Dr. Gregory Riely:
Welcome to The ASCO Post Roundtable Series, Clinical Perspectives on KRAS Mutant Lung Cancer. I'm Dr. Greg Riley. I'm a medical oncologist at Memorial Sloan-Kettering Cancer Center. Joining me today are two of my colleagues. First I'll ask Dr. Heist to introduce yourself.
Dr. Rebecca Heist:
Hi, I am Dr. Becca Heist. I'm a medical oncologist at Mass General Hospital in Boston.
Dr. Riely:
And Dr. Sabari.
Dr. Joshua Sabari:
Hey, I am Josh Sabari. I'm a thoracic medical oncologist at NYU Langone Health Perlmutter Cancer Center in New York. Thanks for having me.
Dr. Riely:
Welcome to you both. Today we'll be discussing the treatment and management of KRAS-mutant non-small cell lung cancer with three patient case studies. Our first installment will focus on the first-line treatment of KRAS-mutant lung cancer.
Our first case is a 58-year-old woman who presents with bilateral lung masses. Past medical history includes hypertension and diabetes. The patient has an initial biopsy, we're going to say this biopsy happened at an outside hospital, and the biopsy reads as non–small cell carcinoma. With a biopsy showing non–small cell carcinoma, Dr. Sabari, what else do you want to know from your pathologist?
Dr. Sabari:
Yeah, first off, we hate these non–small carcinoma NOS. I'd like to know the IHC. Is this a squamous histology? Is this an adenocarcinoma? So I'd love to see the immunohistochemistry for Napsin A, TTF-1. I'd also love to see the PD-L1 expression as well as molecular testing. So it really depends on full broad panel next generation sequencing. But you're right, this NOS non–small cell lung cancer sort of leaves you guessing.
Dr. Riely:
Yeah, it's interesting. Even in this era of things you alluded to like PD-L1 and molecular testing, just the first step of knowing the tumor histology is still critical to lots of our early decision making. So is this a squamous cell? Is this an adeno? Is this large-cell neuroendocrine? All these things are critically important in our thought process. Anything else you want to learn, Dr. Heist, early on in the case?
Dr. Heist:
No, I agree with what's been said. I think the histology makes a big difference when we're choosing our chemotherapies, in particular. It makes a big difference. And then obviously the PD-L1 is a major factor when we think about the role of immunotherapy alone or in combination. And then the tumor genetics. I think that's really critically important. I do think the IHC is important, but I'll just put a plug in for saying let's not do tons of IHC. TTF-1 and Napsin A are great, but then if we are seeing 20 other IHCs worry because we don't want to spend all of our tissue on that and I think it's important to make sure that we have enough to do that tumor NGS that you talked about.
Dr. Riely:
Yeah, absolutely. That is certainly something I think we see a lot less of today, but we've certainly seen over the years these large panels of immunohistochemistry that can take up more unstained slides than we would ever want to lose.
Dr. Heist:
I think one other thing that we sometimes think about is TMB, but to be perfectly honest, I really rely on the PD-L1 and the tumor genetics and not so much the TMB, but that is something that sometimes we will see reported and it's nice to know. But to me that's a little bit less critical to have than the histology and the PD-L1 and the molecular genetics.
Dr. Riely:
All right, great. For this case, after we push a little harder, the pathologist reports this as an adenocarcinoma and they make that decision based on morphology and Napsin A positivity. The PD-L1 testing is done, and the tumor shows 30% PD-L1 expression from a tumor proportion score. Unfortunately, pathology calls back and says they can't send for tumor DNA sequencing because there's insufficient material left. Dr. Heist, what do you do when you get that call from the pathology lab?
Dr. Heist:
That's always a disappointment. Very often if I'm seeing people as we're getting the biopsy things, I actually am very often sending liquid samples from the outset to get the liquid NGS done. We know that there are limitations to what we can see on liquid. For example, if there's a relatively low burden of disease, tumors may not be shedding, but at the same time it can sometimes, in situations like this, be a very nice complementary test where you can detect things and then you kind of have your answer and can move on.
Dr. Riely:
Yeah, I think that question of false negativity is an important one. Dr. Sabari, when you get a result back and maybe it doesn't show a particular mutation, a particular driver, how do you know if this is a true negative or a false negative?
Dr. Sabari:
Yeah, it's a great question and we often say the test is negative for plasma NGS, but I talk to patients about fishing, you go fishing and you don't catch a fish, that doesn't mean there are no fish in the ocean. And very similar with plasma next-generation sequencing. If we test or assay for a driver alteration and we don't find it, we may have missed it. So we do know that there are high false-negative rates, and as Becca mentioned, really, depending on tissue, NGS is the gold standard. Greg, would you rebiopsy this patient? Is there a role for that here or does your pretest probability for a driver have to be much higher? Or a driver that's actionable, should I say?
