Process Development of the Synthesis and Purification of a Reactive Immuno-PET Conjugate Intermediate

Article abstract of DOI:10.1021/acs.oprd.5b00379

We describe a practical multigram scale synthesis of the reactive immuno-PET linker-chelator GNE-605 from the commercially available starting material desferrioxamine B mesylate. A modified three-step synthetic procedure was developed that significantly reduces the amount of solvents and reagents required as compared to previous literature. Additionally, due to the unique handling challenges posed by this reactive intermediate, a liquid chromatography purification and lyophilization procedure was developed to allow for the first isolation of GNE-605 as a solid material.

Full text of DOI:10.1021/acs.oprd.5b00379

Subscriber access provided by UNIV OF WISCONSIN OSHKOSH
Communication
Process Development of the Synthesis and Purification
of a Reactive Immuno-PET Conjugate Intermediate
Dr. Diane E. Carrera, Tina Nguyen, Colin D. Medley, Yi Li, Remy Angelaud, and Francis Gosselin
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.5b00379 • Publication Date (Web): 11 Jan 2016
Downloaded from http://pubs.acs.org on January 11, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 19
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Process Development of the Synthesis and Purification
of a Reactive ImmunoꢀPET Conjugate Intermediate
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Diane E. Carrera,*, † Tina Nguyen, ‡ Colin Medley,‡ Yi Li,‡, Remy Angelaud,† and Francis Gosselin†
†Small Molecule Process Chemistry, Genentech Inc., 1 DNA Way, South San Francisco, CA 94080,
USA.
‡Small Molecule Analytical Chemistry and Quality Control, Genentech Inc., 1 DNA Way, South San
Francisco, CA 94080, USA
CORRESPONDING AUTHOR
*Eꢀmail: carrera.diane@gene.com
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 19
1
2
3
TOC GRAPHIC
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 19
Organic Process Research & Development
ABSTRACT
1
2
3
4
5
6
7
8
9
We describe a practical multiꢀgram scale synthesis of the reactive ImmunoꢀPET linkerꢀchelator
GNE-605 from the commercially available starting material desferrioxamine B mesylate. A modified 3ꢀ
step synthetic procedure was developed that significantly reduces the amount of solvents and reagents
required as compared to previous literature. Additionally, due to the unique handling challenges posed
by this reactive intermediate, a liquid chromatography purification and lyophilization procedure was
developed to allow for the first isolation of GNE-605 as a solid material.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS: ImmunoꢀPET, radioimmunoconjugate, ADC, conjugate
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 19
INTRODUCTION
1
2
3
Positron emission tomography (PET) is a valuable tool for studying the uptake of
4
5
6
molecules within the body with wide ranging applications in oncology, cardiology, neuroimaging and
18
89
67
7
8
9
pharmacokinetics. By using a positronꢀemitting radionuclide such as F, Zr or Ga as a tracer, a
threeꢀdimensional image can be generated to show where the tracer is located within the body. A new
class of nonꢀtherapeutic antibody small molecule conjugates known as ImmunoꢀPET are of increasing
interest to the drug development community due to their potential application as an imaging tool in the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
clinic.1 While a number of platforms have been developed, the Zr chelating PET ligand DFOꢀNSucꢀ
89
mAb (Scheme 1) was of interest for development as a generic platform that could enable real time
imaging of monoclonal antibodies (mAbs) and antibody drug conjugates (ADCs) in the clinic. With
typical dosing of the conjugate ranging from 5ꢀ50 mg, a single gram of the small molecule is sufficient
to provide multiꢀkilogram quantities of conjugate depending on the expected drugꢀantibody ratio (DAR).
A generic scheme of the ImmunoꢀPET synthesis is given in Scheme 1. The first step in the ImmunoꢀPET
synthesis involves preparation of the small molecule linker chelator GNE-605, also referred to as DFOꢀ
SucꢀTFPꢀFe, then conjugation to the antibody of choice, iron stripping, purification and lyophilization
89
afford the nonꢀchelated ImmunoꢀPET conjugate (DFOꢀNSucꢀmAb). Final chelation with Zr gives the
89
active ImmunoꢀPET conjugate ZrꢀDFOꢀNSucꢀTFP. In order to successfully implement the Immunoꢀ
PET platform, we first needed to access the reactive linker chelator GNE-605 and prepare gram
quantities of this compound, which had previously only been synthesized on milligram scale.
