DESCRIPTION
FIELD OF THE INVENTION
The present invention relates to a process for producing 1,4,7,10-
tetraazacyclododecane-1 , ,7,10-tetraacetic acid (DOTA) including
salts and hydrates thereof, macrocyclic compounds comprising metal
ions complexes thereof and compositions comprising said macrocyclic
compounds, which can be used to produce contrast agents for magnetic
resonance imaging.
BACKGROUND OF THE INVENTION
Magnetic resonance imaging (MR ) is a powerful, non- invasive
technique used to produce detailed two or three-dimensional
anatomical images of tissues in the body. Conventional MRI uses the
proton H as its signal source which is highly abundant in tissues
and it has the highest sensitivity of all the biologically relevant
nuclei .
Contrast, which makes the differentiation of internal structures
possible in the image, arises from how the signal decays and is the
difference between the resulting signals from two tissue regions.
The route by which the protons release the energy they absorbed from
the radio- frequency pulse, thus reducing the transverse
magnetisation and causing signal decay, is known as relaxation. In
MRI two independent relaxation processes occur simultaneously: spinlattice
or longitudinal relaxation characterised by the time
constant T and spin-spin or transverse relaxation, characterised by
the time constant ¾ .
Often, when suitable T - or -weighting sequences are used, the
natural contrast between two tissues is enough to produce a
diagnostically-useful image. However, some conditions do not lead to
specific enough changes in the relaxation times of the affected
tissue though and then a contrast agent is used to locally change
the relaxation times of the diseased tissue, improving the image
contrast .
Most contrast agents work by shortening the relaxation times of the
water protons in the targeted tissue. T contrast agents are based on
paramagnetic metal ion chelates which make the tissue appear
brighter on the -weighted image (positive contrast) . T2 contrast
agents are usually superparamagnetic iron oxide nanoparticles which
create dark spots on the T2-weighted image (negative contrast) .
agents are the most widely used and the majority of these are based
on chelates of the gadolinium ion (Gd +) .
To be an effective T agent the gadolinium (III) chelate must
significantly increase the proton relaxation rates in water.
Gadolinium is the seventh element in the lanthanide series and, like
the other lanthanide elements, it is most commonly found in the +3
oxidation state, corresponding to the electronic configuration
[Xe]4f7 . This means that Gd + has seven unpaired electrons, making it
highly paramagnetic i.e. Gd(III) ions have large permanent magnetic
moments (due to electron spin angular momentum) , but in the absence
of an external magnetic field these are randomly oriented. Due to
its large size the Gd(III) ion typically has a coordination number
of nine in its complexes. A s a free ion gadolinium is very toxic for
the tissues but is generally regarded as safe when administrated as
a chelated compound.
The level of toxicity depends on the strength of the chelating
agent, also known as ligand, chelator or sequestering agent.
Usually these ligands are organic compounds which form two or more
separate coordinate bonds with a single central metal ion, in this
case, the gadolinium ion, inactivating it thus reducing or
eliminating its toxic effect in the tissues.
Polyaminopolycarboxylic acid compounds are the ligand type of choice
because they form exceptionally stable complexes with the Gd(III)
ion, which can be explained by a number of reasons. These compounds
can be linear (such as pentetic acid or diethylene triamine
pentaacetic acid also named as DTPA) or macrocyclic (such as
I ,4 ,7,10-tetraazacyclododecane-l , ,7 ,10-tetraacetic acid, DOTA) .
DOTA is used as the ligand in the synthesis of the MRI contrast
agent gadoterate meglumine ([Gd(DOTA) (H 20 )] (meglumine)) .
Several synthetic routes for the production of DOTA have been
proposed, namely by Stetter, Hermann; Wolfram Frank (1976)- "Complex
Formation with Tetraazacycloalkane-N, ,N'',N' ;-tetraacetic Acids
as a Function of Ring Size". Angewandte Chemie International Edition
in English 15 (11) : 686) , by R . Delgado & J.J. Frausto da Silva -
Talanta, Vol. 29, pp. 815-822, Issue 10, 1982, and by J.F. Desreux -
Inorg. Chem. 1980, 19, pp. 1319-1324.
The preparation of DOTA was first reported in 1976 by Stetter &
Frank (full ref. above) through the reaction of 1,4,7,10-
tetraazacyclododecane with chloroacetic acid in aqueous alkali
medium to obtain DOTA wherein the resulting inorganic salts were
separated and purified by treatment with an ion-exchange column
Dowex 2x8 .
The method most widely reported in the literature is typified by
Delgado et al. (full ref. above), where cyclen is reacted with
chloroacetic acid under aqueous basic conditions (pH = -10) to form
DOTA, which is crystallised by acidifying the cooled DOTA solution
to pH 2 with hydrochloric acid and placing it in the refrigerator
overnight .
Desreux (full ref. above) also reported a similar procedure, but
specified sodium hydroxide as being the base used, with a reaction
temperature of 80°C, and stated that upon acidification DOTA
precipitates out of solution at pH 2.5.
E . Clarke & A . Martel (1991) - Inorganica Chimica Acta, 190, pp 27-
36) , describes the preparation of DOTA by alkylation of cyclic
tetraamine ligands with bromoacetic acid at a controlled pH between
II. 2 to 11.3 being the resulting product recovered by treatment with
a ion-exchange column as ammonium salts followed by treatment with a
potassium cation solution at pH of 11.5 and vacuum concentration.
The resulting ligands were then reprotonated by addition of HC1 and
isolated by recrystallization from hot water.
WO9905128 discloses a process for producing DOTA compounds by 2
step-alkylation wherein the alkylation agent is preferably
bromoacetic acid but also includes chloroacetic acid, in aqueous
solution at a basic pH with an excess of said alkylation agent,
followed by hydrolysis and purification with ion exchange resins and
with an optional recrystallization step in order to obtain highly
purified DOTA compounds. In particular, WO9905128 discloses a
multistep process for the preparation of DOTA starting from:
a ) an alkylation reaction of a 2a, a ,6a, 8a-decahydrotetraazacyclo
pent [f g ]acenaphthylene with an acid in aqueous solution and at a
basic pH, followed by
b ) a second alkylation reaction with a different alkylating agent,
and by
c ) the hydrolisis of any ester groups, and
wherein the amount of the first alkylating agent used in step a )
varies between 2 - 2.3 mol of reagent per mol of substrate and from
2 - 3 mol in step b ) and the reaction temperature varies from room
temperature to 80°C, depending on the reactivity of the alkylating
agent .
To be able to be eventually used as a suitable contrast agent
comprising gadoterate meglumine, the concentrations of process
impurities present in the raw DOTA (both organic and the inorganic)
must be removed or significantly reduced. This is so that the
purified DOTA meets the strict specifications for use in a contrast
agent or else it will not be approved for sale by the relevant
medicine regulatory body as it will not be considered safe enough
for human use. Therefore a series of purification steps must be
employed to remove these impurities without introducing too high a
concentration of a new impurity or residual solvent, as these must
also meet the specifications.
However, the DOTA resulting from the above mentioned processes is
still highly contaminated with organic and inorganic impurities, in
particular with chloride and sodium ions, and the conventional
purification steps using ion-exchange resins, as disclosed above,
only solves this problem in some extent .
In fact, G . Hernandez, .F . Tweedle and R.G. Bryant, Inorg. Chem. ,
1990, 29, 5109-5113, disclose the synthesis of the sodium salt of
[Gd(DOTA) (H 20 )] (Na [Gd(DOTA) (H 0 )] .4H20 ) . However, this compound is
unsuitable for use as a contrast agent as it contains sodium.
