Precision, potential and perspective

Personalised medicine is often presented as the future of healthcare: treatments matched to a person’s genes, tumour profile, immune system, environment and medical history rather than prescribed on a broad one-size-fits-all basis. In its strongest form, this approach can improve diagnosis, guide drug choice and reveal disease mechanisms that were previously hidden. But the field is not a single seamless revolution. It is a patchwork of different technologies, each with its own evidence base, risks, costs and social consequences.

Gene therapy sits inside that broader picture. It usually refers to adding, replacing, silencing or editing genetic material in cells in order to alter how a disease behaves. Some therapies act directly inside the body, while others modify cells outside the body before they are returned to the patient. Alongside this are pharmacogenomics, biomarker-guided oncology, cell therapies, genome sequencing, polygenic risk models and newer RNA-based medicines. These areas overlap, but they should not be treated as though they are all equally proven or equally mature.

What personalised medicine actually means

At its best, personalised medicine helps doctors make more precise decisions. A patient with cancer may receive a drug because a tumour contains a specific mutation. A patient taking antidepressants, anticoagulants or pain medication may benefit from pharmacogenomic testing that suggests whether they are likely to metabolise a drug unusually fast or slowly. A person with a rare inherited disorder may receive a more accurate diagnosis because genome sequencing reveals the underlying mutation.

This is a real shift. It can reduce trial-and-error prescribing, identify people who need closer monitoring and uncover biological differences that matter clinically. Yet personalisation does not mean certainty. A genomic marker rarely speaks on its own. Clinical judgement, symptoms, family history, co-existing illnesses and the patient’s wider life still shape whether a treatment is appropriate.

Different technologies, different levels of evidence

It is important to separate technologies that are often blended together in public discussion. Gene replacement therapy, genome editing, CAR-T cell therapy, pharmacogenomics, polygenic risk scores and mRNA medicines are related only in the broad sense that they draw on molecular biology. Their mechanisms and risk profiles differ substantially.

Ordinary mRNA vaccines, for example, are not the same as gene therapy. They do not aim to rewrite the human genome. Therapeutic mRNA platforms, however, are part of a wider RNA-medicine landscape that may enable more individualised treatments, such as personalised cancer vaccines. Similarly, a targeted cancer drug selected through biomarker testing is not the same as a one-time viral-vector gene therapy for a rare inherited disease. Responsible discussion starts by keeping these distinctions clear.

Where gene and cell therapies have delivered real benefits

There have been genuine breakthroughs. Some inherited retinal diseases, spinal muscular atrophy, severe blood disorders and certain cancers have seen outcomes that would have been difficult to imagine a generation ago. Luxturna helped establish the viability of treating some forms of inherited blindness. CAR-T therapies have produced striking responses in some blood cancers. In rare diseases, more accurate genetic diagnosis can also reduce years of uncertainty for patients and families.

These advances matter because they show that molecular medicine is not empty hype. In some cases it has already changed survival prospects, functional outcomes or quality of life. Personalised oncology, in particular, has become a practical part of modern care in many centres, even if the results are uneven across cancer types and health systems.

Why the language of “one-time cures” can mislead

The promotional side of this field often runs ahead of what the evidence can honestly support. Some treatments are described as if they permanently solve a disease in a single intervention. In reality, biology is rarely so neat. A therapy may transform prognosis without fully reversing earlier damage. Benefits may be substantial but incomplete. Long-term durability can remain uncertain for years, and immune responses may complicate retreatment.

That is especially important in severe childhood disorders, where hope is understandably intense. A therapy can be life-changing without being a universal or permanent cure. Serious adverse effects, including liver complications, immune reactions or off-target consequences, remain central to the risk-benefit calculation. One administration does not mean one-time risk.

mRNA medicines after the pandemic

Public attitudes to mRNA-based technologies changed sharply during and after the COVID-19 pandemic. Some of the criticism has been driven by misinformation, but some concerns are legitimate subjects for pharmacovigilance and public debate. The key point is that “mRNA” is not one single settled verdict. Each product must be judged on its indication, dose, formulation, delivery system and patient population.

The broader significance of mRNA technology is that it may speed up the design of highly specific therapies, including cancer treatments based on a patient’s own tumour profile. That potential is meaningful. But potential should not be confused with blanket validation. Safety monitoring, platform-specific side effects, public trust, and the distinction between therapeutic use in very ill patients and preventive use in healthy populations all matter.

Polygenic prediction and its limits

Personalised medicine is not only about treatment. It also includes prediction. Polygenic risk scores try to estimate disease susceptibility by combining many small genetic signals. In theory, these tools could help identify people who might benefit from earlier screening or stronger prevention strategies. In practice, they remain probabilistic and uneven.

A high polygenic score does not mean a person will definitely become ill, and a low score does not guarantee protection. Predictive performance can weaken in populations that were underrepresented in the data used to build the score. Environment, diet, stress, exposure, upbringing and chance still matter. There is also a social danger: genetic information can generate false reassurance for some people and unnecessary anxiety for others.

Cost, patents and access

If this field is discussed only in scientific terms, the picture becomes misleading. Many advanced therapies are extraordinarily expensive. Prices in the high six figures or above are not unusual, especially for rare-disease treatments. Companies justify this by citing development costs, small patient populations and the possibility of long-term savings. Those arguments may sometimes be partly valid, but they do not erase the financial pressure placed on health systems, insurers and families.

This creates a structural tension. The therapies most celebrated as medical breakthroughs may be available only to a fraction of the people who could benefit, especially across poorer countries or underfunded healthcare systems. Access may depend not just on need, but on geography, insurance status, trial enrolment and political bargaining. The result is that personalised medicine can become both more precise and more unequal at the same time.

Data, privacy and discrimination

Personalisation depends on data: genomes, biomarkers, scans, family histories, health records and increasingly behavioural or wearable data. This makes privacy and governance central rather than peripheral. Genetic data is unusually sensitive because it is durable, partly predictive and implicates relatives as well as the individual being tested.

Even where anti-discrimination law exists, concerns remain about data sharing, secondary research use, cybersecurity, corporate partnerships and the long-term storage of genomic information. Patients may gain useful insights from testing while still having limited practical control over how their data travels through medical, commercial and research networks. Trust is therefore a real part of the technology, not merely a communications problem.

Ethics, childhood intervention and enhancement

Some of the hardest questions emerge when interventions are expensive, irreversible or applied to children who cannot give fully informed consent. Parents, clinicians and regulators may have to make decisions under intense emotional pressure, with incomplete long-term evidence and powerful commercial narratives in the background. That does not make the treatments wrong, but it does mean ethical caution is justified.

The same applies to the boundary between treatment and enhancement. Using molecular medicine to treat severe disease is one thing; using similar tools to optimise traits, performance or perceived normality is another. Debates about disability, social pressure, embryo editing and what counts as an acceptable intervention are not side issues. They are part of the future this field is creating.

A cautious but credible future

The strongest case for personalised medicine is not that it will solve everything, but that it can solve some things far better than older medical models could. For rare diseases, some cancers and certain drug-response problems, this is already visible. The weakest version of the story is the one that turns every molecular advance into a sweeping promise of effortless precision and inevitable cure.

A credible future would combine scientific ambition with stronger evidence, long-term monitoring, realistic public communication, fairer access models and clear limits on what is known. Gene therapy, RNA medicine and biomarker-guided care may continue to reshape healthcare, but their value depends on more than technical possibility. It depends on whether society can manage uncertainty, resist hype, protect patients and distribute benefits without deepening existing inequalities.

Gene Therapy & Personalised Medicine