Parikh, Sameer A.

Abstract BACKGROUND: Consistent with the advanced age at diagnosis (median age ~70 years), most patients with CLL have co-existent health problems. These co-morbidities influence the ability of many CLL patients to tolerate aggressive chemotherapy-based treatment and can also contribute to treatment-related side effects. The recent development of novel signaling inhibitors, particularly the Bruton's tyrosine kinase inhibitor ibrutinib, has been a major treatment advance for patients with CLL. While these agents generally have favorable toxicity profiles relative to standard chemotherapy-based treatments, they are chronic therapies which patients typically stay on for an extended period. Preliminary data suggests ibrutinib may be associated with an increased risk of atrial fibrillation (Afib). In one randomized trial comparing ibrutinib to ofatumumab in patient with relapsed CLL, incident grade 3+ Afib occurred in 3% of ibrutinib treated patients compared to 0% of ofatumumab treated patients (NEJM 371:213). Despite these observations, the baseline frequency of Afib in patients with CLL is not well described - particularly incident atrial fibrillation acquired during the course of the disease. METHODS: We used the Mayo Clinic CLL database to evaluate the prevalence of Afib at the time of CLL diagnosis as well as the incidence of Afib during follow-up. All patients with a new diagnosis of CLL after January 1995 who were seen at Mayo within 12 months of diagnosis were included in the analysis. Afib was identified by chart review and by billing search using ICD9 codes. Data on co-morbid conditions associated with risk of Afib was also abstracted (e.g. hypertension, coronary artery disease [CAD], valvular heart disease, cardiomyopathy, diabetes mellitus, pulmonary disease). RESULTS: A total of 2444 patients with newly diagnosed and previously untreated CLL were seen at Mayo Clinic within 12 months of diagnosis between 1/1995 and 4/2015.Median age at diagnosis was 65 years and 1626 (66.5%) patients were men. A history of Afib was present at the time of CLL diagnosis in 148 (6.1%) patients. Four additional patients had Afib documented in the record but the precise date of onset (e.g. prior to or after CLL diagnosis date) could not be determined. Age, male sex and history of CAD, valvular heart disease, cardiomyopathy, hypertension, and diabetes were associated with a greater likelihood of having a history of Afib at the time of CLL diagnosis (all p<0.01). Among the 2292 patients without a history of Afib at CLL diagnosis, 139 (6.1%) had incident Afib during the course of follow-up for their CLL. The incidence of Afib among patients without a history of Afib at diagnosis was approximately 1%/year (Figure 1A). Considering both Afib present at the time of CLL diagnosis or acquired during the course of the disease, 291 (11.9%) of the 2444 patients in this cohort experienced Afib (median follow-up: 59 months). Among patients without Afib at the time of CLL diagnosis, the following characteristics at the time of CLL diagnosis were associated with an increased risk of incident Afib on multivariate analysis: older age (age 65-74 HR=2.4, p<0.001; age ≥75 HR=3.6, p<0.001), male sex (HR=1.8, p=0.004); valvular heart disease (HR=2.4, p=0.007), and hypertension (HR=1.5; p=0.02). A predictive model for acquired Afib was subsequently constructed based on the independent factors in the Cox regression model. An individual weighted risk score was assigned to each independent factor based on the regression coefficients of the HRs. The Afib risk score (range 0-7) was defined as the sum of the scores of these independent factors. The risk of incident Afib among patients with risk scores of 0-1, 2-3, 4, and 5+ is shown in Figure 1B. Rates for these 4 groups were significantly different (p<0.001), with the 10-year Afib rates (95% C.I.) for those with a score of 0-1, 2-3, 4, and 5+: 4% (2-6%), 9% (6-13%), 17% (11-23%), and 33% (20-43%), respectively. CONCLUSIONS: A history of Afib is present in approximately 1 out of every 16 patients with newly diagnosed CLL. Among patients without Afib at diagnosis, the incidence rate of Afib is ~1%/year. The risk of incident Afib in newly diagnosed CLL patients can be predicted based on age, sex, and co-morbid health conditions present at diagnosis. These data provide context to help interpret data on the frequency of Afib in CLL patients treated with ibrutinib and other novel agents. Disclosures Shanafelt: Janssen: Research Funding; Polyphenon E Int'l: Research Funding; Glaxo-Smith_Kline: Research Funding; Cephalon: Research Funding; Genentech: Research Funding; Hospira: Research Funding; Celgene: Research Funding; Pharmactckucs: Research Funding. Ding:Merek: Research Funding. Kay:Tolero Pharm: Research Funding; Hospira: Research Funding; Genentech: Research Funding; Pharmacyclics: Research Funding; Gilead: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding.

