Note: Descriptions are shown in the official language in which they were submitted.
WO g2/tJ333 2 1 ~ 4 9 ll 3 PCT/US92/01577
TECHNETIUM-99m LABELING OF PROTEINS
DESCRIPTION OF THE INVENTION
This invention relates to a procedure for attaching
technetium-99m to antibodies using reducing metal reagents.
These reagents play a dual role in the labeling reaction under
the specified conditions. The method of this invention
l0 overcomes two problems with prior art methods, which are low 7
specific activity and binding of Tc-99m to low affinity binding
sites.
BACKGROUND OF THE INVENTION
Prior art methods for labeling antibodies with technetium-
99m used stannous chloride as a reducing agent to generate
sulfhydral groups on antibodies. At the same time the
antibodies were contacted with technetium and a chelator,
typically DTPA, to achieve binding of the technetium to the
antibodies, while scavenging unbound technetium with the DPTA
present in the reaction medium.
; Paik et al. reported that carrying out technetium-99m
labeling in presence of excess DTPA (MoAb:DTPA = l:l0) one
could selectively attach technetium-99m to high affinity sites.
; 25 Stannous chloride was present in l0-fold excess over the
protein. Their typical reaction conditions (Paik et al.) are
as follows:
[MoAb] = l0~m
[SnCl2] = 100 ~
[DTPA] = l00 ~m
Selective binding to high affinity sites, however, was obtained
only under experimental conditions where both DTPA and antibody
were competing for the reduced technetium ion. Paik et al.
reported that about l0 times molar excess of DTPA was required
to avoid technetium-99m binding to low affinity sites.
Unfortunately the presence of excess DTPA resulted in reduced
specific activity (~mCi/mg). Following their procedures with
antibody 88BV59, an IgG3, the yield was only 0.0l-0.5mCi/mg.
~.'
.:
.... , . ,. . ~ . ~ , .. , .. :
.. ~ . ' . -:
,
: ,:
W0g2/ 21~9,~
15333 PCT/US92/Ot5~17
SUMMARY OF THE INVENTION
This invention relates to a procedure for attaching
technetium~99m to proteins such as monoclonal antibodies using
reducing metal reagents such as tin and zinc according to which
99mTc binds is to high affinity binding sites and high specific
activity is maintained. The reagents play a dual role under
the given experimental conditions by reducing disulfide bonds
in the proteins to sulfhydral groups suitable for binding to
technetium, and reducing pertechnetate from Tc(VII) to Tc(III)
or Tc(V). By the preferred method of this invention, reduction
of the disulfide groups on the protein is conducted initially
with an excess of tin or zinc reagent, a pertechnetate reagent
is added at the end of the protein reduction reaction and
allowed to continue to reduce the technetium. Thereafter a
chelator scavenger is added to remove poorly bound or unbound
9 9mTC .
BRIEF DESCRIPTION OF THE DRAWINGS
Figure l illustrates stability studies of 99mTc bound to
20 IgG3 antibody 88BV59 in saline solution with excess DTPA in a
ratio of IgG:DTPA of l:lOOO at 37C.
Figure 2 shows HPLC radiochromatographs of 99mTc-88BV59 in
the reaction medium after preparation according to the method
of the invention. Figure 2a shows the peak of technetium
antibody conjugate as the major peak. The minor peak is
technetium bound to DTPA. Figure 2b shows the technetium
antibody conjugate purified with all measurable chelator bound
technetium removed.
Figure 3 illustrates the immunoreactivity of the antibody
technetium conjugate prepared according to the invention
compared with the immunoreactivity of the antibody alone.
Immunoreactivity was determined by indirect ELISA on specific
antigen coated wells. The reactivity of the radiolabeled
antibody (Tc-antibody conjugate) was determined by comparison
with the reactivity of native (unbound) antibody by their
ability to bind cognate antigen for which the antibody (88BV59)
has specificity.
Figure 4a illustrates the retention of antibody technetium
conjugate by ~umor xenografts in 6 to 8 week old athymic Balb/c
- .
:' ; ' ., ,
-
.
,
. . .
, wa 92/l5333 2 1 ~ ~ 9 ~1 3 PCT/US92/01577
mice. The xenografts were developed using enzymaticallydissociated human tumor cells containing antigens recognizable
by 88BV59. Ten micrograms (1-2 ~Ci/~g) of labeled antibodies
were injected into the veins of the mice (n=6) for
biodistribution studies. A comparison is made between
conjugates with intact antibodies and conjugates with F(ab' )2-
Figure 4b illustrates serum retention of 88BV59 technetiumconjugates in mice having human colon tumor xenografts.
Figure 4c illustrates the tumor retention of antibody and
F(ab' )2 technetium conjugates in the mice.
Figure 4d illustrates kidney retention of F(ab') 2 and
intact antibody technetium conjugates in the mice.
Figure 4e illustrates liver retention of F(ab') 2 and -
intact antibody technetium conjugates.
; 15 Figure 5 shows a coronal view of the liver SPECT scan of
a human patient who has received 15 mCi/lOmg g9mTc-88BV59 at 4
to 5 hours after administration. Large numbers of lesions in
the liver of a size less than or equal to 0.5 cm can be seen.
These results were later confirmed by CT scan.
i 20
DETAILED DESCRIPTION OF THE ~MBODIMENTS
This procedure describes the protocol for attaching
technetium-99m (9gmTc) to proteins using reagents containing
reducing metals such as tin and zinc. These reagents play a
dual role under the given experimental conditions.
Binding to the protein is through a sulfhydral group (SH)
- obtained by reduction of disulfide in the protein. Thus,
cysteines must be present in the protein for conjugation.
` The reagents contain well known reducing metals bound to
ligands through covalent or coordination bonds. They are
sufficiently powerful enough to reduce disulfide bonds present
in the protein molecule, creating sulfhydryl groups suitable
for attachment to technetium, but not so powerful as to form
metal hydroxide colloids. Examples of the preferred metals are
~5 Sn, Zn, Rn and Co. They are bound to ligands such as
oligosaccharides, polysaccharides and other sugar derivatives
~; by covalent or coordinate bonds.
: ;
.
.,
"
, . ,
: ' : - .; . , :
.
:.'::. . ' ~ , -
: ~ . ., -
.
W092/ls3~3 2:LU'l~ ' ` PCT/US92/Ul57~ ~ ~
The reagents also reduce pertechnetate for attachm~nt to
the protein. Tc~VII) is reduced to either Tc(III) or Tc(V) and
concomitantly coupled to the sulfhydryl group on the protein.
Any loosely bound technetium is chelated with DTPA, EDTA,
S iminodiacetate, cysteine, diaminedithiol or other chelators,
which are added to the reaction mixture after reduction and
binding o~ Tc to the protein to quench the reaction by
scavenging unbound and loosely bound Tc. The ratio of MoAb to
quencher is preferably from about 1:1 to 1:5t and should not to
exceed about 1:8. The chelators may be attached to an immobile
surface, or may be removed by gel filtration chromatography.
Our imaging experiments with Tc-antibody conjugates clearly
show that the presence of small amounts of Tc-DTPA does not
affect the quality of imaging because Tc-DTPA is rapidly
cleared from circulation by renal filtration. Thus, it is not
always necessary to remove chelator bound 99mTc from the
preparation before administration.
This process of making the radiolabeled antibody is
unique. In the preferred embodiment tin or zinc saccharate or
glucarate is used to produce sulfhydryl groups and to reduce
technetium for conjugation to sulfhydryls in the antibody.
Also, the process is unique in using chelators as quenchers,
rather than competing for reduced technetium in the reaction
mixture by adding them earlier. Our reducing reagent is
preferably tin saccharate prepared by adding saccharic acid
(e.g., 20 mg/ml, deaerated) solution to tin chloride solution
(e.g., 5 mg/ml in 0.02M HCl). Tin saccharate may also be
prepared by treating tin chloride with excess saccharic acid,
removing the precipitated tin saccharate and storing the
precipitate in dry nitrogen. It is also possible to combine
the metal chloride and the acid together and add that reaction
mixture to the protein (e.g., combining stannous chloride and
glucaric acid).
The antibody (10 mg/ml or lyophilized powder) in a buffer
solution, or alternatively in a reducing buffer solution is
added to the tin saccharate solution and incubated at about 4
to 60C for 5 to 60 minutes. This incubation leads to
formation of sulfhydryl groups. The period of incubation
varies inversely with temperature. Reaction temperature is
:: -
- .~
- . . . : . ~ . ~
~ ~ 92/15333 21 ~ i 9 l¦ 3 PCT/US92/01577
linlited by the stability of the protein. A temperature of
incubation cannot be used that will denature the protein.
Preferred reaction conditions are about 15 minutes to 60
minutes at about 20 to 37C. Under experimental conditions l
to 3 SH groups are generated per antibody molecule. This
method of labeling has proved to be particularly suitable for
antibodies such as an IgG's. Under the same reaction
conditions use of tin chloride alone, not as a saccharic acid
salt, leads to formation of a colloidal solution not suitable
for further use. Thus the reducing metal must be bound to a
ligand for the method to work.
Reduction of the antibody is followed by addition of
pertechnetate. Incubation to reduce Tc(VII) to TctIII) or
; Tc(V) and to conjugate with the sulfhydrals on the antibody is
carried out at about 20 to 37C for about two minutes to one
hour. Preferably, labeling is accomplished by incubation at
about 23 - 37C for about 30 to 60 minutes. Thereafter, a
chelator is added (e.g., DTPA) to quench the reaction and to
scavenge unbound Tc by conversion to Tc-DTPA. This resulting
pharmaceutical preparation is purified before administering or,
alternatively, directly administered to cancer patients without
removing excess Tc-DTPA. As a general rule, at least 90% of
the Tc should be bound to the antibody. Otherwise it should be
purified. Within 1-2 hour after administration non-antibody
conjugated Tc in the original preparation in the form of Tc-
DTPA will be removed by the kidneys. Patient studies with
radiolabeled antibody preparations containing Tc-DTPA have
shown good tumor localization. If the composition is to be
purified before administration, excess Tc-DTPA is removed by
gel filtration column chromatography, leaving pure radiolabeled
antibody.
Tc labeled antibodies prepared according to this invention
are very stable. Results obtained with cancer patients using
such preparations have clearly shown that even 4 hours after
administration the technetium-99m is firmly bound to the
antibody. Excellent localization of the radiolabeled antibody
was also observed in these cases making it possible to obtain
good radioimmunoscintigraphs. Loosely bound Tc, if any, would
bind to human serum albumin. HPLC analysis of the serum from
:`
~ .,
.
, . . ~. .
W092/15333 21~ PCT/VS92/01571
a patient treated with Tc-99m labeled 88BV59 did not show any
transfer to human serum albumin even 4 hours after
administration.
Another advantage of this method is its ability to label
relatively difficult systems, such as F(ab')z. Reductive
labeling with technetium of F(ab' )2 frequently results in
formation of 99mTc labeled F(ab). In fact many researchers use
the reductive method to obtain 99mTc labeled Fab fragment from
F(ab')z. In this invention, using appropriate concentrations
and reaction conditions, particularly reacting at room
temperature (20-25C), one can mildly introduce technetium in
F(ab' )2 without alteration~
We radiolabeled the F(ab')z fragment of 88~V59, an IgG3,
- using this method and about lO mg/lO mCi of the
radioimmunoconjugate was administered to cancer patients.
Planar and SPECT images showed localization of the radiolabeled
antibody in lesions. Also HPLC analysis of serum from patients
; showed that 9 9mTC was firmly bound to the antibody. The
immunoreactivity of radiolabeled antibody was not affected by
this procedure.
Example
We found out that by treating a concentrated antibody
solution (10-50 ~m solution) with 30 to 50 molar equivalents of
a stannous salt solution (in particular stannous glucarate) for
a short period of time at elevated temperatures (4~60C); one
could generate large numbers of -SH groups (2-3 per molecule,
as determined by DTNB tests using Elmans reagent suitable for
Tc-binding). This method is specifically suitable for -SH rich
proteins. Sodium pertechnetate was added at the end of the
reaction. The reaction was allowed to continue for additional
20-30 minutes in an inert atmosphere (vacuum or nitrogen).
Scavenging solutions containing chelators such as DTPA, ETDA,
cysteine or diaminidithiol chelators were added at the end of
the reaction and incubated at room temperature for about 5 to
lO minutes. This converted any remaining TcO~ unbound to MoAb,
to Tc-DTPA.
. . .
.... .. .
W~92/15333 2 .1~ 3 PCTtUS92/01577
Experimental conditions were as follows:
Stannous Glucarate : 1-2mm
~eaction at 37C. for 15-30 min (alternate condition are
room temperature for 60 min. or 45C, 3-6 min.~ in a
5evacuated vial.
TcO~ (50-100 mCi) was added and reacted at 37C for 15
min. (alternatively 23-25C for 30 min.).
DTPA was then added (1-100 ~m solution). DTPA to MoAb
- ratio was 0.1:1 to 5:1.
10Reaction yields of 10-15 mCi/~g of protein was easily
achieved.
If radiolabeling yields were less than 90%, the
radiolabeled antibody wsuld be purified by gel flltration
chromatography. In general, yields were always >90% (with
1588BV59). Results of purification are illustrated in Figure 2.
In vivo biodistribution data in mice showed that: the
radiolabeled antibody was retained in serum and tumor; uptakes
in normal tissues such as liver, bone, spleen, muscle and
intestine were low (<3% I.D./g); and, depending on the nature
of the antibody, kidney uptakes were low to moderate.
Early studies in colon cancer patients showed that the
radiolabeled antibody localized to tumor metastases (~igure 5).
',
,(
. . .
". .. - ~ , ~ :
,. . . .. - ~ :
- ~ , ,
~laL~
W O 92/15333 P~'r/US92/01577
References
Fritzberg, A.R., Abrams, P.G., Beaumier, P.L. et al.,
Proceedings of National Academy of Services, USA, 85:
~025-4029 (1~88).
Paik, C.H., Pham, L., Hong, J.J., Suhami, M.S., Heald, S.C.,
Reba, R.C., Steigman, J. and Eckelman, W.C., International
Journal of Nuclear Medicine and Biology, 12:3-8 (1985).
Paik, C.H., Eckelman, W.C. and Reba, R.C., Nuc. Med. Biol.,
~ 13:359-362 (1986).
15 Rhode, .A., Torvestad, D.A., Breslow, K., Burchiel, S.W., Reed,
K.A., and Austior, R.W. In: S.W. Burchiel and B.A. Rhodes,
"Tumor Imaging", p. 111, New York, Masson Publishing, USA,
1982.
- , : . . - . .: -:
. .. . , . ... . . . .. :
. , , . . : . :
~ , . : . . : .
: . . . 7 -
:
