Gastric Adenocarcinoma (Intestinal vs Diffuse)

Gastric adenocarcinoma comes in two flavors on USMLE Step 1, and the exam exploits the fact that most students blur them together. Intestinal-type follows a predictable progression — chronic atrophic gastritis → intestinal metaplasia → dysplasia → cancer — driven by H. pylori, nitrosamines, salt-cured foods, and smoking. Diffuse-type skips that sequence entirely: it's poorly cohesive cells (often signet ring cells) infiltrating the gastric wall without forming glands, frequently sporadic or linked to CDH1 (E-cadherin) mutations, and associated with blood group A. The gross result of diffuse infiltration is linitis plastica — a stiff, thickened 'leather bottle stomach.' These two types have different histology, different risk factors, different genetics, and different gross appearances, and Step 1 will test all of it.

The second major angle is metastatic spread — and gastric cancer has more named metastatic signs than almost any other tumor. Virchow node (left supraclavicular), Krukenberg tumor (bilateral ovaries), Sister Mary Joseph nodule (periumbilical), and Irish node (left axillary) are all fair game. The exam loves testing whether you know which eponym goes where, and students routinely swap Virchow and Krukenberg. These aren't just trivia — each one reflects a real anatomical route of spread (hematogenous, lymphatic, transcoelemic, or direct extension through lymphatics), and understanding the route helps you remember the destination.

What makes this topic tricky is the density of interrelated facts. Students memorize signet ring cells but assign them to the wrong subtype. They know Krukenberg is ovarian but forget it's bilateral mucin-secreting metastasis specifically. And Sister Mary Joseph nodule gets dropped entirely because it's covered quickly in most resources. USMLE Step 1 will give you a clinical vignette — a woman with a periumbilical nodule, or bilateral ovarian masses on ultrasound — and expect you to trace it back to gastric (or other GI) cancer and name the mechanism. Lock down the eponyms, lock down the subtypes, and you own this topic.