Wieben, Eric D.

Abstract The εγδβ thalassemias (EGDBT) are infrequently occurring deletions of the β globin gene cluster on chromosome 11. EGDBTs can be an unsuspected cause of severe neonatal anemia that resolves during the first year of life into a normal Hb A2 thalassemic phenotype, presumed to be due to an imbalance of β-like chains during development. First described in 1972, they are categorized into two groups, both with loss of the five DNase I hypersensitivity sites upstream of the β gene (HBB) called the locus control region (LCR). In group 1 deletions, all or most of the β globin gene cluster including HBB is deleted; in group 2 deletions, removal of the LCR silences an intact HBB. Large deletions that do not involve LCR but remove the embryonic (HBE) and fetal genes (HBG2/HBG1) [LCR(ε Gγ Aγ) 0δβ] have not been reported. It is not known whether a deletion with intact LCR as well as δ and β genes would similarly manifest as neonatal anemia that resolves into a normocytic phenotype with normal hemoglobin fraction percentages after infancy. Four individuals with an LCR(ε Gγ Aγ) 0δβ deletion have been evaluated. These include monochorionic, diamniotic twin girls whose newborn screen (NBS) returned significant for Hb A percentage greater than Hb F for both twins. NBS was repeated two weeks later with the same result. The newborns were induced late pre-term (36 weeks and 6 days) after an uncomplicated pregnancy, including no features of twin to twin transfusion. Both infants had no anemia and bilirubin levels remained in the low to low-intermediate risk zones. Neither twin received transfusions or phototherapy during admission. At a follow up visit at 20 months of age a hemoglobin electrophoresis was performed which showed borderline Hb A2 and no Hb F in both twins (Table 1). Ferritin levels were low-normal in Twin A (14 µg/L) and low in Twin B (9 µg/L). Both infants were found to have a heterozygous large deletion affecting HBE, HBG2, HBG1, and HBBP1 loci by Multiplex Ligation-dependent Probe Amplification (MLPA), a method that uses multiple DNA probe pairs to determine the size of the DNA segment deleted. MLPA deletion/duplication testing on the α genes was negative. No β globin variants were identified by DNA sequencing. An evaluation was performed on the 25 year-old healthy mother and showed similar findings including the heterozygous large deletion. Mother and infants continue to be asymptomatic with normal red cell indices for age (Table 1). The fourth was an unrelated 40-year-old female with mild anemia screened by hemoglobin electrophoresis for obstetric evaluation who was found to have borderline Hb A2 (see table 1). MLPA showed a heterozygous deletion involving the same loci. Again, no β globin variants were identified by DNA sequencing. All four cases showed the same MLPA pattern except one non-specific probe commonly altered in common gamma gene conversion events. Long range sequencing performed on one case confirmed a single contiguous 32,599 bp deletion that matched the MLPA data (g.5262276-5294875). The detected deletion has not been reported in literature and differs from EGDBT mutations in that the LCR is intact. As the LCR controls expression of the entire gene cluster, this mutation is expected to display different phenotypic features than classic EGDBT. HBD and HBB are not deleted and therefore adequate transcription of Hb A2 and Hb A are expected with no imbalance of α/β chains after the neonatal period. It is postulated this finding explains the decreased Hb F levels in the NBS test result. The Hb A2 increase is likely a product of this deletion and does not indicate β-thalassemia. The heterozygous absence of HBE and HBG2/HBG1 may result in upregulation of HBD and HBB expression through loss of FKLF and FKLF-2 binding. In addition, HBB-EKLF binding would have less competition as the gamma promoter elements have been heterozygously deleted. In summary, in contrast to classic EGDBTs which cause transient severe neonatal anemia and resolve to normal Hb A2 thalassemic indices, this novel LCR(ε Gγ Aγ) 0δβ deletion was associated with normal CBC values with an absence of severe neonatal anemia, inverted Hb F/Hb A percentages at birth and borderline Hb A2 levels. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

AbstractTo examine the length of a hexanucleotide expansion in C9orf72, which represents the most frequent genetic cause of frontotemporal lobar degeneration and motor neuron disease, we employed a targeted amplification-free long-read sequencing technology: No-Amp sequencing. In our cross-sectional study, we assessed cerebellar tissue from 28 well-characterized C9orf72 expansion carriers. We obtained 3507 on-target circular consensus sequencing reads, of which 814 bridged the C9orf72 repeat expansion (23%). Importantly, we observed a significant correlation between expansion sizes obtained using No-Amp sequencing and Southern blotting (P = 5.0 × 10−4). Interestingly, we also detected a significant survival advantage for individuals with smaller expansions (P = 0.004). Additionally, we uncovered that smaller expansions were significantly associated with higher levels of C9orf72 transcripts containing intron 1b (P = 0.003), poly(GP) proteins (P = 1.3 × 10− 5), and poly(GA) proteins (P = 0.005). Thorough examination of the composition of the expansion revealed that its GC content was extremely high (median: 100%) and that it was mainly composed of GGGGCC repeats (median: 96%), suggesting that expanded C9orf72 repeats are quite pure. Taken together, our findings demonstrate that No-Amp sequencing is a powerful tool that enables the discovery of relevant clinicopathological associations, highlighting the important role played by the cerebellar size of the expanded repeat in C9orf72-linked diseases.

Purpose: To investigate candidate genes associated with familial strabismus and propose a theory of their interaction in familial strabismus associated with early neurodevelopment. Methods: Eighteen families, including 53 patients diagnosed with strabismus and 34 unaffected family members, were analyzed. All patients with strabismus and available unaffected family members were evaluated using whole exome sequencing. The primary outcome was to identify rare occurring variants among affected individuals and investigate the evidence of their genetic heterogeneity. These results were compared with exome sequencing analysis to build a comprehensive genetic profile of the study families. Results: We observed 60 variants from 58 genes in 53 patients diagnosed with strabismus. We prioritized the most credible risk variants, which showed clear segregation in family members affected by strabismus. As a result, we found risk variants in four genes (FAT3, KCNH2, CELSR1, and TTYH1) in five families, suggesting their role in development of familial strabismus. In other families, there were several rare genetic variants in affected cases, but we did not find clear segregation pattern across family members. Conclusion: Genomic sequencing holds great promise in elucidating the genetic causes of strabismus; further research with larger cohorts or other related approaches are warranted.

Introduction: Globin gene sequencing has long been impeded by repetitive regions, homology and structural variation. Currently, long-read sequencing technology has been limited by high cost and error rate. These challenges can be mitigated by sample multiplexing of targeted capture sequencing and the use of consensus reads, respectively. This ongoing study analyzes the utility of long range sequencing technology in identifying historically problematic beta globin cluster mutations. Materials and methods: Long range sequencing was performed on complex beta globin gene cluster cases utilizing the Sequel platform (Pacific Biosciences, Menlo Park, California, United States). Sample input was 1 microgram of DNA extracted from whole blood. Circular Consensus Sequence (CCS) FASTQ sequences files were mapped to reference genome (GRCh37/hg19) using NGMLR, followed by annotation of variants using a custom bioinformatics workflow. Comparison to multiplex ligation-dependent probe amplification (MLPA), Sanger sequencing, array CGH, short-range NGS (300 bp), hemoglobin protein, and clinical data was performed. Results: CCS read length varied from 1 to 10 kb, average 3.5. Analysis of the beta globin cluster region showed impressive superiority and comprehensiveness to Sanger, MLPA, array CGH and short range sequencing technology: 1) single nucleotide variants (SNVs) were identified to a sensitivity and specificity of 99.5%; 2) large structural variants (SVs) such as large deletions, duplications, insertions, crossovers and fusions spanning many kilobases were characterized in the heterozygous, homozygous and compound heterozygous state to a precise genomic coordinate (table 1); 3) phasing SNVs identified Hb S haplotypes (Central African region (CAR), Benin, Senegal, Arab-Indian and Cameroon). Conclusion: 1) Long range sequencing holds significant promise in genotyping hemoglobin disorders. This method improves efficiency with potential for a single test offering with increased resolution, particularly in large deletion analysis. Complex cases greatly benefited from improved range and resolution. 2) The decreased requirement of blood volume for extensive testing is of particular interest for the pediatric population. 3) Genotyping of Hb S patients detected by positive newborn screening will help predict need for prophylaxis or therapy by assessing variant zygosity, phasing and haplotypes. In conclusion, we believe that long range sequencing has immense potential to contribute extensively to thalassemia and hemoglobinopathy diagnostics and is far superior to the currently available technologies for beta globin cluster analysis. Disclosures Jen: Celgene: Employment.