4
ACS Paragon Plus Environment

Page 5 of 19
Organic Process Research & Development
RESULTS AND DISCUSSION
1
2
3
Scheme 1. Immuno-PET Imaging Agent Synthesis
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The published synthesis2 of the linkerꢀchelator GNE-605 (1) is given in Scheme 2. This threeꢀstep
route starts with the commercially available mesylate salt of desferrioxamine B 2,3 a strong chelator of
Fe(III) which has been approved for commercial use in the treatment of acute iron poisoning4. The
terminal primary amine undergoes amidation in the presence of succinic anhydride and base to give the
free acid 3. Chelation with Fe(III) in the form of iron trichloride serves to protect the N-hydroxyamide
functional groups and is followed by a 1ꢀethylꢀ3ꢀ(3dimethylaminopropyl)carbodiimide (EDC) mediated
coupling with 2,3,5,6ꢀtetrafluorophenol (TFP) to furnish the final compound 1. The crude solution from
this final reaction is then purified by preparative reverseꢀphase chromatography and the compound is
taken on to the mAb conjugation step as a solution in acetonitrile.
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 19
Scheme 2. Literature Route to GNE-605 (1)a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a Reagents and conditions: (a) succinic anhydride, pyridine, 25 ˚C, 24 h; (b) FeCl3 in 0.5 M HCl, 0.1 M Na2CO3(aq), 0.9% NaCl (aq), 25 ˚C; (c) 2,3,5,6ꢀ
tetrafluorophenol, EDC•HCl, MeCN, 25 ˚C.
In replicating this literature route in our laboratories, we identified several key issues that would
pose a challenge for multiꢀgram scale synthesis. The first step, installation of the succinate linker, used
the undesirable solvent pyridine, a large excess of succinic anhydride (1000 mol%) and workup was
accomplished by acidifying to cause product precipitation and washing with 1000 volumes of 0.01 M
aqueous HCl solution, which would have equated to 20 L on 20 g scale. Additionally, due to the small
particle size of the solid product 3, the initial filtration and subsequent cake washes with water and
heptane were exceedingly slow and resulted in a very dense cake. In the second step, the low solubility
of the isolated acid 35 again gave rise to the need to use very high solvent volumes (60 mL/g of aqueous
0.9% NaCl and 22 mL/g of aqueous 0.1 M Na2CO3) in order to completely dissolve the starting material
and achieve complete chelation of Fe(III). For the final coupling step, large excesses of both the
coupling reagent EDC•HCl (2900 mol%) and TFP (1000 mol%) were required to achieve complete
conversion to the phenol, presumably as a result of hydrolysis of the coupling reagent as the reaction
was performed in an aqueous media. The final and most critical issue facing the literature synthesis
arose from the fact that compound 1 had only been produced on milligram scale6 and had never been
6
ACS Paragon Plus Environment

Page 7 of 19
Organic Process Research & Development
isolated. As part of a mAb imaging project, we needed to develop synthesis and isolation conditions that
would provide a minimum of 20 g of GNE-605. Our main focus on the process chemistry front was to
develop new reaction conditions amenable to scaleꢀup that would eliminate the use of aqueous reaction
media, reduce solvent volumes and improve reagent stoichiometry. Clearly, both new purification and
isolation methods that could produce gram quantities of 1 in a solid form were much needed.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Process development for the succinate addition step to convert the DFO starting material 2 to the
free acid 3 focused on removing the environmentally unfriendly solvent pyridine, reducing or
eliminating the extensive aqueous workꢀup and avoiding the slow filtration during isolation. We
evaluated a series of organic solvents (Table 1) and found that polar aprotic solvents performed well in
this transformation. Both DMF and NMP performed better than pyridine, achieving 100% conversion to
3 after 16 hours (entries 7–8) as opposed to the 24 hours required in the literature system. These solvents
also exhibited a greater solubilizing effect on the starting material which allowed the amount of solvent
used to be reduced from 15 volumes to 10 volumes. Furthermore, increasing the temperature to 50 ˚C
led to a faster reaction rate with complete conversion achieved in only two hours (entry 9). This increase
in reaction rate allowed for additional process improvements such as reducing the amount of succinic
anhydride used from 1000 mol% to only 115 mol%. The final modification to the reaction conditions
was the addition of 120 mol% of N,Nꢀdisopropylethylamine (DIPEA) to effectively generate the
freebase of the starting material, which had previously been accomplished by using pyridine as the
reaction solvent. The use of large amounts of aqueous media to workꢀup the reaction and precipitate the
product was avoided by telescoping the reaction mixture directly on to the chelation step. This also
served to eliminate the extremely slow filtration that was observed when isolating 3.
