In the JournalsPerspective

Analyses question cost-effectiveness of PCSK9 inhibitors

Dhruv S. Kazi
Dhruv S. Kazi

An analysis of the FOURIER trial did not change previous findings that using PCSK9 inhibitors to treat patients with atherosclerotic CVD was not cost-effective, according to a research letter published in JAMA.

In a separate analysis published in JAMA Cardiology, researchers concluded that for evolocumab (Repatha, Amgen) to be considered cost-effective, the annual net price would need to be reduced to under $10,000, or a higher-risk population would need to be treated with it.

Cost-effectiveness in adults

Dhruv S. Kazi, MD, MSC, MS, attending physician at Zuckerberg San Francisco General Hospital and Trauma Center, and colleagues updated their original analysis of PSCK9 inhibitor cost-effectiveness from 2016 to include updated prices and results from the FOURIER trial.

Researchers used the Cardiovascular Disease Policy Model to estimate cost-effectiveness of PCSK9 inhibitors — evolocumab and alirocumab (Praluent, Sanofi-Regeneron) — or ezetimibe with statin therapy in 8.9 million U.S. adults (mean age, 66 years; 61% men; mean LDL, 104 mg/dL) with atherosclerotic CVD who met the criteria from the trial.

The incremental cost-effectiveness ratio was the primary outcome. The secondary outcome was the cost at which PCSK9 inhibitors would be cost-effective at a willingness-to-pay threshold of $100,000 per quality-adjusted life-year.

Current wholesale acquisition costs were used for PCSK9 inhibitors ($14,542; 1% increase from 2015 to 2017) and ezetimibe ($3,818; 32% increase from 2015 to 2017). Researchers also analyzed data to reflect the results from FOURIER, which showed evolocumab (Repatha, Amgen) lowered risk for MI and stroke but not CV death.

The policy model reproduced major adverse events rates from FOURIER in patients treated with only a statin and patients treated with a statin in addition to PCSK9 inhibitors.

PCSK9 inhibitors reduced an estimated 2,893,500 more adverse events compared with ezetimibe with incremental health care costs of $450,000 per QALY (80% uncertainty interval, 301,000-787,000).

Effects on QALYs

The annual cost of PCSK9 inhibitors would need to be reduced by 71% to achieve a threshold of $100,000 per QALY, Kazi and colleagues wrote, noting that without an effect on CV death, the incremental health care costs increased to $1,795,000 per QALY.

“Although computer simulations that synthesize data from observational studies and clinical trials may not precisely reflect clinical effectiveness that may be observed in practice over time, these updated results continue to demonstrate that reducing the price of PCSK9 inhibitors remains the best approach to delivering the potential health benefits of PCSK9 inhibitors therapy at an acceptable cost,” Kazi and colleagues wrote.

A company spokesperson from Amgen issued a statement by email to Cardiology Today expressing concern that the analysis could help prevent patients from getting access to a needed therapy.

“Several published analyses have shown value-based price ranges that support our current average net selling price, which has been incorporated in several value-based arrangements with payers. Key differences that need to be considered are real-world event rates and value for patient life-years in the U.S.,” the spokesperson wrote. “We continue to be very concerned about these analyses and the impact they may have on patient care by not adequately reflecting the value and becoming a tool for blocking patient access. The Repatha outcomes study taught us that patients with prior [CV] events remain at high risk of additional [CV] events, including [MIs], strokes and the need for revascularization despite optimal statin therapy, so there is an urgency to identify and treat these patients.”

Daniel B. Mark
Daniel B. Mark

“Painful as it is, draconian restrictions on access to drugs that are priced for profit maximization out of proportion to any value proposition and budget tolerances may continue to be the only way medicine can send a strong signal to innovators that their future rewards are tied not just to scientific advancement, but also to affordability,” Daniel B. Mark, MD, MPH, professor of medicine at Duke University Medical Center and director of outcomes research at Duke Clinical Research Institute, and Kevin A. Schulman, MD, MBA, professor of medicine at Duke University and associate director of the Duke Clinical Research Institute, wrote in a related editorial. “This is not just an issue for pharmaceutical companies. Profit maximization behavior in medicine out of proportion to value provided is widespread. Ultimately, this message will need to be heeded by the entire health care enterprise.”

Exceeding cost-effectiveness thresholds

In a separate study in JAMA Cardiology, Gregg C. Fonarow, MD, director of the Ahmanson-UCLA Cardiomyopathy Center, co-director of the UCLA Preventative Cardiology Program, co-chief of the UCLA Division of Cardiology and the Eliot Corday Chair in Cardiovascular Medicine and Science, and colleagues found that the current price of evolocumab ($14,523) in addition to standard background therapy to treat patients with atherosclerotic CVD surpassed “generally accepted cost-effectiveness thresholds.”

Gregg C. Fonarow, MD
Gregg C. Fonarow

The addition of evolocumab was linked to an increase in incremental cost ($105,398). Incremental QALYs improved by 0.39, with an incremental cost-effectiveness ratio of $268,637 per QALY gained. Evolocumab would need to cost an annual net price of $9,669, or with a discounted net price of $10,311 for patients with LDL ≥ 80 mg/dL, to achieve an incremental cost-effectiveness ratio of $150,000 per quality-adjusted life-year gained, Fonarow and colleagues wrote.

“These findings highlight the need for a comprehensive disease management approach for [atherosclerotic] CVD that includes vigorous lifestyle changes, assiduous adherence to all guideline-directed therapies and judicious use of new, more costly therapies,” the researchers wrote.

“This study affirms the clinical benefits and economic value of delivering Repatha to the right high-risk patients, specifically patients who have had [an MI] or stroke with high LDL levels despite maximally tolerated statin therapy,” Joshua Ofman, MD, senior vice president of global value, access and policy at Amgen, said in a company press release. “The actual net prices for payers in the market today after discounts and rebates are quite aligned to the value-based price range identified in this study. While the list price is more often quoted in the media, it is the discounted net price that is actually paid by payers and that should be considered in value assessments.”

In a related editorial, Mark and colleagues wrote, “We must … not lose sight of the fact that money spent on PCSK9 [inhibitor] therapy could be invested by health systems to make better use of the affordable drugs we already have (eg, improving adherence to statins and prescribing the most effective doses of statins). There are several other innovative products in the pipeline that also target PCSK9. The opportunities are tremendous for a company that could produce a new drug that can match the effects of PSCK9 [inhibitor] therapy at a more affordable price.” – by Darlene Dobkowski

References:

Fonarow GC, et al. JAMA Cardiol. 2017;doi:10.1001/jamacardio.2017.2762.

Kazi DS, et al. JAMA. 2017;doi:10.1001/jama.2017.9924

Mark DB, et al. JAMA. 2017;doi: 10.1001/jama.2017.8907

Mark DB, et al. JAMA Cardiol. 2017;doi:10.1001/jamacardio.2017.2911.

Disclosures: The study by Fonarow and colleagues was funded by Amgen. Kazi reports no relevant financial disclosures. Mark reports receiving grants from AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Gilead, Merck and Oxygen Therapeutics, and consulting fees from Janssen and Medtronic. Schulman reports receiving research support from Amylin Pharmaceuticals and Merck, serving as a board member of Bivarus, board of directors for Cue Biologics the scientific advisory board for Cardinal Analytx and the board for Anthelio. Fonarow reports consulting for Amgen, Janssen and Novartis. Ofman is an employee of Amgen. Please see the studies and editorials for all other authors’ relevant financial disclosures.

 

Editor’s Note: This article was updated on August 22, 2017, to add a statement from Amgen. It was further updated on August 24, 2017, to add findings from the JAMA Cardiology study and a statement from Amgen on it, and to add a perspective from Dr. Baum. On August 25, 2017, it was updated to add a perspective from Dr. Malone.

Dhruv S. Kazi
Dhruv S. Kazi

An analysis of the FOURIER trial did not change previous findings that using PCSK9 inhibitors to treat patients with atherosclerotic CVD was not cost-effective, according to a research letter published in JAMA.

In a separate analysis published in JAMA Cardiology, researchers concluded that for evolocumab (Repatha, Amgen) to be considered cost-effective, the annual net price would need to be reduced to under $10,000, or a higher-risk population would need to be treated with it.

Cost-effectiveness in adults

Dhruv S. Kazi, MD, MSC, MS, attending physician at Zuckerberg San Francisco General Hospital and Trauma Center, and colleagues updated their original analysis of PSCK9 inhibitor cost-effectiveness from 2016 to include updated prices and results from the FOURIER trial.

Researchers used the Cardiovascular Disease Policy Model to estimate cost-effectiveness of PCSK9 inhibitors — evolocumab and alirocumab (Praluent, Sanofi-Regeneron) — or ezetimibe with statin therapy in 8.9 million U.S. adults (mean age, 66 years; 61% men; mean LDL, 104 mg/dL) with atherosclerotic CVD who met the criteria from the trial.

The incremental cost-effectiveness ratio was the primary outcome. The secondary outcome was the cost at which PCSK9 inhibitors would be cost-effective at a willingness-to-pay threshold of $100,000 per quality-adjusted life-year.

Current wholesale acquisition costs were used for PCSK9 inhibitors ($14,542; 1% increase from 2015 to 2017) and ezetimibe ($3,818; 32% increase from 2015 to 2017). Researchers also analyzed data to reflect the results from FOURIER, which showed evolocumab (Repatha, Amgen) lowered risk for MI and stroke but not CV death.

The policy model reproduced major adverse events rates from FOURIER in patients treated with only a statin and patients treated with a statin in addition to PCSK9 inhibitors.

PCSK9 inhibitors reduced an estimated 2,893,500 more adverse events compared with ezetimibe with incremental health care costs of $450,000 per QALY (80% uncertainty interval, 301,000-787,000).

Effects on QALYs

The annual cost of PCSK9 inhibitors would need to be reduced by 71% to achieve a threshold of $100,000 per QALY, Kazi and colleagues wrote, noting that without an effect on CV death, the incremental health care costs increased to $1,795,000 per QALY.

“Although computer simulations that synthesize data from observational studies and clinical trials may not precisely reflect clinical effectiveness that may be observed in practice over time, these updated results continue to demonstrate that reducing the price of PCSK9 inhibitors remains the best approach to delivering the potential health benefits of PCSK9 inhibitors therapy at an acceptable cost,” Kazi and colleagues wrote.

A company spokesperson from Amgen issued a statement by email to Cardiology Today expressing concern that the analysis could help prevent patients from getting access to a needed therapy.

“Several published analyses have shown value-based price ranges that support our current average net selling price, which has been incorporated in several value-based arrangements with payers. Key differences that need to be considered are real-world event rates and value for patient life-years in the U.S.,” the spokesperson wrote. “We continue to be very concerned about these analyses and the impact they may have on patient care by not adequately reflecting the value and becoming a tool for blocking patient access. The Repatha outcomes study taught us that patients with prior [CV] events remain at high risk of additional [CV] events, including [MIs], strokes and the need for revascularization despite optimal statin therapy, so there is an urgency to identify and treat these patients.”

Daniel B. Mark
Daniel B. Mark

“Painful as it is, draconian restrictions on access to drugs that are priced for profit maximization out of proportion to any value proposition and budget tolerances may continue to be the only way medicine can send a strong signal to innovators that their future rewards are tied not just to scientific advancement, but also to affordability,” Daniel B. Mark, MD, MPH, professor of medicine at Duke University Medical Center and director of outcomes research at Duke Clinical Research Institute, and Kevin A. Schulman, MD, MBA, professor of medicine at Duke University and associate director of the Duke Clinical Research Institute, wrote in a related editorial. “This is not just an issue for pharmaceutical companies. Profit maximization behavior in medicine out of proportion to value provided is widespread. Ultimately, this message will need to be heeded by the entire health care enterprise.”

Exceeding cost-effectiveness thresholds

In a separate study in JAMA Cardiology, Gregg C. Fonarow, MD, director of the Ahmanson-UCLA Cardiomyopathy Center, co-director of the UCLA Preventative Cardiology Program, co-chief of the UCLA Division of Cardiology and the Eliot Corday Chair in Cardiovascular Medicine and Science, and colleagues found that the current price of evolocumab ($14,523) in addition to standard background therapy to treat patients with atherosclerotic CVD surpassed “generally accepted cost-effectiveness thresholds.”

Gregg C. Fonarow, MD
Gregg C. Fonarow

The addition of evolocumab was linked to an increase in incremental cost ($105,398). Incremental QALYs improved by 0.39, with an incremental cost-effectiveness ratio of $268,637 per QALY gained. Evolocumab would need to cost an annual net price of $9,669, or with a discounted net price of $10,311 for patients with LDL ≥ 80 mg/dL, to achieve an incremental cost-effectiveness ratio of $150,000 per quality-adjusted life-year gained, Fonarow and colleagues wrote.

“These findings highlight the need for a comprehensive disease management approach for [atherosclerotic] CVD that includes vigorous lifestyle changes, assiduous adherence to all guideline-directed therapies and judicious use of new, more costly therapies,” the researchers wrote.

“This study affirms the clinical benefits and economic value of delivering Repatha to the right high-risk patients, specifically patients who have had [an MI] or stroke with high LDL levels despite maximally tolerated statin therapy,” Joshua Ofman, MD, senior vice president of global value, access and policy at Amgen, said in a company press release. “The actual net prices for payers in the market today after discounts and rebates are quite aligned to the value-based price range identified in this study. While the list price is more often quoted in the media, it is the discounted net price that is actually paid by payers and that should be considered in value assessments.”

In a related editorial, Mark and colleagues wrote, “We must … not lose sight of the fact that money spent on PCSK9 [inhibitor] therapy could be invested by health systems to make better use of the affordable drugs we already have (eg, improving adherence to statins and prescribing the most effective doses of statins). There are several other innovative products in the pipeline that also target PCSK9. The opportunities are tremendous for a company that could produce a new drug that can match the effects of PSCK9 [inhibitor] therapy at a more affordable price.” – by Darlene Dobkowski

References:

Fonarow GC, et al. JAMA Cardiol. 2017;doi:10.1001/jamacardio.2017.2762.

Kazi DS, et al. JAMA. 2017;doi:10.1001/jama.2017.9924

Mark DB, et al. JAMA. 2017;doi: 10.1001/jama.2017.8907

Mark DB, et al. JAMA Cardiol. 2017;doi:10.1001/jamacardio.2017.2911.

Disclosures: The study by Fonarow and colleagues was funded by Amgen. Kazi reports no relevant financial disclosures. Mark reports receiving grants from AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Gilead, Merck and Oxygen Therapeutics, and consulting fees from Janssen and Medtronic. Schulman reports receiving research support from Amylin Pharmaceuticals and Merck, serving as a board member of Bivarus, board of directors for Cue Biologics the scientific advisory board for Cardinal Analytx and the board for Anthelio. Fonarow reports consulting for Amgen, Janssen and Novartis. Ofman is an employee of Amgen. Please see the studies and editorials for all other authors’ relevant financial disclosures.

 

Editor’s Note: This article was updated on August 22, 2017, to add a statement from Amgen. It was further updated on August 24, 2017, to add findings from the JAMA Cardiology study and a statement from Amgen on it, and to add a perspective from Dr. Baum. On August 25, 2017, it was updated to add a perspective from Dr. Malone.

    Perspective
    Seth J. Baum

    Seth J. Baum

    The analysis by Kazi and colleagues must be critically evaluated, as it contains several issues. To understand the flaws in their assumptions, we must delve into the weeds.

    In the evaluation of a QALY, one must examine both utility and disutility. Utility is reported on a scale from 0 to 1. Zero is death and 1 is perfect health. Intermediate states are assigned values between them. Let’s say a person has an MI. What does that mean on that scale from 0 to 1? In the general literature, MI typically falls in the 0.7 to 0.8 range; yet in this analysis, Kazi and colleagues considered an MI to have a utility of 0.96. This is 0.04 away from perfect health. For stroke, they used 0.8835. You cannot tell me that a patient has 88% — almost 90% — of a perfect existence after a stroke. In the literature, a stroke is typically assigned a value in the 0.62 range, which is much more appropriate. By establishing their utility values inappropriately high, they negate the effectiveness of the drugs. In a sense, they are suggesting, “MIs and strokes are not that bad. They have minimal impact on people’s lives.” Thus, drugs that prevent MIs and strokes cannot have any value because the conditions prevented by such medications are not important. This viewpoint could not be further from the truth. Simply ask stroke or MI victims whether their lives have changed and you will understand why this is so.

    Then there is disutility, which is how much one transiently suffers after an event. They also minimized the events’ acute impacts, another substantial problem.

    They assigned the CV event rate in their base case an unrealistically low number, one consistent with trial, not real-world, data. These drugs are being evaluated for real-world patients, so real-world statistics should have been used in their model.

    They did not include all CV endpoints. They included only nonfatal MI, nonfatal stroke and CV death. They didn’t include hospitalization for a stent or a hospitalization for ACS or CABG. Such events are important for payers, and they’re important for the patients and the patients’ families. The drug’s valuation would have been entirely different if these important events had been included in their analysis.

    Also, Kazi and colleagues should have broken the groups into subsets, because there’s higher value in treating higher-risk groups. If you use a group in which some people have very high risk and other people do not, and claim that based on this population, the drug doesn’t have value at a certain price point, that does not tell the whole story. What about for higher-risk patients? You can’t just make a blanket statement like that.

    The $100,000 QALY that they use is arbitrary. In the literature, QALYs range anywhere from $50,000 to $200,000, which can make an enormous difference in the valuation of a drug. Commonly in first-world countries such as the U.S., $150,000 has become the norm. Using $150,000 would have made the PCSK9 inhibitors much more valuable and their current price point more acceptable.

    Unfortunately, this paper does have implications for clinical practice. In response to the report by Kazi and colleagues’ in JAMA (doi:10.1001/jama.2016.11004), payers can turn around and say, “No, we’re not covering these medications. We don’t think the drug is valuable.” This is a very slippery slope. The clinician and patient need to determine whether a drug is indicated. It is not the payer who should be determining that. We need to reinstate the sacrosanct nature of the patient-physician relationship. It is not the patient-physician-payer relationship that we’re supposed to be protecting. It is the patient-physician relationship, and unfortunately payers have intruded on that. The paper and its ramifications are certainly disconcerting. 

    The $100,000 QALY that they use is arbitrary. In the literature, that range is anywhere between $50,000 and $200,000, which can make an enormous difference in the effect and the value of the drug.

    Unfortunately, this does have implications for clinical practice. In response to the initial report and Kazi and colleagues; initial paper in JAMA (doi:10.1001/jama.2016.11004), the payers can turn around and say, “No, we’re not covering this. We don’t think the drug is valuable.” It is a really dangerous precedent, because the clinician and the patient need to determine whether a drug is indicated. It is not the payer who should be determining that. We need to reinstate the nature of the patient-physician relationship. It is not the patient-physician-payer relationship that we’re supposed to really be protecting. It is the patient-physician relationship, and this is intruding on that. It is very upsetting and disconcerting. 

    • Seth J. Baum, MD, FACC, FACPM, FAHA, FNLA, FASPC
    • Chief Medical Officer, Excel Medical Clinical Trials Clinical Affiliate Professor of Biomedical Science, Florida Atlantic University President, American Society for Preventive Cardiology

    Disclosures: Baum reports serving on advisory boards for Amgen and Regeneron-Sanofi, conducting research with both drugs and speaking for Amgen.

    Perspective
    Malone

    Malone

    Every drug is cost-effective in the right patient in the right time. The PCSK9 inhibitors have that place. The challenge is identifying that subpopulation of patients with dyslipidemia who would be most beneficial from the medication. The FOURIER analysis looked at a couple of different groups. The clinical trials that both Sanofi/Regeneron and Amgen have conducted have attempted to define those patients with familial hyperlipidemia and patients who are statin intolerant, and those who are not able to reach goal with statins. If you were to define the population correctly, the drugs probably have a great chance of being high-value products.

    On the other hand, if you define the population fairly broadly — eg, potential statin users — then there’s probably not so much value for these drugs, because cost-effectiveness and value is all relative. It’s relative to the next best competitor, and statins do a pretty good job, especially in the more potent statins such as atorvastatin and rosuvastatin. If a potent statin is your comparator for the ability to drive down LDL beyond what they do, it becomes a little more challenging, but for those patients that can’t get to goal for various reasons, then PCSK9s might be valuable.

    The economic analysis in JAMA is pretty much a rehash of what Kazi and colleagues have done previously. There really wasn’t that much new there. They had made some assumptions in the first paper, and with the recent FOURIER publication, they were able to replace some assumptions with some actual data. They kept the model intact.

    Their research letter doesn’t tell us all of the nuances of what they have done with their analysis. It appears they have built it off of what they did previously. Some of the assumptions that they have made previously were somewhat questionable. They didn’t modify their primary assumptions about which patients were treated and how many of them were treated. For the affordability index that they come up with, they failed to take into account that the products just have not been widely used. Their analysis assumes that 100% of eligible patients received those drugs, which clearly is not happening. Then the question becomes, how valid is the analysis if you relax that assumption? They don’t give us any insight into that in the more recent publication.

    The JAMA Cardiology paper, funded by Amgen, appears to tell a completely different story with regards to cost-effectiveness. In this paper, the authors used real-world data to support their model. In general, cost-effectiveness models can be sensitive to the inputs, so you can observe different results based on the parameters and assumptions in the model. It’s not surprising that they found different results.

    • Malone, RPh, PhD, FAMCP
    • Professor of Pharmacy University of Arizona

    Disclosures: Malone reports consulting for Amgen, Mallinckrodt Pharmaceuticals and Sanofi on matters unrelated to PCSK9 inhibitors.