Sharma, Neeraj

Sudden infant death syndrome (SIDS) is the unexpected death of an infant under one year of age that remains unexplained after a thorough investigation. Despite SIDS remaining a diagnosis of exclusion with an unexplained etiology, it is widely accepted that SIDS can be caused by environmental and/or biological factors, with multiple underlying candidate genes. However, the lack of biomarkers raises questions as to why genetic studies on SIDS to date are unable to provide a clearer understanding of the disease etiology. We sought to improve the identification of SIDS-associated genes by reviewing the SIDS genetic literature and objectively categorizing and scoring the reported genes based on the strength of evidence (from C1 (high) to C5 (low)). This was followed by analyses of function, associations between genes, the enrichment of gene ontology (GO) terms, and pathways and gender difference in tissue gene expression. We constructed a curated database for SIDS gene candidates consisting of 109 genes, 14 of which received a category 4 (C4) and 95 genes received the lowest category of C5. That none of the genes was classified into the higher categories indicates the low level of supporting evidence. We found that genes of both scoring categories show distinct networks and are highly diverse in function and involved in many GO terms and pathways, in agreement with the perception of SIDS as a heterogeneous syndrome. Genes of both scoring categories are part of the cardiac system, muscle, and ion channels, whereas immune-related functions showed enrichment for C4 genes. A limited association was found with neural development. Overall, inconsistent reports and missing metadata contribute to the ambiguity of genetic studies. Considering those parameters could help improve the identification of at-risk SIDS genes. However, the field is still far from offering a full-pledged genetic test to identify at-risk infants and is still hampered with methodological challenges and misunderstandings of the vulnerabilities of vital biological mechanisms.

Purpose: Acute myeloid leukemia (AML) is the most common acute leukemia in adults, affecting approximately 20,000 patients annually in the United States. AML genetic subtypes, as defined by the World Health Organization (WHO), are identified through fluorescence in situ hybridization (FISH), conventional chromosome analysis, and sequencing techniques. Mate pair sequencing (MPseq) is a next generation sequencing (NGS) technology optimized to detect genome wide structural variants and copy number alterations at high resolution. Our study goal was to investigate the prognostic value of MPseq in comparison to FISH, chromosome, and sequencing studies in the evaluation of AML patients. Methods: We performed a prospective study using blood and bone marrow samples from 105 patients with a diagnosis of AML, using MPseq, along with chromosome, FISH, and NGS or PCR studies to detect small mutations. Cytogenetic and molecular genetic results were correlated with MPseq findings. We also analyzed the MPseq data for chromoplexy, chromothripsis, and progressive complexity. Junction and copy number burden, the incidence of structural variation in the genome and the percent of the genome with aberrant copy number, were evaluated. Overall survival statistics were stratified by AML subtypes and observed anomalies. Results: Although structural variants in AML were characterized at a high resolution using MPseq when compared to conventional cytogenetic methods, risk stratification using current European Leukemia Net (ELN) guidelines was not improved by MPseq. The cohorts involving 5q and/or 7q deletions exhibited high levels of genomic complexity when compared to normal karyotype AML (NK-AML). The incidence of copy number gains, losses and junctions was greatest in 5q and 7q co-deletions (5q/7q) (16.5, 25.0, 69.3) and 5q deletions (5q) (9.8, 16.7, 31.6) subtypes compared to 7q deletions (3.4, 7.0, 6.7) and NK-AML (2.6, 4.3, 3.8) (p<0.001) subtypes. Chromoplexy, chromothripsis, and progressive structural complexity were detected in most samples with 5q deletions and 5q/7q co-deletions, but absent in samples with 7q or NK-AML. Biallelic inactivation of TP53 by sequencing mutation and/or deletion was common in the 5q/7q co-deletion (14/18 cases) and 5q deletion cohorts (7/10 