Documenting the de-identification process of clinical and imaging data for AI for health imaging projects.

Artificial intelligence (AI) is revolutionizing the field of medical imaging, holding the potential to shift medicine from a reactive "sick-care" approach to a proactive focus on healthcare and prevention. The successful development of AI in this domain relies on access to large, comprehensive, and standardized real-world datasets that accurately represent diverse populations and diseases. However, images and data are sensitive, and as such, before using them in any way the data needs to be modified to protect the privacy of the patients. This paper explores the approaches in the domain of five EU projects working on the creation of ethically compliant and GDPR-regulated European medical imaging platforms, focused on cancer-related data. It presents the individual approaches to the de-identification of imaging data, and describes the problems and the solutions adopted in each case. Further, lessons learned are provided, enabling future projects to optimally handle the problem of data de-identification. CRITICAL RELEVANCE STATEMENT: This paper presents key approaches from five flagship EU projects for the de-identification of imaging and clinical data offering valuable insights and guidelines in the domain. KEY POINTS: ΑΙ models for health imaging require access to large amounts of data. Access to large imaging datasets requires an appropriate de-identification process. This paper provides de-identification guidelines from the AI for health imaging (AI4HI) projects.

Data infrastructures for AI in medical imaging: a report on the experiences of five EU projects.

Artificial intelligence (AI) is transforming the field of medical imaging and has the potential to bring medicine from the era of 'sick-care' to the era of healthcare and prevention. The development of AI requires access to large, complete, and harmonized real-world datasets, representative of the population, and disease diversity. However, to date, efforts are fragmented, based on single-institution, size-limited, and annotation-limited datasets. Available public datasets (e.g., The Cancer Imaging Archive, TCIA, USA) are limited in scope, making model generalizability really difficult. In this direction, five European Union projects are currently working on the development of big data infrastructures that will enable European, ethically and General Data Protection Regulation-compliant, quality-controlled, cancer-related, medical imaging platforms, in which both large-scale data and AI algorithms will coexist. The vision is to create sustainable AI cloud-based platforms for the development, implementation, verification, and validation of trustable, usable, and reliable AI models for addressing specific unmet needs regarding cancer care provision. In this paper, we present an overview of the development efforts highlighting challenges and approaches selected providing valuable feedback to future attempts in the area.Key points• Artificial intelligence models for health imaging require access to large amounts of harmonized imaging data and metadata.• Main infrastructures adopted either collect centrally anonymized data or enable access to pseudonymized distributed data.• Developing a common data model for storing all relevant information is a challenge.• Trust of data providers in data sharing initiatives is essential.• An online European Union meta-tool-repository is a necessity minimizing effort duplication for the various projects in the area.

Functionalization of polymeric nanoparticles with targeting VNAR ligands using vinyl sulfone conjugation.

Actively targeted drug loaded nanoparticles represent an exciting new form of therapeutics for cancer and other diseases. These formulations are complex and in order to realize their ultimate potential, optimization of their preparation is required. In this current study, we have examined the conjugation of a model targeting ligand, conjugated in a site-specific manner using a vinyl sulfone coupling approach. A disulfide-functionalized poly(L-lactide)-b-poly(oligo(ethylene glycol) methacrylate)-stat-(bis(2-methacryloyl)oxyethyl disulfide) (PLA-b-P(OEGMA-stat-DSDMA)) diblock copolymer was synthesized by simultaneous ring opening polymerization (ROP) and reversible addition-fragmentation chain transfer (RAFT) polymerization. Subsequently, the disulfide bonds of the polymer were reduced to thiols and divinyl sulfone was attached to the polymer using thiol-ene chemistry to produce the vinyl sulfone (VS)-functionalized PLA-b-P(OEGMA-stat-VSTEMA) amphiphilic block copolymer. Single emulsion - solvent evaporation was employed using a blend of this polymer with poly(D,L-lactide-co-glycolide) (PLGA) to produce VS-functionalized polymeric nanoparticles. The ability of these novel nanoparticles to attach ligands was then exemplified using a single domain variable new antigen receptor (VNAR) with a free carboxyl terminal cysteine residue. The resulting VNAR-functionalized nanoparticles were found to maintain specific affinity to their cognate antigen (DLL4) for at least 72 h at 4 °C. The simplicity of the degradable amphiphilic block copolymer synthesis and the efficiency of VNAR conjugation to the VS-functionalized nanoparticles show the potential of this platform for therapeutic development.