Nevertheless, the synthetic procedure herein disclosed highlights
that high temperatures (90°C) and long reaction times (6.5 hours)
are required to successfully react DOTA and gadolinium oxide (Gd203,
an ionic salt which is the source of the gadolinium ion) together to
form the thermodynamically stable [Gd(DOTA) (H 20)] . This can be
accounted for by the very slow kinetics of formation of the complex.
It is thus desirable to obtain an optimized process for the raw DOTA
synthesis which ensures not only high yields of this compound, at
least 50% relative to the amounts of starting reagents used, but
also that the raw DOTA is of a suitable quality and in a form that
was easy to work with. Furthermore, it is also desirable to simplify
the method for producing DOTA at an industrial scale .
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a method for
preparing DOTA that allows obtaining high yields with an improved
quality and purity in a simple, straightforward and reliable
process .
This object is realised by providing a five-step process for
preparing DOTA as defined in claim 1 .
It is another object of the invention to provide a process for
producing macrocyclic compounds comprising DOTA and metal ions
complexes, as defined in claim 10.
It is another object of the present invention to provide a process
for producing compositions comprising said macrocyclic compounds
that can be used as contrast agents in magnetic resonance imaging.
Further advantages and embodiments of the present invention will
become apparent from the following description and the dependent
claims .
DESCRIPTION OF THE INVENTION
The present invention relates to a four-step process for producing
1 , 4 ,7 ,10-tetraazacyclododecane-l, 4 ,7 ,10-tetraacetic acid (DOTA),
including salts and hydrates thereof of formula (I):
HOOC / v . , COOH
N
N N
HOOC- V-COOH
Formula I
The process of the present invention comprises the following steps:
reacting the cyclen 1 , 4 ,7 ,10-tetra-azacyclododecane and a haloacetic
acid with a base at a pH > 10;
crystallizing the 1,4,7,10-tetraazacyclododecane-l ,4,7,10-tetra
acetic acid obtained in step a ) by addition of an acid to
achieve a pH < 3 , followed by a heating step and cooling step,
wherein the heating step is performed at a temperature in the
range from 50°C to 100°C, for at least 5 minutes, and the
cooling step is performed at a temperature in the range from 5°C
to 25°C, for at least 5 minutes;
treating the raw material obtained in b ) with a cationic resin
and then desorbing DOTA with a volatile base solution;
further treating the resulting solution of c ) with an anionic
resin;
e ) washing the product of d ) in a two-stage wash, first with an
organic volatile acid with a pKa less than 5 until DOTA starts
to be released from the resin, and a second stage with an
organic volatile acid with a pKa less than and/or with a higher
concentration than the first selected organic volatile acid to
release DOTA from the resin.
1. Synthesis of raw DOTA
In the first step, cyclen (1, , 7 ,10-tetraazacyclododecane) is
reacted with a halo-acetic acid and an excess of a base, at a pH >
10 according to the following reaction:
Cyclene DOTA
The four amine groups on the cyclen molecule react with four
equivalents of a halo-acetic acid in a nucleophilic substitution
reaction .
In the scope of the present invention halo-acetic acid means a
derivative of acetic acid by substitution of one H by iodo, bromo or
chloro. In a preferred embodiment chloroacetic acid is used.
An excess of a halo-acetic acid can be used in this step, preferably
in an amount of at least 4 equivalents (eq.) and more preferably
between 5.0-6.0 eq. with regard to the initial amount of cyclen.
According to the cited prior art, this step is performed at a
temperature of approximately 80°C. However, it was found that using
an excess of halo-acetic acid according to the present invention,
lower temperatures may be used. Therefore, in the scope of the
present invention, this step can be performed at a temperature
ranging from 20 to 100°C, preferably from 20 to 65°C and more
preferably from 20 to 30°C.
The higher temperatures (> 65°C) can be used to speed up the rate of
reaction but they are less preferred in industrial processes due to
the associated higher costs. Moreover, the yield of DOTA is not
better than the one obtained at temperatures ranging from 20 to
65°C.
At the prior art temperatures, this step can be performed during
around 40h. Said reaction times (40 hours or more) can be used to
ensure that all four amine groups undergo substitution and no tri-,
di- or mono -substituted cyclen derivatives are present in the
reaction mixture at the end. However, it was found that the maximum
selectivity for product formation, according to the present
invention, is achieved after 20-24h of reaction. Therefore, this
step is also preferably performed during at least 7h, most
preferably at least 20h, even more preferably at least 24h.
A s in the prior art, the reaction can performed at pH values of
about 10. However, it was now surprisingly found that even when the
reaction was performed at a pH 13 no adverse effects were
observed. By not having to constantly monitor the pH ~ 10 as
described in the prior art, the synthesis procedure is easier to
operate. Therefore, in a preferred embodiment of the present
invention, this step of DOTA synthesis is performed at a pH > 13 by
addition of a base at once with no "on process" monitoring of the pH
values .
In the scope of the present invention, suitable bases are inorganic
bases such as hydroxides of alkali metals, and more preferably
hydroxides of alkali metals selected from the group of KOH, NaOH,
LiOH, RbOH and CsOH. In a most preferred embodiment of the present
invention NaOH is used as a base .
The base can be added in excess, of at least 2 times the amount of
the halo-acetic acid present in the reaction, namely by using
amounts ranging from 8 to 16 equivalents, preferably between 10-12
equivalents with relation to the halo-acetic acid. Higher amounts of
said base may be used but then larger amounts of cations are
introduced into the system making their removal, in a later stage of
the process, more difficult and consequently more difficult to
achieve a better purification of DOTA. Furthermore, by using the
above mentioned amount of base it is also possible to prevent the
reaction from being too exothermic, which would be unsafe,
especially on a large scale. The base can be added in solid form or
as a concentrated solution, e.g. of at least 30%. A possible
explanation for this is that the base activates the cyclen towards
nucleophilic substitution and the resulting carboxylic acid pendant
arms are also deprotonated under the basic conditions, resulting in
the fully deprotonated DOTA molecule (L , L referring to the ligand
DOTA in its neutral form) with positive counterions .
The process of the present invention thus provides a first step
which is easier to operate and takes less time than the ones of the
prior art. Moreover, it also ensures that no undesirable side
compounds are produced thus resulting in higher yields of raw DOTA.
2 . Crystallization of raw DOTA
During the second step, the reaction mixture is crystallized by
addition of a concentrated acid and by performing a heating and
cooling step, preferably followed by a washing step.
The acid is added until a pH < 3 is achieved producing a
precipitate. Suitable acids are inorganic acids and more preferably
acids selected from the group of HC1, H2S04, HN0 , HBr, HI and HC10 4 .
The reaction is then subjected to a heating and cooling step to
obtain an improved yield and quality of the crude DOTA. This is
performed by heating the reaction at a temperature ranging from 50
to 100°C, preferably 50 to 70°C, more preferably 50 to 60°C, for a
short time period of at least 5 minutes, in order to dissolve the
precipitate and obtain a clear solution. Then the reaction is
cooled at a temperature ranging from 5 to 25 °C, preferably 5 to
15°C, more preferably 5 to 10°C, for a short time period of at least
5 minutes, to obtain DOTA in the form of a salt, such as DOTA
hydrochloride or other salt, depending on the acid selected for
lowering the pH of the solution.
Thus, this heating/cooling step allows to obtain the DOTA with a low
content of other cations. In fact, it was surprisingly found that at
pH~3 DOTA can be precipitated from the solution as a solid. The
final pH of the reaction mixture is then low and can be less than
0.5. At this low pH values the raw DOTA is found in its fully
protonated form (H L +) wherein L refers to the deprotonated ligand
DOTA and whereby the counterions, such as chlorides, introduced by
the reaction with the acid, are electrostatically bound to it, so
that its form can be expressed as H L (X ) x refers to the
counterions and n refers to the charge of the counterion. By
precipitating DOTA as a salt it is easier to isolate it by
filtration and to perform further purification. Apart from the
negative counterions the raw DOTA is also contaminated with any
residual ions introduced by the reaction with the base that
precipitate out alongside the DOTA salt.
