 33% 75% 50% 67% 25% 50% Disclosures No relevant conflicts of interest to declare.

Description: Introduction Myelodysplastic Syndromes are diverse group of bone marrow failure syndromes and current treatment options are guided by International Prognostic scoring systems (IPSS and IPSS-R) based on clinical phenotype and cytogenetics. Deletion of long-arm of chromosome 5 is the commonest cytogenetic abnormality and confers distinct biological and clinical implications. A specific subgroup of MDS patients with isolated Del(5q) was described as a separate entity in WHO 2008 classification owing to its unique clinical features, a low risk of leukemic transformation and a relatively good prognosis. Lenalidomide is proven to induce high erythroid response rates (frequently abolishing transfusion requirements) with a median response time of over 2 years in MDS patients with Del(5q). However, with time their anaemia worsens and most patients will become transfusion dependent with subsequent iron overload. Other treatment strategies were explored for those who failed Lenalidomide. Method This retrospective multi-centre analysis involved collecting data for MDS patients with 5q abnormalities (isolated Del(5q), Del(5q) with one additional cytogenetic abnormality, or Del(5q) within a complex karyotype) diagnosed between 2006 and 2014 in the Republic of Ireland. Six haematology units participated and the data of 47 patients were available for analysis. Results The median age of diagnosis was 72(29-91) with male-female ratio of 1:1.6. A range of WHO subgroups were identified and classic 5q Syndrome was documented in only 6.4% of patients. Cytogenetic results showed 47% isolated Del(5q), 17% Del(5q) plus one other abnormality and 36% Del(5q) in a complex karyotype. Patients with isolated Del(5q) or Del(5q) plus one other abnormality had similar haemoglobin, higher neutrophil and platelet count and lower marrow blasts than those with having a complex karyotype as previously described. IPSS-R scores were available in 37 patients as very low(7), low(8), intermediate(6), high(5) and very high(11). IPSS risks were available for 40 patients as low risk(12), intermediate-1(11), Intermediate-2(8) and high risk(9). Treatment options included red cell transfusion(85%), Recombinant Erythropoietin(45%), Lenalidomide(17%), Azacitidine(36%), Intensive chemotherapy(7%), transplantation(5%) and others(21%). All 8 patients in Lenalidomide group had either isolated Del(5q) or Del(5q) plus one other abnormality and received a median of 3 cycles (range 1-50). Responses were as follows: stable disease (3,38%), complete response (3,38%), and no response (2,25%). Three patients (38%) became transfusion independent with Lenalidomide. In 16 patients treated with Azacitidine, 50% had Del(5q) in a complex karyotype and 19% had failed Lenalidomide previously. A median of 10 cycles(range 1-18) were given. Four(25%) patients achieved a complete response, 1(6%) partial response, 1(6%) haematological improvement and 8(50%) stable disease. Response patterns were similar between the two groups with a trend towards improved survival in patients with isolated Del(5q) or Del(5q) plus one other abnormality compared to those with a complex karyotype treated with Azacitidine(23 vs. 15 months, p=0.0862). Transformation to AML was observed in 44% of patients without any difference between different cytogenetic groups. Median time to AML was 14 months. Median overall survival was 15 months(range