- Email: [email protected]
Prognostic impacts of EGFR mutation status and subtype in patients with surgically resected lung adenocarcinoma
Accepted Manuscript The prognostic impact of EGFR-mutation status and subtypes in patients with surgically resected lung adenocarcinomas Kazuya Takamochi, MD, Shiaki Oh, MD, Takeshi Matsunaga, MD, Kenji Suzuki, MD PII:
S0022-5223(17)31574-X
DOI:
10.1016/j.jtcvs.2017.06.062
Reference:
YMTC 11767
To appear in:
The Journal of Thoracic and Cardiovascular Surgery
Received Date: 26 January 2017 Revised Date:
7 June 2017
Accepted Date: 30 June 2017
Please cite this article as: Takamochi K, Oh S, Matsunaga T, Suzuki K, The prognostic impact of EGFRmutation status and subtypes in patients with surgically resected lung adenocarcinomas, The Journal of Thoracic and Cardiovascular Surgery (2017), doi: 10.1016/j.jtcvs.2017.06.062. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Takamochi et al. 1.
ACCEPTED MANUSCRIPT
1
Title: The prognostic impact of EGFR-mutation status and subtypes in patients with
2
surgically resected lung adenocarcinomas
4
RI PT
3
Kazuya Takamochi, MD, Shiaki Oh, MD, Takeshi Matsunaga, MD, Kenji Suzuki, MD
SC
5
Department of General Thoracic Surgery, Juntendo University School of
7
Medicine, 1-3, Hongo 3-chome, Bunkyo-ku, Tokyo 113-8431, Japan
M AN U
6
8
10
Conflicts of interest None declared.
11
TE D
9
Funding statement
13
This work was supported in part by a Grant-in-Aid for Cancer Research from the
14
Ministry of Health, Labor, and Welfare of Japan and the Smoking Research Foundation.
AC C
15
EP
12
16
Corresponding author
17
Kazuya Takamochi, MD
18
Department of General Thoracic Surgery, Juntendo University School of Medicine
Takamochi et al. 2.
ACCEPTED MANUSCRIPT
1-3, Hongo 3-chome, Bunkyo-ku, Tokyo 113-8431, Japan.
20
Tel: +81-3-3813-3111
21
Fax: +81-3-5800-0281
22
E-mail: [email protected]
RI PT
19
Date and number of IRB approval
25
March 19, 2015; No. 2014172
26
Article word count
28
2702 words
29
TE D
27
M AN U
24
SC
23
Abbreviations and Acronyms
31
CEA
32
CT
33
EGFR – epidermal growth factor receptor
34
IASLC – International Association for the Study of Lung Cancer
35
IRB
– institutional review board
36
KM
– Kaplan-Meier
EP
30
AC C
– carcinoembryonic antigen – computed tomography
Takamochi et al. 3.
ACCEPTED MANUSCRIPT
MRI
– magnetic resonance imaging
38
NSCLC – non-small-cell lung cancer
39
OS
– overall survival
40
PET
– 18F-fluoro-2-deoxy-glucose positron emission tomography
41
RFS
– recurrence-free survival
42
TKIs
– tyrosine kinase inhibitors
43
WT
– wild type
M AN U
SC
RI PT
37
44
Central Picture
46
Figure 2A
47
Improved survival associated with EGFR-mutant lung tumors relative to -wild-type
48
tumors.
EP AC C
49
TE D
45
50
Central Message
51
Positive EGFR-mutation status is a favorable prognostic factor in patients with
52
surgically resected lung adenocarcinomas; however, EGFR-mutation subtypes have no
53
prognostic impact.
54
Takamochi et al. 4.
ACCEPTED MANUSCRIPT
Perspective Statement
56
We showed that positive EGFR-mutation status is a favorable prognostic factor
57
independent of TNM stage in patients with surgically resected lung adenocarcinomas.
58
During the next revision of TNM classification of lung cancer, cancer genotype might
59
be adapted as a new category, and in future clinical trials for adjuvant therapy,
60
stratification by EGFR-mutation status might be necessary to avoid bias.
M AN U
SC
RI PT
55
AC C
EP
TE D
61
Takamochi et al. 5.
ACCEPTED MANUSCRIPT
Abstract
63
Objective: Epidermal growth factor receptor (EGFR) gene-mutation status is a
64
well-established predictor of the efficacy of EGFR tyrosine-kinase inhibitors (TKIs) in
65
non-small-cell lung cancer. Recently, differences in EGFR-mutation subtypes were also
66
reported to be associated with the efficacy of EGFR TKIs. However, the prognostic
67
impact of EGFR-mutation status and subtypes remains controversial.
68
Methods: We retrospectively reviewed 939 patients with surgically resected
69
adenocarcinomas who had their EGFR-mutation status analyzed between January 2010
70
and December 2014. Overall survival (OS) and recurrence-free survival (RFS) were
71
compared using the log-rank test according to pathological stage, EGFR-mutation status,
72
and EGFR-mutation subtypes. Independent prognostic factors for OS and RFS were
73
identified by multivariate analyses using the Cox proportional-hazards model.
74
Results: A median follow-up time was 48 months. We found that positive
75
EGFR-mutation status was significantly associated with longer OS and RFS in all
76
patients and associated with longer OS in pathological stage I patients. However, no
77
significant differences were observed in OS and RFS between patients with exon-21
78
L858R mutation and those with exon-19 deletions. In a Cox regression model for OS,
79
EGFR-mutation status was a significant prognostic factor that was independent of
AC C
EP
TE D
M AN U
SC
RI PT
62
Takamochi et al. 6.
ACCEPTED MANUSCRIPT
well-established prognostic factors, such as age, pathological stage, vascular invasion,
81
lymphatic permeation, and serum carcinoembryonic antigen level.
82
Conclusions: Positive EGFR-mutation status is a favorable prognostic factor in patients
83
with surgically resected lung adenocarcinomas; however, EGFR-mutation subtypes
84
(exon-21 L858R mutation or exon-19 deletions) exhibited no prognostic impact.
SC
RI PT
80
AC C
EP
TE D
M AN U
85
Takamochi et al. 7.
ACCEPTED MANUSCRIPT
Mutations in the epidermal growth factor receptor gene (EGFR) represent one of the
87
most frequent driver-oncogene alterations in non-small-cell lung cancer (NSCLC).
88
EGFR mutations are more frequently found in females, those who have never smoked,
89
adenocarcinoma patients, and members of East Asian populations.1 The frequency of
90
EGFR mutations in non-adenocarcinoma NSCLC is quite low (<2%).1 The EGFR
91
tyrosine kinase inhibitors (TKIs), gefitinib and erlotinib, which specifically block the
92
EGFR-dependent pathway, were initially approved among molecular-target drugs for
93
their use in the treatment of NSCLC.2 EGFR-mutated NSCLC has become a clinically
94
relevant molecular subset, because patients with NSCLC harboring activating EGFR
95
mutations (exon-21 L858R mutation or exon-19 deletions) benefit from EGFR-TKI
96
treatment unlike those with EGFR-wild-type (WT) tumors.3, 4
SC
M AN U
TE D
Both the presence of EGFR mutations and differences in EGFR-mutation
EP
97
RI PT
86
subtypes (exon-21 L858R mutation or exon 19 deletions) were reportedly associated
99
with the efficacy of EGFR TKIs. A meta-analysis of eight clinical trials and five
AC C
98
100
retrospective studies using EGFR TKIs for advanced EGFR-mutated NSCLCs
101
demonstrated that exon-19 deletion was associated with longer progression-free survival
102
as compared with exon-21 L858R mutation.5 Therefore, both EGFR-mutation status and
103
EGFR-mutation subtypes are considered significant predictors of the efficacy of EGFR
Takamochi et al. 8.
ACCEPTED MANUSCRIPT
104
TKIs in NSCLC. By contrast, the prognostic impact of both EGFR-mutation status and subtypes
105
in NSCLC patients remains controversial. A systematic review with meta-analysis based
107
on data from 16 studies (n = 3337) showed that EGFR mutations were not a prognostic
108
factor in patients with surgically resected NSCLC.6 However, all of the studies included
109
in the meta-analysis were retrospectively conducted, the number of patients was small,
110
the follow-up period was relatively short, and the treatments for recurrent disease
111
(especially regarding the usage of EGFR-TKIs) were heterogeneous. Therefore, as a
112
matter of course, further study is needed to confirm the results. The prognostic impact
113
of EGFR-mutation subtypes (exon-21 L858R point mutation or exon-19 deletions) in
114
NSCLC patients has not been well established, because the studies investigating the
115
prognostic impact of EGFR-mutation subtypes in early stage NSCLC patients who
116
underwent surgical resection were scarcely reported.7-9 In this study, we elucidated the
117
prognostic impact of EGFR-mutation status and subtypes on surgically resected lung
118
adenocarcinoma patients.
AC C
EP
TE D
M AN U
SC
RI PT
106
119
120
MATERIALS AND METHODS
121
Patients
Takamochi et al. 9.
ACCEPTED MANUSCRIPT
Between January 2010 and December 2014, 1018 lung adenocarcinoma
123
patients underwent complete surgical resection at Juntendo University School of
124
Medicine (Tokyo, Japan). Among these patients, we retrospectively reviewed
125
clinicopathological data for 939 consecutive patients with lung adenocarcinomas who
126
had their EGFR-mutation status analyzed. A flowchart illustrating the selection of study
127
cohorts is shown in Figure 1.
SC M AN U
128
RI PT
122
Preoperative staging was routinely performed with computed tomography (CT) scanning of the chest and upper abdomen and 18F-fluoro-2-deoxy-glucose positron
130
emission tomography (PET)/CT scans. Cranial CT scanning or magnetic resonance
131
imaging (MRI) was performed according to clinical signs and symptoms. The patients
132
were staged based on the 7th edition of the International Association for the Study of
133
Lung Cancer (IASLC) TNM classification for lung cancer.10 Induction chemotherapy
134
and/or radiotherapy and treatments with EGFR TKIs as induction or adjuvant therapy
135
were not performed on any of the patients in this series.
