Aspirin's reduction of cancer risk has been on relatively solid epidemiological footing for a long time, it's simply not a big enough protective effect to balance the risk of bleeds etc:
Now, if you were to consider all cancers plus cardiovascular benefit, and target folks who are more likely to benefit and/or less likely to have adverse events, it might be useful. Trying to sort some of this out with genomic data was the subject of my MSc, but I couldn't make it work.
http://www.uspreventiveservicestaskforce.org/uspstf/uspsasco...
Now, if you were to consider all cancers plus cardiovascular benefit, and target folks who are more likely to benefit and/or less likely to have adverse events, it might be useful. Trying to sort some of this out with genomic data was the subject of my MSc, but I couldn't make it work.