Abstract

Background

Oncogenic driver mutations are responsible for the initiation and maintenance of non-small-cell lung cancer (NSCLC). Elucidation of driver mutation occurrence in NSCLC has important clinical implications.

Patients and methods

NSCLC at various clinical stages were studied for their oncogenic mutations and their association with patients' disease-free survival (DFS).

Results

Of 488 patients with NSCLC, 28 had EML4-ALK fusions. Female, young age (<60 years old), and nonsmoker patients had significant greater mutation frequencies than male, old age (≥60 years old), and smoker patients, respectively (P<0.05). Of 392 patients with NSCLC, 13 had PIK3CA mutations and 3 had MEK1 mutations. EML4-ALK, PIK3CA, and MEK1 mutations were mutually exclusive. EML4-ALK fusion was found to be of coexistence with EGFR and KRAS mutations in two cases. In stage IA NSCLC, EML4-ALK-positive patients had longer DFS than EML4-ALK-negative patients (P = 0.04). However, in stage IIIA NSCLC, EML4-ALK-positive patients had poorer DFS than EML4-ALK-negative patients (P < 0.01). Moreover, multivariate analysis indicated that in stage IIIA NSCLC EML4-ALK fusion was the only significant indicator for poor DFS (P < 0.001). Furthermore, tumors with EML4-ALK fusions had significantly higher levels of ERCC1, a molecule with a key role in platinum drug efficacy, than tumors without EML4-ALK fusions.

Conclusion

EML4-ALK, PIK3CA, and MEK1 mutations occurred in NSCLC with various distinct clinicopathological characteristics. EML4-ALK fusions could serve as a significant prognostic indicator for locally advanced NSCLC.

introduction

Non-small-cell lung cancer (NSCLC) represents the major type (∼80%) of lung malignancies with a poor 5-year survival rate of ∼15%. Multiple genetic and epigenetic aberrations are involved in the development of NSCLC [1]. Of these abnormalities, oncogenic driver mutations cause constitutively activated signaling pathways that lead to uncontrolled cell growth and proliferation. A number of oncogenic driver mutations have been identified in NSCLC, including genes encoding for epidermal growth factor receptor (EGFR), K-ras (KRAS) and anaplastic lymphoma kinase (ALK), etc. These genetic aberrations provide specific molecular targets for therapeutic intervention, in addition to being of prognostic value. Several such targeted therapies, including EGFR tyrosine kinase inhibitors (erlotinib and gefitinib) and ALK inhibitor (crizotinib), have shown significant clinical efficacy in treating patients with NSCLC harboring corresponding gene mutations [2, 3]. While the prevalence and associated clinical characteristics of EGFR and KRAS mutation in NSCLC have been extensively studied in both western and eastern populations [4, 5], newly identified driver mutations such as ALK rearrangements in NSCLC are obtaining great amount of research attention [6–9]. Fully understanding of molecular aberrations of cancer is critical to personalizing treatment.

The fusion of ALK to the gene encoding echinoderm microtubule-associated protein-like 4 (EML4) as a result of inversion of chromosomal 2p leads to enhanced ALK expression and its kinase activity, which in turn activates several downstream signaling pathways including the Ras–extracellular signal-regulated kinase (ERK) pathway [10]. EML4-ALK fusion has been observed in 5.7% of western NSCLC and in 2.9%–6.7% Asian patients with NSCLC [6, 7]. Administration of ALK-specific inhibitor crizotinib to patients with NSCLC with EML4-ALK fusion resulted in an overall response rate of 57% and estimated probability of a 6-month progression-free survival rate of 72% [3].

Activating mutations in PIK3CA, the gene encoding the catalytic subunit p110α of phosphatidylinositol 3-kinase, have been found in many types of malignancies including NSCLC, breast cancer, and colorectal cancer [11, 12]. There are mainly five hot-spots of point mutations in PIK3CA, three in exon 9 and two in exon 20 [13]. As a result of these activating mutations, the downstream AKT pathway is deregulated, leading to enhanced MAPK (ERK1/ERK2) activities that promote cell growth and proliferation. Similarly, as the immediate upstream regulator of MAPK pathway, mutant MEK1 also results in aberrant MAPK pathway activities, often found in NSCLC [14]. Several small molecules specifically targeting mutant PIK3CA and MEK1 have entered into clinical trials and showed promising clinical efficacy [15, 16].

