XClose

UCL Institute of Clinical Trials and Methodology

Home
Menu

Current studentships

Extending the design of multi-arm multi-stage clinical trials

What is the Project?

Randomised controlled trials (RCTs) are the gold standard for testing whether a new treatment is better than the current standard of care. However, traditional RCT designs can take a long time and are often expensive.

Our multi-arm, multi-stage (MAMS) adaptive trial design allows us to test several different treatments in parallel (multi-arm) and allows us to cease recruitment early to treatments that appear futile (multi-stage). The advantages over traditional designs are: 1) saves time compared to testing all the treatments sequentially in a series of two-arm trials; 2) increases the probability of identifying a better treatment; 3) allows testing of more treatments than would ever be performed in two-arm trials; 4) more cost-effective and more efficient than multiple two-arm trials.

The MAMS approach is one of the few adaptive designs being deployed in a number of trials and across a range of diseases, including STAMPEDE (prostate cancer), CompARE (TB), TRUNCATE-TB (TB), RAMPART (renal cancer), and ROSSINI-II (wound surgery) prompting changes to treatment guidelines in patients with prostate cancer in the case of STAMPEDE trial.

The main objective in this PhD project is to extend the design with the aim of increasing its efficiency and application, i.e. to allow for different types of outcome distributions at interim and final analyses, to explore its statistical properties, and to develop both statistical software and practical guidance for practitioners.

 

Who are the ICTM and the MRC Clinical Trials Unit at UCL?

The MRC CTU at UCL is at the forefront of resolving internationally important questions in infectious diseases and cancer, and delivering swifter and more effective translation of scientific research into patient benefits. It does this by carrying out challenging and innovative studies, and developing and implementing methodological advances in study design, conduct and analysis.  You will be joining a team of renowned experts in the field of clinical trials.

Eligibility

Ideally, the candidate would be numerate with a strength for developing statistical methodology and an enthusiasm for applying those methods into practice, e.g. a degree in mathematics, (medical) statistics, or a related quantitative field.

How to apply & Additional Information

Who are the supervisors? Dr Babak Choodari-Oskooei and Professor Max Parmar. You will also be supported by a Thesis Committee (TC), which will provide degree-spanning support and advice about academic and training progress for the successful candidate over the course of the Doctoral study. This provides students with additional research input, improved institutional networking and more centralised management of their skills training.

When can I start? October 2020

What funding is available? Funding may be available for eligible, successful candidates in line with the current UKRI PhD studentship levels.

How do I apply? Please send your cover letter and CV to Dr Babak Oskooei (email: b.choodari-oskooei@ucl.ac.uk).

Deadline for applications: 31 January 2020.

How can I find out more?

https://www.ucl.ac.uk/clinical-trials-and-methodology/education/phd/current-studentships For an informal chat, you can contact Dr Babak Oskooei (b.choodari-oskooei@ucl.ac.uk).
Selected references:

 

Blenkinsop A, Parmar MKB, Choodari-Oskooei B (2019), Assessing the impact of efficacy stopping rules on the error rates under the multi-arm multi-stage framework. Clinical Trials. DOI: 10.1177/1740774518823551 Parmar MKB, et al. (2017), Testing many treatments within a single protocol over 10 years at MRC Clinical Trials Unit at UCL: Multi-arm, multi-stage platform, umbrella and basket protocols. Clinical Trials 14(5):451–461. Choodari-Oskooei B, Parmar MKB, Royston P, Bowden J. Impact of lack-of-benefit stopping rules on treatment effect estimates of two-arm multi-stage (TAMS) trials with time to event outcome. Trials. 2013;14:23 Bratton DJ, Phillips PPJ, Parmar MKB. A multi-arm multi-stage clinical trial design for binary outcomes with application to tuberculosis. BMC Medical Research Methodology. 2013;13:139.

Multiple imputation analysis of clinical trials with missing data

  1.  
What is the Project?

Many methods are now available to handle missing data. Of modern methods, the most popular is multiple imputation (MI), and the theory is well worked out for data sets with small numbers of variables. This project considers two issues which are less well worked out.

