The treatment landscape for gastric cancer is evolving. The use of biomarkers and implementation of testing to improve outcome later on in the course of treatment are being more important.
The treatment landscape for gastric cancer is evolving. The use of biomarkers and implementation of testing to improve outcome later on in the course of treatment are being more important. Daniel Catenucci, MD, associate professor of Medicine, and director, GI Oncology Program at The University of Chicago Medicine in Chicago, IL discussed this topic during a virtual Targeted Oncology Case-Based Roundtable event.
CATENACCI: This has been a very dynamic time [in gastric cancer] with a lot of new changes with data that have come up.
HUBBARD: I think the most important factor—although we consider all the biologic testing—is going to be performance status and goals of the patient when you’re in the palliative setting. With that being said, molecular testing is extremely important for the management of metastatic gastric cancer. We’re going to be checking things like HER2 status, microsatellite instability [MSI], PD-L1 expression with a combined positive score [CPS]. We’re also looking for other things like NTRK fusions. I like to send a complete next-generation sequencing [NGS] panel if I can, to get all this information, because there may be other trials like an FGFR inhibitor trial or something similar that may open up opportunities. But at a bare minimum, I would say HER2, MSI, and PD-L1, for sure.
CATENACCI: I think you stated it right. The bare minimum would be, these days, MSI testing, PD-L1 testing, and HER2 testing. Then, if you’re getting in a panel, whether it’s a tumor-based NGS panel or a circulating tumor DNA [ctDNA] panel, you’re inherently going to get a lot of other things including pan-tumor approvals like NTRK fusion and other things now emerging like FGFR2 amplification.
CHOWDHARY: Those molecular testing variables that were talked about is what we order. We are looking at PD-L1 based on CheckMate 649 [NCT02872116]. The dose levels become even more important. We look at DNA mismatch repair, HER2, and NTRK. These are some of the main molecular markers that we are looking for. The PD-L1 is going to be important because if they have a proficient MMR and are HER2 negative, then based on CheckMate 649, we are going to be looking at chemotherapy and immunotherapy. That’s a paradigm shift that we are looking at. I have a few patients right now—there’s 1 squamous cell carcinoma and 2 adenocarcinoma cases and they’re all starting out with chemotherapy and immunotherapy in combination. With 1 of them, there has been some pushback from the insurances. But those are the molecular markers that are really important to inform how we choose the next regimen for patients.
CATENACCI: What are you getting pushback on right now from insurance?
CHOWDHARY: For the patients with adenocarcinoma, not as much, but I have a patient who had squamous [cell histology] stage IV esophageal cancer. I wanted to use pembrolizumab [Keytruda] plus a platinum therapy drug. I had to do a peer-to-peer [consultation] and the physician on the other line said the patient’s PD-L1 is only 1% or 2%. He was giving me a lot of pushback and I was very surprised by that. I was telling him that we have data with pembrolizumab based on KEYNOTE-590 [NCT03189719]. We have very clear data now that regardless of PD-L1 level...[but] espicially if they have expression of 1% or higher, pembrolizumab with chemotherapy is a very important regimen now.¹ But I’m still in the appeal process. That patient started cycle 1 today with FOLFOX [folinic acid, fluorouracil, oxaliplatin]. I explained to him today that I’m still in the appeal process of trying to get pembrolizumab with cycle 2.
CATENACCI: It’s a bit outside the scope of gastric cancer but still in the realm here in terms of KEYNOTE-590, which was more so in esophageal cancer. What are your thoughts on a new diagnosis of adenocarcinoma, stage IV with CPS of 1? What are you doing for those patients these days?
HUBBARD: [I have] run into similar issues. We’d like to give the pembrolizumab with chemotherapy, but sometimes we are limited and they make us wait until second line to give the pembrolizumab, which we’re anxious to give in the first line. So a lot of times I’ll do FOLFOX for about 8 cycles until they have neuropathy and then I’ll consider them intolerant to oxaliplatin and move to the pembrolizumab quickly.
