Optical Genome Mapping in Routine Human Genetic Diagnostics—Its Advantages and Limitations

In recent years, optical genome mapping (OGM) has developed into a highly promising method of detecting large-scale structural variants in human genomes. It is capable of detecting structural variants considered difficult to detect by other current methods. Hence, it promises to be feasible as a first-line diagnostic tool, permitting insight into a new realm of previously unknown variants. However, due to its novelty, little experience with OGM is available to infer best practices for its application or to clarify which features cannot be detected. In this study, we used the Saphyr system (Bionano Genomics, San Diego, CA, USA), to explore its capabilities in human genetic diagnostics. To this end, we tested 14 DNA samples to confirm a total of 14 different structural or numerical chromosomal variants originally detected by other means, namely, deletions, duplications, inversions, trisomies, and a translocation. Overall, 12 variants could be confirmed; one deletion and one inversion could not. The prerequisites for detection of similar variants were explored by reviewing the OGM data of 54 samples analyzed in our laboratory. Limitations, some owing to the novelty of the method and some inherent to it, were described. Finally, we tested the successful application of OGM in routine diagnostics and described some of the challenges that merit consideration when utilizing OGM as a diagnostic tool.

Automated Peripheral Blood Cell Differentiation Using Artificial Intelligence - a Study with More Than 10,000 Routine Samples in a Specialized Leukemia Laboratory

Background: Cytomorphology is the gold standard for quick assessment of peripheral blood (PB) and bone marrow samples in hematological neoplasms and is used to orchestrate specific diagnostics. Artificial Intelligence (AI) promises to provide an unbiased way of interrogating blood smear data as reproducibility varies across labs. This is a prospective clinical study (ClinicalTrials.gov Identifier: NCT04466059) conducted on our approach outlined at ASH 2020. Aim: Use an AI model to classify cell images to produce differential counts of PB smears side-by-side to routine diagnostics. Methods: We enrolled 10,082 patient samples which were sent to our lab between 01/2021 and 07/2021 for cytomorphology with a suspected hematologic neoplasm. Blood smears were differentiated by highly skilled technicians (median 5y in lab) and all were reviewed by hematologists. In parallel, all samples were scanned on a MetaSystems (Altlussheim, Germany) Metafer Scanning System (Zeiss (Oberkochen, Germany) Axio Imager.Z2 microscope, automatic slide feeder). Areas of interest were defined and leukocyte positions were flagged by pre-scan in 10x magnification followed by high resolution scan in 40x to generate cell images for analysis. We set up a supervised Machine Learning model based on ImageNet-pretrained Xception using Amazon Sagemaker (AS) and trained it on 8,425 carefully annotated color images to identify 21 predefined classes (including 1 garbage class). Overall accuracy of this model against hold-out-set (10%) was 96%. The algorithm consumes 144x144pixel cell images and produces probability scores (PS) for each class in every image. Results: For routine diagnostics in median 100 cells/sample (range 82 - 103) were differentiated manually, overall 988,130. The automated process gathered 500 cell images/sample (range 101 - 500), overall 4,937,389. Average capture times for 500 cells: 4:37 min. Cropped images were uploaded to a cloud storage and exposed to an AS endpoint to initiate classification and the computation of a PS for each of the predefined 21 classes in the model. For the study we only considered images with a probability of at least 90% (n=3,781,670/4,937,389) and excluded normoblasts, smudge cells and images identified as garbage (together n=2,120,258). Final diagnosis included: no lymphoma detectable (2,186), MDS (1,152), AML (369), in these 11 APL, MPN (658), CLL (558), other mature B-cell neoplasms (377), CML (326), multiple myeloma (155), but also rare entities such as hairy cell leukemia variant (2) or PPBL (3). Comparing the benign normal cells in peripheral blood we identified (all values normalized) segmented neutrophils (manual (M): 516,648=52% vs AI: 882,538=53%), eosinophils (M: 24,860=2.52% vs. AI: 55,699=3.36%), basophils (M: 7159=0,72% vs. AI: 11,957=0,72%), monocytes (M: 74,113=7.5% vs. AI: 110,126=6.64%), lymphocytes (M: 313,518=31.7% vs. AI: 399,249=24%). Pathogenic blasts were detected in 16,048 (0.97%) images by AI (M: 16,290=1.65%). In routine diagnostics 536 cases with blast cells, including "questionable blasts" were identified. The AI identified 493 (91%) of these cases. At least one atypical/malignant lymphocyte was found in 2,323 samples manually, out of which the AI identified 2,279 (98%). In few cases manual differentiation relies on the number of pathogenic cells from an immunophenotyping analysis, which the AI does not had. During the course of the study by chance we identified at least 3 instances, were the AI detected pathogenic cells (blasts, atypical promyelocytes (APL) or bilobulated promyelocytes (APL-v)) which were initially missed manually (in some case WBC below .5 G/l) or flagged during subsequent immunophenotyping/molecular genetic analysis. Upon manually revisiting the smear, we could verify the presence of the AI-anticipated cells, revealing the higher sensitivity of the 5 time increase in cells/sample investigated by AI and power of algorithms. Conclusion: We present data of a prospective, blinded clinical study comparing blood smear analysis between humans and AI head-to-head. The concordance is extremely high with 95% for pathogenic cases. Misclassified cells are used for retraining to continuously improve the model and benefit from large datasets even for rare cell types. The model's cloud based implementation makes it easy to connect scanning devices for automated, unbiased classification. Disclosures Haferlach: MLL Munich Leukemia Laboratory: Other: Part ownership. Kern: MLL Munich Leukemia Laboratory: Other: Part ownership. Haferlach: MLL Munich Leukemia Laboratory: Other: Part ownership.

