Understanding Clinical Trials: Methodology, Challenges, and Implications for the Future of Regenerative Medicine

Introduction:

In 1987, Doctor Benjamin Freedman proposed the idea of “clinical equipoise,” a concept inextricably linked to the concept of a clinical trial. Put simply, clinical equipoise demands two conditions to be true before a clinical trial (involving humans) can be conducted:

1. There must be a genuine professional disagreement about what treatment is most effective in targeting a specific condition

2. The clinical trial must be designed in a way such that the results of the trial will definitively resolve the above disagreements

As expected, concern for the ethics of studying humans has been, and continues to be, central to clinical trials. The experiments of Nazi Germany, and the notorious Unit 731, remain as poignant historical examples of why human autonomy must never be overlooked for the sake of research. Meanwhile, the Tuskegee Syphilis study, and, more recently, He Jiankui's experimentation on babies, have reminded the public of the importance of informed consent in experimental research. Ultimately, the goal of a clinical trial is not just to determine if a treatment is effective in treating a disease, but if it is more effective than the current standard of care. Further, this goal of the trial must not be prioritized over ensuring that the patients remain safe, informed, and consensual throughout the whole process.

This article will outline the justification for clinical trials as they are currently designed today, give an overview of the basic structure of clinical trials, and discuss some concerns and recent news regarding the ethics and efficacy of clinical trials. 

From Patent Medicine to Informed Medicine:

While the idea of a trial in which one compares a treated group to a group receiving no treatment dates back to the days before the Bible, clinical trials as we know them date back to only the mid-20th century. Before the 20th century, most remedies marketed as panaceas or “cure-alls” were merely nothing more than sugar pills or unsubstantiated concoctions that used marketing tactics to trick vulnerable and often superstitious buyers. It wasn’t until 1906 that the U.S. government instituted the Pure Food and Drug Act, which required companies to disclose the actual contents of the drugs they were marketing. It wasn’t until 1938 that the government instituted the Food, Drug, and Cosmetic act that required drugs to be tested for safety before marketing.

Even still, it wasn’t until a 1962 amendment to the FDC act was introduced that companies were required to test their products for both efficacy and safety, before marketing them to the public. While they had been done before, it was only around this time that the “gold standard” of clinical trials - a randomized, double-blinded approach - began to become more common. Restrictions on testing, more thorough understanding of statistics, and worldwide communication have steadily grown ever since, leading to the robust form of trials we are used to today.

Structuring of a Clinical Trial:

When we think of a clinical trial, what comes to mind most often is the testing of a drug, such as a cancer therapy. But what does this actually mean? In most cases a drug will target a protein (for example, a mutated enzyme), blocking its effects, in order to restore a normal phenotype. While clinical trials themselves can take upwards of 6 years to complete, there are several steps that must be completed before the first human ever interacts with a drug.

First, a drug must be discovered. This often involves large-scale screening of thousands of different drugs on a protein target. Enzymatic assays are used to determine which of the thousands of synthetically created drugs creates the desired result. For example, when designing a treatment to target cancer caused by a hyperactive enzyme, assays may be used to figure out which drugs reduce activity of the hyperactive enzyme the most, without interfering with the activity of the normal (wild-type) enzyme. 

Next, this drug must be tested in actual cells, which is known as an in vitro experiment. This might involve placing cells on a dish, adding the drug, and observing the results, such as cell growth or death. In the case of cancer, these studies might investigate how well the cells are able to grow, divide, or migrate. If a drug delays the growth of these cells, this could be an indication it might be effective in fighting cancer, and that drug could be selected for further studies. 

Before being tested in humans, drugs are often tested in animals under conditions similar to those of a trial. Consider an animal trial similar to what researcher’s would do in a human trial to measure efficacy, barring ethical concerns. For example, researchers can test different (potentially toxic) levels of a drug, follow the progression of the animals until death, or test against true controls (rather than standard of care, as is required for humans). An important reason to test drugs in animal models and cells is that drugs might not behave the same way in the body as they will on a dish. In a living model, drugs must enter circulation, and may get metabolized and converted to other products. Thus, a non-human living model can be used as an indicator of how the drug will affect a fully functioning organism. 

Phase I: In phase I trials, a drug or other treatment begins to become tested in humans for teh first time. A phase I clinical trial mostly tests for safety, and at what doses volunteers can tolerate the drug. Notably, at this stage, researchers are not looking for efficacy, so usually no placebos are used. Instead, researchers monitor for immediate side effects, changes to the body on the drug, and how the body metabolizes the drug. Because of this, only a few volunteers (<12) are typically recruited for Phase I studies and less rigorous statistical baselines are used.

Phase II: During Phase II of a clinical trial, researchers build on the results from Phase I, looking for effective doses of treatment that do result in response to the disease. Phase II trials usually recruit more volunteers (~25), and assign them various doses of the drug (most often, all patients will take the same dose. The primary goal is to look for response to disease (for example, in the case of cancer, stopping of tumor progression), but placebo groups are typically not used. Additionally, because researchers are studying more volunteers, for longer periods of time, rarer side effects of complications can be observed.

Phase III: The third phase of a clinical trial is often seen as the most robust. In Phase III, researchers build on the results of Phases I/II, but are specifically analyzing whether the target drug is better than what is currently available. Phase III trials typically employ a double-blinded setup. This means patients will either receive the experimental drug or the currently available treatment but will not know which - they are “blinded.” Similarly, the doctors administering the drug also don’t know which treatment they are administering - hence “double-blinded.” 

Notably, Phase III trials do not compare treatment with an experimental drug to no treatment at all (a “true” negative control). This follows the principle of clinical equipoise - guaranteeing that every patient is receiving care that is genuinely thought to improve their condition. It would be unethical to conduct a trial that guarantees some patients will die, even if it might produce statistically stronger results. 

While Phase I and Phase II trials are usually performed in major cancer centers, Phase III trials recruit and are performed worldwide. They recruit far larger numbers of participants than earlier trials, and follow patients for much longer, to observe for long-term side effects and complications. Following the results of a Phase III trial, a company may submit the drug to the FDA for approval. After reviewing the data, and comparing the effectiveness of the new drug to what is currently being performed, the FDA may approve the drug for the illness. Approval can be on a first-line basis (the drug will be the first treatment the patient receives), or as an alternative, if the patient relapses on other therapies. 

Challenges in Clinical Trials:

There are several advantages and disadvantages that result from testing drugs in humans, as opposed to model animals such as mice or primates. Obviously, the end-goal of drug tests is for human use, so the final test for a drug must always be in humans, as no animal model will be 100% accurate to human physiology. However, human models present unique challenges. A primary difference between human and animal models is the ability to control for external factors. Many animal studies, especially in mice, use carefully controlled conditions and inbred strains, to keep as many variables constant as possible. These studies will follow these animals even as their conditions worsen, allowing for more freedom and stranger statistics. Humans are not genetically identical, may drop out for health reasons, or may choose not to continue with trials for personal reasons. Because of this, many challenges arise in creating a robust trial with applicable results:

Volunteer Bias

One obvious facet of clinical trials is that patients must volunteer to be a part of them - no one can be forced to participate. The result of this is that trials are subject to bias due to the fact that the subset of individuals that volunteer may not be representative of the population at a whole. This may be particularly true for mental health studies, or studies involving treatments that require adherence to a strict regimen. Volunteers are more willing than non-volunteers, so these kinds of studies may overstate the effectiveness of treatment. 

Inclusion/exclusion criteria

A profound challenge to clinical trials results from the dichotomy of trying to recruit similar patients, to control for confounding variables, but enough patients to confer statistical significance. In clinical trials, patients are not required to stay until the end of the study, so researchers must often over-recruit to account for dropouts. This presents several issues. Firstly, this conflicts with the desire of researchers to recruit specific individuals (i.e., diseased individuals of similar age, ethnicity, etc). Often researchers must weigh raw numbers of individuals with population heterogeneity, both of which can significantly affect statistical power. Similarly, researchers, when studying diseases such as cancer, often look for individuals who are not already on therapies which may confound the results; this is challenging to find. Thus, the amount of eligible participants often severely limits the viability of the trial.

Exclusion of “True Controls”

As discussed above, there are ethical concerns with using placebos, or “true controls” (individuals who are receiving no treatment at all) in clinical trials, especially if there is a proven therapy available. However, the use of placebos is important for statistical significance in clinical trials, and for providing insight into the treatment’s efficacy and side effects. Even in rare cases where true placebo groups may be used, it is never permissible for an individual to be on a placebo without their knowledge. This necessarily removes the blinded element of a trial, often weakening its power. In clinical trials, researchers must balance ethical considerations with statistical power. 

Efficacy vs Applicability

Another major complication with clinical trials is applicability. In the U.S., most drug development is funded by pharmaceutical companies and venture capitalists, whose primary goal is ultimately to generate profit. Thus, an issue with many clinical trials is that it’s not just enough for a drug to be beneficial, or even better than what’s available. Often, for a drug to fully enter development, it must be profitable for a company. This means it must have other aspects that confer advantage over what is readily available, including route of administration (the general public vastly prefers oral tablets over injections/infusions, for example), must be able to treat a wide variety of individuals (companies are often disincentivize to develop drugs for rare indications), and be acceptable as a first-line therapy. These are complications unrelated to the effectiveness of the drug that may impact clinical trials.

Conclusion:

Ultimately, clinical trials are widely accepted due to their robust nature. However, several issues are still prevalent, especially for countries where a majority of healthcare is largely privatized. While massive advancements in drug development have been realized from the 19th and 20th centuries, further progress must be made to optimize clinical trials, particularly in regards to speed and ethics. Recently, there has been a call for a shift in clinical trials to more broadly sample for the heterogeneous nature of the human population, and to tailor clinical trials to move in the direction of personalized healthcare. 

Much of medicine, particularly the area of regenerative medicine, is moving towards methods of personalized healthcare in which diseases are tackled at the individual level. This of course raises new issues regarding planning, surveillance, and data analysis of clinical trials. Additionally, uncertainty around the current access to funding for research, and a shift toward the use of AI in research, might further complicate the field of clinical trials. Particularly, with a call for speedier trials that might help launch new therapies faster, it’s important not to forego the robust procedures of clinical trials that helped develop medicine beyond the sham-concoctions of the 1900’s, to the innovations we see today.