1 IC 50 ¼ 23 nM), presenting the most promising results from the series. However,
compound ZD1839 (1) with a longer spacer in the C6 side chain was selected as this
analogue presented high and sustained blood concentration after oral administration
and was more potent than 30 in the cellular assay (IC 50 ¼ 80 Â 380 nM, respectively; Fig. 8) [43].
Preclinical studies confirmed the antitumor efficacy of ZD1919 (1), later named
gefitinib (Iressa™), in animal xenograft models, and early phase I clinical trials
established the adequate oral bioavailability of 1 in a once-a-day dose regimen [46].
Since EGFR crystallographic data were not yet available, Palmer and colleagues
constructed an in silico structural model [47] using as background the previous SAR
studies and the crystal structure of a cAMP-dependent protein kinase [48]. According
to this model, the 4-anilinoquinazoline derivatives would interact with the EGFRATP-binding site through hydrogen bonds with specific amino acid residues and
nonpolar interactions within a hydrophobic pocket at the hinge region neighborhood.
H-bonds between the hinge methionine residue 793 (M793) and the quinazoline N1
atom and between the side chain of the gatekeeper threonine residue 790 (T790) and
the quinazoline N3 atom were suggested. In turn, the aniline moiety would fit into a
hydrophobic pocket non-occupied by ATP, and this particular interaction was
supposed to confer selectivity for EGFR inhibition within the kinome [47].
Analyses of previous results also indicated that a polar interaction with M793
would be more important than one with T790 residue, as the 4-anilinoquinoline
derivative 16 presented significant EGFR inhibitory activity while the
2-anilinoisoquinoline derivative 17 did not (Fig. 5).
Five years later, in 2002, a co-crystallized structure of EGFR with a
4-anilinoquinazoline inhibitor was finally obtained and elucidated. Erlotinib (2;
CP-258,774; Tarceva™; Roche, Fig. 2), the inhibitor present in the mentioned
crystal structure (Protein Data Bank code: 1M17), was originally described in
1997 as a potent and selective ATP-competitive EGFR inhibitor (EGFRi) [49].
Compound 2 has an acetylene group at the meta position of the aniline ring,
which is twisted in an angle of 42
in relation to the quinazoline moiety and is placed
inside a hydrophobic pocket in the binding region, performing nonpolar interactions
with T790, K745, and L788 amino acid residues. As already anticipated by Palmer
[47], the quinazoline N1 acts as a hydrogen bond acceptor to the amide hydrogen of
the M793 residue backbone. Additionally, the quinazoline N3 atom interacts with
the side chain of the gatekeeper residue T790 through a water-bridged H-bond, and
both C6 and C7 substituents are directed to a solvent-exposed region in the outer area
of the protein [50] (Fig. 9).
Gefitinib (1; ZD1819; Iressa™; AstraZeneca) and erlotinib (2; CP-258,774;
Tarceva™; Roche) were initially approved in 2003 and 2004, respectively, as
third-line therapies for patients with metastatic or advanced NSCLC who did not
respond to chemotherapy with docetaxel and platinum-based antineoplastic drugs
[51–53]. These EGFRi are currently being approved as first-line therapies for
treatment of metastatic NSCLC characterized by the occurrence of EGFR-activating
mutations.
Case Study on Receptor Tyrosine Kinases EGFR, VEGFR, and PDGFR
165
compound ZD1839 (1) with a longer spacer in the C6 side chain was selected as this
analogue presented high and sustained blood concentration after oral administration
and was more potent than 30 in the cellular assay (IC 50 ¼ 80 Â 380 nM, respectively; Fig. 8) [43].
Preclinical studies confirmed the antitumor efficacy of ZD1919 (1), later named
gefitinib (Iressa™), in animal xenograft models, and early phase I clinical trials
established the adequate oral bioavailability of 1 in a once-a-day dose regimen [46].
Since EGFR crystallographic data were not yet available, Palmer and colleagues
constructed an in silico structural model [47] using as background the previous SAR
studies and the crystal structure of a cAMP-dependent protein kinase [48]. According
to this model, the 4-anilinoquinazoline derivatives would interact with the EGFRATP-binding site through hydrogen bonds with specific amino acid residues and
nonpolar interactions within a hydrophobic pocket at the hinge region neighborhood.
H-bonds between the hinge methionine residue 793 (M793) and the quinazoline N1
atom and between the side chain of the gatekeeper threonine residue 790 (T790) and
the quinazoline N3 atom were suggested. In turn, the aniline moiety would fit into a
hydrophobic pocket non-occupied by ATP, and this particular interaction was
supposed to confer selectivity for EGFR inhibition within the kinome [47].
Analyses of previous results also indicated that a polar interaction with M793
would be more important than one with T790 residue, as the 4-anilinoquinoline
derivative 16 presented significant EGFR inhibitory activity while the
2-anilinoisoquinoline derivative 17 did not (Fig. 5).
Five years later, in 2002, a co-crystallized structure of EGFR with a
4-anilinoquinazoline inhibitor was finally obtained and elucidated. Erlotinib (2;
CP-258,774; Tarceva™; Roche, Fig. 2), the inhibitor present in the mentioned
crystal structure (Protein Data Bank code: 1M17), was originally described in
1997 as a potent and selective ATP-competitive EGFR inhibitor (EGFRi) [49].
Compound 2 has an acetylene group at the meta position of the aniline ring,
which is twisted in an angle of 42
in relation to the quinazoline moiety and is placed
inside a hydrophobic pocket in the binding region, performing nonpolar interactions
with T790, K745, and L788 amino acid residues. As already anticipated by Palmer
[47], the quinazoline N1 acts as a hydrogen bond acceptor to the amide hydrogen of
the M793 residue backbone. Additionally, the quinazoline N3 atom interacts with
the side chain of the gatekeeper residue T790 through a water-bridged H-bond, and
both C6 and C7 substituents are directed to a solvent-exposed region in the outer area
of the protein [50] (Fig. 9).
Gefitinib (1; ZD1819; Iressa™; AstraZeneca) and erlotinib (2; CP-258,774;
Tarceva™; Roche) were initially approved in 2003 and 2004, respectively, as
third-line therapies for patients with metastatic or advanced NSCLC who did not
respond to chemotherapy with docetaxel and platinum-based antineoplastic drugs
[51–53]. These EGFRi are currently being approved as first-line therapies for
treatment of metastatic NSCLC characterized by the occurrence of EGFR-activating
mutations.
Case Study on Receptor Tyrosine Kinases EGFR, VEGFR, and PDGFR
165