Aspirin and cancer treatment: systematic reviews and meta-analyses of evidence: for and against
-
https://www.nature.com/articles/s41416-023-02506-5
A couple sections of this I found particularly interesting, after correcting for publication bias, the benefits of aspirin were still "significant." Also despite the lack of randomized trials, the few early reports also suggest a 9% reduction in cancer related deaths. And finally, gastrointestinal bleeding has been, by far and large, fearmongering.
"Publication bias, arising from the selective publication of positive findings for an intervention such as aspirin, is a most important issue in meta-analyses such as the above. A ‘trim and fill’ testing procedure to restore symmetry in forest plots was therefore applied extensively to the data, and although the benefits of aspirin were reduced, significance of almost all the reductions remained.
In their report the authors identified 23 publications which had focused upon fifteen less studied cancers (naso- and oropharyngeal, oesophagus, gastrointestinal, gastric, rectal, liver, gallbladder, bladder, pancreas, bladder, endometrium, ovary, glioma, head and neck, lung, melanoma). Meta-analysis of these gave pooled reductions of around 30% in deaths associated with aspirin taking (HR 0.67; 0.60, 0.75 in 21 studies, and OR 0.47; 0.26, 0.83 in five studies) [30]. Together with the evidence of favourable effects of aspirin upon a wide range of biological mechanisms relevant to cancer mortality [3], these clinical outcomes suggest that aspirin is likely to be of benefit to patients within a very wide range of different cancers...
...The pooled results of four early trials based upon a total of 722 patients with cancer gave a suggestive pooled reduction associated with aspirin of about 9% in cancer deaths. Currently, a number of randomised trials which test aspirin and mortality are in progress. These focus upon the common cancers: colon, breast, prostate and one in lung cancer."
"In a systematic literature search eleven randomised trials which included data on fatal bleeding were identified . These 11 RCTs included together a total of 121,094 subjects, followed for an average of 2.8 years...
...These data confirm the usual excess risk of all ‘major’ bleeds for aspirin (RR 1.55), equivalent to about one per bleed per 1000 persons per year. Note however that the 50% increase above the background risk of bleeding from a peptic ulcer, stomach infection or other pathology, means that amongst patients who are taking aspirin the risk of a GI bleed being truly attributable to aspirin is only one in every three bleeds.
However, the proportion of ‘major’ bleeds in subjects who had been randomised to aspirin that led to death was 4% in those who had been randomised to aspirin while 8% of bleeds in subjects randomised to placebo were fatal. Clearly, this implies that overall, the bleeds truly attributable to aspirin must be of a much lower severity than other bleeds attributable to stomach pathology. This is further confirmed by the absence of any increased risk of a fatal bleed associated with aspirin taking, as shown in the third cell in the above table (RR 0.77), and this final conclusion has been confirmed in overviews of bleeding reported by other authors.
It is unfortunate however that the scientific literature on the issue of aspirin and bleeding appears to have been swamped by a host of statements about serious dangers of aspirin, most unsupported by any evidence while some are total misinterpretations. Probably the most misleading and most influential item on the web was a report issued by Reuters on the 14th June 2017, stating: ‘daily aspirin causes 3000 deaths from bleeding in Britain every year’. This claim was taken up and very widely and repeatedly publicised in the web and the media across the world. The report by Reuters was however a totally invalid, having been based on a prospective study of 3166 older patients, all of whom (93–97%) were taking aspirin. There were therefore no control subjects and no valid estimate of the independent contribution of aspirin to the fatal bleeds can be made."
Also, I can't really leave without posting this things of beauty. Straight from the belly of the beast itself, the oldie but goldie, the good old Harvard finding that pissed everyone off.
It appears that, like all too often, Ray was right again. Aspirin appears to be a useful addition for cancer patients.