• cancer
• mass-spectrometry
• proteomics
• antibody
• targeted therapy

The experimental activities have been performed in the Philochem laboratories.
Philochem is a fully owned daughter company of Philogen and represents the
Discovery Unit of the Philogen group.
In this thesis project we set up a mass spectrometry-based (MS) method that allows us
to quantify the abundance of antibody-based therapeutics in tumor and healthy tissues
after administration in a mouse model. These in vivo biodistribution experiments are
necessary to evaluate the selective accumulation of our molecules of interest at the
tumor site. This targeted tumor delivery should prevent their off-target effects and
related toxicity. In the company, these experiments are currently performed using
radiolabelled surrogates, however this procedure does require a stable labelled antibody
and the expensive use of facilities that allow to handle mice injected with radioactive
compounds.
The aim of the thesis was to achieve MS-based quantification results as accurate as
radioactivity-based ones. This non-radioactive analytical method will enable
company’s scientists to screen for different antibody-based products without, or with a
reduced use of radioactivity. As starting model for our analysis, we have chosen full
immunoglobulins G (IgGs) and IgG fragments such as the tandem diabody format.
The goal of our experimental workflow was to extract proteins from tissue samples
using mild lysis conditions to avoid conformational changes in the analyte (our IgG or
diabody of interest). Later, an enrichment step was performed using Protein A
(Staphylococcus aureus). Finally, samples were digested with trypsin and resulting
peptides analysed by mass spectrometry. To compensate variability introduced by the
sample preparation, we decided to add an internal standard (IS). As IS we decided to
use a different biotherapeutic in the same format and with low sequence homology to
our target protein. In specific, we used a monoclonal antibody carrying a kappa light
chain as IS for an analyte with lambda light chain and vice versa.
To optimize our methodology, we firstly performed experiments in which we “spikedin”
different amounts of our antibodies in “blank” tissues to assess if the signal detected
by the MS was proportional to the amount of protein injected. The results of these
experiments showed good linearity between injected analytes and measured signals for
all antibody-formats that were tested. For the tandem diabody format, the MS signal
detected was lower when compared to the IgGs signal. This was probably due to the
fact that the affinity of the tandem diabody for protein A is lower than the affinity of an
IgG. For this reason, we also started investigating alternative enrichment protocols.
Then, encouraged by these positive results we moved to in-vivo experiments. To
validate our MS measurements and to benchmark our results with radioactivity-based
quantifications, we choose well characterized antibody (L19) of our company for which
radioactive data were available. The antibody biodistributions were measured at two
different time point (24h and 48h after IgG administration) and three mice per each time
point were analysed. The biodistribution data that we obtain showed a selective
targeting at the tumor site of the antibody. These results were further validated by
comparing our MS biodistribution data with radioactive data.