How do researchers decide early clinical trials?

Launch of clinical investigation represents a substantial escalation in commitment to a particular clinical translation trajectory; it also exposes human subjects to poorly understood interventions. Despite these high stakes, there is little to guide decision-makers on the scientific and ethical evaluation of early phase trials.

In our recent article published in Medicine, Health Care and Philosophy, we review policies and consensus statements on human protections, drug regulation, and research design surrounding trial launch- concentrating on evidentiary factors used to justify launch of clinical development and to evaluate risk and benefit for subjects. We conclude that existing policies grants very wide moral and scientific discretion to research teams and sponsors. We then review what is currently understood about how research teams exercise this discretion, and find that decision-making surrounding trial launch is not simply- or even primarily- centered on proof of principle or concerns about subject welfare. It involves a constellation of commercial, regulatory, and professional considerations. Investigators are adept at establishing and maintaining their authority over decisions surrounding trial launch, and they emphasize that preclinical research is an important resource in legitimizing trial launch and enrolling other actors. However, nothing in this last point suggests that preclinical studies are mere rhetorical devices. If preclinical research is a key resource in enrolling other actors, it is surely because it contains content that resolve certain uncertainties.

We close by laying out a research agenda for characterizing the way investigators, sponsors, and reviewers approach decision-making in early phase research. We suggest that by investigating how various stakeholders describe, reason about, approach and resolve questions about ethics and study validity guidance can be established on a design and review principles for trial launch. Such an approach can pay dividends by improving human protections, reducing attrition in drug development, and reducing costly uncertainties encountered in deciding launch of clinical development.

2016 Mar

Clinical Trial Disaster in France

Days after receiving the experimental medication BIA 10-2474 in a first in human trial, one man was brain dead and another five hospitalized. According to recent reports, three are likely to have neurological deficits.
To my knowledge, this is the first time since 2001 that a healthy volunteer has died in a medical experiment. And it is the first major drug disaster in a phase 1 trial since 2006, when six men were hospitalized after developing a life threatening (but not ultimately fatal) immune response to the drug TGN1412.

Details surrounding the BIA 10-2474 trial are sketchy – and are likely to remain so as long as a manslaughter investigation is underway. Here is what we do know. The drug was a small molecule inhibitor of an enzyme involved in endocannabinoid metabolism, fatty acid amide hydrolase (FAAH). Other FAAH inhibitors have been tested in human beings without incident. Nor have any shown clinical activity. We also know that the men who developed life-threatening toxicities were the first to receive multiple doses of BIA 10-2474. And, based on a study protocol released by Le Figaro, multiple doses within this cohort were not staggered.
Healthy volunteer phase 1 studies are creepy. The realm of phase 1 testing is secretive, and most studies are conducted in private contract research organizations rather than academic medical centers. Few studies are ever published. Indeed, drug regulators exempt companies from even registering them in public databases. This makes it difficult to know anything about their volume, record of safety, the demographics of study participants, or the nature of study procedures.

Another reason phase 1 studies are creepy is that this is one of the few areas where doctors perform medical procedures – including administering unknown substances – that have no conceivable medical benefit for subjects. The risk/medical benefit ratio is infinite. All research on human beings, in a sense, treats people as (consenting) biological objects. But nowhere is this moral dynamic more stark than in healthy volunteer phase 1 trials, where people are valued not for exercising distinctly human capacities like labor or character – but rather for their biological passivity.

Another reason phase 1 studies give one pause is the financial element. They are funded by one of the most profitable sectors of the contemporary economy: the pharmaceutical industry. And some of that economic might is used to recruit volunteers who are probably financially disadvantaged or underemployed. Based on the figures I’ve seen, you can make a handsome sum being a professional “guinea pig.” When I lived in Berlin, I visited a large phase 1 clinic operated by Parexel near Westend. It was one of the few places far outside the museum and Reichstag districts where one saw English signage. The signs were not trying to reach tourists, or native Germans.

