Detection of Mutations in EGFR in Circulating Lung-Cancer Cells.
Detection of Mutations in EGFR in Circulating Lung-Cancer Cells.
Detection of Mutations in EGFR in Circulating Lung-Cancer Cells.
original article
A bs t r ac t
Background
From the Massachusetts General Hospi- The use of tyrosine kinase inhibitors to target the epidermal growth factor receptor
tal Cancer Center (S.M., L.V.S., L.U., B.B., gene (EGFR) in patients with non–small-cell lung cancer is effective but limited by
E.I., S. Diederichs, D.W.B., S. Digumarthy,
A.M., J.S., T.J.L., D.A.H.), Biomicroelec- the emergence of drug-resistance mutations. Molecular characterization of circulat-
tromechanical Systems Resource Center ing tumor cells may provide a strategy for noninvasive serial monitoring of tumor
(S.N., C.V.C., D.I., R.G.T., M.T.), Depart- genotypes during treatment.
ments of Surgery (S.M., S.N., D.I., R.G.T.,
M.T.), Medicine (L.V.S., T.J.L., D.A.H.),
Pathology (A.J.I.), and Radiology (S. Digu- Methods
marthy), and Biostatistics Unit (A.M.), We captured highly purified circulating tumor cells from the blood of patients with
Massachusetts General Hospital and Har-
vard Medical School; Shriners Hospital
non–small-cell lung cancer using a microfluidic device containing microposts coat-
for Children (R.G.T., M.T.); and the How- ed with antibodies against epithelial cells. We performed EGFR mutational analysis
ard Hughes Medical Institute (D.A.H.) on DNA recovered from circulating tumor cells using allele-specific polymerase-
— all in Boston. Address reprint requests
to Dr. Haber at Massachusetts General
chain-reaction amplification and compared the results with those from concurrently
Hospital Cancer Center, CNY-7, Bldg. 149, isolated free plasma DNA and from the original tumor-biopsy specimens.
13th St., Charlestown, MA 02129, or at
[email protected]. Results
Drs. Maheswaran, Sequist, and Nagrath We isolated circulating tumor cells from 27 patients with metastatic non–small-cell
contributed equally to this article. lung cancer (median number, 74 cells per milliliter). We identified the expected
EGFR activating mutation in circulating tumor cells from 11 of 12 patients (92%)
This article (10.1056/NEJMoa0800668) was
published at www.nejm.org on July 2, 2008. and in matched free plasma DNA from 4 of 12 patients (33%) (P = 0.009). We de-
tected the T790M mutation, which confers drug resistance, in circulating tumor
N Engl J Med 2008;359:366-77. cells collected from patients with EGFR mutations who had received tyrosine kinase
Copyright © 2008 Massachusetts Medical Society.
inhibitors. When T790M was detectable in pretreatment tumor-biopsy specimens,
the presence of the mutation correlated with reduced progression-free survival (7.7
months vs. 16.5 months, P<0.001). Serial analysis of circulating tumor cells showed
that a reduction in the number of captured cells was associated with a radiograph-
ic tumor response; an increase in the number of cells was associated with tumor
progression, with the emergence of additional EGFR mutations in some cases.
Conclusions
Molecular analysis of circulating tumor cells from the blood of patients with lung
cancer offers the possibility of monitoring changes in epithelial tumor genotypes
during the course of treatment.
I
ncreasing knowledge of molecular relapse show the acquisition of a secondary EGFR
abnormalities that drive human cancers of- mutation, in which methionine is substituted for
fers the promise of therapies targeted at spe- threonine at position 790 (T790M). This mutation
cific genetic lesions.1,2 Genetic abnormalities hinders drug binding but may be susceptible to
may define a cancer at diagnosis, but mutations, second-generation, “irreversible” tyrosine kinase
some of which lead to acquired drug resistance, inhibitors, which form covalent cross-links with
may emerge during treatment. For many epithe- the receptors.16-18 Other mechanisms of resistance
lial cancers, minimally invasive biopsies provide to tyrosine kinase inhibitors have also been re-
insufficient material for molecular analysis at di- ported.19,20 We tested the ability of microfluidic
agnosis, and tumors typically are not sampled techniques to isolate a sufficient number of cir-
repeatedly during treatment to monitor changes culating tumor cells from patients with non–
in genetic abnormalities. Although tumor cells small-cell lung cancer to permit mutational anal-
are known to circulate in the blood of patients ysis of EGFR.
