This review describes the relationship between the coronavirus-related pandemic and health inequities. The latter are linked to pre-existing social and economic discriminations in terms of access to healthcare for people affected by chronic diseases. We believe that we are living in a “syndemic pandemic”. The term “syndemic” was originally developed by the medical anthropologist Merrill Singer in the 1990s in order to recognize the correlation between HIV/AIDS, illicit drug use, and violence in the United States. This complex interplay exacerbated the burden of the disease and the prognosis of the patient. Similarly, in COVID-19 infection, socio-economic, ethnic, and racial inequities result in higher morbidity and mortality in certain sections of society. Unfortunately, such differences are becoming too common during the COVID-19 pandemic, in terms of the incidence and prevalence of the disease, as well as inequal access to new medical advances and life-saving therapeutics for those with COVID-19, such as vaccines and monoclonal antibody treatment. Lockdown measures, imposed internationally as a response to the COVID-19 pandemic, are causing economic inequities, which complicate the issue even further. An appropriate syndemic anthropological approach is necessary to ensure that this pandemic does not increase health inequities in access to appropriate treatments.

Background The death rate from colorectal cancer is high and affects poor and medically underserved populations disproportionately. In the United States, health disparities are particularly acute in the Lower Mississippi River Delta region. Because many in the region have limited access to basic health care resources, they are not screened for cancer, even though screening is one of the most effective strategies to prevent colorectal cancer. Community-based participatory research is a promising approach to prevent colorectal cancer in this population. Community Context The Empowering Communities for Life program was implemented in 2 underserved counties in the Arkansas Lower Mississippi River Delta. The program arose from a 9-year partnership between the University of Arkansas for Medical Sciences and 9 cancer councils across Arkansas. Methods Empowering Communities for Life is a community-based participatory intervention designed to increase colorectal cancer screening in rural, un...

Background: Studies have shown that nonmedical reading is associated with low burnout and that small group study sections can promote wellness.Burnout and other psychosocial distress are common among health care professionals, necessitating additional measures to promote well-being.The field of narrative medicine is one proposed solution.Observations: We added small narrative medicine group discussions of nonmedical fiction to our hematology oncology clinical program to promote physician resilience and decrease risk for burnout.We explored how reading and reflecting would result in profound changes in thinking and feeling and noted 7 different ways by which reading and reflecting together can increase well-being.We describe how stories led us to increase bonding, improve empathy, and promote meaning in medicine.Conclusions: Our small group discussions showed that the intervention was feasible, improved empathy and fulfillment at work, and resulted in greater appreciation for the human dimensions of health care.

Allogeneic hematopoietic stem cell transplantation (alloHCT) may be associated with significant morbidity and mortality, resulting in increased healthcare utilization (HCU). To date, no multicenter comparative cost analyses have specifically evaluated alloHCT in children with acute leukemia. In this retrospective cohort study, we examined the relationship between survival and HCU while investigating the hypothesis that matched sibling donor (MSD) alloHCT has significantly lower inpatient HCU with unrelated donor (URD) alloHCT, and that among URDs, umbilical cord blood (UCB) alloHCT will have higher initial utilization but lower long-term utilization. Clinical and transplantation outcomes data from the Center for International Blood and Marrow Transplant Research (CIBMTR) were merged with inpatient cost data from the Pediatric Health Information System (PHIS) database using a probabilistic merge methodology. The merged dataset comprised US patients age 1 to 21 years who underwent alloHCT for acute leukemia between 2004 and 2011 with comprehensive CIBMTR data at a PHIS hospital. AlloHCT was analyzed by donor type, with specific analysis of utilization and costs using PHIS claims data. The primary outcomes of overall survival (OS), leukemia-free survival (LFS), and inpatient costs were evaluated using Kaplan-Meier curves and Cox and Poisson models. A total of 632 patients were identified in both the CIBMTR and PHIS data. The 5-year LFS was 60% for MSD alloHCT, 47% for well-matched matched unrelated donor bone marrow (MUD) alloHCT, 48% for mismatched unrelated donor alloHCT, and 45% for UCB alloHCT (P = .09). Total adjusted costs were significantly lower for MSD alloHCT versus MUD alloHCT by day 100 (adjusted cost ratio [ACR], .73; 95% confidence interval [CI], .62 to .86; P < .001), and higher for UCB alloHCT versus MUD alloHCT (ACR, 1.27; 95% CI, 1.11 to 1.45; P < .001). By 2 years, total adjusted costs remained significantly lower for MSD alloHCT compared with MUD alloHCT (ACR, .67; 95% CI, .56 to .81; P < .001) and higher for UCB alloHCT compared with MUD alloHCT (ACR, 1.25; 95% CI, 1.02 to 1.52; P = .0280). Our data show that UCB and MUD alloHCT provide similar survival outcomes; however, MUD alloHCT has a significant advantage in cost by day 100 and 2 years. More research is needed to determine whether the cost difference among URD alloHCT approaches remains significant with a larger sample size and/or beyond 2 years post-alloHCT. (C) 2020 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc.

