Few diseases have been studied more extensively with little to show for it as much as pancreatic cancer. Starting in the 1970’s investigators looked at adding chemotherapy with or without radiation to best available surgery for that minority of patients with pancreatic cancer who were candidates for surgical resection (fewer than 20% in most series). The results have been all over the map, but the trend has been that adding best available chemotherapy has played a role in the modest improvement in outcome seen in the last thirty-five years. Better surgery and better definition of a curative operation, with more attention paid to the posterior margin, may also have played a role in better outcome (with better attention to margins, fewer patients were deemed “cured” after their operation; and they would be expected to do better than those patients formerly included in this group who were not cured by modern standards of margin examination). So what matters, and what does this all mean? A recent study from Japan, published in a German cancer journal, gives a clue (the link is to an abstract; contact Dr. Stark for the full article by filling out the form on the right of this page). The Japanese looked at time to initiation of chemotherapy after surgery. They used a combination of gemcitabine and S1, a 5FU-like drug not used in the US, after curative resection. They found that the most important predictor of outcome was how long it took to initiate therapy after surgery. The break point was twenty days — very short by standards of US clinical trials, where the interval can be up to ten weeks. Patients whose chemo was started within twenty days didn’t just do a little better; they did spectacularly better: 52 vs. 26% five-year survival after surgery. The 52% is a very high number for this disease, perhaps the highest ever reported. There were only 114 patients in the trial but the p value (0.003) was very convincing because the difference was so great. Surgery to remove the primary cancer can, in experimental animal models, stimulate the growth of microscopic metastases; hence the need to add adjuvant chemotherapy quickly. The problem with pancreatic cancer is that the surgery is so difficult that chemo is often delayed. This study may provide a powerful clue as to why results from prior studies have been so disappointing.