RNA-Sequencing Improves Diagnostics and Treatment of Pediatric Hematological Malignancies

Background Diagnosis and treatment of hematological malignancies relies increasingly on the detection of underlying genetic abnormalities. Various laboratory techniques, including karyotyping, SNP-array, FISH, MLPA and RT-PCR are typically required to detect the full spectrum of clinically relevant genetic aberrations. These techniques are also hampered in their sensitivity by their targeted approach or lack of resolution. Ideally, an unbiased genome wide approach like RNA sequencing (RNA-seq) as a one-test-fits-all, could save costs and efforts and streamline diagnostic procedures. In the Netherlands, the care for all children with oncological disorders has been concentrated in a single, national center. Within the Laboratory of Childhood Cancer Pathology, we aim for a comprehensive diagnostic pipeline by implementing RNA-seq to aid diagnosis, prognosis and treatment of all children with cancer in the Netherlands. Methods We have established an RNA-seq based diagnostic pipeline, primarily aimed at detecting gene fusion events. Library prep is performed on 50-300 ng total RNA isolated from fresh (frozen) samples, followed by ribo-depletion and subsequent paired-end sequencing (2x150 nt) using the Illumina NovaSeq platform. Data is analyzed using the StarFusion algorithm for gene-fusion detection. We are prospectively comparing the results with routine diagnostic procedures. In addition, we are validating the detection of single nucleotide variants (SNVs) from RNA-seq data and developing a diagnostic classifier, using a nearest neighbor network approach. Results Based on RNA-seq profiling in diagnostics for all patients entering the Princess Maxima Center, there are several use-cases that highlight the value of RNA-seq. 1) In a prospective cohort of 244 patients (pan-cancer, including 97 hematological malignancies) we have shown that the diagnostic yield for detecting gene fusion events increased by approximately 40% compared to classical methods. An example is the TNIP1--PDGFRB gene fusion in a patient with pre B-ALL, making this patient eligible for imatinib treatment, which was not detected by other methods. 2) Variant calling on RNA-seq shows that activating mutations in e.g. KRAS are detected with high sensitivity, stratifying patients for therapeutic MEK intervention. 3) By expression outlier analysis, we were able to detect various promotor exchanges, e.g. IGH-MYC or IGH--DUX4, which are typically hard to detect by molecular techniques since the genomic breakpoint is highly variable and no chimeric transcript is formed. 4) Preliminary results from our diagnostic classifier show its potential to predict subclasses of hematological malignancies, e.g. high-hyperdiploid or bi-phenotypic ALL patients. 5) Fusion gene breakpoints detected by RNA-seq serve as a target for MRD analysis, allowing us to monitor disease progression and therapy response in individual patients. Currently, RNA-seq data is available for more than 1500 pediatric tumor samples. At the upcoming conference we will present an update of our results and some typical cases highlighting the added value of RNA-seq in routine diagnostics. Conclusion We show that RNA-seq on pediatric cancer samples is feasible and of great value for routine diagnostics. It has a higher sensitivity to detect gene fusion events compared to targeted assays. RNA-seq based gene fusion detection, in combination with mutation and expression analysis, is also promising to improve classification of malignancies, prognosis and stratification of patients for targeted therapies. Disclosures No relevant conflicts of interest to declare.

