Optimization and SAR for dual ErbB-1/ErbB-2 tyrosine kinase inhibition in the 6-furanylquinazoline series

Article abstract of DOI:10.1016/j.bmcl.2006.05.090

Synthetic modifications on a 6-furanylquinazoline scaffold to optimize the dual ErbB-1/ErbB-2 tyrosine kinase inhibition afforded consistent SAR whereby a 4-(3-fluorobenzyloxy)-3-haloanilino provided the best enzyme potency and cellular selectivity. Changes made to the 6-furanyl group had little impact on the enzyme activity, but appeared to dramatically affect the cellular efficacy. The discovery of lapatinib emerged from this work.

Full text of DOI:10.1016/j.bmcl.2006.05.090

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4686–4691
Optimization and SAR for dual ErbB-1/ErbB-2 tyrosine kinase
inhibition in the 6-furanylquinazoline series
Kimberly G. Petrov, Yue-Mei Zhang, Malcolm Carter,à G. Stuart Cockerill,à
Scott Dickerson, Cassandra A. Gauthier,§ Yu Guo, Robert A. Mook, Jr.,
David W. Rusnak, Ann L. Walker, Edgar R. Wood and Karen E. Lackey*
GlaxoSmithKline, 5 Moore Drive, Research Triangle Park, NC, USA
Received 18 April 2006; revised 25 May 2006; accepted 30 May 2006
Available online 13 June 2006
Abstract—Synthetic modifications on a 6-furanylquinazoline scaffold to optimize the dual ErbB-1/ErbB-2 tyrosine kinase inhibition
afforded consistent SAR whereby a 4-(3-fluorobenzyloxy)-3-haloanilino provided the best enzyme potency and cellular selectivity.
Changes made to the 6-furanyl group had little impact on the enzyme activity, but appeared to dramatically affect the cellular
efficacy. The discovery of lapatinib emerged from this work.
Ó 2006 Elsevier Ltd. All rights reserved.
The Type I receptors represent an ideal point for drug
intervention in cancer therapy due to aberrant regula-
tion in certain solid tumors and correlation with disease
progression and outcome.2 The quinazoline scaffold fea-
tures in several well-known examples of potent ErbB-1
(EGFR) tyrosine kinase (TK) inhibitors that are cur-
rently in clinical trials or are used in anti-cancer thera-
py.3 Several reviews have been published covering
these and other small molecule Type 1 receptor inhibi-
tors and provide excellent information regarding the
current state of the art in this class of drug candidates.4,5
Our medicinal chemistry lead optimization efforts in the
quinazoline series to produce a dual ErbB-1/ErbB-2
inhibitor led to the discovery of lapatinib
(GW572016), which is currently in clinical trials.6 The
structure is shown in Figure 1.
O
F
HN
N
Cl
H
N
N
O
O
S
O
Figure 1. Lapatinib.
synthesis, kinase enzyme inhibition, and cellular activity
of this important series of compounds.
The general strategy for synthetic modifications of the
core 6-furanyl quinazoline to establish the SAR is sum-
marized in Scheme 1. The orientation of the substituents
on the furan ring was found to dramatically affect the
cellular activity (discussed below), so most of our work
focused on the 5-substituted-2-furanylquinazoline series.
The introduction of a side chain via a facile reductive
amination provided an excellent scaffold to optimize
the cellular activity of the series with modifications in
chain length, heteroatom linker, and terminal R-group
changes. This chemistry discussion will cover the general
synthesis, and the generation of a focused set of anilines
and selected amines used to optimize dual ErbB-1 and
ErbB-2 TK inhibition.
Many compounds were synthesized in the 6-furanyl qui-
nazoline series using facile syntheses to investigate the
structure–activity relationships (SAR) for dual inhibi-
tion of ErbB-1/ErbB-2 focused on the ATP binding
region of the protein. Herein we will describe the
Keywords: ErbB-1 tyrosine kinase inhibitors; ErbB-2 tyrosine kinase
inhibitors; EGFR kinase inhibitors; Quinazoline; Lapatinib.
*
Corresponding author. Tel.: +1 919 483 6240; e-mail:
karen.e.lackey@gsk.com
See Ref. 1a.
