Editorial
Epidermal Growth Factor Receptor Inhibitors: A Moving Target?
Susan E. Bates and Tito Fojo
It has been over 20 years since the first agents targeting the
epidermal growth factor receptor (EGFR) pathway were
developed by Mendelsohn and Baselga (1). In vitro studies
demonstrating that antibodies inhibiting the pathway could
retard cell growth, and markedly so in some cell lines, led
ultimately to clinical trials testing EGFR inhibition as a
therapeutic strategy. As the EGFR pathway was dissected, it
was recognized that EGFR was a member of a receptor
tyrosine kinase family comprising four members that dimerize
or heterodimerize with activation (2). Diversity in dimeriza-
tion and differences in ligand binding likely result in a
spectrum of biological phenotypes. HER2/ErbB2, the second
member of the family to be identified, engenders an aggressive
subtype of breast cancer but is inhibited by the antibody
Herceptin, generating major responses in breast cancer. A
decade after the introduction of antibodies inhibiting EGFR,
small-molecule inhibitors were identified that impaired EGFR
tyrosine kinase activity (3). Several inhibitors have now been
brought to clinical trials, leading to the approval and
marketing of two of them—gefitinib and erlotinib—in lung
cancer. Much has been written recently about these agents and
it will not be the task of this editorial to provide a detailed
review.
Previous studies, including one by the authors of this
editorial, have noted that the level of expression of EGFR
cannot predict the sensitivity of cells to inhibitors (4). Bishop
et al.’s examination of the National Cancer Institute Drug
Screen database revealed that the 60 cell lines could be grouped
based on sensitivity to EGFR inhibition. Consistent with the
notion that a threshold level of EGFR is required for sensitivity
to agents targeting the EGFR pathway, cells with low levels of
EGFR expression were found to be insensitive to quinazoline
inhibitors. Moreover, cells with high levels of EGFR could be
divided into two groups composed of cells sensitive or
insensitive to EGFR inhibition. Yet, inhibition of EGFR and
mitogen-activated protein kinase phosphorylations could be
shown in both groups following exposure to AG1478, an
antecedent to gefitinib and erlotinib. These and the observa-
tions of others indicated that there were differences in the
EGFR pathway among cell lines and an independence from
mitogen-mediated signaling was inferred as an explanation for
insensitivity to EGFR inhibition. Clinical trials have confirmed
these in vitro findings, demonstrating that inhibition of both
EGFR and mitogen-activated protein kinase phosphorylations
in tumors can occur in the absence of disease response (5 – 8).
Authors’ Affiliation: Center for Cancer Research, National Cancer Institute,
Bethesda, Maryland
Received 8/22/05; accepted 8/22/05.
Requests for reprints: Susan E. Bates, Center for Cancer Research, National
Cancer Institute, Building 10, Room 12N226, Bethesda, MD 20892. Phone: 301-
402-1357; Fax: 301-402-1608; E-mail: sebates@ helix.nih.gov.
doi:10.1158/1078-0432.CCR-05-1845
www.aacrjournals.org
Just a few years ago, it was widely expected that EGFR
inhibitors would be effective in a wide range of solid tumors
with high levels of EGFR including breast, ovarian, lung, renal,
head and neck, pancreatic, and colorectal malignancies (9).
Thus, it was disappointing to find that response rates in phase II
trials in these diseases were low, as in lung cancer, or nonexis-
tent, as in renal, colon, ovarian, or breast cancer (5, 10 – 15).
Why were our expectations so underserved? Our original
paradigm—that EGFR expression could be equated with
dependence on the pathway and could predict eventual drug
sensitivity—was incorrect.
Gefitinib first, and then erlotinib, was licensed for lung
cancer based on results in patients with locally advanced or
metastatic non – small cell lung cancer after failure of front-line
chemotherapy. For erlotinib, approval was based on a study of
731 patients in which the response rate was 8.9% with a
survival advantage of only 2 months (16). Despite the initial
approval of gefitinib based on a 9% to 19% response rate in
non – small cell lung cancer, a survival advantage could not be
shown in later randomized trials and the Food and Drug
Administration has since limited the scope of its approval (17).