New educational methods and structures to improve medical education are needed to face the challenge of an exponential increase and complexity of medical knowledge. Collaborative learning has been increasingly used in education, but its use in medical training programs is in its infancy, and its impact is still unknown; the role of competition in education is more controversial. We introduced these pedagogical methods to the hematology/oncology fellowship programat the University of Arkansas for Medical Sciences to improve attendance and performance at didactic activities and different educational outcomes. One year after the adoption of these methods, the fellowship program has reached many of the expected goals from this intervention without the negative consequences of competition observed in younger learners. The most important conclusion of this project is that collaboration and cross-generational team work provide a healthy and effective learning environment and competition may not add further benefit. Analysis, interpretation, and discussion of our experience are provided. This study was approved by the University of Arkansas for Medical Sciences IRB as a low risk educational intervention not requiring a consent form.

Chronic myelomonocytic leukemia (CMML) is an aggressive neoplasm with sparse data on outcomes at a population level. Using Surveillance Epidemiology and End Results (SEER) database, we identified 2238 patients with CMML diagnosed in the period 2003-2013. We found that the disease incidence was significantly higher with advancing age and lower in females, Blacks, and Asian/pacific islanders. Median OS declined significantly with increasing age (age 20-39 - 25 months, age 40-59 -20 months, age 60-79 -18 months, and age >= 80 - 11 months, p <. 01), but did not vary by gender or race. Median OS has improved in the period 2007-2013 as compared with 2003-2006 (17 months vs. 14 months, p <. 01). In spite of advances in CMML biology and therapeutics, in general, the survival of CMML

We used the Surveillance, Epidemiology, and End Results database to study the incidence and survival of anaplastic lymphoma kinase-positive anaplastic large cell lymphoma. We found that the disease incidence varied significantly by age, gender, and race, whereas survival was influenced by age, race, stage, period of diagnosis, and radiotherapy.Introduction: Systemic ALK-positive anaplastic large cell lymphoma (ALK-positive ALCL) is a T-cell lymphoma. Owing to its rarity, variations in incidence and survival at the population level are not clearly known. Materials and Methods: Using the Surveillance Epidemiology and End Results database (SEER 18), we selected patients aged >= 20 years with ALK-positive ALCL, diagnosed between 2001 and 2013. Incidence rate, overall survival (OS), and its determinants were analyzed with a significance level of P < .05. Results: We identified 1604 patients with a median age of 54 years. The disease incidence increased significantly with advancing age, with higher incidence in Blacks and lower incidence in American Indians and Asian/Pacific Islanders as compared with Whites. The 5-year OS significantly declined as the age advanced (age 20-40 years, 68.7%; age 41-60 years, 53.8%; age 61-80 years, 28.9%; age > 80 years, 15.2%; P < .01) and varied with race (Whites, 49.7% vs. Blacks, 37.7% vs. Asian/Pacific Islander, 42.8% vs. American Indian, 35.8%; P = .03). On multivariate analysis, treatment with radiation (hazard ratio [HR], 0.72; 95% confidence interval [95% CI], 0.59-0.87; P < .01) and year of diagnosis from 2009 through 2013 (HR, 0.77; 95% CI, 0.65-0.93; P < .01) were associated with lower mortality. Advanced age, Black race (HR, 1.37; 95% CI, 1.14-1.65; P < .01), and advanced disease stage (HR, 1.74; 95% CI, 1.51-2.02; P < .01) were associated with higher mortality. Conclusion: Incidence and survival of ALK-positive ALCL varies significantly with patients' demographic characteristics as identified in our study. Treatment strategies need to be tailored accordingly to address these variations and ensure uniform access to care.