Previously we developed a quantitative PCR-based assay to address the prognosis issue of patients with early-stage nonsquamous NSCLC following surgical tumor resection [17]. To gain comprehensive understanding of oncogenic driver mutations and their associated clinicopathological characteristics in various stages of NSCLC, in the present study, we set out to investigate the mutations of ALK, PIK3CA, and MEK1, their coexistence with EGFR and KRAS mutation, and their relationship with clinical outcome in a large cohort of Chinese patients with NSCLC. Furthermore, to examine relationship between EML4-ALK fusion and platinum-based adjuvant chemotherapy efficacy, we investigated the expression levels of excision repair cross-complementation group 1 (ERCC1), a molecule with a key role in platinum drug efficacy, in EML4-ALK fusion-positive NSCLC.

patients and methods

patients and tumor tissue samples

Four hundred eighty-eight consecutive patients with NSCLC who underwent radical surgical resection of primary lung cancer were enrolled in this retrospective study. All of these patients were admitted into the First Affiliated Hospital of Guangzhou Medical University from January 2007 to December 2010 and had pathologically confirmed diagnosis. Tumors were staged pathologically according to the Union International Contre le Cancer (UICC-7) staging system for lung cancer [18]. No patients received any anticancer therapies before surgery. Most of the patients at locally advanced stage received two to four cycles of platinum-based adjuvant chemotherapy following surgery (platinum–gemcitabine, navelbine, taxol, docetaxol, or pemetrexed). Following tumor resection, the patients were followed up every 3 months in the first 2 years and every 6 months in the next 3 years. Tumor recurrence was identified using radiological examination or biopsy. Disease-free survival (DFS) was measured from the day of tumor resection until radiological/biopsy-confirmed tumor recurrence or death. Formalin-fixed and paraffin-embedded (FFPE) primary tumor tissues collected in surgical resection were used, and were evaluated by pathologists to meet the criteria of containing at least 50% of tumor cells. Patients with insufficient or poor-quality tissue for molecular analyses were excluded from this study. This study was approved by the Institutional Review Board of the First Affiliated Hospital, Guangzhou Medical University, Guangzhou, China.

mutation analysis

Genomic DNA and total RNA were extracted from FFPE tissues using QIAamp DNA FFPE Tissue Kit and RNeasy FFPE Kit (Qiagen, Germany), respectively. Mutations of EML4-ALK, PIK3CA, MEK1, EGFR, and KRAS were detected using commercially available kits from Amoy Diagnostics (Xiamen, China). All these kits were based on amplification refractory mutation system (ARMS) real-time PCR technology. The EML4-ALK kit detects nine fusions (variants 1, 2, 3a, 3b, 4, 4′, 5a, 5b, and 5′) [19, 20]. The PIK3CA kit detects five point mutations including E542K, E545K, E545D (all in exon 9), H1047R, and H1047L (both in exon 20). The MEK1 kit detects three point mutations at Q56P, K57N, and D67N (all in exon 2). The EGFR kit detects 29 mutations in exons 18–21, including T790M, L858R, L861Q, S768I, G719S, G719A, G719C, 3 insertions in exon 20, and 19 deletions in exon 19. The KRAS kit detects seven mutations including G12D, G12A, G12V, G12S, G12R, G12C, and G13D. For the detection of EML4-ALK fusion, mRNA was first reverse-transcribed to cDNA. For the detection of PIK3CA, MEK1, EGFR, and KRAS mutations, genomic DNA was used. All detections were carried out following the manufacturer's protocol. To validate the ARMS-PCR results, all samples were tested for these mutations using direct DNA sequencing method. Briefly, DNA (cDNA for EML4-ALK) was PCR-amplified using a commercial kit (TaKaRa Biotechnology, Dalian, China), and the PCR products were sequenced by Sangon Biotech Co., Ltd (Shanghai, China).

immunohistochemistry

Tumor FFPE sections were used for immunohistochemistry (IHC) using an automated immunostainer (Leica Microsystems, Germany). Briefly, slides were heated for antigen retrieval, and endogenous biotin was blocked using a commercial kit (Leica Microsystems). Following incubation with primary anti-ERCC1 antibody (Maixin, China), antibody binding was detected using a commercial kit (Leica Microsystems). A semiquantitative H score was calculated for each slide based on the staining intensity and the proportion of positive nuclei as described by Olaussen et al. [21]. Tumors with an H score exceeding 1.0 were deemed to be ERCC1 positive [21]. All IHC slides were evaluated independently by two pathologists (PH and JJ).

statistic analysis

Fisher's exact test was used to assess the association between genotype and clinical characteristics. The Kaplan–Meier method was used to compare the DFS of patients with different genotypes. Cox multivariate proportional hazard model was used for survival analysis, and the hazard ratio (HR) and 95% confidence interval (CI) were calculated. The correlation between EML4-ALK fusion and ERCC1 positivity was examined using Pearson's chi-square method. All analyses were carried out using PASW Statistics 18.0 (IBM, Chicago, IL).

results

patient characteristics

There were 295 male and 193 female patients with NSCLC enrolled in this study, with median age of 60 years old (ranging from 18 to 85 years). Histologically, there were 349 adenocarcinomas, 101 squamous-cell carcinomas (SCC), and 38 other types (adenosquamous, large-cell, and sarcomatoid carcinomas). There were 258 patients received platinum-based adjuvant chemotherapy (platinum–gemcitabine, navelbine, taxol, docetaxol, or pemetrexed). The clinicopathological characteristics of patients are summarized in Table 1.

Table 1.

