ALBERT

Description: Abstract 460 Epstein-Barr virus (EBV)-associated post-transplant lymphoproliferative disease (PTLD) remains an important complication of allogeneic stem cell transplantation (alloSCT). Monitoring of EBV genomes in blood using quantitative PCR (EBV qPCR) coupled with pre-emptive administration of Rituximab in response to high-level EBV reactivation has emerged as a strategy to reduce mortality from PTLD. However, the effect of pre-transplant Rituximab therapy on the risk of EBV reactivation and survival post-alloSCT is unknown. This retrospective study examined 193 consecutive adult patients undergoing T cell depleted or cord blood alloSCT at University Hospital Birmingham, UK (UHB) and Nottingham University Hospital, UK (NUH) between May 2009 and April 2011. Median age at transplant was 54 years (range 16–73 years). Conditioning was reduced intensity in 84% and myeloablative in 16%. Stem cell source was matched unrelated donor in 70%, sibling in 24% and cord blood in 6%. T cell depletion was with in vivo Alemtuzumab in 89% and ATG in 6%. Patients were monitored by EBV qPCR whole blood assay, performed every 1–2 weeks post-transplant. EBV reactivation was defined as a single positive EBV qPCR result, whilst high-level EBV reactivation was defined according to institutional thresholds; 30,000 and 10,000 EBV genomes/ml for UHB and NUH respectively. All patients with high-level reactivation were pre-emptively treated with Rituximab. Median follow-up was 23 months (interquartile range [IQR] 18–30 months), with EBV qPCR testing for a median of 8 months (IQR 4–13 months) post-transplant. The cumulative incidence of EBV reactivation, adjusting for the competing risk of death, was 41% at 2 years post-transplant. Amongst those reactivating, the median time to EBV qPCR positivity was 120 days (IQR 77–198 days). High-level EBV reactivation was observed in 34/193 (18%) patients, accompanied by PTLD in 10 patients (4 biopsy-proven and 6 probable cases). Of patients developing high-level EBV reactivation, in 30/34 (88%) the interval from first EBV qPCR positivity to high-level reactivation was less than 4 weeks, with 15/34 (44%) exhibiting high-level reactivation at first qPCR positivity. In univariate analysis, significant predictors for EBV reactivation were older age (hazard ratio [HR] 1.02 per year; p=0.04), male sex (HR 1.75; p=0.03) and T depletion with ATG (HR 4.8; p

Description: Abstract 3088 Background: High dose chemotherapy followed by autologous stem cell rescue is superior to salvage chemotherapy alone in relapsed Hodgkin's Lymphoma (HL) and Non-Hodgkin's lymphoma (NHL). A commonly used regimen is BEAM (BCNU, Etoposide, Cytarabine and Melphalan). Unfortunately, BCNU is associated with potentially lethal pulmonary interstitial toxicity, in a dose dependent manner. Therefore, it is often omitted in conditioning regimens for patients who have pre-existing impaired lung function. We aimed to determine the outcomes of patients given this EAM (Etoposide, Cytarabine, Melphalan) schedule in our centre. Method: At our unit, from 2000–2010, 338 patients had been transplanted with BEAM conditioning and 23 patients with an EAM regimen. BEAM chemotherapy involved cumulative doses of BCNU at 300 mg/m2, etoposide at 800 mg/m2, cytarabine at 1600 mg/m2 and melphalan at 140 mg/m2. The EAM regimen was the same, except for the omission of BCNU. We wanted to compare overall survival, event free survival and transplant related mortality between the two groups of patients. A matched-pair analysis based on age and sex on a 1:1 basis was conducted. Survival rates were estimated by the Kaplan-Meier method. Differences between the survival distributions were compared with the log-rank test. Results: 23 patients transplanted with an EAM regimen was matched with an equal number transplanted with a BEAM conditioning regimen. The median age was 53 in both groups. The median follow-up for patients alive was 3.2 years. Further baseline characteristics of the two groups are described in table 1. Other than a slight predominance of more advanced disease in the BEAM cohort, the two groups were evenly matched. Survival estimates were statistically significantly poorer for the EAM group compared to the BEAM group. The median overall survival for the EAM cohort was 29 months compared to 77 months (p=0.03) for the matched BEAM cohort. The median event free survival for the EAM cohort was 11 months compared to 63 months (p=0.02) for the BEAM cohort. The 100 day