Introduction: World Health Organization (WHO) classifies therapy-related myeloid neoplasms (t-MN) into therapy related acute myeloid leukemia (t-AML), therapy related myelodysplastic syndrome (t-MDS), and therapy related myelodysplastic syndrome/myeloproliferative neoplasms (t-MDS/MPN). These diseases are aggressive hematological malignancies and only allogeneic transplant offers the possibility of long-term remission. We performed retrospective analyses of Surveillance, Epidemiology, and End-Results (SEER) database to examine differences in incidence and survival outcomes of t-MN across different races and ethnicities in United States (US).

BACKGROUND:Acute leukemia of ambiguous lineage (ALAL) is a rare leukemia with sparse data availability about the survival and management strategies in elderly patients. METHODS:We used the Surveillance Epidemiology and End Results (SEER)-Medicare database to describe the overall survival (OS) and treatment pattern of elderly patients (age > 65 years) with ALAL. OS analysis was done using the Kaplan-Meier method, and its determinants were analyzed using the Cox proportional hazard regression method with a significant P < .05. RESULTS:We included 705 patients with ALAL and a median age of 80 years. The 2-year OS was 16.4% for patients aged 66 to 70 years, 8.1% for patients aged 71 to 75 years, 5.5% for patients aged 76 to 80 years, and 3.7% for patients aged > 80 years (P < .01). Two-year OS did not significantly vary by race or gender. Among the study cohort, 151 patients received chemotherapy. Two-year OS was 17% in the chemotherapy group and 3% in the no-chemotherapy group (P < .001). On multivariate analysis, age less than 80 years (Age 66-70 years: hazard ratio [HR]; 0.66, 95% confidence interval [CI], 0.52-0.85; age 71-75 years: HR, 0.80; 95% CI, 0.65-0.99; age 76-80 years: HR, 0.80; 95% CI, 0.66-0.98; P = .004) and chemotherapy (HR, 0.51; 95% CI, 0.42-0.62; P = .001) significantly reduced the hazard for mortality. CONCLUSION:Our study suggests that the OS of elderly patients with ALAL remains poor. Although treatment improved the OS, only 21.5% of patients received therapy. The optimal choice of therapy needs to be determined by prospective studies.

7049 Background: AHCT is increasingly employed for the treatment of HM in elderly patients. However, it is unclear if the hospitalization outcomes of this procedure differ by patient’s age. Methods: Using Nationwide Inpatient Sample database, we analyzed the differences in hospitalization outcomes of elderly (age > 60 years) vs. young (age 20-60 years) patients who underwent AHCT for HM between the years 2006-2011. ICD-9 diagnosis and procedure codes were used to identify subjects. Patient’s length of stay (LOS), total charge (TOTCHG), complications and mortality were studied. Descriptive statistics, Chi-Square test, Mann-Whitney U test and multivariate logistic regression analyses were performed with p < 0.05 as significant. Results: A total of 4371 young and 1175 elderly patients who underwent AHCT were included. Baseline characteristics are summarized in table 1. Median LOS was 27 days in the young group vs. 25 days in elderly group (p < 0.01). Median TOTCHG was $ 267,321 in the young group vs. $ 240,269 in elderly group (p < 0.01). In-hospital mortality was 7.4% in the young group vs. 7.9% in elderly group (p = 0.53). Complications like septicemia (21% vs. 19.2% p = 0.16), respiratory failure (9% vs. 9.1%, p = 0.90), renal failure (14.5% vs. 15.4% p = 0.43), shock (3.6% vs. 3.7% p = 0.86) were comparable in both groups. On multivariate analysis, age > 60 was associated lower TOTCHG and LOS (OR 0.69, 95% CI 0.59-0.81, p < 0.01). Cord blood AHCT (OR 4.67, 95% CI 3.68-5.91, p < 0.01) and bone marrow AHCT (OR 1.59, 95% CI 1.39-1.84, p < 0.01) were associated with prolonged LOS and TOTCHG. Conclusions: Compared to younger patients, AHCT in elderly patients is associated similar medical complications and in-hospital mortality. Lower TOTCHG and LOS in elderly patients might be related to the practice of using lesser intensity conditioning regimen and better patient selection. Baseline characteristics. Young Elderly Median age (Range) 47 (20-60) 64 (61-78) Gender Male 2441 742 Female 1930 433 Race Whites 2844 899 Blacks 222 24 Others 568 81 Missing 737 171 Disease AML/MDS 2101 716 ALL 637 59 CML 275 38 CLL 207 102 Lymphoma 875 194 Myeloma 214 28 MPD 62 38

