Steelyard Analytics, Gaithersburg, MD
October 4th, 2019
Written by M Bernstein (Mestrelab Research)

The meeting was well attended by an international group representing regulatory bodies, academia, contract analytical laboratories, NMR instrumentation and software vendors, and industry. There was a wide range of competencies and contributions to the discussions and talks.

Agenda

 

Time Presentation Title Presenter
10:45 Welcome – Opening Statements Kristie Adams (Steelyard Analytics) & Dan Sorensen (Eurofins Alphora)
11:00 Presentation: “We want YOU for the ValidNMR community of practice” Kristie Adams (Steelyard Analytics)
11:20 Presentation: “Qualification of NMR instruments” Christoph Freudenberger (Bruker Biospin)
11:40 Presentation: “Applying the proposed USP <1220> concepts to validation of NMR methods” Dan Sorensen (Eurofins Alphora)
12:00 Lunch (Sponsored by Spectral Service AG)
13:00 Open Forum: “Debate of opportunities and obstacles to use of NMR in regulated industries – based on community Q&A Facilitators: ValidNMR committee
14:30 Keynote Presentation: “Combined Accuracy and Precision Validation – A qNMR Method Example” Carlos Amezcua (FMC Agricultural Solutions)
15:00 Workshop Adjourns

 

Presentations can be downloaded from the members’ area at validnmr.com

 

Discussion topics

 

Sponsorship model (C Freudenberger, Bruker)

The ValidNMR group no longer has a sponsored scientist but employs Ms Elina Hafer (elina(at)validnmr.com) to perform administrative and web-related tasks. Its financial requirements have changed, and this is an opportunity to modify the sponsorship model.

If you have an interest to sponsor the group, please contact us.

The sponsorship model was therefore rethought, and will be as follows for 2019/2020:

Platinum Sponsors (max 3, SOLD OUT)

  • Bruker BioSpin
  • Mestrelab Research
  • Sigma-Aldrich

Gold / Silver / Bronze Sponsors

  • Wako

Sponsors are reviewed annually, and there is an opportunity to sponsor at the highest level in the future.

Past digitised material

There is a lot of talks, etc. lying on servers around – e.g., past qNMR Summit/PANIC talks/EUROLAB docos (or links).

We would like ValidNMR.com to be a central repository for these. A special effort will be made to collect and curate these data and make them available to ValidNMR registered people.

We believe that we can be more thorough than qnmr.org.

 

A Q&A portal

We discussed the possibility for a web method for people to post relevant, qNMR-related questions – a little bit like the Reddit concept. Procedures would be in place to stop advertising, etc.

We discussed that people might agree to be moderators. When a question is posted then moderators receive an email and the question poser will hopefully get a quick first reply.

 

Weighed samples

There is a demand for samples that are pre-weighed, dissolved, and delivered in a sealed NMR tube. Merck and Bruker produce the “qPQ” sample is a good start, but we should encourage other CRM manufacturers.

The qPQ sample is a mixture of 2 reference materials at specified concentrations.

Wako markets a sample having a specified concentration of DSS in D2O.

 

Uncertainty determinations

Software to help with this on the web – see Carlos’ talk.

This is a part of a broader concern/issue: how do you validate qNMR method?

 

Where can metrology-inclined manuscripts be published?

MRC etc. are not keen to publish data like this (Kristie).

Suggested alternatives:

Anal Bioanal Chem Example

Accred Qual Assu Example

J Pharm Anal

J Nat Prod

Tetrahedron

 

Mixed (non) GxP environments

The question was raised, and some attendees described how they are achieving it.

It comes to user controls (esp. to experiments), data archival/permissions. Linux is well suited to this control.

 

Mixed forum

The discussion was dominated by ideas to perform, once and for all, the ultimate set of test spectra that will prove the basics of (qNMR) – linearity, etc. Once done, this should be published and used as a reference. Maybe picked up by USP as a general procedure?

John Edwards has taken the lead on this effort, and agreed to be the study director. ValidNMR will provide support, including mailings.

At a minimum, volunteer labs could do “Carlos’ 45 samples”, or maybe just 1 physical sample/3 solutions/3 replicates for each (“the nine”). The work could be split, respecting QbD criteria

An anonymised data analysis exercise could be an add-on activity.

There should be open access to the data used for the study.

It was proposed that an extension could include simple mixtures (say, 100% method). A standard mixture could be used, e.g. flavonoids in Ginkgo.

John plans to put together a proposal that will be presented to USP. If they like/agree, then the volunteers can be sought, etc. The “prize” would be ensuring acceptance by regulators.