transplant related mortality was 8.7% for both cohorts. During the follow-up period 11 patients died of disease related causes in the EAM group compared to 6 in the BEAM group. Conclusion: Despite the prevalence of pre-existing lung disease in the EAM group of patients, the conditioning was reasonably well tolerated. Our analysis suggests that patient transplanted with an EAM regimen had an inferior survival outcome to patients treated with the standard BEAM regimen. This data provides evidence for the importance of BCNU in lymphoma control; consideration should be given to a reduction in BCNU dosage, rather than simple omission, for patients with borderline lung function results. Collaboration with other centres is warranted to determine if this experience with EAM is replicated elsewhere. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 2361 Introduction: Reduced intensity allogeneic transplants represent a potentially curative therapy in older patients with haematological malignancies. However the upper age limit for transplantation using a sibling or unrelated donor is unclear and few studies have addressed this important issue. In the UK in vivo T cell depletion utilising alemtuzumab is commonly used as a strategy to reduce the risk of acute and chronic graft-versus-host disease (GVHD) but this manoeuvre impairs immune reconstitution and may pose a particular problem in older patients. Aim: We analysed the outcome of patients over the age of 60 after an alemtuzumab based reduced intensity allograft from five UK Transplant Centres with the aim of identifying factors determining long term outcome and also factors affecting the duration of inpatient admission during the first 100 days post transplant. A modified Charlson's comorbidity index score (Sorror modified HCT comorbidity index) was applied to analyse the transplant outcomes of these patients according to the presence of transplant co-morbidities. Patients and Methods: We have studied the outcome of 161 patients (89 male, 72 female) after an alemtuzumab conditioned reduced allograft allograft for a haematological malignancy. The median age of the patients was 62 years(range 60–72). 115 patients had a transplant for myeloid malignancies (87 acute myeloid leukaemia, 21 myelodysplastic syndrome, 2 chronic myeloid leukaemia, 5 myelofibrosis) and 46 for lymphoid malignancies (18 Non_Hodgkin's lymphoma, 14 chronic lymphocytic leukaemia, 4 T-prolymphocytic leukaemia, 7 multiple myeloma and 3 acute lymphoblastic leukaemia). 93 patients received an unrelated donor transplant and 68 had a sibling transplant. The great majority of patients were transplanted using a Fludarabine/Melphalan conditioning regimen(n=118) whilst 13 received a combination of fludarabine and busulphan, 21 BEAM (BCNU, cytarabine, etoposide, melphalan) and 9 a combination of BEAM and fludarabine. The median follow up was 19.1 months (range 2–94 months). Results: The one year overall survival for the whole group was 52% and the predicted two year OS 46%. The transplant related mortality (TRM) was 19% in the first 100 days post transplant and 23% in the first year post transplant. 40 patients (25%) relapsed. 25 patients (21%) developed Grade III-IV acute GvHD and 16 (10%) patients developed chronic extensive GvHD. There was a weakly negative correlation between Age and Bed Days(p:0.05843) but the correlation between high comorbidity index (3 or greater than 3) and increased bed days is significant(P:0.0013). Transplant outcomes were affected by Sorror modified comorbidity index for HCT.A score of 3 and above was significantly associated with decreased overall survival (P:0.001) and interestingly with decreased disease free survival (P:0.02). CMV status and stem cell dose did not have any impact on overall survival and disease free survival and CR1 at transplantation was showed a trend towards increased overall survival(P:0.05). Conclusions: Reduced intensity alemtuzumab based stem cell transplant can be delivered safely in patients above the age of 60. It appears that a Sorror comorbidity score of 3 and above has a negative impact on overall survival and disease free survival of these patients and significantly increases inpatient days.These observations require confirmation in a larger cohort of patients but suggest that the selection of patients above the age of 60 for a reduced intensity transplants will be facilitated by incorporation of the Sorror HCI comorbidity index into a transplant algorithm. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction Graft-versus-Host Disease (GvHD) is one of the leading causes of mortality and morbidity following allogeneic stem cell transplant. In vivo T cell depletion by alemtuzumab as part of the transplant conditioning is an effective strategy to reduce the risk of GvHD. While it is recognised that the overall incidence of GvHD is reduced by alemtuzumab, the incidence of chronic GvHD as defined by the National Institute of Health (NIH) consensus criteria, the impact on outcome, and the pattern of organ involvement have not been defined yet in this transplant setting. Methods Consecutive patients (n = 323) undergoing allogeneic stem cell transplantation at the Queen Elizabeth Hospital, Birmingham, between January 1 2008 and June 30 2012 were reviewed in this retrospective, single centre study. Medical records were examined and data regarding the development of GvHD were collected; NIH consensus criteria for diagnosis and staging of chronic GvHD were stringently applied. Clinical characteristics of GvHD occurring in patients transplanted following T cell depletion by alemtuzumab administration (n=248) were compared with those of patients transplanted with a T cell replete graft (n=75). Patients receiving alemtuzumab were mainly treated with reduced-intensity conditioning protocols, while patients in the no-T-cell depletion group were mainly treated with a myeloablative, sibling transplant. Results After a median follow up of 38.4 months, the cumulative incidence (CI) of grade II-IV classic acute GvHD was 35% and 48% for patients transplanted respectively with or without T cell depletion by alemtuzumab (p= 0.041, Figure 1); with a CI of grade III-IV classic acute GvHD of 13% and 27% (p=0.007). The 2-years CI of grade II-IV late acute GvHD was not significantly different in the two groups (20% and 23% for patients respectively treated with or without alemtuzumab, p=0.589, Figure 2). T cell depletion by alemtuzumab significantly reduces the 3 years cumulative incidence of classic chronic GvHD (5% versus 31%, p

Description: Acute and chronic graft versus host disease (GVHD) represent major causes of morbidity and mortality following reduced intensity unrelated donor allografts, particularly in older patients. Alemtuzumab (AL, humanised anti-CD52 antibody) is highly effective at reducing the incidence of both forms of GVHD. There are a number of reported schedules used in this setting, with doses of between 50-100mg AL administered over 2-5 days. The timing of AL administration relative to stem cell return is also likely to have a significant impact on the depletion of donor T cells, but comparative data for unrelated donor transplants are lacking and the optimal dose and schedule remain unknown. In this three-centre retrospective study, we compared the post-transplant outcomes of patients receiving the same chemotherapeutic backbone of fludarabine 150mg/m2 and melphalan 140mg/m2 (flu-mel) along with the originally described AL regimen (100mg administered as 20mg/day, day -7 to -3, n=79), with those of patients receiving flu-mel and one of two alternative AL schedules: 60mg administered as 30mg/day, day -4 and -2, n=72; or 50mg administered as 10mg/day, day -7 to -3, n=107. In keeping with prior pharmacokinetic studies of de-escalating schedules in sibling donor cohorts, the day 1 serum AL levels in the 60mg cohort (median 4.7 mcg/ml, range 1.75-8.02 mcg/ml, n=14), did not differ significantly from historical data for the 100mg schedule. Donor lymphocyte infusions were given for mixed chimerism in all centres according to similar schedules starting at 6 months post transplantation. Median age at transplantation differed significantly between cohorts at 52, 45 and 56 years for the 100mg, 60mg and 50mg cohorts respectively (p = 0.005). Other statistically significant differences between cohorts were CMV status (p = 0.006) and disease indication (p = 0.0005); with AML patients constituting a majority in the 50mg cohort and lymphoproliferative disorder patients a majority in the 100mg cohort. No significant differences were detected between cohorts for HLA matching status (mismatched in 21/79 [27%], 12/72 [17%] and 33/107 [31%] respectively, p = 0.11), graft type (p = 0.06), or disease status at transplantation (p = 0.09). The incidences of acute grade II-IV GVHD were 35%, 32%, and 38% (p = 0.6) for the 100mg, 60mg and 50mg cohorts respectively, whilst those for grade III-IV GVHD were 4%, 6%, and 15% (p = 0.02). The