with metastatic cancer,3 their use in monitoring
of tumor genotypes has been limited by relatively Me thods
insensitive detection strategies.4,5 The detection
of circulating tumor cells in some patients with Patients and Clinical Specimens
the use of magnetic bead–conjugated antibodies Patients with advanced non–small-cell lung can-
against epithelial-cell adhesion molecule (EpCAM) cer were recruited according to one of two proto-
may be useful as a prognostic marker.6-9 How- cols that were approved by the institutional re-
ever, the small number of circulating tumor cells view board. A total of 31 patients in Group A
isolated by this method is below the dynamic (Patients 1 to 27 and 43 to 46), who were treated
range required for measuring treatment response, at the Massachusetts General Hospital Cancer
and the low purity of such cells prevents reliable Center, donated 10 ml of blood on one or more
molecular analyses.10 occasions for CTC-chip analysis. Blood samples
We recently developed a microfluidic-based de- were analyzed for the quantity of circulating tu-
vice (called the CTC-chip) that can isolate, quantify, mor cells11 (for details, see the Methods section
and analyze circulating tumor cells from a blood in the Supplementary Appendix, available with
sample. In the CTC-chip, blood flows past 78,000 the full text of this article at www.nejm.org). We
EpCAM-coated microposts under controlled con- analyzed circulating tumor cells, free plasma
ditions that optimize the capture of circulating DNA, archived paraffin-embedded tumor tissue,
tumor cells.11 An average of 132 circulating tumor or all three specimens for EGFR mutations using
cells per milliliter (median, 67 cells per milliliter) the Scorpion Amplification Refractory Mutation
are isolated at high purity from virtually all tested System (SARMS) technology (DxS), standard nu-
patients with metastatic cancers — including cleotide sequencing, or both.
non–small-cell lung cancer and prostate, pancreas, The number of tumor-biopsy specimens that
breast, and colorectal cancers — but not from were available for comparison of EGFR sequenc-
healthy controls.11 The prevalence and quantity ing and SARMS analysis was extended by the in-
of circulating tumor cells that are isolated from clusion of 15 patients in Group B (Patients 28 to
patients with advanced cancer may thus provide 42) who had participated in a multicenter clinical
a measure of tumor response, whereas the high trial of gefitinib21 but were not available for the
purity of such cells allows repeated analysis of analysis of circulating tumor cells. We reviewed
molecular markers. the medical charts of all patients, and an inde-
Tumor-associated activating mutations in the pendent radiologist quantified the tumor burden
epidermal growth factor receptor (EGFR) gene at various times as the sum of the unidimen-
identify patients with non–small-cell lung can- sional size of all measurable tumor sites, accord-
cer who have a dramatic response to EGFR ty- ing to the Response Evaluation Criteria in Solid
rosine kinase inhibitors, including gefitinib Tumors (RECIST).22 Patients who had been treat-
(Iressa) and erlotinib (Tarceva).12-14 However, ed with an EGFR tyrosine kinase inhibitor (gefi-
most patients have a relapse within 1 year after tinib or erlotinib) were assessed for the best re-
the initiation of therapy.15 Studies of tumors at sponse to therapy with the use of RECIST.