Neither agent showed a survival benefit in combination with
chemotherapy in randomized trials. The low response rates
underscore the critical question of how to predict which subset
of patients will benefit from the targeted therapy.
How then can differential responsiveness be explained?
Herceptin offered the first paradigm for successful inhibition
of an EGFR family member: successful inhibition required
significant overexpression. Immunohistochemical analysis of
ErbB2 expression showed that a 2+ or 3+ level of expression,
predominantly due to amplification of the encoding gene, was
required for a meaningful response (18). This observation led
to the development of diagnostic tests for HER2 expression
and amplification and to careful selection of patients for
therapy (19).
The applicability of the HER2 paradigm to EGFR is
uncertain. Until recently, EGFR gene amplification had been
principally described in glioma as a cause of overexpression.
About 50% of human glioblastomas display amplification of
EGFR and about 40% of these have amplified EGFRvIII, a
variant receptor with ligand-independent kinase activity due to
deletion of the extracellular ligand-binding domain. This
mutant receptor has been shown to be resistant to gefitinib
(20, 21). In a recently reported phase I trial of erlotinib in
glioma, eight responses were noted in 41 patients (22). Ten of
39 tumors were noted to have EGFR amplification; four
responses were noted in the group with amplification and
four in the group without amplification. Two of the amplified,
nonresponding tumors expressed mutant EGFRvIII. Evaluating
phospho-Akt as a downstream marker, no tumor positive for
phospho-Akt responded to erlotinib. Thus, gene amplification
alone is not sufficient to provoke a response to EGFR
inhibitors.
7203 Clin Cancer Res 2005;11(20) October 15, 2005
Editorial
Before the clinical trials with gefitinib, overexpression of
EGFR was thought to be the principal mechanism of pathway
activation in non – small cell lung cancer and solid tumors
except for gliomas. This fueled speculation that response to
inhibitors would correlate with expression. However, muta-
tions in the EGFR tyrosine kinase domain were then
discovered among a subset of lung cancers in patients with
tumor responses following gefitinib or erlotinib (23 – 26).
These mutations result in a higher level of receptor phosphor-
ylation and activation. In the initial reports, a high concor-
dance was found between the presence of a mutant EGFR and
response to therapy, with mutations reported in 25 of 31
(81%) patients with an objective tumor response (23). By
comparison, no mutations were detected in 29 patients whose
tumors did not respond to therapy. This observation created
an evolving paradigm that EGFR inhibitors might be effective
only in cells dependent on EGFR signaling, such as would
occur in cells with an activating EGFR mutation. Three
subsequent studies have confirmed and extended these
findings (26 – 28). Further, in the studies by Taron et al. from
the Spanish Lung Cancer Group and by Mitsudomi et al., in
which a high response rate was documented in patients with
tumors harboring mutations, a convincing increase in median
survival following gefitinib was also reported (26, 28).
Whereas the presence of EGFR mutations in non – small cell
lung cancer is well documented, similar EGFR mutations have
not been systematically observed in other tumor types,
potentially limiting this paradigm to non – small cell lung
cancer (29).
The discovery of these mutations led to a paradigm shift for
lung cancer that did not resolve the question of whether
uncomplicated high-level or even low-level expression of EGFR
can confer sensitivity to EGFR inhibitors. Two studies that
conflict with the mutation paradigm suggest that in non – small
cell lung cancer, survival following a tyrosine kinase inhibitor
is enhanced in patients whose tumors contain a high copy
number of EGFR (27, 30). What mechanisms might regulate
EGFR expression and be important in other tumors? One
example is an intron 1 polymorphism in the EGFR gene—the
number of CA single sequence repeats (31, 32). This
polymorphism has been correlated with EGFR expression
levels and has been suggested as a mediator of responsiveness
to EGFR inhibition. Without gene amplification, head and
neck cancer cells with a lower number of CA dinucleotides had
higher levels of EGFR and enhanced responsiveness to erlotinib
(33). Although these results suggest that EGFR expression alone
could create an EGFR-dependent environment that would be
susceptible to inhibition, the failure of EGFR inhibitors in
colon and renal cell carcinoma argues against such a simple
explanation. These results do, however, suggest that mutation
or gene amplification is not required to activate the EGFR
pathway.
