The Therapeutics Landscape of Miniproteins

1. Introduction

2026 marks 40 years since the approval of the first monoclonal antibody (mAb), muromononab-CD3 (OKT3), in 1986. Over four decades, the pharmaceutical industry has accumulated an extraordinary body of knowledge on how to discover, engineer, manufacture, and safely administer antibody-based therapeutics. Issues such as immunogenicity, developability, and manufacturability have been extensively mitigated, and regulatory frameworks have matured accordingly. As of September 2025, nearly 200 antibody-based therapeutics have been approved by global health authorities, and five of the top ten best-selling drugs in 2024 were antibody-derived molecules, including pembrolizumab, dupilumab, and daratumumab.

Yet for all their success, monoclonal antibodies carry well-recognised limitations. Their large molecular size (approximately 150 kDa) restricts tissue penetration, prevents crossing of the blood-brain barrier, and makes oral administration essentially impossible due to gastrointestinal degradation. Their complex glycosylated structure necessitates expensive mammalian cell-based manufacturing systems, and cold-chain requirements add further logistical and cost burdens to global access.

It is in this context that miniproteins have emerged as a compelling alternative modality, as a complementary class of therapeutics capable of addressing clinical needs that conventional antibodies cannot efficiently serve.

1.1 Definition and Structural Characteristics

Miniproteins are a diverse group of polypeptide scaffolds generally ranging from 40 to 120 amino acids in length, with molecular weights typically between 4 and 15 kDa. They are characterised by compact, stable three-dimensional structures, stabilized through alpha-helices, beta-sheets, disulfide bonds, or cystine-reinforced frameworks, which confer exceptional thermodynamic stability and resistance to proteolytic degradation.

Despite their small size, miniproteins can achieve antibody-like binding affinity and selectivity toward their targets. Their rigid scaffolds incorporate diversifiable surface loops or helical faces that can be engineered through directed evolution, phage display, mRNA display, yeast surface display, or computational design, to engage molecular targets with nanomolar or sub-nanomolar affinity. 

1.2 Advantages Over Monoclonal Antibodies

The pharmacological and manufacturing advantages of miniproteins are substantial and multifaceted:

Size-driven pharmacokinetics. At 4–15 kDa, miniproteins are 10 to 20 times smaller than conventional monoclonal antibodies. This confers superior access to buried epitopes, increased tissue and tumor penetration, faster biodistribution to target compartments, and, importantly, the potential to access compartments inaccessible to antibodies. Some miniprotein classes, particularly cystine-dense scaffolds and engineered variants, demonstrate the ability to cross epithelial barriers relevant to oral delivery, a property that remains essentially unattainable for conventional antibodies.

Manufacturability. Unlike monoclonal antibodies, which require complex glycosylation patterns achieved only in mammalian cell systems (CHO, HEK), many miniproteins can be efficiently produced in bacterial expression systems such as Escherichia coli, or in yeast. This can dramatically reduce manufacturing cost and complexity. Miniprotein scaffolds typically have no requirement for glycosylation, are not prone to misfolding into aggregates, and can be produced with high yield and strong batch-to-batch consistency.

Oral bioavailability. The potential for orally administered miniprotein therapeutics represents perhaps the most transformative advantage of the class. Computationally designed miniproteins targeting IL-23R and IL-17, cytokine pathways central to autoimmune diseases such as inflammatory bowel disease (IBD) and psoriasis, have demonstrated proof-of-concept for once-daily oral administration in preclinical models of IBD using humanised mice, resulting in significant clinical improvement. Although these findings require validation in clinical studies, they suggest that miniproteins could eventually replace subcutaneous injections for chronic inflammatory conditions, profoundly improving patient quality of life and treatment adherence.

Stability. Many miniprotein scaffolds exhibit exceptional thermostability, maintaining their folded conformation at temperatures that would denature conventional antibodies. Certain scaffolds are room-temperature stable for extended periods, a property with direct implications for cold-chain-independent distribution in low-resource settings and patient self-administration.