Molecular Analysis DNA from plasma was isolated with the use of
DNA that was extracted from captured circulat- plasma preparation tubes (Vacutainer PPT) and
ing tumor cells with the use of a PicoPure DNA the QIAmp DNA Blood Midi Kit (Fisher Scientific)
Extraction Kit (Molecular Devices) was subjected and a standard method using proteinase K. For
to two rounds of linear amplification with a identification of EGFR mutations with the SARMS
TransPlex amplification kit (Rubicon Genomics). assay, 1.5 ng of DNA was analyzed with the use
Table 1. Detection of Circulating Tumor Cells in Patients with Non–Small-Cell Lung Cancer.*
Patient No. and EGFR Time since Previous Systemic Tumor Circulating
Mutation Status Sex Age Histologic Features Diagnosis Therapy† Burden‡ Tumor Cells§
yr mo cm no. per ml
EGFR mutation present
1 M 58 Adeno 3.2 None 19.8 156
2 M 55 Adeno 14.4 C, E 2.4 50
3 F 66 Adeno 18.2 G, C 19.5 9
4 M 59 Adeno/BAC 20.7 G 2.0 771
5 M 57 Adeno 10.8 C 1.5 152
6 F 74 Adeno 18.3 E 9.8 5
7 M 64 NSCLC 13.1 G, E, C 4.6 196
8 F 70 Adeno 1.4 None 5.8 175
9 F 63 Adeno 0.9 None 28.6 143
10 F 66 Adeno/BAC 1.3 None 8.2 112
11 F 74 Adeno/BAC 4.8 G 4.5 74
12 F 62 Adeno/BAC 56.8 G, E, O 7.2 9
13 M 27 Adeno 9.9 G 4.1 47
14 F 55 Adeno 11 G 5.2 241
15 F 70 Adeno/BAC 54.7 G, C 30.2 31
16 M 53 Adeno/BAC 9.2 E 7.0 49
17 F 60 Adeno/BAC 97.0 G 4.3 103
18 F 37 Adeno 29.4 G 13.1 70
19 M 66 Adeno/BAC 3.0 None 8.8 20
20 M 63 Adeno 40.7 G, E, C 7.3 64
21 F 58 Adeno 19.7 C, E 2.1 107
22 M 86 Adeno 6.4 E 1.4 62
23 M 65 NSCLC 5.0 E 3.8 62
EGFR mutation absent
24 F 65 Adeno 1.1 None 22.1 538
25 F 71 Adeno 4.6 C 6.7 219
26 M 47 Adeno 0.1 None 26.6 84
27 F 70 Adeno 15.4 C, E 3.6 43
* Adeno denotes adenocarcinoma, Adeno/BAC adenocarcinoma with bronchoalveolar features, C chemotherapy, E erlotinib, G gefitinib,
NSCLC non–small-cell lung cancer not otherwise specified, and O other or experimental agent.
† Previous systemic therapies are listed in the order they were administered.
‡ The tumor burden was measured by unidimensional diameter, according to the Response Evaluation Criteria in Solid Tumors (RECIST).
§ The number of circulating tumor cells per milliliter was calculated on the basis of the analysis of 1 to 5 ml of whole blood per patient. Patients
are listed in the order of specimen collection. Each blood sample was processed once through the CTC-chip. There was no correlation be-
tween the tumor burden and the number of circulating tumor cells (Spearman’s correlation coefficient, −0.02; P = 0.88).
of ABI 7500 Real-Time PCR System (Applied Bio- surements that were performed close to the time
systems). The assay detects grouped deletions with- of analysis of circulating tumor cells (median,
in exon 19, insertions within exon 20, and muta- 8 days; range, 0 to 38) showed that the quantity
tions affecting codon 719 (G719X), as well as the of circulating tumor cells at a single time point
individual mutations T790M, L858R, L861Q, and was not well correlated with simple tumor vol-
S768I. The rate of amplification of these mutant ume (Spearman’s correlation coefficient, −0.028;
alleles was compared with that of EGFR exon 2 as P = 0.88). This finding suggested that additional
an internal control. Standard bidirectional nucle- tumor characteristics, such as invasiveness and
otide sequencing was performed with the use of vascularity, probably influenced the number of
dye terminator chemistry and a Capillary ABI 3100 circulating tumor cells.
sequencer (Applied Biosystems).
Detection of EGFR Mutations in Tumors
Statistical Analysis We tested the suitability of the allele-specific
The relationship between the quantity of circu- SARMS assay23 for detecting EGFR mutations in
lating tumor cells and tumor burden was ana- rare cell populations. This test is designed to de-
lyzed with the use of Spearman’s correlation co- tect multiple drug sensitivity-associated types of
efficient. Fisher’s exact test was used to compare EGFR mutation, including the multiple in-frame
mutations that were identified in different popu- exon 19 deletions (collectively analyzed as “Del”
lations. The relationship between patients’ base- mutations) and the L858R missense mutation,
line T790M status and progression-free survival which together comprise 90% of EGFR mutations.
(the time between the initiation of therapy with a The test also detects the T790M mutation associ-
tyrosine kinase inhibitor and either tumor pro- ated with resistance to tyrosine kinase inhibi-
gression or death) was analyzed with a multi- tors.16-18 To validate the results of the SARMS
variate Cox model and the Kaplan–Meier method assay, we first analyzed 26 paraffin-embedded
with a log-rank test (for details, see the Supple- tumors of non–small-cell lung cancers previously
mentary Appendix). identified as having the EGFR mutation and 8 spec-
imens reported as having wild-type alleles by se-
R e sult s quencing (Table 2). The SARMS assay and sequenc-
ing identified the same mutation in 25 tumor
Identification of Circulating Tumor Cells specimens, whereas all 8 wild-type tumors were
Blood samples were obtained from 23 patients confirmed as being negative, yielding a sensitiv-
with EGFR mutant tumors, including 5 patients who ity of 96% and a specificity of 100%. The single
had undergone no previous treatment, 10 patients discrepancy was due to a rare deletion mutation
who had previously been treated with erlotinib or that was not within the detection capacity of the
gefitinib, and 8 patients who had previously been SARMS assay.
treated with another chemotherapy agent or mul- Using the SARMS test, we were able to iden-
tiple regimens, including both tyrosine kinase in- tify rare EGFR mutant alleles below the detection
hibitors and chemotherapy. Four patients whose limit of standard sequencing. In addition to the
tumors had wild-type EGFR were also analyzed. known primary EGFR mutation, low levels of
(A schematic depiction of the strategy for micro- T790M were detected in pretreatment tumor sam-
fluidic isolation of circulating tumor cells and ples from 10 of 26 patients (38%). The relatively
representative images of captured cells are shown high number of amplification cycles that were
in Figure 1 in the Supplementary Appendix.) required to detect T790M suggests that the mu-
Circulating tumor cells were identified in all tation is present in only a small number of cells.
patients, with a median of 74 cells per milliliter Indeed, the sequencing of cloned polymerase-
(mean, 133; range, 5 to 771), with a similar num- chain-reaction (PCR) products from one of these
ber in patients with or without EGFR mutant tu- tumors identified only one T790M mutation in
mors (Table 1). The number of circulating tumor 500 EGFR alleles. The presence of the drug-resis-
cells that were isolated from this series of patients tance mutation at such a low frequency did not
with lung cancers enriched for EGFR mutations was preclude significant responses to therapy with ty-
similar to that from patients with other cancers.11 rosine kinase inhibitors among patients with EGFR
The tumor burden on matched radiographic mea- mutant tumors, but it was associated with a strik-
ing difference in progression-free survival, with a kinase inhibitors in patients with sensitizing EGFR
median of 7.7 months in patients with a detect- mutations.
able T790M allele, as compared with 16.5 months
in those without a detectable allele (hazard ratio Detection of EGFR Mutations in Circulating
for progression for the T790M allele, 11.5; 95% Tumor Cells
confidence interval, 2.94 to 45.1; P<0.001) (Fig. Having established the reliability of the SARMS
1). It seems likely that therapy with tyrosine ki- assay, we applied it to the analysis of circulating
nase inhibitors results in the selection of the pre- tumor cells isolated from peripheral blood. We
existing T790M resistance allele. Such selection first compared the EGFR mutations detected in
would be predicted to contribute to variation in circulating tumor cells by SARMS with those re-
the duration of response to therapy with tyrosine ported for the tumor specimen using either stan-
Table 2. Allele-Specific SARMS Analysis of EGFR Mutations in Tumor Samples and Best Clinical Response.*
Duration of Therapy
Patient No. and EGFR EGFR Mutation with Tyrosine Kinase Best Clinical
Mutation Status by Sequencing EGFR Mutation by SARMS Assay† Inhibitor‡ Response
Primary
Mutation T790M Other
mo
EGFR mutation present
2 Del T751_I759 insS None detected§ No 18.2¶ SD
3 G719S G719X Yes 3.9 PR
4 L858R L858R No >36.9 PR
5 Del E746_A750 Del Yes 8.6¶ PR
8 L861Q L861Q Yes 4.0 SD
9 Del E746_A750 Del Yes 6.2 PR
10 L858R L858R Yes 8.3 PR
11 Del E746_A750 Del No >14.0 PR
13 Del E746_A750 Del Yes 8.3 PR
14 Del L747_P753 insS Del No 13.5 PR
15 Del E746_S752 insV Del No 30.8 CR
22 Del E746_A750 Del No >9.2 PR
23 L858R L858R No >7.7 SD
28 Del E746_A750 Del No 14.9 PR
29 L858R L858R No 11.8 PR
30 L858R L858R No Del§ 3.6 PR
31 G719A G719X Yes Del§ 3.9 SD
32 Del E746_A750 Del No >16.7 SD
33 Del E746_A750 Del No 11.4 PR
34 Del E746_A750 Del No 17.5 CR
35 Del E746_A750 Del No 7.6 SD
36 L858R L858R No 9.5 PR
37 Del L747_T751 Del Yes 9.2 PR
38 L858R L858R Yes 6.9 PR
39 Del E746_A750 Del No 14.3 PR
40 Del E746_A750 insIP Del Yes 11.3 SD
Table 2. (Continued.)
Duration of Therapy
Patient No. and EGFR EGFR Mutation with Tyrosine Kinase Best Clinical
Mutation Status by Sequencing EGFR Mutation by SARMS Assay† Inhibitor‡ Response
Primary
Mutation T790M Other
mo
EGFR mutation absent
24 None (wild type) None detected No 0.7¶ PD
25 None (wild type) None detected No None NA
26 None (wild type) None detected No 1.2¶ PD
27 None (wild type) None detected No 7.5¶ SD
41 None (wild type) None detected No >4.4¶ SD
42 None (wild type) None detected No None NA
43 None (wild type) None detected No 0.7 PD
44 None (wild type) None detected No 2.6 SD
* The best clinical response was defined according to the Response Evaluation Criteria in Solid Tumors (RECIST). CR denotes complete re-
sponse, Del deletions in exon 19, NA not applicable, PD progressive disease, PR partial response, SARMS Scorpion Amplification Refractory
Mutation System, and SD stable disease.
† The SARMS assay groups all variant breakpoints of the in-frame (exon 19) EGFR deletion mutations as a single Del mutation and all muta-
tions at codon 719 as G719X. The presence or absence of T790M is indicated. Other mutations that were identified with the use of the
SARMS assay are listed only when they were present.
‡ When the number of months is preceded by “>,” therapy was ongoing.
§ The mutation identified by sequencing in Patient 2 was not within the detection capacity of the SARMS assay. In Patients 30 and 31, a sec-
ond activating EGFR allele was detected at low frequency (other), in addition to the prevalent activating mutation (primary mutation).
¶ In these patients, either gefitinib or erlotinib was administered as second-line or third-line therapy. In all other patients, the drugs were ad-
ministered as first-line therapy for advanced cancers.
dard sequencing or SARMS. Among specimens tumors (Table 3). The SARMS assay identified
from 20 patients that were available for molecu- EGFR mutations in 17 of 18 specimens of circulat-
lar analysis of circulating tumor cells, SARMS ing tumor cells (94%) and in 7 of 18 plasma sam-
identified EGFR mutations in 19 patients (95%) ples (39%). Among 12 patients for whom speci-
(Table 3). In addition to the primary activating mens of the primary tumor, circulating tumor
mutation, T790M was detected in circulating tu- cells, and plasma were all available for analysis,
mor cells from 2 of 6 patients (33%) who had a genotyping of circulating tumor cells had a sen-
response to tyrosine kinase inhibitors and from sitivity of 92% (in 11 of 12 patients), whereas
9 of 14 patients (64%) who had clinical progres- plasma genotyping had a sensitivity of 33% (4 of
sion (P = 0.34). This finding was consistent with 12 patients) (P = 0.009).
the reported prevalence of T790M (about 50%) in The sensitive SARMS assay also identified rare
patients with the EGFR mutation who were under- secondary EGFR activating mutations in a sub-
going repeat tumor biopsy after the development group of tumor samples, circulating tumor cells,
of resistance to tyrosine kinase inhibitors.18 and plasma (Tables 2 and 3 and Fig. 2A and 3,
A recent study reported the detection of EGFR and Table 1 in the Supplementary Appendix). To
mutations with the use of the SARMS assay in assess the significance of these mutations, we
free plasma DNA from patients with metastatic monitored genotypes of circulating tumor cells
non–small-cell lung cancer.24 Therefore, we stud- and quantity over time in a subgroup of patients.
ied the accuracy of mutational analysis in puri-
fied circulating tumor cells and free plasma DNA Serial Measurements
using blood samples from 18 patients with EGFR Detailed serial analyses of the quantity and geno-
mutant tumors and 3 controls with wild-type type of circulating tumor cells were available for
Table 3. Analysis of EGFR Mutations in Circulating Tumor Cells and Free Plasma DNA and the Concordance with Tumor Mutation.*
Concordance
Patient No., EGFR Mutation Status, with Tumor
and Response to Therapy SARMS Assay Results† Mutation‡
Circulating Tumor Cells Free Plasma
primary primary
mutation T790M other mutation T790M other
EGFR mutation present and response
during therapy§
1 Del No UN UN C¶‖
9 Del Yes None detected Yes C
11 Del No None detected No C
21 Del No Del Yes NA
22 Del No G719X None detected Yes C
23 Del Yes L858R Del Yes L858R C,P
EGFR mutation present and disease pro-
gression during therapy§
3 Del No G719X UN UN C¶
6 Del Yes Del No C,P
7 None detected Yes L858R Yes P
10 L858R Yes Del None detected No C
12 Del Yes None detected No C
13 Del Yes Del Yes C,P
14 Del No None detected No C
15 Del No None detected No C
16 Del Yes None detected No C
17 Del Yes Del Yes NA
18 Del No G719X None detected No NA
20 Del No None detected No NA**
45 Del Yes None detected No NA
46 Del Yes Del Yes NA
EGFR mutation absent
26 None detected No None detected No C,P
43 None detected No None detected No C,P
44 None detected No None detected No C,P
* Listed are patients in Group A, for whom genotypes were available from at least two of the following: circulating tumor cells, plasma, and
tumor-biopsy specimens. Molecular analyses of circulating tumor cells and plasma were performed within 6 months after the initial quan-
titation of circulating tumor cells. Del denotes deletions in exon 19, NA not applicable due to unavailable tumor specimen, SARMS Scorpion
Amplification Refractory Mutation System, and UN sample unavailable for analysis.
† The primary activating EGFR mutations are listed when present, including the presence or absence of the specific drug-resistance allele
T790M. In all cases, other activating mutations that are listed as “other” were present at a lower allele frequency than that of the primary
mutation.
‡ Listed is the concordance between the presence of the primary mutation in the tumor specimen and that in circulating tumor cells (C)
and free plasma DNA (P).
§ Patients who were receiving EGFR tyrosine kinase inhibitors were classified as having either a response or disease progression at the time
of the analysis of circulating tumor cells.
¶ A plasma specimen was not available for this patient.
‖ The mutation in the primary tumor was detected by high-performance liquid chromatography but was below the level of detection by stan-
dard sequencing analysis.
** The presence of a mutation of unknown significance (S885L) that was not included in the SARMS assay was reported in the primary tu-
mor. An additional tumor specimen was not available for SARMS analysis.
CTCs/ml
CTCs/ml
60 10 60 5
Del,
40 T790M 8 40 4
Patient 1 Patient 14
180 12 300 3.85
160 Del, Del, 3.80
280 [T790M]
Del T790M 10
140 260 3.75
120 Del, 8 Del, 3.70
L858R, 240
100 T790M 3.65
T790M 6 220
80 3.60
CTCs/ml
CTCs/ml
60 200 3.55
4 Del,
40 180 3.50
T790M
20 2 160 3.45
Del
0 0 0 0
0 50 100 150 200 250 300 150 200 250 300 350 400 450
Gefitinib Chemo Gefitinib
B
1590 Tumor Deletion 1200 CTC at Response 952 CTC at Progression Exon 2
1190 Exon 2
800 652
990 Deletion T790M
552
790 T790M 600 T790M 452
Luminescence
590 352
Downloaded from www.nejm.org on November 9, 2010. For personal use only. No other uses without permission.
400
390 252
190 200 152
1 3 5 7 9 1113151719212325272931333537394143454749 1 3 5 7 9 1113151719212325272931333537394143454749 1 3 5 7 9 1113151719212325272931333537394143454749
Cycle No. Cycle No. Cycle No.