Development of next generation nanomedicine-based approaches for the treatment of cancer: we've barely scratched the surface.

Interest in nanomedicines has grown rapidly over the past two decades, owing to the promising therapeutic applications they may provide, particularly for the treatment of cancer. Personalised medicine and 'smart' actively targeted nanoparticles represent an opportunity to deliver therapies directly to cancer cells and provide sustained drug release, in turn providing overall lower off-target toxicity and increased therapeutic efficacy. However, the successful translation of nanomedicines from encouraging pre-clinical findings to the clinic has, to date, proven arduous. In this review, we will discuss the use of nanomedicines for the treatment of cancer, with a specific focus on the use of polymeric and lipid nanoparticle delivery systems. In particular, we examine approaches exploring the surface functionalisation of nanomedicines to elicit active targeting and therapeutic effects as well as challenges and future directions for nanoparticles in cancer treatment.

Anti-DLL4 VNAR targeted nanoparticles for targeting of both tumour and tumour associated vasculature.

Whilst there is an extensive body of preclinical nanomedicine research, translation to clinical settings has been slow. Here we present a novel approach to the targeted nanoparticle (NP) concept: utilizing both a novel targeting ligand, VNAR (Variable New Antigen Receptor), a shark-derived single chain binding domain, and an under-investigated target in delta-like ligand 4 (DLL4). We describe the development of an anti-DLL4 VNAR and the site-specific conjugation of this to poly(lactic-co-glycolic) acid PEGylated NPs using surface maleimide functional groups. These nanoconjugates were shown to specifically bind DLL4 with high affinity and were preferentially internalized by DLL4-expressing pancreatic cancer cell lines and endothelial cells. Furthermore, a distinct anti-angiogenic effect endowed by the anti-DLL4 VNAR was evident in in vitro tubulogenic assays. Taken together these findings highlight the potential of anti-DLL4 targeted polymeric NPs as a novel therapeutic approach in pancreatic cancer.

In Vitro ELISA and Cell-Based Assays Confirm the Low Immunogenicity of VNAR Therapeutic Constructs in a Mouse Model of Human RA: An Encouraging Milestone to Further Clinical Drug Development.

Anti-drug antibodies (ADAs), specific for biotherapeutic drugs, are associated with reduced serum drug levels and compromised therapeutic response. The impact of ADA on the bioavailability and clinical efficacy of blockbuster anti-hTNF-α monoclonal antibodies is well recognised, especially for adalimumab and infliximab treatments, with the large and complex molecular architecture of classical immunoglobulin antibody drugs, in part, responsible for the immunogenicity seen in patients. The initial aim of this study was to develop solid-phase enzyme-linked immunosorbent assays (ELISA) and an in vitro cell-based method to accurately detect ADA and estimate its impact on the preclinical in vivo efficacy outcomes of two novel, nonimmunoglobulin VNAR fusion anti-hTNF-α biologics (Quad-X™ and D1-NDure™-C4) and Humira®, a brand of adalimumab. Serum drug levels and the presence of ADA were determined in a transgenic mouse model of polyarthritis (Tg197) when Quad-X™ and Humira® were dosed at 1 mg/kg and D1-NDure™-C4 was dosed at 30 mg/kg. The serum levels of the Quad-X™ and D1-NDure™-C4 modalities were consistently high and comparable across all mice within the same treatment groups. In 1 mg/kg and 3 mg/kg Quad-X™- and 30 mg/kg D1-NDure™-C4-treated mice, an average trough drug serum concentration of 8 µg/mL, 50 µg/mL, and 350 µg/mL, respectively, were estimated. In stark contrast, Humira® trough serum concentrations in the 1 mg/kg treatment group ranged from <0.008 µg/mL to 4 µg/mL with trace levels detected in 7 of the 8 animals treated. Trough serum Humira® and Quad-X™ concentrations in 3 mg/kg treatment samples were comparable; however, the functionality of the detected Humira® serum was significantly compromised due to neutralising ADA. The impact of ADA went beyond the simple and rapid clearance of Humira®, as 7/8 serum samples also showed no detectable capacity to neutralise hTNF-α-mediated cytotoxicity in a murine fibrosarcoma (L929) cell assay. The neutralisation capacity of all the VNAR constructs remained unchanged at the end of the experimental period (10 weeks). The data presented in this manuscript goes some way to explain the exciting outcomes of the previously published preclinical in vivo efficacy data, which showed complete control of disease at Quad-X™ concentrations of 0.5 mg/kg, equivalent to 10x the in vivo potency of Humira®. This independent corroboration also validates the robustness and reliability of the assay techniques reported in this current manuscript, and while it comes with the caveat of a mouse study, it does appear to suggest that these particular VNAR constructs, at least, are of low inherent immunogenicity.

