Practical and efficient synthesis of gefitinib through selective O-alkylation: A novel concept for a transient protection group

Article abstract of DOI:10.1080/00397911.2017.1359627

A practical process that includes a simple four-step procedure for the preparation of gefitinib (1), a tyrosine kinase inhibitor that targets the epidermal growth factor receptor, is described. Dramatic improvements over previously reported conventional synthetic procedures were achieved. We found effective coupling conditions to minimize the inevitable production of an N-alkylated side product, N-(3-chloro-4-fluorophenyl)-7-methoxy-6-(3-morpholinopropoxy)-N-(3-morpholinopropyl)-quinazoline-4-amine (3) using a transient trimethylsilyl protecting group. We synthesized gefitinib in an 81.1% overall yield from a commercially available starting material on a multigram scale using a route that did not require work-up of any of the reaction steps.

Full text of DOI:10.1080/00397911.2017.1359627

Synthetic Communications
An International Journal for Rapid Communication of Synthetic Organic
Chemistry
ISSN: 0039-7911 (Print) 1532-2432 (Online) Journal homepage: http://www.tandfonline.com/loi/lsyc20
Practical and efficient synthesis of gefitinib via
selective O-alkylation: A novel concept for a
transient protection group
Sung Kwon Kang, Seung Wook Lee, Daekoo Woo, Jaehoon Sim & Young-Ger
Suh
To cite this article: Sung Kwon Kang, Seung Wook Lee, Daekoo Woo, Jaehoon Sim & Young-Ger
Suh (2017): Practical and efficient synthesis of gefitinib via selective O-alkylation: A novel concept
for a transient protection group, Synthetic Communications, DOI: 10.1080/00397911.2017.1359627
To link to this article: http://dx.doi.org/10.1080/00397911.2017.1359627
View supplementary material
Accepted author version posted online: 31
Jul 2017.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=lsyc20
Download by: [California Institute of Technology]
Date: 31 July 2017, At: 21:44

Practical and efficient synthesis of gefitinib via selective O-
alkylation: A novel concept for a transient protection group
Sung Kwon Kang
College of Pharmacy, Seoul National University, Seoul, Korea
Department of Synthetic Chemistry, Chong Kun Dang Research Institute, Yongin-si, Gyeonggi-
do, Korea
Seung Wook Lee
Department of Synthetic Chemistry, Chong Kun Dang Research Institute, Yongin-si, Gyeonggi-
do, Korea
Daekoo Woo
Department of Synthetic Chemistry, Chong Kun Dang Research Institute, Yongin-si, Gyeonggi-
do, Korea
Jaehoon Sim
College of Pharmacy, Seoul National University, Seoul, Korea
Young-Ger Suh*
College of Pharmacy, Seoul National University, Seoul, Korea
Address correspondence to Young-Ger Suh, College of Pharmacy, Seoul National University, 1
Gwanak-ro, Gwanak-gu, Seoul 151-742, Korea. Tele: + 82 28809174. E-mail: ygsuh@snu.ac.kr
1

Supporting Information: Full experimental detail, 1H and 13Cꢀ NMR spectra, HPLC traces,
HRMS spectra, and EA data. This material can be found via the “Supplementary Content”
section of this article’s webpage.
ABSTRACT
A practical process that includes a simple four-step procedure for the preparation of
gefitinib (1), a tyrosine kinase inhibitor that targets the epidermal growth factor receptor, is
described. Dramatic improvements over previously reported conventional synthetic procedures
were achieved. We found effective coupling conditions to minimize the inevitable production of
an N-alkylated side product, N-(3-chloro-4-fluorophenyl)-7-methoxy-6-(3-morpholinopropoxy)-
N-(3-morpholinopropyl)-quinazoline-4-amine (3) by using a transient trimethylsilyl protecting
group. We synthesized gefitinib in an 81.1% overall yield from a commercially available starting
material on a multigram scale using a route that did not require work-up of any of the reaction
steps.
GRAPHICAL ABSTRACT
KEYWORDS: anilinoquinazoline, EGFR inhibitor, gefitinib, transient protection group
Introduction
Quinazoline derivatives, especially 4-anilinoquinazolines, have been prevalent motifs in
pharmaceutical molecules in recent years, notably as an Epidermal Growth Factor Receptor
2