Abstract INTRODUCTION. Although uncommon, involvement of the central nervous system (CNS) is the most common extramedullary manifestation of Chronic Lymphocytic Leukemia (CLL). Nonetheless, a broad array of conditions can lead to neurologic symptoms in CLL patients and distinguishing between clinically significant CLL involvement of the CNS and other etiologies can be challenging. The actual prevalence of CNS involvement by CLL is unknown, the only available data being case reports and small case series. METHODS. Between 01/1995 and 11/2014, 186 of the 4317 (4%) patients with CLL followed at our center underwent an MRI of the CNS (e.g. brain, spine) and/or a lumbar puncture (LP) with cerebral spinal fluid (CSF) analysis to evaluate neurologic symptoms. As the mere presence of CLL cells in the CSF is not diagnostic of clinically significant CNS disease, a final diagnosis of clinically significant CNS involvement by CLL was established by the collaborative review of the pathologist, neurologist and treating hematologist based on comprehensive work-up and multi-disciplinary evaluation. RESULTS. After evaluation, the etiology of neurologic symptoms was clinically significant involvement of the CNS by CLL in 19 (10%) patients, CNS Richter Syndrome (RS) in 15 (8%), infection in 42 (23%), autoimmune conditions in 31 (17%), other cancer in 9 (5%), and another etiology in 70 (37%). The sensitivity of LP to detect clinically significant CNS involvement by CLL was 89%, however its specificity was only 42% where CLL cells were also frequently observed in the CSF in concomitance with inflammatory processes, such as infections and autoimmune disease. Factors associated with clinically significant CNS involvement by CLL/RS were mutated IGHV (p=0.04), low CSF glucose (p=0.02), elevated CSF total nucleated cells (TNC)(p=0.004), lymphocytes (p=0.02) and CLL cells, both as absolute count (p=0.05) and percentage of TNC (p=0.02). Median OS among patients with clinically significant CNS involvement by CLL or RS were 21 and 11 months, respectively. Median OS for patients with neurological symptoms secondary to infection was 6 months, to autoimmune process was 63 months, to other cancers was 19 months, and to other non-CLL related etiologies was 37 months. CONCLUSIONS. Clinically significant CNS involvement by CLL and RS are rare conditions. Neurologic symptoms in patients with CLL are attributable to non-CLL-related etiologies in the majority of cases, even when CLL B-cells are present in the CSF. Analysis of the CSF has high sensitivity but limited specificity to distinguish clinically significant CNS involvement by CLL from other etiologies. Comprehensive evaluation is required before attributing neurologic symptoms to CLL. Disclosures No relevant conflicts of interest to declare.

Abstract Introduction: The Food and Drug Administration (FDA) approved ibrutinib in February 2014 based on results from a pivotal phase 2 clinical trial that enrolled 85 patients with relapsed/refractory CLL. A subsequent phase 3 trial confirmed the superior progression-free survival (PFS) and overall survival (OS) of ibrutinib compared to ofatumumab in 391 CLL patients. However, there is no information on the efficacy and safety of ibrutinib in CLL patients treated outside the context of clinical trials. Additionally, it is unclear if the reasons for discontinuation of ibrutinib differ in routine clinical practice from those described in patients treated on clinical trials. We conducted a retrospective study to evaluate these aspects in a large cohort of CLL patients treated with ibrutinib off-protocol. Methods: After IRB approval, CLL patients seen at Mayo Clinic, Rochester, MN, and treated with commercial ibrutinib from November 2013 through July 2015 were evaluated. The baseline characteristics, time to therapy discontinuation, reasons for discontinuation and survival following discontinuation were abstracted. Results: One hundred thirty-five patients with CLL were treated during the study interval; 124 received ibrutinib for relapsed/refractory CLL and 11 received ibrutinib for treatment-naïve CLL with TP53 disruption (either del17p on FISH or TP53 mutation). Patients with relapsed/refractory CLL (n=124) Patients were high-risk, with a median of 3 prior therapies (range, 1-15), 81% had unmutated IGHV, and 38% had either del11q or del17p on FISH testing. After a median follow-up of 6.4 months, 101 remain on ibrutinib therapy. The estimated proportion of patients continuing ibrutinib at 6 months was 84% (95% CI: 77-92%) and at 12 months was 70% (95% CI: 59-83%). Of the 23 who discontinued ibrutinib therapy, 8 discontinued for progressive disease (2 progressive CLL; 6 Richter's transformation). The remaining 15 discontinued therapy for reasons other than progression, including: infection (n=3), bleeding (n=3), diarrhea (n=1), atrial fibrillation (n=1), rash (n=1), immune thrombocytopenia (n=1), renal dysfunction (n=1), severe edema (n=1), MDS (n=1) and physician preference (n=2). Table 1 shows the reasons for discontinuation of ibrutinib in patients treated on clinical trials (Maddocks, JAMA Oncology, 2015) compared to the current cohort of patients. The estimated cumulative incidence of discontinuation of ibrutinib therapy due to CLL progression/Richter's transformation at 6 and 12 months was 9% (95% CI: 4-17%) and 12% (95% CI: 5-23%), respectively. The estimated cumulative incidence of ibrutinib discontinuation due to reasons other than progression at 6 and 12 months was 10% (95% CI: 5-17%) and 22% (12-34%), respectively (Figure). The median survival after discontinuation of ibrutinib for patients who had CLL progression was 3.5 months, not reached in patients who developed Richter's transformation and not reached in those patients who discontinued ibrutinib due to reasons other than progression. Patients with treatment-naïve CLL (n=11) All patients had evidence of TP53 disruption (9 had del17p, 3 had TP53 mutation, and 1 had both). After a median follow-up of 2.8 months, no patient has had progression of CLL, and 2 patients have discontinued ibrutinib therapy for reasons other than CLL progression (atrial fibrillation, n=2). The estimated proportion of patients continuing ibrutinib at 6-months was 71% (95% CI: 45-100%). All 11 are still alive, and no patient has undergone allogeneic stem cell transplantation. Conclusion: This is the first study that describes the outcomes of a large cohort of CLL patients treated with ibrutinib outside the context of clinical trials. Patients treated in routine practice appear more likely to discontinue ibrutinib for reasons other than progression of disease, compared to those participating in clinical trials. Our results confirm the poor survival of CLL patients who progress on ibrutinib. Table 1. Comparative analysis of reasons for discontinuation of ibrutinib in patients with relapsed/refractory CLL: Characteristic Current Study (%) Maddocks (JAMA Oncology 2015) (%) No. of patients 124 308 Median follow-up (months) 6.4 20 Reasons for Discontinuation Progression of disease CLL 2 (2) 13 (4) Richter's transformation 6 (5) 15 (5) Non-progression of disease Infection 3 (2) 28 (9) Other 12 (10) 17 (5) Figure 1. Figure 1. Disclosures Ding: Merek: Research Funding. Kay:Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Membership on an entity's Board of Directors or advisory committees, Research Funding; Hospira: Research Funding; Genentech: Research Funding; Pharmacyclics: Research Funding; Tolero Pharm: Research Funding. Shanafelt:Cephalon: Research Funding; Celgene: Research Funding; Glaxo-Smith_Kline: Research Funding; Polyphenon E Int'l: Research Funding; Hospira: Research Funding; Janssen: Research Funding; Pharmactckucs: Research Funding; Genentech: Research Funding.

