A New Approach for Drug Approvals in the Age of “Big Data”


November 3, 2015 by Steven A. Zecola

I. Introduction and Summary

The amount of health-related information available to researchers is expanding exponentially as a result of the technical advancements in the world of “Big Data”. To most people, this signifies a quicker path to cures of debilitating diseases and/or to better care of people on a personalized basis.

But that is not how the current system works. Under its legislative charter, the Food and Drug Administration (“FDA”) stands between every promising cure and the implementation of that cure.  Its approval process imposes an average of 12 years of development and a cost of over $1 billion per approved drug.  As a result, this system creates a valley of death that most new healthcare solutions fail to cross. For example, the FDA approves only about 25 new drugs per year.

In contrast, the Chemical Abstract Service (CAS) adds 15,000 substances – per day – to its registry; and that number is growing.  For example, CAS recently registered a novel therapeutic designed to treat acute myeloid leukemia as its one hundred millionth entry.

While Congress put the FDA’s regulatory structure into motion decades ago, it has done very little to curb the excessive delays and costs or address the challenges of the valley of death.

Rather than a tweak here and a tweak there, the entire system needs a major overhaul.

In particular, to significantly improve healthcare for both healthy and unhealthy people, Congress should leverage the capabilities and resources of the private sector for quicker implementation and more efficient safety procedures.

As described herein, Congress should require the FDA to provide provisional licenses after a 60 day review period to all qualified applicants who put a $5-10 million bond into an escrow account for a Phase 1 trial. These escrowed funds would escalate in size for Phase 2 and Phase 3 trials, depending upon the magnitude of the trials.

Each licensee would provide reports to the FDA for each phase of its research.  The FDA could seek injunctions for any trials that it believed would harm public safety. If granted by the court, the research organization would lose its provisional license and the escrowed funds.

If the FDA and the courts did not block new drugs and treatments from becoming effective, the escrowed funds would be returned to the developer one year after commercial launch and a commercial license for such drug or treatment would be issued by the FDA at that time.

Such an approach would hasten and improve the advancement of effective cures by dramatically reducing the risk, time and cost of drug development, while minimizing the risk of adverse consequences from research and development.

II. The History of FDA Regulation

The original 1906 Food and Drugs Act prohibited the contamination or misbranding of pharmaceuticals, but did not require premarket approval of drugs. As long as drugs were properly labeled and conformed to the applicable guidelines, there was no federal statute to prevent their commercialization.

This legislation stood until 1938 when President Roosevelt signed the Food, Drug, and Cosmetic Act, which revamped oversight of food and drugs and included consumer protection over medical devices and cosmetics.

The provisions of the new law required manufacturers to submit to the FDA evidence of a drug’s safety before it could go on the market, and the agency had two months to approve, reject, or request additional data from the firm. This mandate for premarket evidence of a drug’s safety represented the birth of the new drug application, or NDA.

In 1962 Congress amended the drug provisions of the law to require, among other items, that manufacturers establish not only the safety but also the effectiveness of the drug through adequate and well-controlled clinical trials prior to marketing. This change was responsible for the expansion of the drug application into a massive document of perhaps hundreds of volumes.

Since that time, there have been several legislative adjustments to the process.  For example, in 1988, Congress adopted The Prescription Drug Marketing Act and modified it in 1992 to ban the diversion of prescription drugs from legitimate commercial channels. The law also requires drug wholesalers to be licensed by the states; restricts re-importation from other countries; and bans sale, trade or purchase of drug samples, and counterfeiting of redeemable drug coupons.

In 1997, the Food and Drug Administration Modernization Act reauthorized the Prescription Drug Act of 1992. Its provisions include measures to review medical devices and to regulate advertising of unapproved uses of approved drugs and devices.

III. The Success and Tribulations of the Current Drug Approval System

The FDA has broad responsibilities. For example, it regulates more than $1 trillion worth of consumer goods, about 25% of consumer expenditures in the United States. This includes $466 billion in food sales, $275 billion in drugs, $60 billion in cosmetics and $18 billion in vitamin supplements.

One method of measuring the success of the current system is the number of drug approvals per year.


As the chart shows, the number of new drug approvals per year is shockingly low.

