Skip to main content


Novartis and CAR-T: What’s Going On?

The upheaval in the Novartis CAR-T program happened while I was off at a conference last week and unable to give it a full treatment. So now that it’s happened, let’s think about it a bit.

The folks at Endpoints have been a good source of info on this story, and it appears that Novartis has cut back in this area more than their press releases would have indicated. That’s actually pretty typical – the company doesn’t go in for big layoff announcements, preferring to cut back quietly. The company responded to that report by saying that no, they’re not pulling back from CAR-T, but rather re-integrating that effort into the larger organization. I’m not sure what I think about that. For large companies, this can be a legitimate point, but it can also be re-integration in sense of “dissolved back into the undifferentiated blob”, which is not good. Losing 120 positions, which is the report, does not sound like an orderly retreat.

Overall, it’s kind of hard for me to believe that the CAR-T stuff is just another oncology therapy, part of the broad portfolio, etc., which is the tone I got from the Novartis response. And the company had been making a big deal out of the separate unit and its results up until now (we’ll see how long that link is valid!) It’s also hard for me to picture that the unit had pretty much solved the scientific and manufacturing questions it was created to address, and had these in a state ready to hand off to the rest of the organization. This is still a very wild, wide-open field, and there are doubtless going to be a lot of surprises coming in it. And one of these is Novartis acting (at least a little bit) as if that’s not the case.



28 comments on “Novartis and CAR-T: What’s Going On?”

  1. wlm says:

    “Losing 120 positions, which is the report, does not sound like a ”
    Derek, either my browser is malfunctioning or we’re missing the end of that sentence.

    1. Yep says:

      True – sentence got swallowed up

    2. Derek Lowe says:

      Yep, off into the bit-bucket for some reason. I’ve fixed it – thanks.

  2. anonao says:

    And in few months, Novartis may buy a biotech for hundreds of millions which is specialised in CAR-T

  3. bhip says:

    I wonder if the engineered T cells offer a huge efficacy advantage over unleashing the patients own T cells with the Checkpoint inhibitors (PD1 et al.)? I know that the CAR-T therapies are still susceptible to the immunosuppressive influences of IDO, adenosine, etc. Perhaps, the efficacy seen with CAR-T vs. PD-1 approach is not worth the infrastructure required to generate the cells?

    1. steve says:

      Apples and oranges. One’s own T cells use T cell receptors, which only recognize antigens in the context of MHC. Tumors down-regulate their MHCs as a way of avoiding T cell surveillance. CAR-T uses chimeric antibodies instead of TCRs to circumvent this strategy. Thus CAR-T will attack tumors that one’s own immune system may miss, even in the presence of checkpoint inhibitors.

  4. anon says:

    Rumor has it that they found CAR-T to be ineffective in solid tumors and the CD19/CD22 field is pretty crowded. It wouldn’t be surprising due to the nature of the immunosuppression and antigen heterogeneity on solid tumors.

    1. paperclip says:

      Solid tumors have their challenges, but most people in the field are far from the “Oh well, we tried!” stage. Many in the field are optimistic that these challenges can be overcome.

  5. Anonymous says:

    Is switchable CAR-T in the clinic yet? Seems like it addresses cytokine storm which is the biggest issue associated with the treatment. Hyperlinked in name.

  6. m livingstone says:

    Oxford biomedica Has CAR T 5T4. 5T4 is expressed on many solid tumours

  7. CAR-T_T-RAC -er says:

    I’m not an oncologist, so I’m curious, why would the CAR-T approach not be effective in solid tumors. If someone has a reference or review to point me to that articulates that skepticism, I’d be much obliged.

    1. steve says:

      The reason CAR-T has not been useful in solid tumors so far is because of the difficulties in penetration. Tumors have high hydrodynamic pressures (which also provides entry challenges for small molecule drugs and biologics) and also they down-regulate not only MHC as in my earlier comment but also adhesion molecules like selectins, integrins, etc. that T cells use for entry. Further, they secrete immunosuppressive cytokines and other factors like TGFbeta, IL-10, IDO, NO, etc., etc. that block T cell responses. The difference between leukemias/lymphomas and solid tumors is like attacking an army on the field versus an army that is in a heavily fortified fortress on a mountain with a moat surrounding it where the bridges are all raised.

    2. anon says:

      The tumor microenvironment basically is immuno-supressive with an influx of T-regs, and other cofactors like IL-10, TGF-B, VEGF, adenosine etc. These factors actually in essence shutdown or reprogram effector T-cells. In patients that fail treatment they often find a high T-reg count for example. In a solid tumor these cofactors are contained, not so for something like CD19, a liquid tumor.

    3. Mol Biologist says:

      Cancer is characterized by adverse metabolism which affect tumor and whole organism by different ways. The possibility to activate CAR-T approach can not be used because major problem is denied. There are few steps to withdraw cash from ATM: You can use 7 step by meaning to get money, but it would not be possible if you are in in debt.

