PCSK9 Inhibitors: Is There a Palatable Price Point?

FOURIER: Prior MI and PAD Subgroups

Michelle L. O'Donoghue, MD, MPH; Jennifer G. Robinson, MD, MPH; David J. Cohen, MD, MSc


December 07, 2017

Michelle L. O'Donoghue, MD, MPH: Hi. I am Michelle O'Donoghue of Harvard Medical School and Brigham and Women's Hospital in Boston. I am at the American Heart Association (AHA) Scientific Sessions in Anaheim, where I am pleased to be joined by David Cohen, who is the newly appointed vice chairman in medicine for research at the Beth Israel Deaconess Medical Center in Boston; as well as Jennifer Robinson, who is professor of medicine at the University of Iowa in Iowa City. Welcome, Jennifer, and welcome, David.

Jennifer G. Robinson, MD, MPH: Thanks.

David J. Cohen, MD, MSc: Thanks very much.

Dr O'Donoghue: It's good to have you here. We have had a couple of interesting presentations here at the AHA meeting regarding the PCSK9 inhibitors. Jennifer, I know that this is an area that you have given a lot of thought to. What are your thoughts about the two new analyses that we have from FOURIER?

Getting to a Cost-Effective NNT

Dr Robinson: There were two analyses that I think are important, where they looked at subgroups in FOURIER. FOURIER was a trial of evolocumab, one of the PCSK9 monoclonal antibodies versus placebo on top of high-intensity or moderate-intensity statin therapy in high-risk people[1]—that is, people with cardiovascular disease plus something else. They are already a high-risk population.

It did indeed find that there was an additional risk reduction from adding evolocumab to statin therapy of about 20% for atherosclerotic cardiovascular disease (ASCVD) events over about a 2.5-year period.

"They are not cost-effective in the regular FOURIER population, even with discounting."

To show that you could add anything to statins was critically important because there had been a number of drugs that failed. But it was also underwhelming because, from some of the earlier trials, we were expecting much larger relative risk reductions; we had seen 50% reductions in the early trials. Why was that? The trial was relatively short, so I was not expecting a lot more than a 25% reduction. I think baseline low-density lipoprotein cholesterol (LDL-C) level also has a lot to do with it.

There are two new analyses from FOURIER: one on patients with cardiovascular disease (CVD) plus peripheral vascular disease[2,3] (ie, polyvascular disease), and one on those with CVD plus recurrent myocardial infarction.[4] The patients with polyvascular disease plus CVD have an extrapolated 10-year ASCVD risk of 65%. That is quite remarkable compared to around 35% for the FOURIER population as a whole

As a consequence, there was a slightly greater relative risk reduction of about 25% [with PCSK9 inhibitor therapy]. Putting these two analyses in context, it means that you have much better absolute risk reductions (ARR) in these very, very high-risk patients. That is all very complicated, so it's easier to think of a number needed to treat (NNT). The NNT is calculated from the inverse of the absolute risk reduction (1/ARR). It is a really nice number because it tells you the risk of the patient and the benefit from therapy.

The NNTs for these very high-risk subgroups is 8 to 10 [over 5 years]. The higher the LDL-C, the lower the NNT. Because of the cost of these drugs, it has been a big problem. They are not cost-effective in the regular FOURIER population, even with discounting. On my back-of-the-envelope calculations, PCSK9 inhibitors [in these very high-risk populations] are cost-effective by anybody's parameters, at $50,000 per quality-adjusted life-year (QUALY), probably with sufficient discounting. It makes these drugs much more digestible for payers to be thinking about. They are on par with other drugs; $50,000 is pretty good and is in the range of other on-patent drugs.

The AHA ranked that threshold as a good value.[5] When you start getting to over $100,000, and over $150,000 per QUALY, the ranking is poor value. I was quite excited about these results.

Dr Cohen: That was an NNT over how many years?

Dr Robinson: Five years. That's a great point

Dr O'Donoghue: It is extrapolated.

Dr Robinson: It is all arithmetic, but on the other hand, I think it gives us a much better idea of whom to target with these drugs. Payers do not want to pay more; we discussed it earlier about how much we pay for our premiums. The payer has a fixed amount of money, and if you pay for one thing, it displaces something else.

Dr Cohen: Something is not going to get paid.

Dr O'Donoghue: Certainly, David, you have given a lot of thought to cost-effectiveness and the discussions regarding the price tag for the PCSK9 inhibitor class. How do you think about it?

Dr Cohen: I think Jennifer laid it out pretty well. Again, we have thresholds that the AHA has endorsed as being reasonably cost-effective, and we know from the analyses that have been published that for the overall group that was studied in FOURIER and the cost-effectiveness threshold, ratios were in excess of $300,000.[6]

Dr Robinson: Even the ones done by the company were about $150,000. I will say, though, that FOURIER is the wrong population [to assess cost-effectiveness]. We had to show incremental benefit on top of best therapy. That was the requirement. That is not the population we want to treat. We want to treat people with high cholesterol with these drugs.

Part of an analysis that we did in JACC, in 2016,[7] showed that once you see LDL-C levels over about 100 mg/dL in very high-risk people, you are starting to approach cost-effectiveness with discounting. It is not titrating to goal. It is not treating people to get their LDL-C down to 30 mg/dL; it is actually thinking about the threshold.

Dr Cohen: Starting with people whose LDL-C level is a problem.

Dr Robinson: Right—people with CVD with pretty well-controlled risk factors. They do not have diabetes, chronic kidney disease, or polyvascular disease; the run-of-the mill patient with an LDL over 130 mg/dL is the threshold where things start looking cost-effective. The NNTs are very acceptable—they are about 20 to 25. Even patients would think that, right?

Dr Cohen: We do not usually have drugs that cost $14,000.

Dr Robinson: In the cardiovascular space.

Dr Cohen: Correct. We do in cancer.

CVD-Plus and Risk Scores

Dr O'Donoghue: I had two interesting observations from the peripheral arterial disease (PAD) analysis of FOURIER. One was that the drug appeared to have a profound effect on reducing peripheral vascular events, the major adverse limb events (MALE).

And second, when you looked at the event rates for the different patient groups, it was interesting that patients with PAD who did not have known CAD or other atherosclerotic disease actually had a higher event rate than the patients who had the established CAD. It suggests that PAD really is a powerful marker of risk in and of itself.

Dr Robinson: It is the polyvascular disease, because having CVD plus PAD was associated with a 10-year ASCVD risk of 65%. I was asking the sponsors about how to get this message out. You have so few options for limb preservation in polyvascular disease patients. This is, in a way, a game changer.

Dr O'Donoghue: Jennifer, based on the new data that have been presented, how are you thinking about whom you target for these drugs? You mentioned LDL-C as being one parameter to determine who might benefit more from a PCSK9 inhibitor, but are you using polyvascular disease as a marker? Number of prior myocardial infarctions? Some of the other data that were presented?

Dr Robinson: In the JACC paper from 2016, we identified several groups—we called them "CVD-plus," so CVD plus diabetes, or plus chronic kidney disease, or plus polyvascular disease—we even found the signal back then. All of those high-risk groups have at least a 30% 10-year ASCVD risk on a good dose of a statin. That is where the benefit really is for these drugs. That is the only population that really approaches the realm of what payers are willing to pay. The higher you go on LDL-C, the more cost-effective they are, which makes sense. It is a cholesterol-lowering drug. Absolute risk [is important], but actually, cholesterol plays a role as well.

We did not touch on familial hypercholesterolemia (FH). The patient with CVD and FH is probably a CVD-plus-plus. In our evidence review, they have a 10-year ASCVD risk of about 40% to 45%. I'd like to think that we want to treat a curable genetic fatal disorder regardless of how much it costs, but that being said, PCSK9 inhibitor therapy is actually very cost-effective in those FH patients with CVD. Doctors need to do the prior authorization and keep hammering away at getting it for those patients.

That is the other possibility: If the price came down a lot, so many more patients could be on it.

Dr Cohen: It seems like there is a step further that could be taken, and hopefully the FOURIER investigators are doing this: Where is the risk score? That is what we need. So far we are talking about isolated single factors—PAD or the LDL-C. These days, everyone wants a risk score.

Dr Robinson: There is the TIMI Risk Score for Secondary Prevention that they did after IMPROVE-IT.[8] I imagine that there will be something similar. We are working on something similar in a broader population than FOURIER participants, with a broader range of LDL-C levels. We need to come up with some NNT; it's a way to have a conversation with payers to negotiate some price that makes sense for them, and then we just agree that these are the people we are going to treat based on this value.

Dr O'Donoghue: I think both the clinicians and the payers are looking for that, because it becomes challenging to think about at the current price, to have very universal use.

Dr Cohen: Well, there is another solution—you just said it.

Dr O'Donoghue: Reducing the price?

Dr Cohen: Yes, exactly. That is the other possibility: If the price came down a lot, so many more patients could be on it.

Dr Robinson: Just to be clear, it is about $14,000 straight off the street. We have a very arcane pricing system—I am going to stop there before saying what I really think about it—but there are middlemen who derive value that is not transmitted to the patient. The companies are getting about $9500. There is discounting, but it is not to the level where it needs to be or where it is in other countries; it is about $7500 US. They have different programs to cover people in the donut holes. It would be nice if they would have thought a little bit more about the pricing.

Dr Cohen: They must have thought about it.

Dr O'Donoghue: I am sure they gave a lot of careful thought to it.

Dr Cohen: You would think they would notice that it pushes the limits on cardiovascular drug pricing because the populations are so large. That is why it is the issue. It's not because the value is not there; it's because we have millions and millions of people.

Dr Robinson: You cannot treat 40 million Americans with CVD with a drug that costs $14,000 a year. It is a similar issue with Entresto® (sacubitril/valsartan). That is very expensive and it is not being sold, and it is for a selective population. I am not sure; maybe you have some insights into that.

Dr Cohen: I have not studied Entresto very much. It is being used some, but there is certainly a learning curve for new drugs as they come out. I was very involved in a number of the antiplatelet drugs, like prasugrel and ticagrelor, and all of the companies were similarly discouraged when they first came out. Over time, people become more accepting of them; they understand the risks and benefits. There is more of a lag [in uptake] than the companies think, but the price absolutely matters these days.

Dr Robinson: It is a new price for a chronic drug in the cardiovascular space. It may change as guidelines may change over time [and we learn] the niche for these drugs.

Dr O'Donoghue: It will continue to change. Thank you, Jennifer and David, for a terrific discussion about this important topic.

Dr Robinson: Thank you.

Dr Cohen: Thanks.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.