Whole Genome Sequencing in the Management of Non-Tuberculous Mycobacterial Infections

Infections caused by non-tuberculous mycobacteria (NTM) have been a public health problem in recent decades and contribute significantly to the clinical and economic burden globally. The diagnosis of infections is difficult and time-consuming and, in addition, the conventional diagnostics tests do not have sufficient discrimination power in species identification due to cross-reactions and not fully specific probes. However, technological advances have been made and the whole genome sequencing (WGS) method has been shown to be an essential part of routine diagnostics in clinical mycobacteriology laboratories. The use of this technology has contributed to the characterization of new species of mycobacteria, as well as the identification of gene mutations encoding resistance and virulence factors. Sequencing data also allowed to track global outbreaks of nosocomial NTM infections caused by M. abscessus complex and M. chimaera. To highlight the utility of WGS, we summarize recent scientific studies on WGS as a tool suitable for the management of NTM-induced infections in clinical practice.

Multicenter evaluation of the FilmArray Meningitis/Encephalitis assay in a routine setting

Introduction. The FilmArray Meningitis/Encephalitis (FA-ME) Panel (Biofire, Salt Lake City, Utah, US) enables fast and automated detection of 14 pathogens in cerebrospinal fluid (CSF). Gap statement. The performance of the FA-ME panel in a real routine setting has not yet been described and could lead to better patient management in cases of good performance. Aim. This multicenter study verified the FA-ME panel analytical performance in a routine hospital setting. Methodology. Between April 2016 and April 2018, 454 CSF samples were analysed with the FA-ME panel and compared with routine diagnostics. In cases of discrepancy or lack of a comparator result, a profound analysis based on patient records and other laboratory results was performed. Results. A first analysis of 65 frozen samples, suspicious for meningitis had a 89 % concordance with routine diagnostics. The limit of detection (LOD) was confirmed for all pathogens except for Streptococcus agalactiae and a strain of Haemophilus influenzae (Escherichia coli K1 and Cryptococcus gattii LOD experiments were not performed). The routine evaluation showed a positive result in 114 (25 %) clinical samples for at least one target. In three samples co-infections were found. After discrepancy analysis, overall sensitivity was 98 % (false negative FA-ME results for one HSV2, two HSV1 and two parechovirus). Four FA-ME results were considered false positive (two HHV6, one VZV and one E. coli K1), resulting in an overall specificity of >99 %. A clinical added value of the assay was seen in the diagnosis of eight cases of bacterial meningitis. Conclusion. Because of its rapidity and ease of use, the FA-ME panel has great potential in the diagnosis of central nervous infections. Implementation can improve clinical management, but costs and analytical limitations need to be addressed to convince clinicians and laboratories of its value.

Peptides in COVID-19 Clinical Trials—A Snapshot

Since the beginning of the COVID-19 pandemic, there has been a strong drive and desire to find effective treatments for and protection against the disease. On the webpage ClinicalTrials.gov, a total of 6505 clinical trials currently (September 2021) investigating various aspects of COVID-19 are registered. Of these, 124 studies involving peptides were identified. These 124 were further evaluated, and 88 trials that used peptides only for routine diagnostics were excluded. The remaining 36 trials were classified into 5 different classes according to their function: immunomodulatory (5 trials), regain homeostasis (10 trials), diagnostics/biomarkers (8 trials), vaccination (9 trials), and antiviral activity (4 trials, all overlap with immunomodulatory activities). In the current review, these 36 trials are briefly described and tabularly summarised. According to the estimated finish date, 14 trials have not yet finished. All of the finished trials are yet to report their results. Seven trials were based in the USA, and Egypt, France, the UK, Turkey, and the Russian Federation conducted three trials each. This review aims to present a snapshot of the current situation of peptides in COVID-19 clinical trials and provides a template to follow up on trials of interest; it does not claim to be a complete overview.

Machine learning-based tissue of origin classification for cancer of unknown primary diagnostics using genome-wide mutation features

