Biomarkers SIG

The PSI / EFSPI Biomarkers Special Interest Group has been formed with the aim of developing knowledge and opinions within the statisticians about biomarker usage and the related statistical and study design techniques that are involved. We hope to provide a forum for discussion and the sharing of experiences regarding the use of biomarker data across all stages of pharmaceutical development, in topics such as:

  • Developments in Biomarker Statistical Metholodology related to the pharmaceutical development
  • Statistical issues involving the Translation of biomarkers from pre-clinical in-vivo and in-vitro models
  • The development of the statistical expertise on the  use of Omics (genomics, proteomics, metabonomics  and other multiplexed molecular assay systems) as a means for identifying potential biomarkers
  • Statistical issues relating to Biomarkers of safety and toxicity
  • Statistical Validation  and qualification of biomarkers
  • Biomarkers in clinical development, biomarker clinical study designs, stratification of patient populations, prediction of clinical response, diagnostics and related statistical issues in personalised healthcare 
  • The statistical assessment of benefits and costs of using biomarkers

If you have experience in one of these other areas, or in biomarker usage in general and would like to participate in the SIG, its committee or a future meeting then please contact Athula Herath ( Similarly if you would just like to be kept informed of our activities then please get in touch. More info can be found on our pages in the resources section of psiweb.


Mailing list and LinkedIn Group

To keep up to date with the activities of the SIG, contact us to join the mailing list. This will keep you up to date with our events and other interesting pieces of news. Similarly, we have a LinkedIn group where you can start and participate in discussions on biomarker related statistical topics.

Paper - A Statistician's Perspective on Biomarkers in Drug Development

The SIG's thoughts and advice have been summarised in a review paper in the 10th anniversary edition of Pharmaceutical Statistics (November/December 2011, Volume 10, issue 6). This paper entitled "A Statistician's Perspective on Biomarkers in Drug Development" provides practical advice to statisticians working in the field. It can be found on the journal website here

In addition an overview of the publication was also presented at the PSI Journal Club in September 2012. Slides and a recording of the presentation can be found from the PSI Journal Club webpage here.

A list of biomarker related references from the scientific literature can be found here. This includes many references from the review article, but with further additions too. Weblinks are included in the document.

Omics Web Resources

A list of web based resources for those new to 'omics work, including genomics, mass-spec etc is being established. This list can be found here in our resource folders.

Recent Activity

  • PSI Training Course - Statistical Issues in the Use of Biomarkers - 25th and 26th September 2013

    Sep 25, 2013
    In conjunction with the PSI training committee, the biomarker SIG arranged a training course on statistical issues in the use of biomarkers. This course has now been completed and was attended by 20 statisticians from across Europe.

    This two day course provided an introduction to the use of biomarkers in pharmaceutical drug development. The course provided an overview of general statistical principles in the development of biomarkers and biomarker-based classifiers and in the design and analysis of biomarker studies. Practical examples were given in R.
    We were lucky enough that the course could be led by statisticians with great experience in this area in Athula Herath, Nigel Dallow, Viswanath Devanarayan and Marc Buyse.

    To see the course flyer please click here.

  • Biomarker Validation - Case Studies and Approaches

    Oct 10, 2012

    The SIG's most recent face to face meeting was held on the 10th October 2012 at MedImmune in Cambridge on the topic of biomarker validation and qualification. Various scenarios for biomarker validation were considered and it was discussed how the burden of proof that needs to be demonstrated varies in each instance. Technical aspects of analytical assay validation were also described. This was a successful meeting with some very informative talks and around 40 attendees.

    The following topics were discussed:

    • An Overview of Scenarios for Biomarker Validation

    • The Burden of Proof required to validate a pharmacogenetic biomarker for adverse events

    • A Phase III Alzheimer’s Disease Programme based on Pharmacogenetic markers

    • Difficulties in translation from animal to man: some examples from neuroscience research

    • Biomarker method validation: Fit-for-purpose - statistical considerations

    • Developing Companion Invitro Diagnostic Tests

    For details and to see selected slide sets please click here.

  • Incorporating Biomarkers into Clinical Trial Designs

    Nov 11, 2011
    On the 11th November 2011 The SIG's held a face to face meeting at Amgen in Cambridge. This was one of our most successful meetings yet with 50 attendees and some interesting debate.
    The topic of this meeting was how biomarkers can be incorporated as an intrinsic part of a clinical trial design rather than as a mere "add-on". This topic is particularly relevant to personalised therapies.

    The following topics were expertly covered by the speakers:
    • An Overview of Biomarker-based Clinical Trial Designs
    • Incorporating Biomarkers into Adaptive Clinical Trial Designs
    • A Comparison of Approaches for the Identification of the Optimum Threshold for a Biomarker
    • Prospectively Designing a Confirmatory Trial on the Basis of a Post-Treatment Biomarker
    • Establishing Biomarker Study Designs for Early Phase Oncology Studies
    • Pharmacogenetics: Practices and Opportunities for study design and data analysis
    For details and to see the full slide sets please click here.

  • PSI Conference Session - How to Statisticians Guide Biomarker Research and Development

    May 17, 2011

    Practical Lessons from Clinical Applications

    The SIG arranged a parallel session at the 2011 PSI conference giving practical examples of a variety of biomarker uses. 3 high quality presentations covered the following areas:

    • Genetic variants as predictors of disease risk and efficacy
    • Validating a biomarker diagnostic test for early detection of cancer
    • Development of the safety biomarkers (SAFE-T) statistics working party

    To view slidesets from the day please click here. You may need to log in as a PSI member to see some presentations.

  • High-Dimensional, Multivariate and Composite Biomarkers

    Nov 05, 2010
    A one-day meeting was held at Roche, Welwyn Garden City on the 5th November 2010. The topic for this meeting was "High-Dimensional, Multivariate and Composite Biomarkers". This was our most successful meeting yet with over 40 attendees from across Europe and beyond.

    The extremely informative presentations included:
    • Introductions to the Challenges in Multivariate Biomarker Modelling
    • Development of Biomarker Signatures Within and Between Modalities
    • Group Discussion on Strategies for Developing Signatures
    • Multivariate Approaches to Molecular Biomarkers
    • Considerations in the Design and Interpretation Process
    • Case Studies on High-Dimensional Examples in Both Imaging and Proteomics
    To view slidesets from the day please click here. You may need to log in as a member to see some presentations.

  • The Use of Biomarkers in Clinical Decision-Making

    Dec 03, 2009
    A one-day SIG meeting was held at GlaxoSmithKline, Harlow on the 3rd December 2009. The topic for this meeting was "The Use of Biomarkers in Clinical Decision-Making".

    The meeting was again well attended and the breakout sessions encouraged particpants to openly share their experiences with the group, which led to some interesting discussions.

    Presentations included:
    • Examples of Biomarker Usage in Early Phase Oncology Decisions
    • Developing Biomarker-Based Decision-Making Criteria
    • Group Discussion on the use of Biomarkers in Clinical Decisions
    • Quantifying the Usefulness of PD Biomarkers in Phase II Trials
    • Designing a POC study Based on the Relationship Between Biomarkers and Clinical Endpoints
    • The Use of Biomarkers for Adaptive Treatment Selection in Progressive MS
    The full slide sets can be found in our resources folders here.

  • PSI Conference Session - Preparing Preclinical Biomarkers for Conversion to the Clinic

    May 18, 2009
    At the 2009 PSI Conference in Brighton the SIG ran a successful parallel session on the topic of translating biomarkers from preclinical to clinical applications.

    The outline of the session is as follows:
    Biomarkers play a valuable role in both preclinical and clinical drug discovery and development.

    Clinical biomarkers in humans will first have gone through a thorough development process to  demonstrate their utility. Appropriate preclinical statistical work should be completed with this transition in mind such that reliable endpoints can be taken on to further use, taking full  advantage of the knowledge that has been gained.

    During this session Chris Harbron (AstraZeneca) gave an overview of the challenges and
    techniques involved; Amanda Darekar (Pfizer) presented a case study of how this preclinical to clinical linkage has been successfully demonstrated for a urology marker and Phil Brain  (Pfizer) discussed a promising preclinical method.

    The slides for all of their talks can now be viewed in our resource folders here.

  • Biomarkers in Early Development Meeting

    Nov 18, 2008
    Our first one day face to face meeting took place on the 18th November at the GlaxoSmithKline Clinical Imaging Centre, Hammersmith Hospital. The topic for this meeting was "Biomarkers in Early Development."

    Presentations included:
    • An introduction to biomarkers in early development
    • The cost/performance of imaging biomarkers
    • A Bayesian strategy for early-phase biomarker trials
    • Crossover designs beyond the battenberg cake
    • Methods for comparing measurements when the gold standard is poor
    The meeting was well attended and appreciated. Attendees had the opportunity to see interesting case-studies, network with others who work in the area and have a forum for lively discussion.

    To view all the presentations please click here.

EventsFuture Events

  • A PSI Training Course on Improving Influence and Increasing Impact: Communication Skills for Industry Statisticians - Dates: 16 – 16 Sep, 2019

    This 1-day course will lead participants, in a non-threatening and encouraging atmosphere through steps to focus on improving the key skills of making impactful verbal and written communications.
  • PSI Webinar: Adaptive design: updated draft FDA guidance and its implications - Dates: 04 – 04 Jul, 2019

    In September 2018 the FDA published a guidance on adaptive design for clinical trials of drugs and biologics, updating (not finalizing) its initial draft from February 2010. The main focus of the webinar will be to provide an overview of its main contents, which will be presented by Jürgen Hummel (PPD). In addition, Kaspar Rufibach (Roche) will introduce an open-source statistical software for adaptive designs, RPACT (an R package that enables the design and analysis of confirmatory adaptive clinical trials). Kit Roes will also comment on the guidance from a European Regulatory perspective, which will be followed by a brief panel discussion.
MSD Logo Color

Statistics and Statistical Programming Opportunities at MSD


Costello Logo_RGB JPG

Statistics at Costello Medical


Pharm Research Associates

Principal Biostatistician


Autolus Therapeutics

Head of Statistical and Data Innovation as Head of Biometrics



Statistician Opportunities in the UK