Frank, Ryan D.

Background. The assessment of perioperative risk factors for the development of acute respiratory distress syndrome (ARDS) has been described in various surgical populations. However, there are only limited data among patients undergoing liver transplantation (LT), particularly regarding the influence of intraoperative ventilation parameters. We sought to identify the perioperative risk factors associated with the development of ARDS in LT recipients. Methods. This is a single-center, retrospective cohort study of adult patients who underwent LT at a tertiary academic medical center between January 1, 2006, and January 31, 2016. Postoperative ARDS was identified using the Berlin definition. Multivariable logistic regression analysis was used to identify perioperative risk factors for ARDS. Results. Of 817 eligible patients who underwent an LT during the study period, 20 (2.45%) developed postoperative ARDS. In the preoperative model, ongoing dialysis (odds ratio, 6.41; P < 0.01) was identified as an independent risk factor of ARDS post-LT. A higher mean peak inspiratory pressure per increase of 1 cm H2O (odds ratio, 1.31; P < 0.01) was the only independent risk factor in the intraoperative model. Patients who developed ARDS postoperatively had significantly greater intensive care unit and hospital stay compared to non-ARDS patients (P < 0.001). There were no significant differences in the 30-day (P = 0.16) and 1-year (P = 0.51) mortality between the groups. Conclusions. Dialysis at the time of transplant and elevated intraoperative mean peak inspiratory pressure were associated with the development of ARDS. ARDS post LT was associated with increased intensive care unit and hospital length of stay, but not increased mortality.

Abstract Background: Disparities in MM survival among different racial groups persist despite improvements in overall survival (OS). We report a comprehensive review of the SEER Medicare database to elucidate the causes of differences in disease presentation, management practices, cost of care and their relationships to outcomes among the MM patients (pts) from different racial groups. Methods: Pts diagnosed with MM between 1991-2010 with continuous Medicare coverage (1 year prior to diagnosis-date of death/end of 2012) were included. MM complications (MM-C); hypercalcemia (C), renal dysfunction (R), anemia (An), bone fractures (B), dialysis (D) occurring prior to/within 30 days of MM diagnosis (at diagnosis; AT-D), or any time later (after diagnosis; AF-D) were summarized. Demographics, survival (PEDSAF files) and disease characteristics, drug/stem cell transplant (SCT) utilization (NCH, OUTSAF, PDESAF files) were obtained. Trends in incidence of MM-C, treatment type over time (proportional odds) and their association with race (logistic regressions) were compared. Associations between race, MM-C AT-D and MM related death using Fine and Gray's competing risk methods were analyzed. Medicare claims adjusted for inflation (2013) within the first 6 months (mth)/total MM care were summed by drug and total charges. Median cost of care by race and year of diagnosis were evaluated by Kruskal-Wallis (univariate) and proportional odds models (multivariate). Results: A total of 35842 pts (52.1% males) with median age of 76 yr (range 18-104) were included. Mutually exclusive racial subgroups were: White (W, 72%), Hispanic (H: 7.3%), African-American (AA, 16.2%), Asian (A, 4.1%), other (0.3%). A significantly increasing trend in the prevalence of all MM-C AT-D was seen over time (all p<0.001). For AF-D, all MM-C initially increased but decreased from 2006-2010 except D that showed continued increase (Fig 1A). For race, MM-C were significantly different, with AA having the most C, R, An and D while W having the most B (Fig 1B). Inferior OS was seen with male gender (p=0.006), increasing age, earlier years of diagnosis, belonging to W or H race, C, R and B (all p<0.001), but not with An (p=0.298) or D (p=0.378). Trends in OS over time showed a significant interaction with race in 1991-1995 (p<0.001), 1996-2000 (p=0.004) and 2001-2005 (p<0.001) but not in 2006-2010 (p=0.182), possibly due to shorter follow up. Medicare claims analysis showed that pts receiving any treatment increased from 12.8% in 1991-1995 to 59.3% in 2006-2010 (p<0.001). Specific therapy utilization, including SCT and all drug classes/agents evaluated increased except for cytotoxic chemotherapy utilization (melphalan/cyclophosphamide), which decreased (Fig 2A). Any therapy utilization was significantly different by race (p<0.001) with highest for H (36.5%) and lowest for AA (31.6%). Specific treatments differed significantly by race (Fig 2B). Cost of care analysis showed significant increase over years for median drug claims (0.23k to 3.63k) and total claims (5.11k to 27.8k) during first 6 mth as well as for any time after MM diagnosis (drug: 1.3k to 9.9k, total: 21.9k to 84k) (p<0.001 for all). Median drug claims in first 6 mth were highest among AA (1.3k) and lowest among A (0.63k) (p<0.001). Same was true for total claims in first 6 mth (p<0.001). For any time after diagnosis, H had the highest drug (p<0.001) and total (p=0.081) claims. There were significant interactions between year of diagnosis, race and cost of care with H seeing the highest increase in drug claims in first 6 mth (p<0.001) and any time (p=0.002). Conclusion: MM-C have generally decreased AF-D due to improved therapeutics although AA continue to have the highest rates of MM-C. Our data shows that utilization and cost of MM care have increased tremendously with resulting improved outcomes, but significant racial disparities persist. These disparities may be in part due to impaired access to new agents from lack of supplemental insurance for certain racial groups. Drug utilization and cost of care disparities may also be due to differences in co-morbidities, MM-C or access leading to differential healthcare utilization and should be explored in order to ameliorate disparities in outcomes. Disclosures Ailawadhi: Pharmacyclics: Consultancy; Novartis: Consultancy; Amgen Inc: Consultancy; Takeda Oncology: Consultancy.