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 19
Table 1: Solvent Evaluation for Succinate Amidationa
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Entry
Solvent
THF
Temp (˚C)
Time (h)
conversion(%)b
1
2
3
4
5
6
7
8
9
25
25
25
25
25
25
25
25
50
48
48
48
48
48
48
16
16
2
70
66
IPA
EtOAc
iPrOAc
DCM
MeCN
NMP
49
61
74
82
100
100
100
DMF
DMF
a Reaction performed with 2 (25 mg, 38.1 ꢀmol,100 mol%), succinic anhydride (77 mg, 77.0 ꢀmol, 200 mol%), N,Nꢀdiisopropylethylamine(10 ꢀL, 57.1
ꢀmol 150 mol%) and solvent (0.5 mL, 20 mL/g) b Determined by HPLC analysis with UV detection at 220 nm.
Modification of the chelation conditions focused primarily on replacing the aqueous media used
in the literature procedure with an organic medium that would be compatible with the telescoped
material from the modified succinate amidation procedure. It was found that the addition of 100 mol%
of FeCl3 dissolved in 2.5 volumes of DMF at room temperature resulted in complete chelation to form
chelated 4 after only 15 minutes (Scheme 3). The chelation could be monitored visually by a color
change of the white slurry to dark red and also by HPLC assay showing 100% conversion to a new peak
at 420 nm.7 A corresponding increase in solubility was also noted at this stage. The resulting reaction
mixture was used directly in the next stage.
8
ACS Paragon Plus Environment

Page 9 of 19
Organic Process Research & Development
Scheme 3. Chelation with FeCl3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
For the installation of active TFP ester to generate product 1, process development was focused
on optimization of the coupling reagent, replacing the aqueous reaction media with an organic solvent
and developing workꢀup conditions that would remove process impurities. Examining the reagent used
in the literature procedure, EDC•HCl, it was found that the coupling reaction performed much better in
DMF than NMP (Table 2, entries 1–2). Additional coupling reagents such as CDI and
propanephosphonic acid anhydride (T3P) were also examined but neither resulted in high levels of
product formation (Table 2, entries 3–6). In the case of T3P, the chelated starting material 4 underwent
decomposition. In addition to changing the reaction media to DMF, the amount of EDC•HCl used was
reduced from 2900 mol% to 200 mol% and the amount of TFP used was reduced from 1000 mol% to
200 mol%.8 Neither change had a negative impact on the reaction rate as the reaction reached
completion in 5 hours at 25 ˚C (Table 2, entry 7).9
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 19
1
2
3
Table 2. Evaluation of Coupling Reagentsa
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Entr
y
Reagent
EDC•HCl
EDC•HCl
T3Pb
Solvent
NMP
DMF
NMP
DMF
NMP
DMF
DMF
Time (h)
1(A% HPLC)c
1
12
12
24
24
24
24
5
45
87
0
2
3
4
T3Pb
0
5
CDI
4
6
CDI
3
7d
EDC•HCl
88
a
Reaction performed with 4 (38 ꢀmol, 100 mol%), coupling reagent (76 ꢀmol, 200 mol%), 2,3,5,6ꢀtetrafluorophenol (76 ꢀmol, 200 mol%), N,Nꢀ
diisopropylethylamine (76 ꢀmol, 200 mol%) and 0.5 mL of solvent (20 mL/g). Reaction performed with T3P (57 ꢀmol, 150 mol%), 2,3,5,6ꢀ
tetrafluorophenol (46 ꢀmol, 120 mol%) and 4ꢀmethylmorpholine (57 ꢀmol, 150 mol%). Determined by HPLC analysis with UV detection at 254 nm.
