Guest Column | May 14, 2024

Gene Therapy: How Should We Approach Benefit And Risk?

By Jonathan (John) Finn, Ph.D., chief science officer, Tome Biosciences


Once relegated to the realm of science fiction, gene and cell therapy now has the real potential to transform the lives of many patients living with disease. As with any medical intervention, however, there has to be a careful evaluation of the benefit-risk ratio before deploying these novel therapies.

Understanding the benefit-risk ratio is a complex exercise, and it often has to be done in the absence of having perfect information. This is especially true for new modalities such as gene editing, where there hasn’t been sufficient time to understand the long-term risks associated with these transformative therapies. In the absence of a crystal ball, how should we be looking at benefit-risk with respect to gene and cell therapy applications?

Using Real-World Examples To Map Our Current Understanding

Thankfully, we are not working in a vacuum and have real-world clinical examples we can draw on for guidance. For instance, in sick patients with terminal cancer, there is a clear precedent of taking more risk. Once efficacy and safety with a novel modality have been established in this population, decisions can be made on whether the benefit-risk ratio can justify moving either to earlier lines of treatment or into other indications.

The autologous CD19-CART story is a great example of this. The first patients enrolled in the CD19-CART trials had failed multiple lines of therapy and were essentially out of options. The remarkable response rate seen in these early trials opened the door to move into more patients and potentially move CD19-CART therapy to earlier lines of therapy for B cell lymphomas.

It is important to note that the CD19-CART therapy was not without its own safety risks. Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) are well-established side effects associated with these T cell therapies; these side effects, however, can often be managed in patients and the benefit of these therapies outweighs the risks for these sick patients.

Another safety risk (secondary T cell cancers) has now emerged, resulting in a black box warning on all autologous lentivirus-based CAR-T therapies. It is not expected to have much impact on the approved use cases (it still has a positive benefit-risk ratio), but these safety issues will likely have an impact on the widespread adoption of these therapeutics into other areas, such as autoimmunity, where the benefit-risk ratio can be very different from oncology.

There has been a lot of excitement recently over the impressive results using autologous CD19-CAR-T for refractory systemic lupus erythematosus. This approach capitalizes on a known side effect of CD19 CART (B cell depletion) to eliminate auto-reactive B cells in patients with severe disease. Given the need for lymphodepletion using fludarabine and cyclophosphamide, as well as the risk of secondary T cell cancers, there is a question about how widespread this approach will be used for autoimmunity in general, where the majority of patients are fairly healthy. For this population, next-generation therapies that have a cleaner safety profile will likely be needed for widespread adoption.

Currently there are multiple approaches in development that can potentially address one or many of the safety concerns associated with CD19 CART (NK cells vs. T cells, targeted CAR integration vs. random integration, incorporation of safety switch, non-myeloablative conditioning, transient CAR expression, etc.) and it will be very interesting to see how these next-generation therapies are able to bend the risk-to-benefit curve in favor of bringing these therapies to more patients in need.

With respect to in vivo gene therapy/editing, a similar approach can be taken to help guide how to best apply potentially curative gene therapies to patients in need.

For the past decades, the concept behind gene therapy was pretty simple: If a patient is lacking a gene, we would try and deliver a healthy copy of the missing gene into the right cell. Viruses were harnessed as gene delivery systems, and healthy copies of genes were delivered to target cells. This led to the first wave of clinical trials and approved drugs.

As expected, these experimental therapies were piloted in patients that had severe unmet needs, including lipoprotein lipase deficiency (Glybera), childhood blindness (Luxterna), hemophilia (Hemgenix, Beqvez, Roctavian), and spinal muscular atrophy (SMA).

While many of these treatments have had a transformative effect on patients, safety issues either related to the delivery system (e.g., thrombotic microangiopathy [TMA] and liver enzyme elevations associated with high vector doses) or the specific indication (e.g., thrombosis potentially related to overexpression of clotting factors) have been observed. For the most part these safety issues are transient and well-managed and have not prevented the use of these life-changing therapies.

Should Our Approach To Gene Therapy Benefit-Risk Be “Fit-For-Purpose”?

At the end of the day, the benefit-to-risk ratio acceptable to physicians and patients will depend on the specific indication, the magnitude of benefit, and the severity of the risk. For instance, transient and reversible liver damage following treatment may be acceptable for diseases such as hemophilia, OTC deficiency, SMA, or phenylketonuria (PKU), where the number needed to treat (NNT) is likely close to one (that is, a single treatment could lead to a lifetime cure for that patient). The same side effects, however, may not be acceptable for a more common disease, such as ASCVD, where the NNT may be significantly higher. Ultimately, some transient side effects might be acceptable and tolerable as a trade-off for a lifetime cure, which is a discussion that should be had between the doctor and the patient.

As discussed at the outset, considering gene or cell therapy for a particular condition is a complicated exercise in understanding the benefit-risk ratio associated with treatment – especially now, when the science is still emerging and long-term risks for these therapies may not be sufficiently documented to the satisfaction of all concerned. As more gene editing safety data is compiled, the risk-benefit calculation is likely to change, which will open the door to more – and more important – medical applications for the technology.

About The Author:

John Finn is chief science officer of Tome Biosciences. He can be reached at He has more than 20 years of gene therapy experience with a focus on genome editing and delivery technologies. He was most recently VP of discovery research at Codiak Biosciences, where he led the development of a new class of therapeutics based on engineered exosomes. Previously, Finn was executive director of platform biology and liver discovery at Intellia Therapeutics, and was director of research at Arthrogen. He received his Ph.D. in biochemistry and molecular biology from the University of British Columbia and BS in molecular biology and genetics from the University of Guelph. He completed post-doctoral programs at McMaster University and the Children’s Hospital of Philadelphia.