John M. Mandrola, MD: Hi, everyone. This is John Mandrola from theheart.org on Medscape. I'm here in Rome, Italy, at the European Society of Cardiology (ESC) Meeting. I'm very excited to welcome Brian Ference, who is an associate professor at Wayne State University and the lead investigator of an exciting Hot Line trial. This trial evaluated lower LDL cholesterol (LDL-C) and blood pressure levels, and exposure to these lower levels over the course of long periods of time. Dr Ference, welcome.
Brian A Ference, MD, MPhil, MSc: It's a pleasure to be here. Thank you for having me.
Lower LDL-C and Lower SBP: Does It Make a Difference?
Dr Mandrola: Tell us about the background of this trial first. What got you started?
Dr Ference: As you know, adults who are able to maintain ideal levels of risk factors throughout the whole of their lifetime have very low lifetime risks for cardiovascular disease (CVD). But fewer than 5% of persons are able to maintain those ideal levels, suggesting that this may not be an attainable goal for most people.
By contrast, Mendelian randomization studies have suggested that both LDL-C and systolic blood pressure (SBP) are not only causally associated with the risk for disease, but they have cumulative benefits over time.[2,3,4] Focusing on maintaining low levels of LDL-C and low levels of SBP beginning earlier in life than we do at the moment may be an effective strategy for preventing CVD.
However, the causal effect of combined exposure to both lower LDL-C and lower SBP on the risk for events is unknown. That uncertainty has been exacerbated recently by the results of the landmark HOPE-3 trial,[5,6,7] which was an important and really well-done trial. That study reported that the combination of a lipid-lowering agent and a modest blood pressure-reducing regimen did not reduce CV events more than lipid-lowering agents alone. Some have questioned the synergistic effect between blood pressure and LDL-C on the risk for CV events, particularly among persons who do not have hypertension.
Dr Mandrola: The key aspect of what you are looking at is the combination of blood pressure and LDL-C.
Dr Ference: That is right. The objective of our study was to estimate the causal effect of combined exposure to both lower LDL-C and lower SBP on the risk for CV events. The only way to do that is with either a 2x2 factorial randomized trial, like HOPE-3, or 2x2 factorial Mendelian randomization, which is the approach we took.
Dr Mandrola: Why don't we pause for a second so you can educate us about Mendelian randomization.
What Is Mendelian Randomization?
Dr Ference: Thanks for the opportunity. Mendelian randomization is merely a randomization scheme. It is an attempt to introduce a randomization scheme into observational research to allow us to make causal, unconfounded inferences about the effects of biomarkers, not the effects of polymorphisms. Ironically, it has very little to do with genetics.
Perhaps the easiest way to explain what Mendelian randomization is is by way of analogy with a randomized trial—after all, it has been called "nature's randomized trial." For example, there are many polymorphisms associated with LDL-C. Each of those polymorphisms is allocated approximately randomly at the time of conception in a process sometimes referred to as Mendelian randomization. Inheriting an allele associated with lower LDL-C is analogous to being randomly allocated to an LDL-C–lowering therapy, while inheriting the other allele is analogous to being randomly allocated to usual care. This allocation is indeed random. The only difference between those two groups should be the LDL-C. We can then follow people forward in time and make unconfounded causal estimates of the effect of LDL-C, not the polymorphism. The effect of LDL-C on the risk for CVD [is] analogous to a long-term randomized trial.
Dr Mandrola: In a randomized trial, the randomization process takes care of the random allocation. But this is more of a natural allocation.
Dr Ference: Exactly. The whole purpose of randomization in whatever experimental paradigm one chooses is to evenly allocate all of the potential confounders—those that are known and those that are unknown. We create two groups that are exactly identical except for the factor that we are interested in. A randomized trial is generally done with a randomization scheme using a random number generator or some number of other mechanisms.
Here, we use polymorphisms in combination to randomly allocate people to higher or lower SBP and LDL-C. But it's the same process. The whole purpose is to allocate people into two exactly equal groups. The only difference is the LDL-C or the SBP.
Dr Mandrola: Where did you acquire the patients?
Dr Ference: We had 102,000 patients with individual-level data who had previously been enrolled in one of 14 prospective cohort or epidemiologic cohort studies as part of the National Heart, Lung, and Blood Institute's Database of Genotypes and Phenotypes (dbGaP).
Dr Mandrola: Were these really observational data that you could have turned into a randomized trial?
Dr Ference: We introduced the randomization scheme naturally to make unconfounded causal estimates of the effect of combination lower LDL-C and lower SBP on the risk for events.
Dr Mandrola: What happened with these patients?
Dr Ference: Over the course of the study, allocation to higher or lower blood pressure or higher or lower LDL really constitutes a lifelong exposure. Over that lifetime of exposure to the 102,000 participants, around 14,350 major vascular events accrued. We confirmed that both LDL-C and SBP have causal and cumulative effects on the risk for CVD. When present together, their effects are independent, multiplicative, and cumulative. And because their effects are multiplicative and cumulative over time, the exposure to a lower LDL-C by 1 mmol/L (18 mg/dL) and a lower SBP by 10 mm Hg was associated with an 86.1% reduction in the lifetime risk for CV events.