Reaction performed with 4 (60.9 mol, 100 mol%), EDC•HCl (122 mmol, 200 mol%), 2,3,5,6ꢀtetrafluorophenol (122 mmol, 200 mol%), N,Nꢀ
b
c
d
diisopropylethylamine (122 mmol, 200 mol%) and 400 mL of DMF (10 mL/g)
Workꢀup of the coupling reaction was accomplished by dilution with dichloromethane and washing
with a 13% NaCl aqueous solution to remove the reaction solvent DMF and several reaction byꢀproducts
such as succinic acid, DIPEA•MsOH, DIPEA•HCl and 1ꢀethylꢀ3ꢀ(3ꢀdimethylaminopropyl)urea. Two
reaction impurities, residual acid 4 and succinamide 510 (Scheme 4) resulting from intramolecular
displacement of the TFP leaving group, were also removed by these washes. The use of a 13% brine
solution was necessary to achieve suitable phase separation during these washes as attempts at using
pure water resulted in a large rag layer and high amounts of hydrolyzed product 4.
10
ACS Paragon Plus Environment

Page 11 of 19
Organic Process Research & Development
Scheme 4. By-products Resulting from Attempted Recrystallization
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
All attempts to isolate active ester 1 via crystallization were unsuccessful and resulted in the
isolation of degradation products (Scheme 4). The TFP ester 1 rapidly underwent hydrolysis to give 4
when exposed to adventitious water during solvent swapping procedures and when filtered, stability
studies showed that this hydrolysis occurs rapidly at 25 ˚C (rate > 20%/h in a 70:30 mixture of
acetonitrile/water, Figure 1). Upon exposure to base or extended heating in anhydrous solvent,
compound 1 underwent an intramolecular cyclization reaction to give complete conversion to
succinamide 5 (Scheme 4). Furthermore, no crystalline form for this material has been identified so the
prospects of achieving a successful crystallization procedure were quite low.
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 19
1
2
3
Figure 1. Stability of Linker-Chelator 1 in 70:30 MeCN/water Eluenta,b
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a Determined by HPLC analysis with UV detection at 254 nm. b Purity of fraction immediately after elution from C18 column is the fraction purity at t = 0.
As crystallization was not an option for isolation of the highly reactive linkerꢀchelator 1, we turned
our attention to establishing a chromatographic means of isolation. After observing decomposition
products with normal phase silica, amineꢀmodified silica and alumina, a successful purification was
obtained by using C18 reverseꢀphase silica.11 A key observation was that while chromatography
successfully removed a number of small impurities from the crude product, it also resulted in a
significant amount of hydrolysis as a result of using water as the mobile phase. In light of these results,
the chromatography procedure was modified in several key ways in an attempt to minimize hydrolysis.
The starting ratio of the mobile phase was changed from 5% to 10% acetonitrile in water in order to
reduce the amount of time the material spent on the column. Also towards this end, the continuous
gradient was changed to a step gradient as this enabled high column loadings of up to 14.5%12 using a
12
ACS Paragon Plus Environment

Page 13 of 19
Organic Process Research & Development
capture/release method rather than a true chromatographic separation. Most importantly, we also
examined the amount of hydrolysis that took place in the collected fractions after elution. Not
surprisingly, hydrolysis occurred very rapidly when the fractions were left at 25 ˚C with 12%
degradation observed after only 30 minutes (Figure 1). Holding the fractions at 0 ˚C reduced the rate of
hydrolysis significantly but did not eliminate it completely. Freezing the fractions in dry ice at –78 ˚C
gave the best results with only minimal hydrolysis observed as the material purity decreased by only two
percent from 91 A% HPLC to 89 A% HPLC over 24 hours. Implementing these improvements to the
purification procedure resulted in a dramatic increase in final product purity (Table 3). Whereas the
highest observed purity using the continuous gradient and not freezing the fractions was only 86 A%
HPLC (entry 1), up to 92 A% HPLC purity was achieved using the step gradient and freezing the
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
fractions before combining them for lyophilization (entry 3).13
Table 3. Optimization of the C18 Purification Procedure
Entr
y
1b
Scale
(g)a
Hold Temp
(˚C)b
Yield(%)
Gradient
Purityd
86
c
continuou
20.0
25
23
26
s
continuou
s
2
20.0
25
78
3
4
40.0
40.0
step
–78
–78
14
22
92
90
step
a Amount of amine starting material 2 taken through telescoped synthesis and workꢀup procedure prior to purification; b Fractions were sampled for
HPLC analysis then either held at rt or frozen in a dryꢀice acetone bath until column purification and analysis was complete; c Isolated yield after
lyophilization; d Determined by HPLC analysis with UV detection at 254 nm.
Having developed a telescoped procedure for the linkerꢀchelator synthesis and an efficient C18
reverseꢀphase purification procedure, we next embarked upon a campaign to deliver gram quantities of
GNE-605 for conjugation (Scheme 5). Our improved process was demonstrated on a 609 mmol, 400 g
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 19
scale of DFO mesylate to produce 100 mmol, 86.3 g of active ester 1 as a red solid with 92.3 A% HPLC
purity and 88.7 wt% assay in 16% overall yield after chromatographic purification14. The only residual
solvents detected were acetonitrile at 855 ppm and water at 1.5 wt%. Residual metals were also all
within ICH limits (iron excluded).15 To facilitate the ease of use of this material for conjugations, it was
packaged under nitrogen into individual vials containing 1.2 g of the linkerꢀchelator 1 and stored with
dessicant at –20 ˚C to minimize hydrolysis. Subsequent stability studies showed that this material was
stable for up to two years when stored under these conditions with 0% degradation based on HPLC
analysis.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Synthesis of GNE-605
CONCLUSION
In conclusion, a practical process for manufacturing the ImmunoꢀPET linkerꢀchelator GNE-605
(1) from commercially available desferrioxamine B on multiꢀgram scale has been developed.
Modifications to the previously published synthetic procedure allow for an efficient 3ꢀstep telescoped
process with significantly reduced reaction times, reduced raw material requirements and elimination of
14
ACS Paragon Plus Environment

Page 15 of 19
Organic Process Research & Development
most of the aqueous workup operations. A single straightforward workꢀup procedure to remove reaction
byꢀproducts was introduced and helped to facilitate a subsequent C18 chromatographic purification.
Significant optimization of the column purification minimized the amount of hydrolysis and
significantly increased the purity of the isolated material to > 90 A% HPLC purity. For the first time,
gram quantities of the ImmunoꢀPET linkerꢀchelator GNE-605 have been synthesized and isolated with a
high level of purity. This material has since been successfully conjugated to a number of therapeutic
monoclonal antibodies and the small amount of material required for such conjugations means that this
single campaign will enable the development of many future conjugates of interest.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
General: Solvents and reagents were obtained from commercial sources and used without further
purification. DFO mesylate salt 2 was purchased from ECA with >93 A% HPLC purity. C18
chromatography was performed with a Thomson Single Step Pump® with manual fraction collection
and a lyophilization was performed with a VirTis benchtop lyophilizer. Unless otherwise noted all
reactions were run under a nitrogen atmosphere. Unless otherwise noted, 1H NMR (400 MHz), 13C
NMR (100 MHz) were recorded in DMSOꢀd6. Chemical shifts are reported in ppm (δ units) downfield
of internal tetramethylsilane [(CH3)4Si] or residual DMSOꢀd6; coupling constants are reported in hertz
(Hz). Multiplicities are as follows: s = singlet, d = doublet, t = triplet, q = quartet, dd = doublet of
doublets, m = multiplet. Melting points and Tg are uncorrected and were measured by differential
scanning calorimetry (DSC Q200 from TA Instrument). IR spectra were recorded on solid samples using
Smart Orbit on a Thermo Nicolet 6700. UV spectra were recorded as solutions in MeCN.
