Slager, Susan L.

Abstract Introduction: CLL phenotype (CD5+CD23+CD20dim) MBL is present in 5-10% of the general population. The prevalence and outcomes of adults with non-CLL like MBL (including CD5+CD23-, CD5+CD20bright, and CD5-) are not known. We screened for MBL using highly sensitive flow-cytometry in a clinic-based cohort of adults from the Mayo Clinic Biobank. Methods: The Mayo Clinic Biobank is a large scale bio-repository of 50,000 adults seen in primary care-based clinics. Herein, we focused on those adults ≥40 years who had stored peripheral blood mononuclear cells available (10% of the overall cohort). We screened for MBL using an 8 color (CD38, CD45, Kappa, Lambda, CD19, CD23, CD5 and CD20) flow cytometry assay validated to detect clonal B-cell events to the 0.005% level (1/20,000 events). Chi-square tests were used to assess associations with demographic characteristics. Cox regression models were used to estimate hazard ratios (HR) and 95% confidence intervals (95% CI) for death. Results: A total of 2594 adults were screened; their median age was 63 years (range, 40-96) and 1111 (43%) were male. 2445/2594 (94%) individuals had interpretable flow cytometry results. A total of 336 (14%) individuals had MBL identified, of whom 276 (82%) had CLL-like MBL and 60 (18%) had non-CLL like MBL (including 39 adults with CD5- and 21 with CD5+CD23- or CD5+CD20 bright phenotype). A strong association between age and the prevalence of non-CLL like MBL was observed, with 0/278 (0%) in 40-49 years, 6/613 (1%) in 50-59 years, 20/674 (3%) in 60-69 years, 21/490 (4%) in 70-79 years, and 13/114 (11%) ≥80 years demonstrating non-CLL like MBL (p<0.001). Non-CLL like MBL was more common in men compared to women (5% vs. 1%; p<0.001). Compared to adults with no MBL, a higher proportion of non-CLL like MBL had a first-degree relative with a lymphoproliferative disorder (8% vs. 17%; p=0.03), and prior non-hematologic cancer (22% vs. 38%, p=0.006). The remainder of the present analysis is focused on adults with non-CLL like MBL (n=60). After a median follow-up of 5 years (range 0-8 years), 3/60 (5%) adults with non-CLL like MBL developed a hematologic malignancy (one myelodysplastic syndrome, two mantle cell lymphoma) compared to 5/2109 (0.2%) adults with no MBL (two large cell lymphoma, one multiple myeloma, one acute myeloid leukemia, one myelofibrosis). Among adults with non-CLL like MBL, the estimated 7-year overall survival (OS) was 79% compared to 94% in those with no MBL (p<0.001, HR for death 3.8, 95% CI 1.98-7.3). After adjusting for age, sex, and self-reported prior non-hematologic cancer, the HR for death was 1.50 (95% CI 0.74-3.05, p=0.26). Conclusion: In this large study, ~1 of 40 adults seen in a primary care setting had non-CLL like MBL. The prevalence of the non-CLL phenotype MBL increased with age. With current follow-up, the risk of progression to an overt lymphoproliferative neoplasm appears to be low. Citation Format: Sameer A. Parikh, Sara J. Achenbach, Kari G. Chaffee, Neil E. Kay, Connie E. Lesnick, James R. Cerhan, Curtis A. Hanson, Tait D. Shanafelt, Susan L. Slager. The prevalence of non-chronic lymphocytic leukemia (CLL) phenotype monoclonal B-cell lymphocytosis (MBL) in a population-based cohort of U.S. adults [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3256.

