Suresh S. Ramalingam, MD, FASCO: From that I want to talk about KRAS. This is a target that is common in lung cancer—approximately 25% of adenocarcinoma patients have it. But we’ve not had successful targeted therapy options. Pasi, can you talk a little about KRAS biology and why it’s been hard to find a targeted strategy for that particular subset?

Pasi A. Jänne, MD, PhD: As you mentioned, KRAS is about 20% to 25% of the lung adenocarcinomas, and about half of those are the KRAS G12C mutations, which are most common in lung cancers. It’s been a challenge to find targeted therapies. KRAS is not a kinase like the other molecular targets that we talked about, and finding a drug that can out compete the cellular affinity for GTP [guanosine triphosphate] has been quite difficult. The big breakthrough here came a few years ago with the identification of potentially a drug that can covalently bind KRAS, and that’s why they work for the G12C. Osimertinib is a covalent inhibitor and also binds to cysteine. These new drugs, like the AMG 510, are also covalent inhibitors and binds to cysteine and open the possibility that we will finally be able to have targeted therapies for KRAS-mutant patients.

If you go back and look at all our pie charts that we present, KRAS has been there for a very long period of time, yet it’s only now that we’re able to do something about it. The data that have been presented at the [International Association for the Study of Lung Cancer 2019] WCLC [World Conference on Lung Cancer] meeting on the AMG 510—which is 1 of the KRAS G12C inhibitors—is certainly very exciting. There was approximately a 50% response rate in patients with G12C-mutant, KRAS-mutant cancer. This drug was well tolerated because it’s a highly specific drug and technically shouldn’t have off-target effects. I think it’s really an exciting time for KRAS-mutant lung cancer. This is the big one that, at least in the United States, we see in the clinic. It’s the biggest chunk of the pie, and I think if we can start to make an impact here, it will have real clinical significance, not just for lung cancer but for the other cancers that have KRAS G12C mutations. But lung is really the big one.

Suresh S. Ramalingam, MD, FASCO: Johan, can you talk about how you view the AMG 510. Pasi obviously talked about his level of enthusiasm. What about yours?

Johan F. Vansteenkiste, MD, PhD: One of the nice things is that many people in Europe, as I told you, do sequencing at the DNA level. You usually detect KRAS, so you know at baseline. That’s the first good thing, and you know it was G12C. KRAS has been the problem child for decades, and it’s a big piece of the molecular pie, as just has been said. Any progress there would be very welcome because it will be the case, at least in the western world, for a large proportion of patients—larger than the EGFR fraction. So yes, if they have a response in about half the patients with that drug, they are very early data, but they are most encouraging.

Byoung Chul Cho, MD, PhD: One thing I want to comment on about AMG 510 is that it is targeting only G12C. In Asian countries, we often see non-G12C KRAS-mutant patients because most of KRAS mutation in Asian countries is occurring in never-smoker populations. We often see G12D and G12V KRAS mutations, which cannot be targeted by AMG-510. That is most of our concern in Asian countries.

Pasi A. Jänne, MD, PhD: I’m hoping that the enthusiasm from what we’re seeing so far, targeting G12C, will open up more interest in developing newer therapies for the other KRAS mutants. People have struggled with KRAS targeted therapies for a long period of time, and there’s a nice history of failures. Having successes will drive additional enthusiasm for the other KRAS mutants.

Suresh S. Ramalingam, MD, FASCO: Based on this discussion about the KRAS molecular prevalence and the AMG 510 data, should oncologists be routinely getting patients’ KRAS status moving forward right now, or is that something you would hold off until we have FDA approval?

Pasi A. Jänne, MD, PhD: Well, I think if you do molecular testing as part of a panel, it will come anyway, so that’s a good thing. The second reason to test it would be to potentially enroll patients on to the clinical trials if you have access to those trials, given the preliminary data that we’ve seen so far. But routine, outside that, if you have no access to a trial, then probably not yet. But maybe we’ll be 1 of the next ones to be a target that we test for routinely.

Suresh S. Ramalingam, MD, FASCO: As we wind down the year 2019, is it fair to say that as we think about what are the exciting advances we look forward to in 2020, looking at expanded data sets with AMG 510 and also some of the other KRAS, G12C inhibitors belongs right on the top of the list?

Pasi A. Jänne, MD, PhD: Absolutely. I think we saw more at WCLC than we did at ASCO [American Society of Clinical Oncology] Annual Meeting. I’m sure we’ll continue to see more data. I think it will be seeing what the denominator is, what is the durability of the response, and why some KRAS-mutant patients respond whereas others don’t. All the same questions that we had in other targeted-therapy fields will continue to apply here as well. But I think it’s an exciting time.

Suresh S. Ramalingam, MD, FASCO: The last topic we want to discuss in the realm of targeted therapies is the issue of molecular testing. We saw the results from the B-FAST trial, where they had done peripheral blood-based ctDNA [circulating tumor DNA] NGS [next-generation sequencing] testing and Dr Shirish Gadgeel presented results at the ESMO [European Society for Medical Oncology] Congress. He showed that in over 2200 patients who underwent testing of the blood, they were able to detect ALK in about 5% of patients. These patients were treated with alectinib, and the response rate was 87%. The results from the blood identified a patient subset that derives practically the same amount of benefit that you would have expected from tissue-based testing. Is this 1 more step toward routine incorporation of peripheral blood testing in our frontline algorithm?

Johan F. Vansteenkiste, MD, PhD: In general, we have the policy to try to have as much tissue as possible in as many patients as possible at the baseline. In the European context, and it’s in contrast to some other agents, we use quite a lot of cell blocks because about one-third of the diagnosis comes from EBUS [endobronchial ultrasound] and other samples. If you take into account that you use these samples and there are no sufficient data to say you can do a reliable immuno-testing, you can do reliable targeted-therapy testing on these types of samples. Then if you take these cell blocks together with the biopsies, the majority of the patients will still have a biopsy at baseline. But we discussed it for EGFR. Obviously, it’s welcome in patients in whom it’s difficult to get the biopsy. If you can do something reliable on plasma that leads you to a targeted therapy, it is a plus.

Transcript Edited for Clarity