HPLC Methods: HPLC analysis for compound 3 was performed by diluting the reaction mixture in
0.1M sodium carbonate (aq) with detection at 220nm (UV). HPLC analysis for compound 4 was
performed by diluting the reaction mixture in 1:1 MeCN/water with detection at 420 nm (Vis). HPLC
analysis for compound 1 was performed by diluting the reaction mixture in MeCN with detection at 254
nm (UV). HPLC method for monitoring the formation of 3, 4 and GNE-605(1): ACE 3 C18ꢀHL, 3.0 x
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 19
50 mm, gradient elution from 95:5 to 5:95 0.05% TFA in water/ 0.05% TFA in MeCN over 4.8 min, 1.0
mL/min flow at 35 ˚C with detection at 220 nm, 254 nm and 420 nm. HPLC retention times: 2 = 2.309
min, 3 = 2.481 min, 4 = 2.289 min, GNE-605 = 3.277 min.
1
2
3
4
5
6
7
8
9
Crude Iron (III), [2,3,5,6-tetrafluorophenyl-3,14,25-tri(hydroxy-κO)-2,13,24-tri(oxo-κO)-10,21,24-
trioxo-3,9,14,20,25,31-hexaazapentatriacontan-35-oato(3-)]-(9CI) Coordination Compound (GNE-
605): To a 20 L reactor under nitrogen was charged DMF (3.79 kg, 9.48 kg/kg), 2 (0.400 kg. 0.609 mol,
100 mol%), succinic anhydride (0.070 kg, 0.700 mol, 115 mol%), N,N-diisopropylethylamine (0.0787
kg, 0.609 mol, 100 mol%) and the mixture was heated at 50 °C for at least 1 h. The reaction was deemed
complete when 2 ≤ 5.0 A% by HPLC at 220 nm. An analytically pure sample of 3 for characterization
was isolated by vacuum filtration of the crude mixture and rinsing with DMF and diethyl ether.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Compound 3 was obtained as a free flowing white solid: H NMR (400 MHz, DMSOꢀd6) δ 12.01 (1H,
brs), 9.64ꢀ9.62 (3H, m), 7.79ꢀ7.75 (3H, m), 3.48ꢀ3.44 (6H, m), 3.03ꢀ2.98 (6H, m), 2.60ꢀ2.56 (4H, m),
13
2.43ꢀ2.40 (2H, m), 2.31ꢀ2.26 (6H, m), 1.97 (3H, s), 1.52ꢀ1.20 (18H, m). C NMR (100 MHz, DMSOꢀ
d6) δ 174.31, 172.43, 171.77, 171.20, 170.61, 47.56, 47.26, 30.49, 30.39, 29.69, 29.27, 28.03, 26.49,
23.95, 20.80; mp = 163.85 ºC; IR (cm–1) 3297.2, 2928.3, 2850.1, 1721.7, 1564.3, 1427.5, 1415.3,
1396.1, 1373.9, 1345.5, 1297.1, 1269.5, 1251.3, 1223.1, 1196.9, 1162.9, 1134.2, 1084.9, 1044.7, 968.3,
961.5, 933.0; HRMS (ESI+) calculated for C29H52N6O11 [M+H]+, 661.3772 found 661.3779. The
mixture was cooled to 25 °C and a solution of FeCl3 (0.108 kg, 0.670 mol, 100 mol%) in DMF (0.342
kg, 0.855 kg/kg) was added. The reaction mixture was then agitated at 25 °C for a minimum of 30 min
and deemed complete when 3 ≤ 5.0 A% by HPLC at 420 nm. An analytically pure sample of 4 could not
be obtained as a solid as the compound was unstable during isolation and decomposed to a black tar.
