KIF5B-RET Fusions in Chinese Patients With Non Small Cell Lung Cancer

Transcription

1 KIF5B-RET Fusions in Chinese Patients With Non Small Cell Lung Cancer Weijing Cai, MD 1 ; Chunxia Su, MD 1 ; Xuefei Li, PhD 1 ; Lihong Fan, MD, PhD 1 ; Limou Zheng, PhD 2 ; Ke Fei, MD, PhD 3 ; and Caicun Zhou, MD, PhD 1 BACKGROUND: It has been established that ret proto-oncogene (RET) fusions are oncogenic drivers in non small cell lung cancer (NSCLC). The prevalence and clinicopathologic characteristics of RET fusions in Chinese patients with NSCLC remain unclear. The objective of the current study was to determine the prevalence and clinicopathologic characteristics of KIF5B-RET fusions (fusions of the RET and kinesin family member 5B [KIF5B] genes) in Chinese patients with NSCLC. METHODS: The authors screened for KIF5B- RET fusions in 392 patients with NSCLC using multiplex real-time polymerase chain reaction assay and validated all positive samples using direct sequencing. The relations between KIF5B-RET fusions and clinicopathologic characteristics were analyzed. RESULTS: In total, 6 patients (1.5%) were identified who harbored KIF5B-RET fusions. Of these, 4 had adenocarcinoma, 1 had a malignant neuroendocrine tumor, and 1 had squamous cell carcinoma. All patients who were positive for a KIF5B-RET fusion were never-smokers. There was no statistically significant difference in age, sex, smoking status, pathologic stage, or histologic type between patients with and without KIF5B-RET fusions. Patients without KIF5B-RET fusions had a better prognosis than those with KIF5B-RET fusions (median survival, 52.6 months vs 21.0 months; P ¼.06), with a hazard ratio of (95% confidence interval, ; P ¼.055) on multivariate analysis. Disease stage (hazard ratio, 2.879) and younger age (<65 years; hazard ratio, 1.485) were identified as independent prognostic factors for better survival. CONCLUSIONS: KIF5B-RET fusions were quite rare, with a prevalence of approximately 1.5% in Chinese patients with NSCLC, and they were a little more common in patients with adenocarcinoma than in those with squamous carcinoma (1.73% vs 0.84%). In addition, KIF5B-RET fusions also existed in patients with low-grade malignant neuroendocrine tumors. Cancer 2013;119: VC 2013 American Cancer Society. KEYWORDS: KIF5B-RET, fusion, rearrangement, oncogenic driver, non small cell lung cancer. INTRODUCTION Lung cancer is the leading cause of cancer death worldwide. 1 Although chemotherapy is the main strategy for the treatment of patients with advanced non small cell lung cancer (NSCLC), its efficacy remains unsatisfactory. Some genes with somatic mutations, such as mutated epidermal growth factor receptor (EGFR), v-ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KARS), v-erb-b2 erythroblastic leukemia viral oncogene homolog 2 (HER2), etc, have been proven to be driver genes in some subgroups of lung cancer. Gene translocation or fusions play an important role in carcinogenesis of solid tumors, such as fusions of the echinoderm microtubule associated protein like 4 (EML4) and anaplastic lymphoma receptor tyrosine kinase (ALK) genes (EML4-ALK) in lymphoma and c-ros oncogene 1 receptor tyrosine kinase (ROS1) rearrangement in glioblastoma. Recently, these fusions have been identified in NSCLC. EML4-ALK fusions have been identified in NSCLC and have become useful targets for NSCLC. 2-5 To date, 2 additional gene rearrangements involving ROS1 6 and ret proto-oncogene (RET), 7-10 respectively, also have been identified in NSCLC. It is noteworthy that ROS1 rearrangement has been detected in lung squamous cell carcinoma by Davies and his colleagues. 11 Furthermore, in a phase 1 clinical trial, patients with ROS1 fusions and adenocarcinoma had a good response to crizotinib. 