The new proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors can have significant budget effects, depending on the breadth of the US Food and Drug Administration (FDA)’s approved labeling. American Health & Drug Benefits asked Stephen Gorshow, MD, Regional Medical Director, UnitedHealthcare, and James T. Kenney, RPh, MBA, Manager, Specialty and Pharmacy Contracts, Harvard Pilgrim Health Care, to participate in a teleconference to better understand how payers are approaching the management of these agents.
Q Dr Gorshow, what are the greatest unmet needs today in cholesterol management?
Dr Gorshow: From my standpoint, the greatest unmet need is a solution for patient nonadherence. We have insufficient numbers of patients taking their medications for hypercholesterolemia. After 1 year of therapy, approximately 50% of patients stop taking their statins or other relevant medications. This happens for several reasons, but it is a very difficult problem to surmount. This is especially true when talking about medications for a condition that has no symptoms, and when the medication use itself may actually cause new symptoms.
Not too long ago, I was reading a study in the New England Journal of Medicine that revealed a surprising insight: a significant number of people, especially the elderly, said that simply the act of taking medication made them feel as if they were sick, and that they must be sick if they were taking medication, even though it might have been preventive. With the increasing prevalence of accountable care organizations and similar care models, there are opportunities for enhancing patient adherence to all preventive medications. The fact remains that the best medication in the world will not do any good if patients do not take it.
Q Mr Kenney, from Harvard Pilgrim’s standpoint, have you been involved in any particularly successful or useful programs for nonadherence, especially related to cholesterol management?
Mr Kenney: AstraZeneca offered an interesting program in conjunction with Eliza Corporation. The program offered several patient interventions by Eliza that were designed to target patients at risk for nonadherence or simply to promote adherence by patients taking a particular drug. Eliza would call patients and try to drive adherence, and the data from Eliza suggest that they were able to measurably improve adherence. Although the program offered the opportunity to improve adherence, you may only be focusing on 1 or 2 drugs at a time in each specific therapeutic area. Typically, a patient with a high cholesterol level has multiple comorbidities and is taking additional drugs; the bigger problem is patient adherence and persistence across all therapies. The real needs are to improve adherence for all medications and to increase individual patient engagement in all his or her disease states.
Q What may be the effect of the new injectable PCSK9 inhibitors on the problem of patient adherence?
Dr Gorshow: Based only on the fact that they are injectables, we can expect that many people will be averse to taking them, especially for a disease that is asymptomatic (ie, patients never know whether the medication is working). So I do not think that the PCSK9 inhibitors will address this unmet need.
By contrast, having to pay more out of pocket may actually be an incentive for people to take these medications. The patient may say, “I might as well take it, because otherwise I am just flushing my own money down the drain.”
Q The PCSK9 inhibitors will come with autoinjectors, which will make home use fairly convenient. With regard to the costs involved, how do we try to mitigate potential nonadherence and the potential for wastage of this specialty drug?
Mr Kenney: From a health plan perspective, because the PCSK9 inhibitors are biologics and will likely cost more than $600 monthly, we will opt to distribute them only through specialty pharmacies. This would at least provide an additional adherence check, by the specialty pharmacy contacting the patient a couple of weeks into the therapy to ask, “How are you doing? Are you taking your medication? Are you having any problems?” We do see higher adherence rates through specialty pharmacy than in traditional pharmacy dispensing. The simple fact that the patient gets a phone call helps. I do not mean to downplay the value of retail pharmacy, but if every retail pharmacy called a patient to check on the medications, that person’s adherence would probably improve as well.
The bigger issue is, how many physicians will push the use of the PCSK9 inhibitors? If your patient is taking a statin, and that person’s low-density lipoprotein cholesterol (LDL-C) is down to 120 mg/dL but does not quite reach the target goal of 100 mg/dL, will that be good enough? Or will the doctor want to add yet another drug, and one that is an injectable?
The situation is different for patients who have (homozygous or heterozygous) familial hypercholesterolemia (FH). Clearly, these patients will be good candidates for a PCSK9 inhibitor. The patients who fall in the middle will be the big question. From our perspective, we will manage these agents through the pharmacy benefit because of the self-administration requirements.
Q Mr Kenney brought up a very interesting point. Dr Gorshow, do you think clinicians will consider whether patients will actually take the medications?
Dr Gorshow: I think so. In the beginning, the main prescribers will be cardiologists rather than primary care physicians. In fact, the health plan could even restrict the drugs to being prescribed by a specialist. Specialists, who are seeing more patients with hypercholesterolemia and may be prescribing the PCSK9 inhibitors to more of them, may want to implement their own programs to help ensure adherence.
Q How would you define success in improving cholesterol management?
Dr Gorshow: If you would have asked me that question a few years ago, I would have answered it very simply by pointing to the Healthcare Effectiveness Data and Information Set (HEDIS) and Medicare Star scores. Those are the tools that health plans use to measure their success. HEDIS stopped measuring cholesterol level monitoring in 2014; the Medicare Star program focuses on general cholesterol monitoring and adherence and cholesterol measurement in patients with diabetes. That being the case, it will be 2 years before we see any impact of HEDIS no longer measuring this on the Medicare Star Ratings.
Q Let’s shift gears to a discussion about coverage decision-making. We know the PCSK9 inhibitors are very effective in lowering cholesterol levels. It will be several years before we know the resulting cardiovascular and composite outcomes. This scenario is similar to what happened when statins were introduced but their effects on cardiovascular outcomes came many years later. How will this affect coverage decisions today for the PCSK9 inhibitors?
Mr Kenney: We certainly know that lowering LDL-C levels has positive clinical benefits. When ezetimibe added to simvastatin did not reach its outcomes end point in the ENHANCE trial, we did not stop believing that lowering LDL-C levels was beneficial.
I do not think anyone will deny coverage of these drugs outright because of the lack of cardiovascular outcomes data, and the results are expected in 2016. If these drugs demonstrate good LDL-C lowering, there is little reason not to extrapolate the potential benefits of these agents, as we did with statins, and say, “If you lower somebody’s LDL-C level by 50% or 60%, it should have a positive clinical benefit.” We have to accept the drugs as approved and wait for additional clinical data. We will cover them within our benefit designs and apply the appropriate utilization management criteria to manage their use.
However, it will be interesting to see what the outcomes data show, whenever they are released, because the real question is, “How low is low enough?” Is there a point at which LDL-C levels are too low? There are some theoretical concerns about neurologic effects, including neurocognitive issues, when LDL-C levels are below 50 mg/dL, but this is a piece of the puzzle that is yet to be resolved.
Dr Gorshow: I agree in part, but I am a little more cautious in terms of the potential for harm from the drugs. We do not know a great deal about their longer-term effects, and history is littered with drugs that were thought to be wonderful in the beginning and then had to be withdrawn from the market because of effects that did not show up until much later.
I am especially concerned that adverse effects tend to be found early on only where researchers and clinicians are looking for them. Sometimes, adverse effects appear in surprising areas, and if we are not alert for any possible safety signal, no matter how weak it is, then we may miss something important until a larger number of patients have taken the drug.
As a result of this concern, as well as others that include cost, payers will be using as many tools as possible to restrict access to patients within the US Food and Drug Administration labeling. The patients who are getting these drugs should be the ones who stand to benefit the most from them. In that way, we can optimize the cost–benefit of the PCSK9 inhibitors.
Q Let’s explore that potential adverse effect. What happens if the LDL-C level becomes too low? What type of alerts exist today for patients with hypocholesterolemia and the possible negative effects that may have? Do we have the ability today to detect this?
Dr Gorshow: There have been many reports of low cholesterol levels being associated with depression, suicidal behavior, and even overaggressiveness or psychocognitive issues. These are real concerns, magnified by the possibility of much lower cholesterol levels with the use of PCSK9 inhibitors than we have seen with statins. It certainly is a real concern.
Mr Kenney: I can speak to that from our health plan system. We actually have an outcomes contract in cardiovascular disease, in which we pull LDL-C laboratory data as part of the contract. However, we do not have LDL-C laboratory data for every patient.
I agree that most providers will not necessarily be looking for patients with very low LDL-C levels other than how many patients got to their goal. That is the extent of what we, as health plans, typically monitor. We say, “This person is down to 70 mg/dL or this person is down to 100 mg/dL. They both met their goal.” We do not currently focus on how low the patient gets, and rather we target the number of patients who achieve their LDL-C goal.
The physician will need to consider low LDL-C values, and whether this finding is a major concern. Some patients are genetically predisposed to very low LDL-C levels. Are they a proxy for patients whose LDL-C levels are reduced this dramatically by drug therapy? We do not know. The drug manufacturers may tell us that there are people with very low LDL-C levels who live normal lives. But what about the behavior issues that Dr Gorshow mentioned? Are those associated with chemically lowering LDL-C levels as opposed to genetically lowering them? Those are some of the unanswered questions.
Q How easy or difficult will it be, Dr Gorshow, to make that connection between someone’s behavioral difficulties and low LDL-C levels because they are receiving a PCSK9 inhibitor?
Dr Gorshow: That question is right on the mark. Obviously, it will be difficult. The older the patient, the more difficult it will be to distinguish changes in neurocognitive function caused by other conditions versus those associated with drug therapy. I have real doubts that we will be able to show this link if it exists. Much will depend on one’s level of concern about, or sensitivity to, this possible effect.
Q The 2 prototypical PCSK9 inhibitors were launched at approximately the same time. Outside of the recent sofosbuvir (Sovaldi) versus ledipasvir plus sofosbuvir (Harvoni) scenario in hepatitis C, this does not happen very often. How do you think this will affect competition?
Mr Kenney: Each of the 2 companies has contracting experience in managed care; however, they have not been competing with similar therapies, and this gives us some leverage in negotiation. That being said, I cannot help but compare this category with the hepatitis C situation of AbbVie versus Gilead. In that case, AbbVie had a drug that was not as user-friendly as Gilead’s drug, so AbbVie felt compelled to offer a lower price, which it did. AbbVie also appeared to offer an aggressive contract strategy, which was evidenced by the quick uptake by Express Scripts.
In the case of Sanofi and Amgen, we did see shadow pricing around WAC (wholesale acquisition cost) and each company can provide varying levels of contract discounts to achieve formulary access. In a couple of years, if Pfizer launches its own PCSK9 inhibitor drug, then the market will become even more competitive: Pfizer is more experienced in contracting in a very competitive market than the other 2 companies. I did not expect to see net discounts of approximately 48% across all payers as was the case with the hepatitis C drugs. We chose to select evolocumab (Repatha) as a sole preferred agent and included an outcomes component to the contract for patient response to treatment. We will look forward to the pending cardiovascular outcomes data as an opportunity to validate our own experience with this drug.
Q What about the general reaction to the initial announced launch price of Sovaldi? I do not think that anybody has expressed a negative reaction, based on the initial pricing of either PCSK9 inhibitor.
Mr Kenney: We need to determine how big the population will ultimately be. To Dr Gorshow’s point, we are going to restrict it to specialists as well. We will apply utilization management tools to manage the early uptake and to make sure that the prescriptions are targeted to the most appropriate population.
If the indication is broad, for patients who are statin-intolerant or for patients who are not at goal, this will be more challenging. We will model out our expected population and establish criteria to properly identify these additional groups of patients.
Q Assuming the broadest indication, how will payers structure their prior authorization and step therapy tools to address the expected utilization? And will it be enough?
Mr Kenney: Dr Gorshow said it well: we have to work within the label, so we cannot fabricate steps. I think the statin-intolerant question is an interesting one, because 70% of patients who fail one statin have been shown in studies to be able to tolerate a second statin. We do not know what the term “failure” means specifically—maybe patients did not reach their lipid goal because of pruritus, myalgias, or simply fear of the class.
We are not concerned about cardiologists prescribing PCSK9 inhibitors, because we expect them to do the risk calculation and assess their patients to identify the best candidates. If use expands beyond the specialists, then I think our own utilization management efforts will be challenged to meet the potential increase in demand.
Dr Gorshow: Even though it will have the benefits you have described, Mr Kenney, I think that one unknown is how enthusiastically PCSK9 inhibitors will be accepted by cardiologists. My guess is they will be very enthusiastically accepted, and that the prevalence of prescribing within that specialty may be high.
Q Dr Gorshow, will the FH indication spur any interest in genetic testing as one of the prior authorization requirements for the drug?
Dr Gorshow: Most likely a series of requirements would have to be met. Many people can be either diagnosed with or be determined to have an extremely high probability of having FH based on screening factors other than genetic testing. A positive family history, early-onset heart disease, or early-onset hypercholesterolemia can indicate an FH diagnosis. If patients are not meeting all the other non–genetic testing requirements for the diagnosis of FH, then the genetic test will not settle the issue, at least to my understanding.
Having said that, there probably will be some increase in the use of genetic testing. However, payers will be reluctant to require it to dispense the drug. Rather, they will require that an FH diagnosis is very likely.
[Editor’s note: Subsequent to this interview, the FDA approved an indication for patients with FH and for patients with atherosclerotic coronary disease in whom cholesterol levels are not controlled with statins.]
Q Is that partially related to the cost of the genetic testing as well and the relative ease of making a diagnosis of FH?
Dr Gorshow: Cost is certainly a factor. I think, in general, payers have been reluctant to cover genetic testing unless there is a proved or obvious benefit to patient outcomes.
Mr Kenney: The other issue is that genetic testing is not foolproof either. The population with FH has so many genetic variants. If a cardiologist says that this patient had a total cholesterol level of 600 mg/dL, which was reduced to 400 mg/dL with a statin, I don’t think that we need a genetic test to suggest a diagnosis of FH.
Q What do you hear from employers regarding these agents? What are they asking you about them?
Mr Kenney: I have been to 3 employer meetings last summer, and every one of these employers asked about the PCSK9 inhibitors. They want to know what our plans for utilization management will look like. If they happen to have patients with hepatitis C in their mix, they want to know what we are doing about hepatitis C as well.
Employers are very well-informed about specialty drugs, and they want to know our strategy to manage pipeline agents. Our response is that we do not know for certain how we will manage them, but we can assure them that this class will be managed. Without knowing what the final labeling will be, and what limitations we can place on them, we cannot say much else today. They appreciate our proactive approach, but they are definitely very concerned.
Dr Gorshow: The awareness level is definitely high among the employers that I have heard from, and they are concerned. There also will be a great deal of interest in whether large pharmacy benefit managers will exert their influence over pricing, such as Express Scripts and others have recently done with hepatitis C. Will the 800-pound gorillas have some impact on pricing?
Q The PCSK9 inhibitors are not the only new drug class being evaluated for cholesterol management. Another class is the cholesteryl ester transfer protein (CETP) inhibitors. Are health plans today starting to think about new tools to limit the risk for inappropriate use in these new categories overall?
Dr Gorshow: I would have a hard time answering that question other than to say that I think there are already plenty of other examples of drugs that fall into that paradigm. To that extent, I know that they are a work in progress.
Mr Kenney: We are not developing new tools yet. The CETP inhibitors raise high-density lipoprotein cholesterol levels, which is another parameter that adds complexity to treatment decisions. This could be taken in combination with the PCSK9 inhibitors and statins, or it could become part of a stepped algorithm for cholesterol management.
It is a little too early to figure out that question, but we will apply a similar approach to the CETP inhibitors in the near future.
Author Disclosure Statement
Mr Mehr, Dr Gorshow, and Mr Kenney reported no conflicts of interest.