Some clinical trial data sets include many variables, and statisticians will want to include many of these variables in order to make the missing at random assumption more plausible and hence reduce bias in their analysis. It is not well understood how many variables can be included in a MI procedure before precision is lost, nor how the variables should be selected.

Missing data also occur in baseline variables, and published literature supports the use of simple imputation methods (not MI) for this problem. However, there is still widespread resistance to simple imputation, and its merits have not been demonstrated in realistically complex data sets with many variables. 

This project aims to provide clear guidance on how to specify the variables to be included in a MI analysis of a clinical trial. The results will also be relevant to analysis of observational studies. Particular topics to be addressed are:

1. Reviewing the literature on size and specification of the imputation model.

2. Exploring properties of MI procedures with increasing numbers of covariates in the imputation models. Results are expected to depend on size of data set and type of variables (e.g. continuous, categorical).

3. Comparing MI procedures with their simpler alternatives and deriving rules for what covariates are valuable to include in MI procedures.

4. Comparing simple baseline imputation procedures with alternatives in complex settings with missing data in both baseline and outcome, and in the context of a non-collapsible treatment effect measure.

5. Exploring the convergence of MICE procedures in practice.

 

The student will acquire mathematical understanding of the assumptions and implementation of MI methods and will explore whether this can be used to address the aims in simple settings. They will design and run simulation studies to evaluate MI with different imputation models and compare it with simpler alternatives. They will implement the methods in real data.

 

 

Who are the ICTM and the MRC Clinical Trials Unit at UCL?

The MRC CTU at UCL is at the forefront of resolving internationally important questions in infectious diseases and cancer, and delivering swifter and more effective translation of scientific research into patient benefits. It does this by carrying out challenging and innovative studies, and developing and implementing methodological advances in study design, conduct and analysis.  You will be joining a team of renowned experts in the field of clinical trials. The student will be based at MRC CTU and will be part of a group of more than 50 statisticians and methodologists and about 12 PhD students.

Eligibility

Ideally, the candidate would be numerate with a strength for developing statistical methodology and an enthusiasm for applying those methods into practice, e.g. a degree in mathematics, (medical) statistics, or a related quantitative field.

How to apply & Additional Information

Who are the supervisors? The project will be supervised by Ian White, Professor of Statistical Methods for Medicine, and Dr. Tim Morris, Senior research associate in medical statistics.

Prof. White has 20 years’ experience in developing and disseminating statistical methods to handle missing data in RCTs and observational studies, especially but not only MI. He consults and teaches about these methods. He also has wide expertise in simulation studies and recently published an influential tutorial paper on their use. His other research interests include novel designs for RCTs, handling treatment switches in RCTs, and methods for meta-analysis and network meta-analysis.

When can I start? October 2020

What funding is available? Funding may be available for eligible, successful candidates in line with the current UKRI PhD studentship levels.

How do I apply? Please send your cover letter and CV to Prof Ian White (email: ian.white@ucl.ac.uk).

Deadline for applications: 31/01/2020

How can I find out more?

https://www.ucl.ac.uk/clinical-trials-and-methodology/education/phd/current-studentships

For an informal chat, you can contact Prof Ian White (ian.white@ucl.ac.uk) or Dr Tim Morris (t.morris@ucl.ac.uk)

 

Good outcome measures for monitoring clinical trials

What is the Project?

Clinical trialists monitor trial data in order to protect the rights and well-being of participants, to ensure that the trial data are accurate, complete, and verifiable, and to confirm that the trial is being run in compliance with the currently approved protocol, and following both the principles of good clinical practice (GCP) and the relevant regulatory requirements(1). Academic clinical trials units (CTU) use risk-based monitoring whereby the risks in and to a trial are assessed and the monitoring plans are based on these. The monitoring typically consists of central monitoring (assessing metrics) with on-site monitoring carried out when central monitoring indicates a need (if the metrics are bigger than a threshold).

There has been little research on how best to carry out monitoring. One impediment is the uncertainty in what would be a good outcome measure.  The primary aim of this PhD is to suggest a good outcome measure for monitoring research.

