alternative LDL lowering strategies

 A recent systematic review found that taking a moderate statin dose along with ezetimibe was associated with lower LDL levels, decreased adverse medication effects, and no significant change in cardiac outcomes, when compared to high dose statins alone (see lipid alternative lipid lowering JAMAcardiol2024 in dropbox, or doi:10.1001/jamacardio.2024.3911). thanks to Ian Huntington for bringing this to my attention

Details:

--Two randomized clinical trials compared an "alternative LDL-cholesterol lowering strategy" versus a high intensity statin strategy in patients with atherosclerotic cardiovascular disease (ASCVD):

    --The RACING study (Randomized Comparison of Efficacy and Safety of Lipid-Lowering With Statin Monotherapy versus Statin/Ezetimibe Combination for High-Risk Cardiovascular Disease, quite a mouthful...):

        -- a randomized, open label, noninferiority trial at 26 centers in South Korea of 3780 patients who were randomized to rosuvastatin 10 mg with ezetimibe 10 mg (a moderate intensity statin along with ezetimibe, ie the alternative LDL-cholesterol lowering strategy group) versus rosuvastatin 20 mg by itself (the high intensity statin monotherapy group) in this 3-year study (see lipids RACING study mod intensity statin plus ezet as good as high-intensity Lancet2022 in dropbox, or https://doi.org/10.1016/S0140-6736(22)00916-3)

        -- mean age 64, 75% male, essentially all had prior significant ASCVD events, baseline serum LDL 80 mg/dL with 33% having concentrations less than 70 mg/dL, lots of participants with hypertension and diabetes

        -- the achieved LDL levels in the study were 58 mg/dL in the combination group and 66 mg/dL in the high-intensity group

        -- non-inferiority was met for the alternative strategy vs high-intensity strategy in the composite primary endpoint (cardiovascular death, major cardiovascular events, or non-fatal stroke); essentially all of the components of that composite trended to benefit for the alternative strategy group

    -- The LODESTAR study (Low-Density Lipoprotein Cholesterol-Targeting Statin Therapy versus Intensity-Based Statin Therapy in Patients With Coronary Artery Disease, and another mouthful; it seems that the terse clever prior cardiology study names have been supplanted):

        -- A total of 8180 patients with ASCVD were randomized to 2 groups, one group to treat-to-target LDL to the 50-70 mg/dL range, the other taking just rosuvastatin 20mg or atorvastatin 40mg. But only a few were on ezetimibe; per the actual text of the LODESTAR study, "Nonstatin add-on therapy, such as ezetimibe, was not recommended strongly in the treat-to-target group", and only 12% in the treat-to-target group and 10% in the high-intensity statin group were on ezetimibe. Even in the treat-to-target group, ezetimibe was actually added to those on high-intensity statins more often than those on the moderate dose statin (ie the first change to lower the LDL was from the lower dose statin to a higher one, then add ezetimibe as needed)

        --the achieved LDL in both groups was essentially the same at 69 mg/dL; the target-to-treat approach was non-inferior [but this study really did not test the strategy of moderate dose statin plus ezetimibe vs high dose statin: ie, it was not truly a study of lower dose statin with ezetimibe as a means to achieve good clinical outcomes without the adverse effects of higher dose statins; for a fuller critique of the study but one which endorses the role of treat-to-target approach in general, see https://gmodestmedblogs.blogspot.com/2023/03/statins-treat-to-target-ldl.html]

-- for the combined 8180 patients with ASCVD from both studies: mean age 64.5 years; 2182 (26.7%) female/5998 male (73.3%)

-- the primary end point was a 3-year composite of all-cause death, myocardial infarction, stroke, or coronary revascularization. The secondary end points comprised a few clinical efficacy and safety end points, as per below.