375
ICM
FIGURE: 2 of 3 2nd
REG F
3rd
CASE Revised
EMail Line 4-C
The n e w e ng l a n d j o u r na l of m e dic i n e
19. Engelman JA, Zejnullahu K, Mitsudo- of the response to gefitinib in patients human EGFR kinase domain mutations
mi T, et al. MET amplification leads to with non-small-cell lung cancer. Clin Can- on lung tumorigenesis and in vivo sensi-
gefitinib resistance in lung cancer by acti- cer Res 2006;12:3915-21. tivity to EGFR-targeted therapies. Cancer
vating ERBB3 signaling. Science 2007;316: 25. Bell DW, Gore I, Okimoto RA, et al. Cell 2006;9:485-95.
1039-43. Inherited susceptibility to lung cancer may 31. Asahina H, Yamazaki K, Kinoshita I,
20. Bean J, Brennan C, Shih JY, et al. MET be associated with the T790M drug resis- et al. A phase II trial of gefitinib as first-
amplification occurs with or without tance mutation in EGFR. Nat Genet 2005; line therapy for advanced non-small cell
T790M mutations in EGFR mutant lung 37:1315-6. lung cancer with epidermal growth factor
tumors with acquired resistance to gefi- 26. Inukai M, Toyooka S, Ito S, et al. Pres- receptor mutations. Br J Cancer 2006;95:
tinib or erlotinib. Proc Natl Acad Sci U S A ence of epidermal growth factor receptor 998-1004.
2007;104:20932-7. gene T790M mutation as a minor clone in 32. Inoue A, Suzuki T, Fukuhara T, et al.
21. Sequist LV, Martins RG, Spigel D, et non-small cell lung cancer. Cancer Res Prospective phase II study of gefitinib for
al. First-line gefitinib in advanced non- 2006;66:7854-8. chemotherapy-naive patients with ad-
small-cell lung cancer patients harboring 27. Godin-Heymann N, Bryant I, Rivera vanced non-small-cell lung cancer with
somatic EGFR mutations. J Clin Oncol MN, et al. Oncogenic activity of epidermal epidermal growth factor receptor gene
2008;26:2442-9. growth factor receptor kinase mutant al- mutations. J Clin Oncol 2006;24:3340-6.
22. Therasse P, Arbuck SG, Eisenhauer leles is enhanced by the T790M drug re- 33. Bell DW, Lynch TJ, Haserlat SM, et al.
EA, et al. New guidelines to evaluate the sistance mutation. Cancer Res 2007;67: Epidermal growth factor receptor muta-
response to treatment in solid tumors: 7319-26. tions and gene amplification in non-
European Organization for Research and 28. Greulich H, Chen TH, Feng W, et al. small-cell lung cancer: molecular analysis
Treatment of Cancer, National Cancer In- Oncogenic transformation by inhibitor- of the IDEAL/INTACT gefitinib trials.
stitute of the United States, National Can- sensitive and -resistant EGFR mutants. J Clin Oncol 2005;23:8081-92.
cer Institute of Canada. J Natl Cancer Inst PLoS Med 2005;2(11):e313. 34. Tsao MS, Sakurada A, Cutz JC, et al.
2000;92:205-16. 29. Politi K, Zakowski MF, Fan PD, Schon- Erlotinib in lung cancer — molecular and
23. Whitcombe D, Theaker J, Guy SP, Brown feld EA, Pao W, Varmus HE. Lung adeno- clinical predictors of outcome. N Engl J
T, Little S. Detection of PCR products using carcinomas induced in mice by mutant Med 2005;353:133-44. [Erratum, N Engl J
self-probing amplicons and fluorescence. EGF receptors found in human lung can- Med 2005;355:1746.]
Nat Biotechnol 1999;17:804-7. cers respond to a tyrosine kinase inhibi- 35. Steeg PS. Tumor metastasis: mecha-
24. Kimura H, Kasahara K, Kawaishi M, tor or to down-regulation of the recep- nistic insights and clinical challenges.
et al. Detection of epidermal growth factor tors. Genes Dev 2006;20:1496-510. Nat Med 2006;12:895-904.
receptor mutations in serum as a predictor 30. Ji H, Li D, Chen L, et al. The impact of Copyright © 2008 Massachusetts Medical Society.