Preferably, a washing step is performed after the heating/cooling
step to further remove the remaining cations. This can be done with
a mixture of water and a water miscible low boiling organic solvent
in a ratio ranging from 1:1.5 to 1:3, preferably in a ratio of 1:2
(weight/weight) . Suitable examples of a water miscible low boiling
organic solvent are acetone, ethanol, methanol and iso-propanol . In
a preferred embodiment of the present invention, acetone or ethanol
are used. Due to the low solubility of DOTA in such solvents this
step is performed to precipitate DOTA from a solution in water.
The potential organic impurities present in the raw DOTA include any
unreacted alkylating acid or intermediate cyclen derivatives.
However, it is observed that the reaction goes to completion; the
yield losses are most probably due to solubility issues, i.e. the
DOTA failing to precipitate out of solution completely. It is
possible to improve the yield of raw DOTA by adjusting the reaction
conditions, but this affects adversely and significantly the purity
and thus the synthetic method of the present invention aims to
achieve a compromise between yield and purity necessary to comply
with the requirements for contrast agents.
In this way a raw DOTA can be obtained that has already a low cation
content (<0.5 % ) , which makes further purification easier. Also
unreacted starting material, such as the halo-acetic acid that is in
excess, and intermediates are removed. This also leads to a very
pure final product and to yields of approximately 70%. The
composition of the washing mixture is optimized for minimal loss of
DOTA and low sodium content. Moreover, working at low pH values the
reaction is easier to operate since no exact final pH is required.
3 . Purification of DOTA
3.1. Treatment with cationic resin
The raw DOTA is first treated with a cationic resin to remove non
desirable anions introduced by the addition of the acid in step b )
of the process, such as chloride ions.
The cationic resin is typically a strongly acidic cationic exchange
resin such as Amberlite IR120H, Lewatite S100H, or Purolite UCW9126
H+, preferably a resin in the hydrogen form due to its ready
availability and good theoretical total capacity (2.0 meq/ml) .
The positively charged raw DOTA and the cations are adsorbed by the
resin while the anions remain in solution and can be washed away. To
desorb the DOTA from the resin though requires the use of a volatile
base in an aqueous solution so it can be easily removed and to
increase the solution pH to values higher than . In the scope of
the present invention, a suitable volatile base may be ammonia,
butylamine, triethylamine or diethylamine , ammonia being the
preferred one.
Adding an aqueous volatile base to desorb the DOTA from the resin
causes the remaining cations from the previous step and ions
introduced by said volatile base to enter into equilibrium with one
another. The relative selectivities favouring the adsorption of some
cations to the resin are outweighed by the excess of the volatile
base. This results in desorption of the cations from the cationic
exchange resin during elution with aqueous base. The basic solution
also raises the pH, resulting in converting the DOTA cationic salt
into the neutral DOTA or even a DOTA anionic salt. Both these
species will be released from the cationic exchange resin.
Desorbing DOTA as described introduces the volatile base ion as an
impurity to the system, yielding for instance an ammonium-DOTA
species as the major product from the treatment. The washing step
with said volatile base may be repeated until no more DOTA is found
in the solution.
Although not all the remaining cations resulting from the
crystallization step can be removed during the ion exchange resin
treatments, the concentration can be significantly reduced,
particularly during the cationic resin treatment.
Therefore, the concentration of cations in the freshly synthesised
raw DOTA can be sufficiently decreased, by the first 3 steps of the
process of the invention, before any further purification steps are
performed and thus the product after the optimised ion exchange
resin treatments meets the required sodium content specification.
Using a larger volume of a concentrated acid was shown to help
decrease the cation content of the raw DOTA. However, this step
alone is insufficient for getting the desired low cation
concentrations (around 0.1 w%) , meaning that further purifications
are needed.
Furthermore, the cationic exchange resin is a very effective
treatment for removing the anions from the DOTA cationic salt so
that the concentration present meets the specification without any
further treatment required. It works regardless of the original
concentration of anions in the raw DOTA because the anions simply do
not bind to the cationic exchange resin (or remain electrostatically
bound to DOTA) and can therefore be filtered and washed away off the
DOTA-bound resin.
3.2 Treatment with anionic resin
The treated DOTA resulting from the previous step contains now the
volatile base ions (-5%) , which were introduced as an additional
impurity to the system by the cationic resin treatment and thus must
be substantially decreased. Therefore, the resulting DOTA is then
subjected to an anionic exchange resin treatment to remove said
impurities. On addition of the resin the negatively charged DOTA
binds to the resin, freeing the said ions and other cations so that
they can be washed away (the base ions and the hydroxide ions from
the resin will be in equilibrium with base and water) .
The anionic resin is typically a strong basic anionic exchange
resin, such as Amberlyst A1260H, Lewatite 600OH or M800OH, Purolite
UCW5072 OH-, preferably a resin in the hydroxide form due to its
"ready-to-use" availability and good theoretical total capacity (1.0
meq / ml) . 6 volumes of resin are used with regard to the weight of
the cationic resin treated DOTA. If the content of DOTA in the
solution is too high, an extra volume of resin may be added.
3.3 Washing with organic volatile acids
After treating DOTA according to the procedure described in the
previous step DOTA has to be released from the anionic resin. This
is possible by converting it to the neutral or cationic form through
the use of an acidic washing step.
Normally one would expect that by this step all cations should be
removed as they do not bind to the anion exchange resin. However, it
was found that after washing the resin with a volatile organic acid
the obtained DOTA is still contaminated with cations introduced by
the previous treatments. This might be explained by the fact that at
the pH of the solution obtained after releasing DOTA from the cation
exchange resin, said DOTA is at least partly in a di-anionic form
H2L , where 1 charge is bound to a cationic resin site and the other
charge is neutralised by an other cation that still is present in
the solution after release from the cation exchange resin.
Thus, in order to avoid the problem of DOTA contamination with such
cations, it was surprisingly found that the use of a two step
washing, first with a more diluted solution of a choosen volatile
acid washes away the cations from the DOTA-resin complex. In result,
upon using a more concentrated acid, in the following second washing
step, DOTA is released from the resin, possibly as a DOTA - formate
or - acetate salt, the salt depending on the choosen volatile acid,
but finally free from other cations.
Examples of suitable organic volatile acids for the present
invention are organic volatile acids with pKa less than 5 , such as
formic acid, acetic acid, and fluoracetic acid and oxalic acid.
Formic acid and acetic acids are preferred being the most preferred
the formic acid.
In a preferred embodiment formic acid or acetic acid is used. In a
more preferred embodiment, formic acid is used because it introduces
lower amounts of impurities that must be removed later.
If formic acid is used in the first washing step, a concentration
between 0.01-0.1% may be used, preferably between 0.02-0.03%. If
acetic acid is used in this step, then a concentration between 0.1-
0.3% may be used. In a preferred embodiment of the invention, the
organic volatile acid used in this step is formic acid.
In the second washing step, DOTA is released from the resin by
addition of a higher concentration of the organic volatile acid used
in the previous washing step or by adding an organic volatile acid
with lower pKa than the one used in the previous washing step.
In a preferred embodiment, formic acid is used because it requires
less effort in a later stage to obtain higher purity levels of DOTA
but other acids may be used. In this sense, the formic acid
concentration is between 0.5 and 20%, preferably between 1.0-5.0%,
more preferably between 1.0-2.0% and even more preferably of 1.0%.