  8. steve says:

    One question about Novartis disbanding its CTGU group. Does anyone know if they are continuing with their cord blood expansion program? This also was a cell therapy where they were developing their SR1 compound (Stemregenin-1) to expand stem cells in cord blood for transplantation. If they’re still moving forward with that then it would suggest that they haven’t soured on cell therapy in general and maybe this really was just a strategic move. On the other hand, if that project has been abandoned then maybe they really are retreating from the field.

  9. simpl says:

    As far as I know, Novartis has 3 CART projects in development, and there are large differences, at least as large as in the biotech area, The most advanced one requires the patients’ own cells as raw material, others can use a more general source – more like a vaccine. Likely, a factory needs a rebuild for each product, as was the case with biotech actives, where grouping and site transfer of product manufacture has become clearer only after decades – probably along with the rise of biosimilars.
    It is a strange, almost empty, feeling if you work on the products when research stops and the development continues. I recall it happening in migrane, where the lead model turned out to be okay as an anti-emetic but not for migrane; the whole area was dropped like a hot potato. Another case was a large search program for dopaminergics that took so long to complete that the research lead was in his 80s when the last one, quinagolide, was launched.

  10. Thomas says:

    It’s probably due to Novartis’ leadership change — the supporters of cell and gene therapy left (
    Focus changes often follow management change, especially those “visionary” projects like cell and gene therapy.

  11. GREEDY PANTS says:

    I want it all, the money, the women….I want it all, said the greedy meglomanic surgeon, the insatiable maniac. ME ME ME you stupido, Its all about ME

  12. UndergradChemist says:

    This is a personal one for me as a family member of mine was affected by the layoffs… oh well, what can you do…

  13. Barry says:

    Even if a CAR-T were to work consistently, even on solid tumors, if it requires the patient’s own T-cells, Novartis might drop it. Where the patient’s own T-cells must be modified, there will never be a NCE that can pass (or fail) clinical trials and get FDA approval. The therapy–under the current regulatory regime–will always be “experimental” and therefore not reimbursable.

    1. steve says:

      Sorry, that’s nonsense. Autologous bone marrow transplants receive reimbursment. Although it didn’t do well financially, Dendreon developed an autologous dendritic cell therapy (“Provenge”) that was FDA approved and received reimbursement codes. The idea that just because a therapy is autologous that there’s no way to get reimbursed and “will always be experimental” is just ignorance of the field. There are also companies like Celectis and the newly established collaboration between FATE and Sloan Kettering (and I’m sure others) that are working on allogeneic cell therapies but autologous therapies certainly can get approved and reimbursed.

      1. Barry says:

        autologous therapies are regulated “more like a surgical procedure than like a drug”

        I.e. there’s no NCE for a Drug Company to patent/own and therefore no period of market exclusivity in which to recoup the R&D

        1. steve says:

          Again, that’s absurd. In fact, GSK just got European approval for Strimvelis, an autologous therapy using gene-modified cells ( I very much doubt that GSK would be putting together the infrastructure for autologous gene-modified HSC, that Celgene would have partnered with Bluebird on CAR-T, that Novartis would have partnered with Penn, that Pfizer would have partnered with Cellectis or that Amgen would have partnered with Kite if there was “no period of market exclusivity in which to recoup R&D”. Kite and Juno don’t have market caps in excess of a billion dollars for no reason.

          1. Proteus says:

            Are you suggesting that the first approved CD19 CART will block all others? e.g., if Kite gets approved first, Juno is not approvable for that target/disease due to regulatory exclusivity? I’d hope that is not the case.
            The way I understand it is that each approach is sufficiently unique such that market exclusivity isn’t justified. On the other hand the differences and complexity will not support referencing of data as is utilized by traditional follow-on providing some protection beyond patents.
            Also, a list a Pharma deals is never compelling evidence of rational behavior..

  14. steve says:

    No, it’s not that the first CD19 CART would block others; that’s never the case in medicine. The first statin didn’t block others either but that didn’t stop them from making money. Exclusivity lies in the patents. There is already a patent battle on CAR-T between St. Jude and Penn and between Juno and Novartis; there undoubtedly be will be others. The patent landscape and exclusivity on that front is unclear. Generally the way these things get resolved is that the loser pays the winner a portion of the profits so both products end up on the market (Novartis already settled with Juno paying $12.5M upfront and royalties down the road). However, the original claims were that autologous therapies can’t get reimbursed and that they can’t recoup their R&D costs. Neither claim has any basis in reality.

    1. Dave Hoffman says:

      The Juno/Novartis agreement is actually a consequence of the settlement in the St. Jude/Penn litigation, as Novartis is partnered with Penn and the IP at issue in the St. Jude/Penn case is owned by St. Jude and licensed to Juno.

  15. Proteus says:

    I agree totally. I took the earlier statement as a suggestion that New chemical entity regulatory exclusivity (21CFR314.108 –which blocks approval of the same chemical entity submitted by a second party) to include cd19 CARTs as a class of new chemical entity. That would be silly.
    I think you are correct in that there is sufficient protection / incentive there for a couple winners, although probably not enough for everyone in the race now unless they start getting better data in other tumors.

  16. smallbizmom says:

    ..My 16 year led son had t-lymphoblastic lymphoma. I’m hanging on what these people will do everyday in case we need it…Real people are affected by this stuff…

Comments are closed.