Opinion: Secrecy: A demon of gene therapy’s past bedevils its future

Twenty-three years ago, the field of gene therapy was bursting with the promise of breakthrough treatments. Then it was almost instantly derailed by the death of an 18-year-old clinical trial volunteer named Jesse Gelsinger after he received a genetically engineered virus that had been developed to treat his rare liver condition.

An FDA investigation revealed that the principal investigator and/or the team running the gene therapy trial at the University of Pennsylvania failed to disclose that before Gelsinger was treated, other patients had experienced alarming side effects and that monkeys administered the same engineered virus had died. The incident, coupled with dangerous outcomes involving other gene therapy treatments, had a chilling effect on the field and investors backed away.

Today, new approaches to gene therapy that include advances driven by CRISPR gene editing tools are raising hopes of a gene therapy revival. There are potential breakthroughs in the pipeline, including treatments for different types of cancer and sickle cell disease.


I’m concerned that gene therapy 2.0 is at risk of making the same mistakes that plagued the 1.0 version. Most notably, exciting work to translate gene therapy advances into safe, effective, and commercially viable treatments are at risk of being undermined by a reluctance to share data.

I’m not saying that the field is on the verge of something analogous to the tragedy experienced in 1999. But the road to such extremes can be paved with a series of lesser nondisclosures that inhibit the free flow of scientific data essential for assessing potential risks long before treatments are given to people.


I saw an example in May at the annual meeting of the American Society of Gene and Cell Therapy. A presenter was discussing a new approach for using gene therapy to treat a rare genetic disorder called Leber congenital amaurosis that causes blindness in children. The presenter discussed experiments in mice that involved disabling or deleting certain segments of a gene linked to the disorder, and this treatment appeared to at least slow the process of vision loss.

Assessing the safety and effectiveness of this potentially exciting advance requires knowing which regions of the gene were being modified. But when someone in the audience asked for more details, the presenter indicated that the information was proprietary and he would “not disclose that.” The exasperated questioner pointed out that the presenter’s approach was “all about deleting [a segment of a gene], but you don’t say what you are deleting.” The response? Silence.

Tinkering with a particular gene’s function, even when the goal is to stop it from doing harmful things, can be risky. Knowing which region of a gene is being altered is essential for determining if solving one problem might create an even bigger one. For example, I and other scientists are interested in the potential of using CRISPR gene editing tools to disable a gene called nuclear factor erythroid 2-related factor 2 (NRF2), which produces a protein that protects a certain type of lung cancer tumor from the effects of chemotherapy or radiation. But that same gene confers a range of health benefits, such as withstanding toxic insults like nicotine or radiation or heat stress, and a poorly targeted edit might do more harm than good.

I am also regularly seeing a failure to disclose important details, like studies that use several strains of mice in a gene therapy experiment without clarifying which strain was linked to a particular result. That can make it difficult, or even impossible, to conduct a fundamental exercise in science: reproduce the results of other investigators. Reproducibility is a problem across all areas of science. But the stakes are higher in the fragile world of gene therapy, where investigators need to be doing all they can to protect patient safety.

My initial thoughts about the presenter who was loquacious in describing his success with gene therapy for Leber congenital amaurosis but quiet about revealing his methods was simply, “If you aren’t willing to share your full information, don’t present at scientific conferences.”

I am not against protecting intellectual property or patenting biotechnology advances. Properly used, these strategies help attract investment that accelerates the search for transformative treatments. But researchers shouldn’t come to a place where they benefit from everyone else freely sharing their findings but then refuse to reciprocate.

My broader message is that the field of gene therapy was once severely damaged by not sharing data that may have been able to prevent a young man from losing his life. Transparency in gene therapy research — which can be accomplished without compromising commercial prospects — is vital to success. One high-profile failure would badly hurt the revival of gene therapy; two would send it back into hibernation.

When everyone embraces transparency, all of our projects are likely to advance faster toward safe and effective treatments and everyone wins: scientists, investors and, most importantly, the people who could benefit from gene therapy.

Eric B. Kmiec is the executive director and chief scientific officer of the ChristianaCare Gene Editing Institute in Newark, Del.

Source: STAT