Kay, Neil E.

Abstract Introduction. CLL has a highly heterogeneous clinical course. In fact, while some patients experience a very aggressive disease, others may present with an indolent form, and eventually die of “other” causes. Here we report the findings of a prospective cohort study evaluating the comorbidities and causes of death in newly diagnosed patients with CLL. Methods. Clinical information was collected from 1174 patients with newly diagnosed CLL (within 9 months of diagnosis), who consented to participate in a prospectively observational trial with standardized longitudinal follow-up between 01/2002 and 11/2014. All comorbidities were diagnosed before presentation. Results. Baseline characteristics are shown in the Table. Co-existent health conditions included: cardiac (28% of patients), respiratory (18%), endocrinological (14%), rheumatologic (42%), gastrointestinal (33%), genitourinary (35%), and psychiatric diseases (17%), diabetes (10%), hyperlipidemia (41%), other cancers (20%), stroke (3%), hypertension (40%), venous thromboembolism (3%), substance abuse (5%), sexually-transmitted infections (3%), and obesity (32%). After a median follow up of 4 years, 197 patients died and the cause of death could be accurately determined in 135 patients. Among these, death was due to CLL progression in 69 (51%) patients, infections in absence of CLL progression in 7 (5%), other cancers in 29 (21%), and other comorbid health conditions in 30 (22%). CLL patients (N = 1174)Number (%), median [range]Age (years)63 [23-89]Males791 (67)Females383 (33)Rai stage 0604 (51)I-II512 (44)III-IV54 (4)missing4 (1)B2M < 4 mg/dL717 (61)> 4 mg/dL97 (8)missing360 (31)CD49d negative638 (54)positive277 (24)missing259 (22)CD38 < 30%780 (66)> 30%332 (28)missing62 (6)ZAP70 negative650 (55)positive380 (32)missing135 (13)IGHV mutated506 (43)unmutated394 (34)missing274 (24)FISH del13q395 (34)negative173 (14)+12276 (23)del11q90 (8)del17p50 (4)other2 (1)missing188 (16) Conclusions. Despite the advanced age at CLL diagnosis, the frequent presence of comorbidities, and the indolent nature of the disease, CLL progression is the ultimate cause of death in 51% of newly diagnosed CLL patients, with an additional 26% dying of causes potentially related to CLL, such as infections and second cancers. The influence of comorbidities and other clinical factors at diagnosis (e.g. smoking, Charlson Comorbidity Index) on survival and ultimate cause of death are being abstracted, and will be presented at the meeting. Citation Format: Paolo Strati, Kari Chaffe, Sara Achenbach, Timothy Call, Neil Kay, James Cerhan, Susan Slager, Tait Shanafelt. Comorbidity and cause of death in patients with chronic lymphocytic leukemia (CLL). [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 5267. doi:10.1158/1538-7445.AM2015-5267

