How Individualized Gene and RNA Therapies Could Transform Treatment for Ultrarare Diseases

Photo by Thirdman on Pexels.

How Individualized Gene and RNA Therapies Could Transform Treatment for Ultrarare Diseases

For people with ultrarare genetic diseases, the central obstacle to treatment is often not a lack of biological insight. Researchers may know the affected gene, understand how a particular variant disrupts its function, and even have a plausible way to correct the defect. The problem is that conventional drug development was not built for a population of one person, one family, or a few patients worldwide.

Individualized gene and RNA therapies offer a different model. Instead of finding a medicine and then identifying a large group of eligible patients, researchers begin with the molecular defect in a specific patient. They design a treatment around that defect, using a broader therapeutic platform whose chemistry, delivery method, or manufacturing process may already be understood.

This approach has produced compelling experimental cases, especially with antisense oligonucleotides. It is also beginning to influence gene editing, RNA editing, and other forms of precision medicine. Yet a treatment that can be designed is not necessarily one that can be delivered safely, evaluated rigorously, or financed sustainably. The promise is real, but so are the constraints.

What Makes a Therapy Individualized

The term individualized can describe several levels of personalization.

Some genetic therapies are designed for a disease but work across many variants. A gene replacement therapy, for example, may deliver a functional copy of a gene regardless of the patient's exact pathogenic mutation. Other treatments apply only to a molecular subgroup, such as patients whose variants cause the same splicing error.

At the most personalized end of the spectrum are treatments created for one person or a very small number of people. These are sometimes called N of 1 therapies. The active sequence may be selected specifically for the patient's variant, even if the underlying chemical platform and delivery method have been used before.

This distinction matters. A drug developed for several thousand patients can be studied through conventional trials and supported by a recognizable commercial market. A medicine for one patient requires different approaches to evidence, manufacturing, oversight, and cost.

Individualized treatment also depends on unusually strong genetic diagnosis. Finding a rare variant is not enough. Researchers must establish that it causes the disease, determine how it changes RNA or protein function, and show that the proposed intervention addresses that mechanism. Variant interpretation remains difficult, particularly for changes outside protein coding regions or for disorders involving several genes. The widely used American College of Medical Genetics and Genomics framework provides standards for classifying variants, but many findings remain uncertain even after detailed analysis.

Why RNA Is a Practical Starting Point

RNA therapies are especially attractive for individualized treatment because their sequence can often be changed without redesigning the entire drug platform.

Antisense oligonucleotides, commonly called ASOs, are short synthetic strands of nucleic acid that bind to a selected RNA sequence. Depending on their design, they can alter RNA splicing, promote degradation of a harmful transcript, block translation, or modify how regulatory proteins interact with RNA.

This creates several possible strategies. An ASO might cause a cell to skip an exon containing a damaging variant, restore inclusion of an exon that has been incorrectly removed, or reduce production of a toxic protein. The chemistry, formulation, and route of administration may remain broadly similar across different ASOs, while the nucleotide sequence changes to match the target.

The best known example of a highly individualized ASO is milasen, developed for a child with a rare form of neuronal ceroid lipofuscinosis. Researchers discovered that a unique DNA insertion disrupted RNA splicing. They designed an ASO to correct the splicing defect, conducted laboratory and animal testing, and obtained regulatory authorization for clinical use. The treatment and early observations were described in a 2019 report in The New England Journal of Medicine.

Milasen demonstrated that a patient-specific drug could move from genomic diagnosis to administration within a clinically meaningful period. The report found that treatment was associated with improvements in measured seizure characteristics and did not produce serious adverse events during the reported period. It did not establish that the therapy reversed the underlying neurodegeneration or prove efficacy through a controlled trial. Its importance lies primarily in showing that the development model is technically and operationally possible.

Other RNA approaches may eventually support similar customization. Small interfering RNA can selectively reduce expression of a target transcript. Messenger RNA can provide temporary instructions for producing a needed protein. Experimental RNA editing systems aim to change individual RNA letters without permanently modifying DNA. These methods differ greatly in maturity, delivery requirements, and risk. ASOs currently offer the clearest precedent for therapies whose sequence can be adapted while much of the surrounding platform remains constant.

