Litzow, Mark R.

e18560 Background: Myelodysplastic syndromes (MDS) are rare hematological neoplasms typically seen in the elderly. Young MDS (yMDS) patients ( < 50 years old) have been reported to comprise between 3-6% in the Surveillance, Epidemiology and End Results (Ma X et al, Cancer 2007; Rollison R et al, Blood 2008) with a better overall survival (OS). Methods: 1012 MDS patients from 1993 to 2015 were found after IRB approval was obtained. All cases had their bone marrow slides reviewed at our institution. yMDS Patients were included as cohort 1, while the rest as cohort 2. Survival estimates were calculated using Kaplan-Meier curves and univariate and multivariate analyses was based on log-rank testing using JMP software version 10. Results: We found 68 (7%) yMDS patients with a median age of 42 years (range, 18-49). Female gender was more common (43% vs 31%, p = 0.05) while platelets were lower in yMDS (61 vs 102, p < 0.0001). Therapy related MDS (t-MDS) was more frequent (33% vs 17%, p = 0.005), as was transformation to acute myeloid leukemia (AML) in yMDS (28 % vs 11%, p = 0.0004). Allogenic hematopoietic cell transplantation (HCT) was more frequent in yMDS (42% vs 4%, p < 0.0001). Survival outcome: Median OS was longer for cohort 1 vs cohort 2 but did not reach statistical significance (43 vs 21 months, p = 0.1). Median progression free survival (PFS) was shorter for cohort 1 vs cohort 2 but also did not reach statistical significance (8 vs 12 months, p = 0.3). Median OS for cohort 1 based on R-IPSS was 44, 105, 40, 18, and 12 months for very low, low, intermediate, high and very high risk groups, respectively (p = 0.09). Median OS was shorter in t-MDS vs de novo MDS in yMDS (13 vs 47 months, p = 0.04). yMDS patients who transformed to AML had worse median OS (18 vs 93 months, p = 0.001). On multivariate analysis neither t-MDS nor R-IPSS had a statistically significant impact on OS. Conclusions: MDS is rarely diagnosed under the age of 50 with IPSS-R being less powerful in detecting mOS. Among yMDS patients, there was a higher proportion of t-MDS compared to older patients with subsequent higher rates of AML transformation and higher rates of allogeneic HCT. In our study, we did not find an improved OS for yMDS patients compared to older patients.

e18559 Background: Myelodysplastic syndromes (MDS) are a group of myeloid neoplasms defined by ineffective hematopoiesis, dysplastic morphologic features, and variable risks of progression to acute myeloid leukemia (AML). Refractory anemia with ring sideroblasts (RARS) is a subtype of MDS defined by < 5% basts and > 15% ring sideroblasts (per 2008 WHO). Because RARS is a lower risk disease which does not frequently require HMA therapy, very little is known about the effects of HMA on RARS. Methods: A total of 1325 MDS patients’ data from 1993-2016 at Mayo Clinic were reviewed after appropriate IRB approval was obtained. All cases had their bone marrow slides reviewed at our center. Patients were considered for our study if they received HMA for their RARS. Response was identified based on MDS IWG 2006 criteria. Prognostic factors were analyzed by univariate and multivariate analyses. Survival estimates were calculated using Kaplan-Meier curves. Results: 168 of the initial 1325 patient were RARS (14%); only 14 (8%) were treated with HMA. Median age was 72 years (range, 51-85); half were males, with median overall survival of 91.3 months. The median number of ring sideroblasts (RS) at diagnosis was 40% (range 10-85%). Of the 14 patients, 11 (79%) received azacitidine (AZA), 2 (14%) decitabine (DAC) and one received both (7%). The median number of cycles for AZA was 6.5 (range, 2-28) vs 6 cycles for DAC (range, 1-24). Of the 14 patients receiving HMA, only 3 responded (21%) achieving hematologic improvement. All 3 responders received AZA. 6 of the 14 RARS patients (43%) had their bone marrow biopsy repeated after HMA therapy, 3 of whom were among the responders to AZA therapy and 3 were non-responders. In 5 patients, ring sideroblasts decreased after HMA therapy by a median of 15% (range 0-80); they remained the same in 1 patient; one patient who had developed AML had a decrease in RS from 85% to 5%. Conclusions: RARS is a low-risk subtype of MDS that is infrequently treated with HMA (8% in our study). Response rate to HMA (21%) was inferior to known rates in MDS. HMA therapy seems to decrease the number of ring sideroblasts irrespective of response. RARS patients should be treated on clinical trials if available.

