Zhu, Xiaochun

Abstract Both genome wide association studies (GWAS) of common variation and exome wide association studies (EXWAS) of rare variation have successfully identified disease susceptibility variants for a variety of diseases. One GWAS of inherited susceptibility to Acute Myeloid Leukemia (AML) has been conducted, but no EXWAS have been performed to measure risk of AML attributable to low-frequency constitutional genetic variation. We performed the first EXWAS of risk of AML as a nested case-control study in the DISCOVeRY-BMT (Determining the Influence of Susceptibility Conveying Variants Related to one-Year mortality after BMT) cohorts. The DISCOVeRY-BMT parent study examined transplant-related mortality in leukemia patients undergoing unrelated donor allogeneic BMT. To identify low frequency variants and genes contributing to increased susceptibility to AML we used genotype data from the Illumina HumanExome BeadChip typed in the DISCOVeRY-BMT cohorts; the HumanExome BeadChip contains 242,901 variants, which are mainly protein-coding variants. The optimal sequence kernel association test (SKAT-O) was used to analyze gene-level associations with risk of AML. These gene-based tests evaluate the cumulative effects of multiple single gene variants on risk of AML. Analyses were performed in all European American AML cases and two subtypes: 1) de novo AML, 2) de novo AML with normal cytogenetics. Models were adjusted for age at transplant and principal components to control for population stratification. For gene-based tests at least 2 variants with minor allele frequency (MAF) ≤ 5%, were required to be present in the gene. This yielded a total of 13,687 genes tested, and a Bonferroni corrected significance level of P<3.65 x 10-6. Association tests were performed in 1,189 AML cases reported to CIBMTR 2000-08 (Cohort 1) and 327 AML cases reported to CIBMTR from 2009-11 (Cohort 2). Controls in Cohorts 1 (n=1,986) and 2 (n= 515) were 10/10 HLA-matched unrelated donors who passed a comprehensive medical exam and deemed healthy. We used metaSKAT to combine Cohorts 1 and 2 and obtain p-values of association with AML. We present the results of gene-level tests significant in both cohorts. The likely pathogenicity of these variants was determined in silico using SIFT, PolyPhen and MutationTaster. Patient characteristics are in Table 1. DNMT3A, on chromosome 2, was associated in the gene-based test with risk of AML (Pmeta=1.70x10-9, Table 2). Three missense variants at MAF <1% comprise both overall AML and de novo AML gene-based association: exm177559 (Asn->Ser), exm177507 (Arg->His), and exm177543 (Arg->Trp). Normal cytogenetics de novo AML gene-based assocations consisted of only 2 of these variants: exm177559 and exm177507 (Table 2). While prevalence of exm177507 is <1% for all AML cases, in de novo AML with normal cytogenetics the MAF was higher at 3%. The other 2 variants had a MAF<1% irrespective of subtype. Somatically, DNMT3A is most frequently mutated in hematologic malignancies, with >30% of de novo AML cases with a normal karyotype and >10% of MDS patients having DNMT3A mutations. Although these are germline gene associations all three of the variants found have been reported somatically in hematologic malignancies. In 200 AML cases from The Cancer Genome Atlas (TCGA) p.R882H (represented as exm177507 on the exome chip) was a frequent somatic mutation (25%). Exm177543 (p.R635W) and exm177559 (p.N501S) are reported in the Catalogue of Somatic Mutations in Cancer (COSMIC) as somatic mutations involved in hematopoietic and lymphoid tissue in both cell lines and humans. Exm177507 and exm177543 show evidence of pathogenicity in all three in silico tools, while exm177559 was reported as deleterious and disease causing by Sift and MutationTaster, respectively. Our results show that multiple potentially pathogenic missense germline variants in DNMT3A comprise the gene-based association with AML, specifically de novo AML with normal cytogenetics. Given the functional nature of these variants it is possible germline risk stratification could be informative in determining AML risk, and subsequently development of AML harboring DNMT3A mutations. Confirmation of these findings in additional cohorts could have implications for individualized risk screening, prediction and prognosis. Additional cytogenetic subgroup analyses, including treatment-related AML, are underway. Disclosures Hahn: Novartis: Equity Ownership; NIH: Research Funding. McCarthy:Sanofi: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Onyx: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bristol Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees; The Binding Site: Honoraria, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Honoraria, Membership on an entity's Board of Directors or advisory committees; Gamida Cell: Honoraria, Membership on an entity's Board of Directors or advisory committees. Sucheston-Campbell:NIH/NCI: Research Funding.