Dr. Riely:
I think that there's definitely a role for rebiopsy. I have to admit, I don't typically do rebiopsy upfront until I've exhausted my plasma ctDNA analyses options. So if I get a completely negative plasma DNA, then I think I'm going to try to rebiopsy because I do think it's critical to identify any mutations present at the beginning. Now, clearly whether we wait for that biopsy to happen and wait for subsequent results, that might be a little bit of a different question, but I definitely want to get the ball rolling on identifying tumor mutations. If I can't get it off of plasma, then I think rebiopsy is certainly something to consider.
All right, for this patient, we send off the circulating tumor DNA. It comes back and it shows us that the patient has a KRAS G12D mutation. And in addition to the KRAS G12D, we identified TP53 and STK11 mutations. We've now got enough information. We've got a patient with metastatic lung cancer. I didn't mention it, but the patient has a brain MRI without any evidence of brain metastases, but they do have bilateral lung masses as well as liver metastases. So diffuse metastatic disease, KRAS G12D, PD-L1 is 30%. Dr. Heist, I'll start with you. What's your choice of first-line therapy for this patient with KRAS G12D, PD-L1 30%?
Dr. Heist:
Yeah, my choice here, and the KRAS G12D is interesting because there are definitely clinical trials we can talk about, but certainly in the first-line setting here I would still be doing, my personal first choice is the KEYNOTE-189 regimen. So lots of data showing that the addition of immunotherapy to chemotherapy improves outcomes overall. And so here with the PD-L1 of 30%, assuming this person has a good performance status and can tolerate the combination of chemo and IO, I would really be looking at that combination as the first line of treatment.
Dr. Riely:
Great. So for you, carboplatin, pemetrexed, pembrolizumab. Dr. Sabari, what's your take on this one?
Dr. Sabari:
I hate to agree, but I think I have to agree here with Dr. Heist that I would use carbo, pemetrexed, and pembrolizumab, the KEYNOTE-189. That being said, STK11 is a high-risk alteration. So there is a lot of data coming and retrospective data potentially supporting dual immune checkpoint inhibitor. Greg, I'll throw that to you momentarily. But I agree, a KRAS G12D mutation, very intriguing. We know that 30% of our patients will have KRAS alterations. We're always looking for that KRAS G12C because that's actionable in 2024 in the later-line setting. But unfortunately, G12D, quite rare in lung cancer and not actionable yet. So Greg, how would you treat this dual checkpoint inhibitor or would you use the KEYNOTE-189 regimen?
Dr. Riely:
Yeah, I think there is a role for dual checkpoint inhibitors in patients, sort of all-comers with any diagnosis of non–small cell lung cancer. I think there's an inclination to use dual checkpoint inhibitors such as durvalumab/tremelimumab along with chemotherapy or ipilimumab/nivolumab along with chemotherapy in a variety of situations, and I think one of those situations is where you think the patient has a bit worse prognosis. We do know that these STK11 and p53 alterations that are going along with the KRAS mutation does predict a worse outcome for these patients.
Now, the challenge here is knowing whether this worse outcome is something you can do something about. I think that both of you suggesting carboplatin, pemetrexed, pembrolizumab tells me that you think that maybe this presence of STK11 isn't enough to justify dual checkpoint inhibition. And I think that's probably reasonable based on today's data. We definitely need to explore whether we can alter the outcomes for these patients with STK11, KEAP1, p53 alterations that happen in concert. So I think for this patient, both your choices of carboplatin, pemetrexed, pembro a reasonable one. That's actually what I did for this particular patient and I think we'll certainly look forward to our options going forward.
Now, Dr. Heist, you mentioned that clinical trials may be an option. Can you describe a little bit about clinical trial options in this space? Are there frontline options? Are they more being explored in the later line?
Dr. Heist:
Specifically for the KRAS G12D, there are several clinical trials ongoing of G12D-specific inhibitors, and those typically would be in later lines after people have received a standard treatment. So it wouldn't be something I offer right up front think people should see chemo-IO in some form or another right up front. But in the future if the person were to progress, I would definitely want to be thinking about G12D-specific inhibitors. There are also other kind of KRAS specific-inhibitors that are pan-KRAS. So a lot of work being done. And again, nothing that's FDA approved right at the moment, but I think in the future for this patient, I think exposure to some of these clinical trial options may be worthwhile to think about.
Dr. Riely:
Great. So both options as G12D-specific drugs, but also pan-RAS drugs that are being explored in this setting. So certainly lots more to come in the world of KRAS. Okay, great.
This has been a good case and so maybe I'll emphasize a few of our key clinical takeaways for this case. I'd say, first, up front testing for all genomic alterations is necessary and we do that, whatever means we have available to us, whether that's tumor tissue or circulating tumor DNA. Next, I'll highlight something we didn't talk too much about, which is KRAS mutations occur in about 20% to 25% of lung adenocarcinomas, and about half of those mutations are KRAS G12C. The other half are other KRAS alterations like KRAS G12D, G12V, and a variety of other alterations that we see. Finally, initial therapy for all KRAS-mutant lung cancer patients today is based on tumor histology and PD-L1 status. We're really not using targeted therapies up front for patients with KRAS-mutant lung cancer today.
This brings us to the end of this case. Please see the other segments for further discussion about the latest research in lung cancer or visit ascopost.com.