Clinical characteristics of patients with NSCLC

Patients (N = 488)
 
Characteristics N (%) 
Age (years) 
 ≥60 251 (51.4) 
 <60 237 (48.6) 
Sex 
 Male 295 (60.5) 
 Female 193 (39.5) 
Stage 
 IA 134 (27.5) 
 IB 136 (27.9) 
 IIA 6 (1.2) 
 IIB 43 (8.8) 
 IIIA 165 (33.8) 
 IIIB 1 (0.2) 
 IV 3 (0.6) 
Histopathology 
 ADa 349 (71.5) 
 SCCb 101 (20.7) 
 Othersc 38 (7.8) 
Differentiation 
 Poor 127 (26.0) 
 Good 337 (69.1) 
 Undetermined 24 (4.9) 
Smoking status 
 No 257 (52.7) 
 Yes 156 (32.0) 
 Unknown 75 (15.3) 
Adjuvant chemotherapyd 
 IA 30 (11.6) 
 IB 66 (25.6) 
 IIA 6 (2.3) 
 IIB 29 (11.2) 
 IIIA 124 (48.1) 
 IIIB 1 (0.4) 
 IV 2 (0.8) 
Patients (N = 488)
 
Characteristics N (%) 
Age (years) 
 ≥60 251 (51.4) 
 <60 237 (48.6) 
Sex 
 Male 295 (60.5) 
 Female 193 (39.5) 
Stage 
 IA 134 (27.5) 
 IB 136 (27.9) 
 IIA 6 (1.2) 
 IIB 43 (8.8) 
 IIIA 165 (33.8) 
 IIIB 1 (0.2) 
 IV 3 (0.6) 
Histopathology 
 ADa 349 (71.5) 
 SCCb 101 (20.7) 
 Othersc 38 (7.8) 
Differentiation 
 Poor 127 (26.0) 
 Good 337 (69.1) 
 Undetermined 24 (4.9) 
Smoking status 
 No 257 (52.7) 
 Yes 156 (32.0) 
 Unknown 75 (15.3) 
Adjuvant chemotherapyd 
 IA 30 (11.6) 
 IB 66 (25.6) 
 IIA 6 (2.3) 
 IIB 29 (11.2) 
 IIIA 124 (48.1) 
 IIIB 1 (0.4) 
 IV 2 (0.8) 

aAdenocarcinoma.

bSquamous cell carcinoma.

cIncluding 18 adenosquamous, 12 large-cell, and 8 sarcomatoid carcinomas.

dAll chemotherapy were platinum based.

oncogenic driver mutations in NSCLC

In 488 NSCLC examined for EML4-ALK fusions using ARMS-PCR, we found 28 (5.7%) tumors harboring various forms of fusion. In stage IA patients (N = 134), the frequency of EML4-ALK fusions was 9.0%; and in stage IIIA patients (N = 165), the frequency was 5.5%. The major form of EML4-ALK fusion was E13;A20 (variant 1, 50%), followed by E6;A20 (variant 3a/3b, 29%), E20;A20 (variant 3, 14%), and E18;A20 (variant 5′, 7%) (Table 2). In 392 NSCLC specimens with sufficient DNA for PIK3CA mutation testing, 13 were positive (3.3%), including one at stage IA (1.7%) and eight at stage IIIA (5.1%). About half of the mutations occurred in exon 9 (E545K and E542K) and half in exon 20 (H1047L and H1047R) (Table 2). There were three samples detected with MEK1 mutations in 392 tumors (0.8%), one at stage IB and two at stage IIIA (Table 3). The results obtained using ARMS-PCR were consistent with that obtained using direct DNA sequencing, except one EML4-ALK fusion-positive sample detected using PCR was negative by sequencing analysis.

Table 2.

The mutation types of EML4-ALK, PIK3CA, and MEK1 in patients with NSCLC

EML4-ALK
 
PIK3CA
 
MEK1
 
Mutation N (%) Mutation N (%) Mutation N (%) 
E13;A20 14 (50) E545K 4 (31) K57N 2 (67) 
E6;A20 8 (29) E542K 2 (15) D67N 1 (33) 
E20;A20 4 (14) H1047L 3 (23)   
E18;A20 2 (7) H1047R 4 (31)   
EML4-ALK
 
PIK3CA
 
MEK1
 
Mutation N (%) Mutation N (%) Mutation N (%) 
E13;A20 14 (50) E545K 4 (31) K57N 2 (67) 
E6;A20 8 (29) E542K 2 (15) D67N 1 (33) 
E20;A20 4 (14) H1047L 3 (23)   
E18;A20 2 (7) H1047R 4 (31)   
Table 3.