Abstract Background: Venous thromboembolism (VTE) is an important cause of morbidity and mortality in patients with hematologic malignancies (HMs). The national burden of inpatient VTE in patients with HMs is not known. Methods: We used the National Inpatient Sample (NIS), one of the largest publicly-available inpatient dataset in United States (U.S.), which represents a 20% stratified random sample of discharges from all hospitals, excluding rehabilitation and long-term acute care hospitals. The NIS is drawn from all States participating in Healthcare Cost and Utilization Project, representing more than 95 percent of the U.S. population. Discharge weights are used to generate national estimates. Patients with HMs were identified using ICD9 codes published by NIS and VTE was defined by ICD9 codes 451 - 453 and 415.1. Annual rates of inpatient VTE were calculated from 1993 to 2012 by HM type. Multivariate odds of inpatient VTE was determined by logistic regression controlling for confounders (age, sex, Charlson co-morbidity index, metastasis, chemotherapy, transfusion of blood products, major operating room procedure, primary hypercoagulable state, disseminated intravascular coagulation, heparin induced thrombocytopenia and thrombosis, antiphospholipid syndrome, obesity and calendar year). The malignancies with non-specific designation (e.g. unspecified leukemia) or with less than 1000 patients (weighted number) were excluded from results. Results: 8.42million hospitalized patients (weighted number) with HMs were identified, and 335,240had VTE (3.98%). In patients with HMs, rates of inpatient VTE increased from 2.59% to 5.22% between 1993 and 2012 (p<0.01). A detailed analysis of VTE rates for various HMs is presented in Table 1. The HMs with highest odds of inpatient VTE are peripheral T-cell lymphoma (PTCL), anaplastic large cell lymphoma (ALCL) and primary central nervous system (CNS) lymphoma (OR 1.56, 1.53, and 1.39, p<0.05) as compared to all other HM. Among leukemias, hairy cell leukemia had higher odds of VTE (OR 1.19, p<0.05). The HMs with the least odds of VTE were acute lymphoid leukemia, chronic monocytic leukemia and plasma cell leukemia (OR 0.65, 0.66 and 0.71, p<0.05). The HMs with the highest rates of VTE were PTCL, large cell lymphoma and ALCL (6.96%, 6.35% and 6.27%). The HMs with lowest rate of VTE were megakaryocytic leukemia, acute lymphoid leukemia and acute monocytic leukemia (1.42%, 2.04%, and 2.73%). Conclusions: Our study provides national estimates of inpatient VTE in HM, which are increasing over the last 20 years. Certain HMs have higher rates and odds of VTE as compared to others. Identification of these differences provides valuable information for the use of prophylaxis in HM with high risk of VTE. Table 1. Multivariate Odds and Rates of Inpatient Venous Thromboembolism in Patients with Hematologic Malignancies (Total weighted N = 8.42 million) Hematologic malignancy Inpatient VTE rate (%) Adjusted OR (95% CI) of inpatient VTE compared to all other HM Hodgkin's disease Hodgkin's disease - lymphocyte rich 3.86 1.15 (0.82 - 1.61) Hodgkin's disease - nodular sclerosing 4.04 1.3 (1.18 - 1.43) Hodgkin's disease - mixed cellularity 4.02 1.22 (0.89 - 1.68) Hodgkin's disease - lymphocyte depleted 3.88 1.14 (0.73 - 1.77) Non-hodgkin's lymphoma Nodular lymphoma 4.89 1.2 (1.13 - 1.28) Mycosis fungoides 4.83 1.25 (1.16 - 1.35) Sezary's disease 3.99 1.08 (0.72 - 1.62) Peripheral t-cell lymphoma 6.96 1.56 (1.32 - 1.85) Burkitt's lymphoma 3.44 1.05 (0.97 - 1.15) Marginal zone lymphoma 5.57 1.07 (0.91 - 1.26) Mantle cell lymphoma 4.91 1.03 (0.92 - 1.15) Primary CNS lymphoma 6.18 1.39 (1.18 - 1.64) Anaplastic large cell lymphoma 6.27 1.53 (1.21 - 1.94) Large cell lymphoma 6.35 1.36 (1.22 - 1.51) Sarcoma Myeloid sarcoma 4.24 1.14 (0.9 - 1.44) Myeloma and leukemia Multiple myeloma 4.57 1.14 (1.1 - 1.18) Plasma cell leukemia 2.85 0.71 (0.5 - 1) Lymphoid leukemia - acute 2.04 0.65 (0.61 - 0.69) Lymphoid leukemia - chronic 3.43 0.76 (0.73 - 0.78) Myeloid leukemia - acute 3.71 0.97 (0.93 - 1) Myeloid leukemia - chronic 2.93 0.72 (0.68 - 0.76) Monocytic leukemia - acute 2.73 0.75 (0.59 - 0.96) Monocytic leukemia - chronic 2.74 0.66 (0.44 - 0.98) Erythrocytic leukemia - acute 2.95 0.81 (0.5 - 1.32) Megakaryocytic leukemia 1.42 0.48 (0.23 - 1.02) Hairy cell leukemia 4.72 1.19 (1.07 - 1.32) Disclosures No relevant conflicts of interest to declare.