136
Follow-up and Diagnosis of Recurrence
137
Follow-up examinations were performed every 3 or 4 months for the first 3 years and
138
every 6 months thereafter. Chest X-rays and biochemical examination of blood,
139
including for carcinoembryonic antigen (CEA), were performed at every visit to the
AC C
EP
TE D
129
Takamochi et al. 10.
ACCEPTED MANUSCRIPT
outpatient department. Chest CT scans or PET/CT scans were performed annually.
141
Cranial CT scanning or MRI was performed as necessary. Recurrences were diagnosed
142
based on radiological evidence of cancer relapse upon surveillance imaging and/or
143
pathologic evidence according to a tumor biopsy.
144
EGFR-mutation Analysis
145
We routinely analyzed EGFR mutations for lung cancer patients who underwent surgery
146
for a research purpose. Genomic DNA was extracted from cubes (3–5 mm3) of frozen,
147
fresh lung-cancer tissue samples from surgically resected specimens. The peptide
148
nucleic acid-locked nucleic acid polymerase chain reaction clamp method11 was used to
149
identify EGFR mutations G719A, G719C, and G719A in exon 18, all deletion
150
genotypes in exon 19, T790M in exon 20, and L858R and L861Q in exon 21.
SC
M AN U
TE D
Approval for collection of patient specimens from the tissue bank in our
EP
151
RI PT
140
department was obtained from the institutional review board (IRB) of Juntendo
153
University School of Medicine. Individual patient consent was obtained for
154
procurement of tissue for research purposes prior to surgery. This retrospective study
155
was performed under a waiver of authorization approved by the IRB.
156
Statistical Analysis
AC C
152
Takamochi et al. 11.
ACCEPTED MANUSCRIPT
Overall survival (OS) and recurrence-free survival (RFS) were analyzed by
158
Kaplan-Meier (KM) methods, and curve differences were tested using the log-rank test
159
according to pathological stage, EGFR-mutation status, and EGFR-mutation subtypes.
160
The length of OS was defined as the interval in months between the day of surgical
161
intervention and the date of either death or the last follow-up. The length of RFS was
162
defined as the interval in months between the day of surgical intervention and the date
163
of either recurrence, death from any cause, or the last follow-up. Univariate and
164
multivariate analyses of independent prognostic factors in each cohort were performed
165
using the Cox proportional-hazards model. All significant variables (P < 0.05) in the
166
univariate analyses were entered into the multivariate analyses using a stepwise
167
variable-selection procedure to adjust for potential confounding factors.
168
A P < 0.05 was considered statistically significant. All statistical analyses were
169
performed using the SPSS statistical software package (v20.0; SPSS Inc., Chicago, IL,
170
USA).
SC
M AN U
TE D
EP
AC C
171
RI PT
157
172
RESULTS
173
EGFR mutations were detected in 418 (45%) of the 939 lung adenocarcinoma patients,
174
with the point mutation L858R in exon 21 and exon-19 deletions detected in 222 (53%)
Takamochi et al. 12.
ACCEPTED MANUSCRIPT
and 161 (39%) of 418 EGFR-mutated tumors, respectively. The remaining minor EGFR
176
mutations, including exon-18 G719A in eight, exon-18 G719S in five, exon-18 G719C
177
in three, exon-21 L861Q in seven, and exon-20 T790M in four, were detected in 23
178
(6%) tumors. Double mutations were found in 12 (3%) tumors and included exon- 21
179
L858R and exon 20-T790M in eight, exon-20 T790M and exon-19 deletion in one,
180
exon-20 T790M and exon-18 G719A in one, exon-18 G719A and exon-21 L861Q in
181
one, and exon-18 G719S and exon-21 L861Q in one. Interestingly, exon-20 T790M,
182
which confers resistance to EGFR TKIs,12 was consistently coupled with other
183
mutations. Patient characteristics are summarized in Table 1. Female (P < 0.001),
184
non-smoker (P < 0.001), pathological stage I tumor (P = 0.001), and a tumor without
185
vascular invasion (P = 0.002) were more frequently observed characteristics in patients
186
with EGFR-mutated tumors than in those with EGFR-WT tumors.
SC
M AN U
TE D
EP
To evaluate the prognostic impact of EGFR-mutation status, we analyzed OS
AC C
187
RI PT
175
188
and RFS in 939 overall patients with a median follow-up time of 48 months. The
189
3-/5-year OS rates were 95% [95% confidence interval (CI): 92–97%] /89% (95% CI:
190
84–92%) in 418 patients with EGFR-mutated tumors and 85% (95% CI: 82–88%)/78%
191
(95% CI: 73–82%) in 521 patients with EGFR-WT tumors, respectively. Positive
192
EGFR-mutation status was significantly associated with longer OS in all patients (Fig.
Takamochi et al. 13.
ACCEPTED MANUSCRIPT
2A; P < 0.001). Additionally, the 3-/5-year RFS rates were 85% (95%CI: 81–88%)/81%
194
(95% CI: 76–85%) in 418 patients with EGFR-mutated tumors and 78% (95% CI: 75–
195
82%)/72% (95% CI: 68–77%) in 521 patients with EGFR-WT tumors, respectively,
196
with positive EGFR-mutation status was significantly associated with longer RFS in all
197
patients (Fig. 2B; P = 0.002).
SC
RI PT
193
The 3-/5-year OS rates were 98% (95% CI: 96–99%)/94% (95% CI: 90–97%)
199
in 350 patients with EGFR-mutated pathological stage I tumors and 93% (95% CI: 90–
200
95%)/87% (95% CI: 83–91%) in 391 patients with EGFR-WT pathological stage I
201
tumors, respectively, with positive EGFR-mutation status significantly associated with
202
longer OS in pathological stage I patients (Fig. E1A; P = 0.003). Additionally, the
203
3-/5-year RFS rates were 93% (95% CI: 90–96%)/89% (95% CI: 84–93%) in 350
204
patients with EGFR-mutated pathological stage I tumors and 90% (95% CI: 86–
205
93%)/84% (95% CI: 80–88%) in 391 patients with EGFR-WT pathological stage I
206
tumors, respectively, with positive EGFR-mutation status associated with longer RFS in
207
pathological stage I patients; however, the difference was not statistically significant
208
(Fig. E1B; P = 0.065). Cox regression analysis of prognostic factors for OS in all
209
patients showed that EGFR-mutation status was a significant prognostic factor
210
independent of well-established prognostic factors, such as age, serum CEA level,
AC C
EP
TE D
M AN U
198
Takamochi et al. 14.
ACCEPTED MANUSCRIPT
pathological stage, vascular invasion, and lymphatic permeation (Tables 2 and 3). Cox
212
regression analysis of prognostic factors for RFS in all patients showed that
213
EGFR-mutation status was a significant prognostic factor in the univariate analysis;
214
however, this was not significant in the multivariate analysis (Tables 2 and 3).
To evaluate the prognostic impact of EGFR-mutation subtypes, we analyzed
SC
215
RI PT
211
OS and RFS rates for patients with EGFR-mutated tumors in three groups stratified by
217
EGFR-mutation subtypes: exon-19 deletions, L858R, and other subtypes, including
218
minor mutations or double mutations. In 418 patients with EGFR-mutated tumors, no
219
significant differences were observed in OS and RFS rates according to mutation
220
subtype (Fig. 3A: exon-19 deletions vs. L858R, P = 0.172; exon-19 deletions vs. other
221
subtypes, P = 0.056; L858R vs. other subtypes, P = 0.300; Fig. 3B: exon-19 deletions vs.
222
L858R, P = 0.646; exon-19 deletions vs. other subtypes, P = 0.961; L858R vs. other
223
subtypes, P = 0.696).
TE D
EP
AC C
224
M AN U
216
Postoperative recurrences developed in 57 patients with EGFR-mutated
225
tumors: 10 locoregional only, 31 distant only, and 16 locoregional and distant. Among
226
these, 41 patients were treated with EGFR TKIs for any lines following recurrence, and
227
OS time tended to be longer in these patients as compared with that observed in 16
228
patients not treated with EGFR TKIs; however, the difference was not statistically
Takamochi et al. 15.
ACCEPTED MANUSCRIPT
significant (P = 0.173; Fig. 4). In the 41 patients who received EGFR TKI treatment,
230
both OS and survival periods following recurrence were longer in 19 patients with
231
tumors harboring exon-19-deletion mutations as compared with those in 18 patients
232
with tumors harboring the L858R mutation; however, the differences were not
233
statistically significant (P = 0.256 and P = 0.198, respectively; Fig. E2).
SC
RI PT
229
M AN U
234
DISCUSSION
236
EGFR-mutation status and subtypes indisputably constitute significant predictive
237
factors for the efficacy of EGFR TKIs in advanced NSCLC. However, their prognostic
238
role remains unclear, because only inconsistent results from studies of small and
239
heterogeneous populations are available.6-8 Elucidation of the intrinsic prognostic
240
impact of EGFR-mutation status and EGFR-mutation subtypes in advanced
241
EGFR-mutated NSCLC patients is currently considered impossible, because EGFR
242
TKIs are used as a standard first-line treatment. Furthermore, third-generation EGFR
243
TKIs were recently shown to prolong survival, even after the development of resistance
244
to first- or second-generation EGFR TKIs.13 The prognosis for advanced EGFR-mutated
245
NSCLC patients is expected to become increasingly prolonged. Even in the presence or
246
absence of intrinsic prognostic impacts associated with EGFR-mutation status and
AC C
EP
TE D
235
Takamochi et al. 16.
ACCEPTED MANUSCRIPT
subtypes, their affect would be obscured by a definitive association with the efficacy of
248
EGFR TKIs. Therefore, the intrinsic prognostic role of EGFR-mutation status and
249
subtypes needs to be investigated in surgically resected early stage NSCLC patients who
250
had not received perioperative therapy using EGFR TKIs. To the best of our knowledge,
251
this study constitutes the largest-scale evaluation of patients with surgically resected
252
EGFR-mutated tumors for evaluating the prognostic impact of EGFR-mutation status
253
and subtypes by analyzing both OS and RFS rates.
SC
M AN U
254
RI PT
247
Here, we showed that positive EGFR-mutation status was significantly associated with longer OS time regardless of pathological stage in KM analyses.