Clinical characteristics of patients with NSCLC with ALK, PIK3CA, or MEK1 mutations

Characteristics EML4-ALK
 
PIK3CA
 
MEK1a
 
Mutation, N (%) Wild type, N (%) P Mutation, N (%) Wild type, N (%) P value Mutation, N (%) Wild type, N (%) 
Age, years 
 ≥60 9 (3.6) 240 (96.4) 0.040 5 (2.5) 199 (97.5) 0.319 2 (0.1) 202 (99.9) 
 <60 19 (7.9) 220 (92.1) 8 (4.3) 180 (95.7) 1 (0.5) 187 (99.5) 
Sex 
 Male 9 (3.1) 286 (96.9) 0.002 9 (3.8) 229 (96.2) 0.523 2 (0.8) 236 (99.2) 
 Female 19 (9.8) 174 (90.2) 4 (2.6) 150 (97.4) 1 (0.6) 153 (99.4) 
Stage 
 IA 12 (9.0) 122 (91.0) 0.036b 1 (1.7) 58 (98.3) 0.053c 59 
 IB 4 (2.9) 132 (97.1) 2 (1.5) 133 (98.5) 1 (0.7) 134 (99.3) 
 IIA 2 (33.3) 4 (66.7) 
 IIB 1 (2.3) 42 (97.7) 2 (5.0) 38 (95.0) 40 
 IIIA 9 (5.5) 156 (94.5) 8 (5.1) 150 (94.9) 2 (1.3) 156 (98.7) 
 IIIB     
 IV     
Histopathology 
 AD 25 (7.2) 324 (92.8) 0.053d 7 (2.7) 251 (97.3) 0.505 1 (0.4) 257 (99.6) 
 SCC 2 (2.0) 99 (98.0) 4 (4.1) 94 (95.9) 2 (2.0) 96 (98.0) 
 AD-SCCe 1 ( 5.6) 17 (94.4) 2 (11.1) 16 (88.9) 18 
 Others 20 18 18 
Differentiation 
 Poor 8 (6.3) 119 (93.7) 0.594 3 (2.9) 101 (97.1) 0.774 3 (2.9) 101 (97.1) 
 Good 17 (5.0) 320 (95.0) 10 (3.5) 278 (96.5) 288 
 Unidentified 3 (12.5) 21 (87.5)     
Smoking status 
 No 21 (8.2) 236 (91.8) 0.021f 7 (3.3) 207 (96.7) 0.732 1 (0.5) 213 (99.5) 
 Yes 4 (2.6) 152 (97.4) 6 (4.6) 124 (95.4) 2 (1.5) 128 (98.5) 
 Unknown 3 (4.0) 72 (96.0) 48 48 
Characteristics EML4-ALK
 
PIK3CA
 
MEK1a
 
Mutation, N (%) Wild type, N (%) P Mutation, N (%) Wild type, N (%) P value Mutation, N (%) Wild type, N (%) 
Age, years 
 ≥60 9 (3.6) 240 (96.4) 0.040 5 (2.5) 199 (97.5) 0.319 2 (0.1) 202 (99.9) 
 <60 19 (7.9) 220 (92.1) 8 (4.3) 180 (95.7) 1 (0.5) 187 (99.5) 
Sex 
 Male 9 (3.1) 286 (96.9) 0.002 9 (3.8) 229 (96.2) 0.523 2 (0.8) 236 (99.2) 
 Female 19 (9.8) 174 (90.2) 4 (2.6) 150 (97.4) 1 (0.6) 153 (99.4) 
Stage 
 IA 12 (9.0) 122 (91.0) 0.036b 1 (1.7) 58 (98.3) 0.053c 59 
 IB 4 (2.9) 132 (97.1) 2 (1.5) 133 (98.5) 1 (0.7) 134 (99.3) 
 IIA 2 (33.3) 4 (66.7) 
 IIB 1 (2.3) 42 (97.7) 2 (5.0) 38 (95.0) 40 
 IIIA 9 (5.5) 156 (94.5) 8 (5.1) 150 (94.9) 2 (1.3) 156 (98.7) 
 IIIB     
 IV     
Histopathology 
 AD 25 (7.2) 324 (92.8) 0.053d 7 (2.7) 251 (97.3) 0.505 1 (0.4) 257 (99.6) 
 SCC 2 (2.0) 99 (98.0) 4 (4.1) 94 (95.9) 2 (2.0) 96 (98.0) 
 AD-SCCe 1 ( 5.6) 17 (94.4) 2 (11.1) 16 (88.9) 18 
 Others 20 18 18 
Differentiation 
 Poor 8 (6.3) 119 (93.7) 0.594 3 (2.9) 101 (97.1) 0.774 3 (2.9) 101 (97.1) 
 Good 17 (5.0) 320 (95.0) 10 (3.5) 278 (96.5) 288 
 Unidentified 3 (12.5) 21 (87.5)     
Smoking status 
 No 21 (8.2) 236 (91.8) 0.021f 7 (3.3) 207 (96.7) 0.732 1 (0.5) 213 (99.5) 
 Yes 4 (2.6) 152 (97.4) 6 (4.6) 124 (95.4) 2 (1.5) 128 (98.5) 
 Unknown 3 (4.0) 72 (96.0) 48 48 

aFor MEK1 mutations, no statistical analysis was carried out due to the limited sample size.

bTumors at stage IA versus stage IB.

cTumors at stages IA + IB versus stages IIB + IIIA.

dAD versus SCC.

eAdenosquamous carcinoma.

fSmoker versus nonsmoker patients.

P value in bold indicates a significant difference.

