Applications and challenges in designing VHH-based bispecific antibodies: leveraging machine learning solutions.

The development of bispecific antibodies that bind at least two different targets relies on bringing together multiple binding domains with different binding properties and biophysical characteristics to produce a drug-like therapeutic. These building blocks play an important role in the overall quality of the molecule and can influence many important aspects from potency and specificity to stability and half-life. Single-domain antibodies, particularly camelid-derived variable heavy domain of heavy chain (VHH) antibodies, are becoming an increasingly popular choice for bispecific construction due to their single-domain modularity, favorable biophysical properties, and potential to work in multiple antibody formats. Here, we review the use of VHH domains as building blocks in the construction of multispecific antibodies and the challenges in creating optimized molecules. In addition to exploring traditional approaches to VHH development, we review the integration of machine learning techniques at various stages of the process. Specifically, the utilization of machine learning for structural prediction, lead identification, lead optimization, and humanization of VHH antibodies.

High-affinity, neutralizing antibodies to SARS-CoV-2 can be made without T follicular helper cells.

T follicular helper (TFH) cells are the conventional drivers of protective, germinal center (GC)­based antiviral antibody responses. However, loss of TFH cells and GCs has been observed in patients with severe COVID-19. As T cell­B cell interactions and immunoglobulin class switching still occur in these patients, noncanonical pathways of antibody production may be operative during SARS-CoV-2 infection. We found that both TFH-dependent and -independent antibodies were induced against SARS-CoV-2 infection, SARS-CoV-2 vaccination, and influenza A virus infection. Although TFH-independent antibodies to SARS-CoV-2 had evidence of reduced somatic hypermutation, they were still high affinity, durable, and reactive against diverse spike-derived epitopes and were capable of neutralizing both homologous SARS-CoV-2 and the B.1.351 (beta) variant of concern. We found by epitope mapping and B cell receptor sequencing that TFH cells focused the B cell response, and therefore, in the absence of TFH cells, a more diverse clonal repertoire was maintained. These data support an alternative pathway for the induction of B cell responses during viral infection that enables effective, neutralizing antibody production to complement traditional GC-derived antibodies that might compensate for GCs damaged by viral inflammation.

Longitudinal Immune Profiling of a Severe Acute Respiratory Syndrome Coronavirus 2 Reinfection in a Solid Organ Transplant Recipient.

BACKGROUND: The underlying immunologic deficiencies enabling severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) reinfection are currently unknown. We describe deep longitudinal immune profiling of a transplant recipient hospitalized twice for coronavirus disease 2019 (COVID-19). METHODS: A 66-year-old male renal transplant recipient was hospitalized with COVID-19 March 2020 then readmitted to the hospital with COVID-19 233 days after initial diagnosis. Virologic and immunologic investigations were performed on samples from the primary and secondary infections. RESULTS: Whole viral genome sequencing and phylogenetic analysis revealed that viruses causing both infections were caused by distinct genetic lineages without evidence of immune escape mutations. Longitudinal comparison of cellular and humoral responses during primary SARS-CoV-2 infection revealed that this patient responded to the primary infection with low neutralization titer anti-SARS-CoV-2 antibodies that were likely present at the time of reinfection. CONCLUSIONS: The development of neutralizing antibodies and humoral memory responses in this patient failed to confer protection against reinfection, suggesting that they were below a neutralizing titer threshold or that additional factors may be required for efficient prevention of SARS-CoV-2 reinfection. Development of poorly neutralizing antibodies may have been due to profound and relatively specific reduction in naive CD4 T-cell pools. Seropositivity alone may not be a perfect correlate of protection in immunocompromised patients.

High-resolution epitope mapping and characterization of SARS-CoV-2 antibodies in large cohorts of subjects with COVID-19.

As Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) continues to spread, characterization of its antibody epitopes, emerging strains, related coronaviruses, and even the human proteome in naturally infected patients can guide the development of effective vaccines and therapies. Since traditional epitope identification tools are dependent upon pre-defined peptide sequences, they are not readily adaptable to diverse viral proteomes. The Serum Epitope Repertoire Analysis (SERA) platform leverages a high diversity random bacterial display library to identify proteome-independent epitope binding specificities which are then analyzed in the context of organisms of interest. When evaluating immune response in the context of SARS-CoV-2, we identify dominant epitope regions and motifs which demonstrate potential to classify mild from severe disease and relate to neutralization activity. We highlight SARS-CoV-2 epitopes that are cross-reactive with other coronaviruses and demonstrate decreased epitope signal for mutant SARS-CoV-2 strains. Collectively, the evolution of SARS-CoV-2 mutants towards reduced antibody response highlight the importance of data-driven development of the vaccines and therapies to treat COVID-19.

Immunogenic amino acid motifs and linear epitopes of COVID-19 mRNA vaccines.

Reverse vaccinology is an evolving approach for improving vaccine effectiveness and minimizing adverse responses by limiting immunizations to critical epitopes. Towards this goal, we sought to identify immunogenic amino acid motifs and linear epitopes of the SARS-CoV-2 spike protein that elicit IgG in COVID-19 mRNA vaccine recipients. Paired pre/post vaccination samples from N = 20 healthy adults, and post-vaccine samples from an additional N = 13 individuals were used to immunoprecipitate IgG targets expressed by a bacterial display random peptide library, and preferentially recognized peptides were mapped to the spike primary sequence. The data identify several distinct amino acid motifs recognized by vaccine-induced IgG, a subset of those targeted by IgG from natural infection, which may mimic 3-dimensional conformation (mimotopes). Dominant linear epitopes were identified in the C-terminal domains of the S1 and S2 subunits (aa 558-569, 627-638, and 1148-1159) which have been previously associated with SARS-CoV-2 neutralization in vitro and demonstrate identity to bat coronavirus and SARS-CoV, but limited homology to non-pathogenic human coronavirus. The identified COVID-19 mRNA vaccine epitopes should be considered in the context of variants, immune escape and vaccine and therapy design moving forward.

Phase Ib study of patients with metastatic castrate-resistant prostate cancer treated with different sequencing regimens of atezolizumab and sipuleucel-T.

BACKGROUND: Combining an immune checkpoint inhibitor with a tumor vaccine may modulate the immune system to leverage complementary mechanisms of action that lead to sustained T-cell activation and a potent prolonged immunotherapeutic response in metastatic castration resistant prostate cancer (mCRPC). METHODS: Subjects with asymptomatic or minimally symptomatic mCRPC were randomly assigned in a 1:1 ratio to receive either atezolizumab followed by sipuleucel-T (Arm 1) or sipuleucel-T followed by atezolizumab (Arm 2). The primary endpoint was safety, while secondary endpoints included preliminary clinical activity such as objective tumor response and systemic immune responses that could identify key molecular and immunological changes associated with sequential administration of atezolizumab and sipuleucel-T. RESULTS: A total of 37 subjects were enrolled. The median age was 75.0 years, median prostate specific antigen (PSA) was 21.9 ng/mL, and subjects had a median number of three prior treatments. Most subjects (83.8%) had at least one treatment-related adverse event. There were no grade 4 or 5 toxicities attributed to either study drug. Immune-related adverse events and infusion reactions occurred in 13.5% of subjects, and all of which were grade 1 or 2. Of 23 subjects with Response Evaluation Criteria in Solid Tumors measurable disease, only one subject in Arm 2 had a partial response (PR) and four subjects overall had stable disease (SD) at 6 months reflecting an objective response rate of 4.3% and a disease control rate of 21.7%. T-cell receptor diversity was higher in subjects with a response, including SD. Immune response to three novel putative antigens (SIK3, KDM1A/LSD1, and PIK3R6) appeared to increase with treatment. CONCLUSIONS: Overall, regardless of the order in which they were administered, the combination of atezolizumab with sipuleucel-T appears to be safe and well tolerated with a comparable safety profile to each agent administered as monotherapy. Correlative immune studies may suggest the combination to be beneficial; however, further studies are needed. TRIAL REGISTRATION NUMBER: NCT03024216.

Longitudinal immune profiling of a SARS-CoV-2 reinfection in a solid organ transplant recipient.

