KRAS Drugs Fighting Uphill | Seeking Alpha

Let’s talk about KRAS a bit. It has a lot to tell us about drug discovery. For most of my med-chem career, it was considered a famous “undruggable target”, which was a shame, because that protein is very important in non-small-cell lung cancer and many other tumor types. KRAS is important in the signaling cascade in some big cellular proliferation and growth pathways, which are inappropriately set pedal-to-the-metal in these tumor types, so it seemed clear that you could have a real clinical benefit if you could interfere with KRAS function. But how were you going to do that? It has no druglike small-molecule binding sites at all, really – people ran screens against it, hoping to be wrong about that, but nothing really ever came out. It’s a GTPase enzyme, functionally, so years ago, people hoped that enzyme inhibitors could be directed against this protein and others of that type, but it turns out that making useful GTPase inhibitors is extremely difficult.

But over the last few years, a crack in that “target armor” has been exploited. There are some patients who have a mutated form of KRAS where a glycine at position 12 has been turned into a cysteine instead (G12C), and Cys residues can (in theory) be targeted with covalent drug compounds. That’s not a residue at the active site, but rather at an allosteric one that can still influence the protein’s activity when messed around with. That has led to a great deal of work in both academia and in the biopharma industry, screening for such inhibitors and developing them into actual drug candidates, and that idea has actually worked out. Both Amgen (AMGN) and Mirati (MRTX) have actual drugs that work by this route (Amgen’s is already approved). If you look at those structures, you can see the Michael addition acceptors hanging off them, the business end “covalent warheads”. Realizing this goal has been a triumph of drug discovery against a target that had been defying all efforts for decades. Given the direction I’m going to take this post from this point on, I really do want to stop and acknowledge this.

There are very few unmixed triumphs in this business, though. It was clear from the outset that such drugs were only going to offer hope to the subset of patients with the G12C mutation, but it was also clear that that’s a heck of a lot better than sitting back and watching them die. A second problem – and it’s a big one – is the same thing that bedevils oncology therapies in general: these are fast-dividing cells that are often genetically unstable, so they throw off mutations constantly. Every time you come in with an exquisitely targeted anticancer drug, you always have to wonder how quickly the tumors are just going to mutate right around your sharpened spear point. And that is indeed the case with KRAS G12C – there is unfortunately a long list of ways that tumor cells can slip past this therapy, and they show up rather quickly.

The traditional answer to this problem is, of course, combination therapy. It’s the same in infectious disease treatment, because dealing with bacteria and with viruses is in some ways rather similar to attacking tumor cells – in each case, you’re up against opponents that reproduce quickly and with plenty of mutations. In the same way that successful antiviral therapies are almost invariably combination cocktail regimes, oncology has largely moved past single-agent therapies. The question isn’t whether your new anti-cancer agent is going to be part of some combination therapy; it’s which one of the zillions of possibilities to use. That is not an easy question to answer in the clinic. But the development of resistance against the existing KRAS G12C drugs means that it’s the only real way forward.

Amgen has now reported results of a trial of their drug as a combination therapy with two of the big immuno-oncology drugs (Keytruda and Tecentriq). That’s a good place to start – these are two totally different mechanisms, and it’s less likely that a tumor cell could mutate its way past both of these attacks at the same time. Unfortunately, the readout was not a happy one (Reuters here, and STAT behind a paywall). This was in 58 patients with KRAS-positive lung cancer, and the first problem is that the number of patients who showed tumor shrinkage in the combinations was no better than in the standalone KRAS therapy trials – in fact, it appears to be somewhat worse. The second problem, a nasty one, is that Amgen saw clear signs of liver toxicity under these conditions. Many patients dropped out of the trials for that reason, which had to have hurt the statistics. And it’s not clear what the liver tox is, really – the company says that biopsy sample showed liver inflammation with a lot of immune cells infiltrating the tissue, and no one really knows what’s setting that response off. Amgen is trying to go forward with a lower dose of the KRAS drug, dosing it first and then bringing in the checkpoint inhibitors (a dosing protocol that seemed to lead to fewer liver signals). But none of the treatment groups this time had the responses that you’d need for a viable combination therapy, so they have to hope that they can get the numbers up with these workarounds. Will Mirati’s drug show similar problems? Put your money down before they spin the wheel, rien ne va plus.

It’s still early. There are a lot of things that still need to be tried, and you have to think that some of them are going to benefit some patients. But this isn’t what anyone had envisioned for the long-awaited advent of KRAS-directed therapies, either. It’s hard, hard work, and breakthroughs are few. David Reese, Amgen’s R&D head, told STAT that “Drug development is a daily exercise in humility”, and no one who’s tried doing it can disagree.

Disclosure: None

Original Post

Editor’s Note: The summary bullets for this article were chosen by Seeking Alpha editors.