Skip to main content


Click Chemotherapy

So here’s an ambitious idea that’s about to get a hearing in human clinical trials. A startup called Shasqi is using click chemistry as a drug delivery method, and they have a new manuscript on the idea here at ChemRxiv.

The idea is this: you produce a modified version of a hyaluronate biopolymer, decorated with aryl-tetrazine functional groups. The tetrazine/cyclooctene reaction has been exploited many times (here’s an example) since the group was introduced by Joseph Fox’s group at Delaware some years ago. Under the banner of click chemistry, as introduced by Barry Sharpless, the idea is that the two components have little or no reactivity except for each other, but will react if they merely come into proximity under the right conditions.

So you take the tetrazine-laced biopolymer and inject it into a tumor site, where it is expected to largely sit there. You then inject a chemotherapy agent that has been modified to contain a cyclooctene group – and in this system, that modification is through an ester/carbonate/carbamate linkage to a cyclooctenol group. That does several things for you, ideally: if optimized, it makes the chemotherapy agent less effective until that ester is cleaved, so larger doses of it can be given with a better safety profile. And when it encounters the tetrazine-containing polymer, it does the cycloaddition click reaction which then cleaves the ester and releases the free drug at the site of the tumor.

Update: there’s a company in the Netherlands (Tagworks) that is working on a click delivery idea of their own, using antibody-drug conjugates. More on this as it moves into the clinic itself!

It’s a pretty slick idea. Does it work? The team published a proof-of-concept in a mouse model in 2014 (open access link) using a radioligand, and it really did seem to localize the agent at the site of injection when it was delivered a few hours after the hyaluronate polymer. In the latest paper, cyclooctynol versions of the well-known chemotherapy drugs doxorubicin, paclitaxel, etoposide, and gemcitabine are prepared to give it a try from several directions. This was not trivial chemistry, as a look through the paper will show, but they were able to make example of all four modifications.

Then came the evaluations. The modified versions of doxorubicin and etoposide were indeed less cytotoxic than the parent compounds, but not the other two, unfortunately. Helpfully, the modified doxorubicin also had better solubility than the parent compound. It also had better plasma stability than the modified etoposide, so that made it the obvious choice to proceed with. And it does indeed react with the tetrazine hyaluronate polymer in vitro, releasing doxorubicin itself.

What about in vivo, though? In a mouse model, the modified dox compound could be dosed up to 10x the maximum tolerated dose of doxorubicin itself, so that part checked out. And the pharmacokinetics checked out as well: when mice were injected with the modified doxorubicin and then given an injection of the biopolymer, blood levels of the former compound dropped quickly (by over 2000-fold), with an increase in free doxorubicin at the same time. This phenomenon repeated over multiple daily injections of the modified doxorubicin, albeit with slight decreases in capture over time. These are presumably due to gradually diminished tetrazine sites on the polymer and/or its degradation in vivo, but the effects were still significant all the way through. Exposure to plain dox was significantly lower in peripheral tissue compared to a normal dosing protocol, as were its adverse effects in sensitive tissues such as cardiac muscle.

And here is another preprint in which this system is extended to mouse xenograft models, with effects both on the injected tumor and distal ones. So the idea, up to this point, appears worth trying out. The company has started dosing patients in a Phase I trial in people with various solid tumors who are ineligible for standard-of-care treatment. That’s a tough population to show good effects in, but that will make it all the more interesting if they can deliver. There are several places where things could go wrong, of course. The modified polymer and/or the modified doxorubicin could turn out not to be well tolerated in patients. The pharmacokinetics of the click-capture mechanism could be looser or less dramatic than they were in the mouse model (and this could vary from patient to patient and across different sorts of malignancies). And finally, there’s going to be only so much that doxorubicin itself can do for some of these cases. It’s not an infallible tumor destroyer – we don’t have too many of those – and the current trial will (at best) get the most out of doxorubicin treatment that it has to give.

But if the idea is sound, there are plenty of other applications for it, with more labor on the med-chem synthetic front to produce more modified molecules. Different compounds, cocktail treatment regimes, combinations of click-labeled drugs, even – there will be a lot to investigate, if it looks worthwhile. Let’s see if it does!

45 comments on “Click Chemotherapy”

  1. KazooChemist says:

    Interesting idea. The chemistry actually involves trans-cyclooctenes, not cyclooctynes.