See Ref. 1b.
à
Beginning with commercially available 4-chloro-6-iodo-
quinazoline, the desired compounds can be synthesized
§
See Ref. 1c.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.090

K. G. Petrov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4686–4691
4687
Ar
N
Cl
N
HN
O
NH2Ar
I
B(OH)2
I
O
N
Pd(OAc)2, PPh3
Et3N, DMF
iPrOH
70oC
N
(II)
(I)
Ar
N
Ar
HN
N
H
H2N
HN
R
R N
OHC
N
O
CH2Cl2, HOAc
Na(OAc)3BH
O
N
(IV)
(III)
R1
Ar
N
HN
N
R1CH2X
R
N
O
iPr2EtN
DMF
(V)
Scheme 1. General synthetic route for 4-anilino-6-furanylquinazoline derivatives.
in three convergent steps with the capability to build in
highly functionalized groups at each step as shown in
Scheme 1. Optimization work was necessary to improve
the yield and reproducibility of the palladium coupling
reaction to afford analogs of (III) in sufficient scale for
in vivo evaluation studies.7,8 The furylaldehyde moiety
could easily be functionalized to include substitutions
designed to optimize cellular activity.
While the reductive amination step was amenable to a
variety of amines, the data suggested that basic amines
were much less active in the cellular assays (unpublished
data). The preferred amines were substituted sulfonyl-
ethyl amines and thiomorpholine oxide to form (IV).
Thiomorpholine was easily converted to thiomorpholine
oxide using sodium periodate in water.12 There were two
methods for generating the substituted sulfonylethylam-
ines, selected largely on the basis of available starting
materials.13 In Scheme 3a, the longer route is shown
beginning with a substituted thioethylamino reagent
that must first be N-protected followed by oxidation.
Upon removal of the protecting group, the amine was
then used in the reductive amination reactions with the
furanylaldehyde quinazoline intermediate (III). The
second sequence shown in Scheme 3b begins with
Previous research at GlaxoSmithKline had demonstrat-
ed the need for large anilino substitutions to retain both
ErbB-1 and ErbB-2 inhibition activity.9 Since the most
potent dual enzyme inhibition could be obtained when
the aryl group was phenylsulfonylphenyl, N1-benzylin-
dazolyl, and benzyloxyphenyl, we focused our optimiza-
tion efforts on these anilines or further functionalized
versions of these anilines.10 The most interesting SAR
trends were found in the benzyloxyaniline substituted
series and are summarized in this report. In-house data
had demonstrated that the optimal substitution pattern
involved a para-benzyloxy group.11 Compounds with
substitutions in the 2- and 6-position of the aniline ring
had diminished ErbB family enzyme inhibition, so our
work focused only on combining 3- and 4-position
substituents shown in Scheme 2. These anilines were
synthesized by alkylating the appropriately substituted
nitrophenol followed by a platinum or palladium cata-
lyzed hydrogenation. Care must be taken to use less
than 1% catalyst due to the exothermic nature of some
of the reactions.
commercially available sulfonylacetonitriles. In
a
single-step reduction, the substituted sulfonylethylamine
is generated.
To generate compounds with modifications of the sulfo-
nylethylamine side chain, homologations and ether link-
ages were examined. The representative syntheses are
shown in Scheme 4 each beginning with the advanced
furanylquinazoline intermediate (III). To form the
homolygated compound (VI), a nitrovinyl group was in-
stalled, then subsequently reduced to the aminoethyl.
Standard amide coupling conditions are used to form
the amide intermediate prior to the reduction. To gener-
ate the 3-carbon linked side chain (VII), a four-step syn-
thetic route starting with (III) was used. A vinyl ester
was installed, hydrolyzed to the carboxylic acid, subject-
ed to standard amide coupling, and then reduction con-
ditions applied. The final reaction sequence described in
Scheme 4 is the conversion of (III) to the ether linked
series (VIII) by reducing the aldehyde to the hydroxy-
methyl group followed by treatment with vinyl sulfone.