The addition of a solution with higher concentrations of formic
acid, such as 15-20% may speed up the procedure but it may also
require further purification steps to remove impurities introduced
in the system.
Therefore, lower amounts of said acids are needed to lower the pH of
the mixture and thus fewer ions are introduced in the system
resulting in an easier post treatment procedure for their removal in
a later stage.
By following the above described washing procedure it was
surprisingly found that it is possible to obtain good yields of DOTA
in a high purified grade. This is mainly due first to the lower
cationic content in the presence of the volatile base, probably due
to a mass effect of excess of said base on the divalent anionic DOTA
on the resin given by the first treatment with an organic volatile
acid, and secondly due to the lower anionic content when a anionic
resin was further eluted with an organic volatile acid in a higher
concentration or at lower pH.
3.4 Concentration
The resulting DOTA fractions can be subsequently concentrated, by
known techniques such as in a rotary evaporator under reduced
pressure, and treated with water. This procedure can be repeated
until a glassy oil is obtained, which can be further treated with a
low boiling water miscible organic solvent such as ethanol or
acetone to induce crystallisation. In the production plant, a powder
can be obtained by concentrating to 10% by vacuum distillation and
repeatedly adding the above specified solvent and concentrating
again .
If the base content is still out of the required specifications the
above obtained DOTA can be recrystallised by dissolving in water at
a temperature above room temperature such as at 40-60°C and
precipitating with a water miscible low boiling solvent such as
ethanol, acetone, cooling to room temperature, centrifuging and
drying in a vacuum tray dryer.
4 . Synthesis of gadoterate meglumine
The DOTA of high purity obtained as described above can be used as
the ligand in the formation of the contrast agent gadoterate
meglumine, [Gd (DOTA) (H 20 )] (meglumine) .
For this purpose, DOTA obtained according to the process of the
present invention is added to Gd203 by known methods preferably with
excess of DOTA, most preferably in a molar ratio slightly over 2:1
to form an aqueous solution of a complex DOTA-Gd.
5 . Synthesis of a contrast agent formulation
In order to prepare a contrast agent formulation, other excipients
commonly accepted in pharmacy may be added to the gadoterate
meglumine water solution. Typically the pH is adjusted to values
tolerated by the body such as from 6.5 to 8.0.
DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT
The present invention relates to a process for producing 1,4,7,10-
tetraazacyclododecane-1 ,4 , ,10-tetraacetic acid derivatives (DOTA) ,
macrocyclic compounds comprising metal ions complexes thereof and
compositions comprising said macrocyclic compounds, which can be
used to produce contrast agents for magnetic resonance imaging.
1 . Preparation of 1 , ,7 ,10-tetraazacyclododecane-l , ,7 ,10-
tetraacetic acid (DOTA) including salts and hydrates thereof :
1.1 Synthesis of DOTA
Cyclen 1 ,4 ,7 ,10-tetra-azaclyclododecane and 5 to 6 eq. of
chloroacetic acid (with regard to the initial amount of cyclen) are
reacted with a base, such as NaOH at a pH _> 13 .
The chloroacetic acid is added to a solution of 1 , ,7,10-tetraaza
cyclododecane in water (5 to 15%, preferably 9-11% w/v) at a
temperature ranging from 20 to 25°C. Then the reaction mixture is
cooled to ~5°C. A base (in solid form or as a concentrated solution
of at least 30%) is added slowly to the reaction mass by maintaining
the internal temperature -10 °C. The reaction mass is then slowly
warmed to ~25°C and stirred for 20 to 24h.
1.2 Crystallization of raw DOTA
Crystallized DOTA is obtained by adding an acid, such as HC1 , to the
1, ,7,10-tetraazacyclododecane-l , ,7,10-tetra acetic acid obtained
in step a ) to achieve a pH < 3 , followed by an heating and cooling
step. Preferably, a further washing step with a mixture of water and
a water miscible low boiling organic solvent using a ratio of 1:1.5
to 1:3 is also performed.
The reaction mass obtained in the previous step is first cooled to a
temperature of ~10°C, acidified with an acid to achieve a pH < 3 and
stirred for 30 min producing a precipitated.
The slurry is slowly warmed to -25 °C and heated to a temperature of
60-65°C to obtain a clear solution and stirred for about 10 min. The
reaction mass is slowly cooled to 5-10 °C and stirred for about 10
min. The slurry is filtered or centrifuged, suck dried for -10 min
and the bed was washed on the filter/centrifuge with a mixture of
water and a low boiling water miscible solvent in a ratio of 1.5.
The resulting solid is then suck dried for about 2h and dried under
vacuum conditions, such as in a tray drier, at 60+10 °C under
diminished pressure until dry (for 6-12 h ) . The crude DOTA salt is
thus obtained as a white powder in 70-80% yield and a content of
about 80% by HPLC .
1.3 Purification of DOTA
1.3.1 Treatment with a cationic resin
The raw material obtained according to the previous step is further
treated with a cationic resin followed by filtration, washing and
desorbing with a diluted ammonia solution.
A cationic resin in its hydrogen form ready-to-use (6 volumes with
respect to crude DOTA salt) is washed several times with water
(about 15 volumes) until the supernatant attained a pH of 4.0-6.0.
The crude DOTA salt obtained in the previous step is dissolved in
water (-10 volumes) and added to the pre-washed cationic resin,
taken in the reactor and stirred at ~25°C for 16h, until the
solution is free of DOTA. If necessary, an extra volume of resin is
added to remove DOTA form the solution.
The resin is then washed with water (-15 volumes) several times
until the supernatant attained a pH of 4.0-6.0. The resin is stirred
with diluted aqueous solution of a volatile base (15-20 volumes,
about 3%) for 10-20 min, allowed to settle for 20-30 in and the
supernatant is collected separately. The same is repeated for some
more times, ex. for 5 times, until no DOTA is observed in the
supernatant. The product fraction is then collected together and
concentrated to minimum volume (10 volumes with respect to crude
DOTA) , stripped-off with water (3 times 15 volumes) and the solution
(10 volumes) is unloaded, rinsed the reactor with 5 volumes water
and submitted for analysis to determine the DOTA assay. A small
portion of the product is then completely concentrated and analyzed
for ion content .
1.3.2 Treatment with anionic resin
The resulting solution of the previous step is treated with an
anionic resin.
Thus, an anionic resin (6-7 volumes with respect to assay corrected
cationic resin treated DOTA) is pre-treated by washing with water
(about 6 times 15 volumes) until the supernatant to attain the pH
8.0-10.0.
The treated DOTA obtained in the previous section is then added to
pre-washed anionic resin, taken in the reactor and stirred at 25°C
for 4-6h until the solution is free of DOTA. If necessary, an extra
volume of resin is added to remove DOTA form the solution. The DOTA
containing resin is washed with pure water (several times 15
volumes) until pH of the supernatant solution was 8-10.
1.3.3 Washing step
After treating the resulting solution with an anionic resin a twostage
washing treatment is performed with organic volatile acids,
first with an organic volatile acid at low concentration, and a
second stage with a higher concentration of an organic volatile acid
such as formic acid at 1-20% as preferred organic volatile acid.
Therefore, in the first washing step, the resin is stirred with
about 20 volumes of a low concentration of an organic volatile acid,
such as formic acid, for 20-30 min and allowed to settle for 10-15
min. Then the supernatant is removed. The same can be repeated for 1
to 2 times.