cases), rare in the 7q deletion cohort (1/11 cases), and absent in the NK-AML (n=44) cohort. The median OS was significantly worse for patients with 5q/7q deletions (122.5 days) and 5q deletions (248 days) compared to NK-AML (413.5 days; p<0.001 and p=0.017, respectively) and between 5q/7q deletions and 7q deletions (370.5 days; p<0.001). No significant difference was observed between 5q/7q and 5q deletion subtypes, between NK and 7q deletion subtypes and between 5q and 7q deletion subtypes. The median OS was also significantly shorter for patients with TP53 alterations compared to patients with normal TP53 status. Patients with chromoplexy, chromothripsis and/or progressive structural complexity identified by MPseq had a significantly shorter median OS compared to patients without these features. (p<0.0001) Discussion: Risk stratifications based on current guidelines using cytogenetic and sequencing results were not adjusted due to MPseq results, which is not surprising when primary abnormalities are observed by conventional cytogenetic methods. NK-AML cases did not appear to benefit from a high resolution genomic evaluation. However, MPseq added value when structural variation required additional characterization, detecting novel rearrangements, such as a KAT6A/SORBS3 fusion. Lastly, we recognized common mischaracterizations made by conventional chromosome studies - including missed TP53 deletions in 7 cases, 5q/7q deletions misinterpreted as monosomies, cryptic NUP98 rearrangements, and unappreciated genomic complexity correlating with poor OS. These mischaracterizations challenge the use of conventional chromosome studies as a gold standard without accompanying FISH or MPseq studies. MPseq, similar to other structural methodologies such as optical mapping and long read sequencing, should be considered important complements to standard cytologic techniques given the important additional genomic information obtained. The additional structural variant characterization will be critical in paving the way for genomic discovery with the overall goal of improving prognostication for patient care. Figure Disclosures Vasmatzis: WholeGenome LLC.: Other: Owner; Mayo Clinic: Membership on an entity's Board of Directors or advisory committees, Research Funding.

Background: Multiple myeloma (MM) remains an incurable plasma cell (PC) malignancy. Despite improvements in clinical outcomes, variability in treatment response exists. Some patients experience innate resistance, while others develop resistance over the course of treatment. Identifying signatures of treatment response and resistance is essential to predicting effective treatment strategies. Given the phenotypic clonal heterogeneity observed in MM, we hypothesize that resistant and sensitive subpopulations have unique phenotypic signatures. We used mass cytometry (CyTOF) to characterize primary MM samples and correlated their phenotypic signatures with treatment response. Methods: We characterized 102 primary MM samples by CyTOF using 37 antibodies (see figure). Frozen whole BM specimens were thawed, stained with antibodies, barcoded, and analyzed on a Helios mass cytometer. Data was processed using Cytobank or Pathsetter and analyzed using t-distributed stochastic neighbor embedding (t-SNE) algorithm. T-SNE plots for CD45+ cells were created using lineage defining markers. T-SNE plots of neoplastic PCs (CD138+,CD3-,CD16-,CD19-,cytoplasmic kappa or lambda+) were created using all remaining markers. A differential analysis of each sample's clusters was performed against all other events. Z-score values, representing marker changes from the differential analysis, for each cluster were used to group similar clusters across all samples, allowing us to compare sub populations of PCs between patients (see figure). From these groups of clusters with similar changes in markers, we calculated the proportion of cells with increases or decreases in markers for each sample. Results were compared to IMWG 90-day response criteria using a random forest model. Results: We performed mass cytometry on 102 samples from 68 unique MM patients. Median age was 66 years (range 34-95) with a 1.5:1 male:female ratio. 37 (36%) had NDMM, 55 (54%) had RRMM, 2 (2%) had SMM and 8 cases (8%) had unknown status. FISH studies revealed 47 (46%) had hyperdiploidy without a recurrent primary IgH rearrangement, 31 (30%) had t(11;14), 13 (13%) had t(4;14), 5 (5%) had 14;16 or 14;20, 3 (3%) had hyperhaploidy and 3 (3%) had an undefined primary. 