ALBERT

Beschreibung: Results to date argue compellingly that disruption of FA/BRCA gene expression plays a pivotal role in human somatic carcinogenesis. Melphalan, a DNA cross-linker, is one of the most widely used and effective drugs in the treatment of multiple myeloma (MM). Although most patients respond to standard and high dose melphalan, eventually patients acquire resistance and develop progressive disease. In 1991, our laboratory reported that acquired resistance in a human myeloma cell line was associated with reduced DNA crosslinks, elevated glutathione levels, and increased radiation survival (Cancer Res. 5:993; 1991). Most recently, we reported that the melphalan-resistant myeloma cell lines, 8226/LR5 and U266/LR6, showed a significant increase in several FA/BRCA genes compared to drug-sensitive cells, and that enhanced interstrand crosslink (ICL) repair via this signaling pathway contributes to acquired drug resistance in melphalan resistant cell lines (Blood 10:698; 2005). Here, we report that IKKa is constitutively phosphorylated in unstimulated 8226/LR5 cells, but not in melphalan-sensitive control cells. The specific phosphorylation of IKKa leads to an increase in basal NF-kB DNA binding activity, and 8226/LR5 cells are found to be markedly sensitive to BMS-345541 (a highly selective inhibitor of IkB) relative to control cells. Importantly, a cytotoxic dose of BMS-345541 induces a dramatic decrease in FA/BRCA gene expression, and a concomitant inhibition of NF-kB DNA binding activity in both 8226/S and 8226/LR5 cells. Furthermore, we show that 8226/LR5 cells experience the highest degree of direct binding between FANCD2 promoter and NF-kB/Rel family members, which, in turn, leads to an increase in basal FANCD2-specific NF-kB activity. Small-interfering RNA (siRNA)-mediated depletion of RelB and p50, but not other NF-kB subunits, in 8226 cells results in impaired NF-kB binding activity, and visible decrease in FANCD2 protein expression. Studies designed to dissect the role of NF-kB in acquired melphalan resistance are in progress, and the results will be presented. Our findings suggest that NF-kB functions as a regulator of FA/BRCA expression, and that this pathway represents a new target for preventing acquired drug resistance in myeloma patients.

Beschreibung: Usual cytoreductive therapy for bone marrow transplantation (BMT) for thalassemia major consists of busulfan (Bu) 14–16mg/kg and cyclophosphamide(Cy) 160–200mg/kg +/− anti-thymocyte globulin (ATG). In high risk patients, this is associated with significant regimen related toxicity and graft rejection. Fludarabine (Flu) containing conditioning has been used effectively by adding immunosuppression in RIC with reduced doses of myelosuppressive drugs to lower regimen related toxicities and graft rejection. There is also data to suggest that a 24-hour gap between Bu and Cy doses could reduce toxicities associated with this combination. We therefore treated 25 patients (median age: 7 years, range: 3–14), 12 males and 13 females, with beta thalassemia major with a regime of Flu 150mg/m2 (day-15 to -11), Bu 14mg/kg (day-10 to -7) and Cy 160mg/kg (day-5 to day-2). Most of these patients were heavily transfused (median red cell units:100, range: 21–250) and poorly chelated (median serum ferritin: 2510 ng/ml, range: 1039–6740). Their risk stratification (Lucarelli class) was as follows: Class I: 4 (16%), Class II: 8 (32%) Class III: 13 (52%). Allogeneic BMT was performed using a 6/6 matched related donor and bone marrow as the graft. Graft versus host disease (GVHD) prophylaxis consisted of cyclosporine A and mini-methotrexate. The median cell dose was 4.56 × 108 TNC/kg (range: 2.5–14.1). Three (20%) of 20 assessable patients developed grade I-II acute GVHD, 2 (10%) each had hemorrhagic cystitis and veno-occlusive disease. Mortality within the first 100 days were due to graft failure/rejection in 4 patients (16%) and one each of diffuse alveolar hemorrhage, intracranial hemorrhage and sepsis (4% each). With a mean follow-up of 8 months (range 0–11), 18 (72%) of 25 patients are alive. Among the 13 patients in class III, 5 (38.5%) rejected the graft (3 had primary failure and 2 after initial engraftment). In a historical cohort of 47 patients conditioned with Bu16, Cy200 and