Why I am still very skeptical about stem cells for autism

stem cells for autism

When patient families reach out to me, one of the most common questions is whether stem cells for autism offer real hope.stem cells for autism

The short answer at this time remains “no”. Of course, it’s not as simple as just “no” because there isn’t enough data to be sure, but there is reason for major skepticism about the use of stem cells for autism.

I’ve covered this topic a number of times over the years because patients and families want to know and because it is so often hyped. Sure, there are real clinical trials now and then with FDA approval to test stem cells in autistic patients, and in theory some may ultimately prove to be safe and effective. However, of those recently and currently being tested as well as other “therapies” already being sold at dubious clinics, I doubt any will make it to the finish line as making a real difference.

Why am I so skeptical still? Many reasons.

  • There is no compelling, preclinical evidence that I know of that stem cells can help autism and that they would be definitely safe in kids.
  • The most common route of administration, IV into the blood stream, doesn’t make much sense to me. Do a meaningful number of cells even get into the brain that way?
  • If cells don’t get into the brain, they do something systemically that helps the autistic brain? Some folks believe that there is a definite autoimmune component to autism and that flooding the bloodstream with stem cells might reduce autoimmunity, but there’s little hard evidence for either of these ideas.
  • If yes they do get into the brain, what do the cells that get into the brain do that is helpful and where do they do it inside the brain? Which brings me to the next question.
  • What causes autism spectrum disorder? The field doesn’t really know so how can you treat it if you don’t understand it?
  • Under that umbrella “spectrum” term there are many manifestations and probably many causes so stem cells are unlikely to be an autism panacea.
  • Is there any evidence that once autism manifests that it is reversible?

And the list of concerns and questions goes on and on…

According to clinical trials.gov, there are now 13 studies across the globe and 4 registered stem cells for autism trials in the US (see map above). The one here in California at Sutter has ended and no data are posted or published that I know of at this point. The one listing in Arizona and two in Virginia as well as one of the two in Florida (really apparently four that are just one study together) are also completed. They were kind of different in that they was based on the idea that hyperbaric therapy would help boost stem cells in kids with autism. No data posted. The other one in Florida, with Ageless Regenerative Institute as sponsor, was withdrawn. There the notion was to use fat stem cells to treat autism. I have no idea how fat stem cells could by any stretch of the imagination help autism. Unfortunately quite a few stem cell clinics sell stem cells of autism, which is deeply concerning. It is experimenting on kids for profit with little chance of success.

My clinicaltrials.gov search did not find the specific stem cells for autism trial most in the news of late, the one at Duke by stem cell scholar Joanne Kurtzberg, because “stem cells” are not mentioned in the listing, just umbilical cord blood. Update: They’ve published their findings here, which supports safety but we can’t be sure about efficacy in this open-label, uncontrolled study (HT: commenter Matt). I’ve posted about Dr. Kurtzberg’s trial and did an interview with her (see pieces here) in the past. To specifically read my older two-part interview with Kurtzberg see here and here.

More broadly there are 24 research articles in PubMed with both “stem cells” and “autism” in the title. Some of this is very interesting research, but it is concerning to see that in this PubMed search as of today at least, the very first result on PubMed is a “trial” in India of infusing autistic children with apparently undifferentiated human embryonic stem cells, which in my opinion is extremely dangerous.

Overall, it’s just not adding up to me that this is a fruitful line of clinical research at this point. I hope I’m wrong. Some may say to me, “What’s the harm in trying?” and I’d say that there are risks of various kinds to the children including health-wise (although these may be relatively lower with cord blood products) and in raising parents’ hopes with little to back it up.


  1. I think a better answer would be, “don’t know, not enough data as yet”.

    What we know from our studies is that one does not want to cross genders, need same gender for allogeneic stem cell transplants.

    Also, blood group from donorvneeds to be same or O-negative.

    Also, depends on which cell type they are transplanting. If want ectodermal lineage cells, such as neurons and/or glial cells, Then MSCs won’t work and neither will mesodermal stem cells or endodermal stem cells. Can only use ectodermal stem cells, pluripotent stem cells, or totipotent stem cells.