54 (53%) had 1q gain/amp, 22 (22%) had TP53 deletion, 22 (22%) had a MYC rearrangement and 71% had high mSMART risk. Nearly all patients received an immunomodulatory compound, a proteasome inhibitor and dexamethasone. 2 (2%) had complete response (CR), 10 (9.8%) very good partial response (VGPR), 36 (35%) had partial response (PR), 14 (14%) had minimal response (MR), 13 (13%) had stable disease (SD) and 17 (17%) had progressive disease (PD). Characterization of CD45+ cells identified a relatively conserved and reproducible map from each patient. Notable differences included an ~2-fold reduction in the percentage of total B cells (CD45+,CD3-,CD19+)/total CD45+ cells in cases with high risk mSMART, RRMM and TP53 deletion compared to cases with standard risk, NDMM or normal TP53 status. Similar B-cell reductions were found in RRMM vs. NDMM samples from matched-patients. In contrast to the conserved t-SNE structure of the CD45+ population, the neoplastic PC population displayed dramatically reduced t-SNE structural consistency. Heterogeneity between patient samples and between the same patient samples over time was observed. We focused our analysis on 22 patients with serial samples. The proportion of each t-SNE group present in each patient sample was calculated and then correlated to the 90-day response metric. We identified a phenotypic group whose proportion increased in relation to disease progression (R2=0.85). This group had elevated BCL-2, pSTAT3, CD56 levels and reduced CD45 and CD81 levels. Evaluating individual markers, we found an increase in the proportion of cells with elevated BCL-2, pS6, Ki67, and IKZF1 levels in relation to disease progression. We also observed an increase in the proportion of cells with reduced pCREB levels and a reduction in the proportion of cells with reduced CD147 levels in relation to disease progression. Conclusions: We have utilized CyTOF to characterize phenotypic and functional markers in MM cells on a single cell level and can identify unique subgroups that change in relation to disease progression. Future analysis will correlate phenotypic signatures with ISS stage, mSMART risk, cytogenetic abnormality and type of treatment. Figure 1 Disclosures Kumar: Kite Pharma: Consultancy, Research Funding; Oncopeptides: Consultancy, Other: Independent Review Committee; IRC member; Celgene/BMS: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Takeda: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Sanofi: Research Funding; MedImmune: Research Funding; Karyopharm: Consultancy; Merck: Consultancy, Research Funding; Carsgen: Other, Research Funding; Cellectar: Other; Novartis: Research Funding; Adaptive Biotechnologies: Consultancy; AbbVie: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Genecentrix: Consultancy; Genentech/Roche: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Janssen Oncology: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Amgen: Consultancy, Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments, Research Funding; Tenebio: Other, Research Funding; BMS: Consultancy, Research Funding; Dr. Reddy's Laboratories: Honoraria.

Background: Multiple myeloma (MM) is an incurable plasma cell malignancy and genetic abnormalities contribute to disease heterogeneity and outcome. Primary abnormalities, namely recurrent immunoglobulin (Ig) heavy chain translocations and hyperdiploidy, occur early in disease course. Secondary events, such as MYC abnormalities occur upon progression. Earlier studies showed MYC abnormalities detected by FISH or by capture sequencing were independently associated with poor outcome (Walker, et al., BCJ, 2014), while recent studies using WGS did not support this finding (excluding MYC/IGL) (Mikulasova, et al. Haematologica, 2020; Misund, et al. Leukemia, 2020). We hypothesize these discrepancies are due to differences in methods and sensitivities of detection of MYC abnormalities by FISH vs. WGS. Given that MYC abnormalities often display remarkable genomic heterogeneity with numerous gene partners, reduced detection of MYC abnormalities by FISH is not unexpected. This hypothesis is supported by lower frequencies of MYC abnormalities found by FISH (15%) vs. NGS (30-35%) consistent