ATG, the rejection rate was only 8%. The regimen was very well tolerated by all patients and patients in class I had no rejection. Pharmacokinetic data was available on Bu and Cy from 14 of the 25 patients treated with this protocol. The mean values for Bu kinetics were: Cmax-1 (ng/ml): 1069.2±265.8, Cmin-1(ng/ml): 186.8±73.2, Cl-F-1 (l/h/kg): 0.281±0.081, AUC-1 (ng*h/ml): 3648±918 and Css-1 (ng/ml): 608±153. They are not significantly different from patients conditioned Bu16mg/kg without Flu (data not shown). The Cy kinetic data are as follows: Cmax-1(ng/ml): 441.4±188, AUC-1(ng*h/ml): 1431±791, Cl-F-1(l/h/kg): 0.0378±0.027. The Cmax-1 and AUC-1 were significantly lower compared to patients receiving 200mg/kg and dosed without a 24-hour gap between the Bu and Cy. Overall, these data suggest that Flu does not provide adequate immunosuppression for sustained engraftment after allogeneic BMT even when combined with Bu 14 and Cy160 (with a 24-hour gap between Bu and Cy) in high risk patients with thalassemia major (Lucarelli class III) but could be useful in the others (Class I and II). OUTCOME OF ALLOGENEIC BMT CLASS NUMBER (%) OVER ALL SURVIVAL (months) EVENT FREE SURVIVAL (months) REJECTION (%) MORTALITY (%) ALL PATIENTS 25 69±9.8 62.6±10.0 6 (24) 7 (28) CLASS I 4 (16%) 75±21.7 75±21.7 0 1 (25) CLASS II 8 (32%) 83.3±15.2 83.3±15.2 1 (12.5) 1 (12.5) CLASS III 13 (52%) 57.7±14.7 44.9±14.1 5(38.5) 5 (38.5)

Beschreibung: Abstract 3486 Poster Board III-423 Haemophilia B (HB), an X linked inherited disorder is caused by heterogeneous mutations in the F9 gene. Approximately 3% of hemophilia B patients have major deletions in the F9 gene. Gross and small deletions in the F9 gene in HB affected males are easily detected by PCR but detecting the carrier state of females in the family is challenging due to the presence of the normal allele. Different methods like linkage analysis, real time PCR and MLPA have been used to assess the carrier status in this situation. Linkage analysis is limited by the availability of informative markers and adequate number of family members. Real time PCR involves standardisation and preparation of calibration curves for each run. Although MLPA is a better alternative, it can be time consuming and involves multiple steps. We have therefore developed a fluorescent PCR based gene dosage approach which is simple, rapid and cost-effective for determining the carrier status of females in families with deletions in the F9 gene. 200ng of DNA extracted by standard protocols was used for amplification with primers designed to amplify a 160bp product from exon h in the F9 gene. One of the primers was fluorescently labelled. Amplification was carried out using these primers for 20 cycles only and the amplified product was subjected to capillary electrophoresis on an ABI 310 genetic analyser. A 230 bp fragment in the albumin gene was used as the control. Analysis was done using Genescan and Genotyper software. The ratio between the peak heights of the exon h in the F9 and control genes in the patient samples were normalised to a normal control. Five families with deletional HB were analysed (in toto deletion-1; Ex g-h – 1; Ex g-poly A-1; Ex h-poly A-2). The ratios in the probands and the family members are presented in the table. Out of eight females analysed, 6 were carriers and 2 were normal. The mean ratio in the carriers was 0.49±0.08 and 0.75±0.05 in the normal. Deletions are not uncommon in HB and deletions involving the exon g and h constitute a major group. Among 212 families with HB assessed at our centre, we have identified large deletions in 8 families (3.7%). It is interesting to note that all except one of these deletional mutations involved exon h. This method confirmed the presence of these deletions in the males and helped us to identify the carrier status of the females in the family. Identification of carrier status of females with deletions in F9 gene by gene dosage Subject ID Peak height of Exh in F9 Peak height of albumin Normalised Ratio Interpretation HB5 284 442 0.59 Carrier HB6 305 489 0.57 Carrier HB22 188 372 0.47 Carrier HB63 85 165 0.48 Carrier HB129 247 295 0.78 Normal HB238 94 326 0.4 Carrier HB280 372 679 0.77 Normal HB384 202 670 0.4 Carrier Disclosures: No relevant conflicts of interest to declare.