  2. Anon says:



    1. Mathijs van den Bergh says:



      Because it’s a publication by the company itself. There’s no “conflict” of interest because it’s not purporting to be a 3rd party evaluation. In other words, the interests are just what you would expect them to be from reading author affiliations. It would be like a scientist putting “I did this experiment and therefore I have an interest in it having worked” in their conflict of interest declaration. We all already knew that.

  3. Anon says:

    Why not just directly inject the Doxorubicin at the tumor site?

    1. APAJ says:

      I wondered the exact same thing, until the “multiple daily injections of the modified doxorubicin” was mentioned. This presumably would be the design: one well aimed shot of resident polymer, multiple doses of (IV) doxo. I think..

      1. Derek Lowe says:

        Exactly – that’s the plan.

    2. Derek Lowe says:

      I think the idea is that you get better, sustained exposure on the tumor if the polymer is sitting there all the time, and that way you only have to do one tumor injection (and then give the modified dox systemically, as always). There’s also the idea that the modified chemotherapy agent is intrinsically less toxic, so the exposure to the real agent is much more tumor-directed.

    3. Swedish Chef says:

      Unmodified doxorubicin is incredibly toxic to tissues when directly injected. Doxo solutions that inadvertently extravasate when given to patients by IV cause severe skin ulceration very often requiring skin grafts to heal them.

      1. Oudeis says:

        And that, kids, is why you don’t want cancer. Just think how bad a disease has to be in order for that drug to be an improvement.

        1. Some idiot says:

          Bing! Consider that statement stolen!!!
          When explaining why (apart from other things) antivirals are a “hard” problem (compared to, say, antibacterials), I usually use the argument that as a starting point, you want to kill the baddies without touching the good good guys, so you have to find differences… Your statement here sums up quite nicely why cancer is bad…!

          1. ScientistSailor says:

            Antibacterials are a *much* harder problem than antiviral. Just count the number of new antivirals approved in the last 20 years vs new antibiotics…

          2. Some idiot says:

            Yep, I knew I was headed for that one…! I had decided to only write it in a short way, but… Yes, we agree that getting new antibacterials onto the market at the moment is quite hard…

            My point is that in _general_ terms (when describing the problem to people outside the field), and using a historic perspective, that the less like a human cell a pathogen/whatever is, the easier it is to find something that hits it without bad side-effects on human cells. From that point of view, a bacterium is less “human” than a virus, which in turn is less “human” than a cancer cell. Which is the usual progression I use to describe the concept.

            However, I agree entirely that right here and now it is really tough to find/approve/new antibiotics, for many reasons, all of which I feel you probably know better than I (i.e. need new mechanisms, need to show advantages over the large number of existing antibacterials, etc etc). Yes, we are desperately in need of new and novel antibacterials, but I don’t see a run of them coming any time soon! Definitely an area of need…

          3. Mathijs van den Bergh says:

            On the one hand, viruses are more different than human cells than bacteria are. On the other hand, viruses are obligate intracellular parasites which use a lot of human cell biochemistry to reproduce. So the target for a virus is really the human cell + virus system.

            Antiparasitics are even harder. It’s a good thing that there’s no such thing as a mammalian parasite of humans (except for cancer, I guess) as it would be almost impossible to target it effectively.

            A similar reason why it’s much harder to come up with insecticides which aren’t toxic to humans than a herbicide – insect biochemistry is obviously closer to ours than plants. The big differences are in neurotransmitters – and lo and behold a lot of insecticides are neurotoxins.

  4. David E. Young, MD says:

    Still make sense only for a single solid tumor. What makes cancers bad it their ability and tendency to metastasize and if you have dozens of visible metastases and hundreds of microscoping metastases, then this concept loses attractiveness. Antibody-drug conjugates are more attractive but they can go where-ever there is tumor.

    Researches have done all sorts of things to treat a single, but unresectable tumor. Heat it up. Inject all sorts of things inside. Radiofrequency ablation (works for single liver tumors and perhpas single bone or lung tumors). But a lot of other “inject this drug right into the tumor” strategies have not worked out so well.

    1. Jose says:

      This is Jose, founder and CEO of Shasqi.
      My disclosure is that I am a shareholder, employee and inventor of the technology.