We should pay attention to this creepiness, but we should also discipline it with reason. Almost every modern drug we ingest – including many cancer drugs – began its human career in healthy volunteers. I’m hard pressed to think of any morally preferable alternatives to healthy volunteer testing if we value medical advance. Asking patients who are already debilitated by illness to commit their time and bodies for such studies is hardly more appealing. Based on what little has been published on them, healthy volunteer phase 1 studies are mostly benign medically. And I am not being an apologist when I point out that the medical screening procedures performed by Contract Research Organizations provide a service to precisely those populations that are probably underserved in primary care. Visit a phase 1 healthy volunteer clinic and you’ll see armies of people – of various races and age – listlessly peering at their laptops or plugged into ear buds, waiting around for the next blood draw.
Phase 1 trials can and should be done better. The lack of transparency – including nonpublication – is unacceptable. Indeed, drug companies should be expected not only to publish their phase 1 studies, but also the preclinical research leading to them. In the case of BIA 10-2474, I have been unable to find a single published preclinical study of the compound. Indeed, I only learned of its composition through the leaked study protocol. In addition, there are probably many phase 1 studies whose contribution to the pharmacopeia is marginal, because drugs lack a sound biological rationale or are not directed at medically urgent applications.

It is far too early to draw any specific conclusions about the conduct of regulators, BIAL, Contract Research Organizations, or the physicians involved in the BIA 10-2474 debacle. However, here are two interpretations I urge we avoid.
On the one hand, we should avoid the temptation to explain events like this as inevitable, “long tail” phenomena. Some commentators argued this view after the TGN1412 disaster. It’s wrong, because behind every debacle is a chain of rectifiable human events that led to it. Every realm of risk and technology – airplane travel, nuclear power, chemical manufacture, mining – has proven it is possible to devise systems that render avoidable what some might call “inevitable” disaster. When all is said and done, the BIA 10-2474 debacle will reveal some correctable problem in the incentives, practices, organizational structure, and environment in which drug research is pursued.

On the other hand, we should resist the temptation to vilify the institution of phase 1 healthy volunteer testing. To be sure, trials can be better justified and reported. And there is no obvious way to cleanse them of the taint described above. Healthy volunteer phase 1 studies remind us of ineluctable moral tensions in all human research. But, to quote Paul Ramsey, we should continue to “sin bravely.”

This commentary was also published on Impact Ethics:

2016 Feb

Why clinical translation cannot succeed without failure

Attrition in drug development – that is, the failure of drugs that show promise in animal studies to show efficacy when tested in patients- is often viewed as a source of inefficiency in drug development. Surely- some attrition is just that. However, in our recent Feature article in eLife, my long time collaborator Alex London and I argue that some attrition and failure in drug development directly and indispensably contributes to the evidence base used to develop drugs and practice medicine.

How so? We offer 5 reasons. Among them is the fact that negative drug trials provide a read on the validity of theories driving drug development; and that negative drug trials provide clarity about how far clinicians can extend the label of approved drugs. Another is that it is far less costly to deploy cheap (but error prone) methods to quickly screen vast oceans and continents of drug / indication / dose / co-intervention combinatorials. To be clear- our argument is not that failure in drug development is a necessary evil. Rather, we are arguing that at least some failure is constitutive of a healthy research enterprise.

So what does this mean for policy? For one, much of the information produced in unsuccessful drug development remains locked inside the filing cabinets of drug companies (see our BMJ and BJP articles). For another, even the information that is published is probably underutilized (see, for example, Steven Greenberg’s analysis of how “negative” basic sciences are underutilized in the context of inclusion body myositis). Our analysis also suggests that attempts to reduce certain sources of attrition in drug development (e.g. shortened approval times; use of larger samples or more costly but probitive methods in early phase trials) seem likely to lead to other sorts of inefficiencies.

One question our paper does not address is: what is the socially optimal rate of failure in drug development (and how far have we departed from that optimum)? This question is impossible to answer without information about the number of drug candidates that are being developed against various indications; the costs of trials for those treatments, base rates for success for various indications, and other variables. We nevertheless hope our article might inspire efforts by economists and modellers to estimate an optimum for given disease areas. One thing we think such an analysis is likely to show is that we are currently underutilizing the information generated in unsuccessful translation trajectories.