12 Of the fusions that have been identified in NSCLC to date, RET fusion is the latest 1 and is rarely studied. RET is the receptor for members of the glial cell line-derived neurotrophic factor family of extracellular signaling molecules or ligands. 13 RET gain-of-function mutations are associated with the development of various types of human cancer, including medullary thyroid carcinoma, multiple endocrine neoplasias type 2A and 2B, pheochromocytoma, and Corresponding authors: Corresponding authors: Ke Fei, MD, PhD, Department of Thoracic Surgery, Shanghai Pulmonary Hospital, Tongji University School of Medicine, No. 507, Zheng Min Road, Shanghai, , P.R. China; Fax: (011) ; ffeik@yahoo.cn; Caicun Zhou, MD, PhD, Department of Medical Oncology, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Tongji University Medical School Cancer Institute, No. 507 Zheng Min Road, Shanghai, , P.R. China; Fax: (011) ; caicunzhou@yahoo.com.cn 1 Department of Medical Oncology, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Tongji University Medical School Cancer Institute, Shanghai, People s Republic of China; 2 Translational Medical Center, Xiamen University, Xiamen, People s Republic of China; 3 Department of Thoracic Surgery, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Shanghai, People s Republic of China We thank Amoy Diagnostics Company, Ltd. (Xiamen, China) for technical support. The first 2 authors contributed equally to this article. DOI: /cncr.27940, Received: October 10, 2012; Revised: November 20, 2012; Accepted: November 26, 2012, Published online February 1, 2013 in Wiley Online Library (wileyonlinelibrary.com) 1486 Cancer April 15, 2013

2 KIF5B-RET Fusions in NSCLC/Cai et al parathyroid hyperplasia. RET/papillary thyroid carcinoma (RET/PTC) rearrangement was first detected in papillary thyroid carcinoma 14 and subsequently was proven to be oncogenic. 15,16 RET fusions were identified in 13% to 44% of sporadic papillary thyroid carcinomas and in 50% to 90% of radiation-associated papillary thyroid carcinomas. 17 Recently, 2 fusion partners for RET, kinesin family member 5B (KIF5B) and coiled-coil domain containing 6 (CCDC6), have been identified in lung adenocarcinoma; and it has been demonstrated that RET fusions are novel oncogenic drivers and potential targets for existing small-molecule tyrosine kinase inhibitors (TKIs), including sorafenib, sunitinib, and vandetanib. 7-9,18 However, the prevalence and clinicopathologic characteristics of RET fusions in lung cancer remain unclear, especially among Chinese patients. In fact, most studies have screened for RET fusions in selected populations of patients with NSCLC; and, to date, RET fusions have been identified only in patients with lung adenocarcinoma. In a recent Japanese study 7 in which 433 patients with lung adenocarcinoma were screened, 7 patients with KIF5B-RET fusions were identified, including 6 Japanese patients and 1 American patient. In addition, no RET fusions were identified in patients who had nonadenocarcinomas in another large study in which 1482 patients with NSCLC were screened (including 1119 adenocarcinomas). 18 The above-mentioned studies, which were published earlier this year, indicated that the prevalence of RET fusions was approximately 0.9% in NSCLC and 1.2% to 1.9% in lung adenocarcinomas. 7,18 Therefore, because KIF5B-RET has been identified as the most common fusion gene in lung cancer, we screened for known KIF5B-RET fusions in unselected Chinese patients with NSCLC using multiplex real-time polymerase chain reaction (RT-PCR) to determine the prevalence of KIF5B-RET fusions and the association of KIF5B-RET fusion status with both histologic type and prognosis. Consequently, we identified 6 of 392 Chinese patients with NSCLC patients who were positive for a KIF5B-RET fusion, including 4 patients with adenocarcinoma, 1 patient with squamous cell carcinoma, and 1 patient with low-grade malignant neuroendocrine tumor, for