This is an interesting infographic detailing the frequency and lethality of the most common types of cancer. As the article points out it is lucky that the most common ones are not a dead sentence. A small proportion of patients with prostate cancer will die compared to those with pancreatic cancer.
The article also reminds me of something that we need to take into account when studying the biology of the disease. Pancreatic cancer is so deadly because the (relatively) few cases are detected when it is too late. It is quite possible that many pancreatic tumours grow but never grow to the point where doctors will detect it. One of the issues in treating prostate cancer is that we are getting so good at detecting cases early on. Often we treat patients whose tumours are not life-threatening just to be on the safe side.
Could it be that the numbers in this infographic (and on the cancer.gov website as well as other official databases) do not represent the real probability of developing a cancer but of developing a clinically detectable one?