Abstract Background The mutational landscape in acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) is becoming increasingly apparent. Although some mutations and translocations occur in a substantial fraction of patients, in many, the etiologic mutations are unknown and current cytogenetic testing uninformative. Recent whole genome studies show that the mutational landscape can be very diverse. Genomic studies in families at high risk of developing MDS and AML can be highly informative in identifying potentially oncogenic mutations. We have identified a family with a high incidence of MDS/AML and here report on their unique mutational profile. Patients and Methods An index patient presented to our institution with AML with myelodysplasia related changes (AML-MRC) who required allogeneic stem cell transplantation (SCT). Subsequently, 3 other family members developed MDS. DNA from the index case before and after SCT, 3 affected family members, along with 3 healthy family members (controls) was isolated. Whole exome sequencing was performed using the Agilent V4+UTR capture kit. Sequencing of the index case post SCT served as an additional control (100% donor chimerism). Potentially damaging and novel variations in the index case and additional affected patients were detected by comparing their variant files with those of their unaffected relatives. We further excluded variants with an observed MAF greater than 0.01 using annotated populations in HAPMAP, 1k Genomes, Danish-BGI and ESP5400_EU. All variants with a high or moderate effect based on SNPeff were considered. Results We sequenced an average of 1 billion base pairs per case, with an at least 30 fold exome coverage. The average number of INDELs was 206/patient (range, 187-214), while the average number of SNV was 142/patient (range, 139 – 153). Sixteen mutations were identified in the patients that were not present in controls and in the post SCT sample. This frequency is similar to that recently reported by the Cancer Genome Atlas Research Network. However, none of the mutations we identified have been reported previously in AML/MDS. We identified a heterozygous mutation in the NAPRT1 gene that led to a premature stop codon (R405*). A mutation in the splice site donor of PODNL1 was also identified, together with non-synonymous mutations in IL36RN, ABCA12, GOLGB1, PLXNA1, ABLIM3 and others. We found no correlation between total SNV and total reads mapped (r=0.48). Given the potential role of NAPRT1in genome repair, we evaluated the frequency of SNV in individuals with and without this mutation. There was no difference in SNV frequency in heterozygotes compared to controls (p=0.45) and the frequency of SNV in patients with the R405* mutation did not increase with age. These observations argue against this mutation as a cause of genomic instability. We found a bimodal distribution of SNV (0.14% versus 0.08%, p=0.00024) that did not correlate with the presence/absence of disease. The mechanism behind this distribution is being evaluated. Conclusions The mutational landscape of MDS/AML continues to expand. The functional significance of mutations identified still has to be understood at the molecular level. Although some mutations in AML may be recurrent, it is likely that a substantial number of patients with leukemia have ‘private’ mutations that are oncogenic. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 633 Background: To date, the only well characterized chromosomal translocations among peripheral T-cell lymphomas (PTCLs) are those involving ALK, which occur in about half of anaplastic large cell lymphomas (ALCLs). ALK-negative ALCLs remain poorly understood, despite morphology and phenotype similar to ALK-positive ALCLs. We recently described translocations near IRF4 on 6p25.3 in cutaneous ALK-negative ALCLs. In the present study, we utilized massively parallel (Next Generation) sequencing to answer three questions about these translocations: (1) Were they recurrent in the more lethal systemic form of ALK-negative ALCL? (2) What was the translocation partner? (3) What were the likely genetic consequences? Methods: We used a mate pair library approach to juxtapose DNA sequences originally lying ~5 kb apart: these paired sequences map to distinct areas of the genome if a translocation breakpoint lies between them. Because only the ends of the original DNA fragments are sequenced, this approach is more resource efficient than whole genome sequencing. Genomic DNA from a systemic ALK-negative ALCL with an uncharacterized 6p25.3 rearrangement was fragmented to 2–5 kb, biotinylated, and separated on an agarose gel. Fragments ~5-5.5 kb were excised, purified, circularized, and further fragmented to ~300-600 bp. Fragments containing biotin (representing the ends of the original 5 kb fragments) were immobilized on streptavidin beads and amplified using the Illumina library protocol. Paired end sequencing was performed on one lane of a flow cell on an Illumina GAIIx. Sequencing data were mapped to the genome using a binary indexing algorithm. Non-mapping and duplicate sequences were discarded. Candidate translocations suggested by >4 unique mate pairs were examined by BLAT to exclude false positives explained by mapping to homologous regions. PCR and Sanger sequencing of tumor DNA was used for validation. Fluorescence in situ hybridization (FISH) was performed on additional PTCLs. mRNA and miRNA expression was quantified using real-time PCR. Results: Of 28.90×106 paired sequences, 8.61×106 were non-duplicate pairs in which both ends mapped to the human genome (bridged coverage, ~7x). Only