Abstract #the authors contributed equally to this work While survival outcomes after HLA-matched unrelated donor (URD) blood and marrow transplant (BMT) have significantly improved over the last two decades, about 40% of patients die before 1-year post URD BMT. Previously we performed a genome-wide association study (GWAS) named DISCOVeRY-BMT (Determining the Influence of Susceptibility COnveying Variants Related to one-Year mortality after BMT) to investigate the contributions of common genetic variants on survival outcomes. To address the specific contributions of low frequency (≤ 1%) exonic variants on survival outcomes, we used the DISCOVeRY-BMT cohorts typed with the Illumina HumanExome BeadChip containing 242,901 variants, which are mainly protein-coding variants. We used a gene-based test (the optimal sequence kernel association test (SKAT-O)) to evaluate the cumulative effects of multiple genetic variants within a gene on overall survival (OS), transplant-related mortality (TRM), and disease-related mortality (DRM) 1 year after URD BMT. SKAT-O measures the overall significance of the gene and allows the variants to have different directions and magnitude of effects. We used a 2-stage study design: SKAT-O analysis was carried out in Cohort 1 consisting of 1,972 AML, ALL or MDS patients reported to the Center for International Blood and Marrow Transplant Research from 2000 to 2008 and their 2,006 10/10 HLA-matched URD. Genes with a suggestive association with each outcome (P < 1×10-3) entered the final meta-analysis including Cohort 1 and an independent cohort (Cohort 2), consisting of 503 recipients and 520 10/10 HLA-matched URD from 2009 to 2011. The analysis included recipients and URD of European descent due to the low frequency of other populations. All association models included age at BMT, diagnosis (AML, ALL, MDS), disease status at BMT, cell source (peripheral blood, marrow), year of BMT, and principal components from EIGENSTRAT to control for population stratification. Only coding variants (missense and nonsense) with minor allele frequency ≤ 1% were included in our analysis, which resulted in > 12,000 genes tested in our gene-based analysis in recipients and donors. Exome-wide significance was set at Pmeta< 4.10×10-6 after Bonferroni correction for the total number of genes tested. The likely pathogenicity of these variants was determined in silico using SIFT and PolyPhen. In recipients, we identified OR51D1, a member of the olfactory receptor gene family, as significantly associated with both OS and TRM. The OS association with ORD51D1 is driven by TRM as the same six missense variants contribute to both TRM and OS, but not to DRM. In donors, four (ALPP, EMID1, SLC44A5, LRP1), one (HHAT), and two (LYZL4, NT5E) genes were significantly associated with OS, TRM, and DRM, respectively (Table). ALPP encodes an alkaline phosphatase. SLC44A5 is a member of the solute carrier gene family. LRP1 (low density lipoprotein receptor-related protein 1) encodes an endocytic receptor involved in several cellular processes, including intracellular signaling, lipid homeostasis, and clearance of apoptotic cells. HHAT encodes an enzyme that acts within the hedgehog secretory pathway, which is involved in hematologic malignancy. LYZL4 is a member of a family of lysozyme-like genes, which have a protective role in host defense. The encoded protein of NT5E, CD270, is used as a determinant of lymphocyte differentiation. The likely pathogenicity of several of the functional variants in all significant genes identified was predicted to be both (possibly or probably) damaging and deleterious (Table), thus indicating that the amino acid substitution of the variant may not be well tolerated. One damaging variant in LRP1 has been identified as a somatic mutation in hematopoietic and lymphoid tissue. Our study is the first analysis of the low-frequency coding variant contribution to URD BMT survival outcomes. We found that the missense variants contributing to many of these significant gene associations shows evidence of pathogenicity and thus it is biologically plausible these variants are contributing to survival outcomes. Further confirmation of these findings, and studies of the functional consequences of protein-coding changes in these genes, may provide more individualized risk prediction and prognosis as well as alternative donor selection strategies. Table. Table. Disclosures McCarthy: Gamida Cell: Honoraria, Membership on an entity's Board of Directors or advisory committees; Onyx: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Bristol Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Honoraria, Membership on an entity's Board of Directors or advisory committees; The Binding Site: Honoraria, Membership on an entity's Board of Directors or advisory committees; Sanofi: Honoraria, Membership on an entity's Board of Directors or advisory committees. Hahn:Novartis: Equity Ownership; NIH: Research Funding. Sucheston-Campbell:NIH/NCI: Research Funding.

