What Price Healing?

Recent, dramatic price increases (see Sidebar) by Ariad, Valeant, Mylan, and Turing Pharma, which raised the price for a niche anti-infective by 5000% in 2015, have intensified public criticism of the pharmaceutical industry. In many of these cases, these increases affected older drugs that cost only a few dollars per dose to make and were positioned as “passalong costs” to offset business losses or make up for the cost of the new drug pipeline, or, as Turing’s official response said, “to balance patient access to existing drugs with investment in research and value generation for our shareholders” (1).  In each case, company stock prices fell along with public opinion (2).

While Mylan’s CEO blamed changes in the US healthcare insurance landscape for making the price change visible (3), pharmaceutical manufacturers were quick to distance themselves from Turing’s over-the-top maneuver. However, drug companies often note the cost and time required to develop a new drug, which the Tufts Center for the Study of Drug Development pegs at $2.6 billion and 10 years in its latest study (4), and the high rate of failure for new drug candidates, especially during Phase III, when manufacturing comes into play. 

Other challenges they point out are restrictions set by insurance companies and other payers, and limits to patient access established by such authorities as the National Institute for Health and Care Excellence (NICE) in the United Kingdom. “Overall, companies are under pressure to keep costs down, and organizations like NICE use increasingly sophisticated models to assess net clinical benefit versus cost,” says Gawayne Mahboubian-Jones, a consultant based in Switzerland. “Pressure on public purse-strings enables such organizations to play ‘hardball’ with drug approvals.”

Payers exert more pressure

In September 2016, the Biotechnology Innovation Organization set up a web page to emphasize the insurance industry’s role in high healthcare costs and debunk the idea that industry price increases were out of line with costs (5). But discussions about drug prices also raise other questions, among them, whether, and how, pharmaceutical R&D and manufacturing might become more efficient. Pharmaceutical companies have made changes in the way their research and development programs are set up, moving to collaborative and hub models, and new drug approvals have been trending up.

Many companies have implemented operational excellence and Lean Six Sigma programs, not only in manufacturing but in R&D. “This has had varying results, with the biggest challenge being the balance between the risk and benefit of certain redundant processes,” says Henry Levy, CSO at Veeva Systems, Inc. “This risk becomes complex when considering the needed interpretation of regulation, because guidances [from global regulatory bodies] always allow for individual company decisions.”

However, the industry still wastes billions of dollars each year on inefficient practices using outdated technology. Today, there is little current information available on pharmaceutical manufacturing costs.  In the late 1990s, researchers at the Massachusetts Institute of Technology’s Program on the Pharmaceutical Industry  (POPI) estimated that manufacturing cost the world’s eight largest pharmaceutical companies 27% of their revenue each year (6).

In 2006, one study found that the industry wasted $50 billion or more each year, due to regulatory issues and redundant or time-consuming manufacturing and quality-control practices (7).  In 2007, professors of finance at a number of universities estimated that improvements to pharmaceutical manufacturing alone (i.e. a 30% reduction in manufacturing costs)could result in $1 trillion in savings, if manufacturers reduced prices, and $12.3 billion if they did not reduce prices, allowing some of that money to be used to fund R&D or social programs (8).

Dollars and sense

Manufacturing costs, for the typical pharma company, are only approximately 10% of the outlays made in R&D, says Mahboubian-Jones. R&D remains the most costly part of the pharmaceutical value chain, largely because companies still take what Mahboubian-Jones called a “scattershot,” or one-off approach, to research.  “Pharma companies have only poor control of R&D costs, not because control mechanisms do not exist, but because the mindset at many companies is still to chase the next blockbuster, where R&D cost remains small compared to profit,” he says. “Unfortunately the true blockbusters of the 1990s and 2000s have all but disappeared. Most medicines are more specific, better targeted to a smaller market, and purchasing groups are now much more restrictive on what they are prepared to pay.”  As he notes, companies often respond to this situation by cutting manufacturing costs rather than development costs, typically, because they have already gone through their R&D budgets.

A major challenge is improving processes for clinical trials, which account for most of the cost of R&D (9). Data from the Tufts Center for Drug Development, which are usually used to explain industry costs, offer insights. Critics and patient groups (10, 11) have asked for greater transparency about the methods used to estimate the totals.

High clinical trial costs

In his blog, James Love of Knowledge Economy International has suggested that Tufts’ estimates be supplemented by other data, such as numbers generated by the Pharmaceutical Research and Manufacturers of America and Battelle, which showed the average cost per patient in a clinical trial to reach $36,000 ($59,500 for cancer drugs) (12).

Whether or not one agrees with them, the Tufts Center numbers reflect an extremely challenging environment, showing phase I clinical costs to be $25.3 million ($213.9 million when adjusted for risk), with the cost of failures during this phase reaching $188.6 million. Phase II costs are estimated at $58.6 million ($295 million when adjusted for risk), with cost of failure, $236.4 million, and phase III costs at $255.4 ($456.5 million), with cost of failures, $201.1 million.

These numbers raise some important questions. “Why are companies discovering, at Phase III, that drugs don’t work?” asks Girish Malhotra, consultant in manufacturing process development and optimization and principal of EPCOT International in Cleveland. “In API process development and formulation, it is essential to understand and have a command of the process before it is commercialized,” Malhotra says. “Such command would allow companies to produce quality from the get go rather than testing it in, as is currently the case,” he says.

Predictive models needed

“There is a need for better, more mechanistic models for predicting a drug’s behavior,” says Mahboubian-Jones. “Until we develop and improve that ability to predict behavior, we will continue to have to use the ‘try it and see’ approach,” he says.  One potential solution, he says, would be restricting the extent to which companies can write off the costs of failures against the profits of successful drug launches.  “This would have to be done carefully or it could run the risk of shutting down the drug- development process,” he says.

“Another factor influencing clinical trials and drug development is payer pressure, which demands that trials deliver clear, unequivocal benefit, which is not only increasing costs but is changing the nature of clinical trials,” says Mahboubian-Jones. Data management is also a major problem, and Ken Getz, associate professor and director at Tufts University School of Medicine, has attributed drug development delays and inefficiencies to “clinical trial protocol complexity, operating fragmentation, and the use of disparate IT solutions”(13).

The greatest obstacle to improving pharmaceutical manufacturing, experts agree, is failing to understand the cost of poor quality (COPQ) on overall business. This failure, says Malhotra, has had a crippling effect on the industry’s ability to innovate.

All too often, COPQ is seen as a part of doing business in a regulated environment, a risk that you take in order to get the best benefit, says Mahboubian-Jones. “This ‘ghettoization’ of quality continues to be one of the biggest inhibitors of real progress in the pharma industry,” he says.

Measuring the cost of poor quality

“Sometimes the cost of poor quality is not even measured, or, maybe even worse, is not measured properly,” says Ben Locwin, president of Healthcare Science Advisors. He has seen the cost of poor quality reach 42% of overall production costs at some companies. “It’s typically quoted as 10-30% of sales, but in reality, it’s often sustained at the upper end of this range,” he says.

For some generic-drug manufacturers, there has been too rigid a focus on cost reduction, which can lead to noncompliance, says Mahboubian-Jones, who notes that initiatives such as process analytical technology were never fully embraced in this sector.

Nevertheless, there are signs that progress is being made.  FDA has funded a study measuring product quality in an operational excellence environment, which is being conducted by Prabir Basu in the United States and Thomas Friedli at the University of St. Gallen in Switzerland (14) that will analyze trends.

Several companies have made process, documentation and training improvements, and have also improved regulatory compliance, resulting in fewer deviations and reducing the cost of goods sold, notes Locwin. “These improvements lead businesses to a state where they can better deliver what they claim with fewer nonconformities and employee errors,” he says.

Regulatory issues may be impeding the industry’s desire to invest in new and more efficient IT platforms and process technologies. Filing for changes to existing processes, even changes that would reduce risk and improve efficiencies, has become an ordeal for many global pharmaceutical companies today.

“Consider the product lifecycle,” says Maik Jornitz, who led the Parenteral Drug Association’s Aging Facilities task force. “When we cannot make a process change, and it takes four, five, or six years before the change can be made globally, there will be very little motivation for improvement,” he says.

Despite the challenges, there are signs that improvement is on the way, with research approaches such as risk-based monitoring (15). “When done well, this approach can improve predictive capabilities and success rates for clinical trials,” says Locwin.

References

1. A. Pollack and M. Goldstein, “Martin Skreli All But Gloated Over Higher Prices, Memos Show,” NewYorkTimes.com, Feb. 2, 2016, 
2. D. Bloomfield, “Bernie Sanders Follows ‘Greed” Tweet on Ariad With Letter,” Bloomberg.com, October 20, 2016. 
3. B. Brian, “Mylan CEO Blames Obama Care for Epipen Price Increase,” BusinessInsider.com, August 3, 2016. 
4. J.A. De Masi et al, Journal of Health Economics, 47 (2016).
5. Unsurance: Facts About Insurance, BIO.org, www.bio.org/insurance-facts
6. C. Cooney, “Benchmarking Pharmaceutical Manufacturing Performance,” MIT Program in the Pharmaceutical Industry (POPI), Working Paper, Cambridge, Mass, 1996: pp 33-96.
7. J. Macher and J. Nickerson, “The Pharmaceutical Manufacturing Project."
8. J. Vernon et al., Drug Information Journal, 2007, Vol 41, pp. 229-239.
9. K. Getz, Clinical Trial Complexity, Tufts, 2012.
10. A. Gray et al., Letter to Anthony De Masi and the Tufts Center for Drug Development, csdd.tufts.edu, Feb. 3, 2015.
11. D. Light and R. Warburton, Biosocieties, April, 2011, posted Feb. 7, 2016, on Pharmamyths.net.
12. J. Love, “The 2016 Tufts Estimates of the Risk-Adjusted Out-of-Pocket Costs to Develop a New Drug,” keionline.org, April 12, 2016.
13. “Veeva Announced EDC and eSource to Transform Clinical Data Management” Press Release, October 13, 2016.
14. Choosing Quality Metrics, Pharmtech.com, Oct. 19, 2016.
15. Zurdo et al, Developability of Biotherapeutics,Zurdo et al., Chapter 10, Developability Assessment Workflows to De-Risk Biopharmaceutical Development, (CRC Press,) 2015

Article Details

Pharmaceutical Technology
Vol. 40, No. 11
Pages: 16–20

Citation

When referring to this article, please cite it as A. Shanley, "What Price Healing?” Pharmaceutical Technology 40 (11) 2016.