In NSCLC, the frequency of EML4-ALK fusions varied in different age, sex, smoking status, and histological type subgroups (Table 3). In patients younger than 60 years old, there were 7.9% of tumors harboring EML4-ALK fusions, significantly greater than that in patients at or older than 60 years old (3.6%) (P < 0.05). In female patients, 9.8% had EML4-ALK fusions, significantly greater than that in male patients (3.1%) (P<0.05). In nonsmoker patients, 8.2% had EML4-ALK fusions, comparing with 2.6% in smoker patients (P < 0.05). There were 7.2% of adenocarcinomas harboring EML4-ALK fusions. In contrast, only 2.0% of SCC contained EML4-ALK fusions. Pathological examination of EML4-ALK fusions-positive tumors revealed that 69% of them were invasive mucinous adenocarcinomas (Figure 1). In addition, there was one case of adenosquamous carcinoma being EML4-ALK fusion positive.

Figure 1.

The histopathology of adenocarcinoma of the lung with EML4-ALK fusion. A representative H&E-stained FFPE tumor specimen shows an invasive mucinous pattern of tumor cells (magnification ×400).

Figure 1.

The histopathology of adenocarcinoma of the lung with EML4-ALK fusion. A representative H&E-stained FFPE tumor specimen shows an invasive mucinous pattern of tumor cells (magnification ×400).

In NSCLC tumors, we found that EML4-ALK, PIK3CA, and MEK1 mutations were mutually exclusive. However, one EML4-ALK fusion-positive tumor was found harboring EGFR activating mutation (L858R), and another EML4-ALK fusion-positive tumor was found harboring KRAS mutation (G12A). The concurrent EGFR mutation and EML4-ALK fusion occurred in a female, 64-year-old, nonsmoker patient with adenocarcinoma at stage IA. The concurrent KRAS mutation and EML4-ALK fusion occurred in a male, 57-year-old patient with adenocarcinoma at stage IIA (smoking history unavailable).

Interestingly, unlike EML4-ALK fusions that occurred more often in early stages than late stages of NSCLC, PIK3CA mutations occurred more often in late stages (IIB and IIIA) than in early stages (IA and IB) NSCLC (P = 0.011). In smoker, male, or patients with SCC NSCLC, PIK3CA mutations rates were greater than that in nonsmoker, female, or patients with adenocarcinoma, though the differences did not reach statistical significance (Table 3). As a relatively rare event, MEK1 mutations were detected in three cases of 392 NSCLC: two males and one female, two SCC and one adenocarcinoma, and two nonsmokers and one smoker (Table 3).

clinical outcome of patients with NSCLC with oncogenic driver mutations

To elucidate the relationship between the occurrence of oncogenic driver mutations and clinical outcome of patients with NSCLC, we analyzed the DFS of patients with or without the gene mutations analyzed. In stage IA NSCLC, all of the 12 (100%) EML4-ALK-positive patients were still alive without relapse; however, 29.5% of EML4-ALK-negative patients had recurrences of lung cancer within 5 years following tumor resection surgery. EML4-ALK-positive patients had significant longer DFS than EML4-ALK-negative ones (P = 0.04) (Figure 2A). Interestingly, in stage IIIA NSLCL, EML4-ALK-positive patients had poorer DFS than those EML4-ALK-negative patients (median DFS 6 versus 16 months, P = 0.0057) (Figure 2B). For PIK3CA mutation, because of the small sample size of mutation-positive patients in early stage of NSCLC, we only analyzed the survival of patients in stage IIIA. The result showed that there was no significant difference in DFS between patients with PIK3CA mutant and wild-type NSCLC (Figure 3A). Of the three patients with MEK1 mutations, one case at stage IB without adjuvant chemotherapy had no relapse for 26 months after tumor resection, and two cases at stage IIIA relapsed 2 and 18 months after surgery, respectively.

Figure 2.

The disease-free survival (DFS) of patients with NSCLC at various stages with or without EML4-ALK fusions. (A) Kaplan–Meier survival analysis of patients with stage IA NSCLC with versus without EML4-ALK fusions (N = 133). EML4-ALK fusion-positive patients had statistically significant longer DFS (log-rank test, P = 0.04). (B) Kaplan–Meier survival analysis of patients with stage IIIA NSCLC with versus without EML4-ALK fusions (N = 152). EML4-ALK fusion-negative patients had statistically significant longer DFS (log-rank test, P = 0.006).

Figure 2.

The disease-free survival (DFS) of patients with NSCLC at various stages with or without EML4-ALK fusions. (A) Kaplan–Meier survival analysis of patients with stage IA NSCLC with versus without EML4-ALK fusions (N = 133). EML4-ALK fusion-positive patients had statistically significant longer DFS (log-rank test, P = 0.04). (B) Kaplan–Meier survival analysis of patients with stage IIIA NSCLC with versus without EML4-ALK fusions (N = 152). EML4-ALK fusion-negative patients had statistically significant longer DFS (log-rank test, P = 0.006).

Figure 3.

(A) The survival analysis of patients with stage IIIA NSCLC with versus without PIK3CA mutations (N = 152). There was no statistically significant difference in patients' disease-free survival (DFS) regardless of their tumors' PIK3CA mutation status. (B) The DFS of patients with stage IIIA NSCLC with PIK3CA mutations versus those with EML4-ALK fusions (N = 17). Patients with PIK3CA mutations had significant better DFS than those with EML4-ALK fusions (log-rank test, P = 0.02).