Abstract Introduction Chronic lymphocytic leukemia (CLL) is a B-cell malignancy that is known to have a familial component to disease risk. Although 31 loci have been found to be associated with CLL risk, the functional variant(s) driving these associations is mostly unknown. Here we set out to identify rare, highly-penetrant, cosegregating, susceptibility variants within the known GWAS discovered loci using whole exome sequencing (WES) data in CLL families from the Mayo Clinic family study of B-cell malignancies. Methods We performed WES on germ line DNA of 93 CLL families with two or more members with CLL, using Agilent capture kits and Illumina HiSeq2000. Bioinformatics analyses leveraged the following software packages: Novoalign, Picard, The Genome Analysis Toolkit (GATK), and the Biological Reference Repository (bioR). Quality control filters were implemented; subjects with mis-specified relationships were removed, as were variants with <75% call rate, <8X coverage, and those identified as sequencing artifacts. Each GWAS locus was defined by +/- 1Mb of the top GWAS hit within the locus. Linkage disequilibrium (LD) was calculated among the single nucleotide variants (SNVs) located within each locus. Potentially functional SNVs were identified based on: a) uncommon in public databases (< 5%), b) cosegregating in at least two CLL families, c) being highly conserved and in coding regions, and d) functional prediction status of deleterious (SIFT Score), damaging (PolyPhen Score), and a moderate, or high variant impact (SNP Effect). Results In our 93 CLL families, we sequenced 443 individuals: 160 with CLL, 73 with monoclonal B cell lymphocytosis (MBL), and 210 relatives that were not diagnosed with CLL or MBL at the time of sequencing. Median age of CLL diagnosis was 59 years (range 34-87), and 56% were male. Among the MBL individuals and relatives, the median age at recruitment was 55 years (range 18-93), and 40% were male. A total of 317,666 SNVs passed our sequencing quality control filters of which 10,731 were within +/- 1 Mb of known GWAS hits from 31 loci. Of these SNVs, 91% were in coding regions, 18% were reported to have high or moderate impact, 6% were estimated to be damaging and 6% were predicted to be deleterious. From these SNVs, we identified 76 putatively functional variants distributed across 25 GWAS loci that were cosegregating in the individuals with CLL or MBL in multiple CLL pedigrees. These SNVs were all located in coding regions with high or moderate impact and were predicted to be damaging and deleterious. Of these 76 variants, 56 had a frequency of <0.005 in 1000 Genomes’ European population while the remaining 20 had a frequency of 1%. Conclusions Through WES, we identified a number of rare, penetrant and potentially predisposing SNVs located within 25 of the 31 CLL GWAS-discovered loci. These segregating variants provide a list for future validation and functional studies. Citation Format: Sara Beiggi, Daniel R. O'Brien, Sara J. Achenbach, Kari G. Chaffee, Timothy G. Call, Neil E. Kay, Tait D. Shanafelt, Julie Cunningham, James R. Cerhan, Celine M. Vachon, Susan L. Slager. Cosegregating variants in chronic lymphocytic leukemia (CLL) families that are located in loci discovered by genome wide association studies (GWAS). [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2764. doi:10.1158/1538-7445.AM2015-2764

Although ibrutinib and idelalisib are profound treatment advances, they will dramatically increase individual out-of-pocket and societal costs of caring for patients with CLL.

Abstract Background Previous literature regarding the association between occupational exposure to agricultural pesticides and non-Hodgkin lymphoma (NHL) risk has been inconsistent, particularly when looking at specific lymphoma subtypes. Most recently, a large European case-control study showed an elevated risk of chronic lymphocytic leukemia (CLL) among those ever exposed to inorganic and organic pesticides with the strongest association among those exposed to organophosphates. However, when restricting the analysis to subjects with high confidence of exposure, these associations were attenuated (Cocco P., 2013). We used case-control data from the Mayo Clinic to evaluate whether there is an increased risk of CLL among individuals with known occupational exposure to agricultural pesticides. Methods We evaluated pesticide exposure and CLL risk in a clinic-based study of newly diagnosed CLL cases and frequency-matched controls enrolled at the Mayo Clinic from 2002-2012. 200 CLL cases and 474 controls returned a detailed farming and pesticide exposure questionnaire that was modeled after the Agricultural Health Study, where exposure was based on pesticide-specific application. Unconditional logistic regression, adjusted for age and sex, was used to estimate odds ratios (ORs) and 95% confidence intervals (CI). We evaluated the effect of each pesticide individually, by chemical class of pesticides, and by an aggregate of multiple pesticide exposure on CLL risk. Results Among those participants with detailed farming data, the mean age of CLL diagnosis was 66 years and 76% were male; for controls, the mean age at enrollment was 65 years and 66% were male. Risk of CLL was mildly elevated, but not significant, amongst those with application of any pesticide (OR 1.39; 95% C.I. 0.92-2.10). When broken down by chemical class, risk of CLL was also mildly elevated, but not significant, for application of organophosphates and carbamates (OR 1.21; 95% CI 0.84-1.73; OR 1.21; 95% 0.85-1.73, respectively). Organochlorines, phenoxy compounds, pyrethriods and triazines showed no evidence of an elevated risk, nor did we see evidence of a trend that CLL risk increases with number of pesticides applied. Discussion Our preliminary results provide inconclusive support for the role of pesticide exposure as a possible risk factor for CLL. However our sample size was small. Additional data will be incorporated, including confidence of exposure measures, as well as other modifying factors, to further evaluate these results. Citation Format: Emily J. Hallberg, Sara J. Achenbach, Kari G. Rabe, Timothy G. Call, Cristine Allmer, Tait D. Shanafelt, Mark Liebow, Neil E. Kay, James R. Cerhan, Susan L. Slager. Occupational exposure to agricultural pesticides and CLL risk. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 274. doi:10.1158/1538-7445.AM2014-274

