The University of Reading

 

Research Studentship in Medical Statistics

 

Statistical strategies for phase III clinical trials

of targeted therapies in oncology

 

A collaborative research project between the Medical and Pharmaceutical Statistics Research Unit, University of Reading  and AstraZeneca UK Ltd, Alderley Park

 

 

Applications are invited from UK or EU candidates for a three-year studentship concerning Statistical strategies for phase III clinical trials of targeted therapies in oncology, funded by AstraZeneca UK.  Full tuition fees and research expenses will be covered, and a stipend starting at £15,300 p.a. will be available.  Applicants should have a good Bachelors or Masters degree in statistics, or in a mathematical subject with a substantial statistics content.  The closing date for applications is 8 May, and a starting date of 1 October 2006 is anticipated.

 

 

Targeted therapies in oncology

 

Newer biological cancer treatments are often targeted for the specific ways that cancer cells differ from normal cells, particularly those features that help them to grow and spread.  For example, a cancer cell may have far more receptors for a chemical that acts as a growth factor than a normal cell. A drug that blocks that receptor will stop the growth factor from reaching the cancer cell, and so may stop growth of the cancer. These targeted agents sometimes work best in a particular group of patients, such as people who have a cancer with particularly high levels of the growth factor receptor that is being targeted. You may be aware of the breast cancer drug Herceptin which has received a lot of recent publicity. This is an example of a targeted therapy: women will only respond to the drug if their cancer has high levels of the HER2 protein.

 

 

Clinical development of targeted therapies

 

Once it is known that an agent will be effective only, or especially, in some subgroup of patients, then design of a phase III clinical trial will be straightforward.  Only patients who are members, or are suspected of being members, of the target group will be recruited to the trial. The number of patients recruited would be determined using standard sample size formulae. Within the target group, a larger treatment effect will be anticipated than if the trial were to admit all patients, and so power can be set for a larger treatment effect, leading to a smaller sample size.  

 

The development of a targeted therapy is, however, more complicated.  Not only does the appropriate dose have to be determined and safety and efficacy established, but the ideal targeted group also has to be determined.  This PhD research topic will concern the way in which identification of an appropriate target group can be built into the phase II and III trials forming the drug development programme.

The research project

 

One possible approach to the development of a targeted therapy is to begin by randomising all suitable cancer patients between the experimental treatment and control therapy.  Information on numerous biomarkers that might define a suitable subgroup will be collected from each patient along with their reaction to treatment.  At a series of interim analyses, the efficacy of the drug will be evaluated, and also which subgroups experience the greatest benefit.  As the drug development programme progresses, it will be possible progressively to eliminate different subgroups from the trial.  In the end, there will be a long run of patients in the selected subgroup, providing the definitive evidence of treatment superiority within the targeted group.  An analysis is sought which allows for the repeated interim analysis and the progressive selection of the target group.

 

The project would consist of developing suitable stopping rules and elimination rules, and evaluating their performance via theory and simulation.  Results from existing procedures used to select amongst several treatments will be investigated to see whether they can be adapted for selection amongst subgroups.  Comparisons will be made with alternative approaches.

 

Conduct of the research
The student will be based at the University of Reading, and will be supervised by John Whitehead who is Director of the Medical and Pharmaceutical Statistics Research Unit and a Professor of Applied Statistics.  The student will maintain close contact with AstraZeneca at Alderley Park, principally through Sally Hollis who will liaise with the student's supervisor at Reading in developing the research programme.  In this way, the student will have access to the considerable experience of clinical research in oncology available at AstraZeneca.  The student will spend some time at Alderley Park learning about their clinical research programme, and about the development of new drugs for cancer.  Progress of all research degrees in the School of Applied Statistics is overseen by a Higher Degree Committee, and in this case the Committee will comprise the supervisor, two other members of the School and Sally Hollis.   

 

Further information about the Section of Applied Statistics and the Medical and Pharmaceutical Statistics Research Unit can be obtained from www.reading.ac.uk/statistics or by contacting: 

 

Professor John Whitehead, MPS Research Unit, The University of Reading, 

Harry Pitt Building, Earley Gate, Reading RG6 6FN, UK

 

Tel: +44 118 378 8027, Fax: +44 118 975 3169

email: [log in to unmask]

 

To apply, please download an application form from http://www.rdg.ac.uk/pg/pages/apply/pg-applicationform.asp or send an e-mail requesting an application pack to [log in to unmask]