a prevalence rate of 1.5%. MATERIALS AND METHODS Patients and Tissue Specimens Formalin-fixed, paraffin-embedded (FFPE) tissue sections were collected from patients with histologically confirmed primary NSCLC who underwent resection or biopsy at Shanghai Pulmonary Hospital from 2003 to Pathologic diagnosis and staging were performed according to the current World Health Organization classification and the tumor-lymph node-metastasis (TNM) staging system of the International Association for the Study of Lung Cancer (version 7). All FFPE tissue sections were reviewed by pathologists for confirmation of histology and assessment of tumor content. All clinical data were obtained from inpatient/outpatient medical records. Survival was analyzed for patients who received at least 1 follow-up telephone call or visit. Consecutive patients who met the following inclusion criteria were enrolled in this study: written informed consent; age 18 years; histologically confirmed NSCLC; sufficient FFPE tissue available for KIF5B-RET fusion screening and validation; demographic data, including age, sex, smoking status, histologic type, and disease stage, available for analysis; and patients did not receive preoperative systemic or radiation therapy. This study was approved by the Institutional Review Board of the Shanghai Pulmonary Hospital. RNA Extraction and Reverse Transcription Total RNA was extracted from three or four 3 lm-thick FFPE tissue sections using the RNeasy FFPE Kit (catalog no ; Qiagen, Hilden, Germany). Then, total RNA was subject to reverse transcription using the AmoyDx KIF5B-RET Fusion Gene Detection Kit (Amoy Diagnostics Company, Ltd., Xiamen, China). Total amounts of RNA should be within 0.1 to 5.0 lg. Reverse transcription conditions were as follow: 42 C for 1 hour, then 95 C for 5 minutes. The resulting complementary DNA solutions were used for multiplex RT-PCR. Multiplex Real-Time Polymerase Chain Reaction and Direct Sequencing To rapidly identify RET fusions with high efficiency using small amounts of RNA extracted from archival FFPE sections, we used multiplex RT-PCR to screen for these fusions in the current study. Only 7 known KIF5B-RET fusion variants were tested in this study, including KIF5B exon 15 fused to RET exon 12 (K15;R12), K16;R12, K23;R12, K24;R8, K22;R12, K15;R11, and K24;R11. All multiplex RT-PCR reactions were performed using the Stratagene Mx3000P Real-Time PCR system (Stratagene, La Jolla, Calif) with the AmoyDx KIF5B-RET Fusion Gene Detection Kit (Amoy Diagnostics Company, Ltd.). An internal reference gene (b-actin) and RET-rearranged DNA were used as control. The PCR reaction conditions were as follows: 1 cycle at 95 C for 5 minutes; 15 cycles of denaturation at 95 C for 25 seconds, annealing at 64 C for 20 seconds, and elongation at Cancer April 15,

3 72 C for 20 seconds; and 31 cycles at 93 C for 25 seconds, 60 C for 35 seconds (data collection), and 72 C for 20 seconds. All fusion-positive samples were validated using direct sequencing. Statistical Analysis Categorical variables were compared using Fisher s exact test, and continuous variables were compared using the Mann-Whitney U test. Survival curves were plotted and the median overall survival (OS) was calculated using the Kaplan-Meier method. Univariate and multivariate analyses were performed using a Cox proportional hazards model. Overall survival was calculated from the date of resection or biopsy to the date of death. Patients who remained alive or were lost to follow-up were censored on the date of last follow-up. The log-rank test was used to compare survival curves between patient groups with and without KIF5B-RET fusions. The 2-sided significance level was set at P <.05. All data were analyzed using the Statistical Package for the Social Sciences version 17.0 software package (SPSS Inc., Chicago, Ill). RESULTS Patient