corresponding incidences of chronic GVHD at 2 years were 29%, 28% and 39% respectively (p = 0.38). A trend towards a higher incidence of chronic GVHD in the 50mg cohort was confirmed by comparison with the combined data for the other 2 cohorts (29% vs 39%, p = 0.17). In the high-risk CMV cohort (seropositive recipient), reactivation requiring anti-viral therapy occurred in 97%, 77% and 76% respectively (p = 0.02). The incidences of full donor T-cell chimerism (〉97% donor according to assay sensitivity) at day 100 were 56%, 29% and 44% respectively (p = 0.02). There were no significant differences with respect to non-relapse-related mortality at two years post-transplant (28%, 27% and 29% respectively, p = 0.57). Although univariate analyses of relapse rates and overall survival were confounded by the differences in patient characteristics between cohorts, no significant differences were observed (relapse rate at 3 years 20%, 20%, and 26%, p = 0.39; overall survival at 3 years 59%, 64% and 37%, p = 0.05). In conclusion, this study represents the first attempt to define an optimal dosing schedule for AL in unrelated donor allograft recipients, and suggests that significant reduction in AL dose is achievable for patients receiving flu-mel conditioning with unrelated donor grafts. There was evidence of more acute grade III-IV GVHD with the 10mg/day x5 schedule, and a trend towards a higher incidence of chronic GVHD, whilst the 30mg/day x2 schedule delivered similar results to the 20mg/day x5 schedule, perhaps in keeping with the comparable day 1 serum AL levels. Lower alemtuzumab doses were associated with a modest reduction in the risk of CMV reactivation in high-risk patients. Whilst the usual caveats associated with retrospective cross-institutional comparative studies apply, the data suggest that reduction to 30mg/day x2 is not associated with any significant detrimental impact and may be preferable on economic grounds. Disclosures: No relevant conflicts of interest to declare.

Description: Aims Limitations in donor availability require that ABO blood group incompatible donors are used for hematopoetic stem cell transplantation (HSCT). Previous studies on the influence of ABO mismatch on the outcomes of allogeneic HSCT show conflicting results, and suggest that the nature of peri-transplant immunosuppressive therapy may influence the impact of ABO mismatch. This study examined the impact of ABO blood group mismatch on transplant outcomes and transfusion requirements in a large UK HSCT cohort following reduced intensity conditioning (RIC) with alemtuzumab. Methods Consecutive patients from three UK transplant centres (Birmingham, Glasgow, Oxford) undergoing first alemtuzumab-based RIC HSCT between 2000 and 2010 were included for analysis. Overall survival (OS), relapse-free survival (RFS), relapse incidence (RI), non-relapse mortality (NRM), incidence of acute and chronic graft-versus-host disease (GvHD) and red blood cell (RBC) and platelet (PLT) transfusion requirements were analyzed retrospectively according to the degree of ABO mismatch. ABO mismatch was defined as minor if anti-recipient isohemagglutinins were present in the donor, major if anti-donor isohemagglutinins were present in the recipient, and bidirectional if both anti-donor and anti-recipient isohemagglutinins were present. Results 598 patients underwent alemtuzumab-based RIC HSCT between 2000 and 2010. The median recipient age was 51 (range 17-74) and 342/598 (57%) were male. 361/598 (60%) received stem cells from an HLA- matched unrelated donor. Patients were HLA matched to at least 6/6 in 569/598 (95%), 225/598 (38%) were matched to 10/10. The main indications for HSCT were acute myeloid leukaemia (194/598, 32%), non-Hodgkin’s lymphoma (125/598, 21%) and myelodysplastic syndrome (76/598, 13%). Stem cells were sourced from peripheral blood in 548/598 (92%) and bone marrow in 50/598 (8%). The most frequently used conditioning regimen was fludarabine, melphalan and alemtuzumab in 459/598 (77%). Donor and recipient ABO blood groups were matched (M) in 318/598 (53%). Minor (MN) mismatch was present in 113/598 (19%), major (MJ) mismatch in 128/598 (21%), and bidirectional (BD) mismatch in 35/598 (6%). OS & RFS at 2 years did not significantly differ according to the degree of ABO mismatch (OS 51.9% M vs. 54.8% MN vs. 54.7% MJ vs. 38.6% BD; RFS 41.4% M vs. 42.5% MN vs. 50.2% MJ vs. 29.9% BD, p〉0.05 for all), although recipients of a bidirectionally mismatched transplant showed a trend to reduced OS (p=0.078). 