Regulatory Open Forum

 View Only
Expand all | Collapse all

NDA approval of two API suppliers

  • 1.  NDA approval of two API suppliers

    This message was posted by a user wishing to remain anonymous
    Posted 03-Jun-2022 09:46
    This message was posted by a user wishing to remain anonymous

    We are entering Phase 3 and will use drug substance from supplier A to manufacture tablets for the clinical studies. We are considering qualification of a second API supplier and submit both suppliers in the NDA.  A tech transfer of the process with be performed from supplier A to supplier B so both will use same process and controls and same/similar equipment

    1) For validation purposes, does each API site need to validate independently (i.e. 3 batches at each site)?  Or could we perform a "co-validation", for example make 2 batches at site A and 1 batch at site B since same process?

    2) Can the comparability between the two API sites be demonstrated at the API level? Or would we need to make tablets with API from site B and compare tablets used in the clinical studies using API from site A? I know in the past when I've submitted sNDA's post approval to qualify a second API supplier we did make tablets and include release testing results on the tablets, but I don't if that is a requirement vs just being conservative.

    We are focusing on US market for now, but could pursue ex-US approvals afterward.

    Thank you


  • 2.  RE: NDA approval of two API suppliers

    Posted 04-Jun-2022 07:26
    In answer to your questions about 2 API suppliers: 1) Each API supplier must validate their own processes as they are 2 totally different facilities, equipment , personnel, procedures, etc.  There is no process for "co-validation".
    2) Comparability is generally made at the API (physical and chemistry) level.  However, there should be drug product test runs with the second API to show that the tablets are comparable.  Unless requested by the Center, the second API would not need to be used in clinical studies.  

    Another piece of advice.  Submitting the original NDA with 2 API suppliers presents a risk in that should one supplier fail a PAI, the entire NDA will receive a Complete Response Letter.  The more facilities/sites in an NDA, the greater the chance on will fail. Many companies submit one supplier in the NDA, and do all the prep for the second supplier (validation, comparability, product test runs, etc.), and then immediately upon NDA approval submit an sNDA for the second supplier.  This reduces PAI-failure risk.   ​

    ------------------------------
    Peter Smith
    Principal
    Smith GMP Consulting
    Narragansett, Rhode Island
    USA
    ------------------------------



  • 3.  RE: NDA approval of two API suppliers

    Posted 06-Jun-2022 21:56

    Please remember that the NDA is about your commercial supply and supplier and therefore should focus on your commercial manufacturer and capability to manufacture the product as identified in Module 3 for approval.  Of course the commercial supplier must perform process validation, typically clinical trial manufacturers don't conduct a process validation as expected for NDA review and approval. It will be your commercial supplier that will be inspected as part of PAI as stated by Peter, therefore the commercial manufacturer will have to demonstrate readiness which should include some data.  In addition, you will need to place the commercial material on stability to establish expiry dating of the commercial material unless you get agreement from FDA during your pre-NDA that your clinical process and commercial process are "the same" and therefore you can use the clinical supply data as "commercial" material. Your clinical supplier is not considered a second supplier unless you plan to supply commercial product with it, then you have two submit two DP sections in Module 3 which does create risk.  You should submit any compatibility protocol to the FDA or discuss your strategy at the pre-NDA meeting. 

    Before you transfer the clinical process to the commercial supplier and to define your pre-NDA meeting strategy, conduct a risk assessment to ensure that you have captured any potential differences and possible risks between the two manufacturer's  process. In my experience no two manufacturer's conduct the process "the same" or are "identical" there are always differences that need to be considered and mitigated if needed.  Always have some data when you discuss this type of strategy with FDA, that is some feasibility data or a feasibility batch and supportive stability from the "commercial supplier" to show FDA some "sameness". 

    As suggested by Peter, you want to minimize any risk to your NDA approval and to meet approval in 10 months unless a priority review.  Most importantly, as I indicated above, the FDA is interested in your commercial process and supplier, make that the focus of your NDA. Your clinical supplier and process is in the IND. 

    Good luck,

    Dar



    ------------------------------
    Darlene Rosario MBA, RAC
    Principle Consultant
    Ventura CA
    United States
    ------------------------------



  • 4.  RE: NDA approval of two API suppliers

    This message was posted by a user wishing to remain anonymous
    Posted 07-Jun-2022 09:09
    This message was posted by a user wishing to remain anonymous

    Sorry, you (i.e  Principle consultant) never answered the questions raised by the OP! Please next time read the questions carefully and let your responses be specific to the questions raised! Peter's response and the Anon's response is more specific and valuable!

    Thanks!



  • 5.  RE: NDA approval of two API suppliers

    Posted 07-Jun-2022 12:38
    Dear Anonymous,

    No need to be sorry! We all have a perspective and expectations for advise.  My strategy is always to provide additional input for the request so that the requester has more information to consider.  Often answering questions directly that have limited information may not lead to the ultimate "right" answer for the specific question or most appropriate strategy to the specific situation. We can't always know what the requestor is thinking or hasn't thought about, level of experience or challenges that are being faced. 

    Yes, Peter's response and Anon's response had some specific and valuable input, I included that in my response.  Hopefully in the future you will use this forum to provide insight rather than be critical about advise being provided. This is the best use of this forum.  

    Principle consultant, Dar Rosario

    ------------------------------
    Darlene Rosario MBA, RAC
    Principle Consultant
    Ventura CA
    United States
    ------------------------------



  • 6.  RE: NDA approval of two API suppliers

    Posted 08-Jun-2022 11:17
    I agree strongly about the value of sharing related information around the topic. These conversations are read and appreciated by many besides the original posters, including people with different levels of experience and different specific challenges.

