Perspectives of a Lipidologist: LDL Cholesterol Testing, PCSK9 Inhibitors
OVER THE PAST YEAR, new evidence for the cholesterol- fighting class of therapy called proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors has generated discussion about the usefulness of these powerful drugs in diabetes care. The link between diabetes and cardiovascular risk is well known, and this is central to the FDA’s 2008 requirement that sponsors of all new glucose-lowering therapies conduct cardiovascular outcomes trials to ensure that these drugs do not cause heart attacks or strokes.1
“The whole issue of cardiovascular disease in patients with diabetes is one that we have known of for decades, and it is so important,” said Eliot A. Brinton, MD, FAHA, FNLA, president of the Utah Lipid Center in Salt Lake City, in an interview with Evidence-Based Diabetes ManagementTM (EBDMTM). “It has always been the primary cause of death in patients with type 2 diabetes, and also the primary source of morbidity.”
Cholesterol management in diabetes, meanwhile, can be complicated. Diabetic dyslipidemia is a common condition; the best-known form in type 2 diabetes (T2D) is marked by high levels of triglycerides and decreased high-density lipoprotein (HDL) cholesterol. “Changes [are] observed many years before the onset of clinically relevant hyperglycemia,” wrote Jonathan D. Schofield, PhD, and his co-authors, in Diabetes Therapy. As they explained, these decreased levels of HDL cholesterol are especially relevant clinically for those at high cardiometabolic risk, and this may justify aggressively treating low-density lipoprotein (LDL) cholesterol levels as well.2
When PCSK9 inhibitors burst on the scene, the focus was on dramatically reduced levels of LDL cholesterol—up to 60% in clinical trials.3,4 At first, specific benefits to patients with diabetes were not highlighted. But as Schofield and colleagues wrote in spring 2016, there were suggestions that these patients would benefit as much as patients without diabetes.2
Fast forward to 2017, and more PCSK9 inhibitor results are coming in, both for cardiovascular benefits generally and for patients with diabetes specifically. March brought the first results from the FOURIER trial, presented at the American College of Cardiology Scientific Sessions, which found that Amgen’s evolocumab, sold as Repatha, reduced major cardiovascular events— first heart attacks, strokes, and cardiovascular death—by 20% overall and by 25% after the first year.5
Next, at the American Diabetes Association Scientific Sessions in June, came a pair of results for rival PCSK9 inhibitor alirocumab (Praluent, from Sanofi-Regeneron) that showed a 49% LDL cholesterol reduction for patients with T2D and a 32.5% reduction for patients with dyslipidemia.6 Finally, follow-up results from FOURIER presented in September at the European Association for the Study of Diabetes Annual Meeting showed that those with diabetes who took evolocumab had a 57% reduction in LDL cholesterol.7
An accompanying editorial in The Lancet noted, “The focus of cardiovascular disease prevention has shifted from normalization of risk factors to absolute risk reduction. Reducing LDL cholesterol concentration by 1 mmol/L for about 5 years is consistently associated with a 23%-25% lowered risk of major cardiovascular events, with statin and nonstatin therapies alike, irrespective of baseline LDL cholesterol concentration.”8
The potential to reduce LDL cholesterol to levels never imagined in at-risk patients brought with it the need for more accurate tests, and in September, Quest Diagnostics, the major testing company, announced it would be adopting a test developed at Johns Hopkins University to replace an older method, known as the Friedewald calculation.9 According to Brinton, the Friedewald method is simply not accurate once LDL cholesterol reaches the levels achieved with PCSK9 inhibitors.
EBDMTM spoke with Brinton about the importance of the new test, what the latest results from FOURIER tell us about LDL cholesterol levels, and payer issues with PCSK9 inhibitors.
Advances in LDL Cholesterol Testing
EBDMTM:Can you describe the difference between the Friedewald calculation for measuring levels of LDL cholesterol and the new method developed at Johns Hopkins now in use by Quest Diagnostics?
