Biology Reference
In-Depth Information
using large volumes of serum because of their presumed low affinity for EPO.
These antibodies have not been carefully characterized and neither their affin-
ity for EPO nor the part of the molecule recognized by these antibodies have
been reported. We have tested much serum from asymptomatic rHuEPO-treat-
ed patients by immunoprecipitation of
125
I-EPO and we have never detected
such antibodies. Since this assay required extensive washing of the immuno-
precipitates, we cannot exclude that low-affinity antibodies with a high disso-
ciation rate could have been lost during the washing procedure in our assays
[10]. Interestingly, Pazur et al. reported the induction of antibodies against the
carbohydrate chains of rHuEPO in rHuEPO-immunized rabbits [18]. Neither
the affinity of these antibodies for EPO nor their ability to neutralize the bio-
logical activity of EPO have been reported. Since glycosylated carbohydrates
are not involved in the binding or activation of EPOR, however, it is likely that
such antibodies did not inhibit EPO biological activity. Moreover, since
rHuEPO and endogenous EPO differ at the level of glycosylation, it is tempt-
ing to speculate that the low-affinity antibodies detected with high frequency
by Castelli et al. could be directed against the glycosylated carbohydrate moi-
ety of rHuEPO. Whether these antibodies, if any, decrease the efficiency of
rHuEPO treatment remains to be tested, but the low number of patients with
neutralizing antibodies reported up to 1998 make rHuEPO one of the most
successful therapeutic application of recombinant DNA technology.
The situation suddenly changed after 1998. At the time of writing this chap-
ter, more than 160 cases of pure red cell aplasia have been reported in patients
treated with rHuEPO for anemia associated with chronic renal failure.
Moreover, a study from the Food and Drug Administration (FDA) revealed an
exponential growth of these cases during the studied period (July 1997 to
December 2001) [19]. We have had the opportunity to assay for the presence
of anti-EPO antibodies in several of these cases of pure red cell aplasia, and
we found that most were due to anti-EPO neutralizing antibodies [3]. Most of
these antibodies were very similar to those that we have previously described
in a patient who never received rHuEPO [8] with high titer (EPO-binding
capacities ranging from three to 125 EPO units/mL serum) and high affinity
(Kd ranging from 90 to 400 pM). The antibodies recognized the proteinic part
of the molecule (Fig. 3). Except for one patient, denatured EPO molecules
were no longer recognized by these antibodies, demonstrating that they were
directed against conformational epitopes. Most of these antibodies were
developed in patients treated with Eprex brand of epoetin alfa (see below), but
the antibodies recognized all forms of rHuEPO, as well as the erythropoietic
protein darbepoetin alfa, which is not a form of rHuEPO.
After evidence of neutralizing antibodies, rHuEPO was immediately with-
drawn and the patients received red blood cell transfusions. Depending on their
general status, most patients also received an immunosuppressive therapy
(intravenous immunoglobulins, corticosteroids, cyclosporine, or cyclophos-
phamides). The antibody titers decreased at various rates. In some patients, the
antibodies are no longer detectable and these patients have recovered some red
