Lessons from New IPS Cell Study Showing Tumors: Choose Methods & Validate Cells Wisely

Okano labA cool new paper is out in Stem Cell Reports describing long-term tumorigenicity of human induced pluripotent stem cells (IPSC or IPS cells). See graphical abstract.

The potential tumorigenicity of IPSC is a major concern when teams around the world are translating IPSC technology to the bedside. Past studies including one from my own lab have shown that there are some parallels between cellular reprogramming to make IPSC and tumorigenicity. MYC has often been viewed as the likely culprit but this paper expands the potentially risky factors to include OCT4. Given that the line that was the focus of this study from the Okano lab was made without MYC, the prediction was that differentiated cells (neurospheres) made from it wouldn’t be tumorigenic, but tumors were evident after long-term follow up of the transplant recipients.

Key factors that influence tumorigenicity include the method of reprogramming, the selection of clonal lines, and how the IPSC are treated thereafter.

Perhaps in part the tumorigenicity reported in this paper could stem from the use of viral transduction-based reprogramming, but an important take home message I think is that one can never be sure about the tumorigenic potential of IPCS unless you test them in a clinically relevant transplantation model system and do long-term follow up. It would also be prudent to carefully examine the transcriptome of your IPSC and the epigenome as well and consult with experts as one plans potential clinical translation. Starting with non-integrating reprogramming methods is crucial.

The tumors that developed in this studies had some characteristics of primitive neuroectodermal tumors and had Nestin+ cells. One open question is whether these tumors are high-grade, lethal tumors (seems likely). Activation of the OCT4 transgene was a proposed mechanism of tumorigenicity. The authors also speculated that “Tumor progression may have been caused by mesenchymal transition of grafted cells”.

11 thoughts on “Lessons from New IPS Cell Study Showing Tumors: Choose Methods & Validate Cells Wisely


  1. Paul, How do you view the significance of this study with regard science’s ability to overcome the tendency of tumors to form in iPSC therapies? Is this simply a symptom of a natural learning curve or is there an inherent tumor problem with IPCs that is unlikely to be overcome?


    • Hi Ray,
      I think the former idea is most likely. The IPSC field is really getting much more sophisticated as to producing IPSCs and IPSC derivatives for clinical use so I see this as another stepping stone forward. I wish we saw more studies like this but with IPSCs made using methods that were non-integrating. Long term follow up included.
      Thanks.
      Paul


  2. I had talked with them about this when they were presenting on spinal cord implantation of iPSC-derived cells at SfN a few years back, and interestingly they mentioned that they had never seen tumorigenesis on spinal cord implantation when the implanted cells were derived from an iPSC clone where MYC was included, only when it was not included. I think with MYC, people’s intuition has automatically led them to believe that because it is an oncogene, including it is going to result in tumor formation when the cells are implanted; however, I think that the actual biological regulation may be more complicated than that and what happens when iPSC-derived cells are implanted may be different from what occurs when they are used to generate chimeric mice (where tumor formation has been observed with inclusion of MYC). Along these lines, it is interesting to note that the MYC-less 253G1 clone that showed tumorigenesis for neurospheres within ~100 days after implantation in the spinal cord in this study appears to be the same one that previously did not show tumorigenesis for up to 100 days after birth when used to generate chimeric mice (Nakagawa et al., Nat Biotech 2008). For Nori et al., I think that it is also important to note that they also potentially attribute the tumorigenesis to the use of integrating vectors and that they reached the came conclusion as you regarding the use of non-integrating vectors. Having started my career designing and developing non-viral gene therapy vectors, I agree.