256
Furthermore, multivariate analysis revealed that EGFR-mutation status was a significant
257
prognostic factor for OS time independent of pathological stage. Additionally, although
258
the number of patients who received EGFR TKIs for postoperative recurrence in the
259
present study was small, there was no statistical difference in OS regardless of whether
260
patients with recurrent EGFR-mutated tumors received EGFR TKI treatment. OS time
261
might not be significantly influenced by EGFR TKI treatment for postoperative
262
recurrences. Therefore, we concluded that positive EGFR-mutation status is a favorable
263
prognostic factor in patients with surgically resected lung adenocarcinomas.
264
AC C
EP
TE D
255
By contrast, no significant differences were observed in OS and RFS rates
Takamochi et al. 17.
ACCEPTED MANUSCRIPT
between patients with an exon-21 L858R mutation (n = 222) and those with exon-19
266
deletions in EGFR-mutated tumors (n = 161). In 41 patients who received EGFR TKI
267
treatment, both OS and survival periods following recurrence were longer in patients
268
with tumors harboring exon-19 deletions (n = 19) as compared with that in those with
269
tumors harboring an L858R mutation (n = 18), although the differences were not
270
statistically significant. Therefore, we concluded that EGFR-mutation subtypes
271
(exon-21 L858R mutation or exon-19 deletions) have no prognostic impact on patients
272
with surgically resected lung adenocarcinomas. However, further large-scale studies are
273
needed to confirm these observations.
SC
M AN U
Although the underlying mechanism associated with the favorable prognostic
TE D
274
RI PT
265
impact of EGFR mutations in lung adenocarcinoma remains unclear, EGFR-mutated
276
lung adenocarcinoma may be less likely to be affected by complex oncogenic processes,
277
because EGFR mutations provide strong growth advantages due to their constitutive
278
activation. Therefore, EGFR-mutated lung adenocarcinoma cells need not acquire
279
additional oncogenic mutations to enhance their biological malignancy via cell-based
280
characteristics, such as invasive, anti-apoptotic, and metastatic potentials. As a result,
281
EGFR-mutated lung adenocarcinoma (non-smoking- related lung cancer) may acquire a
282
lower degree of biological malignancy relative to that observed in smoking-related lung
AC C
EP
275
Takamochi et al. 18.
ACCEPTED MANUSCRIPT
283
284
cancers exhibiting high nonsynonymous-mutation burden.14 Recently, TNM classifications for lung cancer were newly revised by the IASLC-staging committee.15 Although the revised TNM-staging system adapted
286
categories based on prognostic factors associated with the anatomical extent of tumors,
287
no regard was given to the biological features of tumors, such as driver mutations.
288
Recent advances in the understanding of NSCLC-related molecular biology would
289
likely offer the potential for therapeutic success by matching systemic target therapies to
290
cancer genotype. EGFR mutations offer not only predictive significance for EGFR TKI
291
treatment, but also improved prognostic significance from a biological standpoint for
292
lung adenocarcinoma. Other alterations in driver oncogenes that predict prognostic
293
impact similar to EGFR mutations might be identified in the future. Therefore,
294
regardless of how precise the classification system becomes, a TNM-staging system
295
based solely on the anatomical extent of tumors will become outdated and almost
296
irrelevant to clinical practice in the era of molecular-target therapy. During the next
297
revision to the TNM classification of lung cancer, cancer genotype might be adapted as
298
a new category.
299
300
AC C
EP
TE D
M AN U
SC
RI PT
285
In several clinical trials,16, 17 higher response rates were reported for chemotherapy in both first-line and second-line advanced-NSCLC settings in the
Takamochi et al. 19.
ACCEPTED MANUSCRIPT
presence of EGFR mutations. In the phase III studies of adjuvant gefitinib (BR19) and
302
erlotinib (RADIANT) for patients with completely resected stage IB to IIIA NSCLC,
303
EGFR TKIs failed to show survival benefits as compared with placebo in subgroups of
304
patients with EGFR-mutated tumors.18, 19 However, D'Angelo et al20 reported that
305
adjuvant therapy with gefitinib or erlotinib was significantly associated with a lower
306
risk of recurrence or death, disease-free survival, and a trend toward improved OS in
307
their retrospective study. In a randomized phase II study, adjuvant therapy involving
308
gefitinib following pemetrexed and carboplatin showed significant improvement in
309
disease-free survival in patients with stage IIIA NSCLC harboring EGFR mutations as
310
compared with pemetrexed and carboplatin treatment alone.21 Furthermore, in
311
opposition to our results, positive EGFR-mutation status was recently reported as a
312
favorable independent prognostic factor for survival periods following recurrence.22
313
Therefore, stratification by EGFR-mutation status might be necessary to avoid an
314
uneven distribution of EGFR-mutant populations that could potentially confound
315
survival data in future clinical trials related to adjuvant therapy.
AC C
EP
TE D
M AN U
SC
RI PT
301
316
Our study must be interpreted in the context of the following limitations. First,
317
although the number of the EGFR-mutant cohort was the largest evaluated to date, this
318
study represents a retrospective comparative study at the single institution. Therefore,
Takamochi et al. 20.
ACCEPTED MANUSCRIPT
patient-selection bias is inevitable. Second, the intrinsic prognostic role of
320
EGFR-mutation (EGFR mutation as a prognostic factor) might not be accurately
321
evaluated, because we could not completely exclude the efficacy of EGFR TKI
322
treatment (EGFR mutation as a predictive factor). Namely, our study did not include
323
any patients who received EGFR TKI treatment in the induction or adjuvant setting;
324
however, 41 (4%) of 939 patients received EGFR TKI treatment for postoperative
325
recurrences. Third, although EGFR TKI treatment for patients with recurrent
326
EGFR-mutated tumors did not show statistical significance for OS, the number of
327
patients who received EGFR TKI treatment for postoperative recurrence in the present
328
study might be too small to draw a definitive conclusion. Therefore, further large-scale
329
studies are warranted in this population.
TE D
M AN U
SC
RI PT
319
In conclusion, our analyses indicated that positive EGFR-mutation status is a
331
favorable prognostic factor in patients with surgically resected lung adenocarcinomas.
332
Additionally, our results showed that EGFR-mutation subtypes (L858R mutation or
333
exon-19 deletions) offered no prognostic impact, although they might be significant
334
predictors of EGFR TKI efficacy.
335
AC C
EP
330
Takamochi et al. 21.
ACCEPTED MANUSCRIPT
336
References
337
1.
338
EGFR mutations in lung cancer. Int J Clin Oncol. 2006;11:190-8.
339
2.
340
treating cancer. Nat Rev Cancer. 2004;4:956-65.
341
3.
342
BW et al. Activating mutations in the epidermal growth factor receptor underlying
343
responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med.
344
2004;350:2129-39.
345
4.
346
mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science.
347
2004;304:1497-500.
348
5.
349
deletion were associated with longer progression-free survival compared to those with
350
L858R mutation after first-line EGFR-TKIs for advanced non-small cell lung cancer: a
351
meta-analysis. PLoS One. 2014;9:e107161. doi: 10.1371/journal.pone.0107161.
352
eCollection 2014.
353
6.
RI PT
Mitsudomi T, Kosaka T, Yatabe Y. Biological and clinical implications of
SC
Herbst RS, Fukuoka M, Baselga J. Gefitinib--a novel targeted approach to
M AN U
Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan
EP
TE D
Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S et al. EGFR
AC C
Zhang Y, Sheng J, Kang S, Fang W, Yan Y, Hu Z et al. Patients with exon 19
Zhang Z, Wang T, Zhang J, Cai X, Pan C, Long Y et al. Prognostic value of
Takamochi et al. 22.
ACCEPTED MANUSCRIPT
epidermal growth factor receptor mutations in resected non-small cell lung cancer: a
355
systematic review with meta-analysis. PLoS One. 2014;9:e106053. doi:
356
10.1371/journal.pone.0106053. eCollection 2014.
357
7.
358
mutations within the EGFR kinase domain: a molecular predictor of disease-free
359
survival in resected pulmonary adenocarcinoma. J Cancer Res Clin Oncol.
360
2009;135:1647-54. doi: 10.007/s00432-009-0611-7. Epub 2009 Jun 11.
361
8.
362
epidermal growth factor receptor mutations in resected lung adenocarcinomas. Med
363
Oncol. 2014;31:771. doi: 10.1007/s12032-013-0771-9. Epub 2013 Nov 19.
364
9.
365
Prognostic Impact of EGFR Driver Mutations on Postoperative Disease Recurrence in
366
Lung Adenocarcinoma. Anticancer Res. 2016;36:3057-63.
367
10.
368
al. The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM
369
stage groupings in the forthcoming (seventh) edition of the TNM Classification of
370
malignant tumours. J Thorac Oncol. 2007;2:706-14.
371
11.
RI PT
354
M AN U
SC
Lee YJ, Park IK, Park MS, Choi HJ, Cho BC, Chung KY et al. Activating
TE D
Liu WS, Zhao LJ, Pang QS, Yuan ZY, Li B, Wang P. Prognostic value of
AC C
EP
Okamoto T, Kitahara H, Shimamatsu S, Katsura M, Takada K, Fujishita T et al.
Goldstraw P, Crowley J, Chansky K, Giroux DJ, Groome PA, Rami-Porta R et
Nagai Y, Miyazawa H, Huqun, Tanaka T, Udagawa K, Kato M et al. Genetic
Takamochi et al. 23.
ACCEPTED MANUSCRIPT
heterogeneity of the epidermal growth factor receptor in non-small cell lung cancer cell
373
lines revealed by a rapid and sensitive detection system, the peptide nucleic acid-locked
374
nucleic acid PCR clamp. Cancer Res. 2005;65:7276-82.
375
12.
376
EGFR mutation and resistance of non-small-cell lung cancer to gefitinib. N Engl J Med.
377
2005;352:786-92.
378
13.
379
growth factor receptor tyrosine kinase inhibitors in the treatment of lung cancer. Cancer.
380
2015;121:E1-6. doi: 10.1002/cncr.29139. Epub 2014 Dec 17.
381
14.
382
Genomic landscape of non-small cell lung cancer in smokers and never-smokers. Cell.