Results:

-- the achieved LDL cholesterol level for the combination of the above 2 studies during treatment:

    -- alternative strategy: 64.8 mg/dL

    -- high-intensity statin strategy: 68.5 mg/dL

        -- P < .001

A graph of the different achieved LDL levels during the 3 years of the study, noting that the curves were parallel after year 1:

 

-- primary end point of composite of all-cause death, MI, stroke, or coronary revascularization:

    -- alternative strategy: 7.5% [304 of 4094]

    -- high-intensity statin strategy: 7.7% [310 of 4086]

        -- hazard ratio 0.98 (0.84-1.15); P = .82, not statistically different

-- secondary end point of composite of cardiovascular death, MI, stroke, or coronary revascularization:

    -- alternative strategy: 6.4% [255 of 4094]

    -- high-intensity statin strategy: 6.6% [264 of 4086]

        -- hazard ratio 0.97 (0.81-1.15); P = .82, not statistically different

-- there was also no significant difference between the groups regarding any of the individual efficacy end points within the composite outcome assessment

-- the above findings were consistent in per-protocol analysis

safety end points:

--new-onset diabetes:

    -- alternative strategy: 10.2% [271 of 2658]

    -- high-intensity statin strategy: 11.9% [316 of 2656]

        -- odds ratio 0.84 (0.71-1.00), P = .047

--initiation of antidiabetic medication for new-onset diabetes:

    -- alternative strategy: 6.5% [173 of 2658]

    -- high-intensity statin strategy: 8.2% [217 of 2656]

        -- odds ratio  0.78 (0.64-0.96), P = .02

--intolerance-related discontinuation or dose reduction of assigned therapy:

    -- alternative strategy: 4.0% [163 of 4094]

    -- high-intensity statin strategy: 6.7% [273 of 4086]

        -- odds ratio 0.58 (0.47-0.71),  P < .001

-- the above safety endpoints were also consistent in the per-protocol assessment

-- and the diabetes assessments above were independent of prediabetes status or BMI

Commentary:

-- It is abundantly clear in the medical literature that intensive lowering of LDL cholesterol levels in patients with ASCVD or at very high ASCVD risk decreases ASCVD clinical events in the future

    -- one issue that has been raised over the years is whether statin dosing should be with a particular target in sight or if patients should be dosed for statins by their underlying condition and risk factors

        -- A few studies have suggested strongly that targeted dosing makes the most sense, as presented in https://gmodestmedblogs.blogspot.com/2023/03/statins-treat-to-target-ldl.html 

            -- The argument here is that people have different relative risks of cardiovascular events and very different LDL responses to different levels of statin intensity, that this one-size-fits-all approach of high intensity statins to anyone with ASCVD may well be overtreating or undertreating many people yet exposing them to different levels of adverse effects from the statins (ie, the goal of therapies in general should be patient-specific and not a simple standardized dose of meds)

                -- the whole concept of medium versus high intensity statin therapy is a bit fraught since the individual responses to the same doses of a statin are quite variable, and their underlying disease and risk factors dictate different lipid goals

                -- and the bundling of statins into a "intensity" groups, for example, includes statins with quite different potency: there is really a big difference in achieved LDL levels from atorvastatin 40 mg versus rosuvastatin 40 mg, both of which considered to be “high intensity”; as well as from atorvastatin 10mg vs either simvastatin 40mg or rosuvastatin 10mg, all of which are considered "moderate intensity"

    -- then there is the question of what is the appropriate goal LDL:

        -- a large meta-analysis found that there was considerable clinical benefit in those with ASCVD to achieve a goal of less than 20 mg/dL: https://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html

        -- a more recent analysis of the Fourier study confirmed that achieving an LDL of less than 20 mg/L added benefit with minimal risk (see cad high risk LDL less than 20 fourier studyCirc2023 in dropbox, or DOI: 10.1161/CIRCULATIONAHA.122.063399)

            -- And, though the studies achieving these very low LDL levels involved the use of the highly potent PCSK9 inhibitors (which have other likely beneficial targets besides LDL, particularly Lp(a)), the Cholesterol Treatment Trialists' Collaborative study found that in the subset of patients with a mean baseline LDL of 66 mg/dL, the relative risk for major vascular events per a 1-mmol/L (38.7 mg/dL) further reduction in LDL by more intensive statin therapy was 22%, RR 0.78 (0.65-0.94) (see lipid CTTC low LDL lancet2010 in dropbox, or Lancet. 2010; 376(9753):1670.). so, statins as well as PCSK9 inhibitor both have a dose response with higher doses leading to further lowering of LDL levels and fewer adverse cardiovascular events in high-risk patients (the CTTC sample did include lower risk patients, though vast majority had ASCVD and subgroup analysis found similar outcomes when stratifying patients by previous vascular disease)

-- Another major issue is what is the actual effectiveness of increasing the intensity of statins versus adding ezetimibe

    -- In general, the incremental effectiveness of statins is most profound with the lower statin doses: adding increased statin intensity does have some diminishing returns:

        -- For example, studies have suggested that doubling the dose of a statin to the high intensity level leads to about a 6% additional lowering of LDL levels

        -- Adding ezetimibe to a moderate intensity statin leads to a 23-24% decrease in LDL levels, with the added effect of a likely reduction in statin-related adverse effects (the combination of of simvastatin 40mg plus ezetimibe is superior in terms of hard cardiovascular endpoints vs simvastatin 40mg by itself, per the IMPROVE-IT trial: https://gmodestmedblogs.blogspot.com/2015/06/improve-it-trial-ezetimibe.html)

        -- it does make intuitive sense that lower dose statins plus ezetimibe would lead to lower doses of statins to achieve the same LDL level and fewer of the dose-dependent adverse effects of statins, since these two medications have very different mechanisms of action (blocking HMG-CoA reductase by statins, vs inhibiting dietary and biliary cholesterol absorption by the intestine by ezetimibe)

-- The idea embodied in the concept of the current lipid study is an old one.  Many decades ago the concept emerged with hypertension treatment, also to see if using multiple low doses of antihypertensives would lead to decreased blood pressure and many fewer adverse effects than maximizing the dose of one medication prior to initiating another one.

    -- a meta-analysis of 42 trials found that there was impressive hypertension benefit of taking one quarter of the dose of the standard antihypertensives, with essentially no adverse events (https://gmodestmedblogs.blogspot.com/2017/06/low-dose-hypertension-therapy.html).

    -- this blog also refers to another study finding that doubling the dose of thiazides, beta-blockers, ACE inhibitors, calcium channel blockers, and a summation of all of these classes of the antihypertensives led to only 22% of the equivalent incremental blood pressure effect of adding a second agent (ie, if a patient on one antihypertensive who has an inadequate lowering of their BP, it is >4-fold as effective to add another low-dose antihypertensive from a different class than to double the dose of the initial med)

-- this current statin study was a systematic review of two studies that had individual-level patient data, allowing evaluation of subsequent clinical outcomes based on their initial treatment allocation

    -- the results of the study suggested that using a low-dose statin with ezetimibe was equally effective to using a high-dose statin (likely because in this study the actual achieved LDL was not so different between the 2 groups); ie, both of the included trials found noninferiority of the alternative strategy

    -- but, the adverse effects were considerably less in the former group

-- by the way, another recent study assessed the comparative effectiveness and safety of atorvastatin vs rosuvastatin (lipid effectiveness of rosuv vs atorv AIM2024 in dropbox, or doi:10.7326/M24-0178). this was a large (285,680 participants) data-mining study of 2 databases (China Renal Data System and UK Biobank), finding that the 6-year all-cause mortality was lower with rosuvastatin (2.57 vs. 2.83 per100 person-years in the CRDS database and 0.66 vs.0.90 per 100 person-years in the UKB database), and rosuvastatin conferred lower risks for major adverse cardiovascular events and major adverse liver outcomes. In the UKB database, the risk of development of type 2 diabetes was higher with rosuvastatin, and the 2 medications carried similar risks for development of chronic kidney disease and other statin-related adverse effects. of course, these were observational retrospective studies and there may well have been unanticipated and uncontrolled confounders, so their results cannot be considered causal, just associations

Limitations:
-- the major limitation to me is that of these two randomized control trials, the LODESTAR one did not assess the actual intervention these researchers stated, which was a real comparison of moderate doses of statin with ezetimibe versus high intensity statins by themselves, as a means to identify both cardiovascular effectiveness and adverse effects of these 2 strategies

    -- in fact, the LODESTAR trial actually did not recommend that clinicians prescribe ezetimibe (the alternative lipid-lowering med), leading to very few patients being on ezetimibe and actually having more of the participants on ezetimibe being on high intensity statins. why were those individuals put on ezetimibe when the study did not encourage them to use this med? what dose of which statin were they on beforehand? the current authors of this systematic review do comment that "the use of ezetimibe was consistently higher in the alternative strategy group compared with the high-intensity statin strategy group". This is really misleading: that was true in the RACING study, but in the LODESTAR study, there were few on ezetimibe and more so in those on high-intensity statins. This latter comment reinforces (misrepresents) that there were 2 studies that used ezetimibe specifically in those on lower intensity statins, when there was pretty clearly only one study in this "meta-analysis", the second study only muddying the waters....