Figure 3.

(A) The survival analysis of patients with stage IIIA NSCLC with versus without PIK3CA mutations (N = 152). There was no statistically significant difference in patients' disease-free survival (DFS) regardless of their tumors' PIK3CA mutation status. (B) The DFS of patients with stage IIIA NSCLC with PIK3CA mutations versus those with EML4-ALK fusions (N = 17). Patients with PIK3CA mutations had significant better DFS than those with EML4-ALK fusions (log-rank test, P = 0.02).

Because of the similar sample sizes for EML4-ALK and PIK3CA mutation-positive patients in stage IIIA NSCLC, we examined the differences in DFS for these two subgroups of patients. The results showed that EML4-ALK-positive patients were associated with poorer DFS compared with PIK3CA mutation-positive patients (6 versus 21 months, P = 0.02) (Figure 3B).

To test the prognostic value of these oncogenic driver mutations, we used multivariate Cox proportional hazard model to analyze the significance of gene mutations in patient's survival. The factors included in the model were sex, age, smoking status, histology type, adjuvant chemotherapy, and ALK or PIK3CA mutation status. The results revealed that in stage IIIA NSCLC, EML4-ALK fusion was the only significant prognostic factor for poor survival (HR = 4.0, 95% CI 1.8–8.6; P < 0.001). No significant prognostic factor was identified in any other stages of NSCLC (supplementary Tables S1 and S2, available at Annals of Oncology online).

the association between EML4-ALK fusion and ERCC1 expression

Because of the significant difference in survival for patients with locally advanced NSCLC with or without EML4-ALK fusions, and most of these patients received platinum-based adjuvant chemotherapy, we examined ERCC1 expression levels in tumors with or without EML4-ALK fusions. ERCC1 protein was mainly detected in the nuclei of tumor cells as revealed by IHC. In EML4-ALK-positive tumors that had sufficient tissues for IHC study (N = 20), 14 of them (70%) were ERCC1 positive. In EML4-ALK-negative tumors that had sufficient tissues for IHC study (N = 360), 157 of them (43.6%) were ERCC1 positive. EML4-ALK-positive tumors had significantly higher levels of ERCC1 than EML4-ALK-negative tumors (P = 0.021) (supplementary Table S1, available at Annals of Oncology online).

discussion

Molecular profiling of cancer is crucial for tailoring therapy for patients. A number of genetic abnormalities have been clinically proven to be excellent therapeutic targets or indicators for drug resistance in NSCLC. In the present study, we investigated the mutations of multiple oncogenes in a relatively large group of patients with NSCLC, and particularly analyzed their associations with clinical outcome of patients at different stages of this malignance.

EML4-ALK rearrangements represent a small but important fraction of oncogenic driver mutations in NSCLC. Similar to previous studies with limited number of patients, our present data showed that EML4-ALK rearrangements occurred more frequently in female, younger age, and nonsmoker patients. The mutation frequency of ALK (5.7%) found in the present study was close to that reported for Japanese and Korean NSCLC, but much lesser than that reported for the western patients with NSCLC (11.3%) [6, 22, 23].

Previous studies have mainly focused on studying ALK rearrangements in adenocarcinomas of the lung, but little has been known in SCC. In current study, we identified that 2% of SCC of the lung harbored EML4-ALK fusions, suggesting that it should not be ignored to detect ALK aberrations in this histological type of NSCLC. Pathologically, we found that 69.5% of EML4-ALK fusion-positive adenocarcinomas showed mucinous pattern, a histological feature of NSCLC for EML4-ALK fusions and with a poor clinical outcome [24, 25]. Although there has been a consensus that in NSCLC, ALK rearrangements occur mutually exclusively with EGFR or KRAS mutations, we found two cases of coexistence of EML4-ALK rearrangements with EGFR or KRAS mutations. Unfortunately, we cannot specify whether the coexistence of these oncogenic mutations occurred in the same or different tumor cells. Nevertheless, in addition to reflecting the complexity nature of tumors, our data might suggest the necessity of simultaneously administrating multiple targeting drugs in treating a single tumor, which harbors multiple oncogenic driver mutations.

The occurrence of oncogenic driver mutations in different stages of malignance may have different clinical significances. In the present study, the majority of EML4-ALK fusion-positive patients were in stage IA and IIIA. In stage IA NSCLC, patients with EML4-ALK fusions had a better DFS than those without EML4-ALK fusions. Surprisingly, it was totally opposite for stage IIIA patients in that those with EML4-ALK fusions had shorter DFS than those without EML4-ALK fusions, and EML4-ALK fusion was the only independent prognostic factor for poor survival. These results were different from those reported by Fukui et al. [8] and Shaw et al. [9], which did not find significant differences in DFS or overall survival between patients with ALK fusion-positive and ALK fusion-negative NSCLC. This discrepancy could be attributed to the fact that, unlike their studies, our study stratified patients into specific stages for the analysis of DFS.