383
2012;150:1121-34. doi: 10.016/j.cell.2012.08.024.
384
15.
385
et al. The IASLC Lung Cancer Staging Project: Proposals for Revision of the TNM
386
Stage Groupings in the Forthcoming (Eighth) Edition of the TNM Classification for
387
Lung Cancer. J Thorac Oncol. 2016;11:39-51. doi: 10.1016/j.jtho.2015.09.009.
388
16.
389
Thatcher N et al. Molecular predictors of outcome with gefitinib in a phase III
RI PT
372
M AN U
SC
Kobayashi S, Boggon TJ, Dayaram T, Janne PA, Kocher O, Meyerson M et al.
Steuer CE, Khuri FR, Ramalingam SS. The next generation of epidermal
EP
TE D
Govindan R, Ding L, Griffith M, Subramanian J, Dees ND, Kanchi KL et al.
AC C
Goldstraw P, Chansky K, Crowley J, Rami-Porta R, Asamura H, Eberhardt WE
Hirsch FR, Varella-Garcia M, Bunn PA, Jr., Franklin WA, Dziadziuszko R,
Takamochi et al. 24.
ACCEPTED MANUSCRIPT
390
placebo-controlled study in advanced non-small-cell lung cancer. J Clin Oncol.
391
2006;24:5034-42.
392
17.
393
carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med. 2009;361:947-57.
394
doi: 10.1056/NEJMoa0810699. Epub 2009 Aug 19.
395
18.
396
Gefitinib versus placebo in completely resected non-small-cell lung cancer: results of
397
the NCIC CTG BR19 study. J Clin Oncol. 2013;31:3320-6.
398
19.
399
Adjuvant Erlotinib Versus Placebo in Patients With Stage IB-IIIA Non-Small-Cell Lung
400
Cancer (RADIANT): A Randomized, Double-Blind, Phase III Trial. J Clin Oncol.
401
2015;33:4007-14. doi: 10.1200/JCO.2015.61.8918. Epub 2015 Aug 31.
402
20.
403
Distinct clinical course of EGFR-mutant resected lung cancers: results of testing of 1118
404
surgical specimens and effects of adjuvant gefitinib and erlotinib. J Thorac Oncol.
405
2012;7:1815-22. doi: 10.097/JTO.0b013e31826bb7b2.
406
21.
407
adjuvant chemotherapy with or without gefitinib in resected stage IIIA-N2 non-small
SC
RI PT
Mok TS, Wu YL, Thongprasert S, Yang CH, Chu DT, Saijo N et al. Gefitinib or
M AN U
Goss GD, O'Callaghan C, Lorimer I, Tsao MS, Masters GA, Jett J et al.
EP
TE D
Kelly K, Altorki NK, Eberhardt WE, O'Brien ME, Spigel DR, Crino L et al.
AC C
D'Angelo SP, Janjigian YY, Ahye N, Riely GJ, Chaft JE, Sima CS et al.
Li N, Ou W, Ye X, Sun HB, Zhang L, Fang Q et al. Pemetrexed-carboplatin
Takamochi et al. 25.
ACCEPTED MANUSCRIPT
408
cell lung cancer harbouring EGFR mutations: a randomized, phase II study. Ann Surg
409
Oncol. 2014;21:2091-6. doi: 10.1245/s10434-014-3586-9. Epub 2014 Mar 1.
410
22.
411
Prognostic Factors for Survival After Recurrence in Patients With Completely Resected
412
Lung Adenocarcinoma: Important Roles of Epidermal Growth Factor Receptor
413
Mutation Status and the Current Staging System. Clin Lung Cancer. 2015;16:e213-21.
414
doi: 10.1016/j.cllc.2015.04.005. Epub
M AN U
SC
RI PT
Kudo Y, Shimada Y, Saji H, Kato Y, Yoshida K, Matsubayashi J et al.
AC C
EP
TE D
415
Apr 20.
Takamochi et al. 26.
ACCEPTED MANUSCRIPT
Figure Legends
417
FIGURE 1. Flowchart describing the selection of the study cohort.
418
EGFR, epidermal growth factor receptor; TKI, tyrosine-kinase inhibitor; WT, wild type.
RI PT
416
SC
419
FIGURE 2. OS (A) and RFS (B) in all patients according to EGFR-mutation status.
421
EGFR, epidermal growth factor receptor; OS, overall survival; RFS, recurrence-free
422
survival; WT, wild type.
TE D
M AN U
420
423
FIGURE 3. OS (A) and RFS (B) in patients with EGFR-mutated tumors according to
425
EGFR-mutation subtype (exon-19 deletions, exon-21 L858R mutation, or other
426
mutations).
427
EGFR, epidermal growth factor receptor; OS, overall survival; RFS, recurrence-free
428
survival.
429
AC C
EP
424
Takamochi et al. 27.
ACCEPTED MANUSCRIPT
FIGURE 4. A comparison of OS time between patients with EGFR-mutated tumors
431
who received EGFR-TKI treatment and those who did not following recurrence.
432
EGFR, epidermal growth factor receptor; OS, overall survival; TKI, tyrosine-kinase
433
inhibitor.
SC
RI PT
430
M AN U
434
FIGURE E1. OS (A) and RFS (B) in patients with pathological stage I lung
436
adenocarcinomas according to EGFR-mutation status.
437
EGFR, epidermal growth factor receptor; OS, overall survival; RFS, recurrence-free
438
survival; WT, wild type.
TE D
435
EP
439
FIGURE E2. OS (A) and survival periods following recurrence (B) in patients who
441
received EGFR-TKI treatment according to EGFR-mutation subtype (exon-19 deletions,
442
exon-21 L858R mutation, or other mutations).
443
EGFR, epidermal growth factor receptor; OS, overall survival; TKI, tyrosine-kinase
444
inhibitor.
AC C
440
Takamochi et al. 28.
ACCEPTED MANUSCRIPT
Video Legend
446
The importance and relevance of the present study are summarized on a video.
RI PT
445
AC C
EP
TE D
M AN U
SC
447
ACCEPTED MANUSCRIPT
Table 1. Patient characteristics. EGFR mutation, n (%)
EGFR WT, n (%)
418
521
Total number
0.266
Median (range)
67 (33-78)
Gender
RI PT
Age 67 (22-88)
158 (38)
299 (57)
Female
260 (62)
222 (43)
Smoking status 281 (67)
Smoker
134 (32)
Unknown
3 (1)
Serum CEA level
5 (1) 0.122
233 (56)
265 (51)
Elevated
179 (43)
250 (48)
Unknown
6 (1)
6 (1)
II–IV Lymphatic permeation
EP
Present
TE D
I
350 (84)
391 (75)
68 (16)
130 (25)
0.001
0.399
69 (17)
97 (19)
349 (83)
424 (81)
Vascular invasion
AC C
<0.001
304 (46)
Normal
Pathological stage
<0.001
212 (53)
M AN U
Non-smoker
SC
Male
Absent
P
0.002
Present
67 (16)
127 (24)
Absent
351 (84)
394 (76)
CEA, carcinoembryonic antigen; EGFR, epidermal growth factor receptor; WT, Wild type.
ACCEPTED MANUSCRIPT
RI PT
Table 2. Univariate analyses (Cox proportional hazards model) of prognostic factors for OS and RFS in all 939 patients. OS N (%)
HR
95% CI
Median (range)
67 (22-88)
1.05
1.03-1.07
95% CI
1.03
Male
457 (49)
1.83
Female
482 (51)
1
1.29-2.60
Smoking status
1.01-1.04 < 0.001
1.83
1.37-2.44
1
< 0.001
493 (53)
1
Smoker
438 (46)
2.18
Normal
498 (53)
1
Elevated
429 (46)
3.45
I
741 (79)
EP
Serum CEA level
1.52-3.11
TE D
Non-smoker
II-IV
198 (21)
6.72
Present
166 (18)
4.61
Absent
773 (82)
1
< 0.001 1 1.87
1.40-2.49
< 0.001
< 0.001 1
2.34-5.08
2.96
2.18-4.01
< 0.001
1
AC C
Pathological stage
< 0.001 1
4.73-9.54
8.26
6.20-11.0
< 0.001 3.27-6.51
P value < 0.001
0.001
M AN U
Gender
Lymphatic permeation
HR
< 0.001
SC
Age
P value
RFS
< 0.001 4.97 1
3.75-6.59
ACCEPTED MANUSCRIPT
Vascular invasion
< 0.001 194 (21)
4.45
Absent
745 (79)
1
3.16-6.27
5.53
RI PT
Present
< 0.001 4.17-7.32
1
EGFR mutation
< 0.001 418 (45)
0.45
Negative
521 (55)
1
0.31-066
0.64
SC
Positive
1
M AN U
CEA, carcinoembryonic antigen; EGFR, epidermal growth factor receptor; OS. Overall survival;
AC C
EP
TE D
RFS. Recurrence free survival; 95% CI. 95% confidence interval.
0.003 0.48-0.85
ACCEPTED MANUSCRIPT
OS N (%)
HR
95%CI
Median (range)
HR
RFS 95% CI
0.001 67 (22-88)
1.03
1.01-1.05
1.02
Normal
498 (53)
1
Elevated
429 (46)
1.78
1.18-2.69
Pathological stage
1.00-1.03 0.008
1 1.56
1.12-2.16
< 0.001
741 (79)
1
II-IV
198 (21)
3.68
Lymphatic permeation 166 (18)
1.94
Absent
773 (82)
1
Vascular invasion 194 (21)
Absent
745 (79)
1.8
AC C
Present
EP
Present
2.38-5.67
TE D
I
418 (45)
0.56
Negative
521 (55)
1
1 4.31
3.02-6.16 0.002
1.49
1.06-2.09
1 0.006
1.19-2.73
< 0.001 2.09
1
Positive
< 0.001
0.002
1.28-2.96
P value 0.041
0.006
M AN U
Serum CEA level
EGFR mutation
P value
SC
Age
RI PT
Table 3. Muivariate analyses (Cox proportional hazards model) of prognostic factors for OS and RFS in all 939 patients.
