Aspirin: one dose does not fit all

A recent review of individual data from 10 trials found that aspirin effectiveness for cardiovascular and colon cancer protection depends on the weight of the person (see aspirin decvasc risk if nl wt lancet2018 in dropbox, or doi.org/10.1016/ S0140-6736(18)31307-2).

Details:

--10 RCTs on primary prevention of cardiovasc disease (9 with weight/height and accessible individual patient data, 7 on low dose 75-100mg aspirin, 2 on high dose ≥325 mg), with 117,279 participants

--body weight varied 4-fold, with median weight ranging from 60-81 kg

--results validated in 5 trials of aspirin for secondary prevention of stroke (4 with individual patient data, one low-dose and 2 higher dose aspirin, and 1 comparing these 2 doses) 

Results:

--patients on 75-100 mg aspirin:

    --25% decrease in cardiovascular events in those weighing 50-69 kg, HR 0.75 (0.65-0.85), p<0.0001. Benefit was 32% in those weighing 50-69 kg, HR 0.68 (0.56-0.83), p=0.0001

    --no benefit in those weighing ≥​70kg, HR 0.95 (0.86-1.04). AND the case fatality rate of a first cardiovascular event in those >70kg was 33% higher, HR 1.33 (1.08-1.64), p=0.0082

    --also no effect in those weighing <50kg (very large confidence intervals, ?related to small number of participants, and does this low weight signify medical/psych comorbidities that decrease longevity??)

    --overall 29% dec in stroke in those <70kg, HR 0.71 (0.59-0.85), p=0.0002. no effect if >70 kg

    --effect is similar in those with early follow-up (<3 yrs) vs later; effect was reduced in those who smoked

--patients on ≥​325 mg aspirin:

    --decreased cardiovasc events only in those ≥70kg

        --in primary prevention trials, aspirin 325mg decreased cardiovasc events by 17%, HR 0.83 (0.70-0.98)

            --in those ≥​90kg, aspirin 500mg dec events 45%, HR 0.55 (0.28-1.09, almost significant) and combo of cardiovasc events plus death by 48%, HR 0.52 (0.30-0.89), p=0.017

--the data-derived, weight-based optimal aspirin dosing was:

    --50-69 kg: aspirin 75-100mg

    --70-89 kg: aspirin 325mg

    --≥90 kg: aspirin 500mg

--using this data-derived scale:

    --if exceeded the optimal aspirin dose: risk of sudden death in primary prevention studies was 2-fold higher, HR 2.03 (1.31-3.15), p=0.0015

--all of the above findings were similar in men and women, those with diabetes, or age ≥​70 and different heights

--the weight-dependent effect of low-dose aspirin was the same independent of aspirin formulation, but the decreased effect in those >70 kg was more pronounced in those on enteric-coated or delayed-release aspirin than standard release

--body weight itself was the most powerful predictor of aspirin response, more so than adding BMI (stratified by BMI> vs <25), height, lean body mass, fat mass, body surface area (though heights were important: shortest did better on low-dose aspirin and tallest better on high-dose)

--for colon cancer reduction:

    --20-year risk of colorectal cancer (1217 cases in the 5 primary prevention trials) was not decreased if >70kg 

    --low-dose aspirin in those <70kg: 36% lower risk, HR 0.64 (0.50-0.82), p=0.0004

    --high-dose aspirin in those up to 80kg: 31% lower risk, HR 0.69 ( 0.55-0.87), p=0.0014

--overall cancer-related deaths: no effect until after 5 years of follow-up in those <70kg

    --aspirin did not affect the overall incidence of first cancer, though there was a 20% increase in the first 3 years of follow-up (esp in those <70kg), then followed by reduced incidence after 5 years in all of the trials

    --those with diabetes who weighed <70kg had the highest increase in cancer associated with aspirin, esp in women <50 yo at baseline, partly due to increase in breast cancers. Overall cancer numbers were small, however: 28 of 1285 on aspirin vs 6 of 1283, though 4-fold risk: HR 4.35 (1.80-10.5), p=0.011; breast cancer in those <65yo: 34 vs 13 cases, HR 2.60 (1.13-2.57), p=0.012

--harms of aspirin were also dose-dependent:

    --risk of sudden death increased in patients at lower weight per dose

    --risk of all-cause death increased in those <50kg on 75-100mg aspirin, HR 1.52 (1.04-2.21), p=0.031

    --in people >70yo, 3-year risk of cancer was increased in those on aspirin, HR 1.20 (1.03-1.47), p=0.02, esp in those < 70 kg and in women, with 44% increase, HR 1.44 (1.11-1.87), p=0.0069

    --the risk of major bleeding decreased in those weighing >90kg​

Commentary:

--their data-derived, weight-based optimal dosing for cardiovascular protection was: 

    --50-69 kg: aspirin 75-100mg

    --70-89 kg: aspirin 325mg

    --≥90 kg: aspirin 500mg

--it should be noted, as in data above, that the actual weight thresholds were somewhat different for the noncardiovascular outcomes: excess bleeding on 75mg of aspirin was not decreased til weight >90kg, and there was no upper limit on bleeding risk found for those on 325 mg. and for colon cancer, low dose aspirin lost its effect at >70kg and high dose aspirin at >80kg.

