Present and Future of Clinical Trials in the Pharmaceutical Field
05-06-2017
During all these years as active members of the industry, we have seen how the clinical trial proposal has been evolving, reflecting changes in the national legislation, and in FDA and EMA regulations. During this time, we have witnessed the evolution of the drug and device development model; we have had the privilege to witness dozens of success stories in the development of new drugs and medical devices, from their earlier phases until their marketing, but above all, we have had the opportunity to see new expectations and hope arise for patients from all social-economic and demographical groups, making us proud of our humble contribution to the progress reached.
Currently, the ultimate purpose of a clinical trial has not changed at all, being the same as yesterday and tomorrow: making new treatments available for patients all over the world to improve their quality of life, reduce their suffering, give them new hope and cure their medical conditions.
This purpose is certainly worth the extraordinary financial effort (investment), human effort (research), and productive effort (work) not only by the pharmaceutical industry but also by absolutely every stakeholder in the process, namely, the regulatory agencies seeking to protect patients' rights and ensure respect for the efforts of others in the search for new methodologies; the companies that develop drugs and devices that are constantly looking to innovate, offer new solutions to both known and future problems, make major financial efforts, develop new ideas that require incalculable effort knowing that what is a new idea today will be a starting point to look for another to top it tomorrow; the investigators and doctors who want to offer their patients better treatments and are always willing to learn new methodologies, drugs or devices that will help them treat their patients' conditions; the providers looking to facilitate access to technologies and/or knowledge to anyone who wishes to develop a research task—CROs, translators, providers of technology known by their initials, such as eCRF, eTMF, CTMS, and many others that symbolize years of development and innovation, etc.; and the patients who undergo tests and controls, and whose participation not only helps to treat others today, but to create new future lines of research.
Today, we can tell you how BioClever CRO makes its contribution to all these processes and how the professionals who are part of it understand that we have to act to keep providing trusted solutions in the future.
Upcoming regulation
Clinical trial regulations are evolving, in an effort to standardize administrative processes to start up clinical trials so that innovation does not get lost in bureaucracy, always ensuring that no mistakes or disturbances occur that may result in the opposite effect to the desired one in detriment of the patient. These points are as simple as they are important, and that is why, changes sometimes cannot be made as fast as we would like; they must be introduced in a controlled manner; their effect must be observed; and, in the process of innovation, participants must have time to understand their importance and adapt to them.
From our start-up department, we trust that change, with time and logic, is always for the best from a regulatory perspective and, for this reason, all processes to start a clinical trial must be perfectly understood, including not only the procedure to follow, but also the rationale underlying said procedure. After years of experience in this start-up process, during which we directly witnessed changes in national and international rules and regulations, we know that the future holds simplified processes and more efficient time frames and costs for us, which will prevent innovation in drugs and devices from becoming excessively bureaucratized.
From our start-up department, we trust that change, with time and logic, is always for the best from a regulatory perspective and, for this reason, all processes to start a clinical trial must be perfectly understood, including not only the procedure to follow, but also the rationale underlying said procedure. After years of experience in this start-up process, during which we directly witnessed changes in national and international rules and regulations, we know that the future holds simplified processes and more efficient time frames and costs for us, which will prevent innovation in drugs and devices from becoming excessively bureaucratized.
Monitoring Change
Monitoring is the quintessential service of any CRO, at the same time the most widely known and relevant to all. That is why, we have asked some of our CRAs about their opinions on the immediate present and future of monitoring, and their answers reflect how good they are at what they do.
Their role is so vital that, in recent years, everybody has heard about remote monitoring and risk-based monitoring (RBM), two types of monitoring that may not only reduce costs, but could also allow CRAs to spend less time traveling and more doing what they do best: monitoring, supporting the investigator, controlling the proper progress of the trials from start-up to closure of each site.
Our CRAs agree that this future is intertwined with the present; change is already here, and BioClever's team of CRAs is prepared to face it with confidence and trust.