That 80%-90% reduction in risk was quite consistent across multiple different composite CV outcomes, including an 84% reduction in CVD death. This translated into a smaller but still robust and significant 36% reduction in all-cause mortality. In addition, the effect of combined exposure to lower LDL-C and lower SBP appeared to be quite consistent across all of the different subgroups we looked at, including, importantly, persons with normal SBP (< 120 mm Hg) and persons with normal LDL-C levels (< 3.4 mmol/L or roughly < 128 mg/dL).
Dr Mandrola: Those are big reductions.
Dr Ference: They are. The apparently outsized effects really highlight the potential to dramatically lower lifetime risk for CVD because the effects are not only causal but also cumulative. When present together, they are multiplicative and cumulative. That suggests that, indeed, CVD is preventable. But we may have to [try to] prevent CVD slightly earlier than we do at the moment.
Dr Mandrola: What is one of the big take-home messages in terms of prevention?
Take Home #1: LDL-C and SBP Independent and Multiplicative
Dr Ference: There are really three take-home messages. The first is that, biologically, LDL-C and SBP have independent and multiplicative effects. This is important because there was some uncertainty about that before we conducted the study.
Dr Mandrola: How does the multiplicative effect relate to HOPE-3?
Dr Ference: As I said, the HOPE-3 trial was an extraordinarily important study. They showed that a minimally lower LDL-C is associated with roughly a 24% reduction in events; it was quite significant and similar to what we see with other statin trials. They showed that blood pressure lowered by 6 mm Hg was associated with a non–statistically significant 7% reduction in CV events. The combination-therapy arm had a 29% reduction in risk that was not statistically significantly greater than the lipid-lowering therapy arm alone. However, if one multiplies the effect observed in the lipid-lowering arm alone by the effect observed in the blood pressure–lowering arm alone, that effect is precisely what was reported for the combination-therapy arm. Although it does not appear so in the first instance, we believe that our data are perfectly consistent with the HOPE-3 trial because it also appeared to report an independent and multiplicative effect of LDL-C and SBP.
Dr Mandrola: One difference you mentioned to me earlier was that your study reflects a longer exposure, perhaps earlier in the stage of the disease than in HOPE-3.
Dr Ference: That is right. HOPE-3 had a short-term follow-up duration—a median of 6 years, I believe, and they had relatively small reductions in SBP. Our data suggest that if one had started the study earlier and had more prolonged exposure, even to that 6 mm Hg lower blood pressure, the observed risk reduction would have been much greater because of the cumulative effects over time.
Take Home #2: LDL-C and SBP and Lifetime CV Risk
Dr Mandrola: What is the second take-home message?
Dr Ference: Because LDL-C and SBP have multiplicative and cumulative effects, long-term exposure to even modestly lower LDL-C and SBP has the potential to result in dramatically lower lifetime risk for CV events even among persons who have normal LDL-C and SBP levels.
Dr Mandrola: How does this relate to drug therapy?
Dr Ference: It's important to note that our study does not contemplate drug therapy, and it does not encourage it either. A good example here is SBP. As we know, SBP starts to rise linearly beginning in middle age and as a result of a progressive cumulative arterial sclerotic process. What we do at the moment is let blood pressure rise to 140 or 150 or 160 mm Hg—representing that cumulative arterial sclerotic process—and then argue about whether we should lower it to 140 or 120 mm Hg. What these data show, and what I did not mention, is that long-term random allocation to lower blood pressure is also associated with a slower rise in blood pressure with age. That implies that if we begin to keep our blood pressure low earlier, we may be able to actually prevent the usual age-related rise in blood pressure and thereby prevent hypertension.
Take Home #3: CV Deaths Are Preventable
Dr Mandrola: What would be your third take-home?
Dr Ference: Our study confirms that CV deaths are largely preventable and that the prevention of CVD can be substantially simplified and improved by designing prevention programs to monitor and keep LDL-C and SBP low beginning very early in adulthood.
Dr Mandrola: Maybe the future will hold some novel ways to do that.
Dr Ference: Inexorably we are moving toward the era of the digital patient where we all wear sensors and we will have continuous biometric monitoring. In some sense, this study provides a roadmap for how we can translate that big data into actionable information. For example, with wearing sensors, rather than simply looking at our sleep patterns or activity levels, we can determine how components of a diet, certain activities, or other lifestyle choices affect our LDL-C and SBP. Informed by that data, we can choose the diet, exercise patterns, or activities that keep our blood pressure and cholesterol at optimal levels, and thereby individualize our long-term strategy to largely eliminate a lifetime [risk] for CVD.
Dr Mandrola: It's remarkable and has remarkable implications for prevention of heart disease going forward. Thank you for joining us.
Dr Ference: It's my pleasure. Thank you for having me.
Dr Mandrola: And that's all from the ESC Meeting. This is John Mandrola from theheart.org on Medscape.
© 2016 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Low LDL-C and SBP Sans Drugs Shows CVD Largely Preventable - Medscape - Sep 15, 2016.