one candidate translocation involved 6p25.3, represented by 10 unique mate pairs. PCR and Sanger sequencing confirmed a balanced translocation: one breakpoint disrupted the dual-specificity phosphatase gene, DUSP22, telomeric to IRF4 on 6p25.3; the other was telomeric to the FRAH7 fragile site on 7q32.3. Additional ALK-negative ALCLs with known 6p25.3 rearrangements were tested using a 7q32.3 breakapart FISH probe, which showed a 7q32.3 rearrangement in 13/29 (45%; 7 systemic and 6 cutaneous). Dual fusion FISH confirmed t(6;7)(p25.3;q32.3) in 11/11 cases tested. None of 120 PCTLs (including 39 ALCLs) lacking 6p25.3 rearrangements had a 7q32.3 rearrangement. Levels of 5′ and 3′ DUSP22 mRNA in cases with 6p25.3 rearrangements were decreased relative to those in non-rearranged cases (5′: 0.02±0.01 vs 1.00±0.74, p=.002; 3′: 0.09±0.08 vs 1.00±0.55, p=.0002; mean±SD, t test). IRF4 mRNA levels were similar in both groups. Levels of MIR29A and MIR29B1 within FRAH7 in cases with 7q32.3 rearrangements were increased relative to those in non-rearranged cases (MIR29A: 2.44±2.20 vs 1.00±1.34, p=.15; MIR29B: 4.93±4.68 vs 1.00±1.05, p=.007). Conclusions: The t(6;7)(p25.3;q32.3) is the first recurrent translocation characterized in systemic and cutaneous ALK-negative ALCLs. This translocation was entirely specific for ALK-negative ALCLs, and was present in 45% of cases with 6p25.3 rearrangements. The t(6;7)(p25.3;q32.3) was associated with down-regulation of DUSP22 and up-regulation of MIR29, but not dysregulation of IRF4. DUSP22 inhibits mitogen activated protein kinase activity in T cells, and may represent a novel tumor suppressor in ALCLs. Increased MIR29 levels may additionally contribute to tumorigenesis, as recently shown in acute myeloid leukemias. Finally, clinical testing for t(6;7)(p25.3;q32.3) may aid in differentiating ALK-negative ALCL from other PTCLs. Our findings highlight the utility of mate-pair library sequencing to discover novel translocations in lymphoma and other cancers. Disclosures: No relevant conflicts of interest to declare.

Abstract Background: Peripheral T-cell lymphomas (PTCLs) are aggressive non-Hodgkin lymphomas with marked clinical, pathological, and molecular heterogeneity. Outcomes following standard therapy generally are poor; however, few candidate therapeutic targets have been identified for precision medicine approaches. Retinoic acid receptor alpha (RARA) is a transcription factor that modulates cell growth and differentiation in response to natural or synthetic retinoids. Retinoids have been used successfully to treat acute promyelocytic leukemia and some cutaneous T-cell lymphomas (CTCLs). However, the function of RARA and the action of retinoids in PTCL have not been defined. Methods:Based on identification of a PTCL patient with a non-synonymous point mutation, RARA R394Q, identified in the Mayo Clinic Center for Individualized Medicine, we sought to characterize the role of RARA in PTCL cells. To investigate the role of wild-type and mutant RARA, we constructed expression vectors containing either wild-type RARA or RARA R394Q coding sequences, and also used siRNAs targeting RARA to study the role of native RARA expression. Cell lines derived from post-thymic T-cell malignancies were used for in vitro studies, including HuT78 and Mac-1 (both derived from circulating tumor cells from CTCL patients) and Karpas 299 (from an ALK-positive anaplastic large cell lymphoma). Following RARA overexpression or knockdown, we measured cell growth, cell cycle regulation, and sensitivity to synthetic retinoids. In addition, RNA sequencing and pathway analysis were performed to profile the transcriptomic response to retinoids in malignant T cells. Results:In two RARAlow cell lines, Karpas 299 and HuT78, overexpression of wild-type RARA or RARA R394Q significantly increased cell growth (p<0.001), with a greater increase observed from mutant versus wild-type RARA in Karpas 299 (136% of control versus 122%; p=0.04). Accordingly, knockdown of wild-type RARA in the RARAhigh cell line, Mac-1, resulted in a 22% inhibition of cell growth (p=0.0002). This inhibition specifically was associated with G1 cell cycle arrest (120% of control; p=0.004) and decreased protein expression of the G1-S-associated cyclin-dependent kinases, CDK2, CDK4, and CDK6. These kinases were up-regulated by overexpression of RARA in RARAlow HuT78 cells. The relatively RARA-specific retinoid, AM80 (tamibarotene), and the less specific retinoid, all-trans retinoic acid (ATRA), resulted in RARA protein degradation, cell growth inhibition that was both dose-dependent and proportional to baseline RARA expression, G1 arrest, and CDK protein up-regulation. Gene-set enrichment analysis (GSEA) of transcriptome data confirmed that genes down-regulated by AM80 were highly enriched for regulators of cell cycle and particularly G1-S transition. Finally, overexpressing RARA in RARAlow Karpas 299 and HuT78 cell lines significantly increased the ability of AM80 to inhibit CDK2/4/6 expression and cell growth (16% to 23% greater growth inhibition than control; p<0.05). Conclusions:RARA drives cyclin-dependent kinase expression and G1-S transition in malignant T cells, and promotes cell growth. These functions may be enhanced by specific RARA gene mutations. Synthetic retinoids inhibit these functions in a dose-dependent fashion, and are most effective in cells with high RARA expression. These data suggest RARA as a candidate therapeutic target in some PTCL patients. Disclosures Nowakowski: Celgene: Research Funding; Morphosys: Research Funding; Bayer: Consultancy, Research Funding.