Abstract Over the past decade, there have been numerous published candidate gene association studies of non-HLA variants, genes and/or pathways in relation to BMT survival outcomes. Most of these studies investigated a small number of candidate single nucleotide polymorphisms (SNPs), chosen based on known gene function, in a limited number of cases without replication. Here we test the association of previously reported SNPs with survival outcomes (overall survival (OS), progression-free survival (PFS), death due to disease (DD) or transplant-related mortality (TRM)) after URD BMT using data from the DISCOVeRY-BMT (Determining the Influence of Susceptibility Conveying Variants Related to one-Year mortality after BMT) study which included 2,052 European descent recipient-unrelated donor pairs (Cohort 1) and 763 recipient-unrelated donor pairs (Cohort 1) reported to the Center for International Blood and Marrow Transplant Research (CIBMTR) between 2000-2009 and 2009-2011, respectively. DISCOVeRY-BMT is a Genome Wide Association Study (GWAS) that includes over 9 million typed or imputed SNPs. This represents the largest study to date and is well powered to test both SNP and gene-based association hypotheses. We conducted an extensive literature review of PubMed using MESH terms related to transplant and hematologic disease which yielded 76 candidate gene association studies that investigated a total of 46 genes with any survival outcome. Of the 46 genes, 11 genes (CTLA4, CYP2B6, CYP2C19, GSTM1, GSTT1, IL23R, IL6, MTHFR, NOD2/CARD15, TNFSRF1B, and VDR) were studied in at least 2 publications. A total of 16 SNPs in 8 of these genes reported a significant (P< 0.05) association with at least one survival outcome (OS, PFS, DD or TRM). We tested these 16 SNPs for an association with the survival outcome originally reported in each article (Table 1). All association models included age at BMT, diagnosis (AML, ALL, MDS), disease status at BMT, cell source (blood, bone marrow) and year of BMT. P-values from Cohorts 1 and 2 were combined using a fixed-effects model in METAL software. Since the candidate gene studies have a gene-based hypothesis, we also performed gene-level testing using the VErsatile Gene-Based Association Study (VEGAS2) software to measure the aggregate effect of all available typed and imputed SNPs from our GWAS data in these 8 genes. The candidate gene association studies included heterogeneous patient populations of autologous and/or allogeneic (related or unrelated donors) BMT for a variety of hematologic diseases. Most candidate gene studies performed genotyping on both donors and recipients, and SNPs were tested either independently for donor or recipient genomes, or considering the total number of variant alleles the recipient-donor pair had at a given locus (0-4) (Table 1). The 16 SNPs from the previous candidate gene association studies were not significantly associated at a nominal P<0.05 to either the survival outcome reported in the original published paper, or to any survival outcome (Table 1). Likewise, the VEGAS2 gene-based analysis did not reveal any statistically significant associations for any of the 8 genes at P<0.05. Our study also corroborated prior reports of SNPs and genes that were not associated with survival outcomes. Candidate gene association studies use a priori biological knowledge to select genes that are likely associated with outcome and subsequently SNPs within the gene thought likely to affect function based on knowledge at the time of the investigation. Our findings suggest that neither the genetic variation tested in these candidate genes, nor other individual SNPs in the gene independently or collectively are not associated with survival outcomes in recipients of URD BMT. Our failure to replicate the prior candidate gene studies may be due to these SNPs being dependent on the underlying disease (ie. AML, ALL, etc), the donor relation (related, unrelated, autologous), or that there is no true association of these SNPs with the outcomes studied. #These authors contributed equally Table 1. Table 1. Disclosures McCarthy: Gamida Cell: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Karyopharm: Honoraria, Membership on an entity's Board of Directors or advisory committees; Onyx: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bristol Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; The Binding Site: Honoraria, Membership on an entity's Board of Directors or advisory committees; Sanofi: Honoraria, Membership on an entity's Board of Directors or advisory committees. Hahn:Novartis: Equity Ownership; NIH: Research Funding. Sucheston-Campbell:NIH/NCI: Research Funding.

Key PointsIKZF1 associations with high-risk B-ALL may differ by age and sex. A novel variant on chromosome 14, rs189434316, is associated with over a 3.5-fold risk of normal cytogenetic B-ALL.