Uveitis Therapy With Shark Variable Novel Antigen Receptor Domains Targeting Tumor Necrosis Factor Alpha or Inducible T-Cell Costimulatory Ligand.

PURPOSE: We assess the efficacy of two next-generation biologic therapies in treating experimental autoimmune uveitis. METHODS: Variable binding domains from shark immunoglobulin novel antigen receptors (VNARs) were fused with a mouse IgG2a constant domain (Fc) to generate VNAR-Fc molecules with binding specificity to tumor necrosis factor alpha (TNFα) or inducible T-cell costimulatory ligand (ICOSL). Treatment with VNAR-Fc fusion proteins was compared to treatment with dexamethasone or vehicle in the Lewis rat model of experimental autoimmune uveitis (EAU). Inflammation control was determined by comparing OCT clinical and histologic scores, and aqueous humor protein concentration. The concentration of 27 inflammatory cytokines in the aqueous humor was measured using a multiplex enzyme-linked immunosorbent assay platform. RESULTS: Administration of S17-Fc significantly decreased clinical, histologic, and aqueous protein levels when compared to vehicle treatment. Inflammation scores and aqueous protein levels in A5-Fc-treated animals were decreased compared to vehicle treatment, but not significantly. The concentration of vascular endothelial growth factor (VEGF), regulated on activation, normal T cell expressed and secreted (RANTES), macrophage inflammatory protein 1 alpha (MIP-1α), interleukin (IL)-1ß, LPS-induced CXC chemokine (LIX), monocyte chemoattractant protein-1 (MCP-1), and interferon (IFN)-γ were significantly decreased in the eyes of animals treated with dexamethasone. VNAR treatment demonstrated a trend towards decreased cytokine concentrations, but only VEGF and RANTES were significantly decreased by S17-Fc. CONCLUSIONS: Treatment with the anti-TNFα VNAR S17-Fc ameliorates EAU as effectively as treatment with corticosteroids. TRANSLATIONAL RELEVANCE: VNAR-Fc molecules are a next-generation therapeutic biologic that overcome the limitations of classical biologic monoclonal antibodies, such as complex structure, large size, and limited tissue penetration. This is a novel drug modality that could result in the development of new therapy options for patients with noninfectious uveitis.

Oriented attachment of VNAR proteins, via site-selective modification, on PLGA-PEG nanoparticles enhances nanoconjugate performance.

Herein we report the construction of a nanoparticle-based drug delivery system which targets a key regulator in tumour angiogenesis. We exploit a Variable New Antigen Receptor (VNAR) domain, conjugated using site-specific chemistry, to direct poly lactic acid-co-glycolic acid-polyethylene glycol (PLGA-PEG) nanoparticles to delta like canonical Notch ligand 4 (DLL4). The importance of site-specific chemistry is demonstrated.