OH
Br
K2CO3
O
X
Y
O2N
X
CH3CN
Y
O2N
X = H, Cl, F, MeO,
Br, CF3
Y=H, F, Br, CF3
H2, Pt/C
EtOH/THF
Inhibition of ErbB-1 and ErbB-2 tyrosine kinase activity
was evaluated in enzyme assays using the purified
recombinant human intracellular domain of each recep-
tor. The expression, purification, and assay methods
were conducted as described by Brignola and co-work-
ers using the scintillation proximity assay detection
method.14 The average standard deviation of the mean
O
X
Y
H2N
Scheme 2. Benzyloxyaniline building blocks used for dual ErbB-1/2
inhibition SAR.

4688
K. G. Petrov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4686–4691
R
O
a
CBZCl
O
O
R
H2, Pd/C
EtOH
O
R
OXONE
S
S
CBZ
S
S
CBZ
NH2
R
NH2
NaHCO3
DMF
N
N
MeOH/H2O
H
H
O
.
O
O
O
BH3 Me2S in THF
b
R=aryl, alkyl, heteroaryl
S
S
NH2
R
N
R
Scheme 3. Substituted sulfonylethylamine or ether functional groups.
Ar
HN
N
CH3NO2,
NH4OAc
O2N
1) LiAlH4, THF
O
N
(III)
2) (i-Pr)2NEt, EDC,
MeCN/DMF
RCO2H
Ar
N
O
Ar
O O
S
HN
N
HN
N
O
S
O
N
H
O
N
O
N
.
BH3 Me2S ,
THF
H
(VI)
Ar
N
HN
N
EtO C
1) NaOH/MeOH
K2CO3, CH3CN
2
(III)
O
2) (i-Pr)2NEt, EDC,
MeCN, RNH2
O
EtO
EtO
P
CO2Et
Ar
N
HN
N
Ar
HN
H
H
N
O
S
O O
N
N
O
.
S
O O
BH3 Me2S ,
THF
O
(VII)
N
O
S
O
Ar
1. NaBH4
O
HN
(III)
O
N
O
O
S
2. NaH,
DMF
N
(VIII)
Scheme 4. Synthesis of furan side-chain substitutions on the 4-anilinoquinazoline scaffold.
for the enzyme results described in Tables 1–5 is equal
to 0.2 units of the log of the IC50 value (ErbB-1 and
ErbB-2).
Table 1. Type I receptor inhibition activity and cellular efficacy results
for representative examples of the modified benzyloxyaniline analogs
O
O
S
O
The cellular assay data shown in Tables 1–5 are included
to demonstrate the efficacy of the substituted quinazo-
lines in tumor lines driven by the ErbB family targets.
A thorough explanation of the assay system and the
methods used are included in published work by Rusnak
and co-workers.10 HN5, derived from a head and neck
cancer line, over-expresses ErbB-1.15 BT474, a breast
line, over-expresses ErbB-2.16 While N87, a gastric tu-
mor line, over-expresses both receptors, ErbB-2 is pres-
ent at significantly greater levels.10 The IC50 values listed
in the tables are the average of 3 or more determinations
where the cells were treated with compound for 72 h.
Effective dual ErbB-1 and ErbB-2 tyrosine kinase inhib-
itors should be efficacious in all three cell lines. We
observed distinct SAR in the correlation between the
kinase enzyme potency and cellular activity.