Engineering flexibility. Miniproteins can serve as modular building blocks. Multiple binding domains can be combined in tandem or "beads-on-a-string" configurations to create bispecific or multispecific agents targeting two or more molecular targets simultaneously. They can also be fused to human serum albumin (HSA), polyethylene glycol (PEG) or FcRN binding miniproteins to extend serum half-life, compensating for the rapid renal clearance that their small size would otherwise produce. Additionally, miniproteins have demonstrated compatibility with radio-conjugate, drug-conjugate, and CAR-T cell engineering platforms.

Targeting previously intractable sites. Their compact size and structural rigidity allow miniproteins to engage concave binding pockets, enzyme active sites, protein-protein interfaces, and epitopes on viral proteins that are cryptic or inaccessible to the larger paratope footprint of conventional antibodies.

Immunogenicity. Many miniprotein scaffolds are derived from, or engineered to closely resemble, human proteins, conferring intrinsically low immunogenic potential. For example, adnectins are derived from the tenth type III domain of human fibronectin, a naturally abundant extracellular protein, virtually eliminating the risk of anti-drug antibody formation that can compromise antibody therapy.

1.3 Scaffold Classes Within the Miniprotein Universe

Several distinct scaffold families have been developed and advanced into clinical or commercial stages:

Adnectins (Monobodies / FN3 domains). Adnectins are derived from the tenth type III domain of human fibronectin (10Fn3), a beta-sheet scaffold of approximately 10 kDa with variable loops analogous to antibody CDRs. Critically, adnectins contain no disulfide bonds and are not glycosylated, enabling efficient bacterial expression with high thermal stability and monomeric solution behavior. The adnectin technology was pioneered by Adnexus Therapeutics (acquired by Bristol-Myers Squibb) and has since been licensed and advanced by multiple organisations.

VHH / Nanobodies (Single-Domain Antibodies). VHH domains are the variable domains of the heavy-chain-only antibodies naturally found in camelids (llamas, camels, alpacas) and cartilaginous fish. At approximately 15 kDa, they are the smallest naturally occurring antigen-binding fragments. VHHs bind targets through three CDR loops, containing a CDR3 loop that can penetrate binding clefts inaccessible to conventional antibodies. They exhibit outstanding solubility, stability, including resistance to extreme temperatures and pH, and ease of production in microbial systems. VHH domains can be readily formatted as bispecific or multispecific constructs, Fc-fused dimers, or as antigen-recognition modules in CAR-T cells.

Affibodies. Derived from the Z-domain of staphylococcal Protein A, affibodies are three-helix bundle proteins of approximately 6 kDa. Their helical binding surface is diversifiable to generate high-affinity binders against a wide range of targets.

DARPins (Designed Ankyrin Repeat Proteins). DARPins are derived from the natural ankyrin repeat protein scaffold and use a repeat protein architecture to present a flat or concave binding surface rather than loops. They are thermostable and highly amenable to protein engineering.

Knottins and Cystine-Dense Peptides. These scaffolds are characterised by a network of disulfide bonds (typically three) that create an exceptionally compact, stable structure resistant to both thermal and proteolytic degradation. Their intrinsic stability is the primary rationale for oral bioavailability studies.

De Novo Computationally Designed Miniproteins. A frontier category that has emerged rapidly with advances in protein design software (Rosetta) and deep learning methods (ProteinMPNN, AlphaFold2, RFdiffusion). Pioneered largely by David Baker's group at the University of Washington, this approach designs miniprotein binders computationally from first principles, without reliance on a naturally occurring scaffold. This has been applied to SARS-CoV-2 spike protein neutralisation, MERS-CoV spike protein, Francisella tularensis virulence factors, and multiple oncology targets.

2. Therapeutic Landscape: Approved Agents

2.1 Approved Miniprotein Therapeutics in the United States

As of early 2026, two miniprotein-based therapeutics have received FDA approval, with additional approvals in other global markets. Importantly, a broader picture of the VHH landscape reveals that five VHH-based therapeutics have been approved globally, demonstrating the maturation of the class.