An Anti-hTNF-α Variable New Antigen Receptor Format Demonstrates Superior in vivo Preclinical Efficacy to Humira® in a Transgenic Mouse Autoimmune Polyarthritis Disease Model.

Tumor necrosis factor-alpha (TNF-α), an established pro-inflammatory cytokine plays a central role in the induction and progression of several chronic inflammatory and autoimmune diseases. Targeting TNF-α as a treatment modality has shown tremendous success, however there are several limitations associated with the current anti-TNF-α biologic drugs including: immunogenicity, life-threatening infections, resistance to treatment, complexity of manufacture and cost of treatment. Here, we report the in vivo efficacy of novel anti-TNF-α formats generated from molecular engineering of variable new antigen receptors (VNARs), originally derived from the immune system of an immunized nurse shark. Two anti-TNF-α VNAR formats, a tandem multivalent trimer, D1-BA11-C4 and an Fc-fused quadrivalent D1-Fc-C4 (Quad-X™) construct were tested in a clinically relevant, preclinical mouse efficacy model of polyarthritis (Tg197) and compared to the commercial anti-TNF-α "best in class" therapy, Adalimumab (Humira®). Both VNAR formats bind and neutralize TNF-α through an epitope that appears to be different from those recognized by other anti-TNF biologics used clinically. All doses of Quad-X™, from 0.5 to 30 mg/kg, significantly blocked the development of polyarthritis. At 0.5 mg/kg Quad-X™, the arthritis score was improved by 76% and the histopathology score by 63%. At 3 mg/kg Quad-X™, control of disease was almost complete at 90% (arthritis) and 88% (histopathology). In marked contrast, 1 mg/kg Humira® saw profound disease breakthrough with scores of 39 and 16% respectively, increasing to a respectable 82 and 86% inhibition at 10 mg/kg Humira®. We have previously reported the superior potency of anti-TNF-α VNARs in vitro and in these studies translate this superiority into an in vivo setting and demonstrate the potential of VNAR formats to meet the requirements of next-generation anti-TNF-α therapies.

Next-generation flexible formats of VNAR domains expand the drug platform's utility and developability.

Therapeutic mAbs have delivered several blockbuster drugs in oncology and autoimmune inflammatory disease. Revenue for mAbs continues to rise, even in the face of competition from a growing portfolio of biosimilars. Despite this success, there are still limitations associated with the use of mAbs as therapeutic molecules. With a molecular mass of 150 kDa, a two-chain structure and complex glycosylation these challenges include a high cost of goods, limited delivery options, and poor solid tumour penetration. There remains an urgency to create alternatives to antibody scaffolds in a bid to circumvent these limitations, while maintaining or improving the therapeutic success of conventional mAb formats. Smaller, less complex binders, with increased domain valency, multi-specific/paratopic targeting, tuneable serum half-life and low inherent immunogenicity are a few of the characteristics being explored by the next generation of biologic molecules. One novel 'antibody-like' binder that has naturally evolved over 450 million years is the variable new antigen receptor (VNAR) identified as a key component of the adaptive immune system of sharks. At only 11 kDa, these single-domain structures are the smallest IgG-like proteins in the animal kingdom and provide an excellent platform for molecular engineering and biologics drug discovery. VNAR attributes include high affinity for target, ease of expression, stability, solubility, multi-specificity, and increased potential for solid tissue penetration. This review article documents the recent drug developmental milestones achieved for therapeutic VNARs and highlights the first reported evidence of the efficacy of these domains in clinically relevant models of disease.

Novel, Anti-hTNF-α Variable New Antigen Receptor Formats with Enhanced Neutralizing Potency and Multifunctionality, Generated for Therapeutic Development.