1.48-2.95
1 0.003 0.38-0.82
− − −
−
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RFS. Recurrence free survival; 95% CI. 95% confidence interval.
RI PT
CEA, carcinoembryonic antigen; EGFR, epidermal growth factor receptor; OS. Overall survival;
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0022-5223(17)31574-X
DOI:
10.1016/j.jtcvs.2017.06.062
Reference:
YMTC 11767
To appear in:
The Journal of Thoracic and Cardiovascular Surgery
Received Date: 26 January 2017 Revised Date:
7 June 2017
Accepted Date: 30 June 2017
Please cite this article as: Takamochi K, Oh S, Matsunaga T, Suzuki K, The prognostic impact of EGFRmutation status and subtypes in patients with surgically resected lung adenocarcinomas, The Journal of Thoracic and Cardiovascular Surgery (2017), doi: 10.1016/j.jtcvs.2017.06.062. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Takamochi et al. 1.
ACCEPTED MANUSCRIPT
1
Title: The prognostic impact of EGFR-mutation status and subtypes in patients with
2
surgically resected lung adenocarcinomas
4
RI PT
3
Kazuya Takamochi, MD, Shiaki Oh, MD, Takeshi Matsunaga, MD, Kenji Suzuki, MD
SC
5
Department of General Thoracic Surgery, Juntendo University School of
7
Medicine, 1-3, Hongo 3-chome, Bunkyo-ku, Tokyo 113-8431, Japan
M AN U
6
8
10
Conflicts of interest None declared.
11
TE D
9
Funding statement
13
This work was supported in part by a Grant-in-Aid for Cancer Research from the
14
Ministry of Health, Labor, and Welfare of Japan and the Smoking Research Foundation.
AC C
15
EP
12
16
Corresponding author
17
Kazuya Takamochi, MD
18
Department of General Thoracic Surgery, Juntendo University School of Medicine
Takamochi et al. 2.
ACCEPTED MANUSCRIPT
1-3, Hongo 3-chome, Bunkyo-ku, Tokyo 113-8431, Japan.
20
Tel: +81-3-3813-3111
21
Fax: +81-3-5800-0281
22
E-mail: [email protected]
RI PT
19
Date and number of IRB approval
25
March 19, 2015; No. 2014172
26
Article word count
28
2702 words
29
TE D
27
M AN U
24
SC
23
Abbreviations and Acronyms
31
CEA
32
CT
33
EGFR – epidermal growth factor receptor
34
IASLC – International Association for the Study of Lung Cancer
35
IRB
– institutional review board
36
KM
– Kaplan-Meier
EP
30
AC C
– carcinoembryonic antigen – computed tomography
Takamochi et al. 3.
ACCEPTED MANUSCRIPT
MRI
– magnetic resonance imaging
38
NSCLC – non-small-cell lung cancer
39
OS
– overall survival
40
PET
– 18F-fluoro-2-deoxy-glucose positron emission tomography
41
RFS
– recurrence-free survival
42
TKIs
– tyrosine kinase inhibitors
43
WT
– wild type
M AN U
SC
RI PT
37
44
Central Picture
46
Figure 2A
47
Improved survival associated with EGFR-mutant lung tumors relative to -wild-type
48
tumors.
EP AC C
49
TE D
45
50
Central Message
51
Positive EGFR-mutation status is a favorable prognostic factor in patients with
52
surgically resected lung adenocarcinomas; however, EGFR-mutation subtypes have no
53
prognostic impact.
54
Takamochi et al. 4.
ACCEPTED MANUSCRIPT
Perspective Statement
56
We showed that positive EGFR-mutation status is a favorable prognostic factor
57
independent of TNM stage in patients with surgically resected lung adenocarcinomas.
58
During the next revision of TNM classification of lung cancer, cancer genotype might
59
be adapted as a new category, and in future clinical trials for adjuvant therapy,
60
stratification by EGFR-mutation status might be necessary to avoid bias.
M AN U
SC
RI PT
55
AC C
EP
TE D
61
Takamochi et al. 5.
ACCEPTED MANUSCRIPT
Abstract
63
Objective: Epidermal growth factor receptor (EGFR) gene-mutation status is a
64
well-established predictor of the efficacy of EGFR tyrosine-kinase inhibitors (TKIs) in
65
non-small-cell lung cancer. Recently, differences in EGFR-mutation subtypes were also
66
reported to be associated with the efficacy of EGFR TKIs. However, the prognostic
67
impact of EGFR-mutation status and subtypes remains controversial.
68
Methods: We retrospectively reviewed 939 patients with surgically resected
69
adenocarcinomas who had their EGFR-mutation status analyzed between January 2010
70
and December 2014. Overall survival (OS) and recurrence-free survival (RFS) were
71
compared using the log-rank test according to pathological stage, EGFR-mutation status,
72
and EGFR-mutation subtypes. Independent prognostic factors for OS and RFS were
73
identified by multivariate analyses using the Cox proportional-hazards model.
74
Results: A median follow-up time was 48 months. We found that positive
75
EGFR-mutation status was significantly associated with longer OS and RFS in all
76
patients and associated with longer OS in pathological stage I patients. However, no
77
significant differences were observed in OS and RFS between patients with exon-21
78
L858R mutation and those with exon-19 deletions. In a Cox regression model for OS,
79
EGFR-mutation status was a significant prognostic factor that was independent of
AC C
EP
TE D
M AN U
SC
RI PT
62
Takamochi et al. 6.
ACCEPTED MANUSCRIPT
well-established prognostic factors, such as age, pathological stage, vascular invasion,
81
lymphatic permeation, and serum carcinoembryonic antigen level.
82
Conclusions: Positive EGFR-mutation status is a favorable prognostic factor in patients
83
with surgically resected lung adenocarcinomas; however, EGFR-mutation subtypes
84
(exon-21 L858R mutation or exon-19 deletions) exhibited no prognostic impact.
SC
RI PT
80
AC C
EP
TE D
M AN U
85
Takamochi et al. 7.
ACCEPTED MANUSCRIPT
Mutations in the epidermal growth factor receptor gene (EGFR) represent one of the
87
most frequent driver-oncogene alterations in non-small-cell lung cancer (NSCLC).
88
EGFR mutations are more frequently found in females, those who have never smoked,
89
adenocarcinoma patients, and members of East Asian populations.1 The frequency of
90
EGFR mutations in non-adenocarcinoma NSCLC is quite low (<2%).1 The EGFR
91
tyrosine kinase inhibitors (TKIs), gefitinib and erlotinib, which specifically block the
92
EGFR-dependent pathway, were initially approved among molecular-target drugs for
93
their use in the treatment of NSCLC.2 EGFR-mutated NSCLC has become a clinically
94
relevant molecular subset, because patients with NSCLC harboring activating EGFR
95
mutations (exon-21 L858R mutation or exon-19 deletions) benefit from EGFR-TKI
96
treatment unlike those with EGFR-wild-type (WT) tumors.3, 4
SC
M AN U
TE D
Both the presence of EGFR mutations and differences in EGFR-mutation
EP
97
RI PT
86
subtypes (exon-21 L858R mutation or exon 19 deletions) were reportedly associated
99
with the efficacy of EGFR TKIs. A meta-analysis of eight clinical trials and five
AC C
98
100
retrospective studies using EGFR TKIs for advanced EGFR-mutated NSCLCs
101
demonstrated that exon-19 deletion was associated with longer progression-free survival
102
as compared with exon-21 L858R mutation.5 Therefore, both EGFR-mutation status and
103
EGFR-mutation subtypes are considered significant predictors of the efficacy of EGFR
Takamochi et al. 8.
ACCEPTED MANUSCRIPT
104
TKIs in NSCLC. By contrast, the prognostic impact of both EGFR-mutation status and subtypes
105
in NSCLC patients remains controversial. A systematic review with meta-analysis based
107
on data from 16 studies (n = 3337) showed that EGFR mutations were not a prognostic
108
factor in patients with surgically resected NSCLC.6 However, all of the studies included
109
in the meta-analysis were retrospectively conducted, the number of patients was small,
110
the follow-up period was relatively short, and the treatments for recurrent disease
111
(especially regarding the usage of EGFR-TKIs) were heterogeneous. Therefore, as a
112
matter of course, further study is needed to confirm the results. The prognostic impact
113
of EGFR-mutation subtypes (exon-21 L858R point mutation or exon-19 deletions) in
114
NSCLC patients has not been well established, because the studies investigating the
115
prognostic impact of EGFR-mutation subtypes in early stage NSCLC patients who
116
underwent surgical resection were scarcely reported.7-9 In this study, we elucidated the
117
prognostic impact of EGFR-mutation status and subtypes on surgically resected lung
118
adenocarcinoma patients.
AC C
EP
TE D
M AN U
SC
RI PT
106
119
120
MATERIALS AND METHODS
121
Patients
Takamochi et al. 9.
ACCEPTED MANUSCRIPT
Between January 2010 and December 2014, 1018 lung adenocarcinoma
123
patients underwent complete surgical resection at Juntendo University School of
124
Medicine (Tokyo, Japan). Among these patients, we retrospectively reviewed
125
clinicopathological data for 939 consecutive patients with lung adenocarcinomas who
126
had their EGFR-mutation status analyzed. A flowchart illustrating the selection of study
127
cohorts is shown in Figure 1.
SC M AN U
128
RI PT
122
Preoperative staging was routinely performed with computed tomography (CT) scanning of the chest and upper abdomen and 18F-fluoro-2-deoxy-glucose positron
130
emission tomography (PET)/CT scans. Cranial CT scanning or magnetic resonance
131
imaging (MRI) was performed according to clinical signs and symptoms. The patients
132
were staged based on the 7th edition of the International Association for the Study of
133
Lung Cancer (IASLC) TNM classification for lung cancer.10 Induction chemotherapy
134
and/or radiotherapy and treatments with EGFR TKIs as induction or adjuvant therapy
135
were not performed on any of the patients in this series.
136
Follow-up and Diagnosis of Recurrence
137
Follow-up examinations were performed every 3 or 4 months for the first 3 years and
138
every 6 months thereafter. Chest X-rays and biochemical examination of blood,
139
including for carcinoembryonic antigen (CEA), were performed at every visit to the
AC C
EP
TE D
129
Takamochi et al. 10.