Their role is so vital that, in recent years, everybody has heard about remote monitoring and risk-based monitoring (RBM), two types of monitoring that may not only reduce costs, but could also allow CRAs to spend less time traveling and more doing what they do best: monitoring, supporting the investigator, controlling the proper progress of the trials from start-up to closure of each site.
Our CRAs agree that this future is intertwined with the present; change is already here, and BioClever's team of CRAs is prepared to face it with confidence and trust.
Quality of Information and the Importance of Data Management
To discuss data management, we propose to address it from the perspective of influence on data quality. On that subject, a colleague told me: “A professor I had in college told us once: "No study is better (...) than the quality of its data." (Prof. Erik Cobo).”
You cannot get the most out of statistics and, therefore, you cannot reach useful conclusions if your data are not “good”, and that is where data management plays a critical role.
Any inaccuracies in data collection mean loss of information and, consequently, limit statistical efficiency. Between the original collection of data in a study and the statistical analysis thereof, data go through a process of transfer, recording and handling known as data management, the final objective of which is to obtain a database that reflects reality as reliably as possible.
Currently, data management could not be performed without the essential use of computers, whether to support data collection or processing. That is why, there is software exclusively designed to manage clinical data; at BioClever, we use cutting-edge tools that not only comply with FDA and EMA regulations, but also ensure the integrity of data at all times; these tools are highly versatile and intuitive without losing any quality; that is why we have been working for years with one of the world leaders in the field of electronic tools for data capture and management in clinical trials.
On the other hand, it is important to highlight that the departments of Data Management and Statistics are completely independent from each other. A few years ago, the Statistics Department was in charge of reviewing and validating the database prior to its analysis. Currently, they must be completely separate; it is the only way to ensure that the database closed by the Data Manager will not be modified at the time of analysis.
Lastly, it is important to mention the new international guidelines to standardize procedures and databases. In the future, we will be working under these guidelines intended to develop and support global standards, which are independent from the data platform and allow data interoperability.
You cannot get the most out of statistics and, therefore, you cannot reach useful conclusions if your data are not “good”, and that is where data management plays a critical role.
Any inaccuracies in data collection mean loss of information and, consequently, limit statistical efficiency. Between the original collection of data in a study and the statistical analysis thereof, data go through a process of transfer, recording and handling known as data management, the final objective of which is to obtain a database that reflects reality as reliably as possible.
Highlight:
- CRF Standardization, Databases, Variables and Formats.
- Close collaboration between biostatisticians and CRAs in key aspects, such CRF design, validation plan and query resolution.
- Close collaboration between biostatisticians and CRAs in key aspects, such CRF design, validation plan and query resolution.
Currently, data management could not be performed without the essential use of computers, whether to support data collection or processing. That is why, there is software exclusively designed to manage clinical data; at BioClever, we use cutting-edge tools that not only comply with FDA and EMA regulations, but also ensure the integrity of data at all times; these tools are highly versatile and intuitive without losing any quality; that is why we have been working for years with one of the world leaders in the field of electronic tools for data capture and management in clinical trials.
On the other hand, it is important to highlight that the departments of Data Management and Statistics are completely independent from each other. A few years ago, the Statistics Department was in charge of reviewing and validating the database prior to its analysis. Currently, they must be completely separate; it is the only way to ensure that the database closed by the Data Manager will not be modified at the time of analysis.
Lastly, it is important to mention the new international guidelines to standardize procedures and databases. In the future, we will be working under these guidelines intended to develop and support global standards, which are independent from the data platform and allow data interoperability.
Statistics as Driver of the Scientific Method
Clinical research needs to combine Medicine and Statistics.
The Scientific Method Model
1. Identify the problem to investigate.
2. Document and define the problem or hypothesis.
3. Deduce contrastable consequences from hypotheses.
In this process, statistics can contribute by means of two highly different strategies:
- Exploring and helping to define the topic you wish to research.
The Scientific Method Model
1. Identify the problem to investigate.
2. Document and define the problem or hypothesis.
3. Deduce contrastable consequences from hypotheses.
4. Design the observation or experiment.
5. Collect the data.
6. Analyze and make statistical inferences.
7. Establish conclusions.
8. Include conclusions in the body of knowledge.
5. Collect the data.
6. Analyze and make statistical inferences.
7. Establish conclusions.
8. Include conclusions in the body of knowledge.
In this process, statistics can contribute by means of two highly different strategies:
- Exploring and helping to define the topic you wish to research.
- Confirming the previously defined hypothesis.
Some questions that may be answered with the help of statistical methodology. The easier example would be to study the distribution of a single variable. How tall is a human being? If there were no variability, if all human beings were the same height, the inference would be immediate: measuring one person's height would be enough to know the height of all human beings. Since there is variability in nature, the statistical methodology proposes to consider and quantify the diversity between people.
Another type of question that addresses statistics is the simultaneous study of two variables: What is the relationship between calorie intake and weight gain?
When the objective is to make a prediction, the most important thing is to quantify the predictive capacity of the relationship under study, i.e., to what extent does knowing the daily calories that a person consumes allow us to know their weight gain? If the intention is to intervene on a variable to change the value of another one, it is necessary to establish a cause-effect relationship. Prospective epidemiology would address the question as follows: "What is the effect of increasing calorie intake?" Whereas retrospective epidemiology would try to answer the question: "What are the causes of the weight gain?"
The challenge for statistical inference is to generalize a phenomenon observed in a few cases to all possible observations. This ability of statistics to make a formal inference from a few sample data to all the population has promoted a spectacular progress in all sciences, particularly in the field of medicine.
To finish this insight into the field of statistics and data management in clinical trials, we would like to mention that, in recent years, ever since it has become possible to determine a person's genetic chain, we have been headed towards an increasingly personalized medicine. This generates a massive amount of data and poses new challenges for statistical science (microarray analysis, gene expression, ultra-sequencing, etc.), which needs specific methods to analyze this type of data. This is already being implemented, but it will become increasingly common.
It should not be forgotten that no statistical analysis method, as sophisticated as it may be, allows to draw proper conclusions from poorly planned experiments.
We hope that you have enjoyed this article; it was certainly a pleasure to write it; for more information, you can contact us and ask about any of its co-authors: Dolores, Monica, Pilar, Mireia B., Mireia V., Laura, y Miguel.
Another type of question that addresses statistics is the simultaneous study of two variables: What is the relationship between calorie intake and weight gain?
When the objective is to make a prediction, the most important thing is to quantify the predictive capacity of the relationship under study, i.e., to what extent does knowing the daily calories that a person consumes allow us to know their weight gain? If the intention is to intervene on a variable to change the value of another one, it is necessary to establish a cause-effect relationship. Prospective epidemiology would address the question as follows: "What is the effect of increasing calorie intake?" Whereas retrospective epidemiology would try to answer the question: "What are the causes of the weight gain?"
The challenge for statistical inference is to generalize a phenomenon observed in a few cases to all possible observations. This ability of statistics to make a formal inference from a few sample data to all the population has promoted a spectacular progress in all sciences, particularly in the field of medicine.
To finish this insight into the field of statistics and data management in clinical trials, we would like to mention that, in recent years, ever since it has become possible to determine a person's genetic chain, we have been headed towards an increasingly personalized medicine. This generates a massive amount of data and poses new challenges for statistical science (microarray analysis, gene expression, ultra-sequencing, etc.), which needs specific methods to analyze this type of data. This is already being implemented, but it will become increasingly common.
It should not be forgotten that no statistical analysis method, as sophisticated as it may be, allows to draw proper conclusions from poorly planned experiments.
We hope that you have enjoyed this article; it was certainly a pleasure to write it; for more information, you can contact us and ask about any of its co-authors: Dolores, Monica, Pilar, Mireia B., Mireia V., Laura, y Miguel.
Source: PMFARMA (http://www.pmfarma.es/)