e18556 Background: RARS is a subtype of myelodysplastic syndromes (MDS) defined by < 5% blasts and ≥15% ring sideroblasts (WHO 2008). Hereditary hemochromatosis is a disorder characterized by dysregulations in iron absorption, largely associated with C282Y and H63D mutations of the HFE gene. Iron levels are elevated in both disorders and pathophysiologic correlations were suggested. HFE gene mutations were previously found higher in MDS compared to controls (50% vs 36%) ( Nearman et al, Am J Hematol 2007). Methods: A total of 168 RARS patients’ data from 1994 to 2015 at Mayo Clinic were reviewed after appropriate IRB approval was obtained. All cases had their bone marrow slides reviewed at our center. We searched patients’ records retrospectively to Identify those tested for HFE gene (C282Y, H62D, S65C) mutations, done inside or outside our institution. Survival estimates were calculated using Kaplan-Meier curves. Results: Out of the 168 RARS patients, only 17 (10%) were tested for HFE gene mutations. Out of the 17 tested, 11 (65%) were found to have mutations; 2 of which (18%) had homozygous H63D mutation, 1 patient (9%) had double heterozygous H63D and C282Y mutations, 5 (45%) had only one H36D heterozygous mutation vs 3 patients (27%) with only one C282Y heterozygous mutation. Only one patient was tested for the additional S65C mutation and it was not detected. H63D mutation was present in a total of 8 patients (73%) vs C282Y mutation which was present in 4 patients (36%). Bone marrow iron stores were increased in all 17 tested patients, except one who had decreased stores, this patient had one heterozygous C282Y mutation. Median overall survival (mOS) was 117 months in the HFE mutated patients vs 75 months in the non-mutated (p = 0.6). Conclusions: Our study found the HFE gene, when tested, to be mutated in higher frequencies among patients with RARS compared to that reported in the general population (65% vs 36%), with H63D mutation in almost three quarters of all mutated patients. Although it did not reach statistical significance, the longer survival observed among HFE mutated patients compared to the wild-type raises the question whether testing for HFE gene mutations among patients with MDS-RARS should be further explored.

e18514 Background: Standard induction therapy for acute myeloid leukemia (AML) includes “cytarabine and anthracycline”. Anthracyclines are known to cause acute, subacute and chronic cardiac complications ( Lefrak , EA et al, Cancer 1973:32(2):302-314; Singal, PK et al, NEJM 1998;339(13):900-905). It is recommended to do base line cardiac function assessment with echocardiogram prior to induction therapy containing anthracycline. This study was done to find the frequency of changes made in avoiding anthracycline usage following this evaluation. Methods: The Mayo clinic data base was interrogated to identify patients with newly diagnosed AML who underwent echocardiogram prior to anthracycline containing induction chemotherapy. Echocardiogram results, including left ventricular ejection fraction (LVEF), diastolic dysfunction, valvular disease, and any resultant treatment alterations were reviewed. Results: Three hundred twenty six patients were identified. The median (range) age was 61 (17-82) years and 62% were male. The median (range) LVEF was 65% (30-87), and 3 (0.9%) patients had LVEF < 50% (range 30-47%). Of these 3 patients, 1 had 4 vessel CABG a decade ago, his EF was 46% and treated with full dose idarubicin; 2 of them presented with blastic crisis, 1 had EF 46% and improved to 51% after 5 days and then given daunorubicin, the other one had an EF of 30% and treated with 50% dose reduced cytarabine alone. Sixty five (23%) patients had previous chemotherapy or radiation exposure, including 24 (7.4 %) with previous anthracycline, 14 (4.3%) had prior radiation to the chest wall. No patient with previous chemotherapy or radiation exposure had a reduced LVEF. Conclusions: Echocardiography prior to anthracycline therapy rarely detects abnormal LVEF. Screening echo may be helpful in those presented with leukemic blastic crisis or had previous exposure to chemo-radiation.