BRINTON: The Friedewald equation is something that we’ve used for decades for the calculation of LDL cholesterol. The parameters that are measured are total cholesterol, total triglycerides, and HDL cholesterol. From those 3 measurements, the Friedewald equation will derive an LDL cholesterol level, which is the primary number of interest in terms of the lipid panel and the treatment of lipid disorders.
What has been known for many, many years is that the Friedewald equation is not very accurate. It’s kind of one size fits all. The triglyceride levels are divided by 5 to calculate [very low-density lipoprotein] cholesterol; then LDL cholesterol is calculated by difference. This we’ve known is not a very good calculation. The researchers at Johns Hopkins looked very carefully at the data, where they actually had directly measured LDL—and they said, ‘If we directly measured LDL, what is the best equation that would give us the best estimation of that directly measured LDL without having to do a separate measurement?’ And they came out with a complicated way of calculating the direct LDL—or calculating LDL as if it were measured directly—and in that process, we’ve solved some problems.
Specifically, we know Friedewald does not work well when LDL cholesterol levels are low. This is important, because we are becoming much more aggressive in treating LDL levels. With the advent of the PCSK9 inhibitor class we now have on-treatment LDLs of 40 [mg/dL] or 20 [mg/dL] or less. So, we get a much better LDL level in that setting of super-aggressive LDL treatment.
Another setting where the Friedewald performs poorly is with triglycerides that are elevated. In fact, nobody uses Friedewald where the triglycerides are above 400 [mg/dL]. Well, the problem is that long before you get to 400 [mg/dL], the Friedewald is already going off track and not giving us a very good LDL level, and then above 400 [mg/dL] you can’t even use it at all. In the setting of high triglycerides, this is a large step forward.
And the reason that’s important is 2-fold: (1) High triglycerides are becoming more common, more prevalent, as we have more obesity and more diabetes. Both are becoming more prevalent, in part because of aging of the US population, in part because of other things that are happening. So, having a more accurate LDL cholesterol in the setting where the triglycerides are approaching, or exceeding, 400 [mg/dL] is extremely helpful. (2) Another development is that we’ve learned more recently about the importance of hypertriglyceridemia as a risk factor for cardiovascular disease. As we’re paying more attention to patients with high triglycerides, we’re looking at that triglyceride level, looking at a more accurate calculation of the LDL cholesterol level, which is very, very helpful.
EBDMTM: Can you discuss the significance of not having to fast before taking this new test?
BRINTON: Another thing is development of nonfasting versus fasting lipid panels, and specifically triglyceride levels. For the longest time, our standard has been a fasting lipid panel. Recently, studies have shown that a nonfasting lipid panel can be helpful— specifically, nonfasting triglycerides. But also realize that one of the problems of using a nonfasting sample is that this tends to skew the LDL cholesterol level results if you calculate by Friedewald. And so, having a better calculation of LDL in the setting of increasingly popular nonfasting lipid testing is another big advance.
EBDMTM: Besides a more accurate measure of LDL cholesterol in a nonfasting setting, why does the new test offer greater convenience? Why is this important?
BRINTON: It’s a convenience for the patient to be able to have the testing done nonfasting. Again, our standard is fasting, and there’s nothing wrong with fasting. I’m a lipidologist—I do a lot of my testing, even most of my testing, fasting. But for the primary care doctor, for somebody not just focused on lipids perse, it can be very helpful to be able to do a nonfasting [blood] draw and then good lipid panel results. So, for the convenience of the physician in terms of not having to order a test in advance of the visit, or for the patient of not having to fast until 3 or 4 in the afternoon, if that’s when the draw is going to take place, it is a major step forward to get accurate results with a nonfasting sample.
EBDMTM:Is there an advantage for payers?
BRINTON: The advantages for payers are more indirect, in the sense that there’s going to be less redundant lipid testing; maybe we’re going to have a better ability to calibrate our treatment. Certainly, with LDL being our primary target of treatment, we’re having a renaissance of LDL goals in the United States. It is helpful to the payer to have the doctor understand what is the best estimate of the LDL cholesterol, and therefore, treatment with various pharmaceutical agents. And of course, diet and lifestyle can be better focused where it needs to be focused, if we have a better understanding of how low that patient’s LDL cholesterol truly is. Then I think it’s easier to use the treatments that are truly necessary or not use the treatments that may not be necessary. So, there is a benefit to the payer for sure, not only a modest benefit in terms of the testing, but also a benefit in terms of the use of the pharmaceuticals.
Perspectives on PCSK9 Inhibitors
EBDMTM:What do the recent results published in The Lancet and presented at the European Society of Cardiology imply about the advantages of PCSK9 inhibitors?
BRINTON: The FOURIER trial, which was the main study of one of the PCSK9 inhibitors, evolocumab, is a landmark trial. It is the first clinical cardiovascular outcomes trial looking at this new class—there’s another on its way; we’re going to have results in the next few months—but just taking the FOURIER trial, first, what we see is proof of principle. Adding a PCSK9 to a statin is helpful in terms of additional risk reduction beyond what is available with a statin alone. But then, a very large piece, and this is both clinically and scientifically important, is the question of the relationship between LDL cholesterol measured by whatever means—usually the Friedewald—the relationship between LDL cholesterol on treatment and cardiovascular outcomes. The biggest question here is: Can LDL be too low? Is there harm, or maybe lack of benefit, in having that LDL go ever lower? And the answer from the FOURIER data is “no.”10 The benefit is there, and the harm is not there. There appears to be no such thing as [an LDL] cholesterol that is too low.
[This is] in contrast to glucose, where we have hypoglycemia, a very serious concern clinically, and hypotension, also very serious. [Those] are life-threatening situations if we are not careful. [But] for LDL cholesterol, we don’t appear to have any downside for getting that LDL below 40 [mg/dL] or even below 20 [mg/dL]. FOURIER has really given us convincing data for additional benefit and no additional harm, even for an LDL in the less-than-20 [mg/dL] range. We’ve been concerned about this. I think FOURIER is of an adequate size and adequate power to give us confidence as we go forward to ever-more-aggressive LDL-lowering treatment.
EBDMTM: At the American College of Cardiology Scientific Sessions in March, physicians took note of the cardiovascular benefits that were seen with Repatha as well as studies that showed the difficulty clinicians experience when prescribing PCSK9 inhibitors.11 Has the reimbursement experience changed in recent months for this class of therapy?
BRINTON: We have a wonderful new tool in the field of lipid management and that is this new class of drugs, the PCSK9 inhibitors. They are marvelous because they are so effective in lowering LDL. They work really nicely on top of a statin; if we have someone who is statin-intolerant we are still allowed to use them as long as we have given statins a fair chance. So, then the question is: Can we actually have access to these drugs? And because they are expensive, access has been a big issue. The managed-care people are able to put a few roadblocks in the way—a little extra paperwork—that does make it harder for clinicians to write a PCSK9 inhibitor. Is that getting better? Is it still an issue?
The quick answer is, in my experience, it has changed very little. Maybe we have changed a little in terms of our approach to this, possibly some of the payers are little more relaxed. We do have convincing trial data now with the FOURIER study with evolocumab; we should have similar data soon with the other drug, which is alirocumab. I’m hopeful that this will become less of an issue. But it remains a very important fact that just simply writing a prescription for a PCSK9 inhibitor does not give one immediate access to the drug. On the one hand, we know that we shouldn’t just be using this drug widely for everyone who happens to have high cholesterol. On the flip side, if we have patients who really need additional LDL lowering, [if] we have done everything possible with a statin, we’re certainly considering, in most of these cases, use of ezetimibe. [That] has now gone generic, so we have greater access to ezetimibe. In the case where patients still have a high residual LDL cholesterol level and they have high residual cardiovascular risk, then adding a PCSK9 inhibitor is a good thing to do, and access remains an important issue.