Regulatory Open Forum

 View Only
  • 1.  Clinical strategy: precedent of accelerated approval for combination drugs

    This message was posted by a user wishing to remain anonymous
    Posted 20-Dec-2021 10:41
    This message was posted by a user wishing to remain anonymous

    We have two NMEs in the clinic: compound A and compound B.  Compound A has shown very preliminary efficacy for oncology indication Z, while B has not been tested for that indication.

    Preclinical studies suggest that A has efficacy for indication Z, B has almost no activity, while the A+B combo, or A+C combo (where C is an approved medication) work better than A alone.

    Indication Z has high unmet medical need, with no approved therapies.  However, there are several single-arm clinical studies from academic groups that show some activities for this disease using marketed drugs, including drug C.  These medications are being used to some extent in clinical practice (off label).

    If we use the A+C combo, we will almost certainly need to do a randomized trial to show A+C is advantageous to C alone.  If we use the A+B combo, we might have a shot on accelerated approval (AA) based on single-arm ORR.

    • We are unable to find regulatory precedents of AA for a combo of two NMEs. If you know of one, please point us to that example.  If precedent does not exist, is this a risky or unrealistic strategy ?
    • When the agency evaluates eligibility for AA, are "available therapies" limited to those that have been approved (e.g. drug C is not considered a treatment option for that indication by the FDA) ?


  • 2.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    Posted 20-Dec-2021 12:24
    Your assumptions are not exactly right in oncology drug development for the so called "unmet medical need" as you state for indication Z.
    Reach out to me privately if you are interested! Just an example, you have not factored the choice of endpoint in your assumptions! There are others but this is not a question that has a text book style response or yes or no answer. I'd consider this as an oncology drug development strategic question requiring RA strategist perspective!

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



  • 3.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    Posted 21-Dec-2021 08:54
    An accelerated approval for A+B is theoretically possible.  One precedent is the combination of Mekinist and Tafinlar for BRAF-mutated melanoma, although this accelerated approval was supported by a randomized trial.  However, an accelerated approval for A+B based on a single arm trial would be difficult as this study would not be supported by contribution of components data, i.e., what is the contribution of A and what is the contribution of B to the combined effect of A+B.  Thus this data would need to come from a separate trial if you were to conduct a single arm trial with A+B.

    Also, FDA defines available therapy in their "Expedited Programs for Serious Conditions - Drugs and Biologics" guidance as:

    - Is approved or licensed in the United States for the same indication being considered for
    the new drug, and
    - Is relevant to current U.S. standard of care (SOC) for the indication


    ------------------------------
    Dan Mannix
    South Haven MI
    United States
    ------------------------------



  • 4.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    This message was posted by a user wishing to remain anonymous
    Posted 23-Dec-2021 11:01
    This message was posted by a user wishing to remain anonymous

    Dear Narayan,

        Thank you so much for the comments.  For regular approval,  PFS or OS is often accepted as primary endpoints.  For AA, I believe ORR is rather common.  Please let me know if you disagree with the above statement.  It will be great if you could also elaborate on your comments on unmet medical needs.  

    Kind regards

    OP


  • 5.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    This message was posted by a user wishing to remain anonymous
    Posted 23-Dec-2021 11:01
    This message was posted by a user wishing to remain anonymous

    Hi Dan

        Thank you so much for your comments.  
        With regard to the AA of Mekinist and Taflinlar combo, there was a clear numerical difference in ORR between the combo and mono, but not statistically significant by investigator assessment.  Neither ORR nor mDoR was significantly different from assessment by the the blinded independent radiologic review committee.  
         In our case, ideally we want to run a 3-arm trial: A, B, A+B.  However, indication Z is a rather rare condition, so the trial will not be easy to conduct.  Does the agency still absolutely require component data for rare conditions ?
         If hypothetically, we ran such a 3-arm trial, and the ORR was 40%, 5%, 55%, and the difference between A and A+B was not statistically significant.  Is the numerical difference sufficient for AA consideration ?
          Thank you again for your insights

    OP




  • 6.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    Posted 28-Dec-2021 11:35
    The Mekinist-Tafinlar example does show that an accelerated approval can be obtained without statistical significance being demonstrated, as long as the data are clinically compelling.  In this example, the data are clinically compelling, as not only was the ORR improved in the combination vs. Tafinlar alone, but so were the # of complete responses and the median duration of response. For your situation, you could still demonstrate contribution of components with an A+B vs. A study (especially if the expected ORR with B was around only 5%).  The one question is whether there is a SOC for this rare condition.  If so, then a SOC arm might be needed.​

    ------------------------------
    Dan Mannix, PhD
    Washington, DC
    ------------------------------



  • 7.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    Posted 28-Dec-2021 12:34
    According to the OP question, "We have two NMEs in the clinic: compound A and compound B"

    You are suggesting a trial design A+B vs A???

    A is a NME also!!! I don't think any IRB would approved this design let alone the FDA.

    Rather I would suggest, A+B versus BOC or A+C (if literature supports its unlabeled use due to unavailability of approved drugs )  Versus BOC (Best of care or investigator's choice of care) only if the rare condition has no other treatment options. 

    To the original poster: Your thoughts on endpoints are fine in general terms but it may vary depending on hematological versus general solid tumors versus targeted solid tumors etc...


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



  • 8.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    Posted 28-Dec-2021 12:53
    My point was that it didn't have to be a "full" factorial design if it is a case where one of the agents has only a modest effect on ORR (as the poster said in the example given that one of agents might only induce an ORR of 5%).  I have seen FDA agree to omitting one of the monotherapy arms if it has been shown already that it has only a modest effect.  While both agents are NMEs, the need for a SOC arm is certainly a question, but if there is no SOC in this rare tumor type (or in the particular line of therapy for this tumor type), then an argument could be made to not include this additional arm.  Not having all of the details, it is hard to estimate what is the probability of FDA accepting an A+B vs. A design (with regards to including a B monotherapy arm and a SOC arm).

    ------------------------------
    Dan Mannix, PhD
    Washington, DC
    ------------------------------



  • 9.  RE: Clinical strategy: precedent of accelerated approval for combination drugs

    Posted 28-Dec-2021 13:10
    My point is your proposed trial design of A (NME unapproved) +B (NME unapproved) vs A (NME unapproved) will not be acceptable! Sorry If I did not make my point clear! Even if this trial is somehow allowed, the recruitment will be a huge challenge

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