      • Yes.
        Theoretical: 1) The adrenal cortex secretes low levels of testosterone and estrogen in both males and females throughout their lifespan. 2) Male (XY) cells are pre-programmed to respond preferentially to testosterone. 3) Female (XX) cells are pre-programmed to respond preferentially to estrogen. 4) Naive totipotent stem cells will form any cell type in the body, including the germ cells. 5) Pluripotent stem cells will form any cell type in the body, except the germ cells. 6) Totipotent stem cells and pluripotent stem cells are located in the stroma of almost every tissue and organ in the body as resident populations of naive dormant stem cells. 7) Neither naive totipotent stem cells nor naive pluripotent stem cells diaply MHC-Class-I and/or HLA-DR on their cell surface. 8) Naive ectodermal stem cells will differentiate into any cell type of the ectodermal embryonic lineage, such as epidermis, glands that secrete to the outside of the body, hair, nails, enamel; neurons and glial cells of the CNS and PNS, and neural crest-derived cell types, etc. 9) Naive mesodermal stem cells will differentiate into any cell type of the mesodermal embryonic lineage, such as tissues of the musculoskeletal system, vasculature, immune system, hematopoietic system, genito-urinary system, stroma, etc. 10) Naive endodermal stem cells will differentiate into any cell type of the endodermal embryonic lineage, such as the cells lining the respiratory system and gastrointestinal system, and parenchyma of glands that secrete to the inside of the body, such as the liver, gall bladder, pancreas, etc. 11) Ectodermal stem cells, mesodermal stem cells, and endodermal stem cells express MHC-Class-I and/or HLA-DR cell surface markers. 12) Naive stem cells after transplantation into an individual (animal or human) will preferentially migrate to areas of tissue damage. 13) Naive stem cells, responding to local cues, incorporate into, and help repair the damaged tissue. 14) Those stem cells that do not directly contribute to tissue repair become permanent residents of the body. 15) Newly resident stem cells can proliferate and migrate to areas of tissue damage to repair the damage.

        Gender: We performed long term animal studies with same gender and cross gender transplants using genomically-labeled totipotent stem cells and pluripotent stem cells, i.e., male donor to male recipient, male donor to female recipient, female donor to female recipient, and female donor to male recipient. In most instances where totipotent stem cells or pluripotent stem cells were used to repair damaged tissue within the brain, lungs, or heart, no discernible gender differences were noted. However, if the ovaries or testicles were removed in younger recipient animals, or as the animals aged, gender-specific differentiation of the donor cells was noted, regardless of the gender of the recipient.

        We have been tracking our human volunteers that have received autologous [self] and/or allogeneic [non-self] stem cell transplants long term. The majority of our clinical volunteers are either post-menopausal or have autoimmune disorders that have affected their reproductive organs. Same gender (autologous or allogeneic) stem cell transplants demonstrated no discernible difference. However, cross gender transplantations demonstrated male pattern hair distribution in females and gynomastia in males.

        Blood Groups: Neither totipotent stem cells nor pluripotent stem cells display cell surface markers to distinguish self from non-self. However, as they differentiate into downstream cell types, i.e., ectodermal stem cells, mesodermal stem cells, endodermal stem cells, progenitor cells, and differentiated cells they acquire self-specific cell surface markers. To ascertain whether the acquisition of self-specific cell surface markers during subsequent differentiation would disrupt stem cell transplants, we performed out-bred to in-bred stem cell transplants in animals as well as cross species transplants, utilizing genomically-labeled stem cells. Neither totipotent stem cells nor pluripotent stem cells elicited a graft versus host response. However, ectodermal stem cells, mesodermal stem cells, endodermal stem cells, and progenitor cells did elicit a graft versus host response.

        We obtain totipotent stem cells and pluripotent stem cells for transplantation that are circulating in the blood stream of the donor (autologous or allogeneic). However, there are also mesodermal stem cells within the preparation harvested from the blood. We are following Red Cross guidelines for blood transfusions, whereby one uses either the same ABO blood group or O-negative blood (universal donor). Therefore, we are following a more conservative route in case potential contaminating mesodermal stem cells would potentially elicit a graft versus host response in the recipient.