2015 Nov

Balancing the Evidence: Animal efficacy studies should have more weight in the risk/benefit calculus ahead of clinical trials


Clinicians, sponsors, ethics review committees, and others are charged with ensuring that risk is in a favourable balance with benefit when patients enrol in trials. Yet how do they make this judgment, when the only evidence available is from preclinical animal studies?

In our recent article published in the Journal of Medical Ethics1, we offer an answer to this question. We argue, first, that review committees and clinicians should evaluate the clinical promise of a new intervention based on preclinical efficacy evidence. Specifically, they should form judgments about the clinical promise of a new intervention based on how well preclinical studies address common threats to clinical generalization. Second, review committees should adjust their estimates of clinical promise based on clinical trials with related drugs (e.g. within the same class of agents). All else being equal, the stronger the clinical promise, the stronger the moral justification for embarking on an early-phase clinical trial.

Before consuming scarce financial and human resources, including exposing volunteers and patients to potentially ineffective treatments in trials, researchers should capitalize on the knowledge gained from animal efficacy studies. Ethics review committees and clinical investigators will likely protest they lack the expertise, skill or time to evaluate preclinical evidence. Or, that drug regulators like FDA already make such judgments.

However, the FDA itself states that “lack of… potential effectiveness information should not generally be a reason for a Phase 1 IND to be placed on clinical hold2,” and it explicitly delegates judgments about risk/benefit to IRBs3. This means that, if ethics reviewers and investigators really do lack the expertise, skill or time to review preclinical evidence- they are surrendering their mandate to protect patients from undue risk in early phase trials.

  1. Kimmelman, J. and Henderson, V. Assessing risk/benefit for trials using preclinical evidence: a proposal. J Med Ethics Published Online First: 13 October 2015 doi:10.1136/medethics-2015-102882 
  2. Guidance for Industry: Content and Format of Investigational New Drug Applications (INDs) for Phase 1 Studies of Drugs, Including Well-Characterized, Therapeutic, Biotechnology-derived Products. In: Services USDoH, ed., 1995. 
  3. 46 FR 8975, Jan. 27, 1981, as amended at 56 FR 28029, June 18, 1991; 66 FR 20599, Apr. 24, 2001.
2015 Nov

Too much of a good thing

By Linda Bartlett (Photographer)

A novel anti-cancer drug is found to shrink every tumour type tested in experimental animal models. Let’s rejoice and start clinical trials without delay!

Well, not so fast.

Preclinical experiments in animal models are aimed at showing that a new drug will be useful in human beings. However, individual animal experiments are often too small to support inferences about clinical utility. Techniques such as meta-analysis offer an attractive method for pooling individual studies and supporting more confident assertions regarding the potential clinical utility of a new drug.

With this in mind, our team undertook a meta-analysis using the anti-cancer drug sunitinib (Sutent). Since its publication in eLife, the meta-analysis has been covered in the BMJ, Nature, The Guardian, and Retraction Watch. We were primarily interested in whether the properties of sunitinib observed in preclinical studies would correlate with those observed in patients

What we found were many avoidable roadblocks to proper meta-analysis. Firstly, poor reporting quality in many experiments made it impossible to include them in the analysis. Examples include lack of error bars or no sample size given. Secondly, poor methodological practices such as lack of blinding and randomization, or reliance on single models were common, calling the quality of the data we were extracting into question. Furthermore, there was no discernible dose-response relationship connected across experiments in the same malignancy. Finally, trim and fill analysis (a tool to detect the possibility of publication bias) suggested that the overall efficacy of sunitinib across all malignancies could be inflated as much as 45%.

Other investigators, including those just published in PLoS Biology, have shown similar results in other diseases and have called for reforms to the way we perform and report animal experiments meant to influence the decision to move into clinical trials. It may be time that we looked more closely at the way we test new anti-cancer agents in animal models to ensure we are getting the highest quality data to make decisions.


2015 Oct

Acting on “Actionable Mutations”

The new buzzword in personalized cancer medicine is “actionable mutation”. This label is given to the genetic aberrations that are present in some patients’ tumors, and that are intended targets of new drugs. Increasingly, treatment decisions in routine clinical care, and enrollment in trials are being guided by the concept of “actionable mutations”.

However, determining “actionablility” is an ongoing challenge. The function of a particular mutation, to what degree it is responsible for driving a given malignancy, and how easily it can be targeted by a specific therapy, all affect how “actionable” it is. Further, a tumor’s genomic profiling can vary depending on which tissues are biopsied (ie primary and metastatic tumor sites), or when they are biopsied (before and after particular treatment regiments). “Actionability” of mutations may not be a stable variable that is easily transferred from one clinical setting to another. Instead, the concept of actionability invites further clarification on where a when and these genes should be screened.

Unfortunately, lack of clarity on the definition of “actionable mutations” has not prevented its uptake in either the commercial or scientific medical communities. For example, upon physician request, diagnostics company Foundation One will sequence over 300 genes that are known or likely targets of a specific therapy and provide clinicians with a report listing all of their patient’s actionable mutations. These are mutations they define as all those that can be targeted by both therapies currently approved for their indication as well as those approved for other malignancies (off-label). Further, some mutations are designated “Equivocal” signifying that there isDSC03420-B3 some, but not confirmed evidence, supporting an aberration in a patient’s sample, or “Subclonal” where an abnormality only exists in less than 10% of a patient’s tumor. However, absent clear guidelines or standards surrounding actionable mutations it can be extremely difficult for oncologists to interpret these often ambiguous results.

A number of next generation clinical trials currently underway are allocating patients to treatment arms using similar targeted strategies. Basket trials, like the NCI MATCH study, are assigning mixed-malignancy, advance cancer patients, to off-label targeted therapies based on the presence of “actionable mutations”. However, here too some concerns have been raised. There is little consensus on how to prioritize certain mutations over others – both in terms of their functional importance and rarity – and has raised issues in dealing with patient’s harboring co-mutations and optimizing allotment to ensure sufficient patient accrual to different arms.

These concerns along with the results of a recently published basket trial in the New England Journal of Medicine should lead researchers and physicians to be cautious in blindly treating “actionable mutations”. The phase 2 study looking at Vemurafenib in BRAF V600 mutation positive nonmelanoma patients found variable response among different malignancies – indicating that genomic signatures should not be the only factor playing into treatment selection.

In efforts to give some clarity to the current situation a couple of collaborations have been recently undertaken. Founded in late 2014, The Actionable Genome Consortium, is a collaboration of biotech company Illumina and four leading cancer centers (Dana-Farber Cancer Institute, Fred Hutchinson Cancer Research Center, MD Anderson Cancer Center and Memorial Sloan Kettering Cancer Center) with the goal to define the “actionable cancer genome” and create robust standards for next-generation sequencing and treatment decision making.

Another initiative, TAPUR (Targeted Agent and Profiling Utilization Registry), led by ASCO, is a prospective, observational, non-randomized clinical study that will track the off-label performance of commercially available targeted drugs in advanced cancer patients. These therapies are commonly prescribed off-label. However, this will be the first attempt to aggregate data to determine the usefulness of this strategy in regards to targeting actionable mutations.

In time these initiatives should go a long way in drawing the boundaries around “actionable mutations.” In
the interim, however, practicing oncologists and researchers alike are left wondering how, exactly, to interpret the “act” in “actionable.”

2015 Sep

Is it ok for patients to pay for their own clinical trials?

Most trials are funded by public sponsors, charities, or private drug developers. Austere research funding environments, and growing engagement of patient communities, has encouraged many to seek alternative funding.  One such alternative is patient funding. In the August 6 issue of Cell Stem Cell, my co-authors Alex London and Dani Wenner ask whether “patient funded trials” represent an opportunity for research systems, or a liability.

Our answer: liability.

Current regulatory systems train the self interest of conventional funders and scientists on the pursuit of well justified, rigorously designed, and efficient clinical trials. These regulatory systems have little purchase on patients or on clinics that offer patient funded trials.  Indeed, patient funded trials create a niche whereby clinics can market unproven interventions in the guise of a trial.  Do a few Google searches for patient funded trials and you’ll see what can flourish under this funding model.

On the other hand, our denunciation of the model is not categorical.  Provided there is a system in place for independently vetting the quality of design and supporting evidence—and for preventing such studies from pre-empting other worthy scientific efforts- patient funded trials may be ethically viable.

Until those mechanisms are in place, academic medical centers should refuse to host such studies.

Edit (2015-09-08): Dani Wenner’s name was mis-spelled as “Danni” in the original posting. We regret the error.

2015 Aug

Predicting Risk, Benefit, and Success in Research

weather-vane-711082_1280 The task set before clinical investigators is not easy. They are supposed to answer pressing scientific questions, using very few resources, and exposing patient-subjects to as little risk as possible. In other words, we expect them to be rigorous scientists, stewards of the medical research enterprise, and guardians of their patients’ interests all at the same time. While the duties that emerge from these various roles are sometimes orthogonal, they are intersecting and aligned at the point of clinical trial design. Insofar as a trial is well-designed–meaning that it is likely to answer its scientific question, make efficient use of research resources, and minimize risk–the investigator has successfully discharged all of these duties.What is more, there is a common activity underlying all of these requirements of good trial design: Prediction. When investigators design studies, they are making an array of predictions about what they think will happen. When they decide which interventions to compare in a randomized trial, they are making predictions about risk/benefit balance. When they power a study, they are making a prediction about treatment effect sizes. The accuracy of these predictions can mean the difference between an informative or an uninformative outcome–a safe or unsafe study.

The importance of these predictions is already implicitly recognized in many research ethics policies. Indeed, research policies often include requirements that studies should be based on a systematic evaluation of the available evidence. These requirements are really just another way of saying that the predictions underlying a study should be as accurate as possible given the state of available knowledge. Yet, trial protocols do not typically contain explicit predictions–e.g., likelihood statements about the various outcomes or events of ethical interest. And this makes it much more difficult to know whether or not investigators are adequately discharging their duties to the scientific community, to their patient-subjects, and to the research system as a whole.

In an article from this month’s Journal of Medical Ethics, I argue that investigators ought to be making explicit predictions in their protocols. For example, they should stating exactly how likely they think it is that their study will meet its primary endpoint or exactly how many adverse events they expect to see. Doing so would then allow us to apply the tools of prediction science–to compare these predictions with outcomes and finally get a direct read on just how well investigators make use of the available body of evidence. This would, in turn, provide a number of other benefits–from facilitating more transparent ethical reviews to reducing the number of uninformative trials. It would also provide opportunities for other research stakeholders–like funding agencies–to better align their decision-making with the state of evidence.

The broad point here is that in the era of evidence-based medicine, we should be using this evidence to design to better trials. Applying the science of prediction to clinical research allows us to take steps in this direction.

2015 Jun

Accessibility of trial reports for drugs stalling in development: a systematic assessment of registered trials

Non-publication of clinical trial results has been recognized as a serious scientific and ethical problem. Underreporting frustrates evaluation of a drug’s utility and safety, and fails to redeem the sacrifice of trial participants.

Thus far, policy measures to counteract non-publication have focused on trials of interventions used in practice. However, 9/10 interventions entering clinical testing never achieve marketing licensure. What happens to the results of those trials?

Figure depicting the rates of publication of trials of licensed drugs compared with trials of stalled drugs--overall and by major subgroup.

Figure depicting the rates of publication of trials of licensed drugs compared with trials of stalled drugs–overall and by major subgroup.

In our most recent publication, my colleagues and I systematically quantified the proportion of trials of unlicensed interventions that are not published.

We used trial registration records to create a sample of clinical trials of drugs that achieved licensure between 2005 and 2009 (“licensed drugs” or “translated drugs”) and drugs that stalled in clinical development (“stalled drugs”) in the same time period. Our sample included registered phase II, III or IV trials that closed between January 1st, 2006 and December 31st, 2008 and tested a drug in the treatment of cancer, cardiovascular disease or neurological disorders. We felt this sample provided a relevant and contemporary look into a wide swathe of drug development activity. We then searched Google Scholar, PubMed and Embase, and contacted investigators to determine the publication status of each trial in our sample at least 5 years after reported primary endpoint collection.