    -- also, both of the included studies in the current meta-analysis were open-label, decreasing their clinical significance by introducing a potential bias in that the participants and researchers knew the participant group allocation

    -- both were done in East Asian patients, which might well limit generalizability to other populations given the different cultural/medical/risk factor differences (including diet, exercise, overweight/obesity etc); and the studies had exclusion criteria that might limit generalizability to the patients we see (LODESTAR excluded "patients with risk factors for myopathy, hereditary muscle disorder, hypothyroidism, alcohol use disorder, severe hepatic dysfunction (3 times the normal reference values), or rhabdomyolysis")

    -- as an aside, this is about my 1760^th blog since i started primary care blogging about 8 years ago. over that protracted time, i have found a few really shocking things:

        -- it was not uncommon to find that references cited in the article being analyzed and were used by the authors to make an important point actually did NOT make that point

        -- the cited reference occasionally made the opposite point

        -- or the authors citing the study really exaggerated the point of the cited article

        -- or the citation was totally irrelevant to the comment (ie, the citation had nothing to do with the comment made)

            -- of course, i do not check out the references routinely, though i often do if the point being made was really a pivotal one and/or it was different from what i had thought

        -- it really is not so common that citations or statements in the article are not actually relevant or true, but i have seen a pretty large number of cases of misrepresentation

        -- and, i would have thought that the editors, especially at highly academic, peer-reviewed journals, would be really making sure the stated "facts" were accurate

        -- and this goes for many of the top-tiered journal.

--as with other meta-analyses, the researchers are combining different studies with different methodologies and study designs, different inclusion/exclusion criteria of the participants, different demographics and different comorbidities, different numbers of participants and different endpoints. clearly, meta-analyses are important to help understand the breadth of the medical literature with the enhanced power of more participants (instead of relying on one study). But different meta-analyses can find very different results depending on what studies they decide to include. a few years ago, there were 2 meta-analyses in the same highly ranked journal (as i remember, both were on non-opioid medical therapies to treat chronic pain) and they came to very different conclusions, likely because they included different primary studies.....  ).  so, meta-analyses are useful but we should still have a critical assessment of them.....

-- there was also "contamination" within the current study: about 30% of those in the alternative strategy group were taking high-intensity statins over the 3 years of the study and about 8% of those in the high-intensity statin group were not taking high-intensity statins; ezetimibe also had some unevenness, only 55% in the alternative strategy group were on it and about 5% in the high-intensity group were (perhaps by adding the LODESTAR group to the combo, which as mentioned above just "muddied the waters")

-- one problem with a short-term study, as in the 3-year outcomes of the current one, is that it tends to underestimate the benefits of the statins since that group of patients will include people with very different stages of ASCVD intensity (and those with less severe disease will likely require more time to have a subsequent cardiovascular event), yet the adverse events from the statin may reveal themselves earlier in the course of therapy. 

-- a large study (https://www.ahajournals.org/doi/full/10.1161/JAHA.118.011320), for example, found that most patients who had diabetes that appeared to be related to statins had the diabetes within the first 3-4 years, with increasing diabetes over time (though there was no real difference in cardiovascular outcomes despite the fact that there was a statistically significant though small different achieved LDL (alternative group 64.8 vs 68.5 mg/dL in the intensive treatment group): 

-- this current assessment included patients with ASCVD (secondary prevention). however, one cannot necessarily assume that the results are generalizable to those people on statins for primary prevention

so,

-- there are very clear limitations to this meta-analysis, since only one study actually performed the appropriate analysis to justify its inclusion (a meta-analysis of one study???)

-- that being said, it is very likely that the actual results of this meta-analysis are reasonable and generalizable based on the above commentary: ezetimibe is quite effective in lowering LDL cholesterol levels and decreasing adverse cardiovascular events when adding it to statins

    -- and, to add my completely anecdotal approach, i have found that adding ezetimibe to less than the most potent statin yields really very impressive improvement of the lipid pattern

-- another issue is that statin intolerance increases with statin dose (about 5% overall, but increasing with statin dose), and this current study found a 42% decrease in statin dose in those patients in the alternate therapy group

-- in terms of my commentary above of the inaccuracy of some of the statements and citations in some articles, this is a real issue. No easy answer, since it seems to happen (in my experience) pretty much across the top-tier of medical journals. we just need to be skeptical of strong assertions that are either pivotal to the argument in the article or are questionable based on our prior knowledge or experience....

geoff

-----------------------------------

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

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

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