In the present study, we found that EML4-ALK-positive tumors had higher levels of ERCC1 than EML4-ALK-negative tumors. ERCC1 is a key molecule involved in DNA damage repair. Several lines of evidences have suggested that high levels of ERCC1 are associated with longer survival of patients with in early-stage NSCLC but resistance to platinum-based chemotherapy in patients with late-stage NSCLC [26, 27]. These are in agreement with the findings of present study that, comparing to EML4-ALK-negative patients, EML4-ALK-positive stage IA patients had better survival, but EML4-ALK-positive stage IIIA patients had poorer survival. Nevertheless, the mechanisms underlying the relationship between ALK rearrangement and ERCC1 expression need to be further investigated. It would also be desirable to validate our results using an independent, large cohort of patients with NSCLC.

Previously Ludovini et al. reported that in advanced-stage NSCLC, PIK3CA mutation was an indicator of poor survival and resistance to EGFR TKI [28]. In the present study, however, we did not find PIK3CA mutation as a significant prognostic factor for patients with stage IIIA NSCLC. While the differences in clinical settings between our study and theirs may have contributed to the discrepancy in results, the prognostic significance of PIK3CA mutation in NSCLC needs to be further investigated.

As a rare event in NSCLC, MEK1 mutations occurred in three (0.8%) tumors in our study. Two of these three tumors were SCC. Together with the data for PIK3CA mutations, our results may indicate that there could be a preference in oncogenic driver mutations in the development of adenocarcinoma versus SCC. Moreover, as several specific drugs against mutant PIK3CA and MEK1 are currently in clinical trials (www.clinicaltrial.gov), it is expected that, as for EGFR mutations, detection for PIK3CA and MEK1 mutations in NSCLC will become imperative in choosing appropriate targeting therapeutics for treating this malignance.

In summary, in the present study, we found that a set of oncogenic driver mutations defined the corresponding subsets of NSCLC with distinct clinicopathological characteristics. Particularly, EML4-ALK fusions could serve as a significant prognostic biomarker for locally advanced stage NSCLC. Our study would offer help in further understanding the molecular aberrations of NSCLC, which is critical to personalizing management for this malignance.

funding

This work was partly supported by the National Natural Science Foundation of China [grant number 81172251 to JXZ] and the Fund for Innovative Research Team of Zhejiang Education Department [T200907 to JXZ]. JXZ is supported by the Qianjiang Scholar Professorship program, Zhejiang Province, China.

disclosure

The authors have declared no conflicts of interest.

references

1
Sato
M
Shames
DS
Gazdar
AF
, et al.  . 
A translational view of the molecular pathogenesis of lung cancer
J Thorac Oncol
 , 
2007
, vol. 
2
 (pg. 
327
-
343
)
2
Lynch
TJ
Bell
DW
Sordella
R
, et al.  . 
Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib
N Engl J Med
 , 
2004
, vol. 
350
 (pg. 
2129
-
2139
)
3
Kwak
EL
Bang
YJ
Camidge
DR
, et al.  . 
Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer
N Engl J Med
 , 
2010
, vol. 
363
 (pg. 
1693
-
1703
)
4
Eberhard
DA
Johnson
BE
Amler
LC
, et al.  . 
Mutations in the epidermal growth factor receptor and in KRAS are predictive and prognostic indicators in patients with non-small-cell lung cancer treated with chemotherapy alone and in combination with erlotinib
J Clin Oncol
 , 
2005
, vol. 
23
 (pg. 
5900
-
5909
)
5
Tam
IY
Chung
LP
Suen
WS
, et al.  . 
Distinct epidermal growth factor receptor and KRAS mutation patterns in non-small cell lung cancer patients with different tobacco exposure and clinicopathologic features
Clin Cancer Res
 , 
2006
, vol. 
12
 (pg. 
1647
-
1653
)
6
Jin
G
Jeon
HS
Lee
EB
, et al.  . 
EML4-ALK fusion gene in Korean non-small cell lung cancer
J Korean Med Sci
 , 
2012
, vol. 
27
 (pg. 
228
-
230
)
7
Takeuchi
K
Soda
M
Togashi
Y
, et al.  . 
RET, ROS1 and ALK fusions in lung cancer
Nat Med
 , 
2012
, vol. 
18
 (pg. 
378
-
381
)
8
Fukui
T
Yatabe
Y
Kobayashi
Y
, et al.  . 
Clinicoradiologic characteristics of patients with lung adenocarcinoma harboring EML4-ALK fusion oncogene
Lung Cancer
 , 
2012
, vol. 
77
 (pg. 
319
-
325
)
9
Shaw
AT
Yeap
BY
Solomon
BJ
, et al.  . 
Effect of crizotinib on overall survival in patients with advanced non-small-cell lung cancer harbouring ALK gene rearrangement: a retrospective analysis
Lancet Oncol
 , 
2011
, vol. 
12
 (pg. 
1004
-
1012
)
10
Takezawa
K
Okamoto
I
Nishio
K
, et al.  . 
Role of ERK-BIM and STAT3-survivin signaling pathways in ALK inhibitor-induced apoptosis in EML4-ALK-positive lung cancer
Clin Cancer Res
 , 
2011
, vol. 
17
 (pg. 
2140
-
2148
)
11
Samuels
Y
Wang
Z
Bardelli
A
, et al.  . 
High frequency of mutations of the PIK3CA gene in human cancers
Science
 , 
2004
, vol. 
304
 pg. 
554
 
