Cancer research in the United States: Dying by a thousand paper cuts


  • Hagop Kantarjian MD,

    Corresponding author
    1. Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas
    • Corresponding author: Hagop Kantarjian, MD, Department of Leukemia, The University of Texas MD Anderson Cancer Center, 1400 Holcombe Blvd, Unit 428, Houston, TX 77030; Fax: (713) 794-4297;

    Search for more papers by this author
  • David J. Stewart MD, RCPC,

    1. Division of Medical Oncology, Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada
    Search for more papers by this author
  • Leonard Zwelling MD

    1. Department of Experimental Therapeutics, The University of Texas MD Anderson Cancer Center, Houston, Texas
    Search for more papers by this author


The authors discuss cancer research in the United States and the implications of bureaucratic burdens on the conduct and cost of research and, consequently, the cost of drugs and health care. They propose potential remedial solutions.

Clinical research is not clinical care. When physicians care for patients, their only concern is improving the health and life of that one patient in any way possible. There are reasonable guidelines for diagnostic testing and therapeutic interventions, but the physician and patient together outline the care pathway to improved health, based on the knowledge of one and the informed preferences of the other.

Clinical research is different. Historically, experimentation with individuals, now called human subjects research, had a checkered past. The 20th century, a time of great medical progress, was also full of abuses of the rights of patients and clinical research subjects alike. Examples include experimentations in the Nazi death camps, testing the effects of atomic bombs on innocent bystanders, the use of LSD on soldiers, and withholding treatment from black Americans infected with syphilis in the Tuskegee experiments long after the benefits of penicillin were known.

After the horrible inhuman experimentations conducted by the Nazis during the 1930s and early 1940s, various codes for the conduct of human experimentation in medical research were implemented. These included the Nuremberg Code in 1947, the Declaration of Helsinki in 1964 (which was revised in 1975), and the 1971 Guidelines of the US Department of Health, Education, and Welfare (which was renamed the Department of Health and Human Services in 1979). The Tuskegee study uncovered weaknesses in the oversight of medical experimentation, resulting in the Belmont Report, which was released in 1978 by the National Commission for the Protection for Human Subjects of Biomedical and Behavioral Research. The Belmont Report identified 3 basic ethical principles for all future research in humans: respect for persons (proper understanding of research, free-will participation); beneficence (maximize benefit and minimize risk); and justice (avoidance of prejudices).[1] Since then, research on human subjects in the United States is overseen by the Office for Human Research Protections (OHRP), which is part of the federal Department of Health and Human Services. It is guided by clear stipulations in the Code of Federal Regulations and overseen at the sites at which it is performed by Institutional Review Boards (IRBs), ethics panels that review and monitor all proposed human subjects research. The IRBs report directly to the federal government and not to the institutions in which the research is performed.

When this system was established by the Belmont Report, it appeared as if human subjects research had finally turned a corner from the darkness to the light. No one would have anticipated what would happen over the next 30 years.

The rules set by the OHRP have not changed in the past 30-plus years. Yet the burden of cancer research has increased multifold in an increasingly risk-adverse society. In a genuine effort to protect patients from adverse events, regulatory burdens and research rigidity in clinical trials have increased to a point at which such protection is outweighing the benefits, and actually harming patients who are unable to be involved in clinical trials. Although a high degree of risk adversity in medical research is reasonable for most medical conditions, it may need to be different for lethal conditions such as cancer. The rapid advances that occurred against the acquired immunodeficiency syndrome (AIDS) as a direct consequence of clinical research regulations being relaxed illustrate the potential. After the principle of “Primum non nocere” or “first do no harm,” we believe that in cancer the worst harm to patients is an inability to be treated for the cancer when standards of care have been exhausted.

How did we get to this unbearable situation in cancer research? Numerous parties were involved: the US Food and Drug Administration (FDA) and other regulatory agencies, OHRP, the National Cancer Institute (NCI), the Health Insurance Portability and Accountability Act (HIPAA), the Office of the Inspector General, the US Patent and Trademark Office, the Joint Commission, pharmaceutical companies, insurance companies, lawyers, contract research organizations (CROs), and others. Critical evaluation of cancer research in the United States and remedial steps are urgently needed for the following reasons: 1) to reduce the cost of research (and consequently the cost of health care and drug prices); 2) to speed access to effective new agents, thereby reducing avoidable loss of life; 3) to improve and encourage patient participation on trials; 4) to simplify research protocols, shorten time to study initiation, and improve time-dependent accrual; 5) to reduce perceived protocol “deviations” and “violations”; 6) to reduce research penalties and encourage research specialization; 7) to reduce interference by financially beneficiary CROs; 8) to encourage conduct of research in the United States (vs elsewhere); and 9) to enhance investigator-initiated cancer research (vs pharmaceutical company-driven research).

The estimated combined costs per patient entered in a clinical trial was reported to be < $10,000 in the 1980s, but for phase 3 trials, it had increased to approximately $26,000 by 2006[2] and to $47,000 by 2011,[3] with costs as high as $87,000 reported for some trials.[4] Costs are rising much faster than the rate of inflation. The estimated cost per life-year saved under these regulations is approximately $2.7 million[5]; the consensus among oncologists is that a reasonable cost per life-year saved by therapies is $100,000.[6] We estimate that the cost is now as much as $100,000 to $120,000 for some clinical trials, and this increases the estimated cost per life-year saved several-fold. While the usual justification for the increasing regulatory burden is patient safety, objective data have demonstrated little if any translation into added safety. This massive increase in the cost of research per patient over the past decade has coincided with the burgeoning of the CRO industry, which may have added layers of unnecessary bureaucracy in cancer research.

The regulatory burden is impeding progress in cancer research. As a result of excessive restriction eligibilities and mandated treatment schedules and plans that are impractical and/or socioeconomically impossible for patients, <5% of adults with cancer participate in clinical trials.[7, 8]

Not only does this massively excessive regulatory burden markedly increase health care costs, but it also costs lives. Compared with the 1960s, the current regulatory burden has increased the average time from drug discovery to marketing by approximately 5 years.[5] Although it is calculated that improved safety as a result of more stringent regulations may save an average of up to 16 life-years during the process of bringing 1 new anticancer agent to market, this 5-year regulation-induced delay in access to effective new agents may cost millions of life-years worldwide.[5] For example, if a new therapy increased the cure rate of lung cancer by just 1% (through improved adjuvant therapy) and increased the average life expectancy of uncured patients by just 3 months, the regulation-induced delay would cost more than 2,000,000 life-years worldwide.[5] In the United States, it would cost 1 life-year for every 9.3 minutes of regulatory delay. The current situation could be considered ethical only if nearly all new anticancer agents being tested were completely ineffective. Is someone asleep at the switch?

The regulatory traffic jam results in many research protocols facing significant delays in activation and in patient accrual. This causes a major waste of researchers' time and patients' resources. Delays in protocols result in failure to accrue. Approximately 40% of all NCI-sponsored protocols never achieve minimum patient accrual objectives, with the major culprit being delays in protocol activation.[9] A range of 300 to 600 regulatory steps, approximately one-half of which are judged unnecessary, are required to activate trials, with a median time of 400 to 700 days to initiation.[10]

Numerous protocols stipulate studies that are too frequent and unnecessary, including frequent physical examinations, measurements of vital signs, performance of electrocardiography and cardiac monitoring, blood draws, etc. These add to the cost, to the patients' social and financial burdens, and to treatment hurdles, without added patient safety. The nightmarish regulatory burden adds not only to the cost of research but to significant increases in the frequency of protocol “violations and deviations,” which in turn penalize investigators for “faulty research.” Suspending protocols and researchers for poor compliance discourages further research, and has resulted in a gradual decrease in bright young academic physicians pursuing research careers, having observed their mentors and senior researchers threatened, censored, and penalized by IRBs and other compliance and regulatory agencies.[5, 11, 12]

Cancer research is drowning in paperwork and paralyzed by red tape. In the 1980s, protocols used to be guidelines for researchers to conduct proper research, and followed the spirit of the law. Protocol documents were 10-page to 20-page–long descriptions; consent forms were 3-pages to 5-pages long. They were informative and concise. The increasing regulatory burden and litigation have now resulted in protocols becoming “contractual agreements” that follow the letter of the law. Any deviations may result in penalties. Today, an average protocol may be 100-pages to 200-pages long and a consent form may be 20-pages to 30-pages long. This makes it very difficult to find pertinent information or to truly inform patients. Sick patients often do not have the energy or the stamina to read the consent forms or understand the research. They often sign consent documents without attempting to read them, with the often-quoted patient statement being “Doctor, just show me where to sign.” Thus, in our desire to inform patients about protocols and details of research, we have made them less informed. There has been a marked shift from “regulatory rationalism” to “regulatory fundamentalism.”[5]

The US FDA is the final arbiter of the usefulness of any clinical trial of a new drug or device. Until the FDA says a new therapy can be marketed, investigators and manufacturers alike are beholden to the federal agency that holds the key to the kingdom of medical commerce. To make the FDA happy, an entire industry has grown in the form of CROs, which administer trials for the pharmaceutical companies at the academic centers and community offices, monitoring studies and gathering the paperwork so that the trial drug or device will look as good as possible to the federal regulators who stand between the manufacturers and the market. We believe CROs may be major contributors to both the increased regulatory burdens and the cost of research. In the early 1980s, the only CRO known to cancer researchers was the Clinical Trials Monitoring Service (CTMS), launched by an NCI contract to an organization (which in 1982 became Theradex Systems Inc) to monitor NCI protocols. Today there are >1100 CROs worldwide, including >300 in the United States. CRO industry revenues were projected to be between $20 billion and $24 billion in 2010, representing approximately one-third of the total spending on research and development by pharmaceutical and biotechnology companies.[13] According to some cancer researchers, CROs may have become a “cancer within cancer research.” They employ more than 66,000 people worldwide, and include armies of monitors who may not be well versed or trained in their knowledge or monitoring of particular cancer trials. They are paid by the hour, a disincentive to cost-efficient monitoring. The added value of CROs in improving the quality of research or in protecting patients from adversity has not been objectively proven.

Last, one of the most serious implications of the burden of bureaucracy is a shift in cancer research from academic centers to pharmaceutical-based research. With the cost of research reported to be $100,000 per patient on trial, the voluminous protocols and consent forms, and the extremely time-consuming processes to nurture a protocol from conception to accrual, it has become nearly impossible for investigators to conduct independent investigator-initiated studies. Much of the cancer research is now conceived, generated, conducted, analyzed, published, paid for, and advertised through pharmaceutical-associated infrastructures and resources. This raises very serious concerns regarding the future and credibility of cancer research.[14] Cancer researchers at academic institutions may have become “glorified” employees and extensions of pharmaceutical companies, rather than being independent researchers. A high percentage of cancer researchers now simply wait for the next protocol offered by the next drug company and CRO, who impose their rules and regulations, with a major focus on cost and narrow financial aims rather than a broader vision of research devoted to curing cancer.

Cancer research in the United States has reached a state of paralysis. Much of the cancer research is moving outside of the country to Europe and other countries, where regulation is better streamlined and the cost of research is lower. We propose to highlight the 10 to 20 most common hurdles in cancer research, analyze them for their cost-benefit and value added to patient safety, and initiate deregulatory processes to shift back research to a more rational approach. Our aim should be to markedly reduce the cost of clinical research (reasonably to < $20,000 per patient), reduce the time from a research concept to implementation to an average of < 3 months, reduce the burdens of regulation on patients themselves so they are more willing to participate in clinical trials, and overall rebalance the risks and benefits perceived in cancer research trials.

As oncologists, we believe a far simpler system is needed to hasten the development, testing, and marketing of drugs and devices for potentially lethal conditions such as cancer. Potential solutions include:

  1. Protocol documents need to be no longer than necessary to adequately describe the research so that peer-review can be completed quickly.
  2. Testing (blood tests or imaging) within trials should be kept to a minimum and only up to the standards of good clinical care. Excessive electrocardiograms, chest x-rays, computed tomography scans, and magnetic resonance imaging increase cost and may not contribute to the patient's well-being or the information generated. More importantly, most tests do not really add to safety at all.
  3. Consent documents should describe the risks and benefits concisely and simply, but should not resemble legal documents that make sure no one is accountable or liable for an adverse effect that could not have been predicted.
  4. The US FDA should focus on meaningful endpoints in cancer trials, mostly quantity and quality of life. Having a fifth or sixth drug with a mechanism that is identical to that of past drugs with equally suboptimal effects is jamming up the regulatory pipeline.
  5. CROs would be less necessary if these changes were made, which alone would save a great deal of time and money.

The rules governing clinical research are necessary to curb the abuses that history has shown can infect the behavior of well-meaning caregivers who see the ends justifying the means. In human subjects research, this should never be the case. However, the current system needs an overhaul to reduce the expense of discovering new treatments and to categorize the rules for research on potentially fatal diseases differently from those that govern research for the next head cold reliever. There is a lot at stake for everyone. Much of the wasted research infrastructure and resources could then be used for more and better research rather than to pay a series of middlemen who contribute little or nothing to the well-being of patients with cancer.

What is the probability of successfully changing the current situation? We have had colleagues tell us “The situation is hopeless. Our current system is incompetent and cannot be changed.” But it is our system. The current situation could not have come to exist without our apathy and complicity. We have the responsibility and the ability to bring about the changes needed, but these changes will only be possible if physicians face up to their responsibility and take the lead.


No specific funding was disclosed.


The authors made no disclosures.