Abstract Background: PRCA is characterized by severe normochromic normocytic anemia, reticulocytopenia, and markedly reduced bone marrow erythroid precursors with intact leukocytic and megakaryocytic lineages. There is limited information about the prevalence and outcomes of PRCA in CLL, with the vast majority of information derived from case reports and small series comprising <10 patients. Here, we describe the clinical features, therapy, and outcomes of CLL patients with PRCA seen at our institution. Methods: The Mayo Clinic CLL Database includes all patients with a diagnosis of CLL evaluated at Mayo Clinic Rochester, MN between January 1995 and December 2014. We identified patients diagnosed with PRCA per the following diagnostic criteria: those who presented with normocytic anemia, had decreased erythroid precursors on bone marrow examination, and reticulocytopenia that could not be explained by any other cause besides PRCA. The baseline demographics, clinical characteristics, therapy, and outcomes were abstracted from the electronic medical records. We generated descriptive statistics and calculated overall survival (OS) from diagnosis date (PRCA, autoimmune cytopenia [AID], or CLL) to last known alive or death date. OS was plotted using Kaplan Meier curves and compared between groups using log-rank statistic. The Mayo Clinic IRB approved this study. Results: Of the 3945 CLL patients seen during the study interval, we identified 30 patients who met the diagnostic criteria of PRCA as described in the Methods section. The baseline characteristics, etiology and therapy for PRCA are shown in Table 1. The median time to PRCA from CLL diagnosis was 4.5 years (range-0.1-14 years). Of the 13 patients with IGHV results available, 12 (92%) were IGHV unmutated. Twenty-one (70%) were male and their median age at PRCA diagnosis was 67 years (range 40-81 years). Parvovirus testing was positive in 6 of 24 (25%) patients who were tested; 3 of these patients were treated with IVIG, 2 were treated with prednisone and IVIg, and 1 was treated with prednisone and rituximab. Eight patients (27%) had both PRCA and autoimmune hemolytic anemia (AIHA), and 1 patient (3%) had concomitant PRCA and immune thrombocytopenia (ITP). Of the 27 patients for whom we have treatment data, 10 required only one treatment regimen and 17 required 2-6 treatment regimens, with two patients requiring splenectomy for concomitant AIHA. The median survival of CLL patients after a diagnosis of PRCA was 4.6 years (Figure 1). There was no difference in survival between PRCA cases with and without parvovirus (p=0.92). Compared to newly diagnosed CLL Rai stage III/IV patients, there was no difference in OS (p=0.53). There was no statistically significant difference in survival from an AID cytopenia diagnosis when comparing PRCA, ITP, AIHA, and Evans syndrome (Figure 1; median 5 years for PRCA and 6 years for other groups; p=0.17). Conclusion: PRCA is a rare complication in CLL, occurring in <1% CLL patients seen at our institution over the past 20 years. Parvovirus was the etiologic factor in ~25% of patients and did not influence outcomes. There appears to be no significant difference in outcomes of PRCA patients compared to patients with other AID in CLL. Table 1. Patient Characteristics Characteristic N [range] or (%) Number of patients 30 Males 21 (70) IGHV status* · Unmutated · Mutated 12 (92) 1 (8) FISH results at CLL diagnosis* · 17p- · 11q- · Trisomy 12 · 13q- · Negative 0 (0) 1 (10) 3 (30) 3 (30) 3 (30) At PRCA Diagnosis Median age 67 [40 - 81] Median hemoglobin (gm/dL)* [normal range 12-15.5 gm/dL] 8.2 [3.5 - 10.8] Median absolute reticulocyte count (x 109/L)* [normal range: 38 - 113 x 109/L] 3.5 [0.0 - 25.3] Patients treated for CLL prior to PRCA diagnosis 25 (83) Etiology of PRCA Parvovirus (n = 24) 6 (25) Therapy-related: · Fludarabine · Alemtuzumab/rituximab · Pentostatin/cyclophosphamide/rituximab 3 (10) 1 (3) 1 (3) Autoimmune (neither therapy nor parvovirus related) 19 (63) Initial Therapy of Non-Parvovirus Related PRCA Steroids alone 2 (8) Steroids and Rituximab 3 (13) Steroids in combination with: · IVIG, Chlorambucil · Cyclosporine · Rituximab, IVIg · IVIg, splenectomy 1 (4) 2 (8) 1 (4) 1 (4) Monoclonal Ab alone · Alemtuzumab · Rituximab 1 (4) 4 (17) Cyclosporine 1 (4) Cyclophosphamide 2 (8) RCP (Rituxan, Cyclophosphamide, and prednisone) 2 (8) IVIg 1 (4) Not available 3 (13) *Not available for all patients Figure 1. Figure 1. Disclosures Ding: Merck: Research Funding. Kay:Gilead: Membership on an entity's Board of Directors or advisory committees, Research Funding; Pharmacyclics: Research Funding; Hospira: Research Funding; Tolero Pharma: Research Funding; Celgene Corporation: Membership on an entity's Board of Directors or advisory committees, Research Funding; Genentech: Research Funding. Shanafelt:Cephalon: Research Funding; Hospira: Research Funding; Glaxo-Smith-Kline: Research Funding; Genentech: Research Funding; Jannsen: Research Funding; Celgene: Research Funding; Polyphenon E International: Research Funding; Pharmacyclics: Research Funding.

