Transcript: Johanna C. Bendell, MD: Manish mentioned molecular profiling and testing. What do you get at baseline? What’s reflexive, and how does this tie in to all this, as we’ve been hearing, from a TCGA [The Cancer Genome Atlas] analysis?

David H. Ilson, MD, PhD: I have 2 additional, quick comments about staging. I think for pure gastric cancers we also do staging laparoscopy because we identify up to 25% to 30% of patients who either have positive cytologies or frank peritoneal disease. For gastric cancer, staging laparoscopy is indicated. Also, the EUS [endoscopic ultrasound]. How do we apply that? For T1, T2 node-negative tumors, they would probably go right to surgery. For T3 or node positive, suggested on endoscopic ultrasound, those would be patients for neoadjuvant treatment, which we’ll talk about. And certainly, to risk stratify who gets a laparoscopy prior to treatment would be the T3 or node-positive gastric cancers.

In terms of additional tissue studies at diagnosis, we don’t really have validated molecular treatments for locally advanced disease at this point. But it’s becoming increasingly common for us to do the 3 key tests from the outset, which would include HER2 testing, PD-L1 [programmed death-ligand 1] testing, and DNA mismatch repair protein testing. We want to know HER2 status. We want to know whether or not the patient is MSI [microsatellite instability]—high or stable. And then PD-L1 testing, really more for the treatment of refractory disease. We’ll also talk a little about MSI as a potential prognostic factor for locally advanced disease. Those patients may actually do better. And does adjuvant treatment really contribute to MSI-high tumors?

Johanna C. Bendell, MD: And what about this TCGA analysis dividing the gastric cancers into 4 different types?

David H. Ilson, MD, PhD: I think TCGA has given us a road map, potentially, for drug development because we have identified 4 molecular subsets in gastric cancer. Fifty percent of gastric cancers are genomically unstable, and about 95% of GE [gastroesophageal] junction esophageal cancers fall into this category. These are the receptor-associated tyrosine kinase amplified tumors, TP53 mutation. And the less common subsets: we have genomically stable tumors, which correlate with diffuse gastric cancers, and then a small subset of the MSI-high cancers and EBV [Epstein-Barr virus]—related cancers, which we see more often in the stomach.

There is an Asian classification that differs a little bit, where they actually make the distinction based on MSI-high and stable; then TP53 mutation, and mutant, and wild type; and then whether or not this is an epithelial mesenchymal subtype, which has a particularly poor prognosis. But there is a clear acknowledgement of the importance of testing for MSI. HER2 status really is more of an issue for advanced disease. We’re awaiting results of adjuvant therapies directed toward HER2-positive patients, but this is not an issue yet.

Johanna C. Bendell, MD: Kohei, in Japan, does that change? And I have a couple of specific questions to ask you. With your HER2 testing, do you prefer immunohistochemistry…followed by FISH [fluorescence in situ hybridization], or do you just look at FISH testing? And do you check PD-L1 status in Japan?

Kohei Shitara, MD: For a patient who is at an advanced stage, we can use trastuzumab if the patient is HER2-positive. In clinical practice, we just follow this protocol. That means the immunohistochemistry first, followed by FISH. Recently, NGS [next-generation sequencing] testing is also approved in solid tumors in Japan, and it will be reimbursed very soon. More and more, cancer patients undergo NGS testing if they have advanced disease. I want to mention the previous comment regarding the GE junction tumors.

Until recently, very few patients had GE junction tumors or esophageal adenocarcinoma. But the frequency seems to be increasing, even in Japan. So it is very important to manage these patients. For management, we usually have a conference, especially in the hospital, to discuss esophageal tumors and gastric cancer because the procedure and the management are different, and the surgery is different. Regarding the diagnostic test, as David mentioned, a staging laparoscopy is very important. In Japan, we usually do it for patients with type III or type IV disease, where there is a high frequency of peritoneal dissemination. A PET [positron emission tomography] scan is done only if the patient has suspected metastases.

Transcript Edited for Clarity