  3. U.S. citizens pay for much of this fascinating research via our tax dollars, and I am an impatient patient in need of stem cell therapy, as it is the only therapy that works for progressive MS.
    Paul, you’ve been warning us about human use of stem cell therapy, and I figured you might be seeing some spooky things in your IPSC lab that prompted your caution. Now there is this new evidence that IPSC is not ready for prime time.
    What advice would you give Jeanne Loring at Scripps, about the human use of IPSC in their planned Parkinson’s trial?
    It would be great if you could explain the larger picture of why research funds should be poured into IPSC, when ho-hum mesenchymal cell therapy is already in use in trials around the world, w/ published phase 1/2 results demonstrating safety. You remember this finding from the PatientForStemCell.org posted in 2013 showing 2,154 subjects in over 70 trials across all conditions, all safe, even w/ lab expanded MSCs. http://www.patientsforstemcells.org/how-safe-are-stem-cells/

    So please explain why IPSC is a better investment than MSC? From the standpoint of fastest, safest clinical translation. There’s a silver tsunami of neurodegenerants building up who are gonna need help. Battling Alzheimer’s, Parkinson’s and all the other aging diseases can really put a tarnish on your golden years.


    • SammyJo,
      It’s not “either or”. I advocate for funding, research and use of many kinds of stem cells including MSCs, ESCs, IPSC, and more. All have certain pluses and minuses.
      The key thing is that whichever type of stem cell that a physician intends to use on a patient, the stem cells should be used responsibly based on pre-existing evidence of safety and efficacy.
      The physician should follow the law and the regulations of place in which they practice. If they disagree with the laws and regs, they should work to change them, not flout them.
      They should obtain all the training possible in stem cells, bioethics, and transplantation.
      They should not practice outside of their speciality (e.g. plastic surgeon should not be generally treating MS or ALS).
      For whatever reason, today in America the physicians selling MSC “treatments” more often than not are not doing the things I’ve listed above and are acting irresponsibly. This doesn’t make MSCs themselves as cells in any way bad. I believe that MSCs and SVF have a lot of potential to help people, but they can hurt people too when used recklessly.


  4. truth4science, that seems to be a pretty general statement about iPSCs considering that only 1 iPSC clone was evaluated in this study, and that they have never observed tumorigenesis in the same transplantation model when using a different clone with a different combination of transgenes. They also suggest that the reason for the tumorigenesis in this particular case was the use of an integrating vector and not the nature of the induced pluripotency itself. Have you actually observed tumorigenesis when transplanting iPSCs, especially with the use of a non-integrating reprogramming system?


  5. SammyJo,

    I know that your questions/comments were directed towards Dr. Knoepfler, but I hope you wouldn’t mind if I also offer a couple of opinions.

    I don’t think that anyone is saying that MSC research should not be funded or should be less prioritized than iPSC research (I actually used to be an MSC researcher and we also operated based largely on federal funding). I also don’t think that anyone is saying that iPSCs are fundamentally unsafe or not ready for clinical trials.

    Personally, I think that it is important to fund both because they each have their own particular applications/strengths. There are things that iPSCs and other pluripotent stem cells can do or be used for that are not possible with MSCs (the ongoing clinical trial for iPSCs in Japan for macular degeneration is an example of this). Similarly, because iPSCs can be differentiated into neural stem cells, neurons, and glia but MSCs cannot, they may offer options or outcomes for treatment of Alzheimer’s and Parkinson’s that may not be possible with MSCs. Also, one big non-transplantation application of iPSCs is in drug screening and the general of cellular models of disease so we can figure out new and more effective ways to treat diseases; this is also something that cannot be done with MSCs. On the other hand, as you pointed out, MSCs may be particularly useful for other diseases such as those with an inflammatory component such as MS, especially because it is simpler to obtain and culture them and because they are further along in clinical development.

    I think that funding of both is important to provide the greatest therapeutic benefit and best outcomes to the largest number of people.


    • Hi Shinsakan,
      I am also all for funding and exploring multiple types of research on as many kinds of stem cells as possible.
      Paul


  6. I can comment separately on all of the quality control assays we’ve developed for human pluripotent stem cells – my mantra is “check your cells! Then check them again!”. Microarray technologies are cheap, easy, and comprehensive.

    But I want to address the question about why we’re using iPSCs for development of a Parkinson’s disease cell therapy. The reason is that we can make the exact cell type, dopamine neurons, whose death is responsible for the major symptoms of the disease. Mesenchymal stem cells have been shown to tune down inflammation, but they can never make neurons of any kind. Or pancreatic islet cells, glial cells, or heart muscle. There are cases in which MSCs can help, but many more in which they simply can’t.

Comments are closed.