12
Berns
K
Horlings
HM
Hennessy
BT
, et al.  . 
A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer
Cancer Cell
 , 
2007
, vol. 
12
 (pg. 
395
-
402
)
13
Yamamoto
H
Shigematsu
H
Nomura
M
, et al.  . 
PIK3CA mutations and copy number gains in human lung cancers
Cancer Res
 , 
2008
, vol. 
68
 (pg. 
6913
-
6921
)
14
Marks
JL
Gong
Y
Chitale
D
, et al.  . 
Novel MEK1 mutation identified by mutational analysis of epidermal growth factor receptor signaling pathway genes in lung adenocarcinoma
Cancer Res
 , 
2008
, vol. 
68
 (pg. 
5524
-
5528
)
15
Friday
BB
Adjei
AA
Advances in targeting the Ras/Raf/MEK/Erk mitogen-activated protein kinase cascade with MEK inhibitors for cancer therapy
Clin Cancer Res
 , 
2008
, vol. 
14
 (pg. 
342
-
346
)
16
Zhang
J
Yang
PL
Gray
NS
Targeting cancer with small molecule kinase inhibitors
Nat Rev Cancer
 , 
2009
, vol. 
9
 (pg. 
28
-
39
)
17
Kratz
JR
He
J
Van Den Eeden
SK
, et al.  . 
A practical molecular assay to predict survival in resected non-squamous, non-small-cell lung cancer: development and international validation studies
Lancet
 , 
2012
, vol. 
379
 (pg. 
823
-
832
)
18
Kassis
ES
Vaporciyan
AA
Swisher
SG
, et al.  . 
Application of the revised lung cancer staging system (IASLC Staging Project) to a cancer center population
J Thorac Cardiovasc Surg
 , 
2009
, vol. 
138
 (pg. 
412
-
418
)
19
Sasaki
T
Rodig
SJ
Chirieac
LR
, et al.  . 
The biology and treatment of EML4-ALK non-small cell lung cancer
Eur J Cancer
 , 
2010
, vol. 
46
 (pg. 
1773
-
1780
)
20
Wong
DW
Leung
EL
So
KK
, et al.  . 
The EML4-ALK fusion gene is involved in various histologic types of lung cancers from nonsmokers with wild-type EGFR and KRAS
Cancer
 , 
2009
, vol. 
115
 (pg. 
1723
-
1733
)
21
Olaussen
KA
Dunant
A
Fouret
P
, et al.  . 
DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy
N Engl J Med
 , 
2006
, vol. 
355
 (pg. 
983
-
991
)
22
Takahashi
T
Sonobe
M
Kobayashi
M
, et al.  . 
Clinicopathologic features of non-small-cell lung cancer with EML4-ALK fusion gene
Ann Surg Oncol
 , 
2010
, vol. 
17
 (pg. 
889
-
897
)
23
Lin
E
Li
L
Guan
Y
, et al.  . 
Exon array profiling detects EML4-ALK fusion in breast, colorectal, and non-small cell lung cancers
Mol Cancer Res
 , 
2009
, vol. 
7
 (pg. 
1466
-
1476
)
24
Jokoji
R
Yamasaki
T
Minami
S
, et al.  . 
Combination of morphological feature analysis and immunohistochemistry is useful for screening of EML4-ALK-positive lung adenocarcinoma
J Clin Pathol
 , 
2010
, vol. 
63
 (pg. 
1066
-
1070
)
25
Carretta
A
Canneto
B
Calori
G
, et al.  . 
Evaluation of radiological and pathological prognostic factors in surgically-treated patients with bronchoalveolar carcinoma
Eur J Cardiothorac Surg
 , 
2001
, vol. 
20
 (pg. 
367
-
371
)
26
Roth
JA
Carlson
JJ
Prognostic role of ERCC1 in advanced non-small-cell lung cancer: a systematic review and meta-analysis
Clin Lung Cancer
 , 
2011
, vol. 
12
 (pg. 
393
-
401
)
27
Jiang
J
Liang
X
Zhou
X
, et al.  . 
ERCC1 expression as a prognostic and predictive factor in patients with non-small cell lung cancer: a meta-analysis
Mol Biol Rep
 , 
2012
, vol. 
39
 (pg. 
6933
-
6942
)
28
Ludovini
V
Bianconi
F
Pistola
L
, et al.  . 
Phosphoinositide-3-kinase catalytic alpha and KRAS mutations are important predictors of resistance to therapy with epidermal growth factor receptor tyrosine kinase inhibitors in patients with advanced non-small cell lung cancer
J Thorac Oncol
 , 
2011
, vol. 
6
 (pg. 
707
-
715
)

Author notes

These authors contributed equally.