Using the Quality Oncology Practice Initiative, an affiliate program of ASCO, we outlined opioid-associated constipation (OAC) as a subject in need of quality improvement (QI) in our fellowship program at the University of Arkansas for Medical Sciences and Central Arkansas Veterans Healthcare System. We initiated a fellow-led QI project to advance the quality of patient care and provide a valuable avenue for QI training of young physicians. Fellows organized meetings with all stakeholders, addressed the scope of the problem, and devised strategies for OAC management. Monthly meetings were organized using Plan-Do-Study-Act principles. Mandatory check boxes were inserted into our electronic medical record templates to remind all physicians to identify patients on opioid medications and assess and address OAC. Final chart audit and patient satisfaction surveys were performed 6 months after project initiation. Assessment of OAC improved from 52% at baseline to 92% ( P < .003). This improvement corresponded with high patient satisfaction scores, with 90% of surveyed patients reporting adequate management of their constipation. In this QI initiative, we showed that participation in ASCO's Quality Oncology Practice Initiative helps identify areas in need of QI, and such fellow-led QI projects can serve as models for QI training of young physicians.

Background Local Tumor Invasiveness (LTI) is among the strongest prognostic factors in Early Stage Nasal Natural Killer/T Cell Lymphoma (ES-NNKTCL), with very poor outcomes. Standard therapy is concurrent or sequential chemotherapy and radiotherapy (RT). SMILE (Dexamethasone, Methotrexate, Ifosfamide, L-asparaginase and Etoposide) is an intensive, highly active protocol in this disease. No prior study has explored SMILE with RT specifically for ES-NNKTCL with LTI. Methods Between 2011 and 2015, all patients with ES-NNKTCL with LTI at our center were treated with a uniform protocol of SMILE for 6 cycles with sandwiched RT (45-50 Gray). LTI was defined as bony invasion/destruction or involvement of the skin. Records of these patients were retrospectively reviewed. Results Sixteen patients (62% male) were identified with median age 31 years (range 19-55). Eleven (69%) were stage I and 5 (31%) were stage II. Facial skin involvement (56%), palatal invasion (69%), or orbital extension (56%) were present in the majority. 12/16 had B-symptoms, and 6/16 had elevated lactate dehydrogenase. International prognostic index was 0, 1, and 2 in 56%, 19% and 25% respectively. Korean prognostic score was 0, 1, 2, and 3 in 19%, 38%, 31% and 12% respectively. All patients received 5-6 cycles (Five patients received 5 cycles, the rest received 6 cycles) barring those who progressed on therapy. RT was delivered after a mean 4 6 1 cycles. At a median follow up of 14.5 months (range 2-63), 2/16 patients are still on therapy. Among the remaining 14, two patients (14%) relapsed while on therapy. Rest all achieved complete remission at the end of the treatment (86%). For the entire cohort, one year progression-free survival was 78% and overall survival was 93%. Grade III-IV toxicity was seen in 81%, most commonly neutropenia (69%), anemia (38%) and thromobocytopenia (31%). Neutropenic fever was seen in 25% despite growth factor prophylaxis. Six patients (38%) required dose adjustments (predominantly in the first 1 or 2 cycles). No treatment-related deaths occurred. Conclusions SMILE with RT is a toxic but tolerable protocol for ES-NNKTCL with LTI with high efficacy. Prospective studies are warranted. Legal entity responsible for the study N/A Funding N/A Disclosure All authors have declared no conflicts of interest.

