Validation process

[KarenA01]

I work for a biotech company that produces biologics (enzymes etc). We are workin on our first product and have 13485 and don't have a design exclusion, however we don't make medical devices or in vitro diagnosis kits. At most company who do such things would buy are products, however other customers (most?) will not be using them for medically related purposes...

Now my question...


We are working to bring our first product to market... What has to be done tduring Validation phase?

I think the R&D people want to just do the testing on material produced during production scale development, when there was no approved validation protoco nor final batch records.

They wanted to declare the last development lots as potential product and only want the validation phase to consist of doing formal release testing on lots produced produced during development - this is before formal approved final product batch records that worked do into the Product File (Device File) are created...

Coming from Pharma by understanding is that the Validation stage of product development is about the whole process... In that once R&D develops a process, the validation phase costs of making the lots specifically to be validated specifically as part of the validation phase and not just testing lots made during the development of the manufacturing processes.

Their view is that that can take the lots created during development just do the formal release testing... and the validation protocol would be just about that.

It seems to me manufacturing biologics has more in common with Pharma processes than devices, and in Pharma we actually had to make the validation lots AND test them, as part of the formal validation. Thsi was after all manufacturing processed were finalized.

Who is correct here? Under 13485 can the development stage lots be used for validation event though the process was not locally finalized and validation protocol not written and the validation phase only consist of testing those lots?
Thanks,
-Karen

 

[yodon]

Was hoping someone with more expertise would jump in but I'll offer up my thoughts.

Their view is that that can take the lots created during development just do the formal release testing... and the validation protocol would be just about that.

There is the concept of validating only processes whose outputs are not 100% verified. If the lot formal / release testing would provide sufficient evidence that each lot is fully verified, additional validation wouldn't seem necessary. Test results could be provided to customers in the form of a CoC, if they need it.

 

[v9991]

a simple short answer would be
* batches from development stages could be made part of the validation, subject to necessary justification - rationale (- approved/authrized through documentation); and formally documented as part of the D&D plan, PV protocol etc., viz., the equivalence of the "process" - "controls" - etc., and the tricky part is the justification and rationale, viz., the equivalence of the equipment reliability, process robustness, including the 100% inspection or rejection/failure rates. etc

There are a couple of points that are implicit ; and this is not about who is right; this is about what is right; and why/how it is right ......with limited exposure to the device industry, but adequate exposure to drug products and combination products, let me keep the discussion going.
which I would like to clarify and elaborate

Who is correct here? Under 13485 can the development stage lots be used for validation event though the process was not locally finalized and validation protocol not written and the validation phase only consist of testing those lots?

in any case, this has to be formally documented, and justified with appropriate rationale.

We are workin on our first product and have 13485 and don't have a design exclusion, however we don't make medical devices or in vitro diagnosis kits. At most company who do such things would buy are products, however other customers (most?) will not be using them for medically related purposes...

in either case, one has to comply with the 13485 requirements; and from a pharma context, its either an excipient/inactive ingredient or accessory. apart from the confidence and confirmation of process performance, its a requirement in either case, process validation is not optional, or good-to-have, but an essential tool.

I think the R&D people want to just do the testing on material produced during production scale development, when there was no approved validation protoco nor final batch records.

They wanted to declare the last development lots as potential product and only want the validation phase to consist of doing formal release testing on lots produced produced during development - this is before formal approved final product batch records that worked do into the Product File (Device File) are created...

Coming from Pharma by understanding is that the Validation stage of product development is about the whole process... In that once R&D develops a process, the validation phase costs of making the lots specifically to be validated specifically as part of the validation phase and not just testing lots made during the development of the manufacturing processes.

Their view is that that can take the lots created during development just do the formal release testing... and the validation protocol would be just about that.

this is a little contrasting because, during development phases, one would certainly expect to have more sampling than typical release testing; and at least equivalent if not more than the process validation sample plan.
we outdo the release testing and validation testing criteria, as the ad-hoc and random samples are determined in the interest of characterizing the process ( viz., endpoints, progress of reaction or process steps etc )
so depending upon the confidence of the product an process and data, it might well be used in PV ; but that version too could be pulled up in review/audit, unless it is not defined prior in the protocol and D&D plan etc

as you rightly pointed out, now it is an altogether different ball game, in the context of 21 CFR part 211, wherein one has to go through many steps for PV.

We are working to bring our first product to market... What has to be done tduring Validation phase?

Quality Management Systems - Process Validation Guidance
https://www.imdrf.org/sites/default...g3-n99-10-2004-qms-process-guidance-04010.pdf

 

[Chrisx]

I think the question is not clear to me. There are two types of validations required in ISO 13485. There is design validation and process validation. Design validation demonstrates the product can meet the specified application or intended use (clause 7.3.7). I suspect you are referring to design validation. Design validation has to be per planned arrangements (i.e protocol). The protocol is going to have to define sample sizes, products selected, tests to be performed and acceptance criteria. Acceptance criteria can't be established after everything is done for obvious reasons. The product has to be representative of actual production. The challenge with development runs is they may not be representative of actual production conditions. If the manufacturing processes were not under change control at the time of the development runs, it may be impossible to demonstrate the validation was performed on product representative of actual production conditions. If there have been changes to production, this may also make it not representative. Generally process validation has to be completed first because this establishes the parameters for the production processes. If the parameters for any special processes are still in flux, then it is difficult to establish that the product tested was representative. I think the key issue may be adequately documenting that the product tested is representative.