Tumor tissue of origin (TOO) is an important factor for guiding treatment decisions. However, TOO cannot be determined for ~3% of metastatic cancer patients and are categorized as cancers of unknown primary (CUP). As whole genome sequencing (WGS) of tumors is now transitioning from the research domain to diagnostic practice in order to address the increasing demand for biomarker detection, its use for detection of TOO in routine diagnostics also starts becoming within reach. While proof of concept for the use of genome-wide features has been demonstrated before, more complex WGS mutation features, including structural variant (SV) driver and passenger events, have never been integrated into TOO-classifiers even though they bear highly characteristic links with tumor TOO. Using a uniformly processed dataset containing 6820 whole-genome sequenced primary and metastatic tumors, we have developed Cancer of Unknown Primary Location Resolver (CUPLR), a random forest based TOO classifier that employs 502 features based on simple and complex somatic driver and passenger mutations. Our model is able to distinguish 33 cancer (sub)types with an overall accuracy of 91% and 89% based on cross-validation (n=6139) and hold out set (n=681) predictions respectively. We found that SV derived features increase the accuracy and utility of TOO classification for specific cancer types. To ensure that predictions are human-interpretable and suited for use in routine diagnostics, CUPLR reports the top contributing features and their values compared to cohort averages. The comprehensive output of CUPLR is complementary to existing histopathological procedures and may thus improve diagnostics for patients with CUP.

Molecular epidemiology of hereditary ataxia in Finland

Background The genetics of cerebellar ataxia is complex. Hundreds of causative genes have been identified, but only a few cause more than single cases. The spectrum of ataxia-causing genes differs considerably between populations. The aim of the study was to investigate the molecular epidemiology of ataxia in the Finnish population. Patients and methods All patients in hospital database were reviewed for the diagnosis of unspecified ataxia. Acquired ataxias and nongenetic ataxias such as those related to infection, trauma or stroke were excluded. Sixty patients with sporadic ataxia with unknown etiology and 36 patients with familial ataxia of unknown etiology were recruited in the study. Repeat expansions in the SCA genes (ATXN1, 2, 3, 7, 8/OS, CACNA1A, TBP), FXN, and RFC1 were determined. Point mutations in POLG, SPG7 and in mitochondrial DNA (mtDNA) were investigated. In addition, DNA from 8 patients was exome sequenced. Results A genetic cause of ataxia was found in 33 patients (34.4%). Seven patients had a dominantly inherited repeat expansion in ATXN8/OS. Ten patients had mitochondrial ataxia resulting from mutations in nuclear mitochondrial genes POLG or RARS2, or from a point mutation m.8561C > G or a single deletion in mtDNA. Interestingly, five patients were biallelic for the recently identified pathogenic repeat expansion in RFC1. All the five patients presented with the phenotype of cerebellar ataxia, neuropathy, and vestibular areflexia (CANVAS). Moreover, screening of 54 patients with Charcot-Marie-Tooth neuropathy revealed four additional patients with biallelic repeat expansion in RFC1, but none of them had cerebellar symptoms. Conclusions Expansion in ATXN8/OS results in the majority of dominant ataxias in Finland, while mutations in RFC1 and POLG are the most common cause of recessive ataxias. Our results suggest that analysis of RFC1 should be included in the routine diagnostics of idiopathic ataxia and Charcot-Marie-Tooth polyneuropathy.

Phenotypic Variants of Bacterial Colonies in Microbiological Diagnostics: How Often Are They Indicative of Differing Antimicrobial Susceptibility Patterns?

Phenotypic variants of bacteria are frequent in routine diagnostics and can display differing antimicrobial susceptibility patterns. We found that the likelihood of different antimicrobial susceptibility is low among PV. To save laboratory resources, only one isolate per PV could be tested to guide the antimicrobial treatment of patients.

CombiFlow: Combinatorial AML-specific plasma membrane expression profiles allow longitudinal tracking of clones

Acute myeloid leukemia (AML) often presents as an oligoclonal disease whereby multiple genetically distinct subclones can co-exist within patients. Differences in signaling and drug sensitivity of such subclones complicates treatment and warrants tools to identify them and track disease progression. We previously identified over 50 AML-specific plasma membrane (PM) proteins and seven of these (CD82, CD97, FLT3, IL1RAP, TIM3, CD25 and CD123) were implemented in routine diagnostics in patients with AML (n=256) and MDS (n=33). We developed a pipeline termed CombiFlow in which expression data of multiple PM markers is merged, allowing a Principle Component-based analyses to identify distinctive marker expression profiles and to generate single cell tSNE landscapes to longitudinally track clonal evolution. Positivity for one or more of the markers after 2 courses of intensive chemotherapy predicted a shorter relapse-free survival supporting a role of these markers in measurable residual disease (MRD) detection. CombiFlow also allowed the tracking of clonal evolution in paired diagnosis and relapse samples (n=12). Extending the panel to 36 AML-specific markers further refined the CombiFlow pipeline. In conclusion, CombiFlow provides a valuable tool in the diagnosis, MRD detection, clonal tracking, and the understanding of clonal heterogeneity in AML.