2 year RI & NRM was not significantly different by ABO mismatch (RI 29.6% M vs. 28.5% MN vs. 27.2% MJ vs. 29.4% BD; NRM 26.5% M vs. 26.1% MN vs 20.7% MJ vs 29.4% BD, p〉0.05 for all). On multivariate analysis, ABO mismatch was not a significant predictor of OS, RFS, NRM or RI. The degree of ABO mismatch did not influence the incidence of grade 2-4 acute GvHD (incidence 19.6% M vs. 30.6% MN vs. 26.8% MJ vs. 31.9% BD). The incidence of extensive chronic GvHD, however, was significantly higher in patients with minor and major ABO mismatch on both univariate and multivariate analysis (14.1% M vs. 21.7% MN vs. 22.1% MJ vs. 14.2% BD; HR 1.74, p=0.032 for MN vs M, HR 1.69, p=0.0036 for MJ vs M). The proportion of patients requiring ≥1 unit red blood cell (RBC) transfusion support did not differ significantly by ABO match (91% M, 94% MN, 93% MJ, 85% BD, p 〉 0.05 when compared to M). A lower proportion of bidirectionally mismatched patients required ≥1 unit of platelets than the matched cohort (94% M, 78% BD, p=0.008 for BD vs. M, 91% MJ, 94% MN, p 〉 0.05 when compared to M). For those requiring transfusion, the median number of RBC and PLT units transfused to day 100 were not significantly different by ABO match (median (range) of RBC: 6 (2-66) M, 7 (1-43) MN, 7 (2-38) MJ, 6 (2-32) BD, p=0.157; median (range) of PLT: 4 (1-92) M, 4 (1-70) MN, 4 (1-46) MJ, 3.5 (1-33) BD; p=0.45). On multiple regression analysis, bidirectional mismatch was associated with fewer RBC transfusions (p=0.034). There was no significant effect of ABO match on PLT requirements. Conclusion This is the largest series to date examining the effects of ABO matching in alemtuzumab-based RIC HSCT. Our data indicate that ABO mismatch in this setting has no clinically significant effect on transplant survival outcomes or on transfusion requirements, but may influence the incidence of extensive chronic GvHD. Disclosures: No relevant conflicts of interest to declare.

Description: Allogeneic stem cell transplantation (SCT) performed using a reduced intensity conditioning (RIC) regimen represents an important new treatment modality in older patients with high risk acute myeloid leukaemia (AML) whose outlook with conventional chemotherapy would be poor. However assessment of its role has been hampered by the absence of long term follow-up data. Furthermore factors determining survival after RIC SCT have not been rigorously defined. In order to examine the anti-leukemic activity of RIC allografts in more detail we have examined the outcome of 170 patients with AML transplanted using a uniform conditioning regimen over a ten year period. Long term follow-up data (maximum 119 months) was collected on patients transplanted using a conditioning regimen consisting of fludarabine (30 mg/m2 x 5 days), melphalan 140 mg/m2 x 1 day and alemtuzemab (10 mg x 5 days). The median age of transplanted patients was 54 years (range 18–71). 88 patients were in CR1 at the time of transplant, 63 CR2/3 and 19 had relapsed or refractory disease. Cytogenetic information was available on 149 patients and of these 33 patients had adverse risk and 116 intermediate risk cytogenetics by MRC criteria. 83 transplants were performed using an HLA identical sibling donor and 87 using a volunteer unrelated donor. The 100 day transplant related mortality was 9%. 29% of patients developed Grade II-IV acute GVHD and 22% chronic GVHD. The 3 year overall survival (OS) for the whole group was 48% and 3 year disease free survival (DFS) 45%. 20 patients remain in remission more than five years post-transplant. Survival was significantly influenced by status at transplant (p=0.01), patient age (p= 0.01) and presentation cytogenetics (p=0.05) as determined by multivariate Cox proportional hazards regression. The 3 yr OS for patients transplanted in CR1 or CR2 was 51% and 52% respectively compared to 13% for patients with relapsed/refractory disease. Three year OS for patients with intermediate risk cytogenetics was 52% compared with 34% for adverse risk patients. Three year OS for patients under 60 years was 51% compared with 36% for older patients. This study demonstrates the ability of RIC allografts to deliver encouraging long term disease free survival rates in high risk AML. Pre-transplant characteristics can be used to predict outcome after a RIC allograft and older patients with active disease at the time of transplant or adverse cytogenetics require novel strategies in order to improve long term survival.

