300 Million Patients, 5% Treated: The AI Opportunity in Rare Disease

A $621 Billion Market With 95% of Diseases Untreated

Rare diseases affect approximately 300 million people worldwide, roughly equivalent to the population of the United States. The term "rare" is misleading. Each individual condition may affect a small number of patients, but collectively rare diseases constitute the largest category of unmet medical need in healthcare and a persistently lucrative segment of pharmaceutical development.

The orphan drug market, valued at $193 billion in 2024, is projected to reach $621.85 billion by 2034, a compound annual growth rate of 12.24%. Three forces are driving this growth: regulatory incentives that reduce development risk, premium pricing enabled by limited competition, and advances in gene therapy and biologics that are making previously untreatable conditions addressable.

The gap is stark. Of the more than 7,000 known rare diseases, only approximately 5% have an FDA-approved therapy. That leaves 95% of rare disease patients with no approved treatment options. Pharmaceutical AI platforms that can solve the formulation challenges specific to rare diseases are positioned to capture a disproportionate share of this market.

Why Smaller Populations Produce Bigger Returns

Conventional pharmaceutical economics favours larger patient populations: more patients, more revenue. Orphan drugs invert this logic. Despite serving small populations, orphan drugs consistently outperform conventional drugs on margins, development cost, and return on investment.

Orphan drugs achieve average gross profit margins of approximately 80%, compared to the pharmaceutical industry average of 16%. Development costs run approximately 27% lower than for non-orphan drugs because clinical trials are smaller and regulatory pathways faster. The Orphan Drug Act guarantees seven years of market exclusivity, giving developers extended periods of monopoly pricing.

This ROI profile has reshaped pharmaceutical R&D strategy. Major pharmaceutical companies that once dismissed rare diseases as commercially unviable now compete aggressively for orphan drug designations. The strategic question has moved from whether to pursue rare diseases to how.

The Orphan Drug Act and Its Incentive Structure

The Orphan Drug Act of 1983 changed the economics of rare disease drug development. Before its passage, fewer than 40 drugs had been developed for rare diseases. In the four decades since, more than 600 orphan drugs have reached patients, with approval rates accelerating each year.

The Act provides four specific incentives: seven years of market exclusivity upon FDA approval (compared to five years for conventional drugs), a 25% tax credit on qualified clinical trial expenses, waiver of FDA application fees (which can exceed $3 million), and dedicated FDA assistance through the Office of Orphan Products Development. The net effect is lower development risk with protected revenue upside.

Orphan drug designation also qualifies sponsors for accelerated regulatory pathways. Orphan drugs are frequently eligible for Fast Track designation, Breakthrough Therapy designation, Priority Review, and Accelerated Approval. These mechanisms can shorten time-to-market by years and permit approval based on surrogate endpoints rather than lengthy survival studies.

Formulation Challenges in Small Populations

The economics that make orphan drugs attractive also make them hard to formulate. Small patient populations mean limited clinical data for dose optimisation. Paediatric predominance requires specialised dosage forms. And the modalities themselves are demanding: gene therapies, enzyme replacement therapies, and antisense oligonucleotides each require formulation expertise that few organisations possess.

Traditional formulation development relies on iterative experimentation across large patient cohorts. With rare diseases, this approach fails. A disease affecting 10,000 patients globally cannot support the large-scale clinical trials that guide conventional formulation optimisation. Formulators must achieve stability, bioavailability, and patient acceptability with a fraction of the data available for common diseases.

Data scarcity compounds modality complexity. Biologics require cold chain management and careful consideration of protein stability; gene therapies require novel delivery vectors and manufacturing processes; mRNA therapeutics need lipid nanoparticle formulations optimised for specific tissue targeting. With limited patient data, trial-and-error formulation becomes prohibitively expensive across all of these modalities.

