Gene editing moves from rare disease rescue to prevention of common cardiovascular disease
A single genetic intervention that lowers LDL cholesterol for years, or even decades, has moved from speculation into serious clinical discussion. The appeal is obvious: one treatment could replace a lifetime of pills for patients who now depend on daily medication to control high cholesterol and lower cardiovascular risk. Yet the same permanence that makes genomic therapy attractive also makes it difficult to reverse if the biology behaves badly later. In a perspective in JAMA Internal Medicine, Eric D. Smith, M.D., and Venkatesh L. Murthy, M.D., Ph.D., argue that medicine is entering a phase in which the question is no longer whether gene editing can alter disease biology, but whether it should be deployed for diseases that already have workable treatments.
The shift matters because the target population is changing. Gene therapy already has a track record in serious inherited disorders, including spinal muscular atrophy and sickle cell disease, where the balance between risk and benefit is shaped by severe illness and limited options. Common cardiovascular conditions present a different calculation. A therapy designed to permanently modify a gene involved in lipid metabolism may be scientifically elegant, but it must still compete with medications that can be titrated, paused, or discontinued if the patient develops side effects.
The difference is not philosophical. It is clinical.
Permanent genetic change carries a different safety burden than standard lipid therapy
Genomic therapies can work by supplying a healthy gene, silencing harmful activity, or changing DNA directly. Some are built to function after a single administration, which is why they are often described as “one-and-done” treatments. Even so, one dose does not guarantee a lifetime of benefit. Modified viral vectors can trigger immune responses, which may make repeat dosing difficult or impossible if the effect wanes.
That limitation becomes especially important when the therapy is aimed at diseases affecting millions of people. A rare adverse event may be tolerable when a treatment is reserved for a life-threatening inherited condition with few alternatives. The calculus changes when the same platform is proposed for heart disease, obesity, diabetes, or high cholesterol. In a broad population, a complication that appears infrequently can still affect large absolute numbers of patients. The higher the reach, the tighter the safety threshold must be.
Smith and Murthy frame that challenge around somatic editing, not germline modification. Current clinical strategies are aimed at somatic cells, meaning the changes are not intended to be passed to future generations. Even with that safeguard, the genetic alteration is still meant to be durable, and durability is exactly what complicates risk management when the target process is finely balanced.
Five questions now shape whether genomic therapy fits a common disease
The framework laid out in the perspective centers on five practical questions. How severe is the disease? Are there existing treatments that already work well? Is the target tissue suitable for a lasting intervention? Are delivery-related harms acceptable in the context of the disease being treated? Can long-term safety and effectiveness be monitored reliably over many years?
Those questions create a gatekeeping function that conventional prescribing rarely requires at this level. A statin can be stopped if liver enzymes rise or muscle symptoms become unacceptable. A permanent gene-editing intervention does not offer the same easy off-ramp. If the biology drifts too far in one direction, there may be no simple way to recalibrate the result.
Some disorders may eventually prove suitable for this approach because the therapeutic target is stable and the disease burden is high enough to justify lasting risk. Others may not. Conditions that require flexible dose adjustment, or where the gap between benefit and harm is narrow, are poor candidates for an irreversible intervention. Biological control systems are rarely linear, and permanent manipulation may overshoot in ways that become obvious only after long follow-up.
Long-term follow-up may determine whether the strategy is usable at scale
The hardest problem is not the engineering of the edit itself. It is the time horizon.
Clinical trials usually include relatively small numbers of participants and limited observation periods. That design can identify immediate toxicities, but it is poorly suited to detecting complications that appear years later or only after widespread use. A signal that is invisible in a modest cohort may become clinically important only when a therapy reaches much larger populations.
For that reason, the perspective argues that long-term monitoring cannot be treated as an administrative afterthought. Registries and other follow-up systems will need to be standardized, independent, and capable of tracking both efficacy and delayed harm over extended periods. Without that infrastructure, clinicians may not know whether a gene-based intervention remains protective, loses effect, or triggers unexpected pathology years after the first dose.
That requirement will become more demanding as the target population expands. The public-health implications are very different when a therapy is used for a rare inherited disorder versus when it is offered for a common cardiometabolic risk factor. A monitoring system that works for a few thousand patients may fail when the number grows by orders of magnitude.
Cost and access could shape who benefits from the technology
Price is another barrier that cannot be separated from safety. Some existing gene therapies carry list prices above $1 million per patient, with several costing even more. For catastrophic rare diseases, those prices may be easier to defend because the alternative may be progressive disability or death. For widespread conditions, the same pricing structure would place enormous strain on healthcare budgets and could deepen inequities in access.
The access problem is not abstract. If a therapy can only be delivered to a small fraction of eligible patients, its clinical promise will be filtered through income, insurance status, and health-system capacity. A highly effective genomic treatment that reaches only a privileged subset of the population would not function as a population-health intervention. It would function as a scarce, expensive specialty product.
Early cholesterol editing studies point to the next frontier, not the finish line
Researchers are already running early studies aimed at producing durable LDL reduction through gene editing. LDL cholesterol drives fatty plaque buildup inside arteries, which is why sustained lowering remains one of the most important strategies in cardiovascular prevention. If a single intervention can maintain low LDL safely over time, the effect could reshape prevention medicine.
But the biology that makes the idea attractive also explains the caution. LDL reduction is a long game, and so is risk. A therapy that permanently changes cholesterol handling must prove not only that it lowers a biomarker, but that it does so without generating late toxicities, immune complications, or metabolic tradeoffs that emerge outside the initial trial window.
Smith and Murthy do not argue for slowing genetic medicine in general. They argue for restraint when the target disease is common, the current therapies are adjustable, and the consequences of an error could last for years. Before gene editing becomes routine for widespread cardiovascular prevention, it will need durable efficacy, acceptable long-term safety, workable monitoring, and a cost structure that does not lock out the very patients it is meant to help.