Moreover, in contrast to what it has done (as shown in the chart above), the FDA doesn’t report what it hasn’t done.  For example, earlier this year a CBS 60 Minutes report on the use of polio virus to kill brain cancer disclosed that the FDA had delayed this promising treatment for almost a decade with pre-clinical study requirements.

The cost, delays and risk of the regulatory scheme are substantial. The result is a hostile investment climate that creates a valley of death for new healthcare initiatives.

IV. The Pace of Technological Advancement in Healthcare

Society, in general, is experiencing an explosion of information, its storage and its processing; commonly referred to collectively as “Big Data”.

For example, more information is produced every two days than was produced from the beginning of time to 2003. And 90% of the available data in the world was created in the last two years. By 2020, the amount of digital information is projected to have increased from 3 zettabytes to 40 zettabytes.

In healthcare, Big Data fundamentally changes the approach to experimental design. Today, researchers can afford to collect virtually unlimited amounts of data and search for significant patterns in the data.

For example, Flatiron Health has developed a service called “OncologyCloud”, based on the idea that 96% of potentially available data on patients with cancer is not yet analyzed. It aims to take this data gathered during diagnosis and treatment, and make it available to clinicians to further their studies.

In genetics, what took 10 years to uncover about the human genome would now take one week.  And the analysis of genomes generates even more data.

The ability and pace for scientists to address today’s many illnesses is exploding.   And with the use of personal genetic information, the treatments can become highly individualized and much more effective.

But the potential benefits from Big Data are not confined to curing only the sick.  Information is becoming available that will enable doctors to alter a patient’s health trajectory.    For example, a mutation in the BRCA1 gene is destined to produce breast cancer in 85% of all carriers.  Thankfully, an effective vaccine is within reach given the ongoing research into Parp Inhibitors.

One way to see the explosion of potential opportunities is in the number of new chemical substances. The following chart shows the growth rate in new substances registered by the Chemical Abstract Service (CAS) each year.

CAS Registry

While the technology is advancing at breakneck speed and producing many potential healthcare solutions, the current drug approval process is extraordinarily slow and expensive and most solutions don’t make it across the valley of death.

V. The Cost and Benefits of FDA Regulation

The greatest cost to society from the current drug approval process cannot be seen.  That is, viable healthcare solutions are being delayed or worse yet, are not being developed at all because of the expense and risk of approval.

For example, in addition to gaining a better understanding of the process of metastasis, scientists are now uncovering new compounds and potential avenues of treatment.  However, research for many of these compounds has not occurred in humans.  The principal reason for this vacuum is that the existing clinical trial design ensures that these compounds will fail.  In particular, the compounds are not designed to meet the FDA’s Phase 2 requirement of shrinking existing tumors.

Nevertheless, in rejecting my Petition to put increased focus and new procedures into place for trials of metastatic cancer, the FDA said it “does not believe that its current structure lacks ‘sufficient organizational focus’ on metastatic disease, is inefficient or otherwise impedes the conduct of clinical trials or products intended to either treat or retard the  development of metastatic cancer.”  Yet the system has failed miserably in bringing viable solutions to the market – and the public is paying the price from delay.

We know that increased focus and new treatments will have important benefits to society.  For example, as new anti-HIV drugs became available, the annual death rate from HIV declined by 63%.   Likewise, new heart drugs correlated with a 45% drop in coronary-artery disease.

There are many targets ripe for additional focus. The following table provides examples of chronic diseases in the United States.

Disease Population Affected Impact
Asthma 26 million 11 million missed school days; 2 million emergency visits
Alzheimer’s 5 million Debilitating illness leading to need for increasing care and to early death
Breast cancer 3 million 50,000 deaths per year
Diabetes 26 million 37 million medical visits per year
Epilepsy 3 million Seizures leading to lost productivity
Heart disease 27 million 12 million annual medical visits; 2 million annual outpatient department visits; 600,000 annual deaths


Taken together, chronic diseases cause 7 of every 10 deaths in America. Moreover, about 25% of people with chronic diseases have some type of activity limitation such as needing help with dressing or bathing or being restricted from work or attending school.

By any measure, chronic diseases take a heavy toll on the nation’s wealth and health. For example, heart disease and stroke are estimated to cost about $432 billion per year. Diabetes costs are estimated to be about $174 billion per year.  And Alzheimer’s is estimated to add about $148 billion per year.  In total, well over 50% of the $3 trillion spent on health care each year is attributable to chronic diseases.

Small improvements in bringing effective treatments to market can have a big impact on society in several different ways.

First, the cost of treatment will decrease. For example, the annual costs for diabetes care have been shown to drop almost 50% for patients who take their diabetes medicines properly.

Second, quicker availability of effective treatments can have a big impact by bringing down costs earlier. For example, we know that the length of time for the FDA approvals began increasing after the 1962 Amendments; from about 1 year to 3 years by the end of the decade and then up to 10 years in the 1970’s. Viewed in this light, the lag introduced by the current FDA drug approval process would appear to be at least 5 years, especially when including pre-clinical reviews.

Third, development costs would be lower under a revised system.  For example, the cost of research and development per approved drug has increased from $100m in 1975 to $300 million in 1987 to over $1 billion more recently.  Total R&D spending in the industry exceeded $800 billion from 1997 to 2011 even with high regulatory involvement and risk.  An improved regulatory system could reduce the average cost per approved drug by up to 50% without an undue risk to public safety, as described in Section VI.

Cumulatively, the implication of these numbers is staggering. For example, if effective new treatments came to market 5 years earlier under a revised regulatory regime and addressed 10% of all chronic diseases, this would save society $750 billion over time.  Similarly, a more streamlined process with industry participation in safety management would save up to $50 billion per year in unnecessary regulatory red tape.

On the other side of the coin, the FDA regulation produces benefits to society by keeping unsafe drugs out of trials and away from the broader public marketplace.  Undoubtedly, there are instances in which this would happen without any oversight. A classic example is the “Elixir Sulfanilamide tragedy” which led to the legislation in 1938. In 1937, a Massengill, a Tennessee drug company, began to market a liquid drug called Elixir Sulfanilamide.  The solvent in this drug was highly toxic variant of antifreeze; as a result, over 100 people died from taking this drug.

Similarly, the 1962 Drug Amendments legislation was passed in response to a therapeutic crisis – in this instance, the discovery that the use of thalidomide (a sedative that was marketed to combat the symptoms associated with morning sickness) by pregnant women caused birth deformities in thousands of babies in Europe.

With complete deregulation, there may be such adverse impacts to thousands of people per year.  If 1000 such instances were to occur per year and we put a punitive value of $1 million on each instance, the benefit of effective regulation in mitigating such circumstances would equate to $1 billion per year.  But even in this contrived example, these benefits would equate to less than one half of one percent (0.5%) of the hidden cost that regulation perpetrates from delays of new healthcare solutions.

More to the point, there is a way to minimize the risk to safety while maximizing the possibility of expedited healthcare solutions, as described in the following section.

VI. Recommendation for an Alternative Approach to the Regulation of New Drug Discovery

The existing legislation for drug discovery puts the FDA smack in the middle of all health-related research.

The FDA monitors and approves new solutions in the lab and from preclinical animal studies through 3 or 4 phases of clinical trials.  Oftentimes, trial design is set by the FDA. Hundreds of thousands of pages of information are provided to, reviewed by, and approved by the FDA before any progress can be made. Given the hostile investment climate created by this system, very few healthcare solutions can cross the valley of death.

However, by adjusting the responsibility for public safety and effectiveness to include industry participation, the number of new, effective solutions would increase dramatically given the improved investment climate. The issue would then become how to minimize undue harm to the public from this adjustment.

In a nutshell, an effective way to expedite scientific progress while curbing the risk of harm to the public is as follows:

1) Congress should limit the FDA’s role in preclinical research to surveillance.  That is, all Phase 1 clinical trials by qualified research organizations should be presumed to be valid by the FDA.

2) Each pre-approved, qualified professional organization filing a Phase 1 drug application would receive a provisional license from the FDA within 60 days from the time it provided the details of its trial and put a $5-10 million deposit into an escrow account.

3) Likewise, applications for Phase 2 and Phase 3 trials would be subject to a 60 day review cycle and would require additional funds to be deposited into an escrow account, the amount of which would be subject to the magnitude of the trials.

4) The industry would be required to take over and expand the function of the website Clinicaltrials.gov to develop a self-reporting and surveillance mechanism.  The resulting industry organization would provide regular reports to the FDA for purposes of oversight (along with the reports provided by individual companies with the results of each phase of its trial). The industry organization would have the authority to audit individual trial results and to maintain a secure whistle blower system.  A $1 million reward would be paid to whistle blowers whose efforts led to the successful prosecution of fraud by a licensee.

5) The FDA would have authority to seek injunctions from the U.S. Court of Appeals to stop any new drug or procedure that it believed was unfit for human application.  If a company were found to be in violation of the FDA’s safety guidelines, the company would lose its license and forfeit its escrowed funds.

6) Drug companies would be required to maintain and submit relevant health statistics on patients for one year from commercial launch.  Sixty days after such filings, the FDA would issue a commercial license and the escrowed funds would be returned to the licensee if a Court had not granted an injunction.

7) Out of an abundance of caution, Congress could phase in such an approach.  For example, it could apply such an approach in one or two research areas such as Alzheimer’s or Parkinson’s disease and/or metastatic cancer.

8) Under a phased approach, Congress would review the costs and benefits of the new approach triennially. The current drug approval process would sunset in 10 years unless Congress found the new approach to be flawed.

This proposal does not represent a radical deregulatory approach.  In fact, it is more stringent than the current practice of doctors prescribing “off-label” uses (that is, uses not approved by the FDA) of drugs approved for other conditions.  For example, doctors routinely prescribe Amoxicillin for stomach ulcers, although this drug has never been through any trials for this application.  Yet there is not any failure of the market to digest and utilize the latest information for off-label drugs to the benefit of consumers with very little safety risk.

Likewise, under the system proposed herein, scientists and doctors, working through professional organizations, would make determinations on the best available treatment for patients.  The FDA’s role would be limited to oversight and enforcement.  As a result, the investment climate would improve considerably and any new healthcare solutions would be able to cross the valley of death.

VII. Conclusion

Under its legislative charter, the FDA‘s actions not only can be quite limiting on research organizations, but they can also be hard to predict.

For example, with respect to personal genetic service, the FDA asserted its jurisdiction over the 23andMe.com offering by finding it to be a “device” that was “intended for use in the diagnosis of disease or other conditions or in the cure, mitigation, treatment, or prevention of disease”. It then found 23andMe to be in violation of the Federal Food, Drug and Cosmetic Act by marketing its product without prior approval from the FDA. The agency noted that 23andMe provided individual reports on 254 diseases and conditions and that the FDA expected 23andMe to get prior approvals for each of these reports. As the 23andMe CEO and co-founder, Anne Wojcicki, explained in response, the FDA system would require it to get approvals for over one million tests. Of course, that would be practically impossible for either party to accommodate.  As a result, it took two years for 23andme.com to get FDA approval for a subset of  its reports.

On the other side of the coin, the FDA can arbitrarily dismiss concerns for public safety (arguably to protect the status quo). For example, in denying my Petition to halt the use of ACT chemotherapy for breast cancer patients with a BRCA1 mutation, the FDA said, in part, “There are side effects with all chemotherapeutic agents (not just ACT)”.

Given the intensive FDA involvement, the delays, costs and risks of the current drug approval process are too great for most healthcare solutions to get  funding to cross through the valley of death.

Viewed in this light, the greatest beneficiaries of the current drug approval process are the large pharmaceutical companies who are protected from competitive entry.  Yet under the approach proposed herein, these pharmaceutical will be able to compete for new cures on an equal footing.

More importantly, by shifting the responsibility for public safety and effectiveness to licensed research organizations and putting strong oversight by the FDA into place with punitive remedies for jeopardizing public safety, the advancement of safe and effective new cures will accelerate without undue risk of harm.  In quantitative terms, the gains to society will be more than one hundred times greater than the cost of the increased risk to public safety.

Likewise, the pharmaceutical companies will play a larger role in the process and stand to benefit from the faster implementation of viable healthcare solutions.

The impetus for change to the drug approval process may not be publicly visible, but it would be a godsend to the public. In actuality, the FDA has done a good job given its expansive charter.  However, the FDA’s legislative charter is the kiss of death for millions of people who will never experience the benefits from the healthcare solutions emanating from the world of “Big Data”. It is time for Congress to recognize this national tragedy and to do something about it.