    Also, I've seen that original posters are pretty good at posting followup questions if the responses don't get to the heart of what they were originally wondering about, whether the first post was objectively well-worded or not.

    ------------------------------
    Anne LeBlanc
    United States
    ------------------------------



  • 7.  RE: NDA approval of two API suppliers

    This message was posted by a user wishing to remain anonymous
    Posted 08-Jun-2022 13:22
    This message was posted by a user wishing to remain anonymous

    When additional information is provided there are chances of being confused and loose the focus of the primary question itself! Besides there are misinterpretation possibilities!
    For example the following statement:
    "unless you get agreement from FDA during your pre-NDA that your clinical process and commercial process are "the same" and therefore you can use the clinical supply data as "commercial" material. " 
    The above statement is not really accurate. First of all, one cannot get agreement at pre-NDA meeting without the FDA reviews the data of clinical process and commercial process are similar and that can happen only after the NDA submission is made! The standard FDA response at the pre-NDA meeting would be likely "the proposal seems fine bur we need to wait until your submission is made we review the data...." something along these lines!

    Second example: the following statement "Your clinical supplier is not considered a second supplier unless you plan to supply commercial product with it, then you have two submit two DP sections in Module 3 which does create risk.  You should submit any compatibility protocol to the FDA or discuss your strategy at the pre-NDA meeting.

    Again the above statement is not accurate and confusing to I am sure (to the OP) especially given that the OP was specific to the validation question from two API suppliers! Simply put, the OP asked can two suppliers (one used for clinical and other with similar process/equipment at different site) of API be included in the NDA without full validation of DP with new API supplier. secondly the comparability between two APIs be demonstrated at API level or require at DP level. It was very straight forward and specific to what the OP was looking for. Both Peter and the Anon made very specific responses indicating two separate validations require and also comparability need to be demonstrated at DP level also in addition to API level.

    When someone gives additional information it also comes across as though the questioners are totally naive and have no basic RA expertise and its kind of insulting to assume that!

    So I'd like to be respectable to all questioners who pose questions and assume they have certain basic information and secondly be specific  in responses to those questions! unnecessary information is not appropriate. People may disagree but everyone has their opinions! Another example is if someone says you need to have SOPs, is that mean the people who are asking questions on this forum are totally naive and they have no QA function in their organizations/ and they have to learn from this forum???

     



  • 8.  RE: NDA approval of two API suppliers

    This message was posted by a user wishing to remain anonymous
    Posted 06-Jun-2022 09:45
    This message was posted by a user wishing to remain anonymous

    I'll answer the 2nd question: FDA would certainly expect drug product be made with supplier B. It's also in the guidance somewhere.

    I am not certain about 1). FDA would expect validation independently and I don't think co-validation, as you propose, would be acceptable. My rationale is that its a new NDA (i.e. NCE/NBE) and not much production experience/stability etc.....


  • 9.  RE: NDA approval of two API suppliers

    Posted 09-Jun-2022 10:35
    Hi,

    For question 1), I concur with the answers already given (i.e., both sites must have a validated process to allow for commercial use of the manufactured API).

    For question 2), I would agree that in general comparability at API level should suffice, as you say; this may include stability data on at least one batch. I recommend referring to the excellent 2018 FDA draft guidance concerning post-approval changes of the drug substance manufacturing, as it provides useful recommendations on the science-based risk assessment that should be the base of the data package when major changes are implemented (including new manufacturing sites).  

    Regarding the point of including only one site in your initial NDA, to minimize PAI-related risks, I would counteract that has not been my experience. Unless you are using "problematic" CMOs, with a history of inspectional issues, or you are rushing things through, the risk for a failed PAI, hence a CRL, is extremely low (thinking of small molecules, obviously, since you are talking about tablets). In fact, it's much easier and far less cumbersome to include both sites in the first application, rather than dealing with time-consuming post-approval changes (which in this case could be a CBE-30 or a PAS). This applies to most regions ex-US, with  few exceptions. As a matter of fact, FDA strongly prefers to see at least two sites each, for DS and DP, because it helps reduce the risk of drug shortages, which is one of their biggest concerns nowadays.
    I have contributed to or seen multiple NDAs with at least two API manufacturing sites in the initial NDA, never experienced issues in any of the major territories because of that strategy.

    ------------------------------
    Maurizio Franzini
    Regulatory Affairs CMC, Director
    San Francisco CA
    United States
    ------------------------------



  • 10.  RE: NDA approval of two API suppliers

    Posted 09-Jun-2022 12:42
    I'd agree with Maurizio and for his precise responses (i.e. Thanks you) to the questions raised similar to Peter's and the initial Anon's response. However, I do agree with Peter of risking approval if the alternate API site is found to be deficient in some form then the entire NDA is in jeopardy! At the same time, my experience was that in this situation the FDA usually asks the sponsor to delete the problematic API site (i.e. before the PDUFA action date) so that the NDA could be approved if this is the only issue holding approval! In my experience we quickly submitted an amendment deleting the problematic API supplier and the NDA was approved before the PDUFA date. Basically this is another line of thinking if the OP wish to submit both API suppliers.
    Good luck!

    ------------------------------
    GRSAOnline
    ------------------------------



  • 11.  RE: NDA approval of two API suppliers

    This message was posted by a user wishing to remain anonymous
    Posted 10-Jun-2022 15:41
    This message was posted by a user wishing to remain anonymous

    Thanks to everyone who replied. Appreciate the feedback.