Abstract Background Severe cytopenias in CLL are typically associated with bone marrow failure due to massive infiltration of the bone marrow by leukemic cells and negatively impact patient morbidity/mortality. Although near complete marrow replacement by CLL is typically observed in patients whose degree of cytopenia meet criteria for advanced Rai stage disease (Hg<11g/dL; platelet <100 x10^9/L), profound variability in the degree of cytopenia and the extent of marrow infiltration exists, and the relationship between these dimensions is not well characterized. Methods We used the Mayo Clinic CLL database to identify all previously untreated CLL/SLL patients who underwent bone marrow biopsies. CLL patients known to have autoimmune-based cytopenias were excluded from this study. Hemoglobin, platelets, absolute neutrophil count (ANC), absolute lymphocyte count (ALC), and absolute monocyte (AMC) at the time of bone marrow biopsy were collected. Characteristics from bone marrow biopsy including marrow cellularity, % CLL involvement, CLL leukemic index (calculated as % cellularity times % CLL), and non-CLL leukemic index (calculated as % cellularity times (1-% CLL)) were also collected. Bone marrow cellularity and % CLL involvement were recorded in the Mayo Clinic CLL database based on clinical estimates made by Mayo hematopathologists. Clinical characteristics and CLL prognostic factors (e.g., Rai stage, CLL FISH, IGHV, ZAP-70, CD38 and CD49d) were also obtained from the database. Age- and sex-adjusted Pearson correlations between the bone marrow measurements and hemoglobin and platelet levels were computed, and scatterplots were created. Results A total of 695 untreated CLL patients diagnosed between 1972 and 2015 who had a CBC test at the time of bone marrow biopsy performed at Mayo Clinic were identified. Median age at diagnosis was 61 years, and 487 (70%) patients were men. Mean time from CLL diagnosis to bone marrow biopsy was 2.2 years. At the time bone marrow biopsy was performed, 20% of patients had advanced stage disease (Rai III/IV) disease, while 44% and 36% had intermediate (Rai I/II) and early stage (Rai 0), respectively. With respect to prognostic parameters, 53% had unmutated IGHV and 20% had high risk CLL FISH (11q- and 17p-). The distribution of other prognostic factors included 43% CD49d+, 35% CD38+, and 42% ZAP-70+ CLL diseases. Median peripheral blood counts were: hemoglobin 13g/dL, platelet 167,000/mL, ALC 26,600/mL, ANC 4100/mL, AMC 700/mL. The median bone marrow cellularity and percent CLL involvement were 70% and 70%, respectively. The median CLL leukemic index was 42%, and median non-CLL leukemic index was 21%. Pearson correlation analysis of bone marrow measurements with peripheral blood hemoglobin found the strongest correlation with the CLL leukemic index (r= -0.49, p<0.001) (Figure), followed by the % marrow CLL involvement (r= -0.45, p<0.001), cellularity (r= -0.40, p<0.001), and non-CLL leukemic index (r= 0.37, p<0.001). Pearson correlation analysis of the bone marrow measurements with platelet count also found the strongest correlation with the CLL leukemic index (r= -0.39, p<0.001), followed by % marrow CLL involvement (r= -0.38, p<0.001), non-CLL leukemic index (r= 0.35, p<0.001), and cellularity (r= -0.28, p<0.001). The peripheral blood ANC levels did not correlate with the four bone marrow measurements. Age- and sex-adjusted Pearson correlations between the bone marrow measurements and hemoglobin and platelet levels were similar to the above unadjusted results. Conclusion To our knowledge, this is the first study to examine the relationship between peripheral blood counts and the extent of bone marrow infiltration in patients with CLL. Although the bone marrow leukemic index had the strongest correlation with peripheral blood Hg and platelet counts, the correlation was modest with an r value of -0.49. These findings suggest that the extent of infiltration of the bone marrow by CLL cells only partially explains of the anemia and thrombocytopenia observed in CLL, indicating other biologic interactions between CLL B-cells and hematopoietic cells contribute to anemia and thrombocytopenia in these patients. More research is needed to determine the nature of these interactions. Disclosures Ding: Merek: Research Funding. Kay:Genentech: Research Funding; Pharmacyclics: Research Funding; Hospira: Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Membership on an entity's Board of Directors or advisory committees, Research Funding; Tolero Pharm: Research Funding. Shanafelt:genentech: Research Funding; Glaxo-Smith-knine: Research Funding; Celgene: Research Funding; Cephalon: Research Funding; Pharmacyclics: Research Funding; Janssen: Research Funding; Hospira: Research Funding; Polyphenon E Int'l: Research Funding.

