Serum pepsinogen in screening for gastric cancer.

To establish a sensitive and efficient screening method for gastric cancer using serum pepsinogen, we investigated the characteristics of serum pepsinogen I and II levels and the I/II ratio and their cut-off points. We found that the pepsinogen I level and the I/II ratio were significantly lower in patients with gastric cancer than in control subjects, especially in patients with cancers of the differentiated type, the elevated type, and the depressed type without ulceration. However, sex, depth of invasion, and location of tumor did not correlate with the pepsinogen levels. A suitable cut-off point in screening for gastric cancer was a pepsinogen I level of less than 50 ng/ml and a I/II ratio of less than 3.0, as determined by receiver operator characteristics curves. The sensitivity, the specificity, and the accuracy of detection for all types of gastric cancer were approximately 55%, 75%, and 72%, respectively. If restricted to cancers of the elevated and the depressed type without ulceration, the sensitivity was approximately 85%, and the specificity and accuracy were approximately 76% and 77%, respectively. These results suggest that, in screening for gastric cancer when using pepsinogen levels and morphological examinations, the suitable cut-off point in regard to specificity is as stated above. However, regarding sensitivity, when the pepsinogen method is used alone, a pepsinogen I level of less than 70 ng/ml and a I/II ratio of less than 3.0 is acceptable.