Using the standardized ASCO Quality Oncology Practice Initiative (QOPI) guidelines for assessing quality cancer care, we identified communication about intent of chemotherapy as an area needing improvement in our program at the University of Arkansas for Medical Sciences (UAMS) and the Central Arkansas Veterans Healthcare System (CAVHS). We organized training in communications on intent of treatment (palliative vs curative) and added optional checkboxes to our electronic templates for progress notes. Afterwards, we conducted a retrospective review of electronic medical records of initially often randomly selected patient charts. The first 10 patient charts after 1 month of implementation showed intent of treatment in 80 % of charts compared to 74 % at baseline. We then changed checkboxes from mandatory to optional and reviewed 30 randomly selected patient charts. Intent of treatment was documented in 96.7 % of cases compared to 74 % at baseline. We also assessed patient satisfaction through surveys distributed in clinic. Patient satisfaction scores were close to 100 % for receiving clear information, understanding the reason for which they were receiving chemotherapy, and willingness of oncologists to listen carefully, to take time to answer questions, to explain things clearly, and to spend adequate time with them. In this study, we showed that training in communication of intent of chemotherapy and use of checkboxes in progress note templates could improve competency in communication of intent of therapy in cancer patients.

178 Background: The American Society of Clinical Oncology considers palliative care an integral part of cancer therapy. Our Hematology and Oncology fellowship at the University of Arkansas for Medical Sciences (UAMS) began a year-long palliative care curriculum to improve symptom management education. In this pilot study we evaluate fellows’ attitude and knowledge in cancer pain management before and after implementing a pain management curriculum. Methods: Hematology and Oncology Fellows were divided into three groups. Each group delivered a one hour lecture in pain management for a total of 3 didactic lectures. We adopted “Evidence based Practice of Palliative Medicine by Goldstein and Morrison” as the main textbook. Fellows answered a 30 item questionnaire to address attitudes and knowledge in pain management. Answers were scored using a 5 point Likert scale (1 = strongly disagree and 5 = strongly agree). Results: 11 fellows participated; six males, five females, median age 34 (R 28-40), one US graduate, and ten foreign graduates. More fellows felt comfortable managing acute (M = 4.3, SD = 0.48) compared to chronic (M = 3.8, SD 0.78) cancer related pain. Most believe that if they were taught the principle of pain management they would feel more comfortable managing pain (M = 4.6, SD = 0.51). Post pain management module, there was a statistically significant improvement in fellow’s knowledge in pain management in the setting of renal failure (P = 0.02) and bone pain (P = 0.006), and a trend towards statistically significant in both opioid rotation and conversion (P = 0.06). Fellows did poorly on opioid-drugs interaction and management of neuropathic pain. Fellows valued palliative medicine service as a great resource for their patients but most believe that they should not refer all their patients to palliative medicine for pain management. Conclusions: Pain management skills are eroding among Oncology fellows and efforts should be made to enhance symptom and pain management education in oncology training programs. This curriculum improved knowledge and self-efficacy in pain management and revealed areas for further improvement. More research is needed to address whether fellows use and apply pain management skills in the clinical setting.