Stability Problems in Biologics and Gene Therapies

Many of the most promising rare disease therapeutics are biologics or gene therapies, and both are difficult to formulate. Unlike small molecules, these therapeutics are inherently unstable: sensitive to temperature, shear stress, pH changes, and oxidation. Maintaining therapeutic activity from manufacturing through administration requires carefully engineered formulation strategies.

Gene therapies face particular problems. Viral vectors must remain viable through manufacturing, storage, and delivery while maintaining their ability to transduce target cells. The manufacturing complexity of gene therapies has limited their commercial viability. Zolgensma, Novartis's gene therapy for spinal muscular atrophy, carries a list price of $2.125 million per patient, a figure that reflects both its clinical efficacy and the difficulty of manufacturing and formulating AAV9 vectors at therapeutic scale.

Cold chain requirements add another layer of complexity. Many biologics require storage at -20C or below, with some gene therapies requiring ultra-cold storage at -80C. This creates logistical challenges for distribution to rare disease patients, who may be geographically dispersed and far from specialised treatment centres.

AI-Enabled Formulation for Data-Scarce Programmes

Limited patient data is the central obstacle in rare disease formulation, and it is also where AI adds the most value. AI systems can apply transfer learning from related compounds, predict formulation behaviour through in-silico screening, and optimise multi-parameter spaces that would be impossible to explore experimentally with small patient populations.

Transfer learning allows AI models trained on large datasets of common drug formulations to be applied to rare disease compounds. By identifying structural and physicochemical similarities between novel rare disease APIs and well-characterised compounds, AI systems can make informed predictions about formulation behaviour without requiring extensive empirical data specific to each rare disease.

In-silico screening enables rapid evaluation of thousands of potential formulation combinations (excipient selections, processing parameters, stability conditions) before any physical experiments are conducted. For rare disease programmes where API availability is limited and clinical trial slots are scarce, this computational pre-screening shortens development timelines and reduces material consumption.

Key AI Capabilities for Rare Disease Formulation

Strategic Implications

The orphan drug opportunity goes beyond premium pricing in smaller markets. AI-enabled platforms that can work around data scarcity hold a structural advantage in this segment, and that advantage compounds as the number of addressable rare diseases grows.

• Prioritise Paediatric Formulation Expertise: With 70% of rare diseases affecting children, the ability to develop age-appropriate formulations (oral liquids, dispersible tablets, mini-tablets) is a competitive differentiator. AI-driven taste prediction and dissolution modelling are required capabilities for this segment.
• Build Transfer Learning Capabilities: Compound-to-compound knowledge transfer is the primary mechanism for overcoming rare disease data scarcity. Platforms that can apply formulation knowledge across therapeutic areas will outperform those restricted to disease-specific datasets.
• Develop Biologics and Gene Therapy Formulation Expertise: The highest-margin orphan drug opportunities increasingly involve complex biologics. Stability prediction, cold chain optimisation, and manufacturing process development for these modalities demand specialised AI capabilities.
• Integrate Regulatory Intelligence: Orphan drug regulatory pathways, from designation through accelerated approval, shape both development strategy and timelines. AI systems that incorporate regulatory requirements directly into formulation decisions reduce rework and accelerate submissions.

How DeepC Addresses Orphan Drug Formulation

DeepC's studies help rare disease programmes work past the data scarcity that has constrained them, by bringing every public source to bear on each formulation decision.

The formulation study designs candidates for a novel rare disease API from a QTPP, weighing its physicochemistry, excipient precedent in the FDA Inactive Ingredient Database, and the formulations regulators have already approved. The literature review searches clinical trial registries, regulatory documents, and published literature so that formulation decisions draw on all available evidence.

For paediatric formulations, the formulation study designs age-appropriate dosage forms with excipients checked against precedent for the route, and the formulation optimization returns variants against the objectives you set. The freedom-to-operate analysis maps the patent space around existing orphan drugs and sets out design-around strategies, each with its risk rated, for innovators reaching the 95% of rare disease patients who currently have no treatment options.