ACCEPTED MANUSCRIPT
outpatient department. Chest CT scans or PET/CT scans were performed annually.
141
Cranial CT scanning or MRI was performed as necessary. Recurrences were diagnosed
142
based on radiological evidence of cancer relapse upon surveillance imaging and/or
143
pathologic evidence according to a tumor biopsy.
144
EGFR-mutation Analysis
145
We routinely analyzed EGFR mutations for lung cancer patients who underwent surgery
146
for a research purpose. Genomic DNA was extracted from cubes (3–5 mm3) of frozen,
147
fresh lung-cancer tissue samples from surgically resected specimens. The peptide
148
nucleic acid-locked nucleic acid polymerase chain reaction clamp method11 was used to
149
identify EGFR mutations G719A, G719C, and G719A in exon 18, all deletion
150
genotypes in exon 19, T790M in exon 20, and L858R and L861Q in exon 21.
SC
M AN U
TE D
Approval for collection of patient specimens from the tissue bank in our
EP
151
RI PT
140
department was obtained from the institutional review board (IRB) of Juntendo
153
University School of Medicine. Individual patient consent was obtained for
154
procurement of tissue for research purposes prior to surgery. This retrospective study
155
was performed under a waiver of authorization approved by the IRB.
156
Statistical Analysis
AC C
152
Takamochi et al. 11.
ACCEPTED MANUSCRIPT
Overall survival (OS) and recurrence-free survival (RFS) were analyzed by
158
Kaplan-Meier (KM) methods, and curve differences were tested using the log-rank test
159
according to pathological stage, EGFR-mutation status, and EGFR-mutation subtypes.
160
The length of OS was defined as the interval in months between the day of surgical
161
intervention and the date of either death or the last follow-up. The length of RFS was
162
defined as the interval in months between the day of surgical intervention and the date
163
of either recurrence, death from any cause, or the last follow-up. Univariate and
164
multivariate analyses of independent prognostic factors in each cohort were performed
165
using the Cox proportional-hazards model. All significant variables (P < 0.05) in the
166
univariate analyses were entered into the multivariate analyses using a stepwise
167
variable-selection procedure to adjust for potential confounding factors.
168
A P < 0.05 was considered statistically significant. All statistical analyses were
169
performed using the SPSS statistical software package (v20.0; SPSS Inc., Chicago, IL,
170
USA).
SC
M AN U
TE D
EP
AC C
171
RI PT
157
172
RESULTS
173
EGFR mutations were detected in 418 (45%) of the 939 lung adenocarcinoma patients,
174
with the point mutation L858R in exon 21 and exon-19 deletions detected in 222 (53%)
Takamochi et al. 12.
ACCEPTED MANUSCRIPT
and 161 (39%) of 418 EGFR-mutated tumors, respectively. The remaining minor EGFR
176
mutations, including exon-18 G719A in eight, exon-18 G719S in five, exon-18 G719C
177
in three, exon-21 L861Q in seven, and exon-20 T790M in four, were detected in 23
178
(6%) tumors. Double mutations were found in 12 (3%) tumors and included exon- 21
179
L858R and exon 20-T790M in eight, exon-20 T790M and exon-19 deletion in one,
180
exon-20 T790M and exon-18 G719A in one, exon-18 G719A and exon-21 L861Q in
181
one, and exon-18 G719S and exon-21 L861Q in one. Interestingly, exon-20 T790M,
182
which confers resistance to EGFR TKIs,12 was consistently coupled with other
183
mutations. Patient characteristics are summarized in Table 1. Female (P < 0.001),
184
non-smoker (P < 0.001), pathological stage I tumor (P = 0.001), and a tumor without
185
vascular invasion (P = 0.002) were more frequently observed characteristics in patients
186
with EGFR-mutated tumors than in those with EGFR-WT tumors.
SC
M AN U
TE D
EP
To evaluate the prognostic impact of EGFR-mutation status, we analyzed OS
AC C
187
RI PT
175
188
and RFS in 939 overall patients with a median follow-up time of 48 months. The
189
3-/5-year OS rates were 95% [95% confidence interval (CI): 92–97%] /89% (95% CI:
190
84–92%) in 418 patients with EGFR-mutated tumors and 85% (95% CI: 82–88%)/78%
191
(95% CI: 73–82%) in 521 patients with EGFR-WT tumors, respectively. Positive
192
EGFR-mutation status was significantly associated with longer OS in all patients (Fig.
Takamochi et al. 13.
ACCEPTED MANUSCRIPT
2A; P < 0.001). Additionally, the 3-/5-year RFS rates were 85% (95%CI: 81–88%)/81%
194
(95% CI: 76–85%) in 418 patients with EGFR-mutated tumors and 78% (95% CI: 75–
195
82%)/72% (95% CI: 68–77%) in 521 patients with EGFR-WT tumors, respectively,
196
with positive EGFR-mutation status was significantly associated with longer RFS in all
197
patients (Fig. 2B; P = 0.002).
SC
RI PT
193
The 3-/5-year OS rates were 98% (95% CI: 96–99%)/94% (95% CI: 90–97%)
199
in 350 patients with EGFR-mutated pathological stage I tumors and 93% (95% CI: 90–
200
95%)/87% (95% CI: 83–91%) in 391 patients with EGFR-WT pathological stage I
201
tumors, respectively, with positive EGFR-mutation status significantly associated with
202
longer OS in pathological stage I patients (Fig. E1A; P = 0.003). Additionally, the
203
3-/5-year RFS rates were 93% (95% CI: 90–96%)/89% (95% CI: 84–93%) in 350
204
patients with EGFR-mutated pathological stage I tumors and 90% (95% CI: 86–
205
93%)/84% (95% CI: 80–88%) in 391 patients with EGFR-WT pathological stage I
206
tumors, respectively, with positive EGFR-mutation status associated with longer RFS in
207
pathological stage I patients; however, the difference was not statistically significant
208
(Fig. E1B; P = 0.065). Cox regression analysis of prognostic factors for OS in all
209
patients showed that EGFR-mutation status was a significant prognostic factor
210
independent of well-established prognostic factors, such as age, serum CEA level,
AC C
EP
TE D
M AN U
198
Takamochi et al. 14.
ACCEPTED MANUSCRIPT
pathological stage, vascular invasion, and lymphatic permeation (Tables 2 and 3). Cox
212
regression analysis of prognostic factors for RFS in all patients showed that
213
EGFR-mutation status was a significant prognostic factor in the univariate analysis;
214
however, this was not significant in the multivariate analysis (Tables 2 and 3).
To evaluate the prognostic impact of EGFR-mutation subtypes, we analyzed
SC
215
RI PT
211
OS and RFS rates for patients with EGFR-mutated tumors in three groups stratified by
217
EGFR-mutation subtypes: exon-19 deletions, L858R, and other subtypes, including
218
minor mutations or double mutations. In 418 patients with EGFR-mutated tumors, no
219
significant differences were observed in OS and RFS rates according to mutation
220
subtype (Fig. 3A: exon-19 deletions vs. L858R, P = 0.172; exon-19 deletions vs. other
221
subtypes, P = 0.056; L858R vs. other subtypes, P = 0.300; Fig. 3B: exon-19 deletions vs.
222
L858R, P = 0.646; exon-19 deletions vs. other subtypes, P = 0.961; L858R vs. other
223
subtypes, P = 0.696).
TE D
EP
AC C
224
M AN U
216
Postoperative recurrences developed in 57 patients with EGFR-mutated
225
tumors: 10 locoregional only, 31 distant only, and 16 locoregional and distant. Among
226
these, 41 patients were treated with EGFR TKIs for any lines following recurrence, and
227
OS time tended to be longer in these patients as compared with that observed in 16
228
patients not treated with EGFR TKIs; however, the difference was not statistically
Takamochi et al. 15.
ACCEPTED MANUSCRIPT
significant (P = 0.173; Fig. 4). In the 41 patients who received EGFR TKI treatment,
230
both OS and survival periods following recurrence were longer in 19 patients with
231
tumors harboring exon-19-deletion mutations as compared with those in 18 patients
232
with tumors harboring the L858R mutation; however, the differences were not
233
statistically significant (P = 0.256 and P = 0.198, respectively; Fig. E2).
SC
RI PT
229
M AN U
234
DISCUSSION
236
EGFR-mutation status and subtypes indisputably constitute significant predictive
237
factors for the efficacy of EGFR TKIs in advanced NSCLC. However, their prognostic
238
role remains unclear, because only inconsistent results from studies of small and
239
heterogeneous populations are available.6-8 Elucidation of the intrinsic prognostic
240
impact of EGFR-mutation status and EGFR-mutation subtypes in advanced
241
EGFR-mutated NSCLC patients is currently considered impossible, because EGFR
242
TKIs are used as a standard first-line treatment. Furthermore, third-generation EGFR
243
TKIs were recently shown to prolong survival, even after the development of resistance
244
to first- or second-generation EGFR TKIs.13 The prognosis for advanced EGFR-mutated
245
NSCLC patients is expected to become increasingly prolonged. Even in the presence or
246
absence of intrinsic prognostic impacts associated with EGFR-mutation status and
AC C
EP
TE D
235
Takamochi et al. 16.
ACCEPTED MANUSCRIPT
subtypes, their affect would be obscured by a definitive association with the efficacy of
248
EGFR TKIs. Therefore, the intrinsic prognostic role of EGFR-mutation status and
249
subtypes needs to be investigated in surgically resected early stage NSCLC patients who
250
had not received perioperative therapy using EGFR TKIs. To the best of our knowledge,
251
this study constitutes the largest-scale evaluation of patients with surgically resected
252
EGFR-mutated tumors for evaluating the prognostic impact of EGFR-mutation status
253
and subtypes by analyzing both OS and RFS rates.
SC
M AN U
254
RI PT
247
Here, we showed that positive EGFR-mutation status was significantly associated with longer OS time regardless of pathological stage in KM analyses.
