Therapeutic Product Immunogenicity Community

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------

Hi Johanna,

                I understand if you are unable to share more information about this study on a public forum, but it would be good to know more details such as the sample size, frequency of administration, and the size of database used to determine immunogenicity of the drug in previous studies. Is Day 15 the first of several ADA timepoints or the only one proposed for the study? If there are more ADA timepoints later on, then dropping Day 15 should be easy to justify. In my experience, it's very unusual to see any impact of ADA 15 days after the first dose. If Day 15 is the only timepoint proposed for ADA assessment, then taking samples from a subset of patients requires needs a strong scientific justification based on risk analysis. You mentioned low ADA incidence and impact, so I assume the immunogenicity risk is also low, but I don't know anything about your drug or your patients. Perhaps you could answer your question by statistical methods: to estimate ADA incidence of x% with ≥ y% confidence N subjects are needed. Or it could be restated like this: how many subjects are needed to confirm with ≥ y% confidence that ADA incidence is the same or lower than that established previously? Unfortunately, the lower ADA incidence the more subjects you need to make confident statements about immunogenicity impact ☹.

                You may also be prepared for something unpleasant on the label like "actual incidence of ADA may be unreported due to inadequate sampling", but we make medicines for patients and not for pretty prescribing labels so it may be risk worth taking.

Best regards,

Robert Kubiak

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------

Hi Robert,

Thanks for your thoughts. Yes, I can't share much. Here is what I can say. Risk is low, based on structure, MoA and patient factors.  FIH has no treatment induced ADAs, but assay does not have adequate drug tolerance for most timepoints, we will improve this for upcoming study. There are other planned collections: pre-dose and D29 (inclusive) onwards that follow industry best practices. I left that information out because I figured others would benefit from information on other programs where perhaps other collections were also dropped/removed. Based on other feedback, shared privately with me, it seems like a sound approach would be to remove D15 from the phase 2 protocol instead of assessing in some participants. We plan to ask the agency but have heard about several not asking and simply removing the D15 collections.

Hi Johanna,

                I understand if you are unable to share more information about this study on a public forum, but it would be good to know more details such as the sample size, frequency of administration, and the size of database used to determine immunogenicity of the drug in previous studies. Is Day 15 the first of several ADA timepoints or the only one proposed for the study? If there are more ADA timepoints later on, then dropping Day 15 should be easy to justify. In my experience, it's very unusual to see any impact of ADA 15 days after the first dose. If Day 15 is the only timepoint proposed for ADA assessment, then taking samples from a subset of patients requires needs a strong scientific justification based on risk analysis. You mentioned low ADA incidence and impact, so I assume the immunogenicity risk is also low, but I don't know anything about your drug or your patients. Perhaps you could answer your question by statistical methods: to estimate ADA incidence of x% with ≥ y% confidence N subjects are needed. Or it could be restated like this: how many subjects are needed to confirm with ≥ y% confidence that ADA incidence is the same or lower than that established previously? Unfortunately, the lower ADA incidence the more subjects you need to make confident statements about immunogenicity impact ☹.

                You may also be prepared for something unpleasant on the label like "actual incidence of ADA may be unreported due to inadequate sampling", but we make medicines for patients and not for pretty prescribing labels so it may be risk worth taking.

Best regards,

Robert Kubiak

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------

Hi Johanna,

We have example(s) of not only dropping time points, but some immunogencity analyses all together (e.g. Nab) for late-stage clinical studies where low incidence was observed (e.g. <5-10%). In addition to incidence, I think an important metric is clinical relevance. If there are no safety events associated with the ADA and no impact to PK/PD/efficacy of the drug, then there is quite strong rationale for minimizing and/or eliminating some (especially early timepoint) analyses. Hope that helps!

Kind regards,

Marie

Hi Immunogenicity Experts,

Do you have examples where you dropped a sample collection (e.g. D15) either in phase 3 or by amendment in an on-going phase 2 because of no ADA +s or low incidence with no impact? If the latter, then what was the low incidence value? equal or lower than 3%? 4%? again with no impact.  I am asking for a study were asking the participants to come back only for ADA (and corresponding PK) seems like a huge burden.

The idea is to keep the D15 for the first 100 participants and depending on incidence drop the collection, but I am unsure what is the highest incidence that we would still consider low enough (and no impact) to drop the collection.

Thanks so much for your feedback!!

Johanna

------------------------------
Johanna Mora Ph.D.
Bristol-Myers Squibb
Princeton NJ
[email protected]
------------------------------