Though this is the main aim, the PhD will cover one or two other aspects of monitoring or other trial conduct which will be developed with the applicant.

Who are the ICTM and the MRC Clinical Trials Unit at UCL?

The MRC CTU at UCL is at the forefront of resolving internationally important questions in infectious diseases and cancer, and delivering swifter and more effective translation of scientific research into patient benefits. It does this by carrying out challenging and innovative studies, and developing and implementing methodological advances in study design, conduct and analysis.  You will be joining a team of renowned experts in the field of clinical trials.

Eligibility

Ideally, the candidate would be numerate with a strength  in Bayesian methodology and an enthusiasm for improving trial conduct, e.g. a degree in mathematics, (medical) statistics, or a related quantitative field.

How to apply & additional information

Who are the supervisors? Associate Professor Sharon Love in the first instance. You will also be supported by a Thesis Committee (TC), which will provide degree-spanning support and advice about academic and training progress for the successful candidate over the course of the Doctoral study. This provides students with additional research input, improved institutional networking and more centralised management of their skills training.

When can I start? October 2020

What funding is available? A 3 year studentship covers tuition fees, research support costs, and a stipend; and is available for UK and EU candidates.

How do I apply? Please send your cover letter and CV to Sharon Love (s.love@ucl.ac.uk).

Deadline for applications: 31 January 2020.

How can I find out more?

https://www.ucl.ac.uk/clinical-trials-and-methodology/education/phd/current-studentships

For an informal chat, you can contact Sharon Love (s.love@ucl.ac.uk).

Selected References

1.         International Conference on Harmonisation of technical requirements for pharmaceuticals for human use (ICH). Guideline for good clinical practice E6(R2) 2018 [Accessed 7Feb2019]. Available from: https://www.ema.europa.eu/documents/scientific-guideline/ich-e-6-r1-guid....

 Towards a global data monitoring plan for clinical trials

What is the Project?

Clinical trialists monitor trial data in order to protect the rights and well-being of participants, to ensure that the trial data are accurate, complete, and verifiable, and to confirm that the trial is being run in compliance with the currently approved protocol, and following both the principles of good clinical practice (GCP) and the relevant regulatory requirements(1). Academic clinical trials units (CTU) use risk-based monitoring whereby the risks in and to a trial are assessed and the monitoring plans are based on these. 

The primary aim of this PhD is to create a global data monitoring plan template. The process will require obtaining template data monitoring plans (or template data management plans or template protocols) from each of the 50 UKCRC registered CTU in the UK. This information will form the basis of a Delphi process for many people to comment on the items in the plan. The aim would be to publish a template data monitoring plan.

Though this is the main aim, the PhD will cover one or two other aspects of monitoring which will be developed with the applicant.

Who are the ICTM and the MRC Clinical Trials Unit at UCL?

The MRC CTU at UCL is at the forefront of resolving internationally important questions in infectious diseases and cancer, and delivering swifter and more effective translation of scientific research into patient benefits. It does this by carrying out challenging and innovative studies, and developing and implementing methodological advances in study design, conduct and analysis.  You will be joining a team of renowned experts in the field of clinical trials.

 

Eligibility

Ideally, the candidate would have experience working in clinical trials, be organised and have skills or willingness to learn STATA.

 

How to apply & Additional information

Who are the supervisors? Associate Professor Sharon Love and Dr Victoria Yorke-Edwards. You will also be supported by a Thesis Committee (TC), which will provide degree-spanning support and advice about academic and training progress for the successful candidate over the course of the Doctoral study. This provides students with additional research input, improved institutional networking and more centralised management of their skills training.

When can I start? October 2020

What funding is available? A 3 year studentship covers tuition fees, research support costs, and a stipend; and is available for UK and EU candidates.

How do I apply? Please send your cover letter and CV to Sharon Love (s.love@ucl.ac.uk).

Deadline for applications: 31 January 2020

How can I find out more?

https://www.ucl.ac.uk/clinical-trials-and-methodology/education/phd/current-studentships

For an informal chat, you can contact Sharon Love (s.love@ucl.ac.uk).