(EGFR) inhibitor.[1] As shown in Figure 1, several representative current drugs contain 4-
anilinoquinazoline as a core part.
It was revealed that the quinazoline moiety interacts with the hinge domain of the kinase
and the aniline moiety inserts into the hydrophobic pocket; these are crucial interactions for EGFR
inhibitory activity. The side chains at the C-6 and C-7 positions are not a part of the
pharmacophore, which makes long ether chains or other corresponding moieties compatibile.[2–5]
Therefore, many novel EGFR inhibitors have been designed and synthesized by incorporating an
alkoxy moiety via O-alkylation of the hydroxyl group at the C-6 or C-7 position of the 4-
anilinoquinazoline scaffold.[6–13] However, in most alkylations of anilinoquinazolines that used an
alkyl halide, a mixture of O- or N-alkylated as well as N,O-dialkylated products were produced
without satisfactory selectivity. In addition, purification of the alkylation product is known to be
tedious. Thus, the development of new synthetic strategies that provide selective O-alkylation of
the 4-anilinoquinazoline moiety is of interest for both laboratory and industrial preparation of
EGFR inhibitors.
Gefitinib,[14] the first selective inhibitor of the EGFR kinase domain, was approved in May
2003 for the treatment of recurrent NSCLC (non-small cell lung cancer).[15] Since then, gefitinib
has been widely used, and synthetic methods for its preparation based on the 4-anilinoquinazoline
moiety have been continuously reported.[16–22] According to a previous report,[19] gefitinib was
synthesized from 6,7-dimethoxy quinazolin-4-one starting material via sequential selective
demethylation, condensation with chlorofluoroaniline, and introduction of a 3-morpholinopropoxy
side chain at the C-6 position of 4-anilinoquinazoline. This procedure afforded an excess of an N-
alkylated side product 3inꢀ the final step. This N-alkylated impurity was removed by column
chromatography, which lowered the yield of gefitinib and limited the utility of this procedure for
3

commercial production. To overcome this problem, an improved process for the synthesis of
gefitinib using 3-hydroxy-4-methoxy benzonitrile has been reported.[20] The procedure included
the introduction of a morpholinopropyl group prior to the formation of the quinazoline ring in
order to suppress N-alkylation, which required extra steps that made this process inefficient and
uneconomical.
Currently, O-alkylation products that contain an amino group are prepared by nitro
reduction or via sequential amine protection, alkylation of the hydroxyl group and amine
deprotection. The amino group is generally protected with an acetyl or phthalyl group, which
requires harsh removing conditions. Thus, expensive tert-butoxycarbonyl (Boc) or
benzoxycarbonyl (Cbz) protecting groups are also utilized.
Recently, we have developed a selective O-alkylation procedure using a transient
trimethylsilyl (TMS) protecting group.[21] The TMS group was attractive to us due to its high N-
alkylation suppressing ability and its facile removal through a simple work-up. As predicted, this
new and unique process minimized the generation of the N-alkylated side product. We also
confirmed that our process is well applicable to a variety of 4-anilinoquinazolines. Herein, we
describe our recent studies on the selective O-alkylation of 4-anilinoquinazolines as illustrated in
Figure 2 and the application of the method to a highly practical and efficient synthesis of gefitinib.
Results and Discussion
We initially investigated the selective O-alkylation of intermediate 2 to prevent undesired
N-alkylation. We explored the effect of the amine salt, catalyst, and protecting group (Table 1) in
order to develop a sustainable and economical process. Use of the acid-amine salt as a masking
group for aniline resulted in a 13.5% yield (entry 1) of the N-alkylation product after completion
4