      Thank you for your comment Dr. Young
      Our preclinical studies, published in pre-print format here (not peer-reviewed yet):
      show that there is an effect in the injected tumor with the biopolymer as well as non-injected tumors.

      Obviously, the translation of these types of effects from mice to humans have been less than reliable and that is one of the many questions that we hope to answer in our clinical study.

      Even, if our clinical study shows that the effect in humans is limited to the injected tumor, there could be still significant benefits, because expanding the therapeutic window may modify the risk / benefit calculus on when and how to use well known cytotoxic agents as monotherapy (e.g. neoadjuvant indications) or combinations with local or systemic immunooncology agents.

      1. Chocolate Seller says:

        Wishing you the very best, Jose. I am involved in manufacture of APIs and sometimes I indulge myself by thinking I am contributing to making the world a better place. But the potential of your technology is just another level!
        Ok, vested interest here – I have a close relative who might be want to enroll in the clinical study. Is there an email that he can reach out?

        1. SS says:

          Please check out the information on the clinical trial at

          It has info on the clinical centers and the email you can contact.

  5. Koss says:

    “It’s not an infallible tumor destroyer – we don’t have too many of those – ”
    How about an alpha-emitter radionuclide with a relatively short half life? Should be pretty reliable for shredding the immediate vicinity of where it binds.

    1. Metaphysician says:

      I would assume that falls into the “so can a handgun” territory. If you deliver an alpha emitter to a tumor with enough potency to kill the tumor outright, how much additional cancer will the radiation cause in other healthy tissue? After all, no deliver system is 100% precise and reliable.

      1. confused says:

        Yeah, I;m sure there are plenty of “Things I Won’t Work With” compounds that would 100% destroy a tumor – but would also be fatal to the host (chlorine trifluoride, etc.)

      2. Koss says:

        I mean it is a matter of dose and binding affinity? A lot of conventional chemotherapy drugs are also quite toxic (and mutagenic) to all cells, no cell really likes having its DNA crosslinked by cisplatin or nitrogen mustards etc. Often, the systemic side effects are dose limiting, some of them like permanent hearing loss and neuropathy can be debilitating.
        Dosing alpha-emitters is not healthy, but then again neither are cytotoxic drugs, and if the dose preferentially accumulates in the tumor it might be a worthwhile trade-off.

        1. Pedwards says:

          One of the issues with the idea of target dosing of alpha emitters is that you can’t really do it. Other targeted-dosing methods (like ADCs or the one presented here) work because they payload molecule isn’t active until a certain event happens to cleave it from the targeting system, at which point it should already be in place. Even if you were to bind an alpha emitter to a perfect targeting system, it’s going to be doing all sorts of damage as it decays on it’s way to the tumor, and any of it that leaks out of the tumor is likewise going to cause serious problems.

          Essentially, if ADCs et al. are fire crackers that you can light with a long fuse/remote detonator once they reach the target, targeted alpha emitters are short-fuse roman candles that you have to ignite well before they ever reach their target.

      3. J. Bojarski says:

        That is called brachytherapy, and is done by directly implanting solid radionuclide beads into the tumor. Prostate cancers are treated that way. It’s effective and very precise, but also very invasive, not to mention the patient will be walking around with an alpha emitter inside the body. It’s also possible to insert the source temporarily, but this requires sticking a tube (often several) into the tumor.

        With this technique, a radionuclide wouldn’t work, because it will radiate regardless of whether it’s on the site or not. The idea is not only to bind the agent in the tumor, but also to allow injecting it in a form that’s not quite as toxic.

        1. DrSAR says:

          Small point: brachytherapy of the prostate mostly involves iodine-125 or palladium-103 or cesium-131 none of which emit alpha particles and which do most of their damage in nearby tissue with low-energy photons (X-rays).

          If you want alpha particle as a targeted radionuclide, have a look at Ac-225. It’s all the rage recently and when bound to PMSA could be quite promising in prostate cancer.

          1. J. Bojarski says:

            Yeah, that’s what I’ve been thinking of. Radium is also used, though alpha emitters are problematic because of their toxicity, if they end up somewhere they shouldn’t be. Most brachytherapy is indeed beta or gamma emitters.