Beschreibung: Abstract 3349 Poster Board III-237 Busulfan in combination with cyclophosphamide is the commonly used conditioning regimen for hematopoietic stem cell transplantation (HSCT). We have studied the pharmacokinetics (PK) of oral busulfan in children with beta thalassaemia major undergoing BMT and observed up to 12 fold variation in PK. Since PK parameters of Bu have been shown to be influencing BMT outcome and regimen related toxicity, we attempted to understand factors influencing Bu PK using a population PK based model. The aim of this study was to explore the effect of several demographic, biological, and pharmacogenetic covariates on the disposition of busulfan in children with beta thalassaemia major undergoing hematopoietic stem cell transplant. One hundred and fifty children with beta thalassaemia major received 16 mg/kg/day of busulfan as part of the pre-BMT conditioning regimen. Plasma busulfan concentrations were measured after first and 13th doses using HPLC based method with UV detection. Non-linear mixed effects modeling analysis was performed with Monolix (version 2.4, www.monolix.org) to evaluate the effect of age, body weight, sex, Lucarelli class, serum ferritin levels, pre-transplant liver function and ten polymorphisms corresponding to GSTA1, GSTM1, GSTP1, GSTT1, CYP2B6, CYP2C9, CYP2C19, CYP3A4 (pharmacogenetic data was available for 133 of 150 patients). A one-compartment pharmacokinetic model with first-order oral absorption was used to describe the data. The pharmacokinetic parameters estimated included apparent clearance and volume (CL/F (L/hr or L/hr/kg), V/F (L or L/kg)), and the absorption parameter (ka (1/hr)). The bioavailability, F, was not identifiable since only oral drug was used. The distribution of the parameters was assumed log-normal. The main covariate which explained the largest portion of the inter-individual variation in busulfan kinetic parameters (45%, 22%, and 15% of clearance (CL), volume of distribution (V), and absorption rate constant (ka), respectively) was body weight. The next most significant covariate was GSTA1 promoter polymorphism. In particular, clearance decreased 124% and 148% and the oral absorption increased 16% and 15% in GSTA1 heterozygous and homozygous patients, respectively, compared to wild type. GSTA1 explained an additional 17% and 1% of inter-individual variation in the CL and ka, respectively, compared to the weight normalized model. By combining these two covariates, the interindividual variability on busulfan CL/F decreased from 45% to17%. We have developed a population pharmacokinetic model of oral busulfan in children with beta thalassaemia major undergoing BMT which explains the inter-individual variation in Bu PK, considering demographic and biological covariates. Disclosures: No relevant conflicts of interest to declare.

