Regulatory Fundamentalism: An overly strict application of the rules governing the conduct of clinical trials that increases research costs and slows the speed of RCTs without improving safety.
Let’s say you are diagnosed to have a life-threatening cancer that is not curable with approved therapies. You might assume (hope) that you would have access to the very latest in experimental therapies, including those under study in randomized clinical trials (RCTs). Faced with an incurable disease you might chose to consent and receive access to such research drugs (likely as part of a RCT), once informed of its potential risks and benefits. As a sovereign person you might be willing to tolerate increased risk of drug toxicity if your tumor were incurable. While there has been substantial progress in curing leukemias and lymphomas, prognosis remains poor for many common solid tumors, such as non small cell lung cancer (NSCLC). Most patients with NSCLC present with advanced disease (large tumors that may have already spread) and, with conventional therapy (cisplatin- or carboplatin-based chemotherapy combinations or new targeted agents such as epidermal growth factor antagonists) most survive only ~1-2 years after diagnosis.
Unfortunately, your assumptions would be overly optimistic. In many cases you will not have the opportunity to access the latest in therapy and the reasons for this are surprising (if complex). Part of the reason that promising therapies are unavailable is that research regulations, which are meant to protect patients, have become so dense and are enforced with such zeal that fewer drugs are being tested (i.e. we are getting fewer shots on goals in our efforts to cure cancer). Many clinical investigators have left the clinical trials arena dismayed by the cumbersome, adversarial and legalistic nature of the study regulation.
As President of the Canadian Association of the Professors of Medicine (CAPM), an organization comprised of all the Heads of Academic Departments of Medicine in Canada, I recently invited Dr. David Stewart, , Head of Medical Oncology at the Ottawa Hospital to speak at our annual meeting about this issue. Dr. Phil Wells, past president of CAPM brought Dr. Stewart to my attention, noting his concern that excessive regulation of clinical research is slowing progress toward a cure for cancer without protecting patients. Dr. Stewart introduced the concept of regulatory fundamentalism.
Dr. David J. Stewart, MD, FRCPC, Head, Division of Medical Oncology at the Ottawa Hospital & University of Ottawa
Let’s start with a statement of need-why is timely access to new cancer therapies such an important public health issue? First, Cancer is common and many forms are lethal. In their 2015 report of cancer statistics the Canadian Cancer Society reports that about 2 in 5 Canadians will develop cancer in their lifetime, and about 1 in 4 Canadians will die of cancer (see below).
Like many oncologists who have to look patients in the eye and deal with our inability to cure their disease, Dr. Stewart is a passionate advocate for accelerating testing of new cancer therapies. He notes that the delay from drug discovery to marketing has increased from an average of 8 years in the 1960s to over 13.9 years in the current era. Paradoxically, fewer patients are being enrolled in clinical trials. If the disease is lethal and the current therapies are often non curative, why are fewer than 5% of cancer patients enrolled in clinical trials? He makes a compelling case that the current regulatory system is overly focused on the pursuit of absolute patient safety and has become so nit picky, risk-averse and rule-bound that progress has been stifled. Ironically neither the patient (who has the incurable disease) nor the doctor (researcher) has a say in deciding how much risk they are willing to assume.
His presentation raised several questions
1) Is a proliferation of regulation and fundamentalist application of these rules contributing to the delay to market and slowing progress toward curing cancer?
2) Is excessive regulation an important driver of the rising cost of getting a new drug to market?
3) Is a fundamentalist interpretation of research regulations adversely affecting the health of patients with lethal diseases by delaying their access to experimental drugs and devices?
4) Are we being too paternalistic in assuming (through our regulatory structure) the balance between risk and benefit that patient’s are willing to accept?
To ensure a common frame of reference let me offer a definition of regulatory fundamentalism (as applied to regulation of RCTs). This definition is based on some etymologic trolling though the Oxford English Dictionary which defines regulate and fundamentalism as follows:
Thus, one can define Regulatory Fundamentalism as: An overly strict application of the rules governing the conduct of RCTs that increases research costs and slows the speed of RCTs without improving safety.
To be clear, rules and regulations to govern testing of new drugs and devices in clinical trials are meant to ensure patients safety and third party oversight is required. The responsibility for enforcement rests in a regulatory matrix that begins with the government (Health Canada) (or in the USA the Food and Drug Administration, FDA). I mention the FDA because as the 800lb gorilla of regulation, their practices and policies reverberate beyond the US border and influence Canadian practices. Locally, clinical research must be approved by each university’s institutional review board (IRB). The protocol for the RCT is reviewed (and painstakingly modified in most cases). In addition, the investigator must pass courses documenting their familiarity with patient safety and research ethics. Parenthetically, because of the complexity of research studies (due in part to regulatory fundamentalism) many study sponsors (Pharmaceutical companies) delegate oversight to Contract Research Organizations (CROs). CROs make their money by implementing the trial and monitoring the trial data that is collected so that it will meet regulatory needs (and of course address the research question). While the government, IRBs and CROS are involved in RCTs the responsibility for any wrong-doing or adverse outcome rests with the principal investigator (PI) and it is the patient who suffers the consequence of an incurable disease. In an era of regulatory fundamentalism, the PI can realistically expect to be visited by monitors from the sponsors who ensure documentation is acquired correctly. They are increasingly likely to be audited by regulators who are increasingly obsessed with minor failures of documentation (as well as serious protocol violations).
You might be thinking, that Dr. Stewart and I are cowboys who are trying to skirt regulation. Surely data accuracy and safety are directly related. This surmise would be true if rules were enforced reasonably (with cognizance of the cost of going too slowly) and if errors were clearly ranked, differentiating the clerical from the substantive. However, a philosophy that is somewhat adversarial (regulatory fundamentalism) embraces excessive regulation and promotes overly rigorous enforcement of the rule apparently assuming that researchers would otherwise fail to make patient safety a primary concern. Thank heavens these same regulations are not guiding approved therapies or the health care system would grind to a halt!
If we make an analogy between the guidance of the RCT (a vehicle to take us to discovery) and an automobile we might say that there are many hands on the steering wheel (government, Pharma, IRB, CRO). It’s hard to steer with all that help!
Artist: Mike Baldwin
Moreover, our partners are “nervous” drivers. The government agencies are understandably risk-averse (think thalidomide), as is the hospital/university IRB (think about bad press). The CRO is paid by sponsors based on their deliverables (ie monitoring performed/files submitted); they are professionals-but are not patient advocates.
Perhaps the regulations would be worth the cost if they saved lives. However, the increased regulatory demands (pages and pages of forms completed on the minutiae of trial experience with little focus on drug effectiveness) has had modest effects on safety.
In an article called Equipoise Lost, Dr. Stewart notes that mortality rate in the riskiest type of clinical trial, the Phase 1 study, in which drug dose and drug toxicity are being established for the first time, has dropped by only 0.3%. This means that all the regulation has reduced death rate minimally even in these high risk early phase RCTs. The cost of the regulation per life/year saved is estimated to be a whopping $2.7million.
Table 1: Mortality rate in Phase 1 trials has declined by 0.3% in the modern era
|Years||# Trials||# Patients||#/% toxic deaths|
|* R. Kurzrock, 2009 ** E. Horstmann et al, NEJM, 2005|
Stewart estimates the cost of our false sense of security (i.e. the cost of the 0.3% reduction in mortality is excessive compared to what we are willing to pay for other interventions that are considered costly-see Table 2.
Table 2: Estimated cost/life-year saved by RCT regulations vs other interventions
|Clinical trials regulations||>$2,700,000|
|Statins for heart disease||$25,000|
|Colonoscopy (colon Ca)||$14,000|
|Adjuvant trastuzumab, breast Ca||$20,000|
|Paclitaxel/cisplatin ovarian Ca||$26,000|
While all life is precious (ignoring the $2.7million/life year saved), one must consider the cost of the delay that excessive regulation precipitates.
Stewart notes, that in Phase I studies, the mean number of safety monitoring procedures, test or tasks required (deficiencies) detected in the first month of the study has increased 133% in the past 5 years from 45 to 105 events (Craft. Cancer 115: 1592, 2009; Kurzrock. Oncologist 18:308, 2013). Moreover, the enforcement of regulation often feels coercive and makes investigators and their universities unwilling to complain for fear of appearing negligent or being branded as unethical. Most of the audited discoveries are neither examples of negligent care or disregard for patient safety; the majority fall better into the category of failure to dot an “i” or cross a “t”.
What is the evidence for this assertion? Let’s look at the USA. The FDA has strong global influence on regulatory tone and practice. The FDA audits of clinical trials are increasingly motivated by routine data verification (88%) rather than for cause (12%) (Morgan-Linnell. Clin Cancer Res 20:3364, 2014). As the rules of audit become more granular, most audits (88.8%) find some “deficiency”. Each deficiency must be addressed and this absorbs investigator time and drains their spirit (69% of audits required investigator action).
Good news! We are making progress toward a cure. Stewart’s offers a list of new potential treatments for cancer showing the magnitude of their benefit (relative to best available standard care), based on RCT data. In aggregate, these new agents improve median survival: by 0.31 years (see Figure 1 below). In the Figure below Dr. Stewart has summarized the life/years added (based on RCT data) from each agent (listed by the tumor it is indicated as therapy for).
Figure 1: Impact of new agents on survival (expressed as life/years saved)
However, the very fact that these experimental drugs are effective means that delays are depriving patients of receiving potentially beneficial therapy. Stewart estimates that if access to these new drugs is delayed, the consequence is a substantial loss of life, estimated below (Figure 2) Admittedly, this figure optimistically assumes that people across the globe would have access to new cancer therapies were they approved-and that’s a topic for another blog!).
Estimated Life Years Lost Per Year of Delay in Drug Approval (globally) The estimates are based on the benefit of the drug on survival (in RCTs), the incidence of the tumor and the assumption that all patients with these tumor types would receive the drug and respond as well as in the RCTs. Persnickety adherence to the letter of the regulation and expansion of regulations also drives cost of the drugs that do make it to market. It is touted that the average cost to bring a drug to market has more than doubled in the past decade, exceeding $2.5 billion USD (although Doctors Without Borders, believe these costs are inflated).
While we can use the rising costs to develop a new drug to pillory Pharma, the Insurers, and/or Physicians, regulatory fundamentalism contributes to this bill. For example, the average cost per patient on a phase III trial has increased from $26,000 in 2006 to $74,800 in 2013 … much of this the result of the time and effort to comply with the documentation of minutiae required y regulators. An industry has sprung up around the legitimate and important regulatory role of clinical research which has slowed progress and contributed to a bloated clinical research enterprise that is expensive, lacks agility and ironically-is not doing much to enhance the research subject’s safety.
What of solutions? In a recent article in the journal Clinical Cancer Research Dr. Stewart and colleagues suggest a variety of reforms to accelerate drug development for patients with lethal diseases.
I suggest additional changes to accelerate and promote clinical trials in lethal diseases
- A change in attitude and tone by regulators: Drug development schedules that require a decade are of little comfort to patients with diseases whose likely survival is measured in months. Perhaps it is time to ask patients with incurable diseases how much risk patients with life-threatening diseases are willing to take in clinical trials and simplify RCT regulations. A less paternalistic regulator model would be welcome by patients with diseases like metastatic melanoma or NSCLC, as well as researchers.
- Improved funding of universities so that the preclinical animal studies can be conducted rapidly and well (i.e. provide overhead funding to universities in support of their research mission). Currently, unlike our US counterparts, the more research we do the more money we lose because universities receive a paltry 12% overhead on a grant and thus struggle to support animal care faculties and offer scientists well equipped labs. Since Pharma has contracted its in house research work force this is the ideal time for Canadian Universities to step up and provide the preclinical research upon which drug development is based.
- Simplify and focus reporting of adverse events by reducing/eliminating focus on documentation all minor toxicity in trials. We should take have a more pragmatic focus of serious toxicity (Grade 1-2) with enhanced post-marketing surveillance for patients with lethal diseases (Redefining cancer: a new paradigm for better and faster treatment innovation).
- Simplify trial design so that CROs and other organizations are not required which drives trial cost. This can only happen once regulatory requirements are modified.
- Refocus our research efforts on break thorough therapies-not me too drugs or drugs with minimal efficacy. I discussed this with Elizabeth Eisenhauer, Head of Oncology at Queen’s University.
She reminded me that delays in approval are often modest if the drug is dramatically effective. As an example she cited the immunological active PD1 inhibitor pembrolizumab (KEYTRUDA). This drug is effective in improving survival in melanoma and moved from Phase I trials in 2011 to FDA approval by 2014. Keytruda® (pembrolizumab). In 2015 Keytruda was approved to treat patients with advanced NSCLC (in patients whose tumors express a protein called PD-L1). The FDA granted Keytruda breakthrough therapy designation because Merck demonstrated its major benefits over available therapies. In contrast, to this 3 year path to approval an agent with minor efficacy, the VEGFR tyrosine kinase inhibitor, cediranib began Phase I trials in 2005 and 88 trials later was found to have a late survival read out showing some benefit. However, all major RCTs of cediranib failed their primary endpoint. So its long road to approval had nothing to do with regulatory fundamentalism and everything to do with poor efficacy. Dr. Eisenhauer also suggested catalysts #6-9 (below).
6. Trials should have simple and meaningful endpoints: e.g. survival
7. The monitoring plan should be based on context (clearly first in human drugs may need more oversight than later trials with a well studied drug class)
8. Avoid paternalism by IRBs: A paternalistic view by IRB members may undervalue the risk the patient faces in the absence of the new device or drug can emerge and slow IRB approval of studies.
9. Adopt different regulations for academic vs. industry trials (particularly for repurposing approved drugs for new indications). Regulations have created burdens for academic trial sponsors such as the NCI CTG. In 2001 regulations changed to require trials testing marketed drugs for reapplication to treat a different disease to meet the same standard requirements as non-marketed drugs. This has been a problem for academic groups. A 2011 Canadian Cancer Research Alliance (CCRA) report on the State of Cancer Clinical Trials In Canada identified the risk that over-regulation holds for oncology clinical trials. They noted that, “Without clinical trials, the outcomes of cancer patients will not continue to improve”. The Report noted, “falling performance metrics, increasing complexity and workload, and an increasingly onerous regulatory environment, clinical trials are at risk”. A national effort to streamline regulations for academic trial sponsors has been initiated to DECREASE this burden (see: http://n2canada.ca/isct/).
10. Reward Pharma for dollars spent on research and development-not marketing and advertising. In an excellent article in the October 2015 Globe and Mail’s Report on Business Paul Christopher Webster nicely summarized how Pharma has (mostly) pulled out of Canada and has (in general) cut back on performing research and discovery. This has occurred while they are making record profits and are expanding expenditure on direct to patient marketing.
11. Make sure MDs remain up to date in the field of science so they are able to understand and evaluate modern clinical trials, which are progressively evaluating biologic agents that require an informed medical consumer. This involves maintaining science as an important foundation of both medical school and residency and incorporating science in our continuing professional development programs.
There is a growing dissociation between discovery scientists who explore complex molecular pathways seeking cures, clinical trialists, who study drugs that they do not themselves discover, and clinicians, who often do not fully understand the proposed molecular mechanisms. It turns out what we don’t understand we tend to neither appreciate nor respect. Thus for investigator initiated trials it is our own colleagues that may be a larger impediment than regulatory fundamentalism.
Let me end with some personal commentary about delays in getting experimental therapy to cancer patients. Inspired by similarities between PAH and cancer, and the loss of a family member to NSCLC, my colleague Dr. Evangelos Michelakis and I began to do cancer research in the early 2000s.
We discovered and patented an invention-namely that inhibitors of an enzyme, pyruvate dehydrogenase kinase (PDK) could treat several human cancers in animal models. The PDK inhibitor, dichloroacetate (DCA) turns on mitochondrial metabolism and regresses cancer by causing cancer cells to undergo programmed cell death (apoptosis). Dr. Michelakis took this work to the bedside and showed that DCA could safely shrink tumors in a lethal human cancer, glioblastoma multiforme.
The PDK inhibitor dichloroacetate (DCA) regresses glioblastoma multiforme
Michelakis E. et al Sci Transl Med 2, 31ra34 (2010)
(Archer and Michelakis Canadian Patent Application No. 2,604,625Title: Method of Treating Cancer Using Dichloroacetate).
With this discovery did industry came knocking? Is dichloroacetate being tested in clinical trials? Is the drug or one like it almost to market? No, NO and NO. While not maligning anyone, the incentives for industry are profit oriented (although they share the patient and physician’s interest in a cure). The regulatory costs for moving a generic molecule to market and the low yield expected from “usage patents” (patients like this one which describe the why and how-rather than the invention of a molecule) make the large investment appear prohibitive. Regulatory fundamentalism is a driver of these costs and suppresses the willingness of investors and Pharma to take risks on new molecules.
In summary-we need change! As patients (and we are all potentially future or present patients) we can voice a demand for accelerated discovery. Patients can advocate for a less paternalistic view of the risk/benefit equation in human research. A view that recognizes the risks of delay or failure to make new agents available to people with incurable diseases would be helpful. As physicians and researchers, we can add our voices to the debate and join Dr. Stewart in his question for a more streamlined approval process as part of a package of reforms required to accelerate progress. We need to lobby our political representatives and patient- focused non governmental organizations, such as the Canadian Cancer Society, to convey the message that a less fundamentalist process for approval of drugs will save lives. As clinicians we also need to stay UptoDate® with Science, as well as Medicine.
Let’s end regulatory fundamentalism!