of the reaction. The use of DMAP or KI as a catalyst gave poor results (entries 2 and 3,
respectively). Interestingly, transient protection with TMSI in the presence of DMAP as a catalyst
afforded only 0.3% of the N-alkylation product (entry 5), compared to approximately 30% for the
previously known method.[19] In addition, the N-alkylated impurity was completely removed by a
simple purification process. The postulated mechanism for the selective O-alkylation is depicted
in Figure 3.
Upon reaction of hydroxyquinazoline (2) with an alkylating agent in the presence of base,
catalyst, and either HCl or TMSI, initial TMS binding to the aniline amine affords the tertiary
amine intermediate (4), which is essentially a TMS-protected form of substrate 2. Due to the steric
hindrance of tertiary amine 4, the amino group seems no longer reactive for alkylation and the less
hindered phenolic hydroxyl group is reactive. Therefore, the O-alkylated intermediate 5 is formed
through selective O-alkylation with morpholinopropyl chloride. The TMS group was easily
removed with water during a simple work-up procedure to obtain gefitinib in high purity. The
transient TMS-protection of the amino group prevented the N-alkylation reaction. We confirmed
the postulated mechanism by analyzing the NMR spectra of the proposed intermediates.
Next, we explored the applicability of the selective O-alkylation reaction to other
anilinoquinazoline moieties. As in the case of gefitinib, selective O-alkylation using transient
protection was possible for a variety of anilinoquinazolines regardless of the aniline substituents
and the alkylation reagent (Table 2).
We further examined the selective O-alkylation of synthetically important aminophenol
moieties under the optimized conditions, as shown in Table 3. Primary amines preferentially
afforded the di-alkylation product resulting from both O- and N-alkylation (10, 11). As anticipated,
3-(phenylamino)phenol showed good selectivity for the O-alkylation product 12. However,
5

alkylation of 3-(benzylamino)phenol produced only the N-alkylated product 13, which is likely
due to the steric hindrance of amino alcohols and the good nucleophilicity of arylalkyl amine.
Deacetylation of 14 according to a reported procedure[19] (r.t., 17 h and then reflux, 1.5 hr)
produced 4-(3-chloro-4-fluorophenylamino)-7-methoxyquinazolin-6-ol (2) along with unreacted
14 (approximately 2%), which ultimately affected the purity of the final gefitinib product. Thus,
an additional purification process to remove the remaining 14 was necessary, which limited the
utilization of the procedure in commercial production. We have also examined other reported
deacetylation conditions[23] as shown in Table 4. Deacetylation of 14 (entry 4) by a lithium
hydroxide treatment was completed in 0.5 h and less than 0.1% of unreacted 14 was observed.
Deacetylation of 14 with sodium hydroxide (entry 2) or potassium hydroxide (entry 3) was
completed in 1 hr, which is shorter than that with NH4OH (18.5 hr), with only approximately 0.1%
of 14 remaining. However, an additional procedure to remove the Na or K adduct was still
necessary because the byproducts were less soluble in the MeOH and water solution used for
deacetylation. Fortunately, the Li adduct was more soluble in the MeOH and water solution, and
the byproduct was easily removed by filtration of the reaction mixture.
Using the established procedures, we completed the synthesis of gefitinib in four reaction
steps starting from 7-methoxy-4-oxo-3,4-dihydroquinazolin-6-yl acetate (15), as illustrated in
Scheme 1. Chlorination of quinazolone 15 followed by condensation with 3-chloro-4-
fluoroaniline at 65°C afforded 14inꢀ quantitative yield. Acetoxyquinazoline 14 was conveniently
converted to hydroxyquinazoline 2 with 90% yield after lithium hydroxide treatment in a mixture
of MeOH and water (1:1). Finally, facile alkylation of hydroxyquinazoline 2 with 4-(3-
chloropropyl)morpholine in the presence of TMSI followed by purification by recrystallization
afforded gefitinib (1) in greater than 99% purity by HPLC.
6

Experimental
1H spectra and 13Cꢀ spectra were recorded using a Bruker DPX 400 Spectrometer. All purity
values were obtained by HPLC analysis using HPLC 1200 Series from Agilent Technologies.
Chemical shifts were expressed in parts per million (ppm, δ) downfield from tetramethylsilane and
were referenced to the DMSO-d6 (2.49ppmꢀ for 1H and 39.5ppmꢀ for 13Cꢀ, respectively). 1H-NMR
data were reported in the order of chemical shift, multiplicity (s, singlet; d, doublet; t, triplet; q,
quartet; m, multiplet and/or multiple resonances), number of protons, and coupling constant in
hertz (Hz). High resolution mass spectra were obtained with a Synapt G2 instrument. Elemental
analysis data were obtained on a Flash 1112 instrument (CE instrument).
HPLC analysis
Hypersil C18, 4.6ꢀ  250 mmꢀ (5 m), λ = 250 nm, flow rate 1.0mLꢀ/min, mobile phase
(60:40) buffer/acetonitrile (buffer 5molꢀ% ammonium acetate aqueous solution).
Selective O-alkylation of 4-anilinoquinazoline moieties with an assistance of
TMSI (Table 2)
General procedure 1 (without TMSI)
To a suspension of a 6-hydroxy quinazoline compound (2.0 mmol) in dimethylsulfoxide
(6.0 mLꢀ) was slowly added potassium carbonate (7.0 mmol) and 4-(3-chloropropyl)morpholine
(2.4 mmol). The mixture was heated to 80°C and stirred for 2 h (termination of the reaction was
confirmed by HPLC and TLC analysis). The reaction mixture was cooled to 20°C and purified
water (12.0 mLꢀ) was slowly added. The mixture was stirred for 30minꢀ and the precipitate was
7