The underlying immunologic deficiencies enabling SARS-CoV-2 reinfections are currently unknown. Here we describe a renal-transplant recipient who developed recurrent, symptomatic SARS-CoV-2 infection 7 months after primary infection. To elucidate the immunological mechanisms responsible for reinfection, we performed longitudinal profiling of cellular and humoral responses during both primary and recurrent SARS-CoV-2 infection. We found that the patient responded to the primary infection with transient, poor-quality adaptive immune responses that was further compromised by intervening treatment for acute rejection of the renal allograft prior to reinfection. Importantly, we identified the development of neutralizing antibodies and humoral memory responses prior to SARS-CoV-2 reinfection. However, these neutralizing antibodies failed to confer protection against reinfection, suggesting that additional factors are required for efficient prevention of SARS-CoV-2 reinfection. Further, we found no evidence supporting viral evasion of primary adaptive immune responses, suggesting that susceptibility to reinfection may be determined by host factors rather than pathogen adaptation.

Cytokine release syndrome in a patient with colorectal cancer after vaccination with BNT162b2.

Patients with cancer are currently prioritized in coronavirus disease 2019 (COVID-19) vaccination programs globally, which includes administration of mRNA vaccines. Cytokine release syndrome (CRS) has not been reported with mRNA vaccines and is an extremely rare immune-related adverse event of immune checkpoint inhibitors. We present a case of CRS that occurred 5 d after vaccination with BTN162b2 (tozinameran)-the Pfizer-BioNTech mRNA COVID-19 vaccine-in a patient with colorectal cancer on long-standing anti-PD-1 monotherapy. The CRS was evidenced by raised inflammatory markers, thrombocytopenia, elevated cytokine levels (IFN-γ/IL-2R/IL-18/IL-16/IL-10) and steroid responsiveness. The close temporal association of vaccination and diagnosis of CRS in this case suggests that CRS was a vaccine-related adverse event; with anti-PD1 blockade as a potential contributor. Overall, further prospective pharmacovigillence data are needed in patients with cancer, but the benefit-risk profile remains strongly in favor of COVID-19 vaccination in this population.

Protein-Based Immunome Wide Association Studies (PIWAS) for the Discovery of Significant Disease-Associated Antigens.

Identification of the antigens associated with antibodies is vital to understanding immune responses in the context of infection, autoimmunity, and cancer. Discovering antigens at a proteome scale could enable broader identification of antigens that are responsible for generating an immune response or driving a disease state. Although targeted tests for known antigens can be straightforward, discovering antigens at a proteome scale using protein and peptide arrays is time consuming and expensive. We leverage Serum Epitope Repertoire Analysis (SERA), an assay based on a random bacterial display peptide library coupled with next generation sequencing (NGS), to power the development of Protein-based Immunome Wide Association Study (PIWAS). PIWAS uses proteome-based signals to discover candidate antibody-antigen epitopes that are significantly elevated in a subset of cases compared to controls. After demonstrating statistical power relative to the magnitude and prevalence of effect in synthetic data, we apply PIWAS to systemic lupus erythematosus (SLE, n=31) and observe known autoantigens, Smith and Ribosomal protein P, within the 22 highest scoring candidate protein antigens across the entire human proteome. We validate the magnitude and location of the SLE specific signal against the Smith family of proteins using a cohort of patients who are positive by predicate anti-Sm tests. To test the generalizability of the method in an additional autoimmune disease, we identified and validated autoantigenic signals to SSB, CENPA, and keratin proteins in a cohort of individuals with Sjogren's syndrome (n=91). Collectively, these results suggest that PIWAS provides a powerful new tool to discover disease-associated serological antigens within any known proteome.

Case Study: Longitudinal immune profiling of a SARS-CoV-2 reinfection in a solid organ transplant recipient.