Caplacizumab (Cablivi: Sanofi/Ablynx) Caplacizumab was approved by the EMA in 2018 and by the FDA in February 2019, marking it as the first approved nanobody-based medicine globally. It is a humanised bivalent VHH consisting of  two identical anti-VWF (von Willebrand Factor) nanobodies joined by a short triple-alanine linker, with a molecular weight of approximately 28 kDa. It is indicated for the treatment of adults with acquired thrombotic thrombocytopenic purpura (aTTP), a rare, life-threatening blood-clotting disorder caused by impaired ADAMTS13 activity leading to ultra-large VWF multimers that trigger platelet aggregation. Caplacizumab inhibits the interaction between ultra-large VWF and platelets, rapidly resolving the thrombotic process. The pivotal HERCULES Phase III trial demonstrated faster platelet count recovery, fewer plasma exchange sessions, and shorter hospital stays compared to placebo. The bivalent design maximises avidity while retaining the compact engineering advantages of nanobodies.

Lerodalcibep (Lerochol: LIB Therapeutics) In December 2025, the FDA approved lerodalcibep-liga (Lerochol), marking a historic milestone: the first approval of an adnectin-based therapeutic anywhere in the world, and the first non-VHH miniprotein scaffold to receive FDA approval. Lerochol is a third-generation PCSK9 inhibitor indicated for adults with hypercholesterolemia, including heterozygous familial hypercholesterolemia (HeFH). Its anti-PCSK9 binding domain is an 11-kDa adnectin engineered for high-affinity subnanomolar binding to human PCSK9, fused to human serum albumin (HSA) to extend plasma half-life. The HSA fusion is a critical pharmacokinetic engineering strategy. Without it, the small adnectin would be rapidly cleared by renal filtration. Lerochol is administered as a once-monthly self-injected subcutaneous dose. The Phase III LIBerate program enrolled over 2,900 patients across five global trials. In LIBerate-CVD, lerodalcibep reduced LDL-C by 55% versus placebo at 52 weeks; in LIBerate-FH, the reduction was 59% versus placebo at 24 weeks, with no treatment-related serious adverse events in the long-term extension. The adnectin scaffold's origins in human fibronectin, a naturally abundant protein in human serum, confer very low immunogenic potential, consistent with minimal injection site reactions compared to antibody-based PCSK9 inhibitors.

2.2 Additional Global VHH Approvals

Beyond the US market, the global regulatory approvals of VHH-based therapeutics reflect the growing mainstream acceptance of the class:

• Ozoralizumab (Nanozora: Taisho Pharmaceutical / AbbVie): Approved in Japan in September 2022 for rheumatoid arthritis. A trivalent "beads-on-a-string" construct with two anti-TNFα VHHs flanking one anti-HSA VHH for half-life extension, with a total molecular weight of 38 kDa.
• Envafolimab (KN-035: Alphamab Oncology): Approved in China in 2021 for adult patients with MSI-H or dMMR advanced solid tumors. Notably, it is the first subcutaneously injectable PD-L1 nanobody approved globally, a format enabled by the compact VHH scaffold.
• Ciltacabtagene autoleucel (cilta-cel, Carvykti: Legend Biotech/Janssen): FDA-approved in February 2022 for relapsed/refractory multiple myeloma. A CAR-T cell therapy whose ectodomain harbors biparatopic anti-BCMA VHHs arranged in tandem, demonstrating the compatibility of VHH domains with advanced cellular therapy platforms.