with ~50% false-negative rate of the MYC FISH probe (Smadbeck, et al. BCJ, 2019). To better understand the role of MYC in myeloma disease outcome, we compared the MYC abnormality subtype identified by FISH or NGS vs. MYC gene expression levels and overall survival. Methods: We performed a retrospective study of newly diagnosed MM patients seen at Mayo Clinic or enrolled in the MMRF CoMMpass trial. For Mayo cases, MYC FISH results (breakapart probe, Abbott) were obtained from the Mayo Clinic Genomics database (N=1342) and mate pair sequencing (MPseq) was performed on 140 cases. For CoMMpass cases, we obtained tumor long-insert whole genome sequencing (WGS), RNA sequencing (RNAseq) for gene expression and clinical outcome data. Overall survival (OS) was defined as time from diagnosis to death from any cause or to last follow up. Survival curves were estimated using Kaplan Meier and compared using the Log-Rank test. Statistical analyses performed using SPSS and JMP with significance determined when P <0.05. Results: We first evaluated the impact of MYC abnormalities on OS when detected by FISH or NGS. In Mayo cases, OS was significantly shorter in patients with MYC abnormalities compared to patients without MYC abnormalities using FISH (5.3 vs. 8.0 years, P<0.001, N=1342). In contrast, there was no significant difference in OS between patients with or without MYC or abnormalities using MPseq or WGS in both the Mayo and CoMMpass cohorts (Mayo: 6.4 vs. and 6.9 years P=0.78, N=140; CoMMpass: 4.9 vs. and 5.1 years P=0.74, N=546). Since FISH-detected MYC abnormalities were associated with poor outcome, we evaluated differences in the types of MYC abnormalities identified FISH and genome sequencing; 270 of 658 CoMMpass cases had a MYC abnormality and 12 abnormality subgroups were identified. In the Mayo cases, FISH preferentially detected translocations and complex abnormalities and missed insertions with flanking duplicating sequences or terminal tandem duplications (TTD) that occur telomeric to MYC. Since the level of MYC expression should be a consequence of the various genomic abnormalities altering the MYC gene region, we compared MYC expression levels in relation to MYC abnormality subgroups. Highest expression was seen with MYC amplification, followed by Ig abnormalities, non-Ig abnormalities, complex deletion/duplications, proximal deletions, non-Ig insertions, terminal deletions, TTD, trisomy 8, no MYC structural variation, monosomy 8 and cases with MAX mutations had the lowest expression. Abnormalities identified by FISH had higher MYC expression (83.5 TPM) compared to cases predicted to be missed by FISH (63.2 TPM). We tested if high MYC expression, irrespective of MYC structural abnormality, was associated with differences in OS. Boxplot analysis was used to categorize MYC expression in 631 CoMMpass patients as top quartile/high MYC expression (Q4≥ 75 TPM, n=159) and bottom quartile/low MYC expression (Q1≤ 16.5 TPM, n= 158) (see Figure). OS was significantly shorter in patients with high MYC expression compared to patients with low MYC expression (4.6 vs. 5.3 years, P <0.038). Conclusion: We show that FISH detects only a subset of the MYC abnormalities detected by genome sequencing, and that FISH-detected MYC abnormalities are associated with higher MYC gene expression and decreased survival. Figure 1 Disclosures Kumar: Kite Pharma: Consultancy, Research Funding; Janssen Oncology: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; AbbVie: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Oncopeptides: Consultancy, Other: Independent Review Committee; IRC member; Dr. Reddy's Laboratories: Honoraria; Cellectar: Other; Takeda: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Novartis: Research Funding; Tenebio: Other, Research Funding; Carsgen: Other, Research Funding; Amgen: Consultancy, Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments, Research Funding; Merck: Consultancy, Research Funding; Genecentrix: Consultancy; BMS: Consultancy, Research Funding; Karyopharm: Consultancy; Celgene/BMS: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Genentech/Roche: Other: Research funding for clinical trials to the institution, Consulting/Advisory Board participation with no personal payments; Sanofi: Research Funding; MedImmune: Research Funding; Adaptive Biotechnologies: Consultancy.