256
Furthermore, multivariate analysis revealed that EGFR-mutation status was a significant
257
prognostic factor for OS time independent of pathological stage. Additionally, although
258
the number of patients who received EGFR TKIs for postoperative recurrence in the
259
present study was small, there was no statistical difference in OS regardless of whether
260
patients with recurrent EGFR-mutated tumors received EGFR TKI treatment. OS time
261
might not be significantly influenced by EGFR TKI treatment for postoperative
262
recurrences. Therefore, we concluded that positive EGFR-mutation status is a favorable
263
prognostic factor in patients with surgically resected lung adenocarcinomas.
264
AC C
EP
TE D
255
By contrast, no significant differences were observed in OS and RFS rates
Takamochi et al. 17.
ACCEPTED MANUSCRIPT
between patients with an exon-21 L858R mutation (n = 222) and those with exon-19
266
deletions in EGFR-mutated tumors (n = 161). In 41 patients who received EGFR TKI
267
treatment, both OS and survival periods following recurrence were longer in patients
268
with tumors harboring exon-19 deletions (n = 19) as compared with that in those with
269
tumors harboring an L858R mutation (n = 18), although the differences were not
270
statistically significant. Therefore, we concluded that EGFR-mutation subtypes
271
(exon-21 L858R mutation or exon-19 deletions) have no prognostic impact on patients
272
with surgically resected lung adenocarcinomas. However, further large-scale studies are
273
needed to confirm these observations.
SC
M AN U
Although the underlying mechanism associated with the favorable prognostic
TE D
274
RI PT
265
impact of EGFR mutations in lung adenocarcinoma remains unclear, EGFR-mutated
276
lung adenocarcinoma may be less likely to be affected by complex oncogenic processes,
277
because EGFR mutations provide strong growth advantages due to their constitutive
278
activation. Therefore, EGFR-mutated lung adenocarcinoma cells need not acquire
279
additional oncogenic mutations to enhance their biological malignancy via cell-based
280
characteristics, such as invasive, anti-apoptotic, and metastatic potentials. As a result,
281
EGFR-mutated lung adenocarcinoma (non-smoking- related lung cancer) may acquire a
282
lower degree of biological malignancy relative to that observed in smoking-related lung
AC C
EP
275
Takamochi et al. 18.
ACCEPTED MANUSCRIPT
283
284
cancers exhibiting high nonsynonymous-mutation burden.14 Recently, TNM classifications for lung cancer were newly revised by the IASLC-staging committee.15 Although the revised TNM-staging system adapted
286
categories based on prognostic factors associated with the anatomical extent of tumors,
287
no regard was given to the biological features of tumors, such as driver mutations.
288
Recent advances in the understanding of NSCLC-related molecular biology would
289
likely offer the potential for therapeutic success by matching systemic target therapies to
290
cancer genotype. EGFR mutations offer not only predictive significance for EGFR TKI
291
treatment, but also improved prognostic significance from a biological standpoint for
292
lung adenocarcinoma. Other alterations in driver oncogenes that predict prognostic
293
impact similar to EGFR mutations might be identified in the future. Therefore,
294
regardless of how precise the classification system becomes, a TNM-staging system
295
based solely on the anatomical extent of tumors will become outdated and almost
296
irrelevant to clinical practice in the era of molecular-target therapy. During the next
297
revision to the TNM classification of lung cancer, cancer genotype might be adapted as
298
a new category.
299
300
AC C
EP
TE D
M AN U
SC
RI PT
285
In several clinical trials,16, 17 higher response rates were reported for chemotherapy in both first-line and second-line advanced-NSCLC settings in the
Takamochi et al. 19.
ACCEPTED MANUSCRIPT
presence of EGFR mutations. In the phase III studies of adjuvant gefitinib (BR19) and
302
erlotinib (RADIANT) for patients with completely resected stage IB to IIIA NSCLC,
303
EGFR TKIs failed to show survival benefits as compared with placebo in subgroups of
304
patients with EGFR-mutated tumors.18, 19 However, D'Angelo et al20 reported that
305
adjuvant therapy with gefitinib or erlotinib was significantly associated with a lower
306
risk of recurrence or death, disease-free survival, and a trend toward improved OS in
307
their retrospective study. In a randomized phase II study, adjuvant therapy involving
308
gefitinib following pemetrexed and carboplatin showed significant improvement in
309
disease-free survival in patients with stage IIIA NSCLC harboring EGFR mutations as
310
compared with pemetrexed and carboplatin treatment alone.21 Furthermore, in
311
opposition to our results, positive EGFR-mutation status was recently reported as a
312
favorable independent prognostic factor for survival periods following recurrence.22
313
Therefore, stratification by EGFR-mutation status might be necessary to avoid an
314
uneven distribution of EGFR-mutant populations that could potentially confound
315
survival data in future clinical trials related to adjuvant therapy.
AC C
EP
TE D
M AN U
SC
RI PT
301
316
Our study must be interpreted in the context of the following limitations. First,
317
although the number of the EGFR-mutant cohort was the largest evaluated to date, this
318
study represents a retrospective comparative study at the single institution. Therefore,
Takamochi et al. 20.
ACCEPTED MANUSCRIPT
patient-selection bias is inevitable. Second, the intrinsic prognostic role of
320
EGFR-mutation (EGFR mutation as a prognostic factor) might not be accurately
321
evaluated, because we could not completely exclude the efficacy of EGFR TKI
322
treatment (EGFR mutation as a predictive factor). Namely, our study did not include
323
any patients who received EGFR TKI treatment in the induction or adjuvant setting;
324
however, 41 (4%) of 939 patients received EGFR TKI treatment for postoperative
325
recurrences. Third, although EGFR TKI treatment for patients with recurrent
326
EGFR-mutated tumors did not show statistical significance for OS, the number of
327
patients who received EGFR TKI treatment for postoperative recurrence in the present
328
study might be too small to draw a definitive conclusion. Therefore, further large-scale
329
studies are warranted in this population.
TE D
M AN U
SC
RI PT
319
In conclusion, our analyses indicated that positive EGFR-mutation status is a
331
favorable prognostic factor in patients with surgically resected lung adenocarcinomas.
332
Additionally, our results showed that EGFR-mutation subtypes (L858R mutation or
333
exon-19 deletions) offered no prognostic impact, although they might be significant
334
predictors of EGFR TKI efficacy.
335
AC C
EP
330
Takamochi et al. 21.
ACCEPTED MANUSCRIPT
336
References
337
1.
338
EGFR mutations in lung cancer. Int J Clin Oncol. 2006;11:190-8.
339
2.
340
treating cancer. Nat Rev Cancer. 2004;4:956-65.
341
3.
342
BW et al. Activating mutations in the epidermal growth factor receptor underlying
343
responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med.
344
2004;350:2129-39.
345
4.
346
mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science.
347
2004;304:1497-500.
348
5.
349
deletion were associated with longer progression-free survival compared to those with
350
L858R mutation after first-line EGFR-TKIs for advanced non-small cell lung cancer: a
351
meta-analysis. PLoS One. 2014;9:e107161. doi: 10.1371/journal.pone.0107161.
352
eCollection 2014.
353
6.
RI PT
Mitsudomi T, Kosaka T, Yatabe Y. Biological and clinical implications of
SC
Herbst RS, Fukuoka M, Baselga J. Gefitinib--a novel targeted approach to
M AN U
Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan
EP
TE D
Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S et al. EGFR
AC C
Zhang Y, Sheng J, Kang S, Fang W, Yan Y, Hu Z et al. Patients with exon 19
Zhang Z, Wang T, Zhang J, Cai X, Pan C, Long Y et al. Prognostic value of
Takamochi et al. 22.
ACCEPTED MANUSCRIPT
epidermal growth factor receptor mutations in resected non-small cell lung cancer: a
355
systematic review with meta-analysis. PLoS One. 2014;9:e106053. doi:
356
10.1371/journal.pone.0106053. eCollection 2014.
357
7.
358
mutations within the EGFR kinase domain: a molecular predictor of disease-free
359
survival in resected pulmonary adenocarcinoma. J Cancer Res Clin Oncol.
360
2009;135:1647-54. doi: 10.007/s00432-009-0611-7. Epub 2009 Jun 11.
361
8.
362
epidermal growth factor receptor mutations in resected lung adenocarcinomas. Med
363
Oncol. 2014;31:771. doi: 10.1007/s12032-013-0771-9. Epub 2013 Nov 19.
364
9.
365
Prognostic Impact of EGFR Driver Mutations on Postoperative Disease Recurrence in
366
Lung Adenocarcinoma. Anticancer Res. 2016;36:3057-63.
367
10.
368
al. The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM
369
stage groupings in the forthcoming (seventh) edition of the TNM Classification of
370
malignant tumours. J Thorac Oncol. 2007;2:706-14.
371
11.
RI PT
354
M AN U
SC
Lee YJ, Park IK, Park MS, Choi HJ, Cho BC, Chung KY et al. Activating
TE D
Liu WS, Zhao LJ, Pang QS, Yuan ZY, Li B, Wang P. Prognostic value of
AC C
EP
Okamoto T, Kitahara H, Shimamatsu S, Katsura M, Takada K, Fujishita T et al.
Goldstraw P, Crowley J, Chansky K, Giroux DJ, Groome PA, Rami-Porta R et
Nagai Y, Miyazawa H, Huqun, Tanaka T, Udagawa K, Kato M et al. Genetic
Takamochi et al. 23.
ACCEPTED MANUSCRIPT
heterogeneity of the epidermal growth factor receptor in non-small cell lung cancer cell
373
lines revealed by a rapid and sensitive detection system, the peptide nucleic acid-locked
374
nucleic acid PCR clamp. Cancer Res. 2005;65:7276-82.
375
12.
376
EGFR mutation and resistance of non-small-cell lung cancer to gefitinib. N Engl J Med.
377
2005;352:786-92.
378
13.
379
growth factor receptor tyrosine kinase inhibitors in the treatment of lung cancer. Cancer.
380
2015;121:E1-6. doi: 10.1002/cncr.29139. Epub 2014 Dec 17.
381
14.
382
Genomic landscape of non-small cell lung cancer in smokers and never-smokers. Cell.
383
2012;150:1121-34. doi: 10.016/j.cell.2012.08.024.
384
15.
