Cancer drugs in the United States have increased 5- to 10-fold since 2000, but several analyses indicate that this increase in cost does not match improvements in health and overall survival.
Peter B. Bach, MD
Pricing a drug based on its efficacy, although a reality in Europe, is still a long way from happening in the United States. Those in the oncology business say that it’s high time for value-based pricing to be introduced here.
“We would benefit from a system in the United States that seeks to fix what is broken in the pricing market, and that involves finding a way to pay for the value of drugs based on transparent parameters,” said Peter B. Bach, MD, director of the Center for Health Policy and Outcomes at Memorial Sloan Kettering Cancer Center and also a developer of the DrugAbacus tool for estimating true drug value.
aThis is a speculative price.
bNon—small cell lung cancer. Source: Bach PB. Indication-specific pricing for cancer drugs. JAMA. 2014;312(16):1629-1630. doi: 10.1001/jama.2014.13235.
Cancer drugs in the United States have increased 5- to 10-fold since 2000,1 but several analyses indicate that this increase in cost does not match improvements in health and overall survival. The results of one 2015 analysis recently published in Health Affairs show that the cost of established first-line therapies for multiple myeloma increased without a corresponding gain in benefits.2 In addition, the results of another cost-benefit analysis show no correlation between cancer drug price and objective patient benefits, such as survival or improved quality of life.3
To make prices better reflect drug performance, insurers and prescription benefit managers would like to set prices based on more concrete indicators of a drug’s success, such as survival, drug toxicity, and novelty. Most oncology drugs are approved for more than one type of cancer, and the cost is the same for each indication regardless of variations in efficacy. For example, even though erlotinib (Tarceva) yields a median survival gain of about 3 months for metastatic non—small cell lung cancer (NSCLC) and about 2 weeks for pancreatic cancer,4 the per-unit cost is the same for both indications.Calculating a price that accurately reflects the value of a given drug requires consideration of multiple factors, including overall survival, drug novelty, safety, disease rarity, development cost, and population burden of the disease (assessed by the number of life years lost due to that cancer). Furthermore, how to translate the relative weight of each component into an overall price still remains to be determined. How much is a year of life worth? How much higher should the price be if the drug treats a rare disease? Should the price be lower if the drug has serious side effects?
Recently developed pricing models in European countries factor treatment benefits into the costs of drugs. Beginning in 2005, the UK National Institute for Health Care and Excellence (NICE) mandated that the National Health Service fund drugs and other treatments according to the NICE technology appraisals. These appraisals compare the clinical and cost-effectiveness of new drugs with existing treatments to determine whether a drug qualifies for inclusion in the formulary, with the goal of focusing healthcare spending on treatments that improve quality and/or length of a patient’s life without substantial side effects.5 In Germany, The Act for the Restructuring of the Pharmaceutical Market took effect in 2011 and allows drug companies to set their own price for a new drug during its first 12 months on the market. During this time, the company needs to prove the drug has an added clinical benefit—which may include reduced disease duration, increased survival, reduced side effects, or improved quality of life—over comparable reference drugs.6 If the drug has no added clinical benefit, it is assigned to a particular drug class and priced accordingly. Drug companies may negotiate with health insurers over pricing (or an arbitration board if they are unable to agree on a price). Whether this new legislation will better coordinate a drug’s patient-centered value with its price remains to be determined. An early benefit assessment in Germany shows that pharmaceutical companies, the Institute for Quality and Efficiency in Health Care, and the Federal Joint Committee differ greatly on whether individual drugs have additional benefits that may warrant higher prices.7 Furthermore, some experts state that the new legislation could lead to pharmaceutical companies inflating their prices during the initial year.8
Whereas efficacy-based pricing is not yet a reality in the United States, physicians and cancer societies have made an effort to provide cost guidance to physicians and patients. In 2014, the ASCO launched an initiative to evaluate the relative value of oncology treatments by closely examining drug trial outcomes and assigning values based on clinical benefit, toxicity, and financial cost. According to ASCO, the initiative, which requires much more data and fine tuning, may provide a framework for a future indication-specific pricing model.9
In October 2015, the National Comprehensive Cancer Network (NCCN) released a series of “Evidence Blocks” that include measures of drug efficacy and safety, quality and consistency of evidence, and affordability of the treatment, all presented in a simple visual block for easy comparison.10
According to the NCCN, the Evidence Blocks help to clarify the clinical and scientific rationale for treatment choices and provide an estimate of the economic impact. Taken together, the ASCO and NCCN initiatives underscore the growing national interest in implementing a more logical basis for establishing drug prices than currently exists. However, industry observers state that market and institutional practices stand in the way of efficacy-based drug pricing. According to Lee N. Newcomer, MD, MHA, senior vice president of UnitedHealth Group (with strategic responsibility for oncology, genetics, and women’s health), regulation on how cancer drug prices are established in the United States must be addressed before any implementation of an indication- specific pricing model.
State and federal policies have driven drug costs upward—namely the Medicare Reform Act of 2003 that prohibits Medicare from negotiating drug prices with pharmaceutical companies.11 Newcomer noted that most states require an insurance company to pay for a drug regardless of its benefit or cost, which restricts the insurance company’s leverage with drug companies.
Source: Memorial Sloan Kettering Cancer Center DrugAbacus (drugabacus.org).
Drug manufacturers “could price the drug at $75; $7500; or $75,000,” and the insurance company would have to pay that amount, said Newcomer. “Insurers can’t choose which drug they are willing to pay for, and pricing does not follow any rational market rules.” He added that the prices of specialty drugs for cancer increased 10% per year on average from 2012 to 2014 even though those drugs may not have been undergoing more scientific research or incurring additional marketing costs.Development of indication-based drug pricing would require standardized methods for calculating a drug’s value. Bach and his colleagues developed the “DrugAbacus,” which suggests prices for several cancer drugs based on user-determined weight of efficacy, tolerability, novelty, development cost, rarity, and population burden.12 In a 2014 article in JAMA, Bach provided a crude, hypothetical example for drug price based on the survival gain it yields compared with conventional treatment.13 If an additional year of life is valued at $150,000, erlotinib would cost $5183 per month for metastatic NSCLC and $1282 per month for pancreatic cancer. Both of those DrugAbacus prices are lower than the actual 2014 monthly prices of $6292 and $5563 (due to the lower dose used for pancreatic cancer), respectively.
The tool also adjusts the price upward if the drug treats a rare condition, has a novel mechanism of action, has high development costs, or addresses a significant population burden. For example, because blinatumomab (Blincyto) is a unique immunotherapy used to treat a rare form of acute lymphoblastic leukemia (Philadelphia chromosome—negative precursor B-cell acute lymphoblastic leukemia),14 it could warrant a higher cost. Without novelty or rarity multipliers included in the algorithm, the price of blinatumomab would be $12,612 per month (assuming a life-year valued at $120,000 and a 15% toxicity discount), according to the DrugAbacus.12 If both the novelty and rarity multipliers are included at their maximum potential value (3- fold), the monthly price would be $113,509 versus the current monthly price of $64,260.To implement a system of indication-specific pricing in the United States, experts agree that several modifications are needed to standardize drug costs, insurance reimbursement, and prescription processes. Newcomer said that whereas the United States is not ready to assign monetary values to factors such as improved survival, creating a more open marketplace for drug pricing would help prices to better reflect their value to patients. He commented that because insurers are unable to choose which drug they are willing to pay for, they have little leverage for substituting an equally effective, less expensive drug.
“I don’t know what the true value of a drug is,” said Newcomer. “But I think if we had a more open marketplace with competition required [among drug companies], we would see prices go down.”
Express Scripts Holding Co, a large manager of prescription drug benefits for US employers and insurers, is currently attempting to arrange deals with pharmaceutical companies to implement indication-specific pricing under a pilot program. The pilot will select oral and self-injectable cancer drugs for its National Preferred Formulary for plan year 2016.15 According to a spokesman for Roche’s Genentech, the company would welcome an indication-specific pricing model, but the current fragmented market and different data systems and reimbursement structures among Medicare, Medicaid, and private insurers are currently an obstacle to implementation. “Reimbursement structures would need to be modified, and data systems need to continue to evolve to ensure that outcomes data can be reliably captured in order to track which indications and which outcomes are achieved on a patient-level basis,” said the spokesman, Andrew Villani.
Bach said that an indication-specific pricing model would also require modifications to the prescription processes, which may include adding indications to prescriptions and categories for ordering, prescribing, price tracking, and reimbursement, as well as upgrading the distribution process. Improving patient access to drugs by reducing barriers such as price tiering would also be important to make indication-specific pricing successful, Bach said. However, he indicated that the benefits of indication-based pricing could extend beyond the central goal of reducing cost to improve patient access. As he stated in his JAMA Viewpoint, routine identification of drug utilization and indications, along with the shift toward more electronic documentation and intricate data on patient encounters, could create a framework for real-world outcome analyses of indication and patient outcome (Tables 1,2).13 Ultimately, this may help increase transparency of oncology care and improve the ability of future patients and clinicians to make informed decisions about treatment, he said.