3. Clinical Pipeline

3.1 VHH / Nanobody Clinical Pipeline

The VHH therapeutic pipeline is now extensive, with over 30 clinical trials ongoing globally, spanning oncology, autoimmune disease, cardiovascular disease, infectious disease, and neurodegeneration. Key areas of investigation include:

• Oncology: VHH-based candidates are in trials targeting PD-L1, EGFR, HER2, BCMA, and tumor-associated antigens. VHH-drug conjugates (analogous to ADCs) and bispecific VHH constructs engaging T cells represent particularly active areas.
• Autoimmune and inflammatory disease: TNFα, IL-6, IL-17, and IL-23 pathway-targeting VHHs are under investigation for rheumatoid arthritis, IBD, psoriasis, and related conditions. Ozoralizumab's success in Japan has validated this space.
• Neurodegeneration: The ability of some VHH formats to cross the blood-brain barrier through transcytosis, or to be delivered intranasally, is driving investigation in Alzheimer's disease, Parkinson's disease, and ALS, areas where conventional antibodies face fundamental delivery barriers.
• Cardiovascular and rare diseases: Following caplacizumab's success, additional VHH candidates targeting coagulation cascade components are in development.
• Infectious disease: Inhaled VHH formats for respiratory pathogens (influenza, SARS-CoV-2, RSV) are being explored, leveraging both the small size for deep lung deposition and the stability advantages for manufacturing.

3.2 Adnectin and Other Scaffold Clinical Pipeline

Taldefgrobep alfa (BMS-986089 / RG6206 — Biohaven / Roche): An anti-myostatin adnectin, the anti-myostatin binding domain fused to the Fc domain of an IgG, originally developed by Bristol-Myers Squibb and subsequently licensed to Roche and Biohaven. The program targets myostatin inhibition for muscle-wasting diseases, with clinical evaluation in spinal muscular atrophy (SMA) and Duchenne muscular dystrophy (DMD). In Phase I studies in healthy adults, weekly subcutaneous doses increased thigh muscle volume and total lean body mass. The DMD Phase II trial did not demonstrate efficacy, but the SMA program is still ongoing. 

AKY-1189 (Aktis Oncology): A miniprotein radioconjugate targeting Nectin-4, granted FDA Fast Track Designation in February 2026 for locally advanced or metastatic urothelial carcinoma. A Phase 1b trial is underway evaluating safety and clinical activity in urothelial carcinoma, breast cancer, NSCLC, colorectal cancer, cervical cancer, and head and neck cancer. This agent exemplifies a growing class of miniprotein-based targeted radiopharmaceuticals, where the favorable pharmacokinetics of miniproteins, rapid tumor accumulation, fast clearance from non-target compartments, are particularly advantageous for minimising radiation dose to healthy tissue.

Orally delivered Th17 antagonist miniproteins (preclinical / early development): Academic and industry groups have advanced computationally designed miniproteins targeting IL-23R and IL-17 to preclinical proof-of-concept for oral delivery. While not yet in Phase I trials, this program represents a potential paradigm shift for the treatment of IBD and psoriasis.

De Novo Computationally Designed Miniproteins and AI Platforms:  A frontier category that has emerged rapidly with advances in protein design software, deep learning methods, and generative AI. Rather than relying on naturally occurring scaffolds, this approach designs miniprotein binders computationally from first principles. This space is increasingly being driven by specialized biotech companies combining AI, synthetic biology, and high-throughput validation to drastically reduce discovery timelines:

• AI Proteins: Founded by David Baker-trained scientist Chris Bahl, AI Proteins has built a platform combining generative AI, synthetic biology, and robotics for de novo miniprotein design. Having developed molecules targeting over 150 different sites with several achieving in vivo proof-of-concept, the company entered a research collaboration with Bristol Myers Squibb in December 2024 potentially worth up to $400 million. A miniprotein tumor antigen binder developed by AI Proteins is reported to have entered clinical trials. The company recently raised a $41.5 million Series A financing round to advance its pipeline, which spans diverse therapeutic areas including a recent focus on targeted radioligand cancer therapies (theranostics).
• Ordaōs Bio: Headquartered in New York, the company utilises a proprietary generative AI Design Engine, powered by multitask meta-learning and reinforcement learning, to engineer novel mini-proteins completely from scratch. Their approach integrates computational design with an unprecedented 5-day wet-lab validation cycle using cell-free expression systems. This rapid iteration allows for the multi-objective optimisation of binding, stability, and safety. The platform is being used to is develop candidates for complex indications, such as cytokine receptor mimetics for inflammatory bowel disease (IBD), metabolic diseases, and biodistribution modules.
• VRG Therapeutics (VRG Tx): This biopharmaceutical company leverages its proprietary AI-powered WISDOM (Wetlab-verified In Silico Design of Miniproteins) platform. VRG Tx pairs in silico scaffold design with a wet-lab sequence enrichment technology (ISEP) to rapidly generate high-affinity miniproteins. Their pipeline targets major unmet clinical needs in autoimmune diseases, inflammation (such as blocking IL-6 trans-signaling), and oncology.
• Evozyne: Driven by their AI-native EvoGen engine, Evozyne learns how protein sequences drive function to design novel therapeutic proteins and miniproteins. The company focuses on overcoming limitations in traditional biologics, advancing candidates optimised for potency, stability, and specificity primarily into their immunology and respiratory disease pipelines.
• DenovAI: An early-stage biotechnology startup developing a dedicated AI-based platform designed to rapidly accelerate drug discovery by predicting and generating high-affinity miniprotein and antibody binders for challenging protein targets entirely from scratch.
• Broader Applications (Quercus Biosolutions): The utility of generative AI in miniprotein design is also expanding beyond human medicine. For instance, startups like Quercus Biosolutions have begun applying AI platforms to design de novo miniproteins for agricultural crop protection, highlighting the exceptional stability and versatility of this structural class.

4. Manufacturing and Developability Considerations

One of the enduring advantages of miniproteins, relative to monoclonal antibodies, is the opportunity for simpler, lower-cost manufacturing. Bacterial expression of miniproteins like adnectins, which lack disulfide bonds and glycosylation, can produce high yields at a fraction of the capital and operating cost of mammalian cell culture. VHH domains, while sometimes requiring disulfide bond formation, can be expressed efficiently in E. coli or yeast. This has implications for both commercial competitiveness and for global access, particularly relevant in Asia-Pacific markets where cost-of-goods remains a critical variable for market uptake.

Pharmacokinetic engineering remains a key developability challenge. The small molecular size of most miniproteins places them below the kidney's glomerular filtration threshold of approximately 50–60 kDa, leading to rapid renal clearance and, in the case of radiolabeled formats, elevated kidney radiation doses. Multiple strategies are employed to address this: PEGylation (as used in the BMS-962476 adnectin PCSK9 program), HSA fusion (as used in lerodalcibep), Fc-fusion, and multimerisation. Each approach carries distinct trade-offs in half-life extension, immunogenicity, and manufacturing complexity.

5. Outlook

The approval of lerodalcibep in December 2025, decades after adnectin technology was first conceived, and the now-established global footprint of VHH-based therapeutics signal that miniproteins have crossed from a research curiosity into a bona fide commercial modality. The combination of AI-driven de novo design, expanding display and screening technologies, and growing clinical validation across multiple scaffold classes positions miniproteins for a period of accelerating development.

In oncology, the emergence of miniprotein radioconjugates and miniprotein-based CAR-T architectures offers differentiated mechanisms building on the pharmacokinetic advantages of small size. In autoimmune disease, the ongoing pursuit of orally bioavailable miniproteins targeting IL-23 and IL-17 pathways, currently dominated by large mAbs administered by injection, represents one of the highest-value opportunities in the field. In neurology, the potential to engineer miniproteins capable of crossing the blood-brain barrier through transcytosis opens therapeutic access to targets that have long been beyond the reach of conventional biologics.

For the Asia-Pacific pharmaceutical market, miniproteins represent a compelling strategic opportunity. Manufacturing simplicity lowers barriers to entry for regional biomanufacturers. Multiple regulatory approvals in Asia, most notably envafolimab in China, have established precedent. And the substantial burden of cardiovascular disease, autoimmune conditions, and oncology across Asian populations creates significant unmet need for cost-effective, convenient biologics. As the class matures from its early approved agents toward a broader pipeline, miniproteins are well positioned to be a defining modality of the next decade of biopharmaceutical innovation.