Abstract Background Although its prevalence is not well defined, transformation to Hodgkin lymphoma (HL) is a rare but well recognized complication in patients with CLL. Prior studies have reported a median survival of 0.8-1.7 years in these patients. Prior therapy of CLL, especially with purine analogs, has been shown to have adverse impact on survival after development of HL. We used the Mayo Clinic CLL database to describe outcomes of CLL patients seen at our institution who developed HL. Methods The Mayo Clinic CLL database includes all consenting patients with a pathologic diagnosis of CLL. Patients who developed biopsy-proven HL were included in this analysis. For patients referred from an outside institution, pathology review at Mayo Clinic was mandatory to be included. Clinical characteristics, therapy and overall survival (OS) were analyzed for all HL patients. Results Among 3124 CLL patients seen at Mayo Clinic from 1/1995 to 7/2011, 26 developed HL. In a nested cohort of 1644 patients who were seen within 12 months of CLL diagnosis and followed prospectively, the prevalence of HL was 0.05%/year (10 year risk=0.5%). All 26 patients had classical HL. The median duration between CLL diagnosis and development of HL was 6.2 years (range, 0-24.5 years). The median age at diagnosis of HL was 68 years (range, 45-88 years) and 21 (81%) were male. At the time of HL diagnosis, the median WBC count was 7.8 x 109/L (range, 1.3-87.3), median hemoglobin was 11 gm/dL (range, 7.2-13), median platelet count was 220x109/L (range, 53-627), median serum lactate dehydrogenase was 249 U/L (range, 133-658), and median serum albumin was 3.4 gm/dL (range, 2.6-4.2). According to the Ann Arbor staging system for HL, 10 (38%) were Stage II, 2 (8%) were Stage III, 11 (42%) were Stage IV, and undetermined in 3 (12%) patients. Sixteen patients (61%) received HL therapy with doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD), seven (27%) received other chemotherapy, and treatment was unknown in 3 (12%) patients. No patient underwent stem cell transplantation as therapy for HL. After a median follow-up of 8.7 years from CLL diagnosis date, 16 (61%) patients have died. The median OS from date of HL diagnosis was 3.9 years (95% CI, 1.0 – 5.0 years). The International Prognostic score (IPS) for HL, originally proposed by Hasenclever in 1998, identified seven prognostic factors: serum albumin <4 gm/dL, hemoglobin <10.5 gm/dL, male sex, age >45 years, stage IV disease, WBC≥15 x 109/L, and lymphocytopenia in patients with newly diagnosed HL. When we applied these prognostic parameters (available in 22/26 patients, 85% in our cohort), the IPS score predicted median survival as follows: score of ≤2 (18% of the patients), median survival not reached; score of 3 (23% of the patients), 6.2 years; score of 4 (36% of the patients), 2.4 years; and score ≥5 (23% of the patients), 0.3 years (p-value=0.005, Figure 1). In contrast, the Richter syndrome prognostic score developed for CLL patients who transform to diffuse large B-cell lymphoma failed to predict survival in these patients (Tsimberidou, JCO, 2006). Of the 26 patients with HL, 16 (61%) had received therapy for their CLL prior to developing HL, including 12 (46%) who had received more than one line of therapy. Purine analogs had been administered for treatment of CLL prior to development of HL in 13/26 (50%) patients. The median survival of HL patients who had previously received CLL therapy was significantly worse compared to those who were untreated for CLL (2.8 years vs. 7.6 years, p-value=0.02). There was a trend toward shorter survival for HL patients who had received previous purine analog therapy for CLL compared to those who received alternate therapies (1.8 years vs. 5 years, p-value: p-value=0.07). Of the 9 HL biopsy samples in which Epstein Barr virus (EBV) in-situ hybridization studies were performed, EBV was detected in 6 (67%) patients. No association was observed between prior purine analog therapy and EBV status (p-value=1.0). There was no difference in survival according to the EBV status of HL (p-value=0.33). Conclusion The 10 year incidence of transformation to Hodgkin transformation among patients with newly diagnosed CLL is 0.5% (∼1/200 CLL patients). In this single-institution analysis, median survival of CLL patients who developed Hodgkin transformation of CLL was approximately 4 years after HL diagnosis. The IPS accurately predicted survival in our cohort of HL patients. Disclosures: Zent: Novartis: Research Funding; GlaxSmithKline: Research Funding; Biothera: Research Funding; Genzyme: Research Funding; Genentech : Research Funding. Shanafelt:Glaxo-Smith-Kline: Research Funding; Cephalon: Research Funding; Hospira: Research Funding; Celgene: Research Funding; Polyphenon E International: Research Funding; Genentech: Research Funding.

