Malignant transformation of benign epithelial gastric polyps

To compare the malignant potential of hyperplastic polyps and adenomas in relation to different histological classifications and to try to follow the natural history of the BEGP-carcinoma sequence.

During a 13-yr period (1981-1993), 811 BEGP were discovered in 432 patients in consecutive esophagogastroscopic examinations in our department. Adequate endoscopic biopsies or polypectomy specimens were histologically diagnosed as hyperplastic polyp in 751 (92.6%) and adenoma in 60 (7.4%) of the lesions, according to WHO classification. Hyperplastic lesions were further divided into two subgroups: 268 were polypoid foveolar hyperplasia (FH) and 483 were typical hyperplastic polyps (HP), according to Elster's classification. Special attention was paid to focal malignancy at the first examination or malignant transformation of BEGP during follow-up. Ninety-six patients with 220 BEGP were followed for 1-11 yr, with an average of 2 yr and 8 months.

According to Elster's classification, there were 10/483 (2.1%) HP and 6/60 (10.0%) adenomas with focal carcinoma. Moreover, in 19/265 patients (7.1%) with HP and in 4/30 patients (13.3%) with adenomas, carcinoma was found elsewhere in the stomach. During our follow-up, 5/131 HP showed different steps of histological transformation: focal intestinal metaplasia in two, focal dysplasia in one, and focal carcinoma in two of them, which is 1.6%, 0.8%, and 1.6%, respectively. In 1/23 adenomas, focal carcinoma developed after 1 yr of observation (4.3%). Separate gastric carcinomas developed outside polyps during follow-up in 2/58 (3.5%) patients with hyperplastic polyps only. None of those with FH had focal carcinoma either at primary biopsy or during long-term observation. All proportions concerning gradual transformation of hyperplastic polyps classified as only one group according to WHO were lower.

In contrast to hitherto existing opinions, our results give support to the idea that gastric HP, like adenomas, are susceptible to malignant transformation. It seems sensible to separate a subgroup of FH from HP, since FH have no malignant potential until they change their histology to HP. The treatment of FH and HP as one group is the main reason why the malignant potential of hyperplastic polyps is still underestimated.