IMPACT & Hype: Pt 3 of The Rise and Fall of “Stem Cells for Autism” at Duke University
This is the third in a series about Duke University’s experiments on autistic children, in which young children, including those with IQs below 70, were injected with stem cells to see if it would stop them from being autistic. This series looks at the 9 years from first FDA approval in 2014, to the addition of Expanded Access (EAP) “pay to play” clinical trials, to pushback from experts and finally the cancellation of the EAP in 2023.
Duke’s IMPACT trials: a brief history
In 2014, a team at Duke University, led by Dr. Joanna Kurtzberg, launched IMPACT, a series of clinical trials on stem cells as an autism treatment, recruiting parents to enroll their young children as the subjects of the stem cell experiments.
In the beginning, the Duke experiments had been solely funded by the Marcus Center for Cellular Cures at Duke, endowed by Bernie Marcus.[1] It reflected Marcus’s interest in stem cells, an interest that’s unsurprising given he had sought treatment at (and endorsed) the Stem Cell Institute of Panama, a direct-to-consumer clinic that is not regulated by the FDA. Marcus also had an interest in autism; in fact, it was he who had given $25 million in seed money to Autism Speaks when it was first founded.[2] An avid philanthropist, Marcus has vowed to leave the bulk of his fortune to charity.
With $40 million in sole-source funding from Marcus, the first of the IMPACT trials began. The results looked weak, with no control group, making it impossible to know whether any child’s progress was due to a natural developmental arc or due to the stem cells. Researchers regrouped and got a few more million in funds from Marcus for a slightly different Phase II study. Meanwhile, the media hype machine got going.
The Duke’s team’s first study of 25 children in 2014, which failed to show any evidence of benefit, was hyped in a CNN feature co-authored by Dr Sanjay Gupta, which described the unproven treatments as potentially “promising”. It featured a personal story about Gracie, a young autistic girl and her family. In the vague style that seems a hallmark of the project’s media coverage, Gupta’s team stated that the family may or may not have benefitted from the clinical trial, but Gracie’s mother told CNN: “She got better, and we’re just thankful for that – whether it be the stem cells or not. We’re just thankful for what changes have happened.”
As noted in my last post, Kurtzberg had told CNN: “We’ll be extraordinarily encouraged if the second trial shows that the cells benefit children when the placebo does not. We will consider that a breakthrough.” While CNN covered the initial hopes of the Duke project participants, the news station didn’t follow up on the results of the second trials, which ended up to be disappointing. Families have continued to share the clip in their online support groups, with some expressing their belief that Dr. Kurtzberg knew the treatments did work and that the trials were more of a formality than an actual test. As bioethicist Jeremy Snyder told me:
“There is no correction and these stories get shared online; they get shared in the crowdfunding campaigns; they get shared in patient groups. Even if more recent data doesn’t support that kind of story, the article is still there. You can look at it today and I think that’s hugely problematic.”
As further trials were launched by Duke, various revenue streams were also being developed. Most notably, in 2020 Cryo-Cell brokered an exclusive patent option agreement with Duke, with an option to potentially get licensing to manufacture and sell autism-related stem cell products based on patents held by Kurtzberg. There seemed to be no end to the monetizing potential and, if they could prove autistic children could benefit from stem cell infusions, it had the potential to be a big win.
No evidence of benefit
There was, however, a big problem. The findings from Dukes Phase II randomized controlled trial , released in May of 2020, showed no convincing evidence of benefit, nor even a mechanism that would explain why the use of stem cells as an autism treatment could ever work, in any future trials. (In the end, none of the IMPACT trials showed meaningful evidence of benefit). The Duke team’s report, authored by co-lead Dr. Geraldine Dawson, documented a lack of evidence for primary or secondary outcomes, concluding: “Overall, a single infusion of [cord blood] was not associated with improved socialization skills or reduced autism symptoms.”
This result may not have surprised Arnold Kriegstein, a neural stem cell expert at the University of California, San Francisco. Kriegstein had been one of the first skeptics of the Duke IMPACT trials, telling Spectrum in 2017 that the Duke study was a “Hail Mary pass” and that there was no plausible mechanism for stem cells to treat autism. As he put it:
“None of their explanations hold water on why it would have any therapeutic value.”
Other studies, such as one conducted in 2018 by scientists at the University of Sacramento, also found no benefits for autistic children given stem cells. The National Institutes of Health clinical trials registry lists 11 other autism-related stem cell trials, but three of these concluded without registering results, three have a status of “unknown” and two were withdrawn. Most are outside the U.S., and none were randomized control trials.
Investigating the researchers’ basic hypothesis (that autism is caused by brain inflammation and that stem cells can relieve the inflammation), Spectrum journalist Brendan Borrell reported that, in addition to no benefit from the treatment itself, “the researchers [also] published no evidence to support their inflammation theory.”
Like Kriegstein, Paul Knoepfler, a professor in UC Irvine’s Department of Cell Biology and Human Anatomy, was watching the IMPACT trials. Knoepfler sees “no convincing rationale” for a stem cells as an autism treatment. “But the Duke team seemed unwilling to accept their own results and said they were ‘encouraged,’” after the Phase II results, he told me.
