News

Below is a transcript of the comments of Jochen Salfeld, vice president, global biologics discovery, AbbVie, at the FDA Arthritis Advisory Committee Meeting on July 12, 2016.

My name is Jochen Salfeld. I am Vice President of Biologics Discovery at AbbVie. I led the team of scientists in the 90s that invented Humira.

Biosimilars may have minor differences in clinically inactive components.  But we are concerned that the structural differences between Humira and 501 are 1) not minor and 2) may not be clinically inactive, especially for inflammatory bowel disease.   

First, 501 has significant structural differences from Humira.  AbbVie tests every batch of Humira for many of the structural attributes you’re looking at today.  We do this because we believe that these attributes can impact the different and less-well-understood mechanisms of action for Humira.  Several batches of 501 fall entirely outside of our experience with Humira since launch. One example is galactose, which impacts the functioning of the Fc region and consequently the molecule's role in inflammation.  In fact, AbbVie has rejected potential manufacturing modifications resulting in smaller structural differences than those you’ve seen today because we are concerned about the potential consequences for our patients.   

Second, the structural differences identified today may be clinically relevant—particularly for IBD.  The way Humira works in IBD is not well understood.  But there are clearly additional mechanisms leading to efficacy for Humira in IBD beyond TNF binding.  For example, full performance in IBD likely requires the antibody to bridge multiple immune cells, using both the antigen binding and Fc regions simultaneously.    

Consequently, an anti-TNF’s sugar profile is critical in determining the performance in complex diseases like IBD.  This profile includes galactose, which I already mentioned, and sialic acid.  All batches of 501 fall significantly outside the sialic acid quality ranges for Humira identified by Amgen.  Sialic acid has been demonstrated to play a role in antibody function beyond Fc binding.  Amgen has not tested this established functioning. 

Trying to minimize the structural differences, Amgen has relied upon other functional assays.  These assays may not be sensitive enough to identify the impacts of structural differences and may not fully capture the complex mechanisms in IBD.  We are also concerned about the robustness of some of these assays.  Most of Amgen’s assays use non-human, engineered cell lines that do not reflect the complexity of human immune cells.  Further, some key functional assays do not meaningfully compare 501 to Humira—using as few as 3 samples.  Which samples were they?  Those inside or outside the Humira quality range?        

AbbVie is not suggesting clinical investigation in every indication.  But there is:

1. uncertainty created by significant structural differences;
2. uncertainty regarding the exact mechanisms of action in IBD; and
3. uncertainty regarding the functional assessments relevant to IBD. 

 
Therefore, we respectfully ask: 
Is there sufficient scientific justification that 501 will perform like Humira in IBD, where they have no clinical data in IBD? 

Thank you.

For further information, please contact Alissa Bolton at alissa.bolton@abbvie.com.