Conducting a clinical research is very delicate and when compared to some other forms of research it findings impact its research subjects directly.
Although I have not been involved in too many clinical researches I read a lot of books and there are three major reasons I think most clinical researches fail:
> At first glance, one could conclude that the reason is the stringent FDA review process designed to ensure that only products that deliver an acceptable efficacy and safety profile ever reach consumers.
Other simple but costly reasons of clinical research failures are the following
> Failure to have the right supplies. Either you don’t have all the supplies you need or the ones you have are outdated or expired.
> Failure to follow the directions. everyone can be guilty of rushing through the directions and missing a key step in research.

I have assisted in many pre-clinical and clinical research opportunities to know a lot of the ways that it can go wrong. I specifically worked with rats and performing surgery to remove their brain and basically freeze them in a certain preservative in order to transfer them to other places or just store them for future use. However, a lot of the times, the brains would not correctly be placed in the preservative, basically making the brain useless to use. Sometimes it could be a malfunction with the equipment, such as the freezer. If the freezer stops working overnight, it can be a problem going back to work and realizing that all the samples are unviable. There has also been times where the rat was improperly disposed leading to regulatory failure. There are also times where protocols are not exactly followed and corners are cut in order to speed up the process, which can lead to false data.

As being a Bioanalytical scientist, I deal with a lot of clinical and pre-clinical studies. My biggest disappointments are when other scientists overdose animals and research are been called off. One time, they inadvertently overdosed monkeys and after testing 580 samples, I came to know company called off the study due to monkeys dying right and left. Also, being GLP study, they had to have certain number of animals as a requirement, so even though only couple of them passed away, they still called it off. Human samples biggest failure is controlling them and their behavior. I feel animals are much easier to control than humans especially in food and water.

I haven't done any type of pre-clinical or clinical research, but if was to assume, the failure of the research would possibly something quite simple. For example, not following the direction. It could also be something as simple as that the approach you are taking actually does not work.

While I have never encountered a catastrophic failure for the clinical research we do in our lab, we have encountered a number of setbacks. The most common reason for these is rushing and failing to follow directions perfectly to maximize speed. One example was a situation where a large amount of clinical data was lost because a user entered the data quickly and proceeded to enter the next sequence of information without confirming that the previous sequences had been saved and processed. The data was lost, and some expensive and time-consuming procedures had to be repeated on our patients.

Of course, this is an example of a cause of failure not limited to just pre-clinical or clinical research. Do any of you have examples of similar reasons for failure?

I agree with @ibraheemshaikh that you need to also be careful in entering data. It is extremely important and crucial to always double and triple check and always keep a cautious mind.
About clinical research, I had worked for 2 years in a clinical lab and we didn't have failures but we definitely ran into many issues which are quite common. For example, subjects/participants commitment (they may run late, miss their window, miss their appointments, quit in the middle of the experiment period or just simply change their mind and don't want to use their data anymore); troubleshooting devices (sometimes, the devices you built just break down, but that's how you learn); human error (entering data, data analysis mistakes, mistakes in script, error in paperwork, IRB forms leading to freezing the projects etc...)...There can be many many challenges so our rule of thumb is to always have a checklist, always double checking, always ask questions and keep organized.

About pre-clinical research, I have interned/worked in this clinical lab and we tested chemo drugs on various cancer cell lines. For various cell lines, even though the drug showed a potential reduction in tumor size and weight previously in colon cancer after retested, it just didn't work as well for all types of cells such as breast cancer or lung cancer. I guess that's just how research is, full of trials and errors until you find the cure.

I have few questions for @rd389.

1. Why was a need to overdose animals?
2. What measures were taken to avoid future overdosing to animals?
3. How were the dead animals disposed?
4. Any care was taken during disposing, considering bio hazards?

Answer these questions only if you are comfortable.

I had worked in a research lab on TBI when we used a blast tube to collect data on pressure being applied to the brain. One of the issues that we experienced was the amount of pressure that were actually being applied to the mice, needing a buffer of sorts. The data showed that the intensity of the force was so great that it was difficult to get the data in the controls that were used as set points. They remedied this with kind of an armor that was built for the mice.

Where I work currently, there was such a failure during the early phase 2 clinical trial period for an investigational drug. The drug had performed remarkably well during animal testing trials. However, when it was used on the intended patient population during limited phase 2 clinical trials, there was a marked increase in adverse events associated with administration of the drug including death. The trial was canceled immediately and the analysis of data was undertaken. Upon further investigation, it appeared that certain nuances in certain IL levels resulted in catastrophic failure of the drug in the at risk population. The ultimate cause for the failure of the study included diminished interleukin levels in the at risk population when compared to healthy controls and discrepancies between animal models and human models.

Interesting. Did they go back and re-do the protocol and initiate the the trails again after realizing the diminished interleukin levels? If they conducted it again, what happened afterwards?

I would like to know the answer to this as well.

After the brain is frozen and transported, I assume to a customer, would the customer try to use the brain, realize it doesn't work, and then go back to you? I assume after that you do an investigation to see why it failed afterward?

I was part of clinical research where I was working on the quality control of medical images from the beginning of the trial to the end of the study. Most of the trials are carefully planned and executed as there are many lives involved and billions of dollars spent in conducting the trial. However there are a few cases where the trial is abruptly terminated from any further doses delivered to the patients. There might be two reasons for this: a. more than an expected minimum number patients are showing adverse reactions to the drug or b. the patients do not show any response to the drug at all (which means the disease is progressing). This might due a lot of factors. Success in animal studies gives only a minimal insight for a 'possibly' similar reaction in humans but does not necessarily mean they work all times. The factors that contribute to failed clinical trials maybe: 1. selecting patients who exhibit variedly different stages of the disease. It is easier to assess the progress of a disease when all the patients selected for the trial show similar activity and therefore possibly similar response to the drug. 2. genetic characteristics of the population of a particular gene pool. Patients belonging to different geographical locations react differently to the same drug. This can be due to the inherent differences in their DNA and also how the environmental factors play a huge rule in shaping the immune system.

I haven't worked on any clinical trials or clinical research. But I came across this website that graphically describes the reasons for clinical trial failures from 2013 to 2015. It seems like "efficacy" is the most common cause of failure in Phase II and Phase III trials. There are more operational and commercial reasons for failure in Phase III trials than in Phase II. However, trials fail for safety and strategy in Phase II than in Phase III trials.

I wish I was able to read this full article. It would be interesting to know a couple of specific examples or understand what the scope of efficacy means for the different therapeutic areas. Check out the link below to see the pie charts.

http://www.nature.com/nrd/journal/v15/n12/fig_tab/nrd.2016.184_F1.html?foxtrotcallback=true