Beschreibung: Abstract 1194 Poster Board I-216 Between January 2001 and June 2009, 120 patients with aplastic anemia underwent HLA identical sibling or family donor transplants using a combination of Fludarabine 150-180 mg/m2 over 5 days and Cyclophosphamide 120 mg/kg over 2 days as the conditioning regimen. Antithymocyte globulin (ATG) 10 mg/kg x 4 days in addition was used in 34 patients. Graft versus host disease (GVHD) prophylaxis consisted of Cyclosporine with low dose Methotrexate. Graft source included peripheral blood stem cells (PBSC) in 108 and G-CSF stimulated bone marrow in 12. Seventy patients (58.3%) were considered as high risk (presence of fever/infection at time of HSCT or 〉20 transfusions prior to HSCT or failed previous immunosuppressive therapy). There were 79 males and 41 females with a median age of 22 years (range: 2 - 51) including 36 children (age 0.5 × 109/L) was 12 days (range: 7-19) while platelet engraftment (Platelet count 〉 20 × 109/L) occurred at a median of 13 days (range: 0 -30). Acute GVHD occurred in 38 patients (33.3%) with grade III-IV GVHD in 13.1%. Acute GVHD was not significantly lower in patients where ATG was used in conditioning (21.8% with ATG vs 38.2% without ATG; p = 0.129). Nine patients (7.5%) had veno-occlusive disease of the liver while 11 (9.1%) had hemorrhagic cystitis; all responded well to supportive therapy. Bacterial infections were documented in 28% of transplant recipients while fungal infections (both probable and definite) occurred in 23%. CMV reactivation was seen in

Beschreibung: Dendritic cells (DC) are antigen-presenting cells involved in induction and regulation of immune responses. We investigated the impact of the number of infused and engrafted (day 28) dendritic cells, DC1 (lin−HLA-DR+CD11c+) and DC2 (lin−HLA-DR+CD123+) on the development of acute and chronic graft-versus-host disease (GVHD). Thirty three consecutive patients with hematological malignancies who underwent HLA-matched related G-CSF mobilized allogeneic PBSCT were included in the analysis. The mean follow up was 293 days (range: 36–417). There were 20 males and 13 females (median age: 29 years, range: 15–55). Conditioning regimen was myeloablative in 25 (Bu/Cy=13; Cy/TBI = 12) and non myeloablative in 8 patients (Flu/Mel = 7; Flu/Cy = 1). All patients received cyclosporine and short course methotrexate as GVHD prophylaxis. Three of them received steroids before day 28 for treatment of GVHD. Ten patients developed acute GVHD (grade II–IV) and 11 patients had chronic GVHD. The median DC2 count in the peripheral blood on day 28 was significantly lower among patients who developed acute GVHD as compared to those who did not and there was a trend to a lower count among patients with chronic GVHD [Table 1]. Based on the day 28 DC2 count patients were divided into a low DC2 quartile (2.3cell/ul, n=25). The two groups were comparable with regard to patient and graft characteristics (CD34, CD133, CD3, CD4, CD4CD45RO, CD4CD45RA, CD8, CD8CD45RO, CD8CD45RA, CD19, NK, DC1and DC2 cell dose). Patients in the low DC2 quartile group had higher probability of developing acute (P=0.000) and chronic GVHD (P= 0.022). Cox regression analysis revealed that low day 28 DC2 counts is associated with higher incidence of acute GVHD (HR=11.4; 95%CI=2.9–44.9; P=0.001) and chronic GVHD (HR=7.07; 95%CI=1.9–26.6; P=0.004). In a multivariate analysis which included other standard risk factors such as female to male transplants, conditioning regimen, CD34 and CD3 cell dose, low day 28 DC2 was found to be a risk factor for acute GVHD (HR=21.1; 95%CI=3–149.9; P=0.002) together with patient age (HR=1.1; 95%CI=1–1.3; P=0.039). DC1, DC2 counts in the graft and day 28 DC1 count did not correlate with the development of acute or chronic GVHD. Excluding patients who had developed acute GVHD before day 28 (n=3), a low day 28 DC2 count (

Beschreibung: The efficacy of single agent arsenic trioxide (ATO) in the management of relapsed and newly diagnosed patients with acute promyelocytic leukaemia (APL) has been established. The impact of FLT3 activating mutations and additional cytogenetic changes in newly diagnosed patients with APL treated with single agent ATO has never been reported. Between January 1998 and April 2005, 98 newly diagnosed cases of PML-RARα positive APL were treated with a regimen of single agent ATO at our centre. FLT3 activating mutations were seen in 33% (FLT3-ITD - 21.3%, D835V- 11.7%). Of the 69 patients evaluated for additional cytogenetic findings, 76.8% had an isolated t(15;17) while 23.2% had t(15;17) along with additional cytogenetic findings. The presence of a FLT3 mutation was significantly associated with a higher mean white cell count (18.11±23.74 vs. 8.04±21.44, P=0.024) and a bcr3 PML-RARα isoform (48.4% vs. 