Two trials examining the activity of gefitinib in breast cancer
showed disappointing results without evidence of objective
responses (8, 15). In one of the studies, the effect of gefitinib
on the EGFR pathway was examined in both skin and tumor
biopsies (8). Inhibition of both EGFR and mitogen-activated
protein kinase phosphorylations was observed in normal and
malignant cells. However, Ki67 staining, a marker of proli-
feration, was reduced in skin but not in tumors. Similarly, a
clinical trial evaluating erlotinib in breast cancer found only
Clin Cancer Res 2005;11(20) October 15, 2005 7204
1 of the 15 tumors examined to be EGFR positive; in this
tumor, both EGFR and mitogen-activated protein kinase
phosphorylations were reduced after erlotinib therapy. Skin
samples from all 15 patients showed inhibition of phospho –
mitogen-activated protein kinase and Ki67 (5). None of the
tumors responded to erlotinib. Both studies concluded that
the absence of tumor responses was due to a lack of EGFR
dependence in the tumors rather than a failure of receptor
inhibition.
Writing in the current issue of Clinical Cancer Research, Van
Schaeybroeck and colleagues offer a surrogate marker for
successful inhibition of the EGFR pathway in colon cancer cell
lines. The identification of a surrogate marker for sensitivity
provides insight into the occasional responses—those tumors
capable of activating the pathway are those likely to be
sensitive to its inhibition. Van Schaeybroeck and colleagues
show that the sensitivity of colon cancer cells to the EGFR
inhibitor gefitinib can be discriminated by the higher basal
level of EGFR autophosphorylation and by the induction of
phosphorylation that results from exposure to the chemother-
apy agent oxaliplatin. These results, if confirmed clinically,
provide both a potential surrogate for successful therapy with
an EGFR inhibitor and a potential insight into another
mechanism whereby the EGFR pathway plays a critical role
in oncology. Beyond mutation or constitutive activation, there
might be tumors in which cellular stress such as DNA damage
results in activation and phosphorylation of the EGFR
pathway. This activation of the pathway could result in
proliferation or enhanced survival so that disruption of the
pathway by an EGFR inhibitor could conceivably enhance
drug sensitivity.
Thus, we can now hypothesize rules for successful EGFR
inhibitor therapy. The gene can be amplified, mutated, or
overexpressed, but it must be critical to the oncogenic
phenotype. The overriding paradigm seems to be that the
pathway must be involved in maintenance of the malignant
phenotype or in cell survival to achieve a successful
antineoplastic effect. Considering the alacrity with which
paradigms have been amended in recent years, this one will
undoubtedly be amended as well and eventually we will get it
right. As a scientific community, we must deliver on the
promise of targeted therapy. Inherent in the often-used (and
abused) term ‘‘targeted therapy’’ is the implication that those
who will benefit from such therapy can be identified
prospectively. We cannot develop targeted therapy and then
fail to meet the challenge of finding the subset of patients who
are likely to benefit. A 2-month survival advantage in a
population derived from a therapy administered to every
patient with lung cancer cannot be considered to be
molecularly targeted therapy. Really, how different is the latter
from vincristine, a therapy that specifically targets tubulin and
has been available to us for nearly 50 years? Let us agree that
by definition, a molecularly targeted therapy will be individ-
ualized therapy. Developing a clinically relevant assay to detect
mutations will be a first step. Developing a surrogate assay that
will detect tumors with activation of the pathway will be a
second step. Then we will see the day when agents such as
erlotinib and gefitinib will be prescribed only for patients
who will benefit and the 85% who will not benefit will be
spared the cost, toxicity, and disappointment of an ineffective
therapy.
www.aacrjournals.org

References
1. Mendelsohn J, Baselga J. The EGF receptor family
as targets for cancer therapy. Oncogene 2000;19:
6550 ^ 65.