collected by filtration, washed with purified water (12.0 mLꢀ), and dried for 3 h at 50°C to give a
6-alkoxy quinazoline as pale yellow powder.
General procedure 2 (with TMSI)
To a suspension of a 6-hydroxy quinazoline (2.0 mmol) in dimethylsulfoxide (6.0mLꢀ) was
added potassium carbonate (7.0 mmol). The mixture was cooled to 10°C and iodotrimethylsilane
(2.0 mmol) was slowly added. The mixture was stirred for 1 h at 15°C and 4-(3-
chloropropyl)morpholine (2.4 mmol) was slowly added. The mixture was heated to 80°C and
stirred for 2 h (termination of the reaction was confirmed by HPLC and TLC analysis). The
reaction mixture was cooled to 20°C and purified water (12.0mLꢀ) was slowly added. The mixture
was stirred for 30minꢀ and the precipitate was collected by filtration, washed with purified water
(12.0mLꢀ), and dried for 3 h at 50°C to give a 6-alkoxy quinazoline compound as pale yellow
powder.
Alkylation of amino alcohols (Table 3)
General procedure
To a suspension of an amino alcohol (2.0 mmol) in dimethylsulfoxide (6.0mLꢀ) was added
potassium carbonate (7.0 mmol). The mixture was cooled to -10°C and iodotrimethylsilane (2.0
mmol) was slowly added. The mixture was stirred for 1 h at 15°C and 4-(3-
chloropropyl)morpholine (2.4 mmol) was slowly added. The mixture was heated to 80°C and
stirred for 2 h (termination of the reaction was confirmed by HPLC and TLC analysis). The
reaction mixture was cooled to 20°C and purified water (12.0mLꢀ) was slowly added. The mixture
8

was stirred for 30minꢀ and the precipitate was collected by filtration, washed with purified water
(12.0mLꢀ), and dried at 50°C for 3 h to give an alkylated amino alcohol as pale yellow powder.
N-(3-Chloro-4-fluorophenyl)-7-methoxy-6-(3-morpholinopropoxy)quinazoline-4-
amine (1, Gefitinib)
To a suspension of 2 (645.0gꢀ, 2.02molꢀ) in N,N-dimethylformamide (6.5 L) were added
potassium carbonate (975.9gꢀ, 7.06molꢀ) and N,N-dimethylaminopyridine (24.7gꢀ, 0.20molꢀ). The
mixture was cooled to -10°C and iodotrimethylsilane (404.2gꢀ, 2.02molꢀ) was slowly added. The
mixture was stirred for 1 h at 15°C and 4-(3-chloropropyl)morpholine (396.7gꢀ, 2.42molꢀ) in N,N-
dimethylformamide (645.0mLꢀ) was slowly added. The reaction mixture was heated to 80°C,
stirred for 2 h (termination of the reaction was confirmed by HPLC and TLC analysis), and cooled
to 20°C. After purified water (14.2 L) was slowly added, the resulting mixture was stirred for
30minꢀ and the precipitate was collected by filtration, washed with purified water (645.0mLꢀ), and
dried for 3 h at 50°C to give gefitinib as pale yellow powder (812.3gꢀ, 90.1%). HPLC purity:
1
99.21% (the N-alkylating byproduct was not detected). H-NMR (400MHzꢀ, DMSO-d6): δ 1.95-
2.02 (m, 2H), 2.38 (br, 4H), 2.45-2.50 (m, 2H), 3.56-3.58 (m, 4H), 3.93 (s, 3H), 4.17 (t, 2H, J =
6.3Hzꢀ), 7.19 (s, 1H), 7.44 (t, 1H, J = 9.1Hzꢀ), 7.76-7.80 (m, 2H), 8.10 (dd, 1H, J = 4.2Hzꢀ, 2.6Hzꢀ),
8.49 (s, 1H), 9.57 (s, 1H). 13Cꢀ NMR (175MHzꢀ, DMSO-d6): δ 156.5, 155.0, 154.3, 153.1, 152.9,
148.8, 147.4, 137.3, 124.0, 122.8, 119.3, 119.2, 117.0, 109.2, 107.8, 103.0, 67.6, 66.7, 56.3, 55.4,
53.9, 26.3. HRMS: Calcd for C22H24ClFN4O3 [M + H] + 447.1594; Found 447.1599. Elemental
Anal. Calcd for C22H24ClFN4O3: C, 59.13; H, 5.41; N, 12.54. Found: C, 59.09; H, 5.48; N, 12.52.
Conclusion
9