Characteristics In total, 392 patients with NSCLC who underwent resection or biopsy were collected consecutively at Shanghai Pulmonary Hospital, Tongji University (Shanghai, China) from 2003 to Of these enrolled patients, 3 underwent biopsy, and 389 underwent surgery. All patients were of Chinese origin. The major subtypes of NSCLC, including adenocarcinoma (58.9%; 231 of 392 patients), adenosquamous carcinoma (9.4%; 37 of 392 patients), and squamous cell carcinoma (30.4%; 119 of 392 patients), were included in this study. Patient characteristics are listed in Table 1. RET fusion status was evaluable in archival FFPE tissue sections from all patients enrolled. KIF5B-RET Fusions In total, 6 of 392 patients (1.5%) were positive for KIF5B-RET fusions by multiplex RT-PCR (Table 2), which was confirmed by direct sequencing. Among the 6 KIF5B-RET fusion-positive patients, KIF5B exon 15 fused to RET exon 12 (K15;R12) was validated in 5 patients, and KIF5B exon 22 fused to RET exon 12 (K22;R12) was validated in 1 patient. Sequences of the KIF5B-RET gene fusion variants are illustrated in Figure 1. TABLE 1. Clinicopathologic Characteristics of 392 Patients With Non-Small Cell Lung Cancer Characteristic All Patients, n ¼ 392 No. of Patients (%) KIF5B-RET Fusion Positive, n ¼ 6 Negative, n ¼ 386 Age, y Median [range] 60 [27-83] 58.5 [46-70] 60 [27-83].784 < (65.1) 4 (66.7) 251 (65) (34.9) 2 (33.3) 135 (35) Sex 1.00 Men 211 (53.8) 3 (50) 208 (53.9) Women 181 (46.2) 3 (50) 178 (46.1) Smoking history Median [range], PY 0 [0-180] 0 [0-0] 0 [0-180].064 Never 244 (62.2) 6 (100) 238 (61.7).091 a Light smoker: <10 PY 6 (1.5) 0 (0) 6 (1.6) Smoker: 10 PY 142 (36.2) 0 (0) 142 (36.8) Histology 1.00 b Adenocarcinoma 231 (58.9) 4 (66.7) 227 (58.8) Adenosquamous 37 (9.4) 0 (0) 37 (9.6) Squamous cell 119 (30.4) 1 (16.7) 118 (30.6) Others c 5 (1.3) 1 (16.7) 4 (1) Stage.703 d Ia 34 (8.7) 0 (0) 34 (8.8) Ib 135 (34.4) 2 (33.3) 133 (34.5) IIa 8 (2) 0 (0) 8 (2.1) IIb 48 (12.2) 1 (16.7) 47 (12.2) IIIa 107 (27.3) 1 (16.7) 106 (27.5) IIIb 28 (7.1) 0 (0) 28 (7.3) IV 32 (8.2) 2 (33.3) 30 (7.8) Abbreviations: KIF5B-RET, fusion of the ret proto-oncogene (RET) and kinesin family member 5B (KIF5B) genes; PY, pack-years. a Never/light smokers versus smokers. b Adenocarcinoma versus nonadenocarcinoma (adenosquamous, squamous cell, and others). c Others included neuroendocrine tumors and mixed tumors. d Stage I/II versus stage III/IV. Clinicopathologic Characteristics of Patients With KIF5B-RET Fusions Six patients were identified who harbored KIF5B-RET fusions. Among these, 4 were adenocarcinomas, 1 was squamous cell carcinoma, and 1 was a low-grade malignant neuroendocrine tumor. All KIF5B-RET fusions were identified in the resected specimens. The histologic type of KIF5B-RET fusion-positive nonadenocarcinoma samples was confirmed by immunohistochemistry (Fig. 2). The low-grade malignant neuroendocrine tumor was positive for the synaptophysin (SYN) and negative for cytokeratin 7 (CK7). The squamous cell carcinoma was positive for CK5/CK6 and negative for CK7. Of 119 patients with squamous cell carcinoma, 1 patient (0.84%) was positive for a KIF5B-RET fusion. Of 231 patients with adenocarcinoma, 4 patients (1.73%) were positive for a KIF5B-RET fusion. KIF5B-RET fusions P 1488 Cancer April 15, 2013