Abstract Background Although the clinical importance of del 17p13 in patients with chronic lymphocytic leukemia (CLL) is well recognized, the implications of when this defect is identified are less clearly defined. In particular, the distinction between the identification of del 17p13 at the time of diagnosis and secondary acquisition of del 17p13 during the course of the disease are poorly characterized. Methods We identified all patients with CLL cared for at the Mayo Clinic between 1/1/2000 and 12/31/2012 who underwent baseline fluorescence in-situ hybridization (FISH) testing prior to receiving treatment. The Results of repeat FISH testing were reviewed to identify cases with clinically ascertained clonal evolution. Results A total of 1757 patients with newly diagnosed CLL were seen at Mayo Clinic prior to receiving treatment. Among these, 1243 had FISH testing performed prior to treatment and within 36 months of diagnosis. The median time from diagnosis to initial FISH was 0.8 months (11.5 to 35.4 months). The Results of baseline FISH testing among these patients is as follows: 486 (39%) had del 13q14, 234 (19%) had trisomy 12, 115 (9%) had del 11q23, 59 (5%) had del 17p13, 15 (1%) had other (e.g., del6q) and 334 (37%) had no abnormalities detected. Among these patients, 344 (28%) underwent repeat FISH testing during the course of their disease. Repeat FISH testing was typically performed for clinical suspicion of karyotype evolution or prior to therapy selection in patients with a long time interval between first prognostic FISH and initiation of treatment. Among these 344 patients, 41 (12%) acquired new cytogenetic abnormalities on repeat FISH. Classification of these 344 patients by the Dohner classification at diagnosis and time of follow-up FISH is shown in Table 1. Among the 41 pts who acquired a new FISH detectable genetic abnormality, the newly acquired defect resulted in a change in Dohner FISH risk category for 39 patients including 15/41 (37%) who acquired a del 17p13. Baseline clinical and prognostic characteristics of patients who developed clonal evolution to those who did not are shown in Table 2. Patients with unmutated immunoglobulin heavy chain (IGHV) gene mutation status (p<0.0001) and CD49d (p=0.003) expression were more likely to experience clonal evolution. Among the 319 patients who did not have del 17p13 on baseline FISH, 15 (5%) acquired del 17p13 during the course of their disease. The overall survival from the date del 17p13 was identified is shown for those with del 17p13 at diagnosis (n=59) compared to those who acquired del 17p13 later in the course of the disease (n=15) in Figure 1A. Similarly, among 339 patients who did not have del 6q23 on baseline FISH, 9 (3%) acquired del 6q23 during the course of their disease. The overall survival from the date del 6q23 was identified is shown for those with del 6q23 at diagnosis (n=11) and those who acquired del 6q23 later in the course of the disease (n=9) in Figure 1B. Conclusion Acquired cytogenetic evolution is clinically ascertained by FISH during the course of the disease in approximately 12% of patients. The newly acquired defects in these patients result in a change in Dohner classification for in >90% of these patients including ∼37% who acquire del 17p13. The clinical implications of del 17p13 is influenced by the timing of ascertainment with markedly shorter survival for those who acquire del 17p13 during the course of the disease relative to those with this defect at diagnosis. Disclosures: Shanafelt: Genentech: Research Funding; Glaxo-Smith-Kline: Research Funding; Cephalon: Research Funding; Hospira: Research Funding; Celgene: Research Funding; Polyphenon E International: Research Funding.

Abstract INTRODUCTION: Hemophagocytic lymphohistiocytosis (HLH) is a rare disorder caused by the pathologic activation of the immune system. In children, either a molecular diagnosis consistent with HLH or five out of the following eight criteria are considered necessary for a diagnosis of HLH (HLH-04 criteria): 1) fever; 2) splenomegaly; 3) cytopenia in two or more cell lines; 4) hypertriglyceridemia (≥265 mg/dL) or hypofibrinogenemia (≤150 mg/dL); 5) hemophagocytosis in the bone marrow, spleen, or lymph nodes; 6) hyperferritinemia (≥500 mcg/L); 7) impaired NK cell function; and 8) elevated soluble CD25 (sCD25). These criteria have been extrapolated to diagnose HLH in adults; however, it's unclear if these same criteria are applicable in the adult population. METHODS: We reviewed the Mayo Clinic electronic medical record for all adult (≥18 years) hospitalized patients with an admission serum ferritin of ≥500 mcg/L from January 2012 through December 2014. Patients' charts were reviewed and those who met the HLH-04 criteria were considered to have HLH. For the remainder of the patients, the etiology for hyperferritinemia was determined based on chart review and discharge diagnoses. Logistic regression models were used to assess the ability of these values in predicting a diagnosis of HLH. The Mayo Clinic IRB approved this study. RESULTS: We identified 1,329 patients with a serum ferritin ≥500 mcg/L. Of these, HLH was diagnosed in 28 (2.1%) patients (malignancy-associated HLH in 11 patients, infection-associated HLH in 4 patients, autoimmune-associated HLH in 7 patients, and idiopathic HLH in 6 patients). Table 1 describes the etiology of hyperferritinemia in the remaining 1,301 patients. In contrast to pediatric hospitalized patients (Allen, Pediatr Blood Cancer, 2008), adults are more likely to have malignancy (28.1% vs 7%; p<0.05); bacterial infection (21% vs. 13%; p=0.001); liver disease (9.9% vs. 2.7%; p<0.05); and cardiac disease (7.2% vs. 1.5%; p=0.0001) as the etiology of hyperferritinemia during their hospitalization. Among all patients in the study, the following variables were associated with higher odds of having HLH compared to hyperferritinemia due to an alternate cause: elevated ferritin (odds ratio [OR] 35.97, p<0.01); thrombocytopenia (OR 15.22, p<0.01), cytopenias defined by HLH-04 criteria (OR 8.04, p<0.01); elevated admission serum bilirubin (OR 1.05, p=0.02); and peak serum bilirubin (OR 1.05, p<0.01). After stepwise selection in multivariate analysis, serum ferritin ≥2,600 mcg/L and platelets ≤100 x 109/L were independently associated with HLH diagnosis (OR 24.9 and 7.8 respectively; p<0.01 for both). The area under the curve (AUC or c-statistic) ranges from 0.5 for no ability to discriminate, to 1.0 for perfect discrimination; this model has an AUC of 0.91, which shows very good discrimination between cases and controls. An adult hospitalized patient with a combination of serum ferritin ≥2,600 mcg/L and platelets ≤100x109/L predicts a ~200 fold increased likelihood of being diagnosed with HLH as compared to hospitalized adult meeting neither of these criteria (Figures 1 and 2). CONCLUSION: In contrast to hospitalized pediatric patients, the etiology of hyperferritinemia in adults is more likely to be malignancy, bacterial infection, liver disease, and cardiac disease. We conclude that a combination of serum ferritin ≥2,600 mcg/L and platelet count ≤100 x 109/L can be used as screening criteria to help identify adult patients most likely to have HLH. The traditional 5/8 criteria to diagnose HLH in pediatric patients do not appear necessary for establishing a diagnosis of HLH in adult patients. These observations need replication in an independent data set prior to broader applicability. Table 1.Patients with ferritin > 500 mcg/L with underlying cause for elevation.Number Diagnosed (%)Cardiac Disease95 (7.2%)Liver Disease131 (9.9%)Renal Disease87 (6.6%)Infectious343 (25.8%)Malignancy373 (28.1%)Autoimmune118 (8.9%)Solid Organ Transplant34 (2.6%)Stem Cell Transplant16 (1.2%)Bone Marrow Failure9 (0.7%)Shock39 (2.9%)Idiopathic53 (4.0%)Hemoglobinopathies3 (0.2%) Figure 1. ROC curve for the multivariable model with ferritin > 2,600 (yes/no) and platelets below 100,000 (yes/no) in the model with a very good discrimination c-statistic of 0.91. Figure 1. ROC curve for the multivariable model with ferritin > 2,600 (yes/no) and platelets below 100,000 (yes/no) in the model with a very good discrimination c-statistic of 0.91. Figure 2. Figure 2. Disclosures No relevant conflicts of interest to declare.

