If you cannot find an answer to your question please complete the contact form.
As of August 2024 I have successfully humanized 161 antibodies. The species breakdown is as follows:
As of May 2023, I have worked on 2 failed humanization projects out of 163 (98.8% success rate). In both cases the affinity of the best humanized antibody was within approximately 3-fold of the parental non-human antibody but this did not meet the customers requirements. In both cases the client was provided with a full refund.
This is a difficult question to answer. As a service provider and consultant all the humanization work I have done has been a simple fee-for-service with no royalty and milestone payments due. This means there is no obligation for clients to update on progress of their molecule. Often antibodies are also provided with no information on clone name or target antigen, again making tracking of such antibodies very challenging.
That being said, I am aware of at least 3 antibodies I have humanized that have reached the clinic. The real number may well but much higher than this.
Really the question is why do service providers get away with charging so much and taking so long! Humanization is an old and very well understood process yet it is often shrouded in mystery with confusing terms only understood by those in the know. For outsiders this makes it appear more complex than it really should be. The in silico sequence design step, which is what I provide a service in, can be performed relatively quickly with a few simple algorithms or even with a pen and paper if you want to be old fashioned about it. As an independent consultant I also have no real overheads and so can provide a low cost option for humanization of your antibody sequence in under 48 hours of receipt of a confirmed order (and often within 24 hours).
This depends on your antibody sequence and how many I believe need to be made to give a high probability of success. Typically this will be a minimum of 4 VH sequences and 4 VL, which is a total of 16 possible humanized antibodies. It could be as many as 25 though. Of course this can all be discussed prior to an order and a custom service offering provided to meet your needs.
This is very much dependent on the starting antibody sequence, its homology to human germlines and how many back-mutations are required to maintain activity. I do not set a specific percent identity to human germline that has to be achieved and neither should you. The idea that a certain "humanness score" needs to be achieved is redundant these days and had no basis in science anyway.
Prior to 2017 when antibodies were assigned an international non-proprietary name (INN) they consisted of a prefix (random letters), infix (indicating target and source) and suffix (-mab for all antibodies). The source substem could be -xi- for chimeric antibodies (human constant domains and non-human variable domains), -zu- for humanized antibodies or -u- for human antibodies. An arbitrary threshold of 85% identity to a human germline sequence was set as the cut-off between chimeric and humanized antibodies. This number had no scientific basis and infact many existing antibodies that had undergone humanization and been approved did not meet this criteria.
Due to the increasing complexity of technologies and difficulties in distinguishing humanized and human antibodies this system was abandoned in 2017 and the source substem was removed. Since then the name of the antibody does not reflect whether it is chimeric, humanized or human and so from this point of view there is no requirement to reach a certain percent identity. There is also no requirement from regulatory authorities for this as many chimeric antibodies have been approved and been shown to be perfectly safe.
As a further note, the INN system has been updated again in 2021. The -mab suffix has now been dropped and replaced with new suffixes that separate antibodies into 4 groups:
-tug for unmodified immunoglobulins
-bart for artificial antibodies containing any point mutations
-mig for multi-specific immunoglobulins
-ment for antibody fragments
For more information on the changes to the INN naming system over the years I recommend reading the following articles:
1. Jones, T. D. et al. The INNs and outs of antibody nonproprietary names. mAbs 8, 1–9 (2016).
I use a subset of human germline sequences rather than mature sequences. This tends to be the most common approach due to the belief that mature sequences can theoretically lead to an increased risk of immunogenicity. Germline sequences have also been shown to have greater plasticity, which in theory makes them more likely to successfully accept new CDR sequences. The germline sequence(s) to be used in the humanization process are then selected based on homology to the non-human antibody.
Parental and humanized sequences are assessed for: N-linked glycosylation sites, free Cysteines, and motifs for isomerization, deamidation, oxidation and cleavage. Any such sites will be highlighted to clients, especially if they occur in CDR regions. Whether or not they are removed during the humanization process will depend on their location. This will be discussed with the client at the time and an appropriate plan formed, such as the design of variants with and without the liability.
The client owns all IP relating to the original antibody and all humanized variants. This is a simple fee-for-service and I claim no IP and ask for no milestone or royalty payments.
Yes. In theory I could humanize any number of antibodies in parallel. Depending on the number of antibodies on a project this could lengthen the timelines. For bulk orders a discount could also be offered to reduce the humanization cost per antibody.
It is very simple. If at least on of the humanized variants does not meet your requirements I will prepare additional sequence variants at no extra cost and/or provide a full refund. Typically success would be defined as a binding affinity within 2-fold of the original antibody in an appropriate assay but this is antibody and client dependent.
Some antibodies are more challenging to humanize than others. This is true for all service providers and sometimes you have to go through multiple rounds of variants to find the critical back mutation(s) that are required to maintain full activity. If you have an antibody that has failed humanization previously I would be very happy to take a look at all the data (i.e. sequences and related activities) to see if I can recommend alternative variants to generate.
I no longer have a lab and so my service simply provides humanized sequences that would then need to be produced in your own lab or through a service provider. I do have contacts with many service providers and would be very happy to make introductions. Depending on the service provider it can cost as little as $300-500 per antibody to produce a small amount of purified antibody for analysis.
I am based in the United Kingdom but if required I am happy to try and arrange video calls at a time that is convenient for you.
All invoices will be in British pounds (GBP) as my bank account cannot accept other currencies. This does not prevent me from working with clients across the world as I have done for many years.
An appropriate currency conversion will be applied but the standard humanization service price is equivalent to 2000 USD per antibody. The project will be invoiced on completion and payment is due within 30 days unless agreed otherwise in advance. Payments to be made by bank transfer only.
No. This is a straight forward humanization service where the aim is to maintain affinity but not necessarily enhance it. I do have connections with companies that may be able to provide affinity maturation services but these are substantially more expensive than humanization.
No. I firmly believe that deimmunization by any of the currently offered services is a waste of time and money. Please see the 'debunking deimmunization' section at the bottom of the demystifying humanization page.
Yes, through my consulting company mAbvice. I provided consulting services in antibody sequencing, expression, engineering and other related topics. For more information please see www.mabvice.com.
Copyright © 2024 Antibody humanization - All Rights Reserved.
Powered by GoDaddy