Prior to the emergence of antigenically distinct SARS-CoV-2 variants, reinfections were reported infrequently - presumably due to the generation of durable and protective immune responses. However, case reports also suggested that rare, repeated infections may occur as soon as 48 days following initial disease onset. The underlying immunologic deficiencies enabling SARS-CoV-2 reinfections are currently unknown. Here we describe a renal transplant recipient who developed recurrent, symptomatic SARS-CoV-2 infection - confirmed by whole virus genome sequencing - 7 months after primary infection. To elucidate the immunological mechanisms responsible for SARS-CoV-2 reinfection, we performed longitudinal profiling of cellular and humoral responses during both primary and recurrent SARS-CoV-2 infection. We found that the patient responded to the primary infection with transient, poor-quality adaptive immune responses. The patient's immune system was further compromised by intervening treatment for acute rejection of the renal allograft prior to reinfection. Importantly, we also identified the development of neutralizing antibodies and the formation of humoral memory responses prior to SARS-CoV-2 reinfection. However, these neutralizing antibodies failed to confer protection against reinfection, suggesting that additional factors are required for efficient prevention of SARS-CoV-2 reinfection. Further, we found no evidence supporting viral evasion of primary adaptive immune responses, suggesting that susceptibility to reinfection may be determined by host factors rather than pathogen adaptation in this patient. In summary, our study suggests that a low neutralizing antibody presence alone is not sufficient to confer resistance against reinfection. Thus, patients with solid organ transplantation, or patients who are otherwise immunosuppressed, who recover from infection with SARS-CoV-2 may not develop sufficient protective immunity and are at risk of reinfection.

Serum and Cervicovaginal Fluid Antibody Profiling in Herpes Simplex Virus-Seronegative Recipients of the HSV529 Vaccine.

Previous herpes simplex virus type 2 (HSV-2) vaccines have not prevented genital herpes. Concerns have been raised about the choice of antigen, the type of antibody induced by the vaccine, and whether antibody is present in the genital tract where infection occurs. We reported results of a trial of an HSV-2 replication-defective vaccine, HSV529, that induced serum neutralizing antibody responses in 78% of HSV-1-/HSV-2- vaccine recipients. Here we show that HSV-1-/HSV-2- vaccine recipients developed antibodies to epitopes of several viral proteins; however, fewer antibody epitopes were detected in vaccine recipients compared with naturally infected persons. HSV529 induced antibodies that mediated HSV-2-specific natural killer (NK) cell activation. Depletion of glycoprotein D (gD)-binding antibody from sera reduced neutralizing titers by 62% and NK cell activation by 81%. HSV-2 gD antibody was detected in cervicovaginal fluid at about one-third the level of that in serum. A vaccine that induces potent serum antibodies transported to the genital tract might reduce HSV genital infection.

Autoantibody Landscape in Patients with Advanced Prostate Cancer.

PURPOSE: Autoantibody responses in cancer are of great interest, as they may be concordant with T-cell responses to cancer antigens or predictive of response to cancer immunotherapies. Thus, we sought to characterize the antibody landscape of metastatic castration-resistant prostate cancer (mCRPC). EXPERIMENTAL DESIGN: Serum antibody epitope repertoire analysis (SERA) was performed on patient serum to identify tumor-specific neoepitopes. Somatic mutation-specific neoepitopes were investigated by associating serum epitope enrichment scores with whole-genome sequencing results from paired solid tumor metastasis biopsies and germline blood samples. A protein-based immunome-wide association study (PIWAS) was performed to identify significantly enriched epitopes, and candidate serum antibodies enriched in select patients were validated by ELISA profiling. A distinct cohort of patients with melanoma was evaluated to validate the top cancer-specific epitopes. RESULTS: SERA was performed on 1,229 serum samples obtained from 72 men with mCRPC and 1,157 healthy control patients. Twenty-nine of 6,636 somatic mutations (0.44%) were associated with an antibody response specific to the mutated peptide. PIWAS analyses identified motifs in 11 proteins, including NY-ESO-1 and HERVK-113, as immunogenic in mCRPC, and ELISA confirmed serum antibody enrichment in candidate patients. Confirmatory PIWAS, Identifying Motifs Using Next-generation sequencing Experiments (IMUNE), and ELISA analyses performed on serum samples from 106 patients with melanoma similarly revealed enriched cancer-specific antibody responses to NY-ESO-1. CONCLUSIONS: We present the first large-scale profiling of autoantibodies in advanced prostate cancer, utilizing a new antibody profiling approach to reveal novel cancer-specific antigens and epitopes. Our study recovers antigens of known importance and identifies novel tumor-specific epitopes of translational interest.

Integrated, multicohort analysis reveals unified signature of systemic lupus erythematosus.