In the second washing step the resin is stirred with about 20
volumes of an aqueous solution of a volatile acid, such as formic
acid, at 1-20% for 20 to 30 min, allowed to settle for 10 to 15 min
and the supernatant is collected. The same can be repeated for
several times (~5) until no DOTA is found in the supernatant. The
supernatant is collected separately and checked for presence of DOTA
by HPLC assay.
Preferably, the product fractions are concentrated to minimum volume
(10 volumes with respect to cationic resin treated DOTA) , strippedoff
with water (4 times, 15 volumes) , then stripped-off with low
boiling water miscible organic solvent (3 times 6 volumes) , cooled
the reaction mass to ~25°C, a low boiling water miscible organic
solvent is then added (8 volumes) , cooled to ~10°C, stirred for 20
to 30 min at 10°C and centrifuged. The reactor is rinsed with a low
boiling water miscible organic solvent (2 volumes) , and the filtrate
is bed washed with reactor rinsed the same low boiling water
miscible organic solvent. The solid is suck dried for 30 min and
dried under vacuum conditions at 60+10°C for 6-12 h .
The obtained DOTA can be further purified by dissolving in water (5
to 10 volumes) at 40-60°C and slowly adding a water miscible low
boiling solvent (12-25 volumes). The slurry is stirred at ~25°C for
~lh, filtered and suck dried for ~3 h and dried, for example in a
vacuum tray dryer at 50-70 °C for 6-12 h .
2 . Synthesis of gadoterate meglumine
The DOTA of high purity obtained as described above can be used as
the ligand in the formation of the contrast agent gadoterate
meglumine, [Gd (DOTA) (H 20 )] (meglumine) .
For this purpose, DOTA obtained according to the process of the
present invention is added to Gd203 by known methods preferably with
excess of DOTA, most preferably in a molar ratio slightly over 2:1
to form an aqueous solution of a complex DOTA-Gd.
The temperature of the reaction solution required to form DOTA-Gd
complex was in the range from 80 to 120°C, preferably from 90 to
100°C, more preferably at a temperature of approximately 95°C.
As the kinetics of formation of the complex are very slow the
reaction typically takes 2-8h, preferably from 3-6h, more preferably
during approximately 4h.
During this time the pH of the reaction solution typically decreases
from ~3 to -1.5-1.6. In order to chelate the Gd(III) ion DOTA must
become fully deprotonated, which releases hydrogen ions into the
solution. Due to its basic properties meglumine is then added after
allowing the solution to cool to between 40 and 50 °C to balance the
negative charge of the complex. Once protonated it electrostatically
binds to the complex forming the meglumine salt and to increase the
solution pH.
Meglumine is added until the pH of the solution is between 6.9 -
7.8, to meet the pH range required to allow the solution to be
safely injected as contrast agent. Meglumine is used as an excipient
in many drugs; however, it can be present in the final solution in
excess because it can be well tolerated by the body. After stirring
for about half an hour, to ensure the reaction has gone to
completion, the reaction solution is then allowed to cool to room
temperature and filtered.
The obtained filtrate was analysed by HPLC-MS and was found to
contain gadoterate meglumine, showing that the quality of DOTA being
synthesised can successfully be used to synthesise a solution of the
contrast agent. The DOTA-Gd complex can be easily identified on the
ESI mass spectrum from the collection of peaks 1 m/z value apart,
centred at m/z 560. There are a number of [M + H ]+ peaks
corresponding to the dehydrated complex because gadolinium has six
stable isotopes, five ( Gd, Gd, 7Gd, Gd and 0Gd) of which all
have relative abundances greater than 14 % . Meglumine is also
evident on the mass spectrum with a [M + H ]+ peak at m/z 196.
Measurements
1 .In-process monitoring (IPC) of DOTA
The determination of content of 1 , 4 , 7 , 10-tetraaza-Cyclododecane
was determined in process control samples by using reversed phase
HPLC (High Performance Chromatography) with a gradient program and
DAD (Diode Array Detection) .
Chemicals and Reagents (as listed or equivalent) :
Acetonitrile - HPLC grade
Water - HPLC grade or Milli-Q-ater
Orthophosphoric acid - HPLC grade
Potassium dihydrogen phosphate - AR grade
Instrumentation and Equipment (as listed or equivalent) :
System: Agilent 1100/1200 series HPLC system with UV detector, or
equivalent .
Pump: Constant flow HPLC pump capable of running a gradient
Detector: DAD detector
Data Acquisition: An electronic data acquisition system is required
Chromatographic parameters :
Column: Prevail Organic Acid, (250 x 3.0) mm, 5 .0pm
Column Temperature: 30 °C
Detector Wavelength: 195 nm
Pump Configuration: Gradient
Flow rate: 0.44mL/min
Injection Volume: 5 L
Run Time: 40 min
Mobile phase A : 20mM KH2P04 in water at pH 2.5 using Diluent (see
below)
Mobile phase B : Acetonitrile : Mobile phase A (60:40)
Mobile phase C : Acetonitrile : Water (60:40)
Mobile phase D : Acetonitrile : Water (90:10)
Diluent: 0.1% Orthophosphoric acid in water
Needle wash: Acetonitrile
Blank: Diluent
Gradient Table:
System suitability preparation
Weigh about 100 g of 1, ,7,10-tetraaza Cyclododecane , chloroacetic
acid and DOTA standards into a 100 L volumetric flask, dissolve and
dilute to volume with diluent.
In process control sample preparation
Weigh about 600 mg of DOTA sample into a 50 mL volumetric flask,
dissolve and dilute to volume with diluent. (Prepare Test solution
in Duplicate) .
Retention time :
Analysis for DOTA
A . Purity and Assay (By HPLC)
Method Outline:
The method described above for In Process Control can be used for
Purity and Assay determination.
B . Chloride and Formate content (by IC)
This method was applied for the determination of Chloride and
Formate content of DOTA sample by IC (Ion Chromatography) .
Chemicals and Reagents :
Sodium bicarbonate: AR grade or equivalent;
Sodium carbonate: AR grade or equivalent;
Water: Milli Q grade or equivalent;
Sulphuric acid: AR grade or equivalent;
Instrumentation and Equipment:
System: 850 professional Ion chromatograph with auto sampler;
Detector: Conductivity detector
Preparations :
Eluent (3.2 mM Sodium Carbonate + 1 mM Sodium bicarbonate) :
Weigh about 0.32 g of Sodium Carbonate and 0.084 g of Sodium
bicarbonate in to a mobile phase bottle containing 1000 mL of milli-
Q-water. Mix well, filter through 0.45 m filter and degas.
Suppressor solutions (for anions) : 50mM sulphuric acid
Pipette out 2.8 mL of sulphuric acid in to a mobile phase bottle
containing 1000 mL of milli-Q-water . Mix well, filter through 0.45 m
filter and degas.
Chromatographic parameters :
Column: Metrosep A Supp 5 (250/4) with guard column
Run Time: 30 min
Flow: 0.7 mL/min
Maximum pressure: 15 MPa
MSM: Active
Per. Pump: Rate 3
Temp. Coefficient: 2.3%/°C
Injection volume: 20
Column temperature: 25°C
Preparation of Standard stock solution 1 :
Weigh accurately 165.0 mg of sodium chloride and 106.0 mg of Formic
acid into a 100 mL volumetric flask. Add about 10 mL of water and
sonicate to dissolve and make up the volume with water. Mix well.
Preparation of Standard stock solution 2 :
Pipette out 2.0 mL of this solution into 100 mL volumetric flask and
make up to the volume with water.
Preparation of Test solution (in Duplicate) :
Weigh accurately 50 mg of sample into a 50 mL volumetric flask and
make up the volume with water.
Inject the following as per the sequence:
Calculation of the content of anion by the following formula:
Using Peak area for quantification of chloride and formate content.