Whereas 75% (72/96) of registered trials of licensed drugs were published, only 37% (30/81) trials of stalled drugs were published. The adjusted hazard ratio for publication was 2.7 (95% confidence interval 1.7 to 4.3) in favour of licensed drug trials–that is, clinical trials of licensed drugs were almost three times as likely to publish findings as trials of stalled drugs. Higher publication rates for licensed drug trials were observed regardless of disease type, sponsorship (industry involvement versus not), trial phase, and location across the globe.

Figure depicting the proportion of trials of licensed and unlicensed interventions that are published as a function of time from reported primary endpoint collection. The publication of stalled drug trials plateaus over time around 37%, whereas the publication of translated drug trials attains 75% in the same time period.

Figure depicting the proportion of trials of licensed and unlicensed interventions that are published as a function of time from reported primary endpoint collection. The publication of stalled drug trials plateaus over time around 37%, whereas the publication of translated drug trials attains 75% in the same time period.

Moreover, a total of 20,135 patients participated in trials of stalled drugs that were never published. In addition to the alarming implications for these patients, trials in unsuccessful translation trajectories contain a wealth of scientific information for research planning, such as validation of pathophysiological theories driving drug development, as well as data about drug safety and pharmacology. All of this information vanishes when trials of unsuccessful interventions are not published.

Our key finding is that much of the information collected in unsuccessful drug trials is inaccessible to the broader research and practice communities. Our results provide an evidence base and rationale for policy reforms aimed at promoting transparency, ethics, and accountability in clinical research. One such potential reform is the Notice of Proposed Rulemaking entitled “Clinical Trials Registration and Results Submission” issued by the US Department of Health and Human Services in November 2014. The proposal, which moves to implement the FDAAA summary results reporting requirements for trials of licensed drugs and to extend them to trials of unlicensed drugs, was closed to public comments on March 23rd, 2015. The rule is now undergoing revision.

2015 May

So How Useful is Basic Science Research?

As part of the STREAM workshop series, Dr. Jeremy Howick gave a seminar on March 25th on the usefulness of basic science research. This a great question to ask, and in his answer he makes two valid observations: 1) there’s a 70/30 split in research funding favoring basic science over clinical research and 2) historically, we have had more scientific advances accidentally than through the current system of finding a protein, characterizing its metabolic pathway, and developing and testing a drug that takes advantage of it. The best example is aspirin – we used salicylic acid for centuries before figuring out how it works.

Dr. Howick argues that we should test the wealth of information we already have – urban legends,  homemade flu remedies, medicinal herbs – in more trials, and depend less on basic science research’s ability to develop drugs that we may (read: probably cannot) end up using. He argues that this is something of a win-win situation. For example, consider the case of chicken soup: Grandmothers are under the impression that it cures colds. If we test its efficacy in trials and it really works, we can figure out how, and perhaps develop a drug that is more effective. Whereas if the soup fails to cure colds, we can then definitively disprove its effectiveness, saving a lot of people, a lot of money in the canned-goods aisle.

Given the billions of dollars wasted each year in drug development, this seems a promising proposal. The presentation sparked an hour-and-a-half-long debate, however, probably because it was given to a room full of people with basic-science backgrounds. Although I agree with Dr. Howick on a lot of points, I don’t think using observational data to generate hypotheses is quite the panacea he makes it out to be. Sometimes testing folk remedies actually ends up creating more confusion. Take, for example, vitamin C: despite the dozens of clinical trials disproving its ability to cure the common cold, I’ve seen people (pharmacology students) swear up and down that it is the best cold remedy ever bottled. This just shows how pseudoscience (I’m looking at you, homeopathy) can be incredibly difficult to disprove.

2015 Apr

Page 1 of 1712345...10...Last »


Old blog posts

Our mission

STREAM Group applies empirical and philosophical tools for addressing scientific, ethical, and policy challenges in the development and translation of health technologies.

Continue reading ...

Who we are

The STREAM Group is a collaboration of researchers who share a common set of principles about the goals and methods for studying clinical translation. Our members work in ethics, epidemiology, biology, psychology, and various medical specialties. The network is centered at McGill University, and has affiliates throughout North America.

Continue reading ...