Systemic lupus erythematosus (SLE) is a complex autoimmune disease that follows an unpredictable disease course and affects multiple organs and tissues. We performed an integrated, multicohort analysis of 7,471 transcriptomic profiles from 40 independent studies to identify robust gene expression changes associated with SLE. We identified a 93-gene signature (SLE MetaSignature) that is differentially expressed in the blood of patients with SLE compared with healthy volunteers; distinguishes SLE from other autoimmune, inflammatory, and infectious diseases; and persists across diverse tissues and cell types. The SLE MetaSignature correlated significantly with disease activity and other clinical measures of inflammation. We prospectively validated the SLE MetaSignature in an independent cohort of pediatric patients with SLE using a microfluidic quantitative PCR (qPCR) array. We found that 14 of the 93 genes in the SLE MetaSignature were independent of IFN-induced and neutrophil-related transcriptional profiles that have previously been associated with SLE. Pathway analysis revealed dysregulation associated with nucleic acid biosynthesis and immunometabolism in SLE. We further refined a neutropoiesis signature and identified underappreciated transcripts related to immune cells and oxidative stress. In our multicohort, transcriptomic analysis has uncovered underappreciated genes and pathways associated with SLE pathogenesis, with the potential to advance clinical diagnosis, biomarker development, and targeted therapeutics for SLE.

Leveraging heterogeneity across multiple datasets increases cell-mixture deconvolution accuracy and reduces biological and technical biases.

In silico quantification of cell proportions from mixed-cell transcriptomics data (deconvolution) requires a reference expression matrix, called basis matrix. We hypothesize that matrices created using only healthy samples from a single microarray platform would introduce biological and technical biases in deconvolution. We show presence of such biases in two existing matrices, IRIS and LM22, irrespective of deconvolution method. Here, we present immunoStates, a basis matrix built using 6160 samples with different disease states across 42 microarray platforms. We find that immunoStates significantly reduces biological and technical biases. Importantly, we find that different methods have virtually no or minimal effect once the basis matrix is chosen. We further show that cellular proportion estimates using immunoStates are consistently more correlated with measured proportions than IRIS and LM22, across all methods. Our results demonstrate the need and importance of incorporating biological and technical heterogeneity in a basis matrix for achieving consistently high accuracy.

Unsupervised Analysis of Transcriptomics in Bacterial Sepsis Across Multiple Datasets Reveals Three Robust Clusters.

OBJECTIVES: To find and validate generalizable sepsis subtypes using data-driven clustering. DESIGN: We used advanced informatics techniques to pool data from 14 bacterial sepsis transcriptomic datasets from eight different countries (n = 700). SETTING: Retrospective analysis. SUBJECTS: Persons admitted to the hospital with bacterial sepsis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A unified clustering analysis across 14 discovery datasets revealed three subtypes, which, based on functional analysis, we termed "Inflammopathic, Adaptive, and Coagulopathic." We then validated these subtypes in nine independent datasets from five different countries (n = 600). In both discovery and validation data, the Adaptive subtype is associated with a lower clinical severity and lower mortality rate, and the Coagulopathic subtype is associated with higher mortality and clinical coagulopathy. Further, these clusters are statistically associated with clusters derived by others in independent single sepsis cohorts. CONCLUSIONS: The three sepsis subtypes may represent a unifying framework for understanding the molecular heterogeneity of the sepsis syndrome. Further study could potentially enable a precision medicine approach of matching novel immunomodulatory therapies with septic patients most likely to benefit.

Interpretation of biological experiments changes with evolution of the Gene Ontology and its annotations.

Gene Ontology (GO) enrichment analysis is ubiquitously used for interpreting high throughput molecular data and generating hypotheses about underlying biological phenomena of experiments. However, the two building blocks of this analysis - the ontology and the annotations - evolve rapidly. We used gene signatures derived from 104 disease analyses to systematically evaluate how enrichment analysis results were affected by evolution of the GO over a decade. We found low consistency between enrichment analyses results obtained with early and more recent GO versions. Furthermore, there continues to be a strong annotation bias in the GO annotations where 58% of the annotations are for 16% of the human genes. Our analysis suggests that GO evolution may have affected the interpretation and possibly reproducibility of experiments over time. Hence, researchers must exercise caution when interpreting GO enrichment analyses and should reexamine previous analyses with the most recent GO version.

Gene annotation bias impedes biomedical research.

We found tremendous inequality across gene and protein annotation resources. We observed that this bias leads biomedical researchers to focus on richly annotated genes instead of those with the strongest molecular data. We advocate that researchers reduce these biases by pursuing data-driven hypotheses.

Methods to increase reproducibility in differential gene expression via meta-analysis.

Findings from clinical and biological studies are often not reproducible when tested in independent cohorts. Due to the testing of a large number of hypotheses and relatively small sample sizes, results from whole-genome expression studies in particular are often not reproducible. Compared to single-study analysis, gene expression meta-analysis can improve reproducibility by integrating data from multiple studies. However, there are multiple choices in designing and carrying out a meta-analysis. Yet, clear guidelines on best practices are scarce. Here, we hypothesized that studying subsets of very large meta-analyses would allow for systematic identification of best practices to improve reproducibility. We therefore constructed three very large gene expression meta-analyses from clinical samples, and then examined meta-analyses of subsets of the datasets (all combinations of datasets with up to N/2 samples and K/2 datasets) compared to a 'silver standard' of differentially expressed genes found in the entire cohort. We tested three random-effects meta-analysis models using this procedure. We showed relatively greater reproducibility with more-stringent effect size thresholds with relaxed significance thresholds; relatively lower reproducibility when imposing extraneous constraints on residual heterogeneity; and an underestimation of actual false positive rate by Benjamini-Hochberg correction. In addition, multivariate regression showed that the accuracy of a meta-analysis increased significantly with more included datasets even when controlling for sample size.

EMPOWERING MULTI-COHORT GENE EXPRESSION ANALYSIS TO INCREASE REPRODUCIBILITY.

A major contributor to the scientific reproducibility crisis has been that the results from homogeneous, single-center studies do not generalize to heterogeneous, real world populations. Multi-cohort gene expression analysis has helped to increase reproducibility by aggregating data from diverse populations into a single analysis. To make the multi-cohort analysis process more feasible, we have assembled an analysis pipeline which implements rigorously studied meta-analysis best practices. We have compiled and made publicly available the results of our own multi-cohort gene expression analysis of 103 diseases, spanning 615 studies and 36,915 samples, through a novel and interactive web application. As a result, we have made both the process of and the results from multi-cohort gene expression analysis more approachable for non-technical users.

A Case Study: Analyzing City Vitality with Four Pillars of Activity-Live, Work, Shop, and Play.

This case study evaluates and tracks vitality of a city (Seattle), based on a data-driven approach, using strategic, robust, and sustainable metrics. This case study was collaboratively conducted by the Downtown Seattle Association (DSA) and CDO Analytics teams. The DSA is a nonprofit organization focused on making the city of Seattle and its Downtown a healthy and vibrant place to Live, Work, Shop, and Play. DSA primarily operates through public policy advocacy, community and business development, and marketing. In 2010, the organization turned to CDO Analytics ( cdoanalytics.org ) to develop a process that can guide and strategically focus DSA efforts and resources for maximal benefit to the city of Seattle and its Downtown. CDO Analytics was asked to develop clear, easily understood, and robust metrics for a baseline evaluation of the health of the city, as well as for ongoing monitoring and comparisons of the vitality, sustainability, and growth. The DSA and CDO Analytics teams strategized on how to effectively assess and track the vitality of Seattle and its Downtown. The two teams filtered a variety of data sources, and evaluated the veracity of multiple diverse metrics. This iterative process resulted in the development of a small number of strategic, simple, reliable, and sustainable metrics across four pillars of activity: Live, Work, Shop, and Play. Data during the 5 years before 2010 were used for the development of the metrics and model and its training, and data during the 5 years from 2010 and on were used for testing and validation. This work enabled DSA to routinely track these strategic metrics, use them to monitor the vitality of Downtown Seattle, prioritize improvements, and identify new value-added programs. As a result, the four-pillar approach became an integral part of the data-driven decision-making and execution of the Seattle community's improvement activities. The approach described in this case study is actionable, robust, inexpensive, and easy to adopt and sustain. It can be applied to cities, districts, counties, regions, states, or countries, enabling cross-comparisons and improvements of vitality, sustainability, and growth.