Anion content (%w/w) = Asm X Wst (mg) x 2 x 50 x AtWt of Anion x P
Ast x 100 x 100 x Wsa (mg) x Mol wt salt std x 100
Anion content (ppm) = Anion content in percentage x 1.000.000
Where ,
Asm is area of the anion peak in the sample;
Ast is area of the anion peak in the standard;
Wst is the weight of the standard;
Wsa is the weight of the sample;
P is the potency of the standard;
At t of Anion is the atomic weight of the anion;
Mol wt . salt std is Mol in weight of the salt used as standard
C . Ammonium content by IC:
Method Outline:
This method is applicable to determine the content and presence of
ammonium ion of DOTA sample. This method uses reverse phase ion
exchange chromatography.
Chemicals and Reagents (as listed or equivalent) :
Dipicolinic acid: AR grade
Water: Milli-Q-Water ;
Nitric acid: AR grade
Instrumentation and equipment (as listed or equivalent) :
System: Metrohm, model 850 compact IC
Pump: Constant flow pump
Detector: Conductivity detector
Data Acquisition: An electronic data acquisition system is required
Chromatographic parameters :
Column: Metrocep C-4 (4.6 x 250) mm x 4.0m , Serial No. 1080. 3137
Column temperature: 25°C
Detector: Conductivity detector
Pump configuration: Isocratic
Flow rate: 0.6 mL/min
Injection volume: 20 ]
Run time: 40 min
Mobile phase: 1.7 mM Nitric acid and 0.7 mM Dipicolinic acid in
water
Diluent: Water
Preparations :
Mobile Phase: 1.7 mM Nitric acid and 0.7 mM 2 ,6-Dipicolinic acid in
water
Weigh accurately about 0.12 g 2 ,6-Dipicolinic acid in to 1.0 L of
Milli-Q water and add 0.15 mL of concentrate Nitric acid (67-69 %
w/w) into it, sonicate to dissolve and filter through 0.45m membrane
filter.
Preparation of stock solution:
Weigh accurately 370 g each of ammonium formate into 100 L
volumetric flask dissolve and dilute with diluent up to the mark.
Preparation of standard:
Pipette out 1.0 mL of above stock solution in to a 100 mL of
standard volumetric flask and dilute with the diluent up to the
mark .
Preparation of Test solution (Duplicate) :
Weigh accurately 50 mg of sample into a 50 mL volumetric flask and
make up the volume with water.
Sequence table :
Equilibrate the HPLC system and column with mobile phase and inject
10 m of the solution as per the below sequence table.
Calculate the content of cation by the following formula
Cation content (%w/w) = Asm x Wst (mg) x 1 x 50 x AtWt of cation x P
Ast x 100 x 100 x Wsa (mg) x ol wt salt std x 100
Cation content (ppm) = cation content in percentage x 1.000.000
Where ,
Asm is area of the Ammonium (Cation) peak in the sample;
Ast is area of the cation peak in the standard;
Wst is the weight of the standard;
sa is the weight of the sample;
P is the potency of the standard;
AtWt of cation is the atomic weight of the cation; and
Mol wt . salt std is Mol in weight of the salt used as standard
Results are reported in weight per weight percentages.
D . Sodium content (By ICP) :
ICP -0ES (Inductively Coupled Plasma - Optical Emission
Spectrophotometry) .
Results are reported in weight per weight percentages.
EXAMPLES
All reagents used in the following examples were readily available
from commercial sources unless otherwise specified.
All reagents used to prepare DOTA and gadoterate meglumine,
including 1,4 ,7,10-tetraazacyclododecane (cyclen) , Gd203 and N -
methyl-D-glucamine (meglumine) , were obtained commercially and used
as received.
All ion exchange resins used were obtained commercially and used as
received unless otherwise stated in the experimental.
Amberlite IR120H was obtained from Fluka;
Purolite UCW912 6H+ from Purolite;
Amberlyst A260H obtained from Fluka;
Lewatit M800OH from Lanxess,-
Purolite UCW5072 OH from Purolite;
Lewatit MonoPlus M 600 resin was obtained from Lanxess in the
chloride form and converted to the hydroxide form by stirring 5 L of
the resin in 12 L of 1M NaOH for 1 hour, washing extensively with
deionised water and then repeating the process once more.
Aqueous solutions of NaOH (29 w%) , HC1 (36 w%) , NH (25 w%) , HCOOH
(85 w%) and CH3COOH (99 w%) were obtained commercially and diluted
using deionised water as required.
In Process Control (IPC) using HPLC, according to the previous
section (measurements) : area percentages of the peaks, with regard
to total peak area, are used to report values for cyclen as starting
material and Dota as product .
For the raw DOTA, yield is reported as the number of moles of
isolated product (DOTA bis hydrochloride, without correcting for
assay) per number of moles of cyclene starting material .
For purification steps, the yield is reported as the weight of
isolated pure Dota per weight of crude DOTA input.
HPLC purity is reported according to the method described in the
measurements section, using a standard sample of DOTA (obtained
according to the International Conference on Harmonisation (ICH
guidelines) , unless otherwise stated.
Example 1 - Kinetic study for DOTA production during the step a ) of
the process
This example illustrates the kinetics of step a ) of the process for
production of DOTA according to the present invention. The reaction
was conducted for 72h at a temperature of 25°C, using chloroacetic
acid as the halo-acetic acid (4.3 eq.) and sodium hydroxide as a
base (11.0 eq.) in water (12x - 12 parts of water for 1 part of
initial amount of cyclen in weight) and samples were taken at
different times for in-process control analysis via HPLCchromatography
(see above) .
The results are shown in Table 1 as surface area % in the HPLC
chromatograms and the conversion value (%) is found according to the
following formula:
Conversion (%) = 100% - X
wherein X is the amount of cyclen.
Measured impurities comprise unreacted alkylating acid, intermediate
cyclen derivatives and counterions introduced by the halo-acetic
acid and by the treatment with a base .
Table 1: Kinetic study data
*RRT = Relative Retention Time
These results show that in the settled conditions the reaction is
completed after 20h, i.e at that time the conversion rate is 100%,
which means that the starting material was completely consumed and
therefore, that the process of the present invention can be carried
out at temperatures as low as 20 °C.
Example 2 - Influence of the temperature on quantity and quality for
DOTA production in step a ) of the process
This example illustrates the temperature ranges useful in the
process for obtaining DOTA, according to the present invention. The
reaction was conducted for 2Oh at temperatures ranging from 20 °C to
100 °C and pH _> 10, using chloroacetic acid as a halo-acetic acid
(4.3 eq.) and sodium hydroxide as a base (11.0 eq.) in water (12x -
12 parts of water for 1 part of initial amount of cyclen in weight) .
Samples were taken for evaluation on the reaction progress and the
amounts of cyclen and DOTA were measured by HPLC.
The yield and the purity of isolated DOTA were measured and used
respectively as indicators of quantity and quality of the product
obtained in each experiment conditions, as shown in Table 2 .
Table 2 : Influence of the temperature on DOTA production
For the experiment at 100°C, the IPC after 6h : Dota = 43.07%,
Cyclene = 0%.
The results of Table 2 show that the process for producing DOTA,
according to the present invention, can be performed at temperatures
from 20 °C and still good yields (> 70%) and good quality (> 80%) are
obtained. In experiment 1 , there was no cyclen left after 2Oh of
reaction and DOTA could be obtained in good yield an quality. At
100°C, the cyclen has already disappeared after 6 hours. However,
the DOTA peak reached its maximum only at about 42%.
Higher temperatures, such as at 60 °C can acelerate the reaction
achieving also good results in terms of yield but they also increase
the costs of producing DOTA, particularly when industrial processes
are concerned. Furthermore, higher temperatures result in longer
times for the precipitation of the product after acidification.
