Note: Descriptions are shown in the official language in which they were submitted.
r~
HOE 84/9 016
~he invention rela~es to a process for the pro-
duction of a pasteurized and purified preparation of
blood coayulation factor V}II o This preparation can be
used for the treatment of blood coagula~ion disturbances.
S ~lood coagulation is a complex process which takes
place via several reaction steps and which involves at
least 13 coagulation ~actors (F~ ~hich are identified
by roman numerals. The coagulation factors are predomi
nantly proteins and, in particular, proteins equipped
~ith the properties of proteases or accelerators~ The
only actual substra~e is fibrinogen ~hich, in cases of
injury, is converted by thrombin into its insoluble form,
fibrin, ~hich forms ~he primary wound closure. If one
of the 13 coagulation factors is missing, the formation
of thrombin and fibrin does not take place; the conse-
quence is hemorrhage. An instance of this is hemophilia A
which is the mos~ ~idespread disease with a tendency to
hemorrhage and which is due ~o a deficiency of factor
VIII. ~oth hemophilia A and 8 (F IX deficiency) can only
be effectively treated by replacement of the factor which
is lacking~
The problem of obtaining F VIII from human plasma
in good yield and high purity, as is necessary, in par-
ticular9 for the self-treatment of patients, still has
no opti~al solution. T-his particularly applies to pas-
teuri2ed F VIII concentrates which have increasingly
displaced the conventional commercial product because
it has been possible with them to eliminate the risk of
transmission of hepatitis.
The isolation of a highly purified F ~iII in good
yield is made difficult by, in particular, the fact that
although F YIII is enriched in the cryoprecipitate, it
is associated ~ith fibrinogen and fibronectin, which are
two proteins which have relatively high molecular weights
and are sparingly soluble and have similar physicochemical
properties to F VIII.
, , :. : ' ' ' ' '~
::
A factor VIII concentrate should be as pure
as possible~ that is to say free of undesired conco~itant
proteins and, in particular, of immunoglobulins, includ-
ing isoagglutinins; this is because there are ;ndications
that administration of non-specific proteins leads to
overstrain of the reticuloendothelial system (RES) and
to impairment of the ;mmune defenses manifes~ed by a
change in the composition of the lymphocy~e population
and of the immunoglobulins~ The significance of this
becomes more evident ~hen conn~cted with the fact that
hemophiliacs have to undergo life-long treat~en~ with
such F VI~I concentrates. This resuL~s in ~he demand
for a native, highly purified, pasteurized F VIII con-
centrate, i.e~ a product ~hich rules ou~ transmission
of hepa~itis viruses and other infectious material and
rules ou~ sensitization to isoantigens.
This invention relates to a product of this type
which rules out the transmission of hepatitis and is free
of isosgglutinins, and to a process for its production.
2û I~ has been found that a solution containing
F VIII but ~hich is virtually free of factors of the
prothrombin complex ~F II, VII, IX and X~ can be pasteur-
ized in a manner known per se after the addition of
stabilizers to protect against thermal inactivation, the
heated solution can be treated with an anion exchanger
in the pH range 5 - 6.5, and the adsorbed F VIII can
be washed free of concomitant pro~eins, especially fibrino-
gen~ fibronectin and immunoglobulins, including isoagglu-
tinins, eluted with a concentrated solution of Na, K or
Ca salts ~ith a halogen, and obtained by, for example,
precipitation from the eluate.
Ecteola-cellulose and an anion exchanger bearing
~AE ~quaternary aminoethyl) groups have already been
used for the purification of F VIII tvan Creveld, S. et
al.~ Thromb. Diath. Haem. t1961) VI, No. Z/3~ 282 and
Baugh, R. et al., Bioch. Biophys. Acta (1974) 371. 360).
However~ ~ransfer to pilot-plant or manufacturing scale
has not been possible. It has merely been possible to
use ion exchange chromatography for the final purifica-
.
~l~&~ 'a~
~ion of prefractionated material (Fay, Ph. J. et al.~Proc. ~atl. Acad. Sci. (1982) 79, 7200). Adsorption with
high specifi~ty takes place only at a pH around 5.5;
however, at this most of ~he proteins contained ;n the
cryoprecipitate~ in particular human fibrinogen, preci~
pi~ate ou~, especially on contact with the adsorbent
in the column~ Moreover, according to the papers quo
ted, relatively large amounts of adsorben~ are neces-
sary. This in turn appears to have had an enormous
effect on ehe yield, ~hich was tiny, apparenely o~ing
to non-spe~ific adsorption of F VIlI - i~s physiologi-
cal ~ask is, after all~ to adhere to non-physiological
surfaces - to the large amounts of adsorbent~ It was
also found that the ma~erial finally purified on a QAE
exchanger rapidly lost activity; for this reason, anion
exchangers have not been used for the manufacture of
F VIII.
However, 1~ has now been found~ surpris1ngly,
that chroma~ography on bas~c ion exchangers is p~rfectly
suitable for the manufacture of F VIII preparations.
It ~as surprising that, starting from a pasteurized cryo-
precipitate~ 1t is possible to obtain a highly purified
F YIII preparation in one step by using anion exchange
chromatography~
Thus the invention relates to a process for the
productîon of a factor VIII preparation, in ~hich a solu-
tion containing factor VIII is, in the presence of
stabilizers, pasteurized and purified, ~hich process
comprises the solution being treated wi~h ~n anion
exchanger ~hich 1s based on carbohydraees~ the exchanger
being ~ashed, and the ~ VIII being eluted~
In one embodiment of the invention, a process for
the production of a factor VIII preparation is provided which
comprises comprises treating a solution containing factor
VIII with an anion exchanger based on carbohydrates in the pH
range 5.0 to 6.5, washing the exchanger; eluting the factor
VIII with chaotropic solutions or CaC12, and pasteurizing the
sol.ution containing factor VIII in the presence of
stabil.izers prior to treating with the anion exchanger or
pasteurizing the eluted Eactor VIII in the presence.of stabilizers.
~e~
': ~ . '. ' ' .' :, ~ '
~: '
.: .
: ` : ' '': ~ : , .
5~3~.
- 4~ -
Suitable s~arting materials are a cryoprecipitate
obtained by the method of Pool et al., Nature (1964) 203
312, Çohn fract10n I (Minot, GoR~ et al. J. Cl;n. I -est.
(1945) Z4, 7047, plasma, F VIII :C-containing cell culture
~
medium and side fractions containing F VIII obtalned from
the latter~ Ho~ever, it is advantageous to use directly
cryoprecipitate or Cohn I fraction.
It is possible to use as the stabilizer, for
::
.
,,
P~
-- 5 --
protection against thermal inactivation of F VIII during
pasteurization, carbohydrates and amino acids, preferably
35~60 9 of sucrose per 100 9 of solution and 1-3 mol
of glycine per li~er of solution and~ whera appropriate,
calcium ions. The method can be that of, for example~
German Offenlegungsschrift Z,916~711 or 3,237,512.
Examples of suitable anion exchangers for the
adsorption of F VIII are exchangers bearing DEAE, QAE
or ecteola groups~ specifically those based on~ in par-
ticular, cellulose, sephadex or sepharose, but DEAE-
sepharose is preferred.
In contradistinction to van Creveld and Baugh
~see above) these exchangers are in fact suitable for
purification of F VIIIo but under conditions which have
not hitherto been disclosed and are described belo~.
The adsorption conditions have proved to be vital.
This is because it has been found, like~ise surprisingly,
that in a physiological saline medium, preferably at
pH 5.5, F VIII is substantially selectively bound to
DEAE-sepharose, whtle concomitant proteins, such as
fibrinogen and fibronectin, remain in the supernatant
~batch process) or pass through a column. Finally, it
has been foun~, again surprisingly, that a pasteurized
solu~ion of the cryoglobulins can in fact be chromato-
graphed a~ pH 5.5 in a physiological saline medium,since certain cryoglobulins precipitate under these condi-
tions, especially fibrinogen. The adsorption and, in
particular, the specificity with ~hich F VIII is bound
to the anion exchangers decrease greatly as neutrality
is approached. Ho~ever, under normal chromatography
conditions, DEAE is only loaded when the pH exceeds
7Ø Nevertheless, precipitation does not take place
in the process described here, because the carbohydrates
which have been present in the cryosolutions since the
~5 pasteurization keep the fibrinogen and cig in solution
in a slightly acid medium~
Thus an advantage of the process described is
that on adsorption of pasteurized cryoglobulin onto the
anion exchanger the fibrinogen and fibronectin remain
. '` ~ . ..
.
-- 6
in the supernatan~ or effluent and can be obtained from
them as pasteurized products. An example of a method for
fibronectin is in German Offenlegungsschrift 2,8480529.
The chromatography on, for example, DEAE- or QAE-
exchangers is carried ou~ in a slightly acid medium
(pH 5.5) and on exchangers which have been appropria~ely
equiLibrated, for exa~ple with 0.1 mol/l Na acetate buffer
containing 0~1 moL/L Lysine~ It is aLso possible to use
this buffer to diLute the pasteurized solu~ion contain-
ing F VIII to ~wice the voLume before it is treated with~he exchanger in a batch or column process. Batch adsorp-
tionO ~hich is preferably used, has the advan~age that
i~ is possible to foLLow the binding of F VIII to the
exchanger during adsorption by functional determination
15 of F VIII in the supernatant, and, after the activity
has disappeared, it is possibLe to separate the non-
adsorbed factors from the ion exchanger by sedimentation
or centrifugation and, immediateLy thereaf~erO to start
with the washing of the exchanger loaded with F VIII.
~hereas the washing is advantageously carried out as a
batch process, for example on a suction filter~ it is
advisable to transfer the exchanger to a column for the
solution since eLut;on by coLumn chromatography has the
advan~age that the F VIII is obtained in a ~elatively
~5 concentrated form - of the order of SD - 100 IU F VIII /mL
under experimental conditions.
The buffers which are suitable for ~ashing the
exchanger which is loaded with F VIII are, in particular,
those which, on the one hand, do not dissolve off the
30 F VIII but, on the other hand, are suitable for removaL,
~hich is as nearly quantitative as possible, of non-
specific proteins, such as the immunoglobulins, and thus
also the isoagglutinins~ Surprisingly, an appropriate
buffer has proved to be a buffer in ~hich the starting
35 material can be dissolved for the adsorption and which
contains 0.1 mol/l Na acetate, 0.1 mol/l lysine and
1 g/l NaCl at pH 5.5. The ion exchanger loaded
with F VIII was ~ashed with this buf~er until the eluate
has free of isoagglutinins~ The highly sensitive Coombs
,
~ 7 -
test was used for testing, this also indicating incom-
pLe~e antibodiesO
Concentra~ed salt solutions, for example those
containing NaCl, are suitable for the desorption of the
5 F VIII from the anion exchangers.
However, other salts of halogens with Na, K or Ca
have proved ~o be more advantageous: KBr~ ~a~r and CaCl2.
They have ~he advantage that the F VIII is eluted as a
relatively sharp peak with a high activity per unit
volume, and this is an important advantage for the
precipitation.
The concentrations are in the range 0.05 mol/l
up to the satura~ion limit.
Finally, the yield also depends quite significantly
on the rate of elution~ which should be of the order of
1-10 ml/cm2/h, preferably 5~7 ml/cm2/hO
The elua~e containing F VIII can be concentra~ed
by the customary me~hods, that is to say by precipita~ion
with neutral salts, such as ammonium sulfate or NaCl,
advantageously with NaCl which can be diaLyzed out more
rapidly than ammonium sulfate and, moreover, need not
be removed completeLyO
The fur~her processing of the precipitate contain-
ing F VIII is carried out in such a manner that the residue
from the precipitation is dissolved in, for e~ample, a
buff~r of p~ 6.9 which contains Na citrate ~0.02 mol/l~,
NaCl tO.06 mol/l), glycine ~20 g/l) and albumin (5 g/l)
~dialysis buffer) and is dialyzed to equilibrium against
the same buffer without albumin. The dialyzsd solu~ion
is diluted wi~h the albumin-containing dialysis buffer
to an activity of 25 to 30 IU F VIII/ml and, after
sterilization by filtration, is dispensed into containers
and, ~here appropriate, freeze-drieq. The final product
is a white lyophilizate ~hich dissolves in Less than one
minute~ has a F VIII activity of 5 10 IU/mg protein~ and
is pasteurized and free of isoagglutinins. Compared with
products of the state of the art, it has ~he advantage
that it contains no undesired proteins, in particular
those which, as isoantigens~ might lead to sensitization.
..
1~5(~51
- 8 -
The product has not given rise to blood-group incompati-
bilities. Transmission of viral diseases, in particular
the various types of hepatitis, appears to be ruled out.
The advantages of the process include the facts that it
is relatively straightforward and thus can be transferred
without difficulty to an industrial scale and that it
comprises few ~orking steps. The small number of working
steps is reflected by the good yield, since experience
has shown that every purification step is associated with
loss of activity to a greater or lesser extent.
Thus the invention also relates to a highly puri-
fied, pasteurized fac~or VIII preparation which is virtualLy
free of immunoglobulins, isoagglutinins, fibronectin and
coagulable fibrinogen and which has a specific clotting
~C) activity (F YIII :C~ of about 100 U/mg protein and
a ratio of F VIII :C to F VIII R:Ag (related antigen) of > 1.
F VIII can be determined by the following proce-
dure: One part, for example 0.1 ml of partial thromboplastin,
for example that prepared by the method of German Offen-
ZO legungsschrift 2,316,43û, is mixed with one part of F VIII-
deficient plasma and one part of diluted normal plasma.
This mixture is ~aintained at 37C for 6 minutes. After
the addition of one part of a 0.025 molar calcium chloride
solution prewarmed to 37C, the time ~hich eLapses be-
ZS tween addition of the calcium chloride solution and theappearance of a clot is determined. A calibration curve
dra~n up using serial dilutions of normal plasma is used
for quantitation.
1 international unit ~= 1 IU) of F VIII is equiva-
lent to the F VIII activity of 1 ml of normal plasma asa substandard to the 3rd International WHO Standard.
The process for obtaining a pasteurized F VIII
preparation ~hich is free of isoagglutinins is illustrated
below~
Example
1. Starting material
250 9 of crude cryoprecipitate were dissolved
by heating at 30-37C in 750 ml of an NaCl solution
(0008 mol/L), which contained glycine ~0.25 mol/l) and
9 _
heparin (1 25 USP-U/ml). rhis resulted in 1,000 ~l of
a solution ~ith a concentration of 0~06 mol/l NaCl,
0.2 mol/l glycine and about 1 USP-U heparin/ml~ The pH
of the solution was adjusted with 1 N HCl ~o 6.5.
2. Aluminium hydroxide adsorption
80 ml of a suspension containing 10 g/l aluminium
hydroxide tBehring~erke Marburg) were added to 10000 ml
of solution from 1. and ~he mixture was stirred for
15 minutes (temperature abou~ 30C). It was then
centrifuged at 3,000 x g for 15 min, the residue was
discarded~ and the supernatant ras pas~eurized after the
addition of stabilizers.
3. Pas~eurization
The follo~ing stabilizers were added~ in this
sequence, to 1~000 mL of supernatant from 20
5 ml of CaCl2 so~ution, 1 mol/~ t5 mmol/l)
1,000 9 of sucrose (SOO g/kg solution)
15û 9 of glycine (2 mol in 1 l of solution)~
The pH was adjusted wi~h 2 N NaOH to 7~3. The
volume increased to 1,700 ml o~ing to the additions.
The solution was kept at 60C in a water bath for 10
hours.
4. Ion exchanger treatment
1,700 ml of solution from 3. ~ere diluted ~ith
1~700 ml of a solution which contained 0.2 mol/l Na acetate,
pH S.5, and 0.2 mol/l lysine. The pH ~as adjusted
, with dilute acetic acid to 5.5, and 70 ml of DEAE-sepharose
?
6 B Cl~ equilibrated with a solution containing ~a acetate
~0~1 mol/l), pH 5.50 lysine (0.1 mol/l) and NaCl (1 g/l)~
were added. The mixture was stirred at room tem-
perature for 2-3 hours, and the binding of the F YIII
~as deter~ined by determination of the activity in the
supernatant. When the F VIII activity had decreased from
4 IU to 0.1 TU/ml, the adsorbent was re~oved by centri-
fugation.
The supernatant was poured off and the adsor-
bent was separated from the remainder of the supernatent~
Entrapped protein ~as remo~ed from the sepharose by wash-
ing with a solution containing 0.1 mol/l Na acetate,
de~ e~ ~r~ n~
,
':
S
pH 5.5, 0.1 mol/l lysine and 1 g/l NaCl~ and it was
then ~ransferred as a slurry in the same buffer to a
column (dimensions: 10x3 cm).
The exchanger was washed in the column un~il there
~as no longer any measurable absorp~ion of light at
280 nm and the isoagglutinin values in the Coombs test
were at the detection limit.
5. Elution
The exchanger from 4O ~as eluted with a solu~ion,
of pH 5.5~ containing 0.1 mol/l Na acetate, 0.1 mol/l
lysine and 0.3 mol/l CaCl2. This resul~ed in the
appearance of a peak measurable at a wavelength of
280 nm. The corresponding fractions were collected~ and
a volume of 180 ml con~aining 40 IU F VIII/ml was obtained.
Precipitation of F VIII
2.2 mol/l glycine and 15û g/l NaCl wer~ added
to 180 ml of eluate from 5., and the mixture was stirred
at room temperature for 30 min.
The precipitation ~as recognizable by a marked
opalescence. The precipitate was separated off by
centrifugation at 30,000 x g in an ultracentrifuge for
30 min and, after the supernatant had been poured off~
it was kept at 4C overnight~
7. ~orking up
The precipitate from 6. was taken up in 145 ml
of buffer, of pH 7.0, ~hich contained 0~02 mol/l ~ri-
Na citrate~ 0.06 mol/l NaCl, 10 g/l glycine and S g/l
human albumin tdissolving buffer). The activity de~er-
mined in the solution was 40 IU F VIII :Ctml. It ~as
dialyzed against the abovementioned buffer containing
no albumin, at room temperature for 3 hours, and the
dialyzate was warmed to 30C and centrifuged at 3û,000 x 9
and 25C for 30 minutes~ After determination of F VIII,
which showed 36 IU/ml, the solution ~as adjusted to
30 IU/ml with dissolving buffer, and the solution was
warmed to 37C and sterilized by filtration through a
membrane filter.
It was dispensed into vials, ~rozen and freeze-
dried.
, '
~ .
, : .