385
et al. The IASLC Lung Cancer Staging Project: Proposals for Revision of the TNM
386
Stage Groupings in the Forthcoming (Eighth) Edition of the TNM Classification for
387
Lung Cancer. J Thorac Oncol. 2016;11:39-51. doi: 10.1016/j.jtho.2015.09.009.
388
16.
389
Thatcher N et al. Molecular predictors of outcome with gefitinib in a phase III
RI PT
372
M AN U
SC
Kobayashi S, Boggon TJ, Dayaram T, Janne PA, Kocher O, Meyerson M et al.
Steuer CE, Khuri FR, Ramalingam SS. The next generation of epidermal
EP
TE D
Govindan R, Ding L, Griffith M, Subramanian J, Dees ND, Kanchi KL et al.
AC C
Goldstraw P, Chansky K, Crowley J, Rami-Porta R, Asamura H, Eberhardt WE
Hirsch FR, Varella-Garcia M, Bunn PA, Jr., Franklin WA, Dziadziuszko R,
Takamochi et al. 24.
ACCEPTED MANUSCRIPT
390
placebo-controlled study in advanced non-small-cell lung cancer. J Clin Oncol.
391
2006;24:5034-42.
392
17.
393
carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med. 2009;361:947-57.
394
doi: 10.1056/NEJMoa0810699. Epub 2009 Aug 19.
395
18.
396
Gefitinib versus placebo in completely resected non-small-cell lung cancer: results of
397
the NCIC CTG BR19 study. J Clin Oncol. 2013;31:3320-6.
398
19.
399
Adjuvant Erlotinib Versus Placebo in Patients With Stage IB-IIIA Non-Small-Cell Lung
400
Cancer (RADIANT): A Randomized, Double-Blind, Phase III Trial. J Clin Oncol.
401
2015;33:4007-14. doi: 10.1200/JCO.2015.61.8918. Epub 2015 Aug 31.
402
20.
403
Distinct clinical course of EGFR-mutant resected lung cancers: results of testing of 1118
404
surgical specimens and effects of adjuvant gefitinib and erlotinib. J Thorac Oncol.
405
2012;7:1815-22. doi: 10.097/JTO.0b013e31826bb7b2.
406
21.
407
adjuvant chemotherapy with or without gefitinib in resected stage IIIA-N2 non-small
SC
RI PT
Mok TS, Wu YL, Thongprasert S, Yang CH, Chu DT, Saijo N et al. Gefitinib or
M AN U
Goss GD, O'Callaghan C, Lorimer I, Tsao MS, Masters GA, Jett J et al.
EP
TE D
Kelly K, Altorki NK, Eberhardt WE, O'Brien ME, Spigel DR, Crino L et al.
AC C
D'Angelo SP, Janjigian YY, Ahye N, Riely GJ, Chaft JE, Sima CS et al.
Li N, Ou W, Ye X, Sun HB, Zhang L, Fang Q et al. Pemetrexed-carboplatin
Takamochi et al. 25.
ACCEPTED MANUSCRIPT
408
cell lung cancer harbouring EGFR mutations: a randomized, phase II study. Ann Surg
409
Oncol. 2014;21:2091-6. doi: 10.1245/s10434-014-3586-9. Epub 2014 Mar 1.
410
22.
411
Prognostic Factors for Survival After Recurrence in Patients With Completely Resected
412
Lung Adenocarcinoma: Important Roles of Epidermal Growth Factor Receptor
413
Mutation Status and the Current Staging System. Clin Lung Cancer. 2015;16:e213-21.
414
doi: 10.1016/j.cllc.2015.04.005. Epub
M AN U
SC
RI PT
Kudo Y, Shimada Y, Saji H, Kato Y, Yoshida K, Matsubayashi J et al.
AC C
EP
TE D
415
Apr 20.
Takamochi et al. 26.
ACCEPTED MANUSCRIPT
Figure Legends
417
FIGURE 1. Flowchart describing the selection of the study cohort.
418
EGFR, epidermal growth factor receptor; TKI, tyrosine-kinase inhibitor; WT, wild type.
RI PT
416
SC
419
FIGURE 2. OS (A) and RFS (B) in all patients according to EGFR-mutation status.
421
EGFR, epidermal growth factor receptor; OS, overall survival; RFS, recurrence-free
422
survival; WT, wild type.
TE D
M AN U
420
423
FIGURE 3. OS (A) and RFS (B) in patients with EGFR-mutated tumors according to
425
EGFR-mutation subtype (exon-19 deletions, exon-21 L858R mutation, or other
426
mutations).
427
EGFR, epidermal growth factor receptor; OS, overall survival; RFS, recurrence-free
428
survival.
429
AC C
EP
424
Takamochi et al. 27.
ACCEPTED MANUSCRIPT
FIGURE 4. A comparison of OS time between patients with EGFR-mutated tumors
431
who received EGFR-TKI treatment and those who did not following recurrence.
432
EGFR, epidermal growth factor receptor; OS, overall survival; TKI, tyrosine-kinase
433
inhibitor.
SC
RI PT
430
M AN U
434
FIGURE E1. OS (A) and RFS (B) in patients with pathological stage I lung
436
adenocarcinomas according to EGFR-mutation status.
437
EGFR, epidermal growth factor receptor; OS, overall survival; RFS, recurrence-free
438
survival; WT, wild type.
TE D
435
EP
439
FIGURE E2. OS (A) and survival periods following recurrence (B) in patients who
441
received EGFR-TKI treatment according to EGFR-mutation subtype (exon-19 deletions,
442
exon-21 L858R mutation, or other mutations).
443
EGFR, epidermal growth factor receptor; OS, overall survival; TKI, tyrosine-kinase
444
inhibitor.
AC C
440
Takamochi et al. 28.
ACCEPTED MANUSCRIPT
Video Legend
446
The importance and relevance of the present study are summarized on a video.
RI PT
445
AC C
EP
TE D
M AN U
SC
447
ACCEPTED MANUSCRIPT
Table 1. Patient characteristics. EGFR mutation, n (%)
EGFR WT, n (%)
418
521
Total number
0.266
Median (range)
67 (33-78)
Gender
RI PT
Age 67 (22-88)
158 (38)
299 (57)
Female
260 (62)
222 (43)
Smoking status 281 (67)
Smoker
134 (32)
Unknown
3 (1)
Serum CEA level
5 (1) 0.122
233 (56)
265 (51)
Elevated
179 (43)
250 (48)
Unknown
6 (1)
6 (1)
II–IV Lymphatic permeation
EP
Present
TE D
I
350 (84)
391 (75)
68 (16)
130 (25)
0.001
0.399
69 (17)
97 (19)
349 (83)
424 (81)
Vascular invasion
AC C
<0.001
304 (46)
Normal
Pathological stage
<0.001
212 (53)
M AN U
Non-smoker
SC
Male
Absent
P
0.002
Present
67 (16)
127 (24)
Absent
351 (84)
394 (76)
CEA, carcinoembryonic antigen; EGFR, epidermal growth factor receptor; WT, Wild type.
ACCEPTED MANUSCRIPT
RI PT
Table 2. Univariate analyses (Cox proportional hazards model) of prognostic factors for OS and RFS in all 939 patients. OS N (%)
HR
95% CI
Median (range)
67 (22-88)
1.05
1.03-1.07
95% CI
1.03
Male
457 (49)
1.83
Female
482 (51)
1
1.29-2.60
Smoking status
1.01-1.04 < 0.001
1.83
1.37-2.44
1
< 0.001
493 (53)
1
Smoker
438 (46)
2.18
Normal
498 (53)
1
Elevated
429 (46)
3.45
I
741 (79)
EP
Serum CEA level
1.52-3.11
TE D
Non-smoker
II-IV
198 (21)
6.72
Present
166 (18)
4.61
Absent
773 (82)
1
< 0.001 1 1.87
1.40-2.49
< 0.001
< 0.001 1
2.34-5.08
2.96
2.18-4.01
< 0.001
1
AC C
Pathological stage
< 0.001 1
4.73-9.54
8.26
6.20-11.0
< 0.001 3.27-6.51
P value < 0.001
0.001
M AN U
Gender
Lymphatic permeation
HR
< 0.001
SC
Age
P value
RFS
< 0.001 4.97 1
3.75-6.59
ACCEPTED MANUSCRIPT
Vascular invasion
< 0.001 194 (21)
4.45
Absent
745 (79)
1
3.16-6.27
5.53
RI PT
Present
< 0.001 4.17-7.32
1
EGFR mutation
< 0.001 418 (45)
0.45
Negative
521 (55)
1
0.31-066
0.64
SC
Positive
1
M AN U
CEA, carcinoembryonic antigen; EGFR, epidermal growth factor receptor; OS. Overall survival;
AC C
EP
TE D
RFS. Recurrence free survival; 95% CI. 95% confidence interval.
0.003 0.48-0.85
ACCEPTED MANUSCRIPT
OS N (%)
HR
95%CI
Median (range)
HR
RFS 95% CI
0.001 67 (22-88)
1.03
1.01-1.05
1.02
Normal
498 (53)
1
Elevated
429 (46)
1.78
1.18-2.69
Pathological stage
1.00-1.03 0.008
1 1.56
1.12-2.16
< 0.001
741 (79)
1
II-IV
198 (21)
3.68
Lymphatic permeation 166 (18)
1.94
Absent
773 (82)
1
Vascular invasion 194 (21)
Absent
745 (79)
1.8
AC C
Present
EP
Present
2.38-5.67
TE D
I
418 (45)
0.56
Negative
521 (55)
1
1 4.31
3.02-6.16 0.002
1.49
1.06-2.09
1 0.006
1.19-2.73
< 0.001 2.09
1
Positive
< 0.001
0.002
1.28-2.96
P value 0.041
0.006
M AN U
Serum CEA level
EGFR mutation
P value
SC
Age
RI PT
Table 3. Muivariate analyses (Cox proportional hazards model) of prognostic factors for OS and RFS in all 939 patients.
1.48-2.95
1 0.003 0.38-0.82
− − −
−
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RFS. Recurrence free survival; 95% CI. 95% confidence interval.
RI PT
CEA, carcinoembryonic antigen; EGFR, epidermal growth factor receptor; OS. Overall survival;
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT