2018-04-24 Project Team Meeting - Agenda & Minutes

Date

26 Apr 2018

Attendees

• Richard Norman (Unlicensed) (Pistoia Alliance)

• Francisco Fernandez (Deactivated) (Abvance Biotech)

• Bryan Jones (Deactivated) (Eli Lilly)
• Helen Li (Unlicensed) (Eli Lilly)
• Chris Lloyd (Deactivated) (MedImmune)
• Carmen Nitsche (Deactivated) (Pistoia Alliance)
• Kieran Todd (Deactivated) (GSK)

Discussion items

Minutes

Feedback on AbVance II doc

See link to document in agenda

>General comment: high-throughput screening (platform) assays are used for flagging potential issues rather than for prediction. This data would then be followed up with lower-throughput more detailed assays to determine specific values and risk, if necessary.

>A wiki page will be created for the project which will allow everyone involved to contribute information.   

Validating the problem statements

>First problem statement; overall agreement that standardisation of particular surrogate assays is of value as this will allow direct comparisons between different sources of the same data. In addition, greater numbers of standardised data will lead to greater confidence when querying this data. Outstanding question is, will enough organisations adopt defined standards?

>Second problem statement; there is much literature and internal data reporting that certain assays do correlate with endpoint data. This holds especially for standard mAbs. Based on this we would need to identify which assays are predictive and we would need to validate that predictability as part of the project.

>Generally industry open to sharing which assays are predictive for mAbs but biggest challenge is understanding predictability for non-mAbs since the variability between organisations and molecules, with respect to the format, can be significant. Outstanding question is, would current predictions hold true for non-mAbs?

Defining the scope

>In scope: definition of SOPs (and guideline metrics) for biophysical assays of choice.

>In scope: assays which provide the best ‘bang for buck’ in terms of predictive power (critical decision-making data) as there are many assays being used and many of them provide overlapping information. Are we limiting ourselves to screening/platform assays? Would we need to define a minimum set of data needed to progress a molecule (aka TPPs)?

>In scope: establishing correlation between human PK and sequence/structure.

>Out of scope: Manufacturability (expression yields). Very system dependent so challenging to do cross comparisons unless you have appropriate controls.

What do we want to achieve by end of June?

>Primary objective is to sell the proposal and idea to get more people on board.

>Stress what we feel the gaps in current knowledge are and what the proposed approach would be, esp. for non-mAb biologics. Avoid going into specifics re. which assays to focus on.

>Collate examples of data sharing precedents and identify stakeholders which could provide guidance.

>This group could aim to play a leadership role in terms of what data should be released in this space based on FAIR principles – keep an eye on the FAIR workshop output.

Additional tasks:

>Identify key relevant literature.

>Prepare and circulate a short questionnaire to help refine the problem statements.

>Create graphic of how platform assays are linked to endpoint data and link to sequence/structure information. Are there any reviews which do this already?

>To understand the size of the data gaps, select examples of available molecules with sequence/structure and assay data associated with them.

 (>Create a list of common assays and criteria used to select these i.e. bang for buck.)

>Chris (registered) and Kieran (or colleagues) will attend the EMBL-EBI workshop.

----------

Richard Norman – 26 April 2018