A Doofus' Guide to Clinical Trials: Why Are They So Hard and Why Do They Take So Long?

In a comfy suburban living room, a precocious young lad and his aging nerdy father are munching on Mint Milanos and watching the evening news. For what seems like the dozen-eth time this year, they listen to a breathless newscaster discuss a possible game-changing cancer treatment, only to be followed by the disclaimer: “But experts say it will be years before a possible drug might be available to patients.”

Son: Why do they always do that? Why do they get people all excited only to disappoint them in the end?

Dad: I know, right? It’s like Seinfeld meets Game of Thrones.

Son: Don’t they want to move faster with possible cancer drugs? Or any potential drugs for terrible diseases that don’t have good available treatments? Time is of essence, right?

Dad: Yes, my boy. But proving that a drug works and is safe isn’t like Harry Potter romancing a stone. You can’t wave a wand, yell “Expecto Patron-tequila” and make it all work.

Son: Why can’t you be a more optimistic human?

Dad: Because I started asking girls for dates after I reached puberty. Anyway, it’s not inefficiency, lack of urgency, or Illuminati conspiracy. The drug development process simply takes time.

Son: I wish I could talk to someone who could explain it to me.

Dad: Hello! You’re looking at 25 years of biopharma experience here. I’m not just a pretty face - I’m a bastion of insight and knowledge.

Son: I thought you work in Finance.

Dad: If a skinny boy from Reseda can become karate champion by waxing cars, a numbers guy can absorb knowledge.

Son: Have you ever caught a fly with chopsticks?

Dad: Less talk, more listen. Now, drug development begins with an idea. Scientists might discover that properties in a compound can stop certain disease cells from proliferating. Or they might discover a disease requires a certain condition to survive, then try to engineer a compound that disrupts that condition.

Son: That makes sense. But why does it take so long to prove the compound works?

Dad: It doesn’t take all that long – at least not in a laboratory. By the way, this is called “in vitro”.

Son: Wasn’t that the bad guy in Ghostbusters 2?

Dad: No, blemished fruit of my loins. “In vitro” simply means research testing that’s done in a laboratory with test tubes and culture dishes. Of course, just because something works in a culture dish doesn’t mean it will work in a living things. So the next step in the process is testing in animals – typically mice. The goal is to obtain evidence that the drug is effective and safe in a living organism.

Son: So if it’s effective and safe in mice, there’s a good chance it’ll work in people, right?

Dad: Ah, as Hamlet’s massage therapist might say – there’s the rub. If lucrative markets exist for curing mice, drug companies could clean up! Over the decades, numerous drugs have been found to cure mice of cancers, Alzheimer’s, and many other maladies. Unfortunately, most mousey successes have not translated into the same results in people. Other than Mickey and Jerry, there are vast differences between mice and humans.

Son: Well, if nothing else, the mouse testing would at least determine safety, right?

Dad: Here’s a fun fact: did you know the amount of bacteria needed to kill a mouse is millions of times more than what’s needed to kill humans? This is why rats can live in sewers and nasty dirty places where humans couldn’t survive, your bedroom notwithstanding. The point is, demonstrating a drug is safe in mice is a positive indicator but still just an indicator. Much more testing is needed to demonstrate adequate safety in people.

Son: You once told me that every drug has side effects, and safety is relative. What did you mean by that?

Dad: What I meant was to keep ourselves mentally safe, we should never spend too much time with our relatives because they’re all nuts. But what I also meant was that when it comes to drug safety, “safety” isn’t always crystal clear and depends on many variables such as the target disease, condition severity, patient characteristics, and so on. For instance, let’s suppose a drug has been shown to cause irregular heartbeat, which may be dangerous or even fatal for some. If that drug is intended for severe acne or migraines, it’s arguably not worth the risk. But if the target is pancreatic cancer, then the risk is very acceptable.

Son: But it’s often not as clear-cut as that, right?

Dad: Very true. It’s often much more gray. I’ll give you another example. Multiple sclerosis is a chronic immune system disease that can cause severe nerve damage, even paralysis. Some newer medications have proven very effective in treating some forms of MS. However, it’s also been shown that a small percentage of patients taking these drugs can contract something called PML, or progressive multifocal leukoencephalopathy. Now, this condition is as bad as it sounds: about 30-50% of people contracting PML die within the first few months of diagnosis, and even the survivors can be left with severe neurological damage. However, less than half of 1% of patients taking these drugs will get PML. So are these MS drugs “safe enough”? Not an easy thing to answer, right? What the FDA typically opts for in these instances is to allow the drug to be on the market with the risks spelled out as clearly as possible, usually with something called a “black box” warning label. This allows doctors and patients to decide whether the risks are warranted.

Son: But I still don’t understand why human clinical trials take SOOO long? Given the importance of finding effective life-saving drugs, wouldn’t it be better to speed things up as fast as possible?

Dad: Of course, son. But there’s good reason why the process takes so long, and this is a process that’s been vetted over many years. Back in the old days when no such requirements existed, it was kind of the Wild, Wild West when it came to medicine. Anyone can claim an efficacious treatment based on the flimsiest of evidence or even no evidence at all. You’ve heard of snake oil? Eye of newt? Guava gonads? To be sure, certain treatments were perhaps effective to a degree. But many weren’t effective in the least, and some were downright toxic and dangerous. So government agencies like the US FDA regulate drug marketing and sales. Drugs must meet very stringent requirements before they can be sold and marketed for their therapeutic targets.

Son: But it just seems that every time a promising treatment comes around, it has to go into this big long black hole of clinical trials. And often, the clinical trials fail and the treatment falls by the wayside. Isn’t it possible that a drug is actually good, but it still fails the study? Wouldn’t that mean a lot of potentially great drugs are being tossed aside?

Dad: It’s possible of course, but unlikely. Clinical trials are designed by very smart people with ample experience. Companies typically have lengthy discussions with the FDA to ensure the study designs are adequate for approval. Also, the studies are usually designed with the “passing” bar set as low as possible, because the main goal is to show the drug works, not that it’s the greatest thing since sliced Twinkies. So if the study fails to hit even this lowest possible bar, it’s much more likely the drug simply isn’t effective.

Son: Well, let’s not wait any longer. I want to learn as much as possible, so please teach me, oh exalted experienced and enlightened enema…I mean, enigma of the pharma world. How does the clinical trials process work?

Dad: Well, there are typically three phases to human clinical trials. The first phase – ingeniously called “Phase 1” – focuses on safety of the drug. Enrolling a relative small number of subjects, usually about 30-50, Phase 1 trials try to determine if the drug is safe, at what doses do they become less safe, possible side effects, and so on. Usually at this stage, the drug isn’t tested for effectiveness. As such, even healthy volunteers like you and me can participate if we meet the study requirements.

Son: Would you be a volunteer?

Dad: Son, with the shape I’m in, you could donate my body to science fiction. Ha ha ha – sorry. That’s an old Rodney Dangerfield joke.

Son: Who’s Rodney Dangerfield?

Dad: Seriously? He was a world class famous comedian - one of Johnny Carson’s favorites!

Son: Who’s Johnny Carson?

Dad: Never mind. So once safety is confirmed, the drug can move on to the second phase. Any guesses what this phase is called?

Son: Phase 2?

Dad: Check out the brain in your father’s son! Yes, exactly. Phase 2 trials enroll up to several hundred patients diagnosed with the target disease. The goal here is to see whether the drug might be effective in that disease. Now typically, there aren’t enough patients in Phase 2 to statistically prove the drug is effective; the goal is to find enough evidence that the drug MIGHT work and thus can move to Phase 3. Phase 3 trials can enroll thousands of patients – enough to statistically show that the drug is therapeutically effective.

Son: But why waste time doing Phase 2? Wouldn’t it be cheaper and faster to just jump right into Phase 3?

Dad: That’s a reasonable question, and there are valid reasons. First, Phase 3 trials are a huge investment in time (several years minimum), money (tens or even hundreds of millions of dollars), and effort by not only the sponsoring company but thousands of patients, doctors, nurses, and other medical and scientific people. So it’s very important to try to get things as “right” as possible before jumping into a Phase 3 trial. A properly designed Phase 2 study will not only show signs of efficacy, it will also provide important characteristics associated with that efficacy. Does the drug work better or worse in people with certain genetic traits or physical characteristics? Are certain dosing regimens better than others? Are there additional safety issues associated with longer use? All of this valuable insight makes doing a Phase 2 a very worthwhile commitment.

Son: I guess I understand. But if there was enough solid early evidence of efficacy, you’d think there would be some wiggle room to speed things up and not have to adhere to the normal timelines?

Dad: Indeed there is! The FDA has approved drugs based only on promising Phase 2 data. They are more likely to do that when the target disease is severe with no viable treatment alternatives, of course. Now, the company may still be subsequently required to complete Phase 3 studies just to prove things out, but this accelerates the drug to patients in need. Also, some companies may design a trial as a “Phase 2/3” study, which means the trial starts off as a Phase 2 but then immediately shifts to a Phase 3 if and when certain conditions are met. In other words, although Phase 1-2-3 is the standard process, there are exceptions that can speed up the timeline.

Son: Did you say Phase 3 trials can take up to several years? Why does it take so long? I thought most treatment regimens for each study patient are less than a year.

Dad: First of all, the study has to be designed properly based on analyzing all previous data. Then investigator sites (these are the doctors’ offices that participate and administer drug to the study patients) have to go through a qualifying process and be trained and contracts drawn up. IRB’s, or Institutional Review Boards, which are independent ethics committees that provide regulatory and ethical oversight on behalf of study patients, have to be arranged. But perhaps the biggest factor affecting a study’s total duration is how long it takes to actually find and enroll the total number of required study patients. Remember that not all patients of the target disease will qualify for the study. Every trial has very specific requirements that a patient must meet in terms of characteristics and traits. This is called a study’s “inclusion criteria.” Also, not all patients who qualify will want to participate in the trial. Some opt for different studies. Some don’t want to participate in clinical trials at all.

Son: What you’re saying then is that for most studies, it’s not like patients are standing in line clamoring to participate. They’re interviewing for patients almost like job candidates.

Dad: That’s right. So to enroll for, say, a 1,000-patient study, companies often have to sign up hundreds of investigator sites in the hopes that each site can enroll several viable patients. Well, finding that many sites – often across multiple countries and continents! – takes an enormous amount of time, as does screening for qualified patients.

Son: When you explain it that way, it does seem like a really huge, long, and drawn-out effort.

Dad: Yes, indeed. Let’s say a clinical study requires 1,000 patients and 100 investigator sites worldwide. It might take up to a year or more to find 100 good sites, and that each sites takes up to a year to find 10 qualified patients. Furthermore, suppose the study regimen requires each patient to participate for a 12-month period. Even in this arguably optimistic scenario, between enrolling the first patient at the first site to the last patient completing the 12-month regimen, that’s already 3 years! And that’s not considering the study’s set-up and close-out times, or the fact that many Phase 3 studies enroll several thousand patients.

Son: Wow - that does sound like a daunting effort. So they have to wait until each and every patient finishes the regimen?

Dad: Generally, yes. However, some studies may build in what’s called an “interim analysis”. That’s when experts are allowed to peek at the data in the middle of the study to see whether there are any pre-specified signs that the drug is effective. If that looks to be the case, it’s possible the FDA may allow the trial to stop early and approve the drug as quickly as possible. But that is a fairly rare occurrence. Most of the time after an interim peek, the experts simply say, “carry on” and the trial continues on as planned.

Son: What if they look at the data and see that the drug isn’t doing any good at all?

Dad: Very good question – and indeed a clinical trial can also be stopped after the interim peek. A “futility analysis” can be done during the peek and the experts may determine it would be “futile” for the study to continue and the trial immediately ends.

Son: So can anyone from the company look at the interim data?

Dad: Absolutely not. Only a predetermined very limited number of people can do the interim analysis. They’re also not allowed to discuss any details of what they see with anyone else. Son: How come?

Dad: Because that could result in bias. Bias is bad jujus when it comes to clinical trials, and unfortunately people are predisposed to being biased, including me.

Son: I thought you were only attracted to women. Dad: Yes, except at times when I hear Julio Iglesias sing “Abrazame”…..wait, what? No, I said BIASED. It means a favorable or unfavorable inclination towards something. Clinical trials must eliminate all potential bias, or the study results could be compromised.

Son: How do they do that? Dad: Well, the best way to ensure non-bias is that Phase 3 trials are typically double-blind, placebo-controlled. Son: The patients have to be blind?

Dad: In a way. They are “blind” to whether they’ve been given the actual drug. You see, in a blinded trial patients are divided into different groups and given either actual study drug or a placebo, which looks just like the study drug but has no active ingredient. Patients don’t know which group they were in until the study ends. That way the study can compare the two under a controlled setting and determine whether there’s a real difference.

Son: But why do they have to use a placebo? Can’t they simply see whether the patient on the drug shows improvement?

Dad: Son, do you remember when Aunt Millie was convinced that eating honey-dipped dandelions would cure her migraines?

Son: Yes, she had fewer migraines and got diabetes.

Dad: Right. There is no scientific basis that honey-coated dandelions does anything for migraines. But because she BELIEVED it, she actually had – or felt she had - fewer migraines. This, my son, is called the “placebo effect”. The brain is very powerful and sometimes if a person believes or expects some physical effect, it actually happens to some degree. Clinical trials must eliminate this confirmation/expectation bias – hence the requirement for placebo control.

Son: But what does “double blind” mean?

Dad: In a well-controlled study, not only are the patients “blind” to whether they’re getting study drug, the doctors administering the drug are too. This way, doctors can’t knowingly or unknowingly treat and observe patients differently depending on whether they’re getting study drug.

Son: Seems like they go to great lengths to prevent bias.

 Dad: Yes they do. Quite often you will hear a phase 3 trial described as “randomized”. That simply means that the process of choosing patients for either the study drug group or placebo group is based entirely by chance. Companies use special systems to ensure a random selection process.

Son: Why is randomization important?

Dad: Well, if the company or investigator can choose whether a patient is put on study drug, they might consciously or subconsciously put the “best” patients – the ones that seem the healthiest or have the best chance at getting better - on the study drug, and the “longer shots” on placebo. Randomization removes this potential bias.

Son: Okay, but knowing it’s very possible they’ll get a placebo, doesn’t that make patients reluctant to participate?

Dad: It definitely could. That’s why companies are required to clearly spell out all the risks and details to every patient, including the fact that they might get placebo. However, study participation often offers the benefit of allowing patients to receive either the study drug or the best available FDA-approved “standard of care” drug when the their participation in the study is over, regardless of whether they were recipients of study drug or placebo. Studies are also designed in such a way that patients on placebo don’t incur greater health risk.

Son: Well, hopefully the study drug proves fantastic so all the study patients can choose that, right?

Dad: That is the hope. However, keep in mind that when it comes to clinical trials, it is very rare that results are spectacular, as in “all the patients who were on the drug were cured, and all placebo patients were not!” Even when the drug proves effective, the observed improvements are typically modest. Good study results are usually more along the lines of, “Patients given study drug improved 35% from baseline on average, while placebo patients improved 15%”. But even these relatively modest improvements are statistically significant and very important. This is also why it’s important to complete the study in its entirety in most cases. Stopping a study too early may result a truly effective drug not showing statistical benefit.

Son: Why aren’t the differences bigger between effective drugs versus placebos? If a drug actually works, wouldn’t you expect a dramatic difference when compared to something that’s no more than a sugar pill?

Dad: First, remember that we humans all have significant differences to a degree. Just as some of us are taller or can run faster or like broccoli more, that’s also true when it comes to how a given drug might affect us. Some patients may have experienced considerable, maybe even spectacular improvements, while some may have little to none. Even the best, most successful drugs on the market don’t work on all patients, and in some cases not even on most patients. But even if a drug only works on relatively small percentage of patients, it can be hugely successful and beneficial.

Son: Okay, I think I get it. But when I see those news stories, they often involve a drug that’s already approved for a different disease. In cases like that, can’t the long clinical trials process be avoided or at least shortened considerably?

Dad: Well, you must realize that just because a drug works in one disease, it may not be effective in a different one. That’s true even if the respective disease have some similarities. Michael Jordan might be the greatest basketball player ever, but that didn’t mean he was a baseball star. So efficacy for a different disease must be proven through clinical trials.

Son: But at least from a safety standpoint, the drug is proven, right? I mean, not only has it gone through previous clinical trials safety testing, it has also been administered to non-study patients for some time.

Dad: Alas, safety can’t be assumed between different disease groups. For instance, are there characteristic differences between the respective patients? What if the patients in the new disease group are older or have more compromised immune systems? Additionally, are there differences in required dosage? If the new disease patients requires higher dosages, are there associated toxicities or side effects? Those are all important questions that need answers. But to your point – yes, if there are no differences in patient characteristic and if required dosages are the same or lower, then there’s much more confidence in drug safety. And in fact, this can be taken into account in terms of clinical trial requirements. To test an approved drug for a different indication, companies can often skip the Phase 1 safety trial entirely and go immediately to Phase 2 or 3.

Son: But in some of the news stories, doctors were giving the drug to patients in the new disease group and the doctors themselves were confirming that patients got better! Doesn’t that prove to some extent the drug works? After all, these are professional doctors. In instances like this, couldn’t all the clinical trials be skipped?

Dad: Son, those are examples of what’s called “anecdotal evidence.” That is, based only on observations.

Son: Right! Patient gets the drug, the doctor observes the patient getting better…there you go!

Dad: Alas, not quite. Yes, the source of the anecdote is important, and what an experienced doctor says will carry more weight than what your Uber driver or mechanic says. But even if the smartest doctors make the anecdotal observations, they still needs to be proven out in a controlled clinical trial setting.

Son: But why?

Dad: Because observations in and of themselves can be very unreliable. If you eat irradiated peanut M&M’s and your acne clears up, you might logically conclude irradiated peanut M&M’s can treat acne. But how do you know for sure? Did you change anything else over that period? Did you happen to stop eating greasy burgers from a particular diner? Did you stop taking a certain allergy medication recently? Did you start swimming 5 miles a day? Did the acne simply clear up by itself – as acne sometimes does? And even if it in fact did work on you, to what extent would it work on others, if at all? Could you have been an anomaly?

Son: But in the cases I’m referring to, it was a bunch of doctors from different places and the results were seen in many patients.

Dad: That definitely makes the case much more compelling, but those aforementioned questions along must still be answered. Also, did those doctors give the drug to other patients who did not respond? If so, what percentage were non-responders? Were there any common characteristics shared by the patients who got better? Perhaps certain genetic or physical traits, fitness condition, or age group? How frequently did the various doctors administer the drug and what dosages were used? Did any patients have serious side effects? And if so, were they actually caused by the drug or by something totally unrelated? Were the severity of side effects dose-dependent?

Son: It’s amazing how you can talk so much without taking a breath.

Dad: I recently started swimming 5 miles a day. The point is, those are very important questions, and unfortunately anecdotal evidence, no matter how compelling, doesn’t provide adequate answers.

Son: So the anecdotal evidence is useless and meaningless?

Dad: Of course not. Anecdotal evidence can be very meaningful and immensely important. In fact, many of history’s most important medicines started out with medical and scientific people making anecdotal observations! All I’m saying is that anecdotal evidence must be followed up with scientifically designed and well-controlled clinical trials in order to demonstrate a drug is truly effective and safe. Even the doctors who made the anecdotal observations in those news stories you saw would say the same thing.

Son: But there are patients who could be dying and can’t wait for a drug to go through the whole study process. They have no other alternatives, and they have nothing to lose. Isn’t there some way to get the drug to them early?

Dad: Yes – there absolutely is! If the drug is already on the market for a different disease, these patients can get the drug right away because their doctors are already allowed to prescribe them those drugs.

Son: That’s already allowed? They don’t need special permission from the FDA?

Dad: Not from the FDA, or CIA, or UPS. LOL SMH. No, because doctors are already allowed to prescribe medicines “off-label” if they see fit. “Off-label is when a medicine is prescribed for a condition other than what it is specifically approved for. In fact, doctors do this all the time. About 20% of all drugs prescribed in the US are prescribed “off-label”.

Son: I didn’t realize that. I’m surprised because I’ve always heard the FDA is so strict when it comes to drugs.

Dad: The FDA regulates drug approvals, but it does not regulate drug prescribing. Doctors are given a lot of leeway, and rightfully so. Other than certain controlled substances like opioids, doctors can prescribe what they feel is best for their patients.

Son: Hey dad, what do you call a doctor who really enjoys going to the gas station?

Dad: I give up.

Son: Doctor Fill. Anyways, that’s great about the “off label” prescribing, but what about experimental drugs that aren’t yet on the market? Doctors can’t prescribe those because they haven’t been approved yet, right?

Dad: Exactly right. However, the FDA does have a program that can potentially get even unapproved experimental drugs to patients who have serious or life-threatening conditions with no other treatment options. This is called “compassionate use” or an “expanded access” program. This program does require FDA review and permission, and not all investigational drugs will be eligible. But it is an existing path to get experimental drugs to patients.

Son: Dad, I must say I feel so much more enlightened now. Who would have thought?

Dad: That’s right. Now when you hear a newscast breathlessly declare a potential breakthrough treatment for a terrible disease, you’ll know how and why it may take years to prove that it actually works and be made available to the world. When you hear reports from even the most reputable sources that an existing drug may work on other conditions, you’ll know why clinical trials are still necessary to confirm. And when you hear someone talk about a “randomized, double-blind, placebo-controlled Phase 3 study”, you’ll know exactly what that all means.

Son: I meant that a Finance person can convey useful information, despite what mom says.