Therapeutic Product Immunogenicity Community

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

This is the most contentious scientific discussion in the field and likely to stimulate quite a bit of discussion. I try to think about this as a difference between analytical validation (what is our analytical sensitivity) and clinical validation (what thresholds are meaningful for safety and efficacy). I think that we can all agree that for low risk of safety, that transient responses generally less critical.

The caveat here is that you said that they are high risk. If that is a high risk for potential safety consequences if seroconversion occurs, then I would argue that you do want to understand the nature of transient responses, particularly in the context of repeat doses and/or drug holidays. I'd also monitor kinetics of seroconversion, both transient and persistent, to understand what might happen in a commercial rollout and on what time line,

If only a high risk for seroconversion but no safety consequences if a patient becomes seropositive, then I am more in agreement with your statement that catching lots of transient responses may not be value added. That can be addressed in the way you present individual study data and integrated summary of immunogenicity. For example, you could plan the analysis to look for a titer threshold for clinical impact, run statistics for all treatment-emergent and for all persistent treatment-emergent to see if only the latter matters, and many other ideas that I am sure others will chime in.

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

This is the most contentious scientific discussion in the field and likely to stimulate quite a bit of discussion. I try to think about this as a difference between analytical validation (what is our analytical sensitivity) and clinical validation (what thresholds are meaningful for safety and efficacy). I think that we can all agree that for low risk of safety, that transient responses generally less critical.

The caveat here is that you said that they are high risk. If that is a high risk for potential safety consequences if seroconversion occurs, then I would argue that you do want to understand the nature of transient responses, particularly in the context of repeat doses and/or drug holidays. I'd also monitor kinetics of seroconversion, both transient and persistent, to understand what might happen in a commercial rollout and on what time line,

If only a high risk for seroconversion but no safety consequences if a patient becomes seropositive, then I am more in agreement with your statement that catching lots of transient responses may not be value added. That can be addressed in the way you present individual study data and integrated summary of immunogenicity. For example, you could plan the analysis to look for a titer threshold for clinical impact, run statistics for all treatment-emergent and for all persistent treatment-emergent to see if only the latter matters, and many other ideas that I am sure others will chime in.

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

This is the most contentious scientific discussion in the field and likely to stimulate quite a bit of discussion. I try to think about this as a difference between analytical validation (what is our analytical sensitivity) and clinical validation (what thresholds are meaningful for safety and efficacy). I think that we can all agree that for low risk of safety, that transient responses generally less critical.

The caveat here is that you said that they are high risk. If that is a high risk for potential safety consequences if seroconversion occurs, then I would argue that you do want to understand the nature of transient responses, particularly in the context of repeat doses and/or drug holidays. I'd also monitor kinetics of seroconversion, both transient and persistent, to understand what might happen in a commercial rollout and on what time line,

If only a high risk for seroconversion but no safety consequences if a patient becomes seropositive, then I am more in agreement with your statement that catching lots of transient responses may not be value added. That can be addressed in the way you present individual study data and integrated summary of immunogenicity. For example, you could plan the analysis to look for a titer threshold for clinical impact, run statistics for all treatment-emergent and for all persistent treatment-emergent to see if only the latter matters, and many other ideas that I am sure others will chime in.

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

This is the most contentious scientific discussion in the field and likely to stimulate quite a bit of discussion. I try to think about this as a difference between analytical validation (what is our analytical sensitivity) and clinical validation (what thresholds are meaningful for safety and efficacy). I think that we can all agree that for low risk of safety, that transient responses generally less critical.

The caveat here is that you said that they are high risk. If that is a high risk for potential safety consequences if seroconversion occurs, then I would argue that you do want to understand the nature of transient responses, particularly in the context of repeat doses and/or drug holidays. I'd also monitor kinetics of seroconversion, both transient and persistent, to understand what might happen in a commercial rollout and on what time line,

If only a high risk for seroconversion but no safety consequences if a patient becomes seropositive, then I am more in agreement with your statement that catching lots of transient responses may not be value added. That can be addressed in the way you present individual study data and integrated summary of immunogenicity. For example, you could plan the analysis to look for a titer threshold for clinical impact, run statistics for all treatment-emergent and for all persistent treatment-emergent to see if only the latter matters, and many other ideas that I am sure others will chime in.

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.

Dear network

Thank you for all the replies that this have triggered. My observation actually comes from the agency review of MOUNJARO (Tirzepatide), a dual glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide 1 (GLP-1) receptor agonist from Ely Lilly. Review

Here is what the agency commented during the review of the application: 

The sponsor categorized ADA+ samples of a patient as treatment-induced if a patient has at least one postbaseline result of ADA with a titer ≥2xMRD (1:20) and tested ADA-negative for preexisting antibodies. Any sample with %inhibition greater than or equal to the CCP were reported as ADA 'detected' and were then tested for ADA titer since titer is expected when ADA assay result is 'detected'. Also, the titer above the MRD was used as a method to determine the ADA status, which seems rational to define treatment-induced ADAs. It appears that the sponsor did not report patients who were tested ADA+ with ADA titer at MRD in Table APP.2.4. This report includes ADA+ patients who had an ADA titer of at least 2-fold greater than MRD (referred to as 'Treatment Emergent ADA or (TE-ADA).

kindly, Per

Dear community

I have been digging into multiple dossiers and OPD reviews of peptides with a theoretical high risk profile. Nevertheless, it appears that this can go hand in hand with increasing the bar and only consider binding antibody positive IF confirmed samples can be diluted at least 2x compared to MRD. It makes perfect sense to me, and the agency only commented on this and had no apparent objections. When that said, they wanted the sponsor to use a more conservative CP than higher study specific cut point. We are spending too much time on transient responses to drug candidates..........any comment/thought are highly appreciated.