            However, it turns out the technique very similar to what was proposed is already in limited use, with an alpha emitter (radium-223), even, to treat bone cancer. It’s called Unsealed Source Radiotherapy (USR), and it is (or was) used in cases where you can get the radionuclide to accumulate somewhere, like iodine-131 in the thyroid. It’s an old technique, not very popular for the exact reasons I described (plus, since it usually uses beta emitters like iodine, it makes the patient a walking radiation hazard), but it apparently does extend the patient’s life somewhat.

            I think it just goes to show how bad cancer is, that we’re willing to go that far in treating it. While I don’t think handguns have been used as a method of tumor resection, I wouldn’t be entirely surprised if someone tried that, too. 🙂 Some of the older chemotherapy treatments basically boil down to “poison yourself, and hope the cancer dies first”.

      4. Charles H. says:

        It can be a bit more specific than a hand gun. I believe that in the not to distant past (currently?) something similar was used against some bone tumors. Then there’s radio-iodine and thyroid cancer. So while it’s limited, and certainly never ideal, it’s not always impossibly bad…at least until you find a better approach.

      5. István Ujváry says:

        It depends. Boron neutron capture therapy (BNCT) is an example of targeted radiotherapy (I did some carborane chemistry in a very different context). BNCT uses non-radioactive Boron-10-containing drugs carrying a nonradioactive drug which accumulates in cancerous cells. Nuclear reaction occurs when the stable isotope Boron-10 is irradiated with basically harmless low-energy thermal neutrons to yield alpha-particles Helium-4 and recoiling Lithum-7. In the fission reaction these particles provide high energy along a very brief pathway: less than 10 micrometer. The short range of this reaction limits the damage to only cancer cells without affecting normal cells allowing BNCT to be used to treat cancerous cells without damaging surrounding tissues.

        1. DrWest says:

          The problem with BNCT is getting enough boron-10 into the cancer cells in vivo to reliably kill them.

      6. David E. Young, MD says:

        There are already examples of intravenous radioactive agents on the market. Luthera, is a good example [LUTATHERA® (lutetium Lu 177 dotatate) :

        These do not seem to cause cancer and do seem to help. Quadramet, is an older, more crude radioactive substance: Quadramet (Samarium SM 153 Lexidronam), for metastatic prostate cancer.

        Xofigo is an intravenous radiation treatment for prostate cancer (radium) that has been more successful and more recently approved than Quadramet.

        Alpha emitters would not be a bad choice to test out. But I like the idea of a chemotherapy drug better.

  6. Wavefunction says:

    Neat idea. When I was in graduate school I tried to submit a similar proposal for a metathesis self-assembling reagent. It unsurprisingly turned out it was hard to think of palladium catalysts that were bioavailable and non-toxic.

    1. Jose says:

      Thank you, WF.
      I agree, it would be impossible to conceive of this approach if we needed copper catalysis as the initial generation of click chemistry required.

  7. mayfin says:

    One other slant on this approach is that it might be interesting to look at testing it not with existing approved drugs, but as a way of resurrecting previous candidates that failed out because they were too toxic.

    1. Jose says:

      Mayfin, thank you for your comment.
      This is an excellent idea. As we gather more information on how the CAPAC platform performs with Doxorubicin in humans, we plan to expand to other generics that need improvement, license experimental therapeutics that may have fallen short because of access to tissue or systemic toxicity, and partner with other companies to create protodrugs out of their experimental or existing therapeutic agents based on our CAPAC platform.

  8. Jose says:

    This is Jose, CEO of Shasqi.
    We are looking to hire a Director of Medicinal Chemistry and a Director of Translational Sciences in the coming months. For more information please email us at or connect to us on LindkedIn.
    We value skills and experience and are open to remote work.

  9. John Hasler says:

    Now if you can just get that biopolymer to preferentially accumulate in tumors from the bloodstream…

    1. Some idiot says:

      My though precisely…! With antibodies, perhaps? Totally out of my field, though…!

      1. Jose says:

        Yes, we believe there is a variant where this may be possible, but we decided early on that it was not the lowest hanging fruit, as you can see from my post on the kinetics of the reaction, and our 2014 manuscript:

        “[…] while a mAb can provide exquisite localization, the system has some limitations. Antibodies require: (i) an identifiable and accessible target; (ii) specificity to said target; (iii) the ability to be chemically modified to carry a TCO cargo and maintain specificity; and (iv) rapid clearance from the blood stream of unbound mAbs to minimize unwanted side effects. The platform we describe in this report circumvents the need for mAbs, providing a modular and flexible platform for in vivo small molecule delivery.”

        I would add a (v) point: the need for the mAb and its click chemistry partner to be exposed in the extracellular matrix long or vascular system long enough to react with the other click chemistry partner.

  10. David Schwartz says:

    Question- are the kinetics of reaction of the tetrazine/cyclooctyne better than, let’s say hybridization? If not just use oligos or PNAs.

    1. Jose says:

      David, here may be more than you wanted to hear about the kinetics.

      My understanding are that the kinetics are much faster than hybridization. Also this is a covalent and irreversible reaction as the diels alder reaction adduct leads to the extrusion of N2. So what binds, remains bound.

      If you want to look a bit deeper into oligos. You can see this paper from the Mooney lab at Harvard / Wyss using alginate and oligos for drug refilling that appeared about 5 days before our seminal 2014 paper using TCO / Tz.

      About 6 months later, Brudno, Mooney and co-workers published an alternative approach using tetrazine and transcyclooctynes (not trans-cyclooctenes). Unclear why they switched from their original oligo approach.

      With regards to the kinetics piece for the TCO / Tz reaction, also dubbed as tetrazine ligation by the team at the Fox lab I would suggest the following paper by the Fox lab, including our collaborator Max Royzen:
      There they mention a k2 of 2000 M−1 s−1.

      Finally, it is interesting to see the manuscript from the Weissleder group at Mass General, where they suggest the challenges of harnessing this reaction in-vivo:

      “A plot of predicted labeling efficiency vs. both the kinetics of cycloaddition reaction and the area under the concentration curve (AUC) is displayed in
      Fig. 2C. The fraction of primary agent reacted is an exponential
      function of the reaction rate and AUC of the secondary agent.
      Small molecule tetrazines, while possessing rapid reaction rates,
      do not possess the requisite AUC for very efficient labeling, explaining our experimental results.”

      The way we got around the fast clearance rate mentioned is by forcing the local concentration of Tz’s available to remain high by attaching them to a biopolymer that remains in a specific location for a prolonged period of time, certainly much longer than a small molecule.

  11. A Nonny Mouse says:

    I do work for a clinical group at a hospital who frequently ask me to make acetylenic modifications of compounds so that they can do click chemistry with labelled fluorine azides which they use for cancer diagnostics.

  12. Anonymous says:

    Very cool idea. Obviously it’s not going to solve all our problems but I think as Jose points out, there is a real potential to help patients. I like how down-to-earth this idea is. Even if this particular drug-click combination turns out not to be effective, I think the concept is sound!

    1. Jose says:

      Thank you for your kind words.
      Yes, this could be just the beginning. If our preclinical results are confirmed by our clinical studies, then the concept could be expanded to many other agents, even if doxorubicin is not the panacea for all tumor types (it likely will not). As Derek mentioned, “different compounds, cocktail treatment regimes, combinations of click-labeled drugs”, could be explored in ways that were not previously possible.
      Our animal studies suggest that the CAPAC platform enables one to unlock biological effects of existing validated therapeutic agents that were not previously observed. But the translational aspect is what remains to be proven.

  13. Rtah100 says:

    Derek, is that reference to cyclooctynol intended, the other references are to cyclooctenols?

    My father used to perform radium therapy for prostate and womb cancer, possibly also breast. He thought is was great for the right cases and was disappointed in his successors who were terrified of the Seeds of Doom and preferred their big machines (he had some of those too, at home and in the hospital). He carried the pellets around for domicillary procedures in a big lead box in the boot of the car. We had a radiation hazard sign to hang out if they were onboard. Somebody stole the car one day: a quick word with the local paper and an article about radiation poisoning and the car was back with an apology note the next day. The box had not even been in the car but the sign had!

  14. SS says:

    For more information on Shasqi’s Clinical Trial, please visit

  15. T says:

    Probably commenting too late to get an answer on this but one thing I always wondered about these “inject the tumor with something” approaches (and there are many): If you have an easy-to-find, accessible, well-defined tumor discrete tumor, then couldn’t you just as well remove it surgically? My understanding was that the real problems start when the tumors have started to spread or they are somewhere really tricky to reach (physically and chemically) like the brain?

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.