7058 Background: The 2016 WHO classification includes myelodysplastic/myeloproliferative neoplasms, unclassifiable (MDS/MPN-U), as an MDS/MPN overlap syndrome not meeting criteria for well-defined entities such as CMML. No standard prognostication or treatment guidelines exist for such patients. Methods: We retrospectively identified MDS/MPN-U cases from 1990-2016 through our myeloid malignancies database. All bone marrow reports were reviewed to ensure compliance with 2016 WHO criteria. Clinical & cytogenetic parameters at diagnosis were assessed & compared with treatment outcomes. Results: Eighty nine patients met study criteria, with a median age of 69 years (range: 37-93); 58 (65%) males. Median follow-up was 22.2 months (range: 0-172), with 41 (46%) deaths & 13 (15%) leukemic transformations. Median OS was 24.8 months (range: 0-172). 43 (53%) patients had an abnormal karyotype, with common abnormalities being trisomy 8 (12%), complex karyotype (9%) & del (20q) (6%). Given the fewer types of abnormalities identified, the IPSS cytogenetic stratification was more effective than IPSS-R, with risk categorization including; 45 good (55%), 20 intermediate (25%) & 16 high risk (20%) respectively (8 unavailable). On univariate analysis, increased age (p = 0.05), decreased hemoglobin (p = 0.02), higher ANC (p = 0.03), circulating immature myeloid cells (p = 0.02), higher LDH (p = 0.009), absence of bone marrow ring sideroblasts (p = 0.001) & higher risk (intermediate & high) IPSS cytogenetic categories (p = 0.01) adversely impacted OS. In a multivariate model that included the aforementioned variables, higher risk IPSS cytogenetics retained a negative prognostic impact (p = 0.04). 28 patients received a median of 6 cycles (range: 1-21) of hypomethylating agent therapy (HMA), with an overall response rate of 18% (CR-3, PR-2). All responders had an abnormal karyotype (p = 0.01). However, HMA did not affect either OS or LFS. Conclusions: Intermediate & high risk IPSS cytogenetic categories independently & adversely impact survival in WHO defined MDS/MPN-U patients. HMA use did not impact OS; however, patients with abnormal karyotypes were more likely to respond.

7003 Background: Gilteritinib, a highly selective FLT3/AXL inhibitor, has displayed antileukemic activity in FLT3 mutation-positive (FLT3mut+) relapsed/refractory (R/R) AML in the CHRYSALIS Phase I/II study (NCT02014558), specifically at doses ≥80 mg/d. This exploratory analysis assessed molecular response to gilteritinib in a CHRYSALIS subpopulation. Methods: Molecular response was assessed from bone marrow aspirates obtained at baseline and at ≥1 additional time point from FLT3mut+ patients (≥18 y) treated with 120 or 200 mg/d gilteritinib. These doses were identified due to their ability to induce consistent, potent FLT3 inhibition and high clinical response rates. FLT3-ITD and total FLT3 were quantified by NGS to assess molecular response. A Cox regression model of overall survival (OS) by Kaplan-Meier estimation established a FLT3-ITD:total FLT3 ratio (ITD signal ratio) of 10−2as the threshold for improved survival. Results: Of 147 FLT3-ITDmut+ patients who received gilteritinib 120 or 200 mg/d, 80 were included in this analysis. Composite response rate for these 80 patients was 55%. During response, 20 patients (25%) had an ITD signal ratio of ≤10−2. Of these 20 patients, 18 had a ratio of ≤10−3 (major molecular response [MMR]) and 13 had a ratio of ≤10−4 (minimal residual disease [MRD] negative). Median time to achieve minimum signal ratio was 54 days. Elimination of morphologic leukemia was observed in 80% of patients with ITD signal ratios <10−2. Patients who had a signal ratio ≤10−2, MMR, or were MRD negative had significantly longer median OS than those who did not (Table). Conclusions: Molecular responses to gilteritinib in FLT3-ITDmut+ R/R AML correlated with clinical response and improved OS. This is the first demonstration of molecular response to a FLT3 inhibitor in AML. These data suggest ITD signal ratio may predict durable clinical benefit of gilteritinib. Clinical trial information: NCT02014558. [Table: see text]

e18091 Background: Allogeneic HCT is a potentially curative option for certain patients with AML. However, in-hospital outcomes and cost haven't been described. We use the NIS, a nationally weighted sample of hospital discharges in the US, to examine these outcomes. Methods: Years 2002-2013 of the NIS were queried for adults with a primary diagnosis of AML undergoing HCT. Mortality, length of stay, hospital charges, and complications were analyzed and predictors of increased charges and mortality were determined with multivariate regression analysis. Results: From 2002-2013 19,965 adults with AML were admitted for HCT. Donor sources included peripheral blood stem cells (PBSCTs) (79%); bone marrow (BMTs) (15%); umbilical cord blood (6%). In hospital mortality was 8%. Major complications included stomatitis (36%), total parenteral nutrition (TPN) (17%), neutropenic fever (14%), bacteremia (12%), mechanical ventilation (6%), and dialysis (3%). Median charges per hospitalization increased (p<0.01) from $161,894 (2002) to $324,765 (2013); median length of stay (27 days) remained stable over time. Multivariable regressions are depicted (Table). Umbilical cord blood transplants were associated with increased mortality (OR 2.25, p<0.01) and charges ($69,105, p<0.01) relative to PBSCTs. Conclusions: This study demonstrates increasing cost of BMT on a national level as well as increased cost and mortality with umbilical cord transplants. [Table: see text]