By Kevin She
6 May 2016
Clinical trials are the modern standards of proof for the effectiveness of new ways to medically prevent, detect, treat, or manage diseases and other medical conditions. However, not all clinical trials are equal. When evaluating clinical trials, a number of considerations must be made as to the design and execution of the trial. Homogeneity of the patient population examined is also of immense importance. For example, neurological conditions such as schizophrenia, autism, or depression are notoriously difficult to formulate strong clinical trials for since many different underlying conditions present with similar symptoms, and by current definition in the DSM-V these disorders are diagnosed purely on the presentation of symptoms; a treatment may be effective against one underlying condition but not others. After a clinical trial has been concluded care must be taken to interpret the results. Criteria such as effect size and “number needed to treat” – how many patient must be treated for one to benefit – need to be considered against how many patients were examined; the smaller the effect the more patients are needed to be examined to strongly demonstrate the effectiveness of a treatment. Proper evaluations of clinical trials are complex and filled with nuance.
This is not that article.
Rather, let’s take a look at a brief history of clinical trials!
Fig. 1 – King Nebuchadnezzar II was a king of the Neo-Bablyonian Empire
The root of a clinical trial is comparing two different conditions (drugs or other treatments) and determining which condition was more effective. Arguably, the first “clinical trial” ever performed was recorded in the “Book of Daniel” in the Christian Bible. King Nebuchadnezzar II was a king of the Neo-Bablyonian Empire who ruled until 562 BCE and was a renowned warrior who conquered Jerusalem and was a wise ruler who commissioned the Hanging Gardens of Babylon. At some point during his reign, wanting his people to be healthy and strong, he ordered them to eat only meat and drink only wine.
However, a group of young nobles objected, being mostly vegetarian and preferred to eat vegetables and beans. Nebuchadnezzar, being a wise ruler, conceded and allowed these young men their dietary preference – but only for ten days in case they weakened and died. At the end of the ten days the vegetarians appeared better nourished than those who were, by his decree, strictly carnivorous and so Nebuchadnezzar permitted his people to eat vegetables and beans as they pleased.
While not as exacting as a modern multi-phase parallel-arm placebo-controlled double-blinded trial, this modest experiment may have been the first recorded medical test that guided public health policy.
Fig. 2 – Avicenna was a Persian polymath who is regarded as one of the most significant thinkers and writers of the Islamic Golden Age.
Avicenna, a famous polymath who lived during the Islamic Golden Age (roughly from the 8th to the 13th century), described some rules of testing medical treatments in his ‘Canon of Medicine’ which was used as a standard medical text worldwide as late as the 17th century. Some of these ideas included only evaluating the effectiveness of an experimental treatment only in examples of disease that was uncomplicated and to see if the treatment results were reproducible – which the first idea helps with by limiting uncontrolled variables. However, there are no records of whether these ideas were actually applied in practice.
Fig. 3 – Ambroise Paré was a French barber surgeon who served in that role for kings Henry II, Francis II, Charles IX and Henry III
Perhaps one of the first nearly-modern clinical trials was conducted by accident. Ambroise Paré was a French barber surgeon who lived in the 16th century and is considered one of the fathers of modern surgery, forensic pathology, and a pioneer in modern battlefield medicine. The standard treatment at the time for gunshot wounds was cauterization with boiling oil. After one especially bloody engagement, Paré was treating wounded soldiers and ran out of cauterizing oil. With wounded soldiers still awaiting emergency treatment, Paré resorted to an old Roman remedy that he had previously read about involving a paste made from eggyolks, oil of roses, and turpentine, which he applied to gunshot wounds.
A conscientious man, Paré spent the night sleepless and arose at first light to tend to the wounded soldiers who had been treated with the old Roman remedy, expecting these soldiers to be dead from ‘poison,’ which was the description of microbial infection and sepsis in those times. To his surprise, these soldiers were in better condition than those treated with cauterizing oil; there was less pain, less swelling and inflammation (infection), and that those soldiers had slept through the night. He vowed never again to use cauterization to treat gunshot wounds. Indeed, he subsequently reintroduced arterial ligature (known during the time of Galen, around the 2nd century BCE, but had subsequently been abandoned) and developed the Bec de Corbeau (crow’s beak) ligature, a direct predecessor to modern techniques using haemostats.
It would be another 200 years before a planned and controlled clinical trial was recorded, though.
Fig. 4 – James Lind was a Scottish physician. He was a pioneer of naval hygiene in the Royal Navy. By conducting the first ever clinical trial, he developed the theory that citrus fruits cured scurvy.
During the 15th century ocean-going technology were sufficiently advanced in Western Europe that explorers were able to travel to ever further places on earth and for longer periods away from land. Vasco da Gama, a Portuguese explorer, became the first European to reach India by sea in 1498. By the time he successfully returned home, two thirds of the crew had died from scurvy.
The 17th century saw the rapid expansion and development of naval warfare in Western Europe and ships stayed in the ocean far away from fresh food for increasing lengths of time. During this era, scurvy caused more losses to the British Navy than were suffered in all engagements of enemy action. Every captain dreaded the terrible scourge of scurvy and every expedition suffered from it.
At first, affected sailors would feel woozy and sluggish, which progressed to achey legs, followed by swelling and bleeding of the gums, the development of lesions on the legs (which typically got infected and led to gangrene), increasingly elevated fevers, and finally death from haemorrhaging of the brain. It was a rather unpleasant experience.
No one at the time knew what caused scurvy, much less how to treat it. Popular opinions at the time was that scurvy was caused by the damp, dark, and cold conditions aboardship or that it was caused by terrible shipboard food putrefying in the stomach.
During the 17th century, universities in England had devolved into little more than prep schools where the offspring of the nobility learned the noble arts and networked with one another. Meanwhile the universities in Scotland followed the traditions developing at universities on continental Europe such as rationalism and humanism. James Lind was a fellow of the Royal College of Surgeons of Edinburgh, wrote his medical degree thesis on venereal diseases (sexually transmitted infections), and earned his medical degree from the University of Edinburgh Medical School and in the early part of his career worked as a ship’s surgeon in the British navy aboard the HMS Salisbury.
After two months at sea, many of the sailors on his ship developed scurvy; Lind saw this as an opportunity to test his theories on the cause of scurvy. Believing that it was caused by putrefaction of food in the stomach – which he thought could be cured by ingesting acid – he took 12 scurvy-ridden sailors and divided them into six groups of two. To each group he assigned a different daily dietary supplement for two weeks: 1) one quart of hard cider, 2) 25 drops of sulfuric acid, 3) spice paste and a drink of barley water, 4) two oranges and one lemon, 5) half a pint of seawater, or 6) six spoonfuls of vinegar.
While the group that drank cider showed some modest improvement, the sailors who were given citrus fruits were almost fully recovered. Upon treating the other groups with citrus fruit, they too recovered from scurvy.
James Lind promptly wrote up “Treatise on Scurvy” – and was roundly ignored by the admiralty, due mainly to the expense of supplying citrus fruits to every seafarer in the navy.
It was nearly 50 years before the British Navy made lemon juice a compulsory part of a sailor’s diet, which was quickly replaced with lime juice since it cost less (but was less effective than lemons, having less vitamin C per unit of fruit).
What about the Chinese exploratory fleets of 1421 led by Zheng He? Apparently live chickens for fresh eggs and meat and, importantly, freshly sprouted bean sprouts helped – a cup of bean sprouts has about 13.7mg of vitamin C along with plenty of vitamins A, B-complex, E, and K.
Fig. 5 – Modern clinical trials have evolved into sophisticated research programs utilizing placebos and new treatments.
Almost a century passed before another important milestone in clinical trials was developed, the placebo. Placebos started appearing in the medical literature in the early 1800s and were given to placate patients as opposed to be intended as a real treatment – the word placebo derives from Latin for “I will please.”
Originally given as a placative, it made its way into clinical trials once researchers started noticing that any treatment – even if it had no real effect – was still somewhat more effective than no treatment at all. Thus was born the idea of “control groups.” Today, especially with serious illnesses like cancer, a well accepted “standard treatment” – available for the vast majority of conditions – is used to compare new treatments against instead of an inert placebo, as it would be a grievous breach of ethics not to provide real treatment if it is available. Indeed, in clinical trials for new treatments of serious diseases, the data is periodically analyzed while trials are still running; there are a few examples of cancer drugs that were so much more successful than the standard treatment at the time that the trials were ended early so the control patients would be able to receive the new drug.
However, limits remain as to the extent that placebos are effective in blinding the patient to the treatment; for example, the historical trials for the antibiotic rifampicin were notable in that it causes its taker’s urine to take on a red colouration, which the placebo did not cause. Many drugs, like Cannabis, have psychoactive effects that are readily detectable by the patient and therefore cannot be effectively placebo controlled.
Fig. 6 – Effective clinical trials need to be unbiased.
With placebos the patient being treated is blind to whether they were treated (or which treatment they received) there remains the issue of bias in the researchers themselves. The 19th century saw an explosion of medical advances – especially in the field of small molecule drugs – which pitted legitimate doctors against hucksters and fraudsters, it was imperative to demonstrate medical outcomes uncorrupted by poor judgment, over-enthusiasm, wishful thinking, and outright fraud.
Blinding, also referred to as masking, is a method to decrease bias in any clinical trial. In single-blind studies, the researchers collecting and assessing the data do not know which group received the experimental treatment until the final results were unblinded. In a double-blind trial the participants are also unaware of which treatment they received.
An additional refinement is randomization; the researchers themselves do not assign which treatment a patient may receive. Typically, patients are assigned treatment by a third party who has no knowledge of the particulars of each patient.
The first double-blind randomized control trial was likely the 1943 Patulin Trial at the height of WWII; the Medical Reseaerch Council (MRC) UK wanted to determine if patulin, an antibiotic extracted from Penicillium patulium, was effective for treating the common cold. As they discovered and as we all now know, antibiotics are ineffective for treating the common cold which is caused by viruses (but are sometimes useful if a secondary bacterial infection establishes itself in a host compromised by a viral infection).
The first widely publicized successful double-blind randomized control trial was for the antibiotic streptomycin for treating pulmonary tuberculosis in 1946, also by the MRC UK. The trial was conducted meticulously; not only were patients truly randomly assigned by a third party coordinator, but the study also relied on objective measures of treatment effectiveness using methods such as chest x-rays that were interpreted by experts who were also unaware of which treatment the patient had received.
Unlike the common cold, tuberculosis is caused by Mycobacteria tuberculosis which is susceptible to some antibiotics.
This trial and another testing the efficacy of immunization against whooping cough created a watershed moment and these models of modern clinical trials have become the standard proof for or against the efficacy of new ways to prevent, detect, treat, or manage diseases and other medical conditions. Since then much more work has been done to improve the practices of conducting clinical trials from stratifying trials into different phases to improve patient safety to ethical considerations on whether, how, and to whom new experimental procedures can be provided.
Types of Clinical Trials
Treatment: tests new drugs, therapies, devices, or surgeries
Prevention: tests new means to prevent disease: medicines, vaccines, lifestyle changes, &c.
Diagnostic: tests new ways to diagnose diseases or conditions
Quality of life: tests new ways of improving the quality of life for people with chronic illnesses
Clinical Trial Phases
Phase I: initial testing of a new drug or treatment on a small group of human subject to evaluate safety, dosage ranges, and identify potential side effects.
Phase II: the therapy is tested on a larger group to determine efficacy.
Phase III: randomized controlled multicentre trials on even larger patient groups to confirm effectiveness.
Phase IV: post-market studies gathering data on whether the drug affects different groups differently, or whether there are any side effects associated with long-term use.