Surveillance Should Be Done Consistently

Roughly one-fifth of all human cancers arise in the gastrointestinal tract. Worldwide, over 3.5 million patients are newly diagnosed with GI tract cancer each year, of whom 310,000 are in the U.S. alone. These malignancies show marked gender, racial and socioeconomic differences in incidence. For instance, esophageal and gastric adenocarcinomas occur more frequently in U.S. males than females. In both men and women, gastric cancer is more frequent in people of African-American, Hispanic and Asian descent than in whites. An opposite pattern exists for esophageal adenocarcinoma.

These tumors all have precursors that exist for a long time, and are eligible for screening, intervention and surveillance. As a result, we as gastroenterologists are likely more than any other medical specialty accustomed to diagnose and manage patients with premalignant lesions. Our approach to different conditions is, however, inconsistent. We first focused on the stomach because gastric cancer was common, relatively easy to visualize by fluoroscopy and endoscopy, and we had non-invasive screening tools. Studies in the 1960s and 1970s already showed that atrophic gastritis and intestinal metaplasia were precursors to gastric adenocarcinoma, with progression to cancer annually occurring in an average one in 200 patients with more advanced atrophy and metaplasia. Recently, very large population studies confirmed the association between atrophy, metaplasia and cancer.1 The risk for cancer parallels the severity and extent of intestinal metaplasia, with the highest risk in individuals with marked intestinal metaplasia affecting both antrum and corpus.

In later years, the interest shifted towards other gastrointestinal malignancies, in particular esophageal and colorectal adenocarcinoma. In Western countries, there were good reasons for this shift. The incidence of gastric cancer declined while the incidence of colorectal and esophageal adenocarcinoma markedly rose. Similar to premalignant gastric lesions, the incidence of esophageal adenocarcinoma in patients with Barrett’s esophagus was initially reported to be approximately one in 200 patients per year. Later population studies consistently reported incidences around one in 800 patients per year. Studies have shown that these risks differ depending on the characteristics of the baseline lesion. While esophageal adenocarcinoma occurs only rarely in patients with short segment and nondysplastic Barrett segments, the incidence is markedly higher in patients with dysplastic Barrett’s mucosa. Guidelines therefore stratify patients according to these characteristics, but nevertheless recommend surveillance for patients with non-dysplastic Barrett’s mucosa.

The inconsistency in our clinical approach to individual patients remains striking.

The story for colorectal cancer is similar. The prevalence of adenomatous and serrated polyps and the resulting incidence of colorectal cancer are much higher than for esophageal and gastric lesions. However, colorectal polyps can be removed. This significantly reduces the risk for colorectal cancer. A large U.S. study followed 314,000 individuals after screening colonoscopy.2 The risk of colorectal cancer within 10 years after adequate colonoscopy approximated one in 2,000 subjects per year. These results and those of similar studies strongly support screening colonoscopy and polyp removal.

This explains why we devote so much of our clinical time and our interest at meetings such as the Digestive Disease Week® to esophageal and colorectal neoplasia. We focus on imaging techniques, endoscopic treatment, quality assurance and guideline development, to name a few. We also focus on specific subgroups at increased risk for cancer such as patients with inflammatory bowel disease. However, the inconsistency in our clinical approach to individual patients remains striking. Let us assume that our Monday morning outpatient clinic includes four patients in their fifties. The first is a woman with a 10-year history of now quiescent colitis. The second is a male with a 3 cm non-dysplastic Barrett’s segment. The third is a female in whom you removed two incipient colorectal adenomas at screening colonoscopy. The fourth is an H. pylori-positive man with dyspepsia in whom endoscopy was normal. We will likely follow guidelines and recommend endoscopic surveillance in the first three patients. When each patient asks about cancer risks, we will reassuringly say that the risk of cancer in the coming years is very small, likely less than one in 500 patients per year, and that they should thus not be worried. We might add that we used high-resolution equipment, we’re trained to detect lesions, spend much effort on quality assurance and follow guidelines. To the fourth patient, we will likely prescribe H. pylori eradication therapy, and remark that the endoscopy showed no macroscopic lesions. However, even though we used the same equipment, we most likely did not use its full potential to scrutinize the gastric mucosa for mucosal abnormalities, like we scrutinized the esophagus and colon in our other patients. As a consequence, if we took biopsies, they were not targeted but random (just consider the idea of doing the same in the colon… ). If these biopsies reveal atrophic gastritis and intestinal metaplasia, we do not risk stratify our patient and do not recommend surveillance even when atrophy and metaplasia appear severe and affect both antrum and corpus. By the time we end our clinic, we will have used our maximal potential in terms of early diagnosis and prevention for three patients, but not for the fourth. This explains why the rate of gastric cancer after a negative gastroscopy remains at a level that we find completely unacceptable for colorectal cancer after colonoscopy.3 In our defense, we can say that we were never properly trained to endoscopically visualize and map gastric atrophy and metaplasia, that we need more data to risk stratify individual patients and determine optimal surveillance strategies, and that there is a need for further guidelines on diagnosis and management of premalignant gastric lesions.

We can conclude that atrophic gastritis and intestinal metaplasia are precursor lesions to invasive gastric cancer.4 Atrophy and metaplasia can be visualized with current standard endoscopic equipment, which also allows for targeted biopsy sampling. Patients with advanced atrophy and metaplasia affecting antrum and corpus have similar cancer risks as others in our practice, and should thus be managed accordingly as already recommended by some international guidelines.5,6 Surveillance is recommended at a three-year interval for subjects with marked intestinal metaplasia in both antrum and corpus. Patients with low- and high-grade dysplasia should receive endoscopic follow-up after 12 and six months, respectively, combined with resection of visible endoscopic lesions.

M. Brian Fennerty, MD, AGAF, provides a different view on surveillance.

Dr. Kuipers has no conflicts to disclose.

1. de Vries, A.C., van Grieken, N.C., Looman, C.W. et al, Gastric cancer risk in patients with premalignant
gastric lesions: a nationwide cohort study in the Netherlands.
Gastroenterology. 2008;134:945–52.

2. Corley, D.A., Jensen, C.D., Marks, A.R. et al, Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med. 2014;370:1298–306.
3. Pimenta-Melo, A.R., Monteiro-Soares, M., Libânio, D. et al, Missing rate for gastric cancer during upper gastrointestinal endoscopy: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol. 2016;28:1041–9.
4. Rugge, M., Genta, R.M., Di Mario, F. et al, Gastric Cancer as Preventable Disease. Clin Gastroenterol Hepatol. 2017; Epub ahead of print.
5. Dinis-Ribeiro, M., Areia, M., de Vries, A.C. et al,Management of precancerous conditions and lesions in the stomach (MAPS): guideline from the European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter Study Group (EHSG), European Society of Pathology (ESP), and the Sociedade Portuguesa. Endoscopy. 2012;44:74–94.
6. Sugano, K., Tack, J., Kuipers, E.J. et al, Kyoto global consensus report on management of Helicobacter pylori gastritis. Gut. 2015;64:1353–67.


  • I agree with Ernst. GIM should be treated analogously to BE. Gastric cancer is just as common and deadly as esophageal cancer in the USA.

    • Completely agree with Ernst Kuipers and Stephen Meltzer except that gastric cancer is actually MORE common than esophageal cancer in the USA.

  • Excellent debate! Stratification incorporating other factors such as ethnic/racial background and H pylori oncologic potential may further help stratify risk among individuals with GIM. Research incorporating ancestry of the individual and H pylori subtype as additional modifiers for the progression of gastric intestinal metaplasia is warrant.

Join the discussion

Your email address will not be published. Required fields are marked *