        Differentiation: With respect to which stem cell type differentiating into specific tissue types, see above under theoretical (#s 4,5,8-10).

          • We are waiting for investor funding before we put our pro bono clinical trials on the website. We are currently under IRB approval for a limited number of endogenous autologous and allogeneic TSC and PSC treatments for neurodegenerative, pulmonary, cardiovascular, autoimmune, and systemic disorders.

  2. “What causes autism spectrum disorder? The field doesn’t really know so how can you treat it if you don’t understand it?”

    This statement is the most compelling reason any parents should be skeptical about treatment and why I agree with you. You don’t have to be a biologist to question how one can expect to cure an affliction unless you know what causes it. It’s like trying to fix a software bug without having access to the underlying code.

    • That’s a good analogy. Trying experiments with “stem cells” of various kinds to attempt to help autism is, to use another figure of speach, a real shot in the dark.

    • Indeed, I have been in exactly that position. I have identified and fixed a software bug without access to the code. Difficult, yes. Sometimes possible…

      The code is not the issue. The issue is understanding the nature of the problem. These things should NOT be confused!

  3. Kurtzberg’s work for autism doesn’t mention stem cells because the believe mechanism of effect is from a different cell source. As Kurtzberg often says, cord blood is a bag of cells, not just stem cells. They recently published their results in Stem Cell Translational Medicine official journal of the Cord Blood Association. While there is certainly more work to do, phase II trails are underway with ‘best unit’ of allogenic or autologous cord blood.


    Dawson, G., Sun, J. M., Davlantis, K. S., Murias, M., Franz, L., Troy, J., Simmons, R., Sabatos-DeVito, M., Durham, R. and Kurtzberg, J. (2017), Autologous Cord Blood Infusions Are Safe and Feasible in Young Children with Autism Spectrum Disorder: Results of a Single-Center Phase I Open-Label Trial. STEM CELLS Translational Medicine, 6: 1332–1339. doi:10.1002/sctm.16-0474

    • Thanks, Matt. Now this rings a bell as something I saw earlier.
      As an open-label, uncontrolled study, can’t really conclude much from it on efficacy. Safety seems good.

      • That is correct. And the authors have noted during conferences that the parents in for the children enrolled are exceptionally involved with their chides therapy. Also the metrics by which improvement is measured have some subjectivity as well. They have moved onto a phase II clinical trial with a larger sample size using ‘best source’ cord blood from either an autologous or allogenic source

        Cord Blood Infusion for Children With Autism Spectrum Disorder (Duke ACT) – NCT02847182

  4. Is there any time limit for the onset of possible complications caused by stem cell treatment (for autism or any other condition)? Even tests that assess the safety of this kind of treatment usually cover a relatively short follow-up period.

    • Not really. With a living biological therapy, the body doesn’t necessarily eliminate it as happens with a defined half-live with chemical drugs. For this reason, those who receive stem cell infusions may have some of the cells the rest of their lives, which is particularly important in children. It is important to note that for many types of stem cell transplants such as IV infusions, it is thought that most cells die rather quickly.

  5. On the other hand, I see that variations in Mecp2 and Gabrb3 are associated with autism.
    Or perhaps I should say autism spectral disorder. Or perhaps not “disorder” since not everyone thinks it is either a disease or a disability. So, perhaps there is the beginning of a theoretical basis for modifying autism…. But will the “patient” be grateful for the “cure”?

  6. The only way I can think of that hematopoietic stem cells can influence the brain is by giving rise to microglia..if autism were a microglial disorder, a cord blood transplant might have some impact. Microglia are hematopoietic lineage cells, similar to macrophages, that originate in the yolk sac during embryogenesis, but there are reports that microglia can arise from bone marrow transplants, perhaps only in cases in which there’s neuroinflammation….worth a discussion.

    • For cord blood, Kurtzberg doesn’t think it is the hematopoietic stem cells but other effector cells.

      Also, it is entirely possible autism has multiple mechanisms in play, and thus is may take a combination of treatment options to effect change.

Leave a Reply