Novel biomarker based designs for late phase clinical trials
Lead Research Organisation:
Lancaster University
Department Name: Mathematics and Statistics
Abstract
The development of personal medicine approaches is one of the major challenges of modern medicine. So far, research has focused on developing the most effective treatment for entire populations. However, the most effective treatment might be different for different subgroups of patients. Recent developments in modern laboratory medicine (like advances in genomics and proteomics) have lead to the discovery of new biomarkers determining subgroups of patients benefitting from a specific treatment. Innovative trial designs are needed that enable researchers to concurrently establish clinical relevance of a biomarker and develop the most effective treatment for the subgroups defined by this biomarker.
While clinical trial designs using a biomarker to identify patient subgroups likely to respond to a treatment have been proposed in the literature, there is currently little information available on the performance of these different designs. Furthermore, currently available designs typically evaluate only a single treatment and/or a single biomarker at a time. Another limitation of existing designs is that the effectiveness of the treatment can often only be observed after a long follow-up time. For example, in oncology every patient needs to be observed for at least five years before the effectiveness of the new drug (measured by the percentage of patients surviving at least five years after diagnosis) can be determined. In conjunction with an often large number of patients that need to be enrolled into the trial, this leads to prolonged development times of novel drugs.
The overall aim of my research is to develop innovative biomarker-based trial designs that overcome the above mentioned limitations. One of my main aims is to derive novel data analysis strategies maximising the information obtained within a trial. By using more information, fewer patients have to be enrolled into the trial. Thus, fewer patients are exposed to an ineffective drug or possible side effects.
I will also focus on deriving statistical methods shortening the time-to-market for new treatments. This can be achieved, for example, by incorporating surrogate endpoints into the trial design. A surrogate endpoint is a measurement used as a substitute for a clinically meaningful endpoint directly measuring the outcome of interest. Changes induced by a therapy on a surrogate endpoint reflect changes in the clinically meaningful endpoint. As surrogate endpoints can be observed after a short time period, they allow for a more timely decision about the effectiveness of the treatment.
Another objective is to develop new designs that allow evaluating several treatments, biomarkers and/or subgroups of patients within the same trial. The main challenge is to ensure that extending, for example, the number of treatments does not increase the probability of falsely claiming the effectiveness of an ineffective treatment. This can be achieved by implementing multiple testing procedures. Different procedures have been proposed in the literature and I will investigate which of these procedures can be implemented into a biomarker-based trial design and how well they solve the problem at hand.
Within this project, I will also implement adaptive design methods allowing changes of the trial design within an ongoing study. Traditionally, once a study has been started, no changes with respect to sample size, number of treatments, and other characteristics of the trial are allowed as these changes can compromise the validity of the results. Recent advances in statistics now allow for some adaptations of the trial design as, for example, stopping a trial early if the (in-)effectiveness of a treatment becomes overwhelmingly apparent.
To maximise the relevance and impact of my research, I will work closely with other statisticians as well as with clinical oncologists and clinicians. Results will be presented at conferences and published in scientific journals.
While clinical trial designs using a biomarker to identify patient subgroups likely to respond to a treatment have been proposed in the literature, there is currently little information available on the performance of these different designs. Furthermore, currently available designs typically evaluate only a single treatment and/or a single biomarker at a time. Another limitation of existing designs is that the effectiveness of the treatment can often only be observed after a long follow-up time. For example, in oncology every patient needs to be observed for at least five years before the effectiveness of the new drug (measured by the percentage of patients surviving at least five years after diagnosis) can be determined. In conjunction with an often large number of patients that need to be enrolled into the trial, this leads to prolonged development times of novel drugs.
The overall aim of my research is to develop innovative biomarker-based trial designs that overcome the above mentioned limitations. One of my main aims is to derive novel data analysis strategies maximising the information obtained within a trial. By using more information, fewer patients have to be enrolled into the trial. Thus, fewer patients are exposed to an ineffective drug or possible side effects.
I will also focus on deriving statistical methods shortening the time-to-market for new treatments. This can be achieved, for example, by incorporating surrogate endpoints into the trial design. A surrogate endpoint is a measurement used as a substitute for a clinically meaningful endpoint directly measuring the outcome of interest. Changes induced by a therapy on a surrogate endpoint reflect changes in the clinically meaningful endpoint. As surrogate endpoints can be observed after a short time period, they allow for a more timely decision about the effectiveness of the treatment.