Example 3 - Influence of the halo-acetic acid amount on time,
quantity and quality for DOTA production in step a ) of the process
Cyclen was reacted with different amounts of chloroacetic acid, as
an example of halo-acetic acid, at 20°C to check for the product
formation selectivity (IPC), whilst the other production factors
were maintained. The amount of chloroacetic acid is indicated in
Table 3 as mole equivalents with regard to cyclen and varied between
4.3 eq as in the conventional process (COMP = Comparative), 5.0 eq.,
6.0 eq. and 8.0 eq. The quantity and quality of DOTA was determined
as described in the previous section, and shown in Table 3 .
Table 3 : Influence of the equivalents of chloroacetic acid on DOTA
production
The reactions with 6 and 8 equivalents of chloroacetic acid,
respectively experiments nr. 3 and 4 , showed maximum selectivity
after 2Oh. These conditions afforded the good yield and high purity
levels of isolated DOTA. In experiment 4 , the purity was lower than
the one obtained in reactions with lower amounts of halo-acetic
acid. The reaction nr. 2 showed short reaction time (-24 h ) and
produced the best DOTA yield with the highest purity level. Reaction
nr.l was the slowest (72h) resulting in reasonable yield of DOTA
with high purity level.
Therefore, results of Table 3 clearly show that it is possible to
obtain good purity and yield values of isolated DOTA and that the
best results are achieved when 5.0-6.0 eq. of chloroacetic acid are
used .
Example 4 - Influence of the amount of the base on time, quantity
and quality for DOTA production in step a ) of the process
Cyclen was reacted with 4.3 equivalents of chloroacetic acid, using
different amounts of NaOH, as an example of a base, at 20°C, in
order to achieve pH values > 10 and evaluate the pH role on DOTA
yield, whilst the other production factors were maintained. The
amount of NaOH is indicated in Table 4 as mole equivalents with
regard to cyclen. The quantity and quality of DOTA was determined as
described in the previous section, and shown in Table 4 .
Table 4 : Influence of the amount of the base on DOTA production
Results of Table 4 show that using larger excess of the base (11
equivalents compared with 9 equivalents) , and thus generating pH
values above the prior art given values (-10) and particularly
generating pH values above 13, has no adverse effects in what
regards purity and yield of DOTA and therefore, the process can be
developed at higher pH values. Adding the base as a 30% solution
yields comparable results with solid addition.
The advantage of this is that since it is no longer necessary to
keep strict control of pH values and the base is added all at once,
this step is performed in a more easy and smooth manner. It was also
possible to observe that the amount of base is dependent on the
amount of halo-acetic acid and should be at least 2 equivalents of
said base per mole of acid to achieve good yield and purity levels
of DOTA.
Example 5 - Influence of the amount of the acid on quantity and
quality for DOTA crystallization in step b ) of the process
This example illustrates the effect of adding an acid on the quality
of crude DOTA, especially on the cation impurity-related content
(Table 5 ) .
Experiments were conducted on DOTA obtained according to step a ) of
the process of the present invention and at 20°C during 20/24 h . In
the prior art process, HC1 was used to acidify the reaction mass to
pH of 2-2.5. Several reactions were conducted with different amounts
of concentrated hydrochloric acid (7.3, 9.3, 11.4, 15.6, 20.7, 31.1
equivalents respectively with regard to cyclen) .
Table 5 : Influence of different ratios of concentrated acid on
the quantity and quality for DOTA crystallization
The results of Table 5 show that in the conditions of experiments
to 4 , respectively with 7.3, 9.3, 11.4 and 15.6 eq. of HC1,
possible to obtain crystallized DOTA with low contents of CI .
Table 5 also shows that in the conditions of experiments 2 to 4 ,
respectively with 9.3, 11.4 and 15.6 eq. of HC1 , it is possible to
obtain crystallized DOTA with particular low contents of Na+. The
yields in the tables are not corrected for contaminants, this
explains the figures of more then 100 % for the items 5 to 7 .
It is also clear, from Table 5 , that when increasing or decreasing
the HCL concentration results in higher sodium content in the crude
DOTA. Furthermore, it is possible to observe that increases in HCL
above 15.6 eq. result in higher chloride content and decrease the
purity of crude DOTA.
Thus, by using values of HCL ranging from 9.3 eq. to 15.6 eq. it is
possible to obtain crystallized DOTA with low sodium content and
good purity (above 80%) .
Example 6 - Effect of heating and cooling the reaction mixture
during the crystallization step on the quantity and quality DOTA
This example illustrates the effect of heating and cooling the
reaction mixture on the yield and quality of crude DOTA (Table 6 ) .
The reaction was carried out with 5.0 equivalents of chloroacetic
acid as the halo-acetic acid and 11.0 equivalents of NaOH as the
base at 20°C during 20/24h. In experiments 2 , 3 and 4 , the slurry,
containing precipitated solid after acidifying to pH 0.5, was heated
to obtain a clear solution at a temperature of 65 °C, which was then
slowly cooled and stirred for 10 min at 5°C. The precipitated solid
was filtered and taken forward for further purification. Experiment
1 was performed without the heating/cooling step.
Table 6 : Effect of the heating and cooling step on DOTA
quantity and quality
1 0% 59.6% 91, 00% 75 .72% 6 .6% 24 .4%
2 0% 67 .79% 90, 20% 83 .34% 1 .15% 18 .21%
3 0% 68 .61% 86, 60% 85 .13% 1.57% 17 .57%
4 0% 67 .13% 81,90% 83 .89% 1.57% 16.47%
These results show that by performing the heating and cooling step
during the crystallization of DOTA the quality of the crude DOTA
obtained was better, in comparison to the crude DOTA obtained using
the conventional procedure, where no heating/cooling process was
performed, as HPLC purity (determined by area % ) was higher and both
sodium and chloride were lower.
Example 7 - Effect of heating and cooling step for longer times
during the crystallization step on the quantity and quality DOTA
The previous experiment was repeated with different times for the
cooling and heating step in order to investigate whether is was
possible to obtain even better results for the crude DOTA yield.
The reaction was heated to 65 °C during different times, then cooled
to 5-10°C or to temperatures below 0°C, at different times, and the
slurry was then filtered, dried and the results measured (Table 7 ) .
Table 7 : Effect of the heating and cooling step for longer times on
DOTA quantity and quality
Heating: 65°C Effect of
for 10 min short
3 66.97% 1.23% 16.20% 100%
Cooling: -5 to cooling at
0°C for 10 min <0°C
Heating: 65°C slurry
for 10 min
4 81.04% 1.3% 11.39% 67% stirred at
Cooling: 25°C
RT
for 30 min
Heating: 65°C
slurry
for 10 min
5 83 .56% 0.58% 16.25% 93% stirred at
Cooling: 10°C
10°C
for 10 min
The results of Table 7 show that heating or cooling the reaction
mixture for longer duration decreases the purity of the obtained
crude DOTA. Similar results were obtained with stirring at RT (room
temperature) for longer times.
When heating the reaction for longer times (~2h) , whilst maintaining
the cooling procedure, similar purities are obtained showing that
there is no benefit on prolonging the heating period (comparison
between Exp. 1 and 5 ) .
When cooling the reacting for longer times (~2h) , whilst maintaining
the heating procedure, lower purities are obtained (comparison
between Exp. 2 and 5 ) .
When cooling the reaction with higher temperatures (-25 °C) , whilst
maintaining the heating procedure, similar purities are obtained
(comparison between Exp. 4 and 5 ) but the yield is higher at the
lower temperature .