20.6%, P=0.006). There was also a significant delay in achieving a molecular remission (MR) among patients with a FLT3 mutation as noted by the number of patients who were in MR prior to onset of consolidation therapy (60.9% vs. 85.7%, P=0.031). There was no significant difference in the hematological remission rate (CR), time to CR or early mortality between the group with and without a FLT3 mutation. At a median follow up of 20 months (range: 4 – 97), the 3 year Kaplan-Meier estimate of OS, EFS and DFS for patients with a FLT3 mutation was 82.51±7.24, 77.01±8.6, 88.82±7.48 percent while for those without a FLT3 mutation it was 92.06±3.4, 80.45±5.86, 89.02±5.31 percent. Statistical analysis of these survival curves by a log rank test did not reveal any significant difference between the two groups [Figure 1]. The presence of additional cytogenetic changes was not associated with any significant differences in the clinical or laboratory parameters at presentation. It also did not have an impact on the CR, MR, EFS, DFS or OS. Neither FLT3 activating mutations nor additional cytogenetic changes have a significant impact on the outcome of newly diagnosed patients with APL treated with single agent ATO in the short term. Whether either of these two parameters will affect long term outcome remains to be seen. Figure 1: Kaplan-Meier product limit estimate of EFS based on FLT3 mutation status (n=94). Figure 1:. Kaplan-Meier product limit estimate of EFS based on FLT3 mutation status (n=94).

Beschreibung: The current risk stratification of patients with β thalassemia major undergoing an allogeneic stem cell transplantation (SCT) is based on liver size (〉2cm), presence of liver fibrosis and inadequate iron chelation (Lucarelli et al, NEJM 1990). Our clinical observation suggested that patients in Class III (presence of all three adverse features) were a heterogeneous group and included a large number of patients who would otherwise have a good prognosis. We therefore undertook a retrospective analysis to study the pre-transplant variables that have an impact on outcome. Between 1991 and 2005, 189 patients underwent 196 HLA matched related allogeneic SCT for a diagnosis of β thalassemia major at our center. Except for two cases, all patients were less than 18 years of age at the time of transplant. The majority (97.5%) of patients received myeloablative (BuCy) conditioning regimen. The median (±SD) age of this cohort was 7±4.1 years with 68% males. There were 11(5.6%), 81(41.1%) and 105(53.3%) in Lucarelli Class I, II and III respectively. The Kaplan-Meier 5 year event free survival (event defined as rejection, relapse or death) for Class II and III patients was 78.53±4.53 and 51.97±5.14, respectively. (Table 1) summarizes the impact of pre transplant variables on the EFS. Patient age and liver size as continuous variables were significantly associated with an adverse outcome. Using a receiver operating characteristic (ROC) curve plot analysis, cutoff values of 7 years and 5 cms respectively for age and liver size gave the highest likelihood ratios for an adverse effect on EFS (1.6 and 2.7 respectively). These cut off values significantly discriminated patients’ EFS on a univariate analysis. Table 1: Unadjusted adverse effect of pre-transplant variables on EFS Pre-transplant variable RR (95% CI) P-value Age (≥ 7 years) 2.9 (1.6– 5.2) 0.000 Sex (F) 1.5 (0.9 – 2.6) 0.082 F〉M transplant 0.9 (0.5 – 1.5) 0.715 Liver size (≥5 cm) 3.5 (2.1 – 5.9) 0.000 Chelation (inadequate) 2.9 (0.7 – 12.2) 0.130 Liver fibrosis (yes) 1.7 (0.8 – 3.3) 0.106 SGPT 1.0 (1 – 1.006) 0.080 Ferritin 1 (0.8 – 1.2) 0.056 On a forward stepwise multivariate analysis only age ≥7 years and liver size ≥ 5 cms retained their significance (RR 2.2 and 3.6, P-values 0.014 and 0.000 respectively). Using these two variables patients were categorized as high risk if they were ≥ 7 years and had a liver size ≥ 5 cms. There were 41 cases in this sub group (all were Class III). The 5 year EFS and OS in this high risk group (n=41) was 23.93±6.88 and 39.01±7.96 respectively, while in the remaining Class III patients (n=64) the 5 year EFS and OS was 73.23±5.56 and 81.22±4.89. Statistical analysis of these survival curves by a log rank test revealed that they were both statistically significant (P=0.000 for both EFS and OS). The majority of the events in the high risk group happened in the first 100 days [TRM=17(41.4%), rejection=3(7.3%) and death from GVHD=3(7.3%)]. Using age ≥ 7years and liver size ≥ 5 cms we were able to identify a significant subset of patients in class III (39%) who have a poor outcome with allogeneic SCT and could benefit from novel approaches while the others with clinical outcomes comparable to those in class II should probably be classified with them and managed accordingly.