Selected references:

1.         International Conference on Harmonisation of technical requirements for pharmaceuticals for human use (ICH). Guideline for good clinical practice E6(R2) 2018 [Accessed 7Feb2019]. Available from: https://www.ema.europa.eu/documents/scientific-guideline/ich-e-6-r1-guid....

 

Statistical methods for analysing clinical trials with treatment changes

What is the project?

Participants in randomised trials often change treatment during the course of the trial, particularly in long-term trials with survival-based outcome measures. These treatment changes may be specified by the protocol, or may respond to unexpected clinical or personal events. For example, some protocols specify a treatment change at a first disease event, but subsequent treatments are at the investigator’s discretion. In trials for patients with late-stage cancers, placebo arm patients may commonly be given the experimental treatment after their disease progresses. Many other forms of treatment changes occur in a wide range of trials.

In most cases, treatment changes affect the power of a trial, but they also raise questions about what quantity should be estimated. The standard intention-to-treat analysis compares the effect of “treatment now” with “possible treatment on progression”. However, funders typically want to compare the effect of “treatment now” with “no treatment”. Meanwhile, understanding the true effects of treatments may require accounting for differences in second-line treatments.

These problems are sometimes tackled through standard methods. In particular, per-protocol analysis is widely used in non-inferiority trials, because intention-to-treat analysis is seen as anti-conservative. However, per-protocol analysis is typically subject to selection bias.

Alternative causal inference methods – the rank-preserving structural failure time model (RPSFTM), inverse probability of censoring weighting (IPCW), and the two-stage method (described here) – can reduce bias and provide useful inferences. However they are statistically complex and poorly understood by wider audiences, they make their own untestable assumptions, and they can be hard to adapt to more complex settings.

 

Aims

The student will explore ways to improve and extend the statistical methods for analysing trials with treatment changes. Particular topics to be addressed are:

1. Reviewing the literature on standard methods and causal inference methods for handling complex treatment changes.

2.  Exploring when IPCW methods are suitable in NI trials: identifying any changes that may need to be made to design to facilitate IPCW methods, and identifying challenges in the analysis.

3. Developing methods that are appropriate when departure from protocol is not simply yes/no.

4. Building software for general application of the three methods and disseminating methods through tutorial-type articles.

Possible extra topics are:

1. Extending a recently proposed weighted log‑rank test, which improves the power of the intention-to-treat analysis and the RPSFTM analysis, to handle baseline covariates.

2.  Inverse probability of censoring weighting and the two-stage method are based on the same assumption of no unmeasured confounders, so when do they differ?

3. Extending methods to handle missing data about the treatment changes.

The student will acquire mathematical understanding of the assumptions and implementation of existing methods, and will use this to extend the methods to the new settings considered. They will design and run simulation studies to evaluate the proposed methods and compare them with existing methods. They will implement the methods in real data or write programs for others to do so. Three academic papers will be submitted during the course of the project.

 

Who are the ICTM and the MRC Clinical Trials Unit at UCL?

The MRC CTU at UCL is at the forefront of resolving internationally important questions in infectious diseases and cancer, and delivering swifter and more effective translation of scientific research into patient benefits. It does this by carrying out challenging and innovative studies, and developing and implementing methodological advances in study design, conduct and analysis.  You will be joining a team of renowned experts in the field of clinical trials. The student will be based at MRC CTU and will be part of a group of more than 50 statisticians and methodologists and about 12 PhD students.

 

Eligbility

Ideally, the candidate would be numerate with a strength for developing statistical methodology and an enthusiasm for applying those methods into practice, e.g. a degree in mathematics, (medical) statistics, or a related quantitative field.

 

How to apply & Additional Information

Who are the supervisors? The project will be supervised by Ian White, Professor of Statistical Methods for Medicine. Co-supervisor Nick Latimer (University of Sheffield) will contribute broad knowledge of treatment switching analysis from a health economic perspective. Co-supervisor Victoria Yorke-Edwardes (MRC CTU at UCL) will contribute a practical perspective. Statisticians at MSD will contribute to determining the overall direction of the project.

Prof White has 25 years’ experience in developing causal inference methods to account for changes from randomised treatment in RCTs. As an established leading academic in the field, he consults and teaches about these methods, including the rank preserving structural failure time model (RPSFTM) and inverse probability of censoring weighting (IPCW). He also has wide expertise in simulation studies and recently published an influential tutorial paper on their use (Morris et al 2019). His other research interests include novel designs for RCTs, handling missing data, and methods for meta-analysis and network meta-analysis.

When can I start? October 2020

What funding is available? Funding may be available for eligible, successful candidates in line with the current UKRI PhD studentship levels.

How do I apply? Please send your cover letter and CV to Prof Ian White (email: ian.white@ucl.ac.uk).

Deadline for applications: 31/01/2020

How can I find out more?

https://www.ucl.ac.uk/clinical-trials-and-methodology/education/phd/current-studentships

For an informal chat, you can contact Prof Ian White (ian.white@ucl.ac.uk)

Recognising contributions to clinical trials

What is the Project?

There is a grave challenge in acknowledging the efforts of everyone involved in delivering a clinical trials, including collecting and preparing data. Hundreds or thousands of people will be involved, particularly for a long-term, international clinical trial with many sites. Many research staff, centrally and at sites, are overlooked for explicit recognition, despite commonly having creative input into solving the challenges in conduct and delivery of trials. People working on the trial when a paper is written are more likely to be credited with authorship at the expense of previous workers. ICMJE has clear and restrictive guidelines for who should qualify as authors which are unlikely to change. All told, paper authorship is not sufficient to reflect the efforts of all of the people involved in the trial. These efforts need to be recognised so that individuals can demonstrate involvement for career progression and take pride in their work.

The credits at the end of a movie list everyone involved production by name and role and the Internet Movie Database (IMDb)  has served as an online repository for making clear cross-movie credit. In this project, we would seek to determine whether an appropriately-developed and -presented credit list for clinical trials would provide recognition in a way acceptable to researchers. If so, an IMDb-like database for clinical trials would be developed.

Questions to be addressed would include: How are trialists currently choosing authors for large scale trials? How are other research staff recognised? Would “film credits” be received favourably as a way to recognise people? Would this allow individuals to demonstrate involvement for career progression, in addition to just taking pride in their work? Can a database be developed for a single trial to help researchers in the future select authors for trial papers? What data about individuals would need to be captured, including ORCIDs? Can a database be developed to hold information for a single trial about people who work on a trial? Can the database be developed to hold information across trials allowing research staff to be linked across all of their projects, thereby building an easily-referenced portfolio? If so, how would contributions be verified?

The PhD researcher would already have some of the following skills and develop the others: literature reviewing skills; qualitative interviewing skills; database mapping skills; ability to engage external organisations; a passion for recognising effort.

 

Who are the ICTM and the MRC Clinical Trials Unit at UCL?

The MRC CTU at UCL is at the forefront of resolving internationally important questions in infectious diseases and cancer, and delivering swifter and more effective translation of scientific research into patient benefits. It does this by carrying out challenging and innovative studies, and developing and implementing methodological advances in study design, conduct and analysis.  You will be joining a team of renowned experts in the field of clinical trials. You will be supported by a Thesis Committee (TC), which will provide degree-spanning support and advice about academic and training progress for the successful candidate over the course of the Doctoral study. This provides students with additional research input, improved institutional networking and more centralised management of their skills training.

 

How to apply & Additional information

Who are the supervisors? Professor Matthew Sydes. The members of Thesis Committee will also provide support and advice.

When can I start? October 2020

What funding is available? A 3 year studentship covers tuition fees, research support costs, and a stipend; and is available for UK and EU candidates.

How do I apply? Please send your cover letter and CV to Prof Sydes (m.sydes@ucl.ac.uk).

Deadline for applications: 31 January 2020.

How can I find out more?

https://www.ucl.ac.uk/clinical-trials-and-methodology/education/phd/current-studentships

For an informal chat, you can contact Prof Sydes (m.sydes@ucl.ac.uk).