Abstract Abstract 2902 Background: Although chronic lymphocytic leukemia (CLL) is a low grade B-cell malignancy, 2–8% of patients (pts) are reported to undergo transformation to diffuse large B-cell lymphoma (DLBCL; Richter's syndrome [RS]). Whether the risk of transformation is related to the underlying biology of CLL or is a consequence of treatment for CLL is uncertain. Also, the role of prognostic markers such as immunoglobulin heavy chain (IGHV) gene mutation status, ZAP-70 and CD38 expression, and genetic abnormalities detected by fluorescence in situ hybridization (FISH) in predicting the development of RS is not well defined. The available data on stem cell transplant (SCT) in pts with RS is based on retrospective case series of selected RS pts and is of uncertain relevance to the majority of CLL pts experiencing RS. Therefore, we conducted a cohort study of newly diagnosed CLL pts to explore the incidence and natural history of RS. Methods: The Mayo Clinic CLL database includes all consenting pts with a pathologic diagnosis of CLL. Information on date of diagnosis, treatment history, and disease-related complications is abstracted from clinical records on all pts and then maintained prospectively. We used this database to identify newly diagnosed pts with CLL from January 2000 through July 2011, who were seen at our institution within 12 months of diagnosis. Pts with biopsy-confirmed RS were included in this analysis. For pts referred from an outside institution, pathology review at Mayo Clinic was mandatory to be included in this study. The Mayo Clinic Institutional Review Board approved the study. Results: Between January 2000 and July 2011, 1641 pts with newly diagnosed CLL were seen at Mayo Clinic within 12 months of diagnosis, and this cohort had a median follow-up of 4.0 yrs (range, 0–12.3 yrs). Of these pts, 37 (2.3%) were diagnosed with RS. The median time to development of RS after a diagnosis of CLL was 1.8 yrs (range, 0–11.7 yrs). The incidence of RS at 2 and 5 yrs post CLL diagnosis were 1.3% and 2.1% respectively (Figure). Pts who developed RS had more advanced (Rai III/IV) stage disease (22% vs. 6%; p<0.001) and higher beta 2-microglobulin levels (median 3.1 mg/dL vs. 2.3 mg/dL; p=0.003) at diagnosis of their CLL compared to pts without RS. Pts who developed RS were also more likely to be positive for ZAP-70 (58% vs. 33%; p=0.03), CD49d (77% vs. 31%; p=0.0004), and CD38 (57% vs. 29%; p=0.001), have unmutated IGHV (76% vs. 44%; p=0.003) and have either del(11q22.3) or del(17p13) on FISH analysis (46% vs. 13%; p<0.0001). RS transformation occurred prior to treatment of CLL in 17 (46%) pts and after treatment of CLL in 20 pts (54%). Treatment for CLL was associated with an increased risk of RS (92% vs 34%; p<0.0001). The odds ratio (OR) of RS was increased for pts who received purine nucleoside analogues (OR=3.0; 95%CI=[1.6, 5.9]; p=0.001), alkylating agents (OR=2.5; 95% CI=[1.3, 4.9]; p=0.004), or both (OR=3.3; 95% CI=[1.7, 6.4]; p=0.0003) for treatment of their CLL. Median overall survival after development of RS was 2.1 yrs. Most pts (29/37, 78%) received R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone) therapy. Two (5%) pts underwent an autologous SCT and 5 (14%) underwent an allogeneic SCT. Additionally, 1 (3%) patient received an autologous SCT followed by an allogeneic SCT. By the end of the study period, 17 (46%) pts with RS had died. Conclusion: The overall rate of biopsy-proven RS in this large cohort of pts with newly diagnosed CLL was 2.3% (5 year incidence: 2.1%). Clinical characteristics (Rai stage III/IV) and biologic characteristics of the CLL B-cell (positive expression of ZAP-70, CD38 and CD49d; unmutated IGHV; and either del(11q22.3) or del(17p13) on FISH) at the time of diagnosis were associated with future risk of RS. Exposure to purine nucleoside analogues and/or alkylating agents for treatment of CLL was also associated with increased risk of RS. Disclosures: Zent: GlaxoSmithKline: Research Funding; Novartis: Research Funding; Genentech: Research Funding; Genzyme: Research Funding; Biothera: Research Funding. Shanafelt:Genentech: Research Funding; Glaxo-Smith-Kline: Research Funding; Cephalon: Research Funding; Hospira: Research Funding; Celgene: Research Funding; Polyphenon E International: Research Funding.