Solution phase characterization was obtained by filtering the crude to remove solids and analyzing the
DMF solution: UV (nm) 202, 430; HRMS (ESI+) calculated for C29H49FeN6O11 [M+H]+, 714.2887
found 714.2890. The mixture was charged with EDC•HCl (0.233 kg, 1.22 mol, 200 mol%), N,N-
diisopropylethylamine (0.157 kg, 1.22 mol, 200 mol%) and a solution of 2,3,5,6ꢀtetrafluorophenol
16
ACS Paragon Plus Environment

Page 17 of 19
Organic Process Research & Development
(0.111 kg, 0.670 mol, 110 mol%) in DMF (0.342 kg, 0.855 kg/kg). The mixture was agitated at 25 ˚C
for a minimum of 5 h and monitored by HPLC until completion (4 ≤ 10.0 A% by HPLC at 254 nm). The
contents of the reactor were cooled to 5 ˚C and DCM (5.32 kg, 13.3 kg/kg) was charged. The mixture
was warmed to 25 ˚C and charged with a 13% aqueous NaCl solution (13.2 kg, prepared from 4.52 kg
NaCl and 34.8 kg H2O). The mixture was agitated for a minimum of 1 min and the layers were
separated. The organic phase was collected and dichloromethane (5.32 kg, 13.3 kg/kg) was charged to
the reactor. The mixture was agitated for a minimum of 1 min then separated and the organic layer was
collected and combined with the first organic phase. The combined organic phase was washed twice
more with 13% aqueous NaCl (8.8 kg per wash) and recharged to the reactor. Na2SO4 (0.600 kg, 1.5
kg/kg) was charged and the resulting suspension was agitated for a minimum of 5 min. The slurry was
filtered and the filtrate was concentrated under vacuum with a bath temperature of 25 ˚C.
Dichloromethane (3.00 kg, 7 kg/kg) was charged to the reactor, and the resulting solution was filtered
and concentrated under vacuum with a bath temperature of 25 ˚C to give crude 1 as a dark red, viscous
oil (0.746 kg, 88.5 A% HPLC). The crude product was flushed with nitrogen, sealed and stored at 2–8
˚C until it could be used in the purification.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General Procedure for Purification: A 413 g RediSep® gold column was conditioned with a 50%
MeCN/water solution (2.2 L) with a StepPump motor setting of 42 (86 mL/min). The column was then
equilibrated with a 10% MeCN/water solution (0.75 L) and loaded with crude GNE-605 (56 mL). The
column was eluted with 10% MeCN/solution at a StepPump motor setting of 25 (50 mL/min) until the
eluent became visibly red and continued to be eluted until the eluent once again became colorless. All
eluents up to this point were discarded. The column was eluted with a 30% MeCN/water solution at a
StepPump motor setting of 35 (70 mL/min), and fractions of approximately 150 mL were collected in
250 mL glass bottles with screw tops. The fractions were labeled in sequential order, sampled for IPC
analysis, capped securely and stored frozen in dry ice. In order to minimize hydrolysis, each fraction
from the column was sampled and then frozen in dry ice until all the IPC data had been collected. To
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 19
ensure accurate IPC data, each IPC sample was itself diluted 10ꢀfold with acetonitrile and frozen until it
could be analyzed by HPLC. A total of 1.7 L of eluent was collected. The column was eluted with 100%
MeCN until the eluent was no longer visibly colored and the eluent was discarded. The column was
equilibrated with a 10% MeCN/water solution (0.75L) then loaded with another 55 mL of crude GNE-
605. The overall process was repeated until a total of 11 column purifications had been performed. The
fractions that passed IPC (GNE-605 ≥ 85.0 A% HPLC at 254 nm) were thawed, combined and
lyophilized to give GNE-605 (86.3 g, 92.3 A% HPLC, KF: 1.5%, 16% yield). The material was
packaged in vials containing approximately 1.2 g apiece. The vials were flushed with nitrogen, placed in
LDPE bags with desiccant, sealed and stored at –20 ˚C. Compound 1 was obtained as a freeꢀflowing
bright redꢀorange powder, Tg = 108.7 ˚C; IR (cm–1) 3295.7, 2927.4, 2857.9, 1786.4, 1644.0, 1572.8,
1522.3, 1486.4, 1453.6, 1356.1, 1260.5, 1217.5, 1177.7, 1104.0, 1066.2, 952.1, 841.6, 756.1; UV (nm)
198, 434; HRMS (ESI+) calculated for C35H49F4FeN6O11 [M+H]+, 862.2818, found 862.2811; Fe
content (ICPꢀMS) calculated for C35H49F4FeN6O11, 6.5%, found 7.3%. Because of the paramagnetic
nature of the Fe(III) chelate, NMR spectroscopic data could not be obtained on GNE-605.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOWLEDGMENTS
The authors wish to acknowledge Herman Gill for providing background information on the PET
imaging process and for helping with the initial synthetic efforts on GNEꢀ605. We would also like to
thank Dr. Haiming Zhang for his help with the preparation of this manuscript and Samuel H. Yang for
collecting the ICPꢀMS data for iron content.