Abstract B cell chronic lymphocytic leukemia (CLL) is an incurable disease and represents a significant health problem in the western world. We and others have reported that primary CLL B cells spontaneously produce increased levels of proangiogenic basic fibroblast growth factor (bFGF) in vitro and that CLL plasma contained elevated levels of bFGF. However, the precise role of bFGF in CLL pathobiology is not clearly understood. In this study we investigated functional implication of the FGF/FGFR signaling axis in CLL B cell biology. We have detected expression of FGFR1 and FGFR3 with comparatively higher levels of the latter receptor, but not FGFR2/FGFR4, in primary CLL B cells. Interestingly, constitutively phosphorylated FGFR and its downstream signaling adaptor protein fibroblast receptor substrate 2-α were also detected in CLL B cells. However, in vitro stimulation of FGFR with recombinant bFGF was unable to increase endogenous phosphorylation levels of FGFR in CLL B cells from most CLL patients tested, suggesting existence of an autocrine FGF/FGFR signaling loop in the leukemic B cells. We know that Axl a known predominant RTK in CLL B cells seems to act as a docking site for multiple cellular kinases/lipase. Therefore, in an effort to determine whether Axl is functionally associated with FGFR in CLL B cells, we first examined if these two RTKs exist in the same molecular complex. Indeed we found that Axl forms a complex with FGFR3 in CLL B cells and is therefore likely to regulate FGFR signaling. Our hypothesis in this regard was further supported when we inhibited Axl phosphorylation in CLL B cells using a high-affinity Axl inhibitor (TP-0903; Tolero). Axl inhibition resulted in significant reduction of FGFR phosphorylation in CLL B cells. Finally, inhibition of FGFR signaling using a high-affinity FGFR inhibitor induced significant levels of apoptosis in CLL B cells (n=20) with a mean LD50 dose of ∼2.5 μM. Of interest, LD50 dose for CLL B cell apoptosis was much lower for the Axl inhibitor (mean LD50=140nM; n=20) compared to the FGFR inhibitor, which we believe occurs partly because TP-0903 seems to exert its inhibitory effect on P-FGFR likely via P-Axl inhibition in CLL B cells. In conclusion, we have detected expression of constitutively active FGFR in primary CLL B cells. Interestingly, our preliminary findings suggest that Axl forms an active RTK complex with FGFR and thus likely regulates FGFR signaling in these leukemic B cells. This is evidenced by the finding that Axl inhibition significantly reduced FGFR phosphorylation accompanied by massive induction of CLL B cell apoptosis, suggesting that targeting dual RTK signaling pathways (Axl and FGFR) in CLL B cells warrants further investigation as a way to develop a more effective and efficient therapeutic intervention for CLL patients. Citation Format: Sutapa Sinha, Justin Boyson, Charla Secreto, Steven L. Warner, Neil E. Kay, Asish K. Ghosh. Fibroblast growth factor receptor is expressed as a constitutively active receptor tyrosine kinase in chronic lymphocytic leukemia B cells and exists in an active complex with Axl: Dual targeting in CLL. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3448. doi:10.1158/1538-7445.AM2014-3448

BACKGROUND Although hypogammaglobulinemia is a well recognized complication in patients with chronic lymphocytic leukemia (CLL), its prevalence at the time of CLL diagnosis, and association with novel prognostic markers and clinical outcome is not well understood. METHODS All patients at the Mayo Clinic between January 1999 and July 2013 who had newly diagnosed CLL and had a baseline assessment of serum immunoglobulin G (IgG) were included. The relation between hypogammaglobulinemia at diagnosis and the novel prognostic parameters time to first treatment (TFT) and overall survival (OS) were evaluated. RESULTS Of 1485 patients who met the eligibility criteria, 382 (26%) had hypogammaglobulinemia (median IgG, 624 mg/dL), whereas the remaining 1103 patients (74%) had normal serum IgG levels (median IgG, 1040 mg/dL). Patients who had hypogammaglobulinemia at diagnosis were more likely to have advanced Rai stage (III‐IV; P = .001) and higher expression of CD49d ( P < .001) compared with patients who had normal IgG levels. Although the median TFT for patients who had hypogammaglobulinemia was shorter compared with that for patients who had normal IgG levels (3.8 years vs 7.4 years; P < .001), on multivariable analysis, there was no difference in OS between these 2 groups (12.8 years vs 11.3 years, respectively; P = .73). Of 1103 patients who had CLL with normal IgG levels at diagnosis and who did not receive CLL therapy, the risk of acquired hypogammaglobulinemia was 11% at 5 years and 23% at 10 years. CONCLUSIONS Hypogammaglobulinemia is present in 25% of patients with newly diagnosed CLL. Approximately 25% of patients who have CLL with normal IgG levels at diagnosis will subsequently develop hypogammaglobulinemia on long‐term follow‐up. The presence of hypogammaglobulinemia does not appear to impact overall survival. Cancer 2015;121:2883–2891. © 2015 American Cancer Society .

Abstract Purpose: B-cell chronic lymphocytic leukemia (CLL) is an incurable disease despite aggressive therapeutic approaches. We previously found that Axl receptor tyrosine kinase (RTK) plays a critical role in CLL B-cell survival. Here, we explored the possibility of using a high-affinity Axl inhibitor as a single agent or in combination with Bruton's tyrosine kinase (BTK) inhibitors for future clinical trial to treat patients with CLL. Experimental Design: Expression/activation status of other members of the TAM (e.g., Tyro3, Axl, and MER) family of RTKs in CLL B cells was evaluated. Cells were treated with a high-affinity orally bioavailable Axl inhibitor TP-0903 with or without the presence of CLL bone marrow stromal cells (BMSCs). Inhibitory effects of TP-0903 on the Axl signaling pathway were also evaluated in CLL B cells. Finally, cells were exposed to TP-0903 in combination with BTK inhibitors to determine any synergistic/additive effects of the combination. Results: CLL B cells overexpress Tyro3, but not MER. Of interest, Tyro3 remains as constitutively phosphorylated and forms a complex with Axl in CLL B cells. TP-0903 induces massive apoptosis in CLL B cells with LD50 values of nanomolar ranges. Importantly, CLL BMSCs could not protect the leukemic B cells from TP-0903–induced apoptosis. A marked reduction of the antiapoptotic proteins Mcl-1, Bcl-2, and XIAP and upregulation of the proapoptotic protein BIM in CLL B cells was detected as a result of Axl inhibition. Finally, combination of TP-0903 with BTK inhibitors augments CLL B-cell apoptosis. Conclusions: Administration of TP-0903 either as a single agent or in combination with BTK inhibitors may be effective in treating patients with CLL. Clin Cancer Res; 21(9); 2115–26. ©2015 AACR.

Although transformation to Hodgkin lymphoma (HL) is a recognized complication in patients with chronic lymphocytic leukemia (CLL), its incidence, clinical characteristics and outcomes are not well defined. We used the Mayo Clinic CLL and Lymphoma Databases to identify CLL patients who developed biopsy‐proven HL (CLL/HL) on follow‐up, as well as cases of de novo HL (i.e., without prior CLL). Among 3887 CLL patients seen at Mayo Clinic from January 1995 through August 2011, 26 (0.7%) developed HL. In a nested cohort of 2,465 newly diagnosed CLL patients followed prospectively, the incidence of HL was 0.05%/year (10 year risk = 0.5%). The median overall survival (OS) from date of HL diagnosis in patients with CLL/HL was 3.9 years compared to not reached for de novo HL patients ( n = 709) seen during the same time interval ( P < 0.001). The shorter OS of CLL/HL patients persisted after adjusting for differences in age and Ann Arbor stage of disease. The International Prognostic score (IPS) developed for de novo HL stratified prognosis among CLL/HL patients with median survival of not reached, 6.2, 2.4, and 0.3 years ( P = 0.006) for those with IPS scores of ≤2, 3, 4, and ≥5, respectively. In summary, approximately 1 of every 200 CLL patients will develop HL within 10 years. Survival after HL diagnosis in patients with CLL is shorter than de novo HL patients. The IPS for de novo HL may be useful for stratifying survival in CLL/HL patients. Am. J. Hematol. 90:334–338, 2015. © 2015 Wiley Periodicals, Inc.

Key Points BCL2 mutations in FL correlate with activation-induced cytidine deaminase expression and frequently alter the amino acid sequence of the protein. Mutations in the BCL2 coding sequence at diagnosis are associated with shortened time to transformation and earlier death due to lymphoma.