Biologics database collaboration

Owned by Richard Norman (Unlicensed)
Last updated: Oct 24, 2019


Aim(s):

  1.  To compile a 'gold standard' data set of PK, Immunogenicity and Phys Chem properties for Biologics (primarily antibodies) based on information in the public domain, in the first instance. 'Gold standard' dataset defined as one which allows the direct comparison of different studies and contains a minimum set of required parameters.
  2. Beyond this, would like to model PK from primary sequence.


Contributors:

Vincent Dubois (MedImmune), Dave Fairman (GSK), Anna Gaulton (EMBL-EBI), Theresa Goletz (EMD Serono), Boris Grinshpun (EMD Serono), Anne Hersey (EMBL-EBI), Tushar Jain (Adimab), Andrew Leach (EMBL-EBI), Chris Lloyd (MedImmune), Yves Fomekong Nanfack (EMD Serono), Friedrich Rippmann (Merck)


Drivers, value proposition and contribution(s):

Drivers

(based on organisational needs)

Value proposition

(what difference it will make)

Contribution

(what you/your company are prepared to do to support the initiative)

  • Have well curated, more diverse (different animal models, different Abs, etc.) and evolving data on reference molecules.
  • Have all clinical PK data in one place and move away from comparing the same parameter across different molecules irrespective of the way this has been measured. (Dave)
  • Collect enough data to derive meaningful PK predictions - risk is there is not enough public data available so companies must be open to sharing non-public data. (Terry)
  • Derive biophysical data from suitably classified mAbs and understand potential correlations with clinical PK. (Yves)
  • Have a repository of well documented and high quality large molecule PK data in one place and help drive a better targeted data generation and interpretation across compound developers (Vincent)
  • Use the dataset to improve the likelihood of developing new predictive models
  • Use the dataset for better standardisation; allowing for robust comparison, cross referencing, identifying outliers and further analyses.
  • Time and effort
  • Generating new biophysical data (Yves)
  • Sharing newly generated biophysical data (Yves)
  • Sharing non-public data


Scope:

  • Initial focus is on mAbs
  • Include information on ADCs (this needs to be contextualised in the database)
  • Other IgG-like molecules to be included based on absence of technical limitations


Approach:

<Link to Meeting Minutes>

<Link to Google drive folder with additional publications of interest>

<Link to draft publication “Guiding principles for antibody PK data quality and classification”>

  • Collate PK on approx. 10 priority marketed Ab drugs which are know to have high quality published data and check whether this meets the above 'gold standard' definition.
  • Propose 'data quality' tiers into which data from published molecules can be placed: Link to PK quality tiers page
  • Collate PK and other data from additional Ab molecules and fit these into proposed 'quality tiers'
  • Once we identify T3 data - can we use internal data to move this to T1 or T2?
  • Once we have identified molecules in each Tier can we characterise these from a biophysical perspective and see whether there are correlations between PK and early data


Further needs:

  • Discussion on sources of data e.g. Jain et al., (2017) publication, FDA drug approval packages
  • Discussion on technicalities of database build, setup, etc. bearing in mind future inclusion of data from other biologic drug modalities beyond standard mAbs