CATENACCI: What are you doing for first-line treatment in patients just given diagnoses who are MSI high?
HUBBARD: I usually try to use pembrolizumab first line. Although, technically, the FDA approval is for at least the second line. I suspect that it’s going to be like colorectal cancer, where we’re going to want to use it first line versus chemotherapy, but until we have those data, sometimes we’re stuck. As long as we’re talking about MSI high, [I will] also mention that tumor mutational burden [TMB] is another factor that we’ll look for on the NGS panel due to the recent pan-approval for TMB-high patients.
CHOWDHARY: For patients with adenocarcinoma, I have used FLOT [fluorouracil, leucovorin, oxaliplatin, and docetaxel] a couple of times in very healthy patients in the stage IV setting. But I was wondering whether that’s some-thing that you [all] have ever considered using. FOLFOX has been my go-to for most of them, but I wanted to get your thoughts on that as well.
HUBBARD: If I have a really young patient and they have good performance status, I’ll offer FLOT to them with the understanding that it really hasn’t been tested in this specific situation—particularly if I need a response. But the majority of our patients have poorer performance status and are older. In that case, when I’m in the palliative setting, I’m thinking more about quality of life. FLOT can be a pretty tough regimen for people. I’m pretty judicious with whom I give it to.
CATENACCI: KEYNOTE-590 was a study that enrolled patients with esophageal squamous cell carcinoma, esophageal adenocarcinoma, and gastroesophageal junction [GEJ] adenocarcinoma,1 as opposed to CheckMate 649, which was gastric adenocarcinoma, GEJ adenocarcinoma, and esophageal adenocarcinoma but no squamous cell disease.2 So there was overlap in those studies with GEJ and esophageal cancer; those patients were eligible for either study. Both of those studies, KEYNOTE-590 and CheckMate 649, overlap in their eligibility.
Their primary end point was in all-comers, not by PD-L1 score. But if you look at the subgroup analysis, the benefit appears to be in patients with PD-L1 CPS of 10 or higher. In the National Comprehensive Cancer Network [NCCN] guidelines, [both] are now currently listed for esophageal adenocarcinoma and squamous cell histology [if the PD-L1 CPS] is 10 or higher with chemotherapy for now.3 ...The NCCN guideline recommendations are CPS of 5 or higher for gastroesophageal for nivolumab in adenocarcinoma, and 10 or higher for esophageal cancers that are adenocarcinoma or squamous cell carcinoma.
[Editor's note: The FDA has approved nivolumab combined with chemotherapy for patients with metastatic gastric cancer and esophageal adenocarcinoma as of April 16, 2021.⁴ Keeping pembrolizumab's accelerated approval was voted against by the FDA's Oncologic Drugs Advisory Committee for patients with PD-L1–positive recurrent or advanced gastric or GEJ adenocarcinoma who had received 2 or more lines of therapy, but the FDA decision has not been made yet.⁵]
CATENACCI: Based on that, does anyone have a strong opinion as to whether PD-L1 should be tested at all? Should we be giving it to everybody? Or does anyone have the other opinion that we should be really stringent and make it a CPS of 10 instead of 5?
HUBBARD: I think CheckMate 649 is a game changer. I’m interested to see whether the FDA is going to make CPS 1 versus 5. I envision a lot more people being able to get pembrolizumab with chemotherapy first line. Most of us will use FOLFOX first line anyway, so I think it’s a nice pairing. I’m anxious to see whether we’re going to have to go by CPS at all, whether they’ll make us do CPS of 5.
BADAMI: I agree with everything that’s been said already. I test everybody. Ideally, I would like to give [immunotherapy] to everybody up front. But for right now, I’m just using the CPS cutoff of 5 or higher. I’m showing people data when I try to get insurance approval, and the NCCN guidelines. So at least if it’s an NCCN [recommendation], I won’t get insurance pushback right now.
CATENACCI: If CPS is above 5, I think you’ll be fine. It’s the ones less than 5 that I think you’re going to get push-back on like Dr Chowdhary was saying. When and where do you get PD-L1 testing done?
