Proprietary algorithms might be a significant hurdle though
----- Original Message -----
From: "Herremans, Paul [JanBe]" <PHERREMA@janbe.jnj.com>
To: "PharmWeb Computer Validation" <CompVal@pharmweb1.man.ac.uk>
Sent: Wednesday, April 11, 2001 3:50 PM
Subject: RE: Validation Requirements for Risk Software?
> If I can give my angle at this.
>
> The issue at stake is that the registrant wants to prove something.
> As a "consumer" possibly affected by the negative effects of a
> product put on the market, I would like the registrant to have
> correct evidence so that the agency can make a correct judgement.
>
> What I would expect the registrant to do, is to be certain about the
> evidence. That means: methods by which the evidence is gathered and
> processed.
>
> This implies that the registrant would validate the system that
> produces the results. As far as I am concerned, it means that the
> registrant has to prove the data used and the software (algortithms,
> ...) are correct and retain integrity through the whole process. And
> than I only have one answer: the methods and system must be
> validated, and provisions of part 11 regarding data integrity apply.
>
> Another story is how to validate. GAMP will help.
>
> Regards,
>
> Paul Herremans
> Janssen Pharmaceutica
>
> -----Original Message-----
> From: Christopher Cook [<mailto:ccook@gxp.co.uk>mailto:ccook@gxp.co.uk]
> Sent: Monday, 9 April 2001 19:08
> To: PharmWeb Computer Validation
> Subject: Re: Validation Requirements for Risk Software?
>
>
> My only experience with simulation software is using process/batch
modelling
> software for optimisation of processes during development (Aspen Tech
etc.).
> The results are then tested for real in reactors so we don't bother
> validating the software. However due to the need to distribute the data
> accurately we have thoroughly tested that these aspects work.
>
> If you were using the system to generate items for submission to the FDA
> they would expect Part 11 compliance which includes validation.
>
> Best Regards
>
> Chris Cook
> GxP Ltd
>
> ----- Original Message -----
> From: "David S. Farrier" <DFarrier@SummitPK.com>
> To: "PharmWeb Computer Validation" <CompVal@pharmweb1.man.ac.uk>
> Sent: Saturday, April 07, 2001 10:26 AM
> Subject: Validation Requirements for Risk Software?
>
>
> > Dear Colleagues:
> >
> > Assume that a pesticide manufacture registrant submits risk assessment
> > results to the EPA that are based on using available risk analysis
> software
> > (e.g., @Risk, DEEM, Crystal Ball, etc.). To what extent must the
software
> > be validated?
> >
> > In other words, is simulation software governed by the same guidelines
as
> > for GLP and GALP? Is any validation of the software needed? To what
> extent?
> > Following what guidelines?
> >
> > Your comments will be appreciated.
> >
> > David
> >
> >
> > ==================================================
> > David S. Farrier, Ph.D. Phone: 970-249-1389
> > Summit Research Services Fax:: 970-249-1360
> > 68911 Open Field Dr. Email: DFarrier@SummitPK.com
> > Montrose, CO 81401 Web:
<http://www.SummitPK.com>http://www.SummitPK.com
> > ==================================================
> >
> >
> >
>
>
>
|