“They've gone on to start another autism trial,” Knoepfler wrote on his blog The Niche. “I have to wonder… if this new study also says there's no benefit, will they start yet another trial?” The key show across the board that there is no benefit of cord blood for these patients.” Knoepfler was one of the earliest voices, keeping track of the project and raising red flags, through his blog and in media coverage.
Clear evidence of harm
Not only was there no evidence of benefit, there was also evidence of harm. From early on in the decade-long trials, there were clearly documented risks to the children. In Phase II of Duke’s trials, with participants aged 2-11 years, two children experienced severe allergic reactions, with one suffering anaphylaxis. Some children have had to be sedated while receiving stem cell infusions, because the process caused them trauma and meltdowns. Other children were held down, according to parents. (On the Facebook group, one parent even shared her relief that there were enough staff to help hold her child down during the infusion).
Duke’s researchers described emotional reactions by the children as “aggression, agitation, anxiety, defiant behavior, depression, emotional lability, insomnia, intentional self‐injury, and stereotypies.” Describing the Phase II study, researchers noted:
“The most frequently occurring event, agitation during the infusion procedure, was associated with the requirements of placing and maintaining an IV and being confined to a hospital room for the infusion.”
“These children are making it clear they don’t want to be in the lab. They are saying no.” says Melissa Eaton, an advocate inside the groups, who saw post after post after post from parents concerned about their children’s behaviour after the infusions. We gathered scores of screenshots of these discussions, sharing them with FDA investigators.[3] Says Eaton:
Parents in these groups describe their kids coming back from stem cells and having meltdowns. We need to talk about trauma risk because [they] may be experiencing severe distress from these procedures, which could have long term psychological risks such as PTSD.
In the Facebook group, parents also shared documents that Duke had given to them with the hope of calming their children. One was a ‘social story’ for children, explaining the infusion procedure with photos of a teddy bear going through the motions of being infused. “I will feel a little pinch for a few seconds. I am brave!” says one caption next to a photo of the stuffed bear. “My new friend will put a sticker on me. Next, they will put some tubes near my sticker!”
It's interesting that the Duke team made an effort to ease children’s anxiety, but as any parent of an autistic kid will tell you, there are some problems a social story just can’t fix. The stress of travel, pain and a new environment (the clinic), together with the experience of being held down and injected with stem cells and/or sedated creates the perfect storm for a long-term meltdown. Add to that the fact stem cell infusions leave an awful taste in one’s mouth and an enduring smell exuding from one’s skin that recipients describe as “the creamed corn smell.” For an autistic person with sensory sensitivities, the entire experience sounds like a living nightmare.
Indeed, some parents in the group described their children remaining agitated long after the sessions ended. As just one example of many, one parent wrote that six weeks after stem cell infusion, things were getting worse with their child’s behaviour, with their son climbing and stimming to self-regulate all day long. Other parents wrote about their children becoming more aggressive.
There is no evidence that the stem cells themselves caused the agitation; rather, it was the conditions of the infusion process and its aftermath. In fact, Duke team’s assessment of anxiety in Phase II trials was that children’s anxiety remained at relatively the same level same whether children received stem cells or placebos in the clinic. It seemed the experience of going to the clinics had no documented effects—except for trauma. To get through the process, parents would need, somehow, to believe it was worth it.
It is a chilling thought, but perhaps Duke’s social story booklet was more for the parents’ sense of security than it was for the children.
Despite criticisms and negative results, Marcus (still the sole funder) was willing to continue bankrolling new trials and committed to doing so until 2024. The Duke team also began to explore another funding stream--not the National Institute of Health or any other health-related government granting agency. With no evidence of benefit from these or other trials, those applications probably didn’t look promising.
The next level of funding would come from the parents themselves.
Coming Tuesday. Part 4: “Compassionate care” or exploitation? Duke’s pay-to-play stem cell program.
[1] The issue of sole-source funding for research is beyond the scope of this chapter, but it is worth noting that it’s an issue being discussed, in part because such projects can impact the credibility of an institution as a whole. As Harvard History of Science professor Naomi Oreskes has asserted: “It undermines the integrity of the research enterprise when individuals can pick and choose lines of inquiry that appeal to them simply because they can pay for them.”
[2] It is interesting that Marcus was connected with Autism Speaks. As noted elsewhere, its founder, Bob Wright was a VP at NBC, and the group was successful at gaining media attention in a way no other autism organization ever has before—or since. The Duke project was able to build synergy with local news affiliates and major networks, with the latter accepting the team’s pre-packaged “good news stories” about a promising autism treatment.
[3] This was an interesting process. Each time I emailed a cache of documents, the investigator would respond with a message to the effect of: “I am in possession of the information you’ve sent.” I continued to send in the documents, hoping they weren’t falling into a black hole. Then I’d check the FDA’s public site for warning letters and actions, although I realized the agency might choose a more low-key or even off-the-public-record approach to addressing issues with the project.