--also the effect on overall cancers was mixed, with increased risk in the first 3 years, with benefit after 5 years but only in those with weighing <70kg.  this suggests that low-dose aspirin might accelerate cancer growth in the short-term. I am not sure how to square this with some studies finding that aspirin decreases the risk of ovarian cancer (see http://gmodestmedblogs.blogspot.com/2014/02/aspirin-decreases-risk-of-ovarian-cancer.html ​ , as well as decreasing the risk of metastases in those with adenocarcinomas, especially those on low-dose aspirin (see http://gmodestmedblogs.blogspot.com/2014/02/2-articles-on-aspirin-and-colonbreast.html  ). Part of the explanation as to why aspirin might increase cancer detection is that the likelihood of bleeding in early cancers may be greater (though this study did not find the anticipated lower cancer detection at later years)

--this study did find that those <70kg on higher doses of aspirin have more cardiovascular events, reaffirming mechanistically that lower doses of aspirin inhibit thromboxane A2 with its attendant decreased platelet aggregation (also a potent vasoconstrictor), but higher doses also inhibit prostacyclin which has the opposite effects of thromboxane A2. so, those on 325mg of aspirin who are >70 kg, for example, may have the equivalent bioavailability of those on 75-100mg of aspirin but are <70kg; and those on 325 mg of aspirin but are <70kg may have much higher aspirin bioavailability and might have increased prostacyclin inhibition leading to more clinical vascular events

--one other concern is that men tend to weigh more than women (eg: 80% of men and 50% of women in above study were >70kg), and this might explain the finding in secondary prevention of stroke trials that men do benefit by aspirin alone (vs women)

--one intriguing conclusion to this study is that older studies finding that other platelet inhibitors outperformed aspirin might have had a very different result if the "appropriate" dose of aspirin were given (eg the CAPRIE trial showed that clopidogrel was somewhat better than aspirin in preventing cardiovascular events). Perhaps we should restudy several of these newer (and way more expensive drugs) against an aspirin dose that is titrated to weight??? Or that some of the variable results found in the aspirin prevention trials may reflect the weight differences in the studies. For example, one of the best negative studies I have seen is the POPADAD trial, which had 1276 diabetic adults randomized to aspirin vs placebo (plus an ineffective antioxidant wing), finding no benefit from aspirin.  Is it because of the diabetes (though aspirin is still recommended by guidelines)?? Is it that diabetics have stickier platelets with higher turnover requiring different aspirin doses? Notably, the above study found no decrease in aspirin’s cardiovascular protection in diabetics when looking at weight-based dosing.

--because of some older studies comparing enteric-coated vs standard aspirin, I have been prescribing only the regular 81mg dosage for the past several years. some studies suggested that there was erratic absorption of the enteric-coated aspirin, and other studies suggested that there was no difference in gastric benefit from the enteric-coated over regular low-dose aspirin. Other studies have found that enteric-coated aspirin overall had a 40% decreased systemic bioavailability, perhaps explaining the decreasing effect of enteric-coated aspirin as weight increases >70 kg in the above study.

--the mechanism explaining the above study is likely reduced bioavailability of aspirin with increasing weight as found in many other studies. Though obesity is associated with increased platelet activation, in the above study they  did find that weight was still significant even after controlling for BMI (> vs <25), as well as the finding that enteric-coated aspirin (with known decreased bioavailability) is associated with less cardiovascular protection.

--there are many limitations of the above data, including the fact that several of the studies were quite old (and there have been impressive changes in some risk factors and effective meds subsequently used), pretty limited data on the use of standard dose aspirin (all but one study used enteric-coated or extended release aspirin), they used the secondary stroke prevention trials as their validation cohort (and this might not reflect other secondary prevention sites as well as newer primary prevention trials), body weight might well have changed over the course of these studies (so the reported weight may have changed over the time of the study), and these are secondary analyses many years after completion of the studies (with the attendant risk of unaccounted for confounders).  So, would be important to have actual current RCTs which document the effects prospectively on standard 75mg dose vs weight-based dose

So, this study is really great in that it challenges our widely-held belief that patients at high cardiovascular risk should be on a single standard dose of aspirin, and even that recommendations for colon cancer prevention include a standard aspirin dose (eg, see USPSTF aspirin recommendations: https://www.uspreventiveservicestaskforce.org/Page/Document/RecommendationStatementFinal/aspirin-to-prevent-cardiovascular-disease-and-cancer​ ). Though it is reasonable and appropriate to wait for confirmatory prospective studies, I am really concerned that some patients <70kg on higher doses of aspirin do worse (I do have a couple of them who had CABG and the surgeon insisted they be on 325 mg aspirin) and will consider increasing the dose of aspirin in those at really high cardiovascular risk who are really overweight. and I will  continue to prescribe the non-enteric coated aspirin (though there is push-back from some pharmacies) since to me this is reaffirmed in the above study

geoff​

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to blogs since 8/15/17:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just type in a name in the search box and get all the blogs with that name in them

to access older blogs from the BMJ website, from October 2013 until 8/15/17: go to http://blogs.bmj.com/bmjebmspotlight/category/archive/ 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list