Abstract Background Next Generation sequencing (NGS) is a powerful tool to identify somatic mutations associated with tumor onset and drug response. While it is well suited for high quality fresh/frozen samples, NGS is not proven for FFPE tissue which is the most common type of clinical specimen. Since the nucleic acids can be readily extracted from FFPE samples for a variety of genomic analyses, a comparative mutational analysis of paired frozen and FFPE tissues is urgently needed. Our long term goal is to establish a lab protocol to detect mutations in FFPE tumors using a targeted capture and sequencing approach for genes of interest. This pilot study focuses on the comparison of FFPE and frozen samples to test the validity of using FFPE tissues in such application. Methods Gene Selection: 128 genes associated with known pathogenic mutations in lymphoma Sample Selection: 9 diffuse large B-cell lymphoma (DLBCL) cases with FFPE, frozen and germline samples, as well as 10 frozen normal lymphatic tissues as references for CNV detections Capture Probe Design: We targeted coding exons and UTR, as well as the evolutionarily conserved intronic regions. The capture probes were designed using the Agilent eArray tool. The titling density of the probes was set to 3 probes overlapping with every base in the target region to improve the capture efficiency in FFPE samples. The least stringent masking of the repeat regions was allowed to include regions with small repeats that are shorter than the length of the sequencing reads (100-bp). In addition, boosting parameters were picked to set various levels of probe replication in different regions in order to minimize the local coverage differences (e.g. between regions of different GC contents) Sequencing and Bioinformatics: The target capture and sequencing were performed by the Mayo Clinic Medical Genome Facility. The reads were mapped to Human Reference Genome Build 37 using Novalign, and SNVs were called using GATK. The CNVs were identified using an in-house developed algorithm, patternCNV. Results The designed probes covered 99.65937% of the target regions. We generated 2.2-6.7 Gbp of reads per sample, 57.4-71.5% of which were on target. This equalled an average coverage of 2100-6700 folds which is 10-30 times higher than the minimal coverage recommended by Agilent. Due to this high coverage, we observed duplicate reads that accounted for 7.7-73.5% of the total reads. When we analysed the data with and without the duplicated reads, the concordance of the called SNVs was between 84-93% out of 207-249 mutated positions per trio-sample. There were 7.8-8.9% and 1.1-2.2% unique SNVs per sample by excluding or including duplicate reads, respectively. The dis-concordances were mostly missed calls, where a SNV was observed in only 1 or 2 of the trio samples. The missed calls from frozen samples ranged from 0-10.4% compared to 1.4-10.4% from the FFPE tissues, with 0.88-2.4% more SNVs missed in FFPE. Further analyses showed that all of the missing calls came from the lack of or low coverage of the corresponding positions. There were also differences of the called SNVs between the trio samples. However, this was extremely rare. Only 2 out of the 9 trio samples at a total of 3 positions had disagreements in called SNVs between FFPE and frozen tissues, all due to the allelic imbalance where the percentage of reads supporting the alternative alleles were below 20%. Therefore, this dis-concordance can be removed by back-filling of the read-level information for each position. Unfortunately only 11.9-47.4% of the CNVs called in frozen tissues were identified in FFPE samples, due to the widely various coverage in FFPE samples. The consequent large noises of the log ratio values between the FFPEs and normal references significantly reduced the sensitivity for CNV calling. Conclusions This pilot study compared the performance of SNV and CNV detection in FFPE and paired frozen tissues using a target capture and sequencing approach. With a capture probe design strategized to benefit FFPE samples, we observed SNV detection rates in FFPE that were only slightly lower (0.88-2.4%) than those of frozen tissues due to poor coverage of some positions in FFPE samples. With a proper back-filling step, there was no dis-concordance of the called SNVs between FFPE and frozen samples. However, CNV detections in FFPE were more problematic due to the un-predictable regional coverage in FFPE samples. Disclosures: No relevant conflicts of interest to declare.