Abstract Background: Outcomes in multiple myeloma (MM) have improved significantly over the years but disparities in outcomes exist among MM patients (pts) belonging to different racial subgroups. Previous studies have explored healthcare access and utilization disparities for pts of different races as potential causes, but genomic differences, which could alter disease biology and clinical behavior, have not been studied. We utilized the MMRF CoMMpass database, which prospectively captures clinical and molecular characteristics of MM patients to explore race-based differences at the genomic level. Methods: The publically available CoMMpass trial database (data release version IA8) was utilized. Pts with race category of White (total n=698) and non-White (total n=223, comprising 129 Blacks, 19 Asians and 45 others) were analyzed for genomics including cytogenetics (fluorescent in situ hybridization; FISH), tumor diploidy (conventional karyotyping), gene expression profiling (GEP), gene copy number variations (CNV), gene single nucleotide variations (SNV) and genomic translocations analysis focused on genes from the FoundationOne Heme panel¨ (402 genes). GEP was considered significantly different between categories for more than or equal to a 2-fold change for a given gene (p<0.05). Categorical variables for FISH probes and diploidy status were compared by Chi-square or Fischer exact test, wherever applicable. Results: GEP clustering analysis uncovered 67 genes differentially expressed between Whites vs. non-Whites (Fig 1A). Baseline cytogenetics aberrations between both race cohorts showed statistically insignificant differences. Multiple cytogenetic abnormalities were seen most commonly in both race groups (~31%) with the distribution of others (in descending order) as such: Del13 > No abnormalities > t(11;14) > 1q amplification > Del17p or p53 > t(4;14) > Del1p > t(14;16) > t(14;20). Aneuploidy status showed insignificant differences between race groups with ~34% of patients from both groups having no aneuploidy. SNV data was present for 90% of Whites and 57% of non-Whites. Overall, 84% of genes had>1 SNV (84%), 2% translocations only and 14% had neither SNVs or translocations. SNVs were classified as Non-specific Non-Synonymous coding (NNSCs), Splice-site (SS) or Tier 1 (T1, codon deletion, insertion, frame-shift, etc.). A total of 880 NNSCs SNVs were observed: 670 in Whites across 227 genes, 210 in non-Whites across 120 genes. In whites, 57% of NNSC SNVs consisted of transition-type mutations and 43% as transversions. In non-Whites, transition mutations comprised 64% of all NNSC SNVs and transversions in the remaining 36%. A total of 38 SS SNVs were observed, 30 in Whites and 8 in non-Whites observed in 25 and 6 genes, respectively. In Whites, 57% of SS SNVs were transition type and remaining 43% were transversion. In non-Whites, an equal % of SS transition and transversion mutations were noted. A total of 240 T1 SNVs were noted; 182 in Whites seen in 82 genes; 58 in non-Whites seen 6 genes. In both Whites and non-Whites, frameshifts (41% in Whites, 47% in non-Whites) and stop-gained (46% in Whites, 45% in non-Whites) SNVs were most common. A total of 17 genes with>10 distinct SNVs were identified in Whites; 5 genes in non-Whites and 4 genes commonly seen in both cohorts (Fig 1B). In Whites, 17 genes were found to have translocations (intrachromosomal TRAF3 and PIK3R2 most common); in non-Whites, 5 genes had translocations (intrachromosomal CIC and EMSY most common). CNV analysis demonstrated 21 genes that were significantly altered with either copy number gain or loss between Whites vs. non-Whites (Fig 1C). Conclusions: We present the first race-based comprehensive genomic analysis utilizing the public interface of the CoMMpass trial. We discovered genomic differences between Whites and non-Whites at several levels including GEP, SNV and CNV, but not at the level of cytogenetics, which are in fact the most commonly performed clinical genomic analysis. Correlation of these variations with MM pt outcomes would be invaluable in understanding disease biology and hopefully mitigating some of the outcome disparities by pt race. As the database grows and becomes more enriched, separate analyses for Blacks, Asians and Hispanics would be attempted. Acknowledgments: We would like to thank the MMRF, Ms. Mary Derome and Dr. Jonathan Keats for providing scientific overview and technical support. Disclosures Fonseca: Millennium, a Takeda Company: Consultancy; Millennium, a Takeda Company: Consultancy; AMGEN: Consultancy; AMGEN: Consultancy; Patent: Patents & Royalties: Prognostication of MM based on genetic categorization of FISH of the disease; Patent: Patents & Royalties: Prognostication of MM based on genetic categorization of FISH of the disease; Patent Pending: Patents & Royalties: The use of calcium isotopes as biomarkers for bone metabolisms; Patent Pending: Patents & Royalties: The use of calcium isotopes as biomarkers for bone metabolisms; Celgene: Consultancy; BMS: Consultancy; Bayer: Consultancy; Novartis: Consultancy; Sanofi: Consultancy; Celgene: Consultancy; Janssen: Consultancy; Millennium, a Takeda Company: Consultancy; BMS: Consultancy; AMGEN: Consultancy; Bayer: Consultancy; Patent: Patents & Royalties: Prognostication of MM based on genetic categorization of FISH of the disease; Novartis: Consultancy; Patent Pending: Patents & Royalties: The use of calcium isotopes as biomarkers for bone metabolisms; Sanofi: Consultancy; Janssen: Consultancy; Millennium, a Takeda Company: Consultancy; AMGEN: Consultancy; Patent: Patents & Royalties: Prognostication of MM based on genetic categorization of FISH of the disease; Patent Pending: Patents & Royalties: The use of calcium isotopes as biomarkers for bone metabolisms. Ailawadhi:Takeda Oncology: Consultancy; Amgen Inc: Consultancy; Novartis: Consultancy; Pharmacyclics: Consultancy.

Abstract Introduction: Ibritumomab tiuxetan (Zevalin¨) is a radioimmunoconjugate agent that targets CD20-expressing B cell lymphomas. Currently, it is the only radioimmunoconjugate agent available on the market and is FDA approved for relapsed/refractory follicular lymphoma (FL). A retrospective study of patients with low grade FL (WHO grade 1 and 2) treated with ibritumomab tiuxetan at Mayo Clinic Florida from 2000 to 2016 was performed. Methods: The patient population consisted of previously untreated as well as relapsed low grade FL. We evaluated patient characteristics, response rate, duration of response, progression free survival, overall survival and time to next treatment. Previously untreated patients were treated with ibritumomab tiuxetan due to ineligibility for standard chemotherapy and prolonged immunotherapy, or their preference. Comparison between previously untreated FL patients treated with ibritumomab tiuxetan as the first line of treatment versus patients with relapsed FL treated with ibritumomab tiuxetan as second, third or fourth line therapy was made. Results: Low grade FL patients (n=41) that received ibritumomab tiuxetan were identified; 13 patients (32%) with previously untreated FL and 28 patients (68%) with relapsed FL. The data on response to treatment is presented in Table 1. Ibritumomab tiuxetan use in previously untreated FL was associated with more complete responses than in the relapse setting (92% vs 50% at 6 months, p=0.036). Previously untreated FL patients that received ibritumomab tiuxetan as first line therapy had better progression free survival (80% vs 35% at 3 years). Ibritumomab tiuxetan use in relapsed FL was associated with significant shorter time to relapse compared to previously untreated FL (median 1.3 years vs 4.8 years, p=0.017, figure 1) and shorter time to next treatment (65% at 4 years compared to 0% in previously untreated, p=0.007). 15% of newly diagnosed FL treated with ibritumomab tiuxetan relapsed compared to 64% of relapsed FL patient. There was no significant difference in overall survival between the two groups. The incidence of grades 3-4 neutropenia (Absolute Neutrophil Count <1000 x109/L), Anemia (Hb <8 g/dL) and Thrombocytopenia (Platelets <25 x109/L) were 47%, 7.7% and 48.7% respectively. Only 6/41 patients (15%) had cytopenias that required transfusion of blood products. More significant cytopenias were seen in relapsed group compared to previously untreated group. Conclusions: Ibritumomab tiuxetan as first line treatment in previously untreated FL patients at our institution was associated with improved response rate, better progression free survival and shorter time to next treatment when compared to its use in relapsed FL. Ibritumomab tiuxetan has significant therapeutic efficacy for both previously untreated as well as relapsed low grade FL. It was well tolerated without significant treatment-related morbidity or mortality. Moreover, treatment schedule was very convenient for the patients. Our results suggest that treatment with ibritumomab tiuxetan is an effective option for those previously untreated low grade FL patients who cannot take standard chemoimmunotherapy. Disclosures No relevant conflicts of interest to declare.

Abstract Background: Cost of cancer care is projected to reach $173 billion by 2020, a 39% increase from 2010. Several factors including psychiatric (psych) comorbidities contribute to this increase. Within the oncology setting, 29-38% of the patients (pts) are reported to have mood disorders and 15% have major depression. Depression alone is associated with increased healthcare utilization in pts with breast, colon, lung and prostate cancers. A 2015 report noted that the presence of at least one psychiatric comorbidity in 300 Leukemia pts was associated with an extra $55,000 per pt in just one year. Similarly, in pts treated with systemic steroids, the incidence of neuropsychiatric disorders can be as high as 75%. However, no such data is available for MM, where more than 90% of pts are treated with steroids, likely increasing risk for mood problems and impacting treatment cost. As such, the aim of our study was to analyze the SEER-Medicare database for healthcare utilization trends and acute cost of care (cost incurred during 6 months after MM diagnosis) in MM pts with or without psych comorbidities. Methods: Pts diagnosed with MM between 1991-2010 with continuous Medicare coverage (1 year prior to diagnosis-date of death/end of 2012) were included. Pts were categorized as: MM with any psych disorder (MM+P), MM with depression (MM+D) and MM only. Presence of ³1 inpatient (ipt) or ³2 outpatient (opt) ICD9 diagnosis codes were used to assign pts to the psych categories. Within MM+P and MM+D groups were subdivided by presence of psych or depression diagnosis prior to MM (MM+P PRE or MM+D PRE). Medicare claims adjusted for inflation (2013) within the first 6 months (mth)/total MM care were summed by drug and total charges. Univariate and multivariate logistic regression models (adjusted for age, year, sex, race, and the Charlson Comorbidity Index; CCI) were performed to determine associations with ipt, opt, and any emergency department (ED) charges after MM diagnosis. Associations between psych conditions prior to MM diagnosis and costs of care after MM diagnosis were assessed using univariate and multivariate proportional odds models. Results: The study population included 36,007 eligible MM pts with a median follow-up of 1.8 years. 15168 (42%) pts had a psych condition at any time (MM+P), while 9355 (26%) were diagnosed prior to MM diagnosis (MM+P PRE). Depression was present in 8421 pts (23%), 4546 (13%) of those occurring prior to MM diagnosis. In comparison to MM pts, MM+P and MM+D pts tended to be female, White, and had a higher CCI (all p<0.001). When compared to MM pts, those with MM+P PRE and MM+D PRE had significantly higher incidence of MM-related complications (hypercalcemia, renal dysfunction, anemia, fractures and dialysis) at the time of or after MM diagnosis and also required increased overall care (all p<0.001). Both, MM+P and MM+D had higher odds of ipt visits (OR 1.48 and 1.41, resp., p<0.001), ED care (OR 1.48 and 1.37, resp., p<0.001) and opt visits (OR 1.25 and 1.22, resp., p<0.001) as compared to MM only pts. Cost of care analysis showed that MM+P and MM+D pts had a significantly higher cost of opt (OR 1.36 and 1.39, resp., p<0.001), ipt (OR 1.49 and 1.54, resp., p<0.001) and total care (OR 1.52 and 1.55, resp., p<0.001) as compared to MM only pts during first 6 mth after MM diagnosis (Figure 1). Total costs of care for MM+P and MM+D were also higher than MM only but the differences were less significant. Cost of care differences existed within first 6 mth of MM diagnosis by pt race as well with MM+P among Hispanic and Asian pts being more strongly associated with higher costs than Whites and African-Americans (AA) (p<0.001). MM+D had similar trends but not significant after adjustment for multiple comparisons. Conclusion: Psych comorbidities are associated with significant increase in healthcare utilization and cost of care in MM pts and may contribute to higher MM-related complications. More research is needed to study whether a multidisciplinary approach to identify and manage MM pts with psych conditions may help mitigate these trends. Disclosures Ailawadhi: Pharmacyclics: Consultancy; Novartis: Consultancy; Amgen Inc: Consultancy; Takeda Oncology: Consultancy.

Abstract Background: Patient (Pt) outcomes have improved significantly in multiple myeloma (MM) due to widespread use of novel therapeutic agents and stem cell transplant (SCT) butoutcome disparities among racial/ethnic groups exist. Differences in access to and utilization of novel therapeutics are likely contributing factors. We performed an analysis of the SEER-Medicare database to explore such differences. Methods: Pts with a confirmed MM diagnosis between 2007-2009 and continuous Medicare coverage from 1 year prior to diagnosis through end of 2012 were included. Medicare Part D data (oral medications) is not available prior to 2007 so utilization analysis of lenalidomide (LEN), thalidomide (THAL) and bortezomib (BORT) was performed from 2007-2009. Demographic/survival data (PEDSAF files) and drug/SCT utilization (NCH, OUTSAF, PDESAF files) were obtained (HCPCS/NCH codes). Categorical and continuous variables were compared using /Fisher tests and Wilcoxon tests, respectively. Gender, age, and year-adjusted p-values were calculated using logistic regression. A proportional odds model was used to detect time-dependent trends in drug usage within 12 months of MM diagnosis across year of diagnosis by race. Separate multivariate models were fit for each drug adjusting for gender, age, drug use and race. Associations between drug use and overall survival (OS) were examined using a time dependent proportional hazards model. Results: A total of 5358 MM pts were included (any Medicare claims) with 2733 (51%) males. Mutually exclusive race categories were White (W; 67%), African-American (AA; 18%), Hispanic (H; 10%) and Asian (A; 5%). Median follow up was 2.4 yrs. Baseline pt characteristics were significantly different for age with AA being the youngest at diagnosis (p<0.001) and for gender (p<0.001) with W having the largest (53.4%) and AA the smallest (45.2%) proportion of males. Medicare claims capture only pts without any supplemental insurance and so drug utilization in this database was much lower than expected for all subgroups. Year-wise comparison showed that LEN use within the first year of MM diagnosis increased from 2007 to 2009 (16.5% to 27.7%; p<0.001) and there was a significant decrease in median days to first dose (p=0.002). Over the same period, use of THAL decreased (26.6% to 14.6%; p<0.001) while use of BORT (7.8% to 15.4%; p<0.001), SCT (3.3% to 6.2%; p<0.001) and more than 1 novel agent (2.5% to 5.7%; p<0.001) increased significantly. Utilization patterns within first year of MM diagnosis were significantly different among racial subgroups for LEN, THAL, BORT, SCT and >1 novel agent use. (Figure 1) No significant difference was seen for days to first dose of LEN/THAL for any race but for BORT, H got first dose in median 117 days after diagnosis as compared to median 46-51 days for other races (p=0.025). During the first year after diagnosis, W and AA had higher BORT only usage while H and A had a higher LEN/THAL only usage (p<0.001). For drug use by year and race, a significant increase in LEN use was seen from 2007 to 2009 in W (p<0.001), AA (p=0.001) and A (p<0.001) but not H (p=0.26) while THAL use decreased significantly for all races. BORT use increased significantly for all except A (p=0.602) and SCT use increased significantly for all except AA (p=0.068). Inferior OS was significantly associated with older age (p<0.001) and males (p<0.001). LEN/THAL use within the first year of diagnosis did not affect OS while BORT use within the first year of diagnosis increased mortality risk (HR 1.18, p=0.021). SCT use improved OS (HR 0.52, p<0.001). Conclusions: In this comprehensive analysis of MM therapeutics utilization, we noted significant variability among different racial groups. The use of LEN, BORT and SCT has increased while THAL has decreased over time but disparities exist in utilization among different races, with lower use of LEN in AA, lower use of BORT in A and significantly delayed use of BORT in H, which are all notable since racial/ethnic minorities are fast growing in the US population mix and seem to have inequitable utilization and/or access. Inferior survival with early BORT use may represent selection bias of pts with more advanced/aggressive disease. These trends need to be examined in larger sample sizes especially from commercial payers to eliminate drug access and utilization disparities and achieve equitable benefit of therapeutic advances across all racial subgroups. Figure 1 Figure 1. Figure 2 Figure 2. Disclosures Ailawadhi: Pharmacyclics: Consultancy; Novartis: Consultancy; Amgen Inc: Consultancy; Takeda Oncology: Consultancy.