4 KIF5B-RET Fusions in NSCLC/Cai et al TABLE 2. Characteristics of the 6 Patients With KIF5B-RET Fusion-Positive Non-Small Cell Lung Cancer Patient No. Age, y Sex Smoking Stage Histology KIF5B-RET Fusion Pattern Survival 1 55 Woman Never Ib Adenocarcinoma K22;R12 Dead 2 65 Man Never IIb Squamous cell K15;R12 Dead 3 70 Man Never IIIa Adenocarcinoma K15;R12 Dead 4 46 Woman Never Ib Low-grade malignant K15;R12 Alive neuroendocrine tumor 5 54 Man Never IV Adenocarcinoma K15;R12 Dead 6 62 Woman Never IV Adenocarcinoma K15;R12 Dead Abbreviations: K15;R12, KIF5B exon 15 fused to RET exon 12; K22;R12, KIF5B exon 22 fused to RET exon 12; KIF5B-RET, fusion of the ret proto-oncogene (RET) and kinesin family member 5B (KIF5B) genes. Figure 1. KIF5B-RET (fusion of the ret proto-oncogene [RET] and kinesin family member 5B [KIF5B] genes) variants were identified by multiplex real-time polymerase chain reaction assay and were validated by direct sequencing. (A) KIF5B exon 22 fused to RET exon 12 (K22;R12) was identified in 1 patient. (B) K15;R12 was identified in 5 patients. were present in 2 of 161 patients (1.24%) with nonadenocarcinomas (including adenosquamous carcinoma, squamous cell carcinoma, neuroendocrine tumor, mixed tumor, etc). The ratio of men to women was 1:1 in the KIF5B-RET fusion-positive group. All KIF5B-RET fusion-positive patients had no smoking history (P ¼.064 compared with the KIF5B-RET fusion-negative group). There was no statistically significant difference in age (median age, 58.5 years vs 60.0 years; P ¼.784), sex (P ¼ 1.0), pathologic stage (P ¼.703), and histology (P ¼ 1.0) between the KIF5B-RET fusion-positive patients and the KIF5B-RET fusion-negative patients (see Table 1). Clinical Outcomes Survival analyses were performed in 384 of 392 patients (98%) who received at least 1 follow-up telephone call or visit, and the longest follow-up was 108 months. Among the patients who were available for survival analyses, in total, 202 of 384 deaths (52.6%) occurred during followup, including 197 of 378 deaths (52.1%) in KIF5B-RET fusion-negative patients and 5 of 6 deaths (83.3%) in KIF5B-RET fusion-positive patients. The median OS for evaluable patients in this study was 51.2 months (95% confidence interval [CI], months), with a median OS of 52.6 months (95% CI, months) for KIF5B-RET fusion-negative patients and Cancer April 15,

5 Figure 2. These are histology images from patients who had KIF5B-RET fusion-positive nonadenocarcinoma confirmed by immunohistochemistry (IHC). (A) Images from a patient who was positive for KIF5B-RET fusion reveal (a) the morphology of a lowgrade, malignant neuroendocrine tumor that was (b) positive for SYN (synaptophysin) and (c) negative for cytokeratin 7 (CK7) by IHC. (B) Images from another patient who was positive for KIF5B-RET fusion reveal (d) the morphology of squamous cell carcinoma that was (e) negative for CK7 and (f) positive for CK5/CK6 by IHC. Figure 3. This Kaplan-Meier curve illustrates overall survival (OS) in the groups that were negative and positive for the KIF5B-RET fusion (fusion of the ret proto-oncogene [RET] and kinesin family member 5B [KIF5B] genes) months (95% CI, months) for KIF5B- RET fusion-positive patients. KIF5B-RET fusion-negative patients tended to have a longer OS than KIF5B-RET fusion-positive patients (P ¼.06). The survival curve is provided in Figure 3. Univariate analysis indicated that age <65 years (median OS, 59.0 months vs 43.0 months; P ¼.01), being a woman (median OS, 61.1 months vs 42.1 months; P ¼.02), and low pathologic stage (median OS for stage I/II vs stage III/IV, not reached vs 31.4 months; P <.001) were associated with longer survival. In addition, negative KIF5B-RET fusion status (median OS, 52.6 months vs 21.0 months; P ¼.06) and never/ light smoking history (median OS, 56.9 months vs 42.8 months; P ¼.07) demonstrated a strong trend toward better OS without reaching statistical significance. Multivariate analysis identified low pathologic stage (hazard ratio, 2.879; 95% CI, ) as the strongest independent prognostic factor for better survival. The other independent factor that affected survival was age <65 years (hazard ratio, 1.485; 95% CI, ) (Table 3). DISCUSSION In this study, 6 of 392 Chinese patients with NSCLC were identified as positive for KIF5B-RET fusions by 1490 Cancer April 15, 2013

6 KIF5B-RET Fusions in NSCLC/Cai et al TABLE 3. Univariate and Multivariate Analysis of