Beschreibung: Abstract 2798 Poster Board II-774 Multiple Myeloma (MM) is an incurable malignancy of mature clonal B cells. The refractory nature of MM has long been attributed to the acquisition of drug resistance. Traditionally, mechanisms of drug resistance have been defined by acquired changes in the expression or function of specific gene products. To this end, we have recently demonstrated that selected resistance to the cytotoxic agent melphalan correlated with increased expression of components of the Fanconi Anemia (FA)/BRCA DNA repair pathway and a concomitant increase in repair of DNA interstrand cross-links (ICLs).(Hazlehurst et al Cancer Res 2003; Chen et al Blood 2005) Further, the exogenous expression of specific FANC components in RPMI 8226 cell lines enhanced ICL repair, favored the release from melphalan-induced S-phase delay, and rendered these cells partially resistant to melphalan treatment. Together, these results suggest a causal relationship between increased expression of FA DNA repair components, increased DNA repair, and acquired resistance to melphalan. Over the past decade a large body of evidence has emerged demonstrating that in addition to drug resistance mechanisms intrinsic to the cancer cell, there exist dynamic, de novo mechanisms coordinated by the tumor microenvironment resulting in an environment-mediated drug resistance (EM-DR). As such, we examined the potential role of the microenvironment in regulating the FA/BRCA DNA repair pathway. FA pathway protein expression was evaluated with anti-sera to FANCD1/BRCA2, FANCC, FANCD2, FANCI, FANCG and BRCA1 in drug sensitive RPMI 8226 cells and melphalan resistant 8226/LR5 cells in co-culture with the HS-5 bone marrow stromal cell line. With these preliminary results we present three novel findings. First, we demonstrate that expression of FA/BRCA pathway components is regulated by intracellular interactions in both MM cells and bone marrow stromal cells (BMSCs). Second, we show that the acquisition of drug resistance alters FANC protein expression profiles upon co-culture. Third, in the HS-5 BMSCs, mono-ubiquitinated FANCD2 is observed in the absence of detectable FANCG. In RPMI 8226 cells, Western blot analysis demonstrated an acute (within 30minutes) and prolonged (up to 48hours) time-dependent increase in expression of FANCD2/BRCA2, FANCC, FANCD2, and BRCA1 upon incubation with BMSCs relative to MM cells incubated alone. However, no appreciated increases in FANCI or FANCG were noted under the same conditions. Incubation of 8226/LR5s with HS-5 BMSCs demonstrated a slightly different up-regulation of FA/BRCA pathway protein expression with addition of increased FANCI expression and no increase in FANCD2 or FANCC expression. We also examined FANC protein expression in the HS-5 cells. Interestingly, in the BMSCs significant differences were noted in the FANC expression profiles. Co-culture of RPMI 8226 cells with HS-5 cells demonstrated only modest elevations in FANCD2; however, co-culture with drug-resistant 8226/LR5s resulted in increased levels of FANCD2, FANCI and BRCA1. These data indicate that different tumor cells may alternately influence FA/BRCA-mediated DNA repair and potentially drug resistance in juxtaposed bone marrow stroma. Curiously, we also observed mono-ubiquitinated FANCD2 in the absence of any detectable levels of FANCG protein under co-culture conditions. As the FA/BRCA DNA repair pathway has been associated with cell cycle progression, we evaluated cell cycle kinetics under the co-culture conditions. The results of BrdU analysis demonstrated that the observed changes in FA/BRCA protein expression in MM and BMSC could not be fully explained by cell-cycle distribution. Therefore, within this report we demonstrate for the first time that microenvironmental interactions can modulate the FA/BRCA DNA repair pathway in MM and BMSCs. These results suggest that the FA/BRCA DNA damage repair pathway may be an important modulatory component of EM-DR. Importantly, the potential de novo drug resistance likely involves both the MM tumor cell and adjacent stromal cells. Current and future studies will attempt to examine a causal relationship between increased FANC expression and melphalan (and other drug) resistance seen in co-culture conditions, as well as to identify specific signaling molecules and mechanisms controlling the enhanced expression in both cell models. Disclosures: No relevant conflicts of interest to declare.

Beschreibung: Abstract 846 We had previously reported a well tolerated regimen using single agent arsenic trioxide (ATO) (Blood 2006:107; 2627) leading to durable remissions in patients with newly diagnosed acute promyelocytic leukemia (APL). Briefly, the regimen consisted of ATO (10mg/day for adults and 0.15mg/kg/day for pediatric patients) for up to 60 days in induction; this was followed by a 28 day consolidation after a 4 week break. Four weeks after completion of consolidation, patients received ATO for 10 days/month for 6 months. A concern with the previous report was the relatively short duration of follow up. Here we report the long term follow-up data of the same cohort. As previously reported, 72 newly diagnosed cases of APL were enrolled. 62 patients (86.1%) achieved hematological remission. The remaining died prior to achieving remission. There were no primary induction failures. Twenty two (30.6%) of these patients were considered good risk group (WBC count at diagnosis 20×109/L), the rest were considered high risk. Since publication of the last report an additional 7 patients have relapsed to give a total of 13 relapses, 2 were in the good risk group and the remaining 11 in the high risk group. The relapses in the good risk group were salvaged with an autologous SCT and have durable continued second remissions. The median time to relapse was 1.5 years. Five (38.52%) of these relapses occurred beyond 2 years and included both relapses in the good risk group. At a median follow-up of 58 months the 5-year Kaplan-Meier overall survival (OS), event free survival (EFS) and disease free survival (DFS) of the entire cohort was 74.22±5.26%, 68.93±5.52% and 80.00±5.17% respectively. The 5-year OS and EFS of the good risk and high risk group was 100±00% vs. 63.30±6.9% and 90.00±6.71% vs. 59.66±6.99% respectively. Beyond induction, all deaths followed relapse of disease. There were no second malignancies reported. Besides the previously reported toxicities, which were mild and transient in most cases, there were no new toxicities that were reported on continued follow up of these cases. Since completion of therapy, in spite of counseling and advising against pregnancy, 3 males and 4 females in the reproductive age group have had 8 normal children. No abortions, still births or fetal defects were reported among patients in the reproductive age group in this cohort. Hair and nail samples from 5 cases that had completed maintenance therapy more than 24 months earlier have been collected for analysis, the results of which are awaited. At our center the cost of administering this regimen is a quarter of that of a conventional ATRA plus anthracycline based regimen. Additionally, after the initial induction therapy the rest of the treatment did not require hospital admission nor did it result in any Grade III/IV hematological toxicity. Single agent ATO based regimen as reported previously is well tolerated, results in durable remissions and does not have any significant late side effects. In the good risk group it is associated with excellent clinical outcomes while in the high risk group additional interventions are probably required to reduce the risk of late relapses. In a resource constrained environment it is probably the best option. Disclosures: No relevant conflicts of interest to declare.