BADAMI: For all of my metastatic patients, I send NGS on, whether it’s FoundationOne or Tempus. I send it up front. But our pathologist will add it on if we ask to test the tumor itself. If there’s enough specimen, they’ll test it on the tumor. Otherwise, we’ll revert to ctDNA. But I try to get all the information up front.
CATENACCI: Anyone else have their pathologists doing PD-L1 internally for them? Or is everyone relying on an outside vendor, like FoundationOne or Tempus?
HUANG: I do a lot of in-house NGS panels. I think we have everything important included in our panel. We have a faster turnover, so if I need results fast I would do in-house. If I can start people on first line and wait for other options down the line, I’d probably send for FoundationOne or Tempus.
CATENACCI: Let’s clarify all of this first, because PD-L1 testing is not going to be given to you in an NGS panel. It’s the chemistry with various antibodies. Nivolumab is paired with one and pembrolizumab is paired with a different one. They have different properties, and they have different cutoffs for what’s positive. That’s where the question comes from [regarding whether we] are relying on our pathologists to give that to us or we are clicking the box on the Tempus form and asking for NGS and the PD-L1 results.
HUANG: Basically, when I’m doing testing, I do both together. If I do in-house, I’ll do both in-house. If I send out, I’ll send both out. Sometimes I need it immediately. If I really need to know the results to put orders and get insurance approval, I may turn to our pathologists for help.
KOZLOFF: Do you do it in-house or do you send it to Tempus?
CATENACCI: That’s a good question. Essentially, if there’s a test that’s internal, like HER2, we do it internally because it’s reflexive and it’s standardized. They’re using kits that are FDA approved and companion diagnostics for HER2. These newer PD-L1 antibodies and scoring systems are very specific with specific platforms. There are many studies that show if you tinker with it and you’re doing your internal laboratory-developed tests, you can get different results. For example, we don’t have the [PD-L1 immuno-histochemistry] 22C3 [pharmDx kit using the] Dako platform from Merck to do that assay [at my institution]. So they have antibodies to score “PD-L1” but I don’t know what that means in terms of its validation. This was not what was studied prospectively in these phase 3 trials.
To answer your question, we send it out until we can get that panel properly done internally. Now the other complicating factor is that the CheckMate 649 study with nivolumab has a different antibody and a different platform. We expect a different companion diagnostic approval from the FDA for that one. The incidents of CPS 5 or higher in the study was 60% of the patients. But when we do CPS 5 or higher with the Merck antibody test with pembrolizumab, it’s about 30% of the patients.
We’re not sure which ones to use, so this is becoming a little messy in a way. But for now, we’re using the 22C3 antibody [test], because that’s what we’ve been using all these years. This is the first time now that the Bristol Myers Squibb nivolumab assay is coming on the scene. The next months are going to be interesting [regarding] how that interplays.
HUBBARD: The Mayo Clinic in-house testing does this for us. I don’t really have to make the decision too much. They figure it out and do it, so we do an in-house one, usually.
HUBBARD: If I use FOLFOX in the first-line setting, I usually scan after every 4 cycles, just knowing that the progression-free survival isn’t that long. I’m also going by patients’ symptoms. Sometimes changes in the tumors are harder to anticipate, especially if we’re using checkpoint inhibitors. We have to worry about pseudoprogression. So I’m going by their symptoms; if tumors look a little bigger but patients are doing great and feeling great and I have them on immunotherapy, I might continue that in the first line and make sure it’s to progression instead of pseudo-progression. My main triggers are going to be the scans and how they’re feeling. We’ve all treated these patients where it’s obvious when they’re progressing because they’re feeling crummy. That’s typically a sign to get a scan and move onto the next line.
KALAKOTA: My [triggers are] pretty similar in that sense. Definitely symptoms because most of these patients at least are usually symptomatic up front. Usually with using FOLFOX, we’re seeing them frequently, so we have a chance to check for toxicities. But immediately after the first cycle, we usually have a nurse practitioner who checks in on them. That’s where telehealth has been a bit more helpful, where we’re able to check in on them for toxicities, especially early on.