Aim/Background: A pilot study to standardize MRD estimation in AML by flowcytometry at a cancer centre in India.Methods: 46 patients of AML in CR post-induction were enrolled during the period December 1, 2012 to May 13, 2014 (median follow-up 13.2 months,1.7 -19 months). For flow cytometric analysis of BM aspirate, at least 20,000 events at diagnosis and 200,000 at CR were acquired in all cases from each tube and data were stored. An extensive panel of fluorochrome-conjugated antibodies, selected based on published literature, was applied to bone marrow samples, at two time points (baseline and post-induction) to identify leukemia associated abberant immunophenotypes at baseline and quantify these post-induction. We used seven combinations, each containing five antibodies, with CD34 and CD45 as backbone (CD65FITC/CD2PE/CD34ECD/CD13PE-Cy5/CD45PE-Cy7, CD9FITC/HLA-DRPE/CD34ECD/CD33PE-Cy5/CD45PE-Cy7, CD11b/CD15PE/CD34ECD/CD117PE-Cy5/CD45PE-Cy7, CD34FITC/CD56PE/CD19ECD/CD33PE-Cy5/CD45PE-Cy7, CD15FITC/CD7PE/CD34ECD/CD33PE-Cy5/CD45PE-Cy7, CD38/CD117PE/CD34ECD/CD4PE-Cy5/CD45PE-Cy7 and CD36FITC/HLA-DRPE/CD14ECD/CD64PE-Cy5/CD45PE-Cy7).Serial dilution experiments with normal bone marrow samples were conducted at 10%, 1%, 0.1%, 0.01% and 0.001%. Ten BM aspirates were used as controls.Results: 42/46 samples (91%), showed LAIPs (median= 4, range = 1-8). Cross lineage infidelity was the commonest LAIP (78%), followed by underexpression(71%), overexpression(69%) and asynchronous expression (43%).The individual most common aberrancies were underexpression of CD45 ( 52%), followed by cross-lineage expression of CD56 ( 43%) and CD 4(36%), asynchronous expression of CD65(40%), and underexpression of CD33(36%) and CD38 (36%). Median MRD value was 0.055% (range =0.001-4.1). 31/42 (74%) patients had detectable MRD post-induction.Five patients relapsed, of which four had a positive MRD value post-induction (0.04, 0.3, 0.01 and 0.2%). OS and RFS of the cohort was 87% and 72% respectively. The relapse-free survival was 90.9% in patients with no detectable MRD post-induction and 65.9% in patients with detectable MRD (p = 0.6, HR = 1.7).Conclusions: MRD estimation by flowcytometry is feasible and applicable in our patients of AML.Disclosure: All authors have declared no conflicts of interest. Aim/Background: A pilot study to standardize MRD estimation in AML by flowcytometry at a cancer centre in India. Methods: 46 patients of AML in CR post-induction were enrolled during the period December 1, 2012 to May 13, 2014 (median follow-up 13.2 months,1.7 -19 months). For flow cytometric analysis of BM aspirate, at least 20,000 events at diagnosis and 200,000 at CR were acquired in all cases from each tube and data were stored. An extensive panel of fluorochrome-conjugated antibodies, selected based on published literature, was applied to bone marrow samples, at two time points (baseline and post-induction) to identify leukemia associated abberant immunophenotypes at baseline and quantify these post-induction. We used seven combinations, each containing five antibodies, with CD34 and CD45 as backbone (CD65FITC/CD2PE/CD34ECD/CD13PE-Cy5/CD45PE-Cy7, CD9FITC/HLA-DRPE/CD34ECD/CD33PE-Cy5/CD45PE-Cy7, CD11b/CD15PE/CD34ECD/CD117PE-Cy5/CD45PE-Cy7, CD34FITC/CD56PE/CD19ECD/CD33PE-Cy5/CD45PE-Cy7, CD15FITC/CD7PE/CD34ECD/CD33PE-Cy5/CD45PE-Cy7, CD38/CD117PE/CD34ECD/CD4PE-Cy5/CD45PE-Cy7 and CD36FITC/HLA-DRPE/CD14ECD/CD64PE-Cy5/CD45PE-Cy7).Serial dilution experiments with normal bone marrow samples were conducted at 10%, 1%, 0.1%, 0.01% and 0.001%. Ten BM aspirates were used as controls. Results: 42/46 samples (91%), showed LAIPs (median= 4, range = 1-8). Cross lineage infidelity was the commonest LAIP (78%), followed by underexpression(71%), overexpression(69%) and asynchronous expression (43%).The individual most common aberrancies were underexpression of CD45 ( 52%), followed by cross-lineage expression of CD56 ( 43%) and CD 4(36%), asynchronous expression of CD65(40%), and underexpression of CD33(36%) and CD38 (36%). Median MRD value was 0.055% (range =0.001-4.1). 31/42 (74%) patients had detectable MRD post-induction.Five patients relapsed, of which four had a positive MRD value post-induction (0.04, 0.3, 0.01 and 0.2%). OS and RFS of the cohort was 87% and 72% respectively. The relapse-free survival was 90.9% in patients with no detectable MRD post-induction and 65.9% in patients with detectable MRD (p = 0.6, HR = 1.7). Conclusions: MRD estimation by flowcytometry is feasible and applicable in our patients of AML. Disclosure: All authors have declared no conflicts of interest.

