LDL less than 40 in high risk patients??

A recent evaluation of the FOURIER study found cardiovascular benefit by lowering LDL to less than 40 mg/dL (see cad high risk LDL less than 40 FOURIER circ2021 in dropbox, or doi.org/10.1161/CIRCULATIONAHA.121.056536)

Details:

-- this study was an exploratory analysis of the FOURIER study, which compared cardiovascular outcomes in those with stable atherosclerotic cardiovascular disease on statins, randomized to evolocumab (a PCSK9 inhibitor) vs placebo

    -- 27,564 patients were enrolled in FOURIER

    -- mean age 63, 75% men

    -- prior MI 81%, prior ischemic stroke 19%, peripheral artery disease 13%

    -- diabetes 37%, smokers 28%, hypertension 80%

    -- statin therapy 100% (70% high-intensity, 30% medium-intensity); 5% on ezetimibe; 92% on aspirin, 75% on b-blocker, 78% ACE/ARB

    -- median baseline LDL 93 mg/dL (interquartile range 80-109 mg/dL)

--patients randomized to evolocumab: 65% achieved an LDL <40 mg/dL

Results:

top graph: y-axis = achieved LDL in mg/dL at 48 weeks, finding that the achieved LDL was lower in those with lower baseline LDL. The shaded area represents the amount of LDL lowering between the treatment arms (bottom of shaded area = evolocumab, top = placebo) for a given baseline LDL (blue shaded area = achieved LDL above 40 mg/dL, red is below)

bottom graph: y-axis = Hazard ratio for evolocumab vs placebo for cardiovascular death, MI, or stroke, per 39 mg/dL (1 mmol/L) reduction in LDL

interpretation of these graphs:

-- there were essentially no persons with an initial LDL on statins who achieved an on-treatment LDL < 40mg/dL unless they were randomized to evolocumab

-- the further the baseline LDL was lower than 93 mg./dL, the lower the achieved LDL, decreasing from 0% of the patients at 93 mg/dL to 38% who started at 58 mg/dL

-- AND, the HR for vascular events continued to decrease as the baseline LDL was lower (reflected by the HR actually improving more with those with the lower baseline LDL who were much more likely to have much lower achieved LDLs). that is to say: though >1/3 of the patients had LDL <40 mg/dL when they had baseline LDL nearer to 58 mg/dL, the clinical benefit did not decrease as the LDL continued to decrease below 40 mg/dL

-- there was also clear benefit for those with higher baseline LDLs above 100mg/dL or so

    -- eg, those with baseline LDL of 130 mg/dL did achieve an LDL of about 120 mg/dL on placebo but LDL of around 60 mg/dL on evolocumab (though there was only had a 10% reduction in clinical events, per 39 mg/dL decrease in LDL)

--there was a similar continuous decrease in cardiovascular events by assessing ApoB or non-HDL levels (graphs not shown, see article if interested)

Commentary:

-- the American College of Cardiology/American Heart Association/Multisociety cholesterol 2019 guidelines recommend an LDL <70 mg/dL in high-risk individuals, with addition of other agents if statins are unable to achieve this level 

-- the European Society of Cardiology/European Atherosclerosis Society 2019 guidelines recommend that for secondary prevention in patients at very high risk, a goal LDL reduction of >50% and an LDL level of <55 mg/dL (see doi:10.1093/eurheartj/ehz455 ). And for those who have a second vascular event while on maximally tolerated statin-based therapy, an LDL-C goal of <40mg/dL "may be considered"

-- a meta-analysis of 4 trials (including the FOURIER study, but with only 2,023 participants at that time) found that LDL levels down to 21 mg/dL were associated with improvements in major vascular events, and in a linear fashion (see http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html ). not dissimilar to the results in the larger FOURIER study above

-- in this current study, the LDL in the placebo group did not change, around 90 mg/dL, over the course of the study; those on evolocumab averaged LDL in the low 30 mg/dL range over the study, an average decrease of 56 mg/dL

-- overall reductions with evolocumab at 48 weeks: LDL <70 mg/dL in 87%, LDL <40 in 67%, LDL <25 in 42%

-- for these high-risk patients, 30% were on a moderate-intensity statin...  seems like they were undertreated and should have been maximized in comparing the benefit with evolocumab (see below)

-- rather curious that there was so little protection in those with baseline LDL >110 mg/dL or so. (see above example of baseline LDL of 130 mg/dL). A few issues:

    -- perhaps the reason that there was not much change in the benefit of evolocumab in those with baseline LDL from 110 to 150 mg/dL is because of unmeasured confounders
    -- in the "efficacy in key subgroups" analysis in the original FOURIER trial (page 55 of the supplement): there was only a strong but not statistically significant trend to efficacy of evolocumab in the primary endpoint in those with baseline LDL >109 mg/dL, HR 0.89 (0.77-1.02), similar to the above result. there was no further analysis of results comparing baseline 130-150 vs 110-130 mg/dL, or other subdivisions. But...  evolocumab is so effective in lowering LDL!!!! other studies have found benefit from statins in lowering LDL in this range

    -- post hoc analysis of the Treating to New Targets trial found that baseline HDL was a major predictor of cardiovascular events in patients with baseline cardiovascular disease, with those on atorvastatin 10mg but HDL >55 mg/dL having the same 5-year risk of a major cardiovascular event as those on atorvastatin 80mg but HDL <38 mg/dL, and overall there were more cardiovascular events in those with lower HDLs (see cad TNT with hdl nejm 2007 in dropbox, or Barner P. NEJM 2007; 357: 1301-10).  No stratification by HDL in the FOURIER trial.  

        -- ie, the clinical effectiveness of going from a moderate-intensity statin (atorvastatin 10mg) to a high-intensity one (atorvastatin 80mg) in the TNT trial depended on the baseline HDL!!!  see http://gmodestmedblogs.blogspot.com/2015/06/improve-it-trial-ezetimibe.html . here is their graph:

-- FOURIER trial (see lipids PCSK9 evolocumab nejm2017 in dropbox, or DOI: 10.1056/NEJMoa1615664), or http://gmodestmedblogs.blogspot.com/2017/03/treating-ldl-to-target-more-evidence.html: overall evolocumab was associated with a 59% decrease in LDL form median baseline of 92 mg/dL, associated with a 15% decrease in vascular events with no increase in adverse events (including new-onset diabetes and neurocognitive events). of note, this decrease in vascular events was less than expected for the level of LDL decrease, leading to a significant reduction in the value of the drug company in the stock market, a more significant reduction in their stock than in clinical events with their drug...

    -- this should serve as a reminder that LDL is just a surrogate endpoint, and the really important result is actual clinical events in people. For example, there are lots of different LDLs: this is just an electrophoretic pattern combining LDLs below a certain arbitrary density. And there are lots of studies suggesting that all LDLs are not the same: the small dense ones are more atherogenic than the big fluffy ones. so, do these PCSK9's decrease the better ones more than the bad ones, leading to less clinical benefit???? they did measure the apoB levels, and apoB is a surrogate marker of the surrogate marker of LDL size (one apo B per LDL, so small dense LDLs are associated with relatively higher apoB levels. and apoB level is a stronger marker of clinical events than LDL levels!!!). But, again, in this study it does not appear that there was a sufficiently large decrease in clinical events with lowering the apoB levels: looks even less effective than lowering overall LDL levels on their graph, supporting a change in LDL type:                                                          

Limitations:

-- this is a retrospective analysis of one trial (FOURIER), assessing achieved LDL levels and finding benefit of a lower LDL. and they did find that the lower initial LDL was associated with a lower achieved LDL by the powerful PCSK9 drugs. But, we can only assert that this is an association and not a causal relationship. For the latter, we would need a randomized controlled trial with specific LDL goals targeted. maybe those able to achieve a lower LDL were different from those who could not?

--there was not much granular data on key clinical aspects of cardiovascular events. maybe there were important differences between the groups in their lipids (eg, subgroups of LDL, as above) or other non-measured risk factors (exercise, diet, supplements, stress, other sociodemographic or psychosocial variables)

    -- as mentioned above, one important variable might well be stratification by HDL levels. there was some increase in them by evolocumab in FOURIER trial overall, but no individual-level data/analysis

    -- in theory, an RCT of this number of patients should equalize these variables to those randomized into the different groups, but would be useful to actually know...

--the FOURIER study was of very high-risk people, limiting generalizability to lower risk ones. For example, none of their patients on statins had an achieved LDL <40 mg/dL (and likely few had achieved LDL<40-50 mg/dL, given the interquartile range of the median LDLs was 80-109 gm/dL; and the median for the lowest LDL quartile was 74 mg/dL)). on my anecdotal note, i have been very aggressive in cholesterol management, as mentioned below, and have several patients achieving LDL levels in the 20-40 mg/dL range. my guess is that the FOURIER group was in part so high risk because they had quite high initial LDLs and/or bad LDL phenotypes or were less sensitive to statin-induced LDL lowering than less high-risk people???

-- as noted above, these high-risk patients were not optimally treated at the time of the start of the study: the presumed target LDL was <70 mg/dL by the least aggressive AHA guidelines, yet:

    -- the baseline median LDL was 93 mg/dL

    -- even the lowest quartile LDL at baseline was 74 mg/dL

    -- a full 30% were not even on "high-intensity" statins

        -- and not all "high-intensity" statins are created equal: both rosuvastatin 40mg and atorvastatin 40mg both make the list, yet the former is way more powerful than the latter (per the medial literature, and confirmed in my practice). the FOURIER study itself did not differentiate the "high-intensity statin" used

    -- only 5% were on ezetimibe, further reinforcing the lack of aggressive management of LDL at the start of the study. for a prior blog documenting the clinical improvement with a lower LDL goal with ezetimibe, as well as a brief review of FOURIER, see http://gmodestmedblogs.blogspot.com/2017/03/treating-ldl-to-target-more-evidence.html

    --so, for our clinical practice, this study is ill-conceived: i strongly believe that we should maximally manage this very vulnerable high-risk patients with aggressive lipid targets, and that the primary approach should be statins (grinding up the specific statin/dose as needed), then adding ezetimibe, and then considering a PCSK9 inhibitor (i believe optimizing non-PCSK9 therapy is required by insurers anyway to approve the PCSK9 add-on). So, the relevant clinical study would have been placebo vs med in high-risk patients with LDL maximally controlled by the traditional meds (ie really high-intensity statin with ezetimibe, as needed). but, this type of study would probably have been less appealing to the sponsor of the current study, the drug company....)

so, seems like the lower the LDL the better, as per the prior blog http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html . and without any clear adverse events. so, what does this mean for us in primary care?

--patients at risk, and especially for primary prevention in those with no known prior vascular events but who happen to achieve a very low LDL on just statins: i think they can be left at that level with likely increasing benefit and no clear harm.

    -- i must admit that for years i have been pushing statins quite aggressively in those in higher cardiovasc risk, pushing them well below the 70 mg/dL mark, given that former study of benefit down to 21 mg/dL, by cranking up the statin potency and/or dose as needed

    -- and, i do see patients with achieved LDLs in the 30 mg/dL range on statins, all tolerating these LDLs without a problem (and none have had vascular events, by the way, over many years of follow-up....), so i am leaving them on the med.

--i do think it makes sense to really decrease the LDL as much as can easily be achieved on statins in those with multiple ongoing risk factors (smoking, diabetes, uncontrolled hypertension)

--but the question comes up as to when we should add another med (eg ezetimibe) for further reduction. my guess is that might be reasonable when one cannot achieve the goal of <70 for primary prevention in higher risk patients. but unstudied, and at this point i am still hesitant to add another med

--and, those with much higher risk (secondary prevention: esp recent prior vascular event) should probably be on statin plus ezetimibe to lower their LDL levels if the statin is not achieving an LDL in the <55 mg/dL range as per the European guidelines, and move to a PCSK9 inhibitor (and deal with the prior approval) if the patient cannot achieve a goal <55mg/dL, with the hope of getting <40 mg/dL

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 all of the blogs (2 options):

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 click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

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