Similarly, I usually scan after about 3 to 4 cycles and use that to assess disease progression. To change to second-line therapy, definitely [once we see] both symptoms and then radiographical progression because the more we’re going into further lines of therapy, the less effective and the less likely we’re going to have an impact. I try to keep that first-line therapy going and, with small changes, I’ll give it a bit longer—maybe do some short interval imaging to make sure before I would switch to second line, especially if their symptoms are well controlled.
INOSHITA: Essentially [we look at] symptoms and do x-ray study or CT scan as needed. I generally don’t do every 3 or 4 weeks. I just go by symptoms.
CATENACCI: How often do you see the patient?
INOSHITA: I use cancer antigen, but in gastric cancer it’s not as useful as, say, colon cancer. But I still use it.
KOZLOFF: I follow the tumor markers, but I don’t make decisions on the tumor markers. The decision I may make is to get a scan earlier on the tumor marker.
CATENACCI: Only about half of patients go on to next line [of] therapy with this disease. Despite knowing that there are options that have improved survival advantages, patients aren’t getting there. One of the thoughts is that some patients may be on therapy too long and you’re waiting too long to make the switch. That’s one concept. Many times, patient scans don’t show overt progression, but they’re progressing. Because they may have peritoneal-only disease, as an example. You’re not picking it up on the scan, but they’re progressing.
Many of you said symptoms are important too. I use tumor markers. I use carcinoembryonic antigen and cancer antigen 19-9. I check them, at minimum, once a month, because very often those are going up long before any changes in symptoms or scans. I tend to do a scan every 2 months on my patients generally with FOLFOX. Four doses, then scan and reassess them. But if I’m seeing the tumor markers skyrocketing, I might scan them after 3 [weeks] to pick it up early and switch so that you don’t miss your window before they fall off [with therapy] because then you can’t treat them anymore. That’s how we deal with it to try and get as many patients on to effective later-line therapies. There are more now that we just don’t use enough, I think.
1. Kato K, Shah MA, Enzinger PC, et al. LBA8_PR - Pembrolizumab plus chemotherapy versus chemotherapy as first-line therapy in patients with advanced esophageal cancer: The phase 3 KEYNOTE-590 study. Ann Oncol. 2020;31(suppl 4):S1142-S1215. doi:10.1016/annonc/annonc325
2. NCCN. Clinical Practice Guidelines in Oncology. Gastric cancer, version 2.2021. Accessed March 17, 2021. https://bit.ly/3qXLusW
3. Tabernero J, Alsina M, Shitara K, et al. Health-related quality of life associated with trifluridine/tipiracil in heavily pretreated metastatic gastric cancer: results from TAGS. Gastric Cancer. 2020;23(4):689-698. doi:10.1007/s10120-020-01053-9
4. Fuchs CS, Doi T, Jang RW, et al. Safety and efficacy of pembrolizumab monotherapy in patients with previously treated advanced gastric and gastroesophageal junction cancer: phase 2 clinical KEYNOTE-059 trial. JAMA Oncol. 2018;4(5):e180013. doi:10.1001/jamaoncol.2018.0013
5. Moehler M, Shitara K, Garrido M, et al. Nivolumab (nivo) plus chemotherapy (chemo) versus chemo as first-line (1L) treatment for advanced gastric cancer/gastroesophageal junction cancer (GC/GEJC)/esophageal adenocarcinoma (EAC): first results of the CheckMate 649 study. Ann Oncol. 2020;31(suppl 4):S1191. doi:10.1016/j.annonc.2020.08.2296
6. Satoh T, Xu RH, Chung HC, et al. Lapatinib plus paclitaxel versus paclitaxel alone in the second-line treatment of HER2-amplified advanced gastric cancer in Asian populations: TyTAN--a randomized, phase III study. J Clin Oncol. 2014;32(19):2039-2049. doi:10.1200/JCO.2013.53.6136