Y
HN
N
X
HN
N
O
Compound
X
Y
IC50 (lM)
ErbB-2 ErbB-1 HN5 BT474 N87
1
2
Cl
H
H
H
H
Cl
Br
F
0.010
0.026
0.031
0.10
0.012
0.027
0.024
0.043
0.091
0.018
0.025
0.10
0.12
0.65
0.84
2.56
6.9
0.08 0.08
0.28 0.87
0.77 0.86
2.13 3.46
H
F
3
4
Br
5
CF3 0.36
4.2
5.2
6
H
H
F
0.022
0.025
0.09
0.25
0.26
3.33
0.25 0.28
0.27 0.37
2.07 1.90
7
8
MeO
CF3
CF3
9
10
H
F
0.20
0.22
0.16
0.08
5.90 10.14 6.50
1.94 4.49 2.89
Well over 1000 compounds were generated with the 4-
benzyloxyaniline group or a substituted analog thereof
on a quinazoline core, it was determined that the 3-ani-
lino (X) and 3-benzyloxy (Y) positions were optimal for
dual ErbB-1 and ErbB-2 inhibition designated in the
structure for Table 1. Illustrative examples have been
selected for this report to demonstrate the SAR. Using

K. G. Petrov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4686–4691
4689
Table 2. Type I receptor inhibition activity and cellular efficacy results
for representative examples of terminal side-chain modifications
Table 5. Type I receptor inhibition activity and cellular efficacy results
for representative examples of thiomorpholine side chain with benz-
yloxyaniline changes
O
O
R
S
O
O
F
O
HN
N
Cl
Y
S
HN
N
X
HN
N
O
N
N
O
Compound
R
IC50 (lM)
Compound X, Y
IC50 (lM)
ErbB-2 ErbB-1 HN5 BT474 N87
ErbB-2 ErbB-1 HN5 BT474 N87
1
11
12
13
14
–CH3
n-Pr
i-Pr
0.010
0.026
0.024
0.030
0.022
0.021
0.018
0.019
0.019
0.12 0.08
1.6 3.14
0.08
8.66
0.13
0.87
0.79
25
26
27
28
CF3, F 0.141
0.089
0.028
0.016
0.012
1.11
0.06
1.0
3.00
0.10
1.20
0.42
2.09
0.08
1.13
0.46
0.28 0.20
1.07 1.04
1.55 0.97
Cl, F
F, H
Cl, H
0.018
0.035
0.023
–Ph
2-Pyridyl 0.030
0.32
N
15
0.026
0.017
0.50 0.39
0.23
N
are listed in Table 1. Chlorine at position X afforded
insignificant increases in enzyme potency, but appeared
to offer improvements in cellular efficacy. Larger groups
in position X generally diminished activity. In combina-
tion with suitable X substituents, fluorine at position Y
yielded compounds with greater potency in the cellular
assay panel. Larger groups at Y also reduced activities.
The binding pocket for this moiety in the ErbB-1 crystal
structure appears as a well-defined lipophilic pocket
(Fig. 2) which tends to support the observed size toler-
CH3
Table 3. Type I receptor inhibition activity and cellular efficacy results
for regiochemical analogs of 2
O
3'
HN
4'
H
H3C
N
2'
S
N
5'
O
O
O
ance
and
position
of
the
benzyloxyaniline
N
substitutions.17
Compound
Position
IC50 (lM)
The exploration of the terminal substitution on the
6-position side chain was performed to determine if
Average
ErbB-2/ErbB-1
Average
tumor
HN5/BT474/N87
16
2
None
50
0.074
0.027
0.023
0.068
2.98
0.60
2.68
3.81
17
18
40
30
Table 4. Type I receptor inhibition activity and cellular efficacy results
for representative examples of side-chain modifications
O
O
R
S
O
Y
HN
N
Cl
Z
N
O
Compound
R
Y
Z
IC50 (lM)
ErbB-2 ErbB-1 BT474
19
20
21
22
23
24
Me
Ph
F
F
H
F
F
F
–O–
–O–
0.017
0.028
0.026
0.025
0.050
0.085
0.027
0.090
0.17
NT
Ph
–NCH2CH2- 0.038
–N-(n-Pr) 0.02
–N-(CH2CN) 0.022
–N-(Bn) 0.141
NT
Me
Me
Me
0.25
0.067
0.61
Figure 2. A view through the N-terminal lobe of ErbB-1 of the bound
conformation of 1 as determined by X-ray crystallography.17 The
benzyloxyaniline moiety binds in a well-defined lipophilic pocket,
the quinazoline makes a hydrogen bond to the hinge region, and the
methylsulfonylethylamino group binds at the solvent interface. Image
created with Maestro from Schrodinger, LLC.
NT, not tested.
a fixed 6-position modification, 5-methylsulfonylethyl-
amino-2-furanyl group, many combinations of substitu-
ents were explored and some representative examples

4690
K. G. Petrov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4686–4691
Table 6. Cellular activity and selectivity for representative substituted
6-furanylquinazolines
changes would significantly affect the enzyme and tumor
cell inhibition potency. The crystal structure of lapatinib
places the side-chain pointing toward and extending into
the protein/solvent interface.17 Enzyme inhibition IC50
values for all analogs stayed within the range of 10–
35 nM for both ErbB-1 and ErbB-2, as shown for repre-
sentative examples in Table 2. While little or no changes
in potency were observed for the enzyme assay, the inhi-
bition of tumor cell proliferation varied significantly. No
substituent proved to be superior to the simple methyl
sulfone.
Compound
IC50 (lM)
Average tumora
Selectivity N/T
HFF
1
12
19
23
26
0.09
0.20
0.16
0.11
0.08
9.9
>30
>30
111
>150
>188
>273
102
>30
8.18
a Average IC50 value for N87, BT474, and HN5 cellular data
combined.
Another question we sought to understand was the reg-
iochemistry of the linking furan ring with the side-chain
substituent. Table 3 lists a set of analogs synthesized in a
similar fashion as described in Scheme 1 and tested to
determine the optimal connection chemistry. While the
substitution at the 40 and 50 position of the furan ap-
peared to be equally potent at inhibiting ErbB-1/ErbB-
2 enzymes, the 50 substituted analog was 4–5 times more
potent against the tumor cell lines than either the 30 or 40
regioisomer.
bolstered our confidence in the therapeutic potential of
this series. A more thorough analysis of the enzyme/cell
activity was done for lapatinib, and the results suggest
that the improved cellular activity may be due to the
slow binding kinetics.16
The 6-furanyl-4-(4-benzyloxyanilino)-quinazoline scaf-
fold afforded the necessary drug-like properties and dual
ErbB-2/ErbB-1 tyrosine kinase inhibition to discover a
potential anti-cancer therapeutic agent. The halogen
substitution on the benzyloxyanilino group was key to
improving the enzyme/cell ratio of activity, with 4-(3-
fluorobenzyloxy)-3-chloroanilino providing the most
promising cellular efficacy. The substitutions on the
furanyl ring were important to be in a 2,5-orientation
for the desired cellular activity, while there was no
apparent difference in the enzyme activity. Quite a diver-
sity of amine substitutions were tolerated, presumably
due to the binding mode of these inhibitors where the
aniline is tucked into the back of the ATP binding pock-
et, and the side chain on the furanyl portion extends out
toward solvent. Overall, GW572016 (lapatinib, 1) pos-
sessed the desired enzyme potency, cellular activity in
a panel of tumor cell lines, and selectivity.
Linker side-chain SAR showed modifications were well
tolerated in the kinase enzyme inhibition assays. It is
interesting to note that there is virtually no difference
in the enzyme potency of a small linkage group, oxygen,
and larger groups of n-propylamino or benzylamine
(data shown in Table 4). The ether 19 and N-CH2CN
23 analogs in particular were relatively potent in the cel-
lular assays. Because of the mixing and matching of
functional group combinations and the space required
to depict the compounds, only BT474 cellular assay data
are shown in the Table 4. However, it is noteworthy that
the values in the HN5 and N87 cell lines were compara-
ble (unpublished). No advantage was observed in rela-
tive potency in the in vitro assay systems of the
tertiary amines (e.g., 21–24) compared with the second-
ary amino linker (e.g., 1, 12), except possibly improved
selectivity for tumor versus normal cells (see discussion
below). The extra synthetic steps and increased molecu-
lar weight were deemed non-beneficial for optimizing
the properties of our best compounds that contained
the methylsulfonylethylamine with a one-carbon tether
to the furan ring.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2006.05.090.
Since the disubstituted amine was tolerated in the exam-
ples above, a cyclic version was investigated for poten-
tial value. Some examples of those derivatives are
shown in Table 5. Similar trends in activity were ob-
served for the substitution of the benzyloxy aniline
moiety in this series and the optimally substituted
compound (26) compared very favorably to 1.
References and notes
1. (a) YZ: J&J Pharm, 1000 Route 202, Raritan, NJ 08869;
(b) GSC & MC: Arrow Therapeutics, Britannia House, 7
Trinity Street, London SE1 1DS, UK; (c) CAG: Array
Biopharma, Boulder, CO.
2. Klapper, L. N.; Kirschbaun, M. H.; Yarden, Y. Adv.
Cancer Res 2000, 77, 25.
3. Tiseo, M.; Loprevite, M.; Ardizzoni, A. Curr. Med. Chem.
2004, 4, 139.
4. Cockerill, S. G.; Lackey, K. E. Curr. Top. Med. Chem.
2002, 2, 1001.
5. Ransom, M. Br. J. Cancer 2004, 90, 2250.
6. Rusnak, D. W.; Lackey, K. E.; Affleck, K.; Wood, E. R.;
Alligood, K. J.; Rhodes, N.; Keith, B.; Murray, D. M.;
Mullin, R. J.; Knight, W. B.; Gilmer, T. M. Mol. Cancer
Ther. 2001, 1, 85.
A control cell line, derived from human foreskin fibro-
blasts, HFF, was used to assess the selectivity for tumor
cells (T) versus normal cells (N).13 Table 6 contains rep-
resentative examples from each of the groups of modifi-
cations covered in the discussion thus far. The average
of the tumor cell IC50 values was used to assess the dual
ErbB-1 and ErbB-2 inhibitory properties in assays that
are run under similar conditions. Greater than 100-fold
selectivity was observed for all of these analogs which

K. G. Petrov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4686–4691
4691
7. McClure, M. S.; Roschangar, F.; Hodson, S. J.; Millar,
A.; Osterhout, M. H. Synthesis 2001, 11, 1681.
8. McClure, M. S.; Glover, B.; McSorley, E.; Millar, A.;
Osterhout, M. H.; Roschangar, F. Org. Lett. 2001, 3,
1677.
9. Cockerill, G. S.; Stubberfield, C.; Stables, J.; Carter, M.;
Guntrip, S.; Smith, K.; McKeown, S.; Shaw, R.; Tapley,
P.; Thomson, L.; Affleck, K.; Jowett, A.; Hayes, D.;
Wilson, M.; Wollard, P.; Spalding, D. Bioorg. Med. Chem.
Lett. 2001, 11, 1401.
12. Khanna, I. K.; Weier, R. M.; Yu, Y.; Collins, P. W.;
Miyashiro, J. M.; Koboldt, C. M.; Veenhuizen, A. W.;
Currie, J. L.; Seibert, K.; Isakson, P. C. J. Med. Chem.
1997, 40, 1619.
13. Liebowitz, S. M.; Lombardini, J. B.; Allen, C. I. Biochem.
Pharmacol. 1989, 38, 399.
14. Brignola, P. S.; Lackey, K.; Kadwell, S. H.; Hoffman, C.;
Horne, E.; Carter, H. L.; Stuart, J. D.; Blackburn, K.;
Moyer, M. B.; Alligood, K. J.; Knight, W. B.; Wood, E.
R. J. Biol. Chem. 2002, 277, 1576.
10. Rusnak, D. W.; Affleck, K.; Cockerill, S. G.; Stubberfield,
C.; Harris, R.; Page, M.; Smith, K. J.; Guntrip, S. B.;
Carter, M. C.; Shaw, R. J.; Jowett, A.; Stables, J.; Tapley,
P.; Wood, E. R.; Brignola, P. S.; Kadwell, S. H.; Reep, B.
R.; Mullin, R. J.; Alligood, K. J.; Keith, B. R.; Crosby, R.
M.; Murray, D. M.; Knight, W. B.; Gilmer, T. M.;
Lackey, K. E. Cancer Res. 2001, 61, 7196.
15. Modjtahedi, H.; Affleck, K.; Stubberfield, C.; Dean, C.
Int. J. Oncol. 1998, 13, 335.
16. Pasleau, F.; Grooteclaes, M.; Gol-Winkler, R. Oncogene
1993, 8, 849.
17. Wood, E. R.; Truesdale, A. T.; McDonald, O. B.; Yuan,
D.; Hassell, A.; Dickerson, S. H.; Ellis, B.; Pennisi, C.;
Horne, E.; Lackey, K.; Alligood, K. J.; Rusnak, D. W.;
Gilmer, T. M.; Shewchuk, L. A Cancer Res. 2004, 64,
6652, PDB ID 1XKK.
11. Cockerill, G. S.; Carter, M. C.; Guntrip, S. B.; Smith, K.
J. PCT Appl. WO9802434, 1998.