Prognosis for patients with pancreatic ductal adenocarcinoma (pancreatic cancer) is notoriously bad. In a study published in this issue of Gut, Kim and colleagues¹ make it clear that K-ras mutation in apparently benign pancreatic tissue at some distance from the tumour contributes to poor survival (see page 1598). This could indicate the presence of occult tumour cells in the resection margin but could also indicate that tumours developing in a large field of mutation carrying cells are more aggressive. This offers hope to patients where resection leaves behind no mutant cells in the margin but also suggests that the most aggressive tumours might prove to be the easiest to detect in asymptomatic patients.

A cynical reader might be forgiven for passing over the article by Kim and colleagues.¹ The paper talks of an unfavourable prognosis for patients with pancreatic cancer—cynics will respond with the observation that prognosis for pancreatic cancer is death, how much more unfavourable can it be? The paper also talks about K-ras gene mutations—cynics may groan and remember the many reports on K-ras mutations going back to the 1980s. Some may even remember a happier time when they believed that these mutations might serve as a marker for cancer, a time before it became apparent that such mutations are common in people with benign conditions. Some readers will be tempted to read on when they notice that the abstract refers to K-ras mutations in resection margins histologically free of invasive cancer; these readers will be rewarded with an optimistic report that may prove of immediate clinical value and furthermore gives some exciting new insights into the different ways in which pancreatic cancer could develop.

Kim and colleagues …