e18072 Background: CMML is a rare hematological malignancy with characteristics of both myelodysplastic and myeloproliferative disorders. Population level data about its overall survival (OS) in the recent years is sparse. In this study, we aimed to delineate the characteristics and trends in OS of CMML in the US. Methods: The Surveillance, Epidemiology and End Results (SEER-18) database was used to identify patients with age > 20 years diagnosed with CMML (ICD-O-3 code –9945) between 2000 to 2011. Baseline characteristics were described in frequencies. OS was calculated using Kaplan-Meier method and compared by log rank test. Multivariate analysis was performed using Cox proportional hazard regression method. Results: 3520 patients with CMML had a median age of 76 years (range 22-94). Patient characteristics included age > 60 (89.3%), 62.4% males, 88.4% Whites, 6.2% African Americans and 5.2% other racial groups. About 74.7% patients had primary CMML and 25.3% had secondary CMML. Incidence of CMML has increased from 0.5/100,000 to 0.6/100,000 population from 2000 to 2011. Median OS was worse in patients with age > 60 vs. age < 60 (13 months vs.18 months, p < 0.001), males vs. females (13 months vs. 14 months, p = 0.04) and secondary CMML vs. primary CMML (10 months vs. 14 months, p < 0.001). Median OS did not differ significantly with race (Whites -14 months, African Americans- 8 months and other races -15 months; p = 0.07). We found that OS has improved in patients diagnosed from 2007-2011 as compared to diagnosis from 2000-2006 (Median OS – 15 months vs. 12 months, 2-year OS – 35.7% vs. 32.5%, p < 0.01). On multivariate analysis age > 60 years (HR-1.47, 95% CI 1.29-1.68, p < 0.01) and secondary CMML (as compared to primary CMML, HR-1.20, 95% CI 1.10-1.31, p < 0.01) were significantly associated with higher mortality. Diagnosis from 2007-2011 was associated with lower mortality (HR 0.89, 95% CI 0.82-0.96, P < 0.01). Conclusions: Our data suggests that age > 60 and secondary CMML portend worse prognosis. Although the OS of CMML in US seems to be improving in the era of hypomethylating agents, stem cell transplant and supportive care, its dismal prognosis indicates the need for novel therapies.

8597 Background: Studies on multiple myeloma demonstrate an improvement in overall survival (OS) and increased risk of second primary malignancy (SPM); Population level data on OS and the risk of SPM in primary plasmacytoma (PP) is sparse. Hence, we aimed to determine the trends in OS and SPM in patients with PP in US. Methods: The Surveillance, Epidemiology and End Results (SEER-13) database was used to identify patients with age > 20, diagnosed with solitary plasmacytoma of bone (SPB) or extramedullary plasmacytoma (EMP) (ICD-O-3 codes –9731, 9734) as the first primary malignancy between 1992 to 2011. OS analysis was performed using Kaplan-Meier method and compared by log rank test. SPM was identified using SEERstat software (version 8.1.5) with a latency period of 6 months from radiotherapy (RT). Multivariate analysis of OS was performed using Cox proportional hazard regression method. Results: 2064 patients with SPB/EMP had a median age of 62 years (range 20-96). Baseline characteristics included 63.6% males, 54.2% with age > 60, 79.2% Whites, 14% African americans and 6.8% other races. About 48.4% of patients received RT alone, 23.5% had surgery and RT, 10.5% had surgery and 17.6% did not receive surgery or RT. OS at 10 years was higher in patients with age < 60 vs. age > 60 (62.5% vs. 22.2%, p < 0.01) and for other races vs. Whites vs. African americans (51.1% vs. 40.1% vs. 34.5% respectively, p = 0.01). Use of RT and surgery resulted in better 10-year OS (51.2%) than surgery alone (44.1%), RT alone (39.8%) or no RT/surgery (21.2%) (p < 0.01).On multivariate analysis, age > 60 (HR 3.01, 95% CI 2.61-3.46; p < 0.01), Whites vs. other races (HR 1.35, 95% CI 1.02-1.77; p = 0.03) and African americans vs. other races (HR 1.52, 95% CI 1.11-2.08; p < 0.01) predicted worse OS, whereas the use of RT along with surgery Vs. RT alone (HR 0.82, 95% CI 0.70-0.97; p = 0.02) predicted better OS. Patients treated with RT had more SPM as compared to general population- all sites standardized incidence ratio (SIR) 1.49, with more leukemia/lymphoma (SIR 6.37) and bone tumors (SIR 19.93) (p < 0.05). Conclusions: Our study demonstrates that the use of RT and surgery for PP significantly improves the OS, despite an increased risk of SPM with RT. Our study also shows a significant racial disparity in the long-term OS of PP.