We have developed an efficient and practical large-scale process for the preparation of
gefitinib with high purity (>99% by HPLC). The key step of the synthesis includes selective O-
alkylation with significantly suppressed production of the N-alkylation product by using a transient
TMS protecting group. Our process ultimately provided a significant improvement in both the
yield and purity of the final gefitinib. We further confirmed that our procedure is applicable to a
variety of anilinoquinazolines and can be widely utilized by synthetic and medicinal chemists.
Acknowledgments
This work was supported by the National Research Foundation of Korea (NRF) grant for
the Global Core Research Center (GCRC) funded by the Korea government (MSIP) (No. 2011-
0030001) and by the Global Frontier Project grant of National Research Foundation funded by
Korea government (MSIP) (NRF-2015M3A6A4065798). We acknowledge the Korea Basic
Science Institute, Ochang, Korea, for providing HRMS.
References
[1] Xi, L.; Zhang, J. Q.; Liu, Z. C.; Zhang, J. H.; Yan, J. F.; Jin, Y.; Lin, J. Org. Biomol. Chem.
2013, 11, 4367–4378.
[2] Yun, C. H.; Boggon, T. J.; Li, Y. Q.; Woo, M. S.; Greulich, H.; Meyerson, M.; Eck, M. J.
Cancer Cell 2007, 11, 217–227.
[3] Stamos, J.; Sliwkowski, M. X.; Eigenbrot, C. J. Biol. Chem. 2002, 277, 46265–46272.
[4] Wood, E. R.; Truesdale, A. T.; McDonald, O. B.; Yuan, D.; Hassell, A.; Dickerson, S. H.;
Ellis, B.; Pennisi, C.; Horne, E.; Lackey, K.; Shewchuk, L. Cancer Res. 2004, 64, 6652–6659.
[5] Solca, F.; Dahl, G.; Zoephel, A.; Bader, G.; Sanderson, M.; Klein, C.; Kraemer, O.;
Himmelsbach, F.; Haaksma, E.; Adolf, G. R. J. Pharmacol. Exp. Ther. 2012, 343, 342–350.
[6] Zhang, H. Q.; Gong, F. H.; Li, C. G.; Zhang, C.; Wang, Y. J.; Xu, Y. G.; Sun, L. P. Eur. J.
Med. Chem. 2016, 109, 371–379.
[7] Cheng, W.; Zhu, S.; Ma, X.; Qiu, N.; Peng, P.; Sheng, R.; Hu, Y. Eur. J. Med. Chem. 2015,
89, 826–834.
10