SUPPORTING INFORMATION
Additional details on optimization of the purification procedure along with1H NMR, 13C NMR, IR, DSC
and UV spectra are available in the supporting information.
18
ACS Paragon Plus Environment

Page 19 of 19
Organic Process Research & Development
REFERENCES
1
2
3
4
5
6
7
8
1
For reviews on ImmunoꢀPET platforms and their use in the clinic see: (a) van Dongen, G. A. M. S.;
Visser, G. W. M.; Lubꢀde Hooge, M. N.; de Vries, E. G.; Perk, L. R. The Oncologist 2007, 12, 1379. (b)
Knowles, S. M.; Wu, A. M. J. Clin. Onc. 2012, 30, 2884.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2 Verel, I; Visser, G. W. M.; Boellard, R.; Stigterꢀvan Walsum, M.; Snow G. B.; van Dongen, G. A. M.
S. J. Nuc. Med. 2003, 44, 1271.
3 Also referred to in the literature as desferoxamine, DFOꢀB, DFOA, DFB, desferal and DFO
4 Tenenbein, M. J. Toxicol. Clin. Toxicol. 1996, 34, 485.
5 Solubility of compound 3 in 0.15 M NaOH is < 1 mg/mL.
6 In reference 2, the amount of compound 1 produced is estimated based on an unquantified HPLC trace
7 See experimental section, HPLC methods.
8 Further optimization showed that these reagents could be further reduced to 1.8 molar equivalents and
1.1 equivalents respectively.
9
Attempts were also made to accomplish the esterification on the nonꢀchelated linker 3 with a number
of solvents (NMP, DMF, THF, MeCN, DMSO) and reagents (EDC•HCl, T3P, CDI). However, in all
cases no reaction was observed as chelation of the hydroxyamides appears to be a necessary condition
for successful functionalization of the terminal acid.
10
Residual acid 3 was observed at up to 10 A% by HPLC and succinimide 5 was observed at up to 5
A% by HPLC using the standard HPLC method described in the experimental section.
11
Several brands of columns were compared for their purification effectiveness and recovery efficiency
with the RediSep® Gold brand column giving the highest level of purity and material recovery. See
Supporting Information for further details.
12 See experimental for details. Up to 60 g of crude 1 was loaded onto a 413 g C18 column.
13
For lyophilization it is important to maintain sufficiently strong vacuum (< 1000 mtorr) in order to
avoid meltꢀback. In cases where vacuum was not sufficiently strong, dark black, glasslike particles
consisting of the hydrolysis product 4 would form on the surfaces where meltꢀback had occurred.
14 Crude ester 1 was stored at 2–8 ˚C under a nitrogen atmosphere in an attempt to minimize hydrolysis
while awaiting column purification. Even so, after 24 hours of storage the purity of the crude material
was found to have degraded down to 87 A% purity from 88 A% purity and degradation continued at a
rate of approximately 1% every 24 hours with 77 A% HPLC purity after 8 days of storage.
15 Cd < 5 ppm; Pb < 10 ppm; As < 3ppm; Hg < 3 ppm.
19
ACS Paragon Plus Environment