Description: The ability of reduced intensity conditioned (RIC) allografts to deliver long term disease free survival is dependent on the genesis of an immunologically mediated graftversus tumor (GVT) effect. Post-transplant immunosuppression plays an important role in limiting graft-versus-host disease (GVHD) but also modulates a GVT effect. Although cyclosporine A (CsA) is the most commonly utilised immunosuppressive agent in patients allografted using a RIC regimen the impact of CsA dose intensity on disease relapse and overall survival (OS) has not been studied. We have therefore measured CsA exposure in the first 21 days post-transplant and correlated individualised CsA area under the curve (AUC) values with OS and disease relapse in patients undergoing a RIC allograft. 132 patients with a diagnosis of acute myeloid leukemia (AML) (n=41), myelodysplasia (MDS) (n=17), Non-Hodgkin’s lymphoma (NHL)(n=51) or Hodgkin’s disease (n=23) were transplanted using an alemtuzumab containing RIC regimen. All patients with a myeloid malignancy (AML or MDS) were transplanted using a regimen consisting of fludarabine, melphalan and alemtuzumab (FMA). Patients with an underlying lymphoid disease (NHL or HD) were transplanted using FMA (n=31) or a regimen consisting of BCNU, etoposide, cytosine arabinoside, melphalan and alemtuzumab (BEAMA) (n=43). 39 patients had chemoresistant disease at the time of transplant. All patients received intravenous CsA at a loading dose of 5 mg/kg on day -1 followed by 2.5 mg/kg b.i.d. Patients were switched to oral CsA prior to discharge. Trough CsA levels were measured thrice weekly for the first three weeks after stem cell infusion and the dose of CsA adjusted to achieve levels in the region of 200–300 mg/l during this period. Trough levels obtained during the first 21 days post-transplant were used to calculate the CsA AUC for each patient. The median age of patients studied was 48 years (range 17–68). 71 patients were transplanted from HLA identical siblings and 61 from volunteer unrelated donors. The incidence of acute GVHD (Grades 2–4) was 34%. The median CsA AUC was 3682 mg.hr/l (range 2162- 8084 mg.hr/l). In univariate analysis the presence of chemoresistant disease at the time of transplant and a high CsA AUC were both associated with a decreased OS. In multivariate analyses chemoresistant disease (HR=2.60, 95% CI 1.44–4.65, p=0.002), increased age (HR=1.04, 95% CI 1.01–1.07, p=0.002) and linearly increasing CsA AUC were associated with an increased hazard of death (HR=1.1, 95%CI 1.02–1.24, p=0.02). The two year OS for patients with a CsA AUC less than 3682 mg.hr/l was 77% compared to 30% for patients with CsA of 3682 mg.hr/l and over (p

Description: Abstract 4098 Relapse after allogeneic hematopoietic cell transplantation (HCT) is usually incurable for acute myeloid leukemic (AML) patients. Residual disease (MRD) monitoring pre- and post-HCT may improve relapse prediction, allowing the targeted implementation of post-HCT interventions to at risk patients when disease burden is sufficiently low for these to be effective. Previous studies have shown that MRD detection by multiparameter flow cytometry (MRD MFC) at either pre- (Leung et al 2012, Walter et al 2011) or post- transplant (Yan et al 2012) timepoints is prognostic for myeloablative HCT outcome in AML. Quantification of hematopoietic populations enriched for leukemic stem cells such as CD34+CD38low or lymphoid-primed multi-potential progenitor-like (LMPP-like) (Lin-CD34+CD38lowCD90-CD45RA+) (Goardon et al 2011) may improve the specificity of MFC MRD assays. In this study we retrospectively evaluated the predictive value of MRD MFC at both pre- and post- HCT timepoints in a cohort of unselected AML/high risk myelodysplasia (MDS) patients (n=44) who underwent reduced intensity conditioning (RI n= 32, median age 59, range 