Meet Science: How clinical trials work

Did you know that, with a properly conducted series of clinical trials, it can take upwards of 20 years before a medical discovery makes it from the lab to the hospital?

Judy Stone, an infectious disease specialist who does clinical research, has a guest post on the Scientific American blog network today, explaining the basics of clinical trials—where they came from, and how they can go wrong.

She's going to be publishing a series of posts on this topic, and is looking for input on what you want to know about clinical trials. Disclaimer: As a clinical researcher, Stone has a goal here. She'd like to see more people volunteering for clinical research, and part of what she's interested in is the gaps in knowledge that make people wary of participating, or leave them unaware that they can participate. Your input would be helpful.

Image: Pills Phial, a Creative Commons Attribution Share-Alike (2.0) image from luca_volpi's photostream

Clinical trials seek to learn whether a drug (or device) works as expected—it’s unknown, until tested in people. That’s why early phase trials use only a few people, and more are added as experience is gained. Sometimes unexpected discoveries are made along the way. For example, Rogaine was discovered by an astute clinician researcher during a clinical trial studying high blood pressure. The drug, minoxidil, originally under study as an anti-hypertensive medication, was serendipitously found to have the unexpected side effect of stimulating hair growth, prompting a whole new line of products for baldness.

Similarly, Viagra was discovered by accident. Sildenafil, the generic form, was being studied as a treatment for angina, as it dilates blood vessels by blocking an enzyme, phosphodiesterase (PDE). While not very effective for angina, it was found to prolong erections, stimulating the whole “life-style drug” industry. Fortunately, PDE inhibitors are now being found useful for a host of important medical conditions, ranging from pulmonary hypertension to asthma and muscular dystrophy.

Of course, not all inadvertent discoveries have such rosy outcomes.

For example, Diethylstilbesterol (DES), a synthetic estrogen, was commonly prescribed in the US 1938-1971, to help prevent miscarriages. It was only after many years that DES was found to cause a rare type of vaginal cancer in daughters of exposed women. Later, other types of cancers showed up as well, in small numbers.

Via Aaron Rowe


  1. Hmm, no questions in particular, but I would highly recommend telling the story of the early HIV retrovirals, and how the participants basically ruined the study.  The harsh truth, after all, is that you need some people on the placebo, or the study is meaningless.

    1. kattw – I agree completely – as hard as it is for people who are in the middle of a disease to consider, the trials do have to be done properly for the result to be meaningful. The history of cancer treatment is filled with cases where insufficient research was done, and treatments that seemed promising were given to many people who ended up being made much worse off (and driving their families into bankruptcy).

      Breast cancer is one area where alliances between pharmaceutical companies and patient advocacy groups has been able to create useful compromises between terminal patients who want early access to experimental drugs, and drug companies who need to have well-run trials in order to demonstrate real effectiveness. Those groups can easily torpedo each other’s efforts, which has done nothing but delay drug development. When they work together realistically, and make realistic compromises then the process can be sped up.

  2. I work in clinical research (specifically compliance billing)  I would not say that studies themselves take 20 years, but there are tons of rules etc. (thank god the republicans have not gotten their way and we still have phase I drug studies).

    The main problem we have is that it is often hard to find enough people that fit the study criteria and get all the regulatory “paperwork” done before the sponsor (who usually have multi site trials running) decide to pull the plug on our site because they have reached their enrollment metrics through other sites.

    it is very fun though to hear about all the studies and what they are doing from across the entire UMHS research wing (30+ departments and divisions)

  3. I was introduced to the world of clinical trials 10 years ago when my son was diagnosed with leukemia. I immediately read a book (strangely titled “Into the Age of Miracles”) that described how clinical trials were the only thing that made it possible for my son to have a hope of surviving this disease. Naturally, when the doctors asked us to enroll in a clinical trial, I said “Yes” without hesitation.

    He did survive quite nicely, and I’m eternally thankful to those kids who endured all sorts of nasty treatments (and to the ones selected for the non-treatment arm!) so that others after them would have better lives.

  4. A very interesting new book that goes into this topic quite extensively (though isn’t specifically about it) is The Emperor of All Maladies, which is a biography of cancer, and a history of cancer treatment from ancient Egypt to last year. It is a bit dense, but very well written.

  5. This is embedded in a link in the article, and I know all of us can google, but if you hadn’t considered it, there may be a trial out there appropriate for you or a loved one. This is the NIH clearing house site for ongoing trials, and also has a database on results of earlier trials:
    Not all studies are based solely in the US.

    If you’ve no medical condition, that’s great, as some studies also need healthy volunteers.  Some studies compensate with money, some with some free health care in terms of treatment/exams or monitoring.  This is also a classic way to obtain superpowers, though YMMV.

  6. glad to see that an important subject like this is being discussed in mainstream blogs. Only 14% of americans have heard of clinical trials. 50% of clinical trials are delayed because they cant find participants for the study, even though 76% of americans when asked say they would like to know more about clinical trials and participate in them.  Without such studies medicine cant proceed. 

    To help address this we have built to make it easy for patients to find clinical trials and connect with the doctors. We are also blogging about clinical trials on our blog CureTalk so as to increase awareness and participation in this important endeavor. 

  7. People are cautious to participate in studies for the same reasons they tend not to swallow random drugs; they don’t know if it’s safe.  Therefore a primary motivator is the knowledge that they’re not safe anyway (in other words, participating out of desperation). 

    At least, that’s how I made my decision not to participate yet.

    1. I appreciate you being honest.
      It’s true that risk can not be absolutely ruled out, though neither can benefit. 

      Risk/benefit for these are as well described as possible, and informed consent happens more commonly in studies than it does in routine medicine.

      The first folks to ever try statins may be around longer than those who didn’t feel safe trying something new.  You could still participate in studies that involve only monitoring/sampling, and not treatment.  There are no guarantees that no harm can be ruled out in alsmost any study, but given the paucity of data on many exisiting meds, there’s no guarantee that the current regimen is safer than the new route.  Consider how long cold medicines were used on young kids, how often HRT was used on women before the research decided it was overall a bad idea.  Many studies are also variations on a theme, either testing the extant therapy, higher or lower dosages, or combinations that are often commonly prescribed.  Many studies involve non pharmalogical treatment.  The risk for these is less likely to have unforeseen medical side effects. 

      Someone will likely point out that if everyone felt your way, homeopathy would be as legit as antibiotics, but you do have to do what’s comfortable for yourself. 

      1. There may come a time when I am desperate and willing to take that kind of risk.  In the meantime, I’m participating in as much as I can without risk: donating blood and tissue, surveys, etc. 

        I have immense respect for people participating in clinical trials.  I’m just sort of hoping that currently proven treatments can keep me around long enough for their sacrifices to yield results. If not, then I guess it’ll be my turn.

        As for the legitimacy of homeopathy, that really depends on who you ask. Hmm. Maybe I should volunteer after all…

  8. Judy Stone’s first article is rather more about the prehistory of drug discovery and the early days of regulation of the industry.  I’m looking forward to reading the next one.

    Managing clinical trials is complicated.  I’ve worked with clinical trial management systems (CTMS) and clinical data management systems (CDMS) over the last ten years, mainly for biotech and pharma companies.  I’m working for a med device company now, who, being a start-up on a budget, hand-crafted both their CTMS and CDMS, unfortunately before I got here, or I would have been able to say “Nooooo!” … So part of my job now is to investigate alternatives for our next trials.  I’m in favour of using external hosted solutions for both systems.  No point in reinventing the wheel, after all.

    Nature also ran a short article last week on the issues with clinical trials with some suggestions for improvements:

  9. Must mention publication bias, which, over time inflates positive results to seem more positive than they are.  That’s because all the negative and neutral-effect trials don’t ever get published.  So the rosy world isn’t as rosy as we think it is.

  10. Well-run clinical trials are invaluable.  But anyone thinking of participating in one should be wary.  Definitely do some homework before signing up. The business of clinical trials is a competitive, profit-based, multinational industry backed by well-funded lobbyists like ACRO, the Association of Clinical Research Organizations (“providing a strong voice for Clinical Research Organizations in Washington”.) 

    Before participating in any Phase I “human guinea-pig” trial, make sure the terms include coverage of your medical treatment in case the untested drug screws you up in some way.  Most trials will NOT protect you.  Fewer than one in six will guarantee free medical treatment for any injury sustained during the trial.  And those are the very best trials, run by academic research institutions.  The private, for-profit testing sites are far less likely to help.  To the contrary, they require you to sign away all your rights to recompense in case of injury.

    Judy Stone, the Scientific American guest blogger, says “Most drug trials are closely regulated and safe to participate in”.   So why are the testing organizations so eager to duck responsibility in the rare case something goes wrong? If Stone wants “to see more people volunteering for clinical research”, addressing this issue is a good place to start.   Maybe it isn’t “gaps in knowledge that make people wary of participating”.  Maybe it’s common sense…

    At the moment the clinical trial system looks very much like the banking system did before the financial meltdown.  The regulatory agencies aren’t up to the job (the FDA inspects only one clinical trial out of every hundred, and the Office for Human Research Protections has no jurisdiction over private trials).  And ethical oversight is often hopelessly compromised, because the reviewers are paid by the very organizations they oversee.  It’s like the credit ratings agencies (S&P, Moody’s, Fitch) who gave golden assessments of garbage investments because they were rating their own paymasters.

    So participant beware.  Look for trials conducted by universities, not private, for-profit organizations in the self-styled “CRO industry”.  Ask if the FDA has inspected the trial.   And only participate in trials that put their money where their mouth is.  It’s a simple, effective test of legitimacy.  Are you thinking of  joining a trial that is supposedly “really safe”, but refuses outright to help you if they injure you?   Think about it.  Then turn around, and walk away.  Fast. 

    (More info:  New England Journal of Medicine 2006; 354:1871-3.  Also: Department of Health and Human Services, Office of Inspector General. “The Food and Drug Administration’s oversight of clinical trials”)

    1. There’s a little confusion here.  The FDA ultimately requires information on the outcomes of all clinical trials in the submissions process, and that information had better be detailed and all the ducks in a row, or the submission will fail.  It’s true that the FDA doesn’t directly oversee the trials themselves (although they may perform a site inspection) – that is the job of an Ethical Review Board (ERB), which is granted authority by the FDA and OHRP.  It’s also incorrect that the regulations governing ERBs do not apply to private, for-profit organizations – they most certainly do.

      1. Please read my comment more carefully.  I did not say “that the  regulations governing ERBs do not apply to private, for-profit organizations”.  I said that the Office for Human Research Protections (OHRP) has no jurisdiction over private trials.  This is exactly correct.  The OHRP only has jurisdiction over research funded federally by the Department of Health and Human Services.  

        As for what you call an “ERB”, it is actually an IRB, or Institutional Review Board.  An IRB is supposed to “ensure that clinical investigators take appropriate steps to protect the rights and welfare of human subjects.” But they have become terribly compromised as the management of clinical trials is increasingly performed by for-profit corporate “contract research organizations”, or CROs.  

        The problem was already bad in the 1990s:

        “Previous Office of Inspector General (OIG) reports documented weaknesses in the oversight that FDA and IRBs provide for clinical trials. In 1998, a series of reports concluded that IRBs lacked the time and the expertise to sufficiently monitor the research taking place under their jurisdiction.  The reports found that IRBs often conducted minimal review of studies after the initial approval of the research.

        “A 2000 report documented weaknesses in clinical trial oversight. The report found that data integrity concerns, more than human subject protection, drove FDA’s oversight of clinical investigators…”  

        In short, the very regulatory bodies charged with ensuring the welfare of the participants are concerned more about the data than the human beings in the trial.  People < Data.

        The situation is only getting worse, sadly.  I stand by my analogy with the banking system.  And again, anyone considering participating in a trial would do well to examine what medical treatment they will receive should they suffer injury as a result of the trial.  Usually, they end up on their own.  Although some places actually DO look after them if it goes badly.  Washington State, I believe.  And, oh yeah… all of Europe.
         (Quotations are from the 2007 OIG report on FDA oversight of clinical trials, ).

        1. ERB = IRB.  Same thing.

          Now please excuse me.  I have to go and help one of our researchers prepare for a site visit!

  11. Maggie K-B: 

     It looks like there is someone in your own backyard with some expertise on the dodgy aspects of clinical trials and their ongoing privatization.  Carl Elliott, an M.D. and  Professor in the Center for Bioethics at the University of Minnesota,  has written extensively on this subject in magazines like the New Yorker, and in a book as well, White Coat, Black Hat: Adventures on the Dark Side of Medicine.  (I don’t know him, but I’ve read some of his pieces).

    I’m sure Judy Stone is a fine researcher, but as you rightly note in your disclaimer, she has an agenda, and she is someone who makes her income in part from privatized clinical trials.  She even says (in her book) that she doesn’t think the trend towards for-profit contract research organizations is a problem.  This is somewhat at odds with the Inspector General reports cited above.  

    Clearly, we all benefit from those who participate in clinical trials– good clinical trials– but we also need people to be safe.  Dr Stone does a good job presenting the upside of trials.  But it can also be an exploitative business.  An interview with Dr Elliott might balance people’s understanding, and help them avoid  dubious clinical trials… of which  there are plenty.

Comments are closed.