How Gene Therapy Expands the Possibilities

Gene based interventions could address defects that are difficult to treat at the RNA level. The main categories include gene addition, gene editing, and regulation of gene activity.

Gene addition delivers a working gene, often through a viral vector. It is not usually individualized to a single mutation because one construct may help many patients who lack sufficient function of the same gene. Even so, gene addition can be relevant to ultrarare conditions if a shared vector and manufacturing system can be adapted to different genes.

Gene editing offers more precise forms of correction. CRISPR systems can cut DNA at a selected site, while base editors and prime editors can make certain changes without relying on a conventional double strand break. In principle, an editor could be designed for a particular pathogenic variant.

That precision does not make the process simple. A customized editor may require a new guide RNA, a new donor template, or modifications to the editing enzyme. Researchers must assess unintended editing elsewhere in the genome, unwanted changes near the target, variable editing among cells, and the possibility of immune responses to the delivery system or editing components.

Delivery remains one of the largest barriers. Editing blood stem cells outside the body is different from reaching neurons, muscle cells, or cells distributed across several organs. A successful molecular correction has limited value if too little of the relevant tissue receives it.

Unlike most RNA treatments, some gene editing interventions may create permanent changes. That could reduce the need for repeated dosing, but it also raises the consequences of an error. For a treatment intended for one person, uncertainty cannot be resolved by studying large groups before exposure.

A Platform Rather Than a Completely New Drug

The long term viability of individualized medicine may depend on treating each product as a variation within a validated platform rather than as an unrelated drug.

A platform can include a common therapeutic chemistry, manufacturing process, delivery vehicle, quality control system, and framework for safety testing. The patient-specific element might be limited to an ASO sequence or an editing guide. Reusing well characterized components could reduce duplication while preserving scrutiny of risks that arise from the new sequence.

This does not mean that changing a sequence is trivial. A new ASO can bind unintended RNA targets, stimulate immune responses, accumulate differently in tissues, or behave unpredictably because of its base composition. A new editing guide can alter genomic sites that were not affected by a previous guide. Manufacturing consistency must be established for every administered product.

Platform evidence could nevertheless help regulators and researchers decide which tests must be repeated and which findings can reasonably be carried over from earlier products. The United States Food and Drug Administration has issued guidance addressing investigational submissions for individualized ASO products, including administrative and procedural recommendations. The existence of such guidance reflects growing recognition that conventional development pathways do not fully fit medicines intended for extremely small populations.

Regulatory flexibility does not imply a lower safety standard. It reflects a need to make decisions with different kinds of evidence, limited patient numbers, and diseases that may progress rapidly while development is underway.

Measuring Benefit in One Patient

Traditional randomized trials estimate average treatment effects across groups. An individualized therapy may have only one eligible recipient, making conventional statistical comparisons impossible.

Evidence must therefore be assembled from several sources. Laboratory studies can show that the treatment changes RNA processing, protein expression, or cellular function. Animal studies may provide toxicology and distribution data, although an animal model may not reproduce the patient's exact variant or clinical condition. Biomarkers can indicate whether the therapy engages its intended target in the body.

Clinical interpretation often depends on detailed natural history. Researchers need to understand how the disease would probably have progressed without treatment. Repeated measurements before therapy can sometimes establish a patient's individual trajectory, although progression may not be linear. Comparisons with affected siblings or historical cohorts can be informative but are vulnerable to differences in age, care, genetic background, and disease severity.

Outcome selection is also critical. A molecular correction may be measurable long before a change in daily function becomes apparent. Conversely, a treatment may stabilize function without producing obvious improvement. In a rapidly progressive disease, stabilization could be clinically meaningful, but demonstrating that it resulted from treatment rather than natural variation can be difficult.

For many neurological disorders, timing further complicates evaluation. Correcting the causal defect may not restore cells or developmental processes that have already been lost. A therapy can be biologically active yet produce limited clinical recovery because it was delivered too late. This is one reason rapid diagnosis and newborn screening may become increasingly important as targeted therapies emerge.

Safety Under Conditions of Uncertainty

Individualized therapies compress difficult decisions into a short period. Patients may have severe, progressive illnesses with no established treatment. At the same time, clinicians and regulators may have little direct evidence about the customized product.

Preclinical testing can identify some hazards, but it cannot eliminate uncertainty. For ASOs, relevant concerns include unintended binding, inflammation, organ toxicity, and complications related to repeated administration. For gene editing, concerns include unintended genomic changes, variable editing efficiency, immune reactions, and persistent effects that may not be reversible.

The balance of risk depends partly on the disease. Greater uncertainty may be considered acceptable for a rapidly fatal disorder than for a condition that is disabling but stable. That judgment should not be reduced to a technical calculation. It involves the patient's circumstances, expected disease course, treatment burden, available alternatives, and the quality of the supporting evidence.

Long term follow-up is essential, particularly after permanent or durable interventions. A single treatment can generate information that shapes the safety assessment of an entire platform. Consistent data collection and transparent reporting are therefore both scientific and ethical obligations.

The Economic Problem

Conventional pharmaceutical development distributes research and manufacturing costs across a market. An individualized medicine may have no market in the usual sense.

The direct costs can include genomic analysis, functional experiments, sequence design, toxicology, regulatory work, manufacturing, hospital procedures, and long term monitoring. Repeating this process independently for each patient would be difficult to sustain, even if the technical steps became faster.

A platform model could lower some costs through standardized production, shared testing methods, and reusable regulatory documentation. Nonprofit organizations, academic centers, public agencies, philanthropic funders, and biotechnology companies may each play a role. Centralized manufacturing networks could also prevent every hospital from having to build its own specialized infrastructure.

Financing remains an unresolved policy question. If access depends primarily on a family's ability to raise money or attract institutional attention, scientifically eligible patients will not have equal opportunities. Geographic disparities may be especially severe because advanced genomic diagnosis and experimental therapeutics are concentrated in a limited number of centers.

The data generated from one patient may benefit future patients and improve the wider platform. That public value provides an argument for shared investment, but it does not by itself determine who should pay or how resources should be allocated.

Ethical Questions Beyond Consent

Informed consent is unusually complex when a therapy is being designed for a specific patient. Families may understand that a treatment is experimental while still viewing it as their only realistic source of hope. Researchers must communicate the difference between correcting a molecular defect in the laboratory and producing meaningful clinical benefit.

Children account for many patients with severe genetic diseases. Parents may need to authorize an irreversible intervention before long term outcomes are known. Delay can also carry serious consequences if the disease is progressive. Ethical review must therefore consider both the risks of treatment and the risks of waiting.

Fair patient selection is another concern. Researchers may be able to develop only a small number of customized therapies. Decisions about which cases to pursue can be influenced by biological feasibility, disease severity, available funding, institutional expertise, and public visibility. Clear selection criteria can reduce, but not eliminate, these tensions.

Data sharing also requires care. Detailed genomic and clinical information is valuable for understanding rare diseases, yet patients with unique variants may be identifiable even when names are removed. Governance systems must support scientific learning while respecting privacy and family preferences.

What Could Change Clinical Care

If individualized therapies become more practical, their greatest impact may extend beyond a series of isolated rescue attempts. They could reshape the relationship between diagnosis, research, and treatment.

Genomic sequencing would need to connect more directly with functional laboratories capable of testing whether a variant disrupts splicing, protein production, or gene regulation. Manufacturing facilities would need processes suited to small batches. Regulators would need to evaluate product-specific risks while drawing appropriately on platform experience. Clinical teams would need standardized methods for collecting pretreatment histories and long term outcomes.

Earlier diagnosis would become more valuable because many genetic interventions are likely to work best before irreversible damage occurs. At the same time, broader sequencing would identify more variants whose significance is uncertain. Better diagnosis would therefore have to include stronger functional evidence, not simply more genetic data.

The most plausible near term future is not a unique therapy for every genetic variant. Many disorders will remain inaccessible because the relevant tissue cannot be reached, the disease mechanism is unclear, or damage occurs too early. Other conditions will be better served by treatments that address a shared pathway rather than one mutation.

Individualized gene and RNA therapies nonetheless establish a new therapeutic principle. The rarity of a variant does not necessarily make it biologically untreatable. With reusable platforms, careful evidence standards, equitable funding, and transparent oversight, some patients who fall outside conventional drug development may gain a realistic path from molecular diagnosis to targeted intervention.