34–70) or myeloablative (MA n=12, median age 28, range 19–47) HCT between June 2010 and November 2011 (Table 1). MFC MRD was assessed both by detection of standard leukemic- aberrant-immunophenotypes (LAIPs) (identified at presentation and/or relapse) and quantification of CD34+CD38low and LMPP-like progenitors (LSC-enriched progenitors, LSC-EP). Pre HCT, 37 patients (MA = 11, RI = 26) were assessable for MFC MRD. 15 (41%) were MRD positive (LAIP MRD+) and 47% (7/15) (MA = 37.5%, 3/8; RI = 57%, 4/7) of these relapsed post HCT compared to 8% (MA = 0%, RI = 9%) of MRD negative patients (LAIP-MRD-), (Fig 1 p 0.03). Post HCT, 34 patients (MA = 9, RI =25) were assessable for MFC MRD. 10 (37%) had detectable LAIP MRD positivity between 2 and 9 months post HCT. 80% of these relapsed (MA = 60%; RI = 100%) with a median disease free survival (DFS) post HCT of 5 months (MA = 4 months; RI = 5 months); there were no relapses in the 17 patients who remained LAIP MRD- at a median follow-up of 20 months (range 9–26). (p=0.0006, Figure 2a). Presence of MRD post-transplant was associated with significantly poorer overall survival (p=0.005). Although 1 patient with high MRD (in CR, LAIP 〉1%, LSC-EP +) pre HCT relapsed 1 month (median time to relapse from MRD detection of 1.5 months, range 1–6; OS, median 4 months, range 1 - not reached). CD34+CD38low progenitors (34+38low) were 〈 0.03% of bone marrow nucleated cells in the majority of patients. Detectable 34+38low were mainly CD45RA+ so in most cases correlated with LMPP-like quantitation. Pre HCT, 34+38low were detectable in 40% of patients who went on to relapse and in only 9% of those who have not yet relapsed. Post HCT, 34+38lowpositivity preceded frank relapse by ≥1 month in 60% of patients who relapsed. Only 6% of patients who have not yet relapsed had detectable 34+38low. LSC-LEP positivity appears prognostic for DFS and OS (Figure 2b) but for a lower frequency of relapses compared to LAIP MRD positivity (60% v 80%). Conclusions: These data suggest that post HCT MFC detection of LAIP MRD is predictive of relapse in RI as well as MA HCT. LSC-LEP quantitation may be prognostic in a subset of patients. Pre HCT MRD might be more predictive of relapse in RI than MA HCT. However, post HCT MRD positivity precedes most clinical relapses by a time window which may be sufficient for interventions such as azacytidine or donor lymphocyte infusion (DLI) when disease burden is still low. These results provide a basis for the use of MFC residual disease detection pre and post HCT to inform treatment decisions in reduced intensity as well as myeloablative HCT. Disclosures: No relevant conflicts of interest to declare.

Description: Automatically identifying chemical and drug names in scientific publications advances information access for this important class of entities in a variety of biomedical disciplines by enabling improved retrieval and linkage to related concepts. While current methods for tagging chemical entities were developed for the article title and abstract, their performance in the full article text is substantially lower. However, the full text frequently contains more detailed chemical information, such as the properties of chemical compounds, their biological effects and interactions with diseases, genes and other chemicals. We therefore present the NLM-Chem corpus, a full-text resource to support the development and evaluation of automated chemical entity taggers. The NLM-Chem corpus consists of 150 full-text articles, doubly annotated by ten expert NLM indexers, with ~5000 unique chemical name annotations, mapped to ~2000 MeSH identifiers. We also describe a substantially improved chemical entity tagger, with automated annotations for all of PubMed and PMC freely accessible through the PubTator web-based interface and API. The NLM-Chem corpus is freely available.