When the scientists examined records for those with a 20 year follow-up, they found that the "aspirin group" had 20 % fewer cancer deaths overall, 35% fewer deaths from GI cancers but no decrease in what we call "hematologic cancers," which are known to you as lymphomas and leukemias among others. The greatest benefits were seen in those people who were treated with aspirin for 7.5 years or longer. In this 20 year group, there was a "delayed" or latent period until the deaths declined, with decreases in deaths from cancers of the esophagus and pancreas seen at about 5 years onwards, while decreases in deaths from stomach and colorectal cancer did not appear until about 10 years. The researchers also reported that in absolute terms, the risk of dying from cancer was decreased by a significant 7.08% at age 65 and older.
Most of the decrease in deaths in patients who had cancers other than GI cancers and were followed for at least 20 years occurred in patients with lung cancer, and most of those had the type of lung cancer called adenocarcinoma, which is the most common form of lung cancer in the United States. There was some decrease in prostate cancer deaths late in the follow-up, but this was not statistically significant and was seen in patients who had taken aspirin for 7.5 years or more.
One observation that I think deserves note is the fact that most of the decreases in deaths were seen in cancers that are what we call adenocarcinomas, which are cancers that form in "glandular" organs.
Another point that is important is that most of the participants in these trials were male, and therefore no comment can be made one way or the other about common cancers in women, such as breast cancer or gynecologic cancer.
Another very important point is that the dose of aspirin was 75 mg a day or greater, and increasing the dose did not increase the benefit.
Does this study mean that everyone age 40 or 50 and over should start taking aspirin?
Although the study authors make the case that there are several reasons to believe that this trial can be generalized to larger populations, I don't think that this can be turned into a recommendation for everyone to start taking aspirin.
Yes, there is research that shows aspirin can reduce the risk of colorectal cancer. Yes, this is an excellent study that breaks new ground and provides us with important and valuable information. Yes, it may turn out sometime in the future that aspirin will become a valuable addition to our ability to reduce the risk of death from cancer-for which we have few effective proven alternatives at present.
But aspirin has side effects, my friends. It causes bleeding in the GI tract and can cause bleeding in the brain. Before we doctors started making recommendations about the use of aspirin in people at risk of heart attacks or those who had one, there were prospective trials done so we would understand not only the benefits but also the risks. That information is not available from this particular report (that is not a criticism of the authors' work, just an observation).
We have faced these issues again and again. We plow the fields to find information where we can when we don't have the right types of clinical trials to answer the questions. But the problem is that we frequently try to point the way through more or less indirect means when we don't have direct information. And then there are the situations that have occurred not infrequently where such "retrospective" analyses produce results that don't stand up when carefully designed prospective trials are conducted.
What would be more definitive and exact would be trials that are done specifically to address the question at hand, randomized going forward, and controlled for all sorts of factors. That is what we call a randomized, prospective trial.
What we have here-and let me be very, very clear: I am not criticizing the quality of research done by the authors-is an analysis of other studies done originally for measuring other outcomes. What we need is a prospective trial asking the question specifically about the impact of aspirin on cancer deaths. But the real question is whether any such trial will or ever could be done, especially when you consider that it would take over 20 years to complete.
So we are left in a bit of a quandary, just as we are with vitamin D and have been in the past with other vitamins (until the randomized trials were actually performed). That means experts are going to have to digest this information, and decide whether or not it justifies subjecting millions of people to a recommendation that they take a baby aspirin every day.
I don't know when that is going to happen, yet I am certain there are some/many of you out there are going to charge ahead and take aspirin every day once you hear about this study. And there are some of you out there who do that and will end up in a hospital emergency room or worse as a result of that decision. But we will never know who you are.
So that is the dilemma we face as we write and talk and broadcast the results of trials such as these. I can be as cautious as I want to be, and tell you I would not recommend you start taking a daily aspirin at this time, yet there will be the inevitable headlines and news stories that will shortchange the cautious part and propel you forward to an action that could be harmful, based on evidence that is good in a scientific sense but not sufficient in a "patient recommendation" sense.
As I have said before, and will say again: proceed with caution with "eyes wide open." I have no doubt there will be more information forthcoming on this topic that may make the decision just a bit easier and perhaps a bit better informed.