Another objective is to develop new designs that allow evaluating several treatments, biomarkers and/or subgroups of patients within the same trial. The main challenge is to ensure that extending, for example, the number of treatments does not increase the probability of falsely claiming the effectiveness of an ineffective treatment. This can be achieved by implementing multiple testing procedures. Different procedures have been proposed in the literature and I will investigate which of these procedures can be implemented into a biomarker-based trial design and how well they solve the problem at hand.
Within this project, I will also implement adaptive design methods allowing changes of the trial design within an ongoing study. Traditionally, once a study has been started, no changes with respect to sample size, number of treatments, and other characteristics of the trial are allowed as these changes can compromise the validity of the results. Recent advances in statistics now allow for some adaptations of the trial design as, for example, stopping a trial early if the (in-)effectiveness of a treatment becomes overwhelmingly apparent.
To maximise the relevance and impact of my research, I will work closely with other statisticians as well as with clinical oncologists and clinicians. Results will be presented at conferences and published in scientific journals.
Technical Summary
Recent developments in genomics and proteomics enable the discovery of biomarkers that allow identification of subgroups of patients responding well to a treatment. Although trial designs using a biomarker to identify these subgroups have been proposed in the literature, methodological research into the design of biomarker trials is in its early stages. Currently available designs are restricted in the way that they typically evaluate only a single treatment or biomarker and the performance of only some of the designs has been studied in detail.
The overall aim of this project is to develop and evaluate novel biomarker designs overcoming the above mentioned limitations. The main objectives are: First, to derive novel analyzing strategies maximizing the information obtained from the data. This will be achieved by efficiently using subgroup data and deriving optimal randomization rules. Second, to extend existing biomarker designs and to develop new designs allowing evaluation of several treatments and/or biomarkers while controlling the overall type I error rate. Multiple testing procedures like the Bonferroni or the gate keeping procedure will be evaluated. Third, to incorporate short-term endpoints allowing for early stopping for futility/efficacy. In many disease areas the primary endpoint is only observed after a long follow up. Hence, at the time of the interim analysis little data is available for the primary endpoint. Methods like the double regression which incorporates information on short-term endpoints in order to improve the estimate of the long-term endpoint offer one way of solving this problem. Fourth, to allow inclusion of patients whose biomarker cannot be assessed by, for example, adding another arm randomising these patients between the treatments.
To maximise the impact of my research, I will work closely with statisticians, clinical oncologists, and clinicians. Results will be presented at conferences and published in scientific journals.
The overall aim of this project is to develop and evaluate novel biomarker designs overcoming the above mentioned limitations. The main objectives are: First, to derive novel analyzing strategies maximizing the information obtained from the data. This will be achieved by efficiently using subgroup data and deriving optimal randomization rules. Second, to extend existing biomarker designs and to develop new designs allowing evaluation of several treatments and/or biomarkers while controlling the overall type I error rate. Multiple testing procedures like the Bonferroni or the gate keeping procedure will be evaluated. Third, to incorporate short-term endpoints allowing for early stopping for futility/efficacy. In many disease areas the primary endpoint is only observed after a long follow up. Hence, at the time of the interim analysis little data is available for the primary endpoint. Methods like the double regression which incorporates information on short-term endpoints in order to improve the estimate of the long-term endpoint offer one way of solving this problem. Fourth, to allow inclusion of patients whose biomarker cannot be assessed by, for example, adding another arm randomising these patients between the treatments.
To maximise the impact of my research, I will work closely with statisticians, clinical oncologists, and clinicians. Results will be presented at conferences and published in scientific journals.
Planned Impact
Innovative statistical methodology for clinical trial designs that incorporate information obtained from biomarkers has been recognized as being of strategic importance to increase the chances of an efficacious drug successfully completing development and reaching needful patients. Stratified medicine is one of the MRC priority areas and a highlight notice calling for research in methodology for stratified medicine has been issued. The U.S. Food and Drug Administration also encourages the integration of biomarkers in drug development and their appropriate use in clinical practice. It is believed that this approach will diminish stagnation and foster innovation in the development of new medical products leading to more and more personalised medicine.Developing medicines using information from biomarkers is a highly relevant problem. However, methodological research into the design of biomarker-based designs is in its early stages.The development of sophisticated statistical methods for biomarker-based trial designs is the subject of this Fellowship
The overall aim of this project is to improve our understanding of how to tailor treatments and interventions to the individual needs of people living with a wide range of diseases and conditions, thus improving the health of thousands of patients within the UK and worldwide. The benefits of my research would be felt in the short-term as new medicines will be developed and their efficacy will be evaluated. Long-term benefits would become apparent as new drugs are approved and biomarkers are identified allowing prescription of the most efficient treatment for each specific patient.
The findings of my research activity will be of interest to all parties involved with developing and prescribing medicines based on biomarker results, particularly publicly funded clinical trials units, medicine regulatory bodies, cliniciansand clinical oncologists, and the pharmaceutical industry. Especially in the context of industry sponsored studies where biomarker -based drug development is regulated by agencies like the EMA or FDA, my designs may play a part in contributing high-quality evidence to support drug approval. The medicalcommunity would benefit as there would be new tools to improve the evidence base for prescribing decisions. Furthermore, efficient late phase designs may allow trials to be carried out in circumstances hitherto unfeasible due to for example sample size constraints. In general, the results of my research will make clinical trials more cost- and time-efficient.
The rate of uptake of my novel biomarker-based designs and analysis methods will be increased if user-friendly open-source packages exist for statisticians to implement them. As such I will provide such software which will be made available on a website informing about the status and the latest results of the project.
This project would have a positive impact on my skills and my future research career. Whilst undertaking this research, I will develop transferable skills which can be applied in all areas of my work, these include: leadership and project-management skills as I will lead the project from inception to completion, communication skills as I will communicate statistical results to non-statistical collaborators, and skills of managing relationships with inter-disciplinary collaborators.
The overall aim of this project is to improve our understanding of how to tailor treatments and interventions to the individual needs of people living with a wide range of diseases and conditions, thus improving the health of thousands of patients within the UK and worldwide. The benefits of my research would be felt in the short-term as new medicines will be developed and their efficacy will be evaluated. Long-term benefits would become apparent as new drugs are approved and biomarkers are identified allowing prescription of the most efficient treatment for each specific patient.
The findings of my research activity will be of interest to all parties involved with developing and prescribing medicines based on biomarker results, particularly publicly funded clinical trials units, medicine regulatory bodies, cliniciansand clinical oncologists, and the pharmaceutical industry. Especially in the context of industry sponsored studies where biomarker -based drug development is regulated by agencies like the EMA or FDA, my designs may play a part in contributing high-quality evidence to support drug approval. The medicalcommunity would benefit as there would be new tools to improve the evidence base for prescribing decisions. Furthermore, efficient late phase designs may allow trials to be carried out in circumstances hitherto unfeasible due to for example sample size constraints. In general, the results of my research will make clinical trials more cost- and time-efficient.
The rate of uptake of my novel biomarker-based designs and analysis methods will be increased if user-friendly open-source packages exist for statisticians to implement them. As such I will provide such software which will be made available on a website informing about the status and the latest results of the project.
This project would have a positive impact on my skills and my future research career. Whilst undertaking this research, I will develop transferable skills which can be applied in all areas of my work, these include: leadership and project-management skills as I will lead the project from inception to completion, communication skills as I will communicate statistical results to non-statistical collaborators, and skills of managing relationships with inter-disciplinary collaborators.
Organisations
People |
ORCID iD |
Cornelia Ursula Kunz (Principal Investigator / Fellow) |
Publications

Kunz CU
(2018)
An alternative method to analyse the biomarker-strategy design.
in Statistics in medicine

Wan F
(2019)
Confidence regions for treatment effects in subgroups in biomarker stratified designs.
in Biometrical journal. Biometrische Zeitschrift
Description | Conference of the International Society for Clinical Biostatistic, Netherlands |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Around 60 statisticians attended the session at the conference, in which I presented my research. During the discussion after my talk, I gained several helpful comments, which resulted in me pursuing another extension of my research. |
Year(s) Of Engagement Activity | 2015 |
URL | http://www.iscb2015.info/ |
Description | Heidelberger Kolloquium Medizinische Biometrie, Informatik und Epidemiologie, Germany |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | Around 30 statisticians and informaticians and epidemiologists attended my talk, which sparked a long discussion afterwards. I also was able to set up a new collaboration with one of the attendees and we plan to undertake some joint research which is related to my award. |
Year(s) Of Engagement Activity | 2015 |
URL | https://www.klinikum.uni-heidelberg.de/Kolloquium.6256.0.html |