Skip to main content


An Approval Like No Other

Merck got an unusual approval at the FDA recently, for its Keytruda antibody in oncology. What’s new about it is that it is the first approval that’s based on the molecular biology of the tumors rather than their location. Keytruda is now approved for high microsatellite instability and/or mismatch repair deficient tumors, no matter where such a phenotype might occur.

We’re used to referring to cancer with a body part signifier, because that’s been the most natural classification, but no longer. Eventually, I wonder if there will be acronyms for the various biomarker/molecular biology approvals that might come into general use. Will people refer to a diagnosis of “MIMR” cancer, say, and will calling it by a body part come to sound old-fashioned?

Biology, chemistry, and physics make their way into the popular consciousness in unpredictable ways. I’m not sure if people could have predicted the way that DNA has made itself a visual and verbal shorthand, for example – but I wonder when the first big example was of the double helix as a graphic design element? Chemistry hasn’t had quite as many inroads into the general vocabulary, other than people having good or bad chemistry themselves, but medical acronyms get picked up pretty broadly. I would not be surprised if the way that we talk about cancer has just started to change, as of this week.


15 comments on “An Approval Like No Other”

  1. steve says:

    The issue here is the immunogenic nature of the tumor. High microsatellite instability and/or deficient mismatch repair result in a high level of neoantigen expression which correlates with efficacy of PD-1/PD-L1 inhibitors. Despite their enormous success, only about 20-30% of patients respond to checkpoint inhibitors. Identifying that subset is helpful but the key will be how to turn the 80% of cold tumors into hot ones. That’s where a lot of the energy in the field is currently focused.

    1. UudonRock says:

      A specific pathway targeting drug like PD-1/PD-L1 inhibitors seem to be the way we are moving recently. It is a different development considering the focus on cell of origin based treatment. While it is interesting the simple fact is these inhibitors typically only work for a few months. Targeting genomic mutation could be a better option, but we will need to expand our understanding of what a tumor is.

      1. steve says:

        Sorry, but you don’t understand the mechanism. PD-1/PDL-1 inhibitors remove a block to the immune response to tumors. It’s not a “specific pathway drug” and invokes multiple immune mechanisms as well as direct activity on some tumors. It also doesn’t just work for a few months and has been responsible for cures in otherwise uncurable tumors. You might want to read up a bit more on PD-1 and on immunotherapy in general.

        1. UudonRock says:

          Programmed Death-1 (or CD279) is a specific genetic mutation pathway that is used in some cancers as a means to make themselves “death-proof” by evading the immune system. I do understand the mechanism. Perhaps my wording was poorly chosen. I say pathway as this is one of many possible mutation pathways required by a cell (though not all “choose” this path) to become malignant. Cancer requires a perfect storm of mutations though, not just one pathway but many. What I was referring to was in spite of excellent response in early treatment, cancer immune evasion mechanisms and toxicity with continual treatment is sadly real. The journal Oncoimmunology has some excellent papers in this area.

  2. Unchimiste says:

    About the double helix: at the onset of the 16th century, Leonardo Da Vinci designed the double spiral staircase that is at the center of the Chateau de Chambord in France. It looks a lot like a DNA molecule, except that the staircase has no minor-major grooves.

    1. NJBiologist says:

      That’s OK–most popular representations of DNA don’t have major-minor grooves, either.

      [quietly grinds teeth]

  3. Peter S. Shenkin says:

    “I wonder when the first big example was of the double helix as a graphic design element? Chemistry hasn’t had quite as many inroads into the general vocabulary”

    Gotta disagree with the “Chemistry hasn’t…” assertion, Derek.

    The CPK-style depiction of molecules is ubiquitous now. I know you were around when it wasn’t, but you’ve probably gotten so used to it by now that you’ve forgotten! Back in the ’80s, they were rare even in C&E News. I was working on the MacroModel program then. I remember when we saw such an image on the cover for the first time. Wow! One of our graphical developers commented that the image made with MacroModel. He could tell because of a subtle flaw in the representation that nobody else would have noticed. 😉

    1. anon the II says:

      That brings up an unrelated question. Whatever happened to Clark Still? As a student and early in my career, he was one of my heros. He seems to have disappeared around 2000.

      1. former columbia student says:

        Last I heard, he was retired flying home-built airplanes around the country.

      2. Peter S. Shenkin says:

        Hi, @anon the II,

        I have been in touch with Clark. We get together for dinner from time to time, though we haven’t over the past year. I was lucky to work with him (and in fact, I’ve been lucky throughout my career to have worked with great people). He’s still one of my heroes. 🙂

        When MacroModel was licensed to Schrödinger and the development group moved over, he saw it as a way to retire, which had been on his mind for some time. So he travels a lot, does a lot of scuba diving, and (as far as I know) has not been involved with science. At the time, he said he’d probably go crazy after a while not being involved with science, but that seems not to have happened.

    2. Derek Lowe says:

      You know, you have a point there. You used to see those things back in the old days sometimes, but those were photos of the actual plastic models. Being able to crank them out on a computer certainly raised their profile. I wonder, though, if you showed a hundred random people a CPK molecule and asked them what it was, how many of them would say “Some kind of chemical” or some such answer, compared to the number of people who could identify a double helix as DNA?

  4. Barry says:

    Wow, it’s been a long time coming. Richard Pazdur acknowledged publicly already in 2005 that defining cancers by their tissue of origin is unhelpful, but that was (and still is) the way our FDA defines them. “Pancreatic cancer” is one disease, and “kidney cancer” is another disease, and if you want to test a potential drug in both, you have had to run two clinical trials, asking for two approvals, although the two may be driven by the same biological target(s).
    This newest approval sounds like a new and more hopeful chapter in the regulation of Onco research.

  5. PLR says:

    I don’t think it will go all the way to defined by genetics as the tissue of origin has a significant impact in many cases. For example, my thyroid cancer may have been BRAF V600E, but the treatment with 131-iodine would not have worked very well for melanoma BRAF V600E. All depends on how differentiated the cancer remains as to how tissue-source like it is and how the treatment should progress. Now should it come back and be resistant to radioiodine treatment, the BRAF status may make all the difference.

  6. oncodoc says:

    This is very interesting. I don’t think that it a first. Lymphoma treatments have tended to be cell specific rather than tissue or organ specific. I don’t think that the approval for rituximab specified an organ. Also, let’s keep in mind that there have been some blind alleys in this path. Bronchogenic carcinoma frequently expresses estrogen receptors which may have some prognostic significance, but has not been a therapy opportunity. Small cell lung cancer typically expresses c-kit, but anti c-kit treatment has little impact. Several centers offer testing for a variety of potential targets, but clinical successes have been uncommon.

  7. Jasmin says:

    I finished my biochemistry undergrad in 2012 thinking (naively) that everybody classified cancers according to their molecular properties; tissue of origin almost never came up in my classes. When I started working in a hospital pathology department not long after I was surprised to find that wasn’t the case at all. (-:

Comments are closed.