The management of chronic inflammatory diseases, such as inflammatory bowel disease, psoriasis, and rheumatoid arthritis has significantly improved over the last decade with the clinical availability of anti-TNF-α biologics. Despite this undoubted treatment success, a combination of acquired resistance together with an increased risk of systemic complications, means that a significant number of patients either fail to find a suitable targeted therapy or frustratingly discover that an approach that did work is no longer efficacious. Here, we report the isolation and characterization of a new class of super-neutralizing anti-TNF-α biologics formats, the building blocks of which were originally derived as variable new antigen receptor (VNAR) domains from an immunized nurse shark. These parental small, stable VNAR monomers recognize and neutralize tumor necrosis factor (TNF)-α, in cell-based assays, at nanomolar concentrations. However, the simple, single-chain molecular architecture of VNARs allows for easy and multiple reformatting options. Through reformatting, we achieved a 50,000-fold enhancement in in vitro efficacy with super-neutralizing fusion proteins able to block TNF-α induced cytotoxicity in the 2-5 pM range while retaining other functionality through the addition of fusion proteins known to extend serum half-life in vivo. In an in vitro intestinal epithelial barrier dysfunction efficacy model, the lead VNAR domains, restored barrier function and prevented paracellular flux with comparable efficacy to adalimumab (Humira®). In addition, all multivalent VNAR constructs restored trans-epithelial electrical resistance (TEER) to approximately 94% of the untreated control. Reformatted VNAR domains should be considered as a new class of biologic agents for the treatment of hTNF-α driven diseases; either used systemically with appropriate half-life extension or alternatively where site-specific delivery of small and stable neutralizers may provide improvements to current therapy options.

Shark novel antigen receptors--the next generation of biologic therapeutics?

Over recent decades we have witnessed a revolution in health care as new classes of therapeutics based on natural biological molecules have become available to medical practitioners. These promised to target some of the most serious conditions that had previously evaded traditional small molecule drugs, such as cancers and to alleviate many of the concerns of patients and doctors alike regarding adverse side effects and impaired quality of life that are often associated with chemo-therapeutics. Many early 'biologics' were based on antibodies, Nature's answer to invading pathogens and malignancies, derived from rodents and in many ways failed to live up to expectations. Most of these issues were subsequently negated by technological advances that saw the introduction of human or "humanized' antibodies and have resulted in a number of commercial 'block-busters'. Today, most of the large pharmaceutical companies have product pipelines that include an increasing proportion of biologic as opposed to small molecule compounds. The limitations of antibodies or other large protein drugs are now being realized however and ever more inventive solutions are being sought to develop equally efficacious but smaller, more soluble, more stable and less costly alternatives to broaden the range of drug-able targets and therapeutic options. The aim of this chapter is to introduce the reader to one such novel approach that seeks to exploit a unique antibody-like protein evolved by ancestral sharks over 450 M years ago and that may lead to a host of new therapeutic opportunities and help us to tackle some of the pressing clinical demands of the 21 st century.

Antibody-targeted myofibroblast apoptosis reduces fibrosis during sustained liver injury.

BACKGROUND/AIMS: Myofibroblast apoptosis promotes the resolution of liver fibrosis. However, retaining macrophages may enhance reversal. The effects of specifically stimulating myofibroblast apoptosis in vivo were assessed. METHODS: A single chain antibody (C1-3) to an extracellular domain of a myofibroblast membrane protein was injected as a fluorescent- or gliotoxin conjugate into mice with liver fibrosis. RESULTS: C1-3 specifically targeted alpha-smooth muscle actin positive liver myofibroblasts within scar regions of the liver in vivo and did not co-localise with liver monocytes/macrophages. Injection of free gliotoxin stimulated a 2-fold increase in non-parenchymal cell apoptosis and depleted liver myofibroblasts by 30% and monocytes/macrophages by 50% but had no effect on fibrosis severity in the sustained injury model employed. In contrast, C1-3-targeted gliotoxin stimulated a 5-fold increase in non-parenchymal cell apoptosis, depleted liver myofibroblasts by 60%, did not affect the number of monocytes/macrophages and significantly reduced fibrosis severity. Fibrosis reduction was associated with increased metalloproteinase-13 levels. CONCLUSIONS: These data demonstrate that specific targeting of liver myofibroblast apoptosis is the most effective anti-fibrogenic therapy, supporting a role for liver monocytes and/or macrophages in the promotion of liver fibrosis reduction.