2. Ritter CA, Arteaga CL. The epidermal growth factor
receptor-tyrosine kinase: a promising therapeutic tar-
get in solid tumors. Semin Oncol 2003;30:3 ^ 11.
3. Osherov N, Levitzki A. Epidermal-growth-factor-
dependent activation of the src-family kinases. Eur J
Biochem 1994;225:1047 ^ 53.
4. Bishop PC, Myers T, Robey R, et al. Differential sensi-
tivity of cancer cells to inhibitors of the epidermal
growth factor receptor family. Oncogene 2002;21:
119 ^ 27.
5. Tan AR,Yang X, Hewitt SM, et al. Evaluation of bio-
logic end points and pharmacokinetics in patients with
metastatic breast cancer after treatment with erlotinib,
an epidermal growth factor receptor tyrosine kinase
inhibitor. J Clin Oncol 2004;22:3080 ^ 90.
6. Daneshmand M, Parolin DA, Hirte HW, et al. A phar-
macodynamic study of the epidermal growth factor
receptor tyrosine kinase inhibitor ZD1839 in meta-
static colorectal cancer patients. Clin Cancer Res
2003;9:2457 ^ 64.
7. Baselga J, Rischin D, Ranson M, et al. Phase I safety,
pharmacokinetic, and pharmacodynamic trial of
ZD1839, a selective oral epidermal growth factor re-
ceptor tyrosine kinase inhibitor, in patients with five
selected solid tumor types. J Clin Oncol 2002;20:
4292 ^ 302.
8. Baselga J, Albanell J, Ruiz A, et al. Phase II and
tumor pharmacodynamic study of gefitinib in patients
with advanced breast cancer. J Clin Oncol 2005;23:
5323 ^ 33.
9. Salomon DS, Brandt R, Ciardiello F, Normanno N.
Epidermal growth factor-related peptides and their
receptors in human malignancies. Crit Rev Oncol
Hematol 1995;19:183 ^ 232.
10. Perez-Soler R, Chachoua A, Hammond LA, et al.
Determinants of tumor response and survival with
erlotinib in patients with non-small-cell lung cancer.
J Clin Oncol 2004;22:3238 ^ 47.
11. Jermann M, Stahel RA, Salzberg M, et al. A phase II,
open-label study of gefitinib (IRESSA) in patients with
locally advanced, metastatic, or relapsed renal-cell
carcinoma. Cancer Chemother Pharmacol 2005;1 ^ 7.
www.aacrjournals.org
12. Dawson NA, Guo C, Zak R, et al. A phase II trial of
gefitinib (Iressa, ZD1839) in stage IV and recurrent
renal cell carcinoma. Clin Cancer Res 2004;10:
7812 ^ 9.
13. Kindler HL, Friberg G, Skoog L, Wade-Oliver K,
Vokes EE. Phase I/II trial of gefitinib and oxaliplatin in
patients with advanced colorectal cancer. Am J Clin
Oncol 2005;28:340 ^ 4.
14. Schilder RJ, Sill MW, Chen X, et al. Phase II study of
gefitinib in patients with relapsed or persistent ovarian
or primary peritoneal carcinoma and evaluation of epi-
dermal growth factor receptor mutations and immuno-
histochemical expression: a Gynecologic Oncology
Group Study. Clin Cancer Res 2005;11:5539 ^ 48.
15. von Minckwitz G, Jonat W, Fasching P, et al. A
multicentre phase II study on gefitinib in taxane- and
anthracycline-pretreated metastatic breast cancer.
Breast Cancer Res Treat 2005;89:165 ^ 72.
16. Cohen MH, Johnson JR, Chen YF, Sridhara R,
Pazdur R. FDA drug approval summary : erlotinib
(Tarceva) tablets. Oncologist 2005;10:461 ^ 6.
17. Giaccone G. Epidermal growth factor receptor inhib-
itors in the treatment of non-small-cell lung cancer.
J Clin Oncol 2005;23:3235 ^ 42.
18. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of
chemotherapy plus a monoclonal antibody against
HER2 for metastatic breast cancer that overexpresses
HER2. N Engl J Med 2001;344:783 ^ 92.
19. Ross JS, Fletcher JA, Bloom KJ, et al.Targeted ther-
apy in breast cancer: the HER-2/neu gene and protein.
Mol Cell Proteomics 2004;3:379 ^ 98.
20. Learn CA, Hartzell TL, Wikstrand CJ, et al. Resis-
tance to tyrosine kinase inhibition by mutant epidermal
growth factor receptor variant III contributes to the
neoplastic phenotype of glioblastoma multiforme. Clin
Cancer Res 2004;10:3216 ^ 24.
21. Li B, Chang CM, Yuan M, McKenna WG, Shu HK.
Resistance to small molecule inhibitors of epidermal
growth factor receptor in malignant gliomas. Cancer
Res 2003;63:7443 ^ 50.
22. Haas-Kogan DA, Prados MD, Tihan T, et al. Epider-
mal growth factor receptor, protein kinase B/Akt, and
glioma response to erlotinib. J Natl Cancer Inst 2005;
97:880 ^ 7.
23. Lynch TJ, Bell DW, Sordella R, et al. Activating
7205 Clin Cancer Res 2005;11(20) October 15, 2005
EGFR Inhibitors: A MovingTarget?
mutations in the epidermal growth factor receptor un-
derlying responsiveness of non-small-cell lung cancer
to gefitinib. N Engl J Med 2004;350:2129 ^ 39.
24. Paez JG, Janne PA, Lee JC, et al. EGFR mutations
in lung cancer: correlation with clinical response to
gefitinib therapy. Science 2004;304:1497 ^ 500.
25. Pao W, Miller V, Zakowski M, et al. EGF receptor
gene mutations are common in lung cancers from
‘‘never smokers’’ and are associated with sensitivity of
tumors to gefitinib and erlotinib. Proc Natl Acad Sci
U S A 2004;101:13306 ^ 11.
26. Mitsudomi T, KosakaT, Endoh H, et al. Mutations of
the epidermal growth factor receptor gene predict
prolonged survival after gefitinib treatment in patients
with non-small-cell lung cancer with postoperative
recurrence. J Clin Oncol 2005;23:2513 ^ 20.
27. Cappuzzo F, Hirsch FR, Rossi E, et al. Epidermal
growth factor receptor gene and protein and gefitinib
sensitivity in non-small-cell lung cancer. J Natl Cancer
Inst 2005;97:643 ^ 55.
28. Taron M, Ichinose Y, Rosell R, et al. Activating
mutations in the tyrosine kinase domain of the epider-
mal growth factor receptor are assocated with im-
proved survival in gefitinib-treated chemorefractory
lung adenocarcinomas. Clin Cancer Res 11:5878 ^ 85.
29. Lee JW, SoungYH, Kim SY, et al. Absence of EGFR
mutation in the kinase domain in common human can-
cers besides non-small cell lung cancer. Int J Cancer
2005;113:510 ^ 1.
30. Tsao MS, Sakurada A, Cutz JC, et al. Erlotinib in lung
cancerHmolecular and clinical predictors of outcome.
N Engl J Med 2005;353:133 ^ 44.
31. Buerger H, Gebhardt F, Schmidt H, et al. Length and
loss of heterozygosity of an intron 1 polymorphic se-
quence of egfr is related to cytogenetic alterations
and epithelial growth factor receptor expression.
Cancer Res 2000;60:854 ^ 7.
32. Gebhardt F, Zanker KS, Brandt B. Modulation of epi-
dermal growth factor receptor gene transcription by a
polymorphic dinucleotide repeat in intron 1. J Biol
Chem 1999;274:13176 ^ 80.
33. Amador ML, Oppenheimer D, Perea S, et al. An epi-
dermal growth factor receptor intron 1 polymorphism
mediates response to epidermal growth factor recep-
tor inhibitors. Cancer Res 2004;64:9139 ^ 43.