FUSION PROTEINS WITH TOXIC ACTIVITY - Fusion Protein Technologies for Biopharmaceuticals: Applications and Challenges - page 245

IIB TOXIN: CYTOTOXIC FUSION PROTEINS
17

FUSION PROTEINS WITH TOXIC ACTIVITY

S TEFAN R. S CHMIDT

Rentschler Biotechnologie GmbH, Laupheim, Germany

17.1 Introduction
17.2 Toxins
17.3 Immunocytokines
17.4 Human enzymes
17.5 Apoptosis induction
17.6 Fc-based toxicity
17.7 Peptide-based toxicity
17.8 Conclusions and future perspectives
References

A number of mechanisms are available to destroy cells in
a targeted therapy. It can be distinguished between mole-
cules that have to be internalized and those that work
directly at the cell surface or through interactions with other
cells or extracellular molecules (Figure 17.1). Functional
screening with an antibody fragment library can be applied
to identify internalizing surface antigens [2].
Two parameters are important for the binding molecule:
the affinity and the ability to cross-link, since cross-linking
triggers receptor-mediated internalization. Therefore, scFv
might not be the best choice if internalization of a mono-
meric toxin is required. Often a high affinity is not desirable
because a binding site barrier effect may arise. This means
high affinity ligands bind extremely tight to the first binding
sites they reach, thus creating a zone with blocked access
that limits further penetration. Affinities above 10 7 M are
required for sufficient tumor uptake. Saturation can be
observed around 10 9 M, but at 10 11 M, a model single-
chain (scFv) antibody traversed only 2-3 cell diameters [3].
Overall, the ideal cytotoxic fusion protein has to be opti-
mized for half-life and tissue penetration ability, parameters
that can often be addressed through fine-tuning the hydro-
dynamic radius of the molecule (see Chapter 6). Usually, a
relatively short half-life is desirable to minimize off-target
effects that can increase by prolonged exposure. In general,
three strategies have been employed: improving affinity,
decreasing size, or increasing doses. Since most of the natural
toxins have intracellular activity, an efficient passage from the
surface to the cytosol must be established as well. Frequently,
endosomal escape strategies are required [4].
Besides the direct elimination of malignant cells through
toxic molecules, a similar effect can be achieved indirectly
through the attraction and stimulation of immune cells. This
requires refined strategies since many tumor mechanisms
evolved to escape the immune response of the host. Much

17.1 INTRODUCTION

Targeted therapy was first defined in the second half of the
nineteenth century when Paul Ehrlich described his concept
of a “magic bullet” that selectively targets a disease-causing
organism and delivers a toxin to eliminate the pathogen. This
paradigm is still an ideal many scientists strive after, par-
ticularly to treat cancer without affecting normal tissue [1].
Initially, the targeting moiety and the toxin were chemically
conjugated, but this was causing manufacturing problems.
Other difficulties were the often-unacceptable high off-
target effects or immunogenicity due to toxins of nonhuman
origin. The existence of only a single approved immuno-
toxin, Ontak 1 just underlines the many hurdles that must be
overcome. Since its market introduction in 1999 no other
toxin-based fusion protein was accepted by regulatory
authorities, despite the many clinical trials that have been
ongoing during the last 20 years. Currently, selective admin-
istration, for instance, into bladder by catheters, direct
injection into brain, or application under positive pressure,
is tried to minimize systemic toxicity. In addition, genetic
engineering of immunotoxins addresses these issues.