David H. Ilson, MD, PhD: Just 1 other point about the global incidence is that there’s a shift globally to more proximal cancers of the GE [gastroesophageal] junction and proximal stomach, which are more aggressive, with much worse outcomes, and there is potentially a relation between reflux and Barrett’s esophagus. But in terms of familial risks, we always have to take a family cancer history. We now have clearly defined genetic syndromes that put patients at risk for developing gastric cancer. Probably the most significant is hereditary diffuse gastric cancer, or CDH1-related gastric cancer. Although it’s very uncommon, it has an autosomal dominant inheritance, and patients have up to a 75% to 80% risk of developing gastric cancer by age 40 to 50. They’re multifocal cancers. Once we identify individuals who have this, we have to screen the family and recommend prophylactic gastrectomy. There’s not really appropriate screening for these patients because they develop multifocal diffuse gastric cancers.

Other relevant syndromes: Lynch syndrome is associated with the risk of developing gastric cancer. Lynch families should undergo periodic screening with an upper endoscopy. I’m also an advocate in Lynch patients for prophylactic aspirin, because I think aspirin therapy may potentially reduce the risk of future cancer development. Even BRCA-related breast cancer patients are at increased risk for developing gastric cancer. We don’t routinely recommend screening through an endoscopy in BRCA families, but occasionally you will see gastric cancers developing in these families, and they probably should consider periodic screening. And then the more esoteric polyposis syndromes like juvenile polyposis, for which a screening upper endoscopy is considered. The 1 exception would be a CDH1-hereditary diffuse gastric cancer, for which we recommend prophylactic gastrectomy.

Johanna C. Bendell, MD: Very interesting. I’m going to take you back to the Barrett’s, which you made the comment on, because I think a lot of people wonder about this. You get a patient with Barrett’s esophagus. What’s the risk, and what’s the screening procedure for those patients?

David H. Ilson, MD, PhD: I think there has always been controversy about the appropriate screening. We do screen patients with chronic reflux, and we identify Barrett’s in about 1% of the population. Surveillance endoscopies are typically performed every 2 years because we’re looking for dysplasia. Now even low-grade dysplasia is an indication for radiofrequency ablation. High-grade dysplasia is now treated with radiofrequency ablation and endoscopic resection. But on a positive note, even though Barrett’s is common, only 1% of patients develop cancer over 10 years of surveillance. So we do recommend identifying Barrett’s and putting patients on surveillance, but the actual risk of developing cancer over years of surveillance is relatively low.

Johanna C. Bendell, MD: That’s good news. That’s very good news. And Kohei, you mentioned that in Japan there is more broad-based screening rather than for people at increased risk. How often are patients screened? What age does it start?

Kohei Shitara, MD: From the government, screening is recommended for patients older than age 40.

Johanna C. Bendell, MD: Older than 40 years.

Kohei Shitara, MD: Yeah. The usual procedure is endoscopy, but a contrast study by x-ray is also used. More and more, patients undergo endoscopic screening.

Johanna C. Bendell, MD: It’s not like a colonoscopy.

Kohei Shitara, MD: Yeah.

Johanna C. Bendell, MD: And how often are they getting screened?

Kohei Shitara, MD: Every 2 years.

Johanna C. Bendell, MD: Every 2 years.

Kohei Shitara, MD: Is recommended.

Johanna C. Bendell, MD: Very good. Now we have a patient who comes in and has a gastroesophageal cancer diagnosis. Talk to us, Manish. What’s your initial work-up? What tests need to be done? What tests could be done? What’s your recommendation?

Manish A. Shah, MD: For a gastroesophageal cancer, we typically start with both an endoscopy and a [CT; computed tomography] scan. We need tissue. Like the old adage, tissue is the issue. Tissue is important for making the diagnosis, the subtype of the disease, and for any future molecular testing that we’ll need to do. And then a CT scan is perhaps the most economical and best first step. About half of patients will be diagnosed with more extensive disease, metastatic disease. For those patients, you don’t need any further local testing, like an endoscopic ultrasound or anything like that. But if the CT scan doesn’t show metastatic disease, often an endoscopic ultrasound might be important to determine the extent of local spread of disease, the depth of invasion, and whether nodal disease is present. That may be important for how we approach the patient initially. And then, a PET [positron emission tomography] scan is also important, particularly for esophageal cancer but even for gastric cancer. It can identify about 10% of occult metastatic disease. This is disease that might be in a lymph node that seems to be of normal size, or bone metastases, things like that. That’s your typical screening procedure.

Johanna C. Bendell, MD: We’re hearing so much about locally advanced esophageal cancers versus GE junctional cancers versus gastric cancers. Tell us, when I’m reading that endoscopy report, what am I paying attention to? And how does that help influence what we’re going to do?

Manish A. Shah, MD: Sure. Whereas for metastatic patients the treatment might be overlapping and very similar, the approach to a patient with localized disease can be very different for esophageal tumors versus gastric tumors. And then GE junction tumors: I think it depends on where the predominant disease is. Is it more esophageal, or is it more gastric? The pure type II GE junction tumors could be treated both ways, but people have preferences.

On the endoscopy report, we try to look for the location of the disease. Often they’ll say that there’s a mass located at certain centimeters from the incisors. Typically the GE junction is around 38 cm to 42 cm from the incisors. So tumors that are more proximal to that, so a shorter distance, will be more esophageal, and more distal would be more gastric. That’s not entirely a perfect way to do it, because patients are of different sizes. Sometimes the GE junction is higher or lower.

If you’re lucky, the endoscopy report might say the GE junction is at a certain level or the Z-line is at a certain level. That’s where there’s a curve that’s made in the stomach. But you want to try to find the location. You also want to find other characteristics that may help you manage the patient later. Is it a partially obstructing lesion? Is it not? Is it cratered? Is there a risk for bleeding? Things like that. Those things might help as well for management.

Transcript Edited for Clarity