Abstract Background: We previously identified 5 SNPs at 4 susceptibility loci for diffuse large B-cell lymphoma (DLBCL) in individuals of European ancestry through a large genome-wide association study (Cerhan et al, Nat Gen 2014;46:1233-8); however, much of the heritability remains unexplained. To further elucidate genetic susceptibility to DLBCL, we sought to validate 2 loci at 3q13.33 and 3p24.1 that were suggestive in the original report. Methods: We selected two SNPs rs9831894 (3q13.33) and rs6773363 (3p24.1) for de novo replication from studies at the Mayo Clinic, MD Anderson, and Memorial Sloan Kettering. Logistic regression was used to estimate odds ratios (ORs), using a log-additive model, and adjusting for age, gender and significant eigenvectors for each genotyping center separately. For both SNPs, we then conducted a meta-analysis of all replication studies (n=3) and our original GWAS studies (n=4), encompassing in total 5662 cases and 9237 controls for rs9831894, and 5510 cases and 12,817 controls for rs6773363. Meta-analysis was conducted using the fixed-effects inverse variance method based on the β estimates and standard errors from each study. Results: In a meta-analysis of the 4 original GWAS scans (3856 cases and 7666 controls), rs9831894 (MAF=0.40) was associated with DLBCL risk (OR=0.84, P=4.52 x 10-9) and this replicated in a meta-analysis of the 3 studies with de novo genotyping (OR=0.80, P=4.17 x 10-5); the overall meta-analysis showed a strong association with DLBCL risk (OR=0.83, P=3.62 x 10-13). This locus maps near a plausible candidate gene, CD86, a protein coding gene and a member of the immunoglobulin superfamily that encodes a type I membrane protein. Binding of CD86 with cytotoxic T-lymphocyte-associated protein 4 negatively regulates T-cell activation and diminishes the immune response, while binding of CD86 with CD28 antigen is a costimulatory signal for activation of the T-cell. In a meta-analysis of the same 4 GWAS scans, rs6773363 (MAF=0.45) was tentatively associated with DLBCL risk (OR=1.17, P= 3.68 x 10-7) and this replicated in a meta-analysis of the studies with de novo genotyping (OR=1.27, P=3.78 x 10-7); the overall meta-analysis showed a strong association with DLBCL risk (OR=1.20, P=2.31 x 10-12). This locus also maps near a plausible candidate gene, eomesodermin (EOMES), a transcription factor crucial for embryonic development of the central nervous system and also putatively involved in T-cell differentiation in viral infection defense. Conclusion: In this follow-up analysis of our initial GWAS, we have identified two additional loci associated with risk of DLBCL, the most common lymphoma subtype. These loci provide additional evidence for the role of immune function in the etiology of DLBCL. Citation Format: Geffen Kleinstern, Michelle Hildebrand, Vijai Joseph, Sonja I. Berndt, Hervé Ghesquières, James McKay, Sophia S. Wang, Alexandra Nieters, David Cox, Alain Monnereau, Angela R. Brooks-Wilson, Qing Lan, Mads Melbye, Rebecca D. Jackson, Lauren R. Teras, Mark P. Purdue, Claire M. Vajdic, Demetrius Albanes, Anne Zeleniuch-Jacquotte, Simon Crouch, Yawei Zhang, Susan L. Slager, Xifeng Wu, Karin E. Smedby, Gilles Salles, Christine F. Skibola, Nathaniel Rothman, Stephen J. Chanock, James R. Cerhan. Inherited variants at 3q13.33 and 3p24.1 influences risk of diffuse large B-cell lymphoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 221.

Abstract CLL is a highly heritable cancer with first degree relatives of CLL cases having a 7.5-fold increased CLL risk. Genome-wide association studies (GWAS) and linkage studies have been performed to study inherited predisposition; however a larger proportion of heritability to CLL remains unexplained. Rare coding variants might account for the missing heritability information. Inherited loss of function variants in shelterin complex genes (POT1, ACD, TERF1, TINF2, TERF2, TERF2IP- involved in telomere regulation), CDK1 (critical for cell division) and ATM (tumor suppressor gene) have been found to co-segregate in CLL families and be enriched in CLL cases using exome-wide sequencing data. Our study evaluates rare germline variants from these suspect genes segregating in CLL families who are followed at the Mayo Clinic. Using whole exome sequencing (WES), we sequenced 93 CLL families with at least 2 reported CLL cases consisting of 443 individuals: 160 with CLL, 73 with monoclonal B-cell lymphocytosis (MBL), and 210 relatives. DNA was extracted from buccal cells, coding exons were selectively captured using Agilent 50Mb and SureSelect Human All Exon V4 capture kits; sequencing was performed using Illumina HiSeq 2000. Mayo Clinic's DNASeq pipeline uses Novoalign (initial read alignment), Picard (marking duplicate reads), and the Genome Analysis Toolkit (GATK) for local realignment, recalibration, and variant calling. The variant discovery step leverages GATK's HaplotypeCaller in per sample mode and all of the samples across the cohort are jointly genotyped together. All called variants are evaluated with GATK's Variant Quality Score Recalibration tool and annotated for biological relevance (BIOR). Quality control included removing variants that had <75% call rate across the two capture kits, <8x coverage, or phred score<10, resulting in 317,666 remaining variants. Of these, over 80% of the coding sequence had a median read depth of 23 reads. In our pedigrees, we searched for rare variants within the genes described above. We identified suspect variants with the following criteria: 1) enriched in CLL and MBL samples compared to unaffected samples; 2) multiple affected members with the variant within a family; 3) variants present in all sequenced affecteds within the family; 4) rarely seen in an in-house database of non-cancer controls or 1K Genomes; and 5) predicted to have a functional damaging effect (using SIFT). We identified three novel rare missense variants, defined as functionally deleterious, which each co-segregated within a CLL family. Specifically, these variants from shelterin complex genes; POT1 (rs116916706), TERF2IP (rs138458227), and TERF2 (rs749171225), met the criteria. This study further highlights telomere dysregulation as a key process in CLL development. Investigating rare variants within CLL pedigrees with WES can help identify germline variants impacting predisposition to familial CLL. Citation Format: Alyssa I. Clay-Gilmour, Daniel R. O'Brien, Sara J. Achenbach, Celine M. Vachon, Kari G. Chaffee, Timothy G. Call, Jose F. Leis, Aaron D. Norman, Brian F. Kabat, Sameer A. Parikh, Neil E. Kay, Esteban Braggio, James R. Cerhan, Susan L. Slager. Rare germline variants segregating in chronic lymphocytic leukemia (CLL) families [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1226.

Abstract Chronic lymphocytic leukemia risk variants are associated with alteration of epigenetic state in regulatory regions The vast majority of disease-associated single-nucleotide polymorphisms (SNPs) identified by genome-wide association studies (GWASs) reside in noncoding genomic regions. Functional impact of noncoding casual variants on target gene expression often involves regulatory mechanisms, typically through altering local chromatin accessibility, histone modifications, transcription factor (TF) binding affinity and chromatin interactions. In chronic lymphocytic leukemia (CLL), we and others have identified 41 GWAS risk variants, whose mechanism of action in CLL etiology remains largely unknown. In this study, we aim to decipher the regulatory potential of these variants and to better understand the mechanisms contributing to dysregulation of genes implicated in CLL pathogenesis. We performed ChIP-seq for H3K4me3, H3K4me1, H3K27ac and H3K27me3 in CLL. In addition, we also analyzed over 250 public ChIP-seq, ATAC-seq, DNase-seq, FAIRE-seq, and chromatin interaction data in CLL, B cells and/or lymphoblastoid cell lines. Comparing the 41 index SNPs and the nearby SNPs they associated with in linkage disequilibrium (R2 at least 0.5 in the EUR ethnic group in 1000 Genomes Project) to the above epigenetic data sets revealed overlaps with regulatory regions in 38 of the risk loci, with 28 of the loci overlapping super-enhancers. We used two approaches to elucidate the mechanisms of potential SNP causality. First, we used GATK UnifiedGenotyper to infer genotype for the SNPs that overlapped open chromatin regions and ChIP-seq peaks, by which evidence of imbalance between reference and alternative allele was identified at heterozygous sites. Second, for SNPs located within open chromatin regions or ChIP-seq peaks, we used FIMO program to scan the 100bp sequences centered on the SNPs for significant matches (p=1e-05) to the position weight matrices in five TF binding motif databases; SNPs that had TF motifs altered by the alternative alleles were retained. Together, these two approaches revealed evidence for allele imbalance or TF binding motif alterations at 36 of the 38 CLL risk loci, suggesting the generality of epigenetic regulation in CLL pathogenesis. Further analysis of chromatin interaction, gene expression and expression quantitative trait loci data will likely link the casual SNPs to their target genes. Citation Format: Huihuang Yan, Shulan Tian, Geffen Kleinstern, Susan Slager. Chronic lymphocytic leukemia risk variants are associated with alteration of epigenetic states in regulatory regions [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1221.

SummaryThe effects of ibrutinib on the natural history of autoimmune cytopenias (AIC) among chronic lymphocytic leukaemia (CLL) patients treated in routine clinical practice require further investigation. Using the Mayo Clinical CLL Database, 193 CLL patients treated with ibrutinib between November 2013 and January 2017 outside the context of a clinical trial were identified; complete review of their medical records was performed for details of past history of AIC and treatment‐emergent AIC. We identified 29/193 (15%) patients with history of AIC prior to ibrutinib start. Of 12 patients requiring AIC therapy at ibrutinib start, 8 (67%) were able to discontinue or de‐escalate AIC treatment, and no patient had worsening of their AIC after initiating ibrutinib. Eleven (6%) patients developed treatment‐emergent AIC after a median of 59 (range, 6–319) days following the initiation of ibrutinib, 7 of whom (64%) were able to continue ibrutinib. Overall and event‐free survival from time of ibrutinib start were not significantly different between patients with history of AIC and those with no history of AIC. Treatment‐emergent AIC were seen exclusively in patients with unmutated IGHV and were associated with a shorter EFS. These results suggest a low rate of treatment‐emergent AIC and improvement in patients with existing AIC.

95 Background: While PA is known to improve quality of life in lymphoma pts, its impact on OS is not known. We studied the effect of PA on OS in lymphoma pts enrolled at diagnosis in the Iowa/Mayo Molecular Epidemiology Resource. Methods: Godin Leisure Score Index (LSI; Range 0-50) was calculated based on level of PA pre-diagnosis (baseline) and 3 years post diagnosis (FU3), with pts grouped by tertile into high, moderate, or low PA. At FU3, pts also reported perceived change in PA since diagnosis. OS was measured as time from assessment (baseline or FU3) until death from any cause. Associations between LSI tertiles and OS were assessed using Cox models adjusted for lymphoma subtype, age and sex. Results: PA level was available for 3060 pts at baseline and 1371 pts at FU3. 95% had an ECOG performance status < 2, 37% had an aggressive subtype, 29% had BMI ≥ 30, and 54% were age > 60. Indolent lymphoma pts with higher PA at baseline had no difference in OS. Higher PA at FU3 had improved OS (moderate vs low HR 0.63, CI 0.42-0.93; high vs low HR 0.53, CI 0.33-0.86). Pts with BMI < 30 who had higher PA at baseline showed no difference in OS. Higher PA at FU3 had improved OS for pts with BMI < 30 (moderate vs low HR 0.68 CI, 0.48-0.98; high vs low HR = 0.41, CI 0.26-0.64). For pts > 60 years, high vs low baseline PA showed improved OS (HR 0.77, CI 0.63-0.93). Higher PA at FU3 (moderate vs low HR 0.68 CI 0.48-0.96, high vs low HR 0.53 CI 0.34-0.83) was associated with OS. At FU3, compared to no change, perceived reduction of PA was associated with worse OS for indolent lymphoma (HR 2.13, CI 1.56-2.91), BMI < 30 (HR 1.95, CI 1.45-2.62), BMI > 30 (HR 2.00, CI 1.21-3.30), age ≤ 60 (HR 3.06, CI 1.72-5.46), and age > 60 (HR 1.98, CI 1.52-2.58). Perceived increase in PA was not associated with OS. Higher PA in pts with aggressive lymphoma, BMI ≥30, and age < 60 was not significantly associated with OS but trended similarly. All results were similar when analyzing by continuous LSI score. Conclusions: Higher levels of PA, especially at FU3, were associated with superior OS. The effect was more pronounced in pts with indolent lymphoma, BMI < 30, or age > 60. Perceived reduction in PA was associated with worse outcomes. PA should be incorporated into lymphoma survivorship plans.

SummaryWe evaluated the association of Human Pegivirus (HPgV) viraemia with risk of developing lymphoma, overall and by major subtypes. Because this virus has also been associated with better prognosis in the setting of co‐infection with human immunodeficiency virus, we further assessed the association of HPgV with prognosis. We used risk factor data and banked plasma samples from 2094 lymphoma cases newly diagnosed between 2002 and 2009 and 1572 frequency‐matched controls. Plasma samples were tested for HPgV RNA by reverse transcription polymerase chain reaction (RT‐PCR), and those with RNA concentrations <5000 genome equivalents/ml were confirmed using nested RT‐PCR methods. To assess the role of HPgV in lymphoma prognosis, we used 2948 cases from a cohort study of newly diagnosed lymphoma patients (included all cases from the case‐control study). There was a positive association of HPgV viraemia with risk of lymphoma overall (Odds ratio = 2·14; 95% confidence interval [CI] 1·63–2·80; P < 0·0001), and for all major subtypes except Hodgkin lymphoma and chronic lymphocytic leukaemia/small lymphocytic lymphoma, and this was not confounded by other lymphoma risk factors. In contrast, there was no association of HPgV viraemia with event‐free survival (Hazard ratio [HR] = 1·00; 95% CI 0·85–1·18) or overall survival (HR = 0·97; 95% CI 0·79–1·20) for lymphoma overall, or any of the subtypes. These data support the hypothesis for a role of HPgV in the aetiology of multiple lymphoma subtypes.

Abstract A growing number of loci within the human leukocyte antigen (HLA) region have been implicated in non-Hodgkin lymphoma (NHL) etiology. Here, we test a complementary hypothesis of “heterozygote advantage” regarding the role of HLA and NHL, whereby HLA diversity is beneficial and homozygous HLA loci are associated with increased disease risk. HLA alleles at class I and II loci were imputed from genome-wide association studies (GWAS) using SNP2HLA for 3,617 diffuse large B-cell lymphomas (DLBCL), 2,686 follicular lymphomas (FL), 2,878 chronic lymphocytic leukemia/small lymphocytic lymphomas (CLL/SLL), 741 marginal zone lymphomas (MZL), and 8,753 controls of European descent. Both DLBCL and MZL risk were elevated with homozygosity at class I HLA-B and -C loci (OR DLBCL = 1.31, 95% CI = 1.06–1.60; OR MZL = 1.45, 95% CI = 1.12–1.89) and class II HLA-DRB1 locus (OR DLBCL = 2.10, 95% CI = 1.24–3.55; OR MZL = 2.10, 95% CI = 0.99–4.45). Increased FL risk was observed with the overall increase in number of homozygous HLA class II loci (P trend &lt; 0.0001, FDR = 0.0005). These results support a role for HLA zygosity in NHL etiology and suggests that distinct immune pathways may underly the etiology of the different NHL subtypes. Significance: HLA gene diversity reduces risk for non-Hodgkin lymphoma. Cancer Res; 78(14); 4086–96. ©2018 AACR.

Purpose Selection bias in clinical trials has consequences for scientific validity and applicability of study results to the general population. There is concern that patients with clinically aggressive disease may not have enrolled in recent diffuse large B-cell lymphoma (DLBCL) trials due to the consent process and the inability to delay therapy for eligibility evaluation. We have examined the diagnosis-to-treatment interval (DTI) and its association with clinical factors and outcome in a clinic-based observational cohort of patients with DLBCL from the United States. Validation of results was performed in an independent, clinical trial-based cohort from Europe. Patients and Methods Patients were prospectively enrolled in the University of Iowa and Mayo Clinic Specialized Programs of Research Excellence Molecular Epidemiology Resource (MER; N = 986) or the Lymphoma Study Association (LYSA) LNH-2003 clinical trials program (N = 1,444). All patients received anthracycline-based immunochemotherapy at initial diagnosis. Associations of DTI with clinical factors and outcome were examined. Outcome was assessed using event-free survival at 24 months from diagnosis (EFS24). Results Median (range) DTI was 15 days (0 to 155 days in the MER and 23 days (0 to 215 days) in LYSA. Shorter DTI was strongly associated with adverse clinical factors, including elevated lactate dehydrogenase levels, poor performance status, B symptoms, and higher International Prognostic Index in both cohorts (all P < .001). Longer DTI was associated with improved EFS24 in both the MER (per-week odds ratio, 0.80; 95% CI, 0.74 to .0.87) and LYSA (per-week odds ratio, 0.90; 95% CI, 0.86 to 0.94); association with EFS24 remained significant after adjustment for International Prognostic Index. Conclusion DTI is strongly associated with prognostic clinical factors and outcome in newly diagnosed DLBCL. DTI should be reported in all clinical trials of newly diagnosed DLBCL and future trials should take steps to avoid selection bias due to treatment delay.

Key Points PRS, based on the known CLL loci, predicts CLL risk with high discrimination. This PRS predicts risk of monoclonal B-cell lymphocytosis, a precursor to CLL and a condition that has clinical impact beyond risk for CLL.