Abstract HCT entails substantial TRM justifying continued efforts to better risk stratify patients. Single institutional studies have suggested several clinically available biomarkers of CRP, ferritin and albumin influence TRM if not overall survival (OS). We sought to confirm the independent prognostic value of these biomarkers in a large multi-institutional cohort. The study population consisted of 784 adults with AML in remission (80%) or MDS (20%) undergoing unrelated donor HCT between 2008 and 2010 and available cryopreserved samples through the Center for International Blood and Marrow Transplant Research (CIBMTR) repository. CRP and ferritin were centrally quantified by ELISA from cryopreserved plasma whereas albumin levels obtained from center reported data. Correlation studies for biomarkers required log transformation of CRP and ferritin to produce a normal distribution. Multivariate models were fitted separately for each biomarker applying protocol specified thresholds generated from the literature of CRP > 10 mg/L, ferritin > 2500 ng/mL, albumin < 3.5 g/dL on TRM. Further analysis explored optimal cutpoints for this cohort for all significant clinical variables for TRM. HCT characteristics included a median age of 50 (range 18-78) years, HCT-CI (co-morbidity index) 3 or more in 35%, single allele/antigen mismatch (7/8) in 23%, PB as stem cell source in 83%, and myeloablative conditioning in 72%. Biomarker data were available in 783, 781 and 695 cases for CRP, ferritin and albumin respectively. The median values and ranges for each biomarker were as follows: 5.0 mg/L (0.3 - 316) for CRP, 1148 ng/mL (51 – 14,298) for ferritin and 3.6 g/dL (0.6-5.3) for albumin. Log transformed CRP and ferritin showed a modest correlation (r=0.35, P<0.001), and log CRP was marginally associated with albumin (r=-0.12, P=0.002). HCT-CI had no correlation to CRP and albumin. Higher ferritin was associated with HCT-CI (P=0.014) and disease (P<.001). In the entire population, TRM and overall survival (OS) at 2 years were 23% and 54%, respectively. The table shows the independent association of each biomarker with TRM and OS. Ferritin had no association with these outcomes but only 12% had levels above 2500 by ELISA. The optimal cutpoints for TRM were defined as follows: CRP > 3.67 (HR=1.75, P=0.001); albumin < 3.4 (HR 1.70, P <0.001); and ferritin as a continuous variable (HR 1.30, P = 0.008). A risk score was created modeling optimal biomarker thresholds to predict TRM based on the following formula: risk score = 0.44633*I(CRP>3.67)+0.18330*ln(Ferritin)+0.44730*I(albumin<3.4), where “I” is an indicator function and represent 1 or 0 depending on the presence of the abnormal biomarker level in parenthesis. The risk score included adjustment for HCT-CI. The biomarkers did not influence the incidence of relapse or graft-versus-host disease with any of these models. Elevated CRP and ferritin and decreased albumin prior to HCT were associated with impaired transplant survival, primarily through higher TRM. Integration of a biomarker panel in clinical practice once validated can enhance risk-stratification, improve adjustment for comparative studies and point towards the design of biomarker driven trials. Abstract 422. Table: Multivariate analysis of TRM and OS for biomarkers, and biomarker panel risk score* Abnormal, N (%) Transplant related mortality Overall Survival Biomarker HR 95% CI P HR 95% CI P Protocol defined CRP >10 mg/L 184 (24) 1.52 1.11 – 2.07 0.008 1.22 0.98 – 1.53 0.072 Ferritin >2500 ng/mL 93 (12) 1.26 0.84 – 1.88 0.27 1.15 0.86 – 1.54 0.35 Albumin < 3.5 g/dL 290 (42) 1.48 1.10 – 2.0 0.010 1.39 1.13 – 1.72 0.002 Optimal threshold, combined model ** CRP >3.67 mg/L 497 (64) 1.56 1.11 – 2.19 0.010 1.24 1.00 – 1.55 0.054 Log(Ferritin), linear n/a 1.20 0.99 – 1.46 0.07 1.20 1.04 – 1.38 0.010 <3.4 g/dL 203 (29.5) 1.58 1.16 – 2.15 0.004 1.37 1.10 – 1.72 0.005 Biomarker Score ** N Low (<1.5) 200 1.0 - - 1.0 Intermediate (1.5-2.0) 331 1.66 1.10-2.49 0.015 1.38 1.06-1.80 0.017 High (>2.0) 159 2.72 1.77-4.19 <0.001 2.01 1.51-2.70 <0.001 * biomarkers adjusted for significant covariates of BMI, disease, mismatch, and CMV serostatus and stratified on conditioning regimen ** combined model includes all biomarkers and HCT-CI Disclosures No relevant conflicts of interest to declare.

Key Points We conducted the first exome-wide association study between germ line variant genotype and survival outcomes after unrelated-donor BMT. A number of novel genes were found to significantly affect survival outcomes.

Key PointsCandidate SNP associations with survival outcomes after URD transplant are most likely false-positive findings. Over 85% of candidate SNPs are not linked to a biochemical function; of those that are, about half are not linked to the candidate gene.

Key Points Over the last decade, allogeneic HCT has been increasingly administered in the United States to adults aged 70 and older with hematologic malignancies. Allogeneic transplant outcomes were reasonable; high comorbidity and ablative conditioning regimens were associated with inferior outcomes.

Key Points In combination with cyclophosphamide, intravenous busulfan is associated with better leukemia-free and overall survival in AML than TBI.