When cooling the reaction with even lower temperatures (~ -5°C) ,
whilst maintaining the heating procedure, lower purities are
obtained (comparison between Exp. 3 and 5 ) .
The procedure by heating at 65 °C for a short period of time (-15
minutes) followed by a cooling step at 10°C during 10 minutes (Exp.
5 ) gives the best results in terms of yield (2.58 g/g cyclene) and
quality (assay 85.36 % , Na+ 0.58%).
Example 8 - Treatment with a cationic resin
This example illustrates the role of treating the crystallized DOTA
with a cationic resin to reduce the content of anion impurities
(Table 8 ) .
The crude DOTA obtained in the previous steps was dissolved in 10
volumes water and adsorbed onto pre-washed cationic resin Amberlite
IR-120 (Fluka) at 20 °C during ~16h. The resin was washed with water
until a pH of 4.5-6.0 (supernatant) was attained. Then the resin was
washed with 3% aqueous ammonia solution (6 4 volumes with respect
to crude DOTA) . The product fractions were concentrated and dried to
yield 75-85% DOTA.
Table 8 : Treatment with cationic resin
The results of Table 8 show that the anion content of DOTA (chloride
ion) was reduced to an acceptable limit (below 0.1%) . Further
reduction is realised in the next steps. However the treatment
introduces ammonium ion into DOTA (as ammonia was used for elution)
and sodium content is only marginally decreased proving that any
cations must be removed in early stages of the process in order to
achieve higher yields and the required purity of DOTA.
Example 9 - Treatment with an anionic resin
This example illustrates the role of treating the obtained DOTA, as
described in the previous step, with an anionic resin to reduce the
content of ammonium impurities.
The cationic treated DOTA dissolved in 10 volumes water was adsorbed
onto pre-washed anionic resin Amberlyst A26 OH at 20°C during 4-6h.
The resin was washed with water to attain the pH 8-10 (supernatant) ,
followed by washing with diluted aqueous formic acid solution as the
preferred organic volatile acid to remove residual ammonium,
followed by 1% aqueous formic acid solution. The product fractions
were concentrated and dried (Table 9 ) .
Details of the washing procedure are given in Example 10, Table
10.1.
Table : Treatment with an anionic resin
These results show that the ammonium content was reduced within the
acceptable required limit (below 0.1%) . The level can be further
reduced by recrystallizing in water/solvent if necessary. However,
it introduced formate ion into the system (formate amount not
presented) . The yield obtained with this process was 80-99% and is
possibly mainly influenced by the scale of the reaction. Better
results are obtained when using higher amounts of reactants which
are easier to be accurately measured.
Example 10 - Effect of dilute volatile acid wash in the first step
of the washing procedure d ) of the process
In experiments 2 and 3 in the previous example, the volatile acid
(dilute) used in the first stage of the washing was formic acid and
the concentration was 0.1% for Exp. 2 and 0.02% for Exp. 3 . For the
second stage of the washing, 1% formic acid was used also for the
mentioned experiments. In experiment 4 , the dilute volatile acid
used in the first stage of the washing was acetic acid, 0.26%,
calculated to have the same pH as the 0.02% formic acid. Details on
the procedure are given in table 10.1.
Table 10.1: Procedure for anionic resin wash
For experiments 1 and 2 in table 10.1, the different fractions
obtained from the first and second stages of the washing step were
also analysed for DOTA content, to prove that the dilute volatile
acid used in the first stage of the washing does not remove DOTA
from the anionic resin. The result is presented in tables 10.2 and
10 .3 .
Samples of each stage of the washing step of the above mentioned
experiments 1 and 2 were collected and analysed for DOTA content by
HPLC. In both instances it is shown that the diluted formic acid
washes used in the first step only remove minor quantities of DOTA
from the resin. The second stage using higher concentration of the
volatile acid releases the bulk of DOTA from the resin washing it
out to the eluate. The releasing and washing of DOTA from the resin
may continue during the several repetitions of the second stage
until only a marginal amount of DOTA is present in the eluate.
Table 10.2: Exp. 1 from table
10.1
These results of Tables 10.2 and 10.3 show that the first stage of
the washing step performed with 0.1% formic acid does not elute DOTA
from the anionic resin - no product (DOTA) was detected in the
eluate, thus only removes the ammonium.
The purified DOTA was obtained by elution from the resin only with
higher concentrations of formic acid (1.0%) during the second stage
of the washing step allowing to obtain good yield of DOTA as
required to produce contrast agents.
Similar results were obtained when acetic acid was used in the first
stage of the washing step instead of formic acid. The first stage
was performed with 0.26 % acetic acid (calculated to have the same
pH as a 0.02% formic acid solution) .

CLAIMS
A process for preparing 1,4,7,10-tetraazacyclododecane-l ,4,7,10-
tetraacetic acid (DOTA) including salts and hydrates thereof of
general formula (I)
H OC COOH
N N
HOOC -COOH
Formula (I
comprising the following steps:
a ) reacting the cyclen 1 , ,7 ,10-tetra-azacyclododecane and a haloacetic
acid with a base at a pH > 10;
b ) crystallizing the DOTA obtained in step a ) by addition of an
acid to achieve a pH < 3 , followed by a heating step and a
cooling step, wherein the heating step is performed at a
temperature in the range from 50°C to 100°C, for at least 5
minutes, and the cooling step is performed at a temperature in
the range from 5°C to 25°C, for at least 5 minutes;
c ) treating the raw material obtained in b ) with a cationic resin
and then desorbing DOTA with a volatile base solution;
d ) further treating the resulting solution of c ) with an anionic
resin;
e ) washing the product of d ) in a two-stage wash, with a first
organic volatile acid with a pKa less than 5 until DOTA starts
to be released from the anionic resin, and a second stage with
an organic volatile acid with a pKa less than or equal to the
pKa of the first organic volatile acid and/or with a higher
concentration than the first organic volatile acid to release
DOTA from the resin.
2 . The process, according to claim 1 wherein the step a ) is
performed at a pH > 13 .
3 . The process, according to claim 1 or 2, wherein the amount of
the halo-acetic acid in step a ) is at least 4 equivalents with
regard to the initial amount of cyclen and the amount of said
base is at least two times the number of equivalents of the
halo-acetic acid.
4 . The process, according to any of the claims 1 to 3 , wherein the
halo-acetic acid is selected from the group of iodoacetic acid,
bromoacetic acid and chloroacetic acid.
5 . The process, according to any of the claims 1 to 4 , wherein the
base in step a ) is an alkali metal hydroxide.
6 . The process, according to claim 1 , wherein the acid added in
step b ) is selected from the group of HCl, H2S0 , HN03, HBr, HI,
HC10 4 .
7 . The process, according to claim 1 , wherein the heating and
cooling step of b ) occurs respectively at a temperature in the
range from 50 to 60 °C and 5 to 10 °C.
8 . The process, according to claim 1 , wherein a washing step with a
mixture of water and a water miscible low boiling organic
solvent in a ratio from 1:1.5 to 1:3 (weight/weight) is
performed between the step b ) and step c ) .
9 . The process, according to claim 1 , wherein in step e ) the
organic volatile acid used in the first washing-stage is formic
acid or acetic acid in aqueous solution at a concentration
between 0.01-0.1% or 0.1-0.3% respectively, and the organic
volatile acid used in the second washing-stage is formic acid in
aqueous solution at a concentration in the range from 1 to 20%.
10. A process for preparing gadoterate meglumine comprising the
process as described in any of the previous claims and further
comprising:
f ) adding Gd 03 to the DOTA obtained according to any of the
claims 1-9;
g ) adding meglumine to the complex DOTA-Gd obtained in the step
a )