[8] Cheng, W.; Yuan, Y.; Qiu, N.; Peng, P.; Sheng, R.; Hu, Y. Bioorg. Med. Chem. 2014, 22,
6796–6805.
[9] Zhao, F.; Lin, Z.; Wang, F.; Zhao, W.; Dong, X. Bioorg. Med. Chem. Lett. 2013, 23, 5385–
5388.
[10] Li, R. D.; Zhang, X.; Li, Q. Y.; Ge, Z. M.; Li, R. T. Bioorg. Med. Chem. Lett. 2011, 21,
3637–3640.
[11] Chandregowda, V.; Kush, A. K.; Reddy, G. C. Eur. J. Med. Chem. 2009, 44, 3046–3055.
[12] Barbosa, M. L. C.; Lima, L. M.; Tesch, R.; Sant’Anna, C. M. R.; Totzke, F.; Kubbutat, M.
H. G.; Schächtele, C.; Laufer, S. A.; Barreiro, E. J. Eur. J. Med. Chem. 2014, 71, 1–14.
[13] Pham, H. T.; Hanson, R. N.; Olmsted, S. L.; Kozhushnyan, A.; Visentin, A.; Weglinsky, P.
J.; Massero, C.; Bailey, K. Tetrahedron Lett. 2011, 52, 1053–1056.
[14] Baselga J.; Averbuch, S. D. Drugs, 2000, 60, 33–40.
[15] Federal Drug Administration. Approved labeling for gefitinib. Available at:
http://www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm (accessed July 12, 2004).
[16] Chandregowda, V.; Rao, G. V.; Reddy, G. C. Org. Process Res. Dev. 2007, 11, 813–816.
[17] Chandregowda, V.; Rao, G. V.; Reddy, G. C. Synth. Commun. 2007, 37, 3409–3415.
[18] Marzaroa, G.; Guiottoa, A.; Pastorinib, G.; Chilina, A. Tetrahedron, 2010, 66, 962–968.
[19] Gibson, K. H. PCT Int Appl. WO 1996/033980.
[20] Gilday, J. P.; Moody, D. PCT Int Appl. WO 2004/024703.
[21] Lee, S. U.; Woo, D. G.; Kang, S. K.; Kim, D. J.; Park, D. J. PCT Int Appl. WO
2013/180403.
[22] Kumar, N.; Chowdhary, A.; Gudaparthi, O.; Patel, N. G.; Soni, S. K.; Sharma, P. Indian J.
Chem. Sect. B, 2014, 1269–1274.
[23] Cai, X.; Qian, C.; Gould, S.; Zhai, H. PCT Int Appl. WO 2008/33747.
11

Table 1. O-alkylation of 2 with or without an assistance of TMSI as a transient protection group
Entry N-
protecting
Catalyst
N-alkylation products N-alkylation
Yieldc
after (%)
before purificationa products
agent
HCl
–
(%)
purificationb (%)
1
2
–
13.5
1.2
1.0
64.5
61.6
DMAP (0.1 10.0
eq.)
3
4
5
–
KI (0.1 eq.)
20.0
2.2
1.3
63.9
78.5
90.1
TMSI
TMSI
–
0.2
DMAP (0.1 0.3
eq.)
N.D.
aDetermined by HPLC after completion of reaction.
bDetermined by HPLC after purification.
cIsolated yield after purification.
12

Table 2. Selective O-alkylation of 4-anilinoquinazoline moieties with an assistance of TMSI
Entry R1
R2
N-protecting
N-
Yieldb
agent
alkylation
producta
(%)
(%) (product)
1
2
3
4
5
6
7
8
–
23.2
64.5
TMSI
–
0.2
88.6 (6)
53.9
39.7
1.3
TMSI
–
83.5 (7)
54.4
30.6
9.5
TMSI
–
75.0 (8)
62.0
21.5
4.4
TMSI
80.2 (9)
aDetermined by HPLC after completion of reaction.
bIsolated yield after purification.
13

Table 3. TMSI-assisted alkylation of amino alcohols
Amino
Alcohol
Major
di-alkylation
product (10)
56.2
(11)
O-alkylation product N-alkylation product
Product
Isolated
yield (%)
(12)
(13)
61.5
78.3
69.7
14

Table 4. Examination of deacetylation conditions for the preparation of intermediate 2
Entry Reagent
Temperature
r.t. to reflux
Time (h) Remaining 14a (%)
Yieldb (%)
90.5
1
NH4OH
18.5
2.0
2
NaOH
r.t.
1
0.1
90.2
3
4
KOH
LiOH
r.t.
r.t.
1
0.1
0.1
91.6
90.0
0.5
aDetermined by HPLC after completion of reaction.
bIsolated yield.
15

Figure 1. Current drugs consisting of 4-anilinoquinazoline.
16

Figure 2. Strategy for the synthesis of 6-alkoxy quinazolines.
17

Figure 3. Postulated mechanism for selective O-alkylation of 2 using a transient protecting
group.
18

Scheme 1. Synthesis of gefitinib (1).
19