Clinical trials rely on the cooperation of individuals who, due to extenuating circumstances, seek out novel treatments to their conditions. These people willingly cross the line between patient and subjects to serve the development of new biomedical techniques. Although these trials can certainly result in health benefits for test subjects, such benefits are secondary to the bigger goal of developing a new drug or procedure. Experiments at this stage are testing not only the efficacy of a given treatment, but also the presence of unintended effects which might have been overlooked during in vitro and in vivo studies. However, while exploring and expanding the impact of a treatment in non-human models, the translation of any discovered effects are far too frequently impossible or even blatantly misrepresented. The translatability of research is further complicated by difficulties in accurately representing the prior work when applying its results to clinical studies with humans.
A large emphasis is placed on validating theoretical results by applying medication to animal models, but recent studies have revealed that the real applications of such proofs of concept are, in fact, limited (1). This discrepancy results from shortcomings in the statistical validity and the objective reporting of results from animal studies. Animal studies often have small sample sizes and lack proper blinding methodology to prevent bias, which contributes to inaccurate conclusions instead of innovative discoveries. In fact, a German study found that fewer than 5 percent of animal studies were constructed to exclude researcher biases from impacting the results (2). Such experimental issues are concerning in the context of animal studies, which do not directly impact human life, but inaccuracies in such pre-clinical stages or research seriously influence the danger human subjects are exposed to.
However, the damage stemming from inaccurate representations extends beyond needlessly endangering human lives. In addition to promoting ineffective treatments beyond reasonable stages of development, improper treatment screening can critically harm the development of potentially lifesaving procedures. Incorrectly applying treatments can lead to mixed results during clinical trials and thus delay the acceptance of an otherwise invaluable treatment option for wider medical application. A notable example of this effect is seen with the uncertain and drawn out development of deep brain stimulation (DBS) to treat major depressive disorder (MDD). Originally introduced as a tool for treating the debilitating movement symptoms of Parkinson’s disease (PD) by subthalamic nucleus (STN) stimulation (3), DBS has been thoroughly examined in clinical trials and has proven successful in alleviating MDD symptoms in some patients (4). However, DBS for MDD failed a crucial futility analysis during its clinical trials with device manufacturer St. Jude Medical, leading to the study’s discontinuation. The study was terminated prematurely due to seemingly unsuccessful results in the data from the first six months. Following this premature conclusion, conflicting reports about the DBS treatment’s efficacy began to circulate. While some subjects reported that their depression worsened to the point of nonfunction, others said that receiving the implant immediately alleviated their depressive symptoms (5). There were numerous issues in the statistical analysis of DBS’s efficacy, most notably the small sample size. The longitudinal nature of the study made it prone to subjects leaving the study midway and the control group was so small that individually high depression scores were amplified disproportionately. Furthermore, there were unforeseen technical issues which further muddled the results of the study. In a recent study of DBS lead implants for PD patients, it was found that a 10 percent of subjects experienced significant postoperative lead migration, which resulted in adverse outcomes for these patients (6). As the features of the general procedures for DBS implants are generalizable across conditions, with only the target brain region changing, it can be assumed that similar difficulties would have been encountered in the St. Jude DBS study with treatment-resistant depression (TRD). An additional issue to consider with DBS for TRD is that even as of 2018, the precise mechanisms by which DBS affects TRD are largely unknown. Vast discrepancies continue to be identified between trials to this day due to vast discrepancies in the design of these studies. A study examining six potential DBS target regions for TRD found notable differences in the measured efficacy to DBS treatment by number and location of target brain region(s), a factor which is only further complicated by the interpersonal variability of neuroanatomical structures (7).
The aforementioned clinical and pre-clinical difficulties can certainly stifle biomedical progress, but each failed or contradictory study still produces a silver lining or a new line of questioning which may be produced. In the effort to apply DBS to TRD, considerable progress has been made despite the statistical and retrospectively identified technical flaws of past DBS studies. One of the doctors at the forefront of DBS research, Dr. Helen Mayberg, views the variability of experimental design and the resulting conclusions with some optimism. The different results in open-label studies compared to randomized, controlled studies can be used as indicators on what method is more effective in DBS study and application and illustrate that “[a]n open-label adaptive design may be the most prudent approach for a high-risk procedure performed in a vulnerable population” (7). These results could even explain the enormous disparity in patient response to treatment rates between multicenter, randomized studies (failed to confirm DBS efficacy for TRD) and open-label clinical studies (reported 60 to 78 percent response rates). Additionally, identifying the neuroanatomical variations and lead electrode migrations that contributed to discrepancies in and even adverse DBS effects has benefitted the development of DBS in the long run. By anticipating electrode movement post-operation, surgeons can consider that when they implant them initially and later when patients return for check-ups. On the other hand, the noted effect of neuroanatomy on DBS efficacy has also pushed for the development of another preventative measure to aid in DBS application. Improved neuroanatomical mapping of individual patients would assist in accurately implanting DBS electrodes for optimal anti-depressant effects. Such beneficial developments in the face of difficulty and even abject failure occurs in every part of biomedical research and allows progress to continue in the face of unfortunate, legitimate results and despite bad practice. Further measures are also being taken by many non-governmental organizations (NGO), such as the World Health Organization (WHO), and numerous research endeavors to protect patient rights and identify factors affecting patient outcomes. Such efforts aid in the development of ethical government policies and institutional guidelines for biomedical research to be efficient, accurate, and beneficial to patients.
As patients seek out novel treatment, experimental biomedical procedures gain new samples for their clinical studies. However, the transition of a patient to subject does not alleviate the necessity of ensuring safe and ethical practice in the application of treatments, experimental and otherwise. This commitment cannot begin only at the level of clinical trials. Instead, it must be ever present in the practices and goals of researchers across all levels, from basic science to animal studies to clinical studies and all stages of approval and review in between. Research can help innovate treatments and cures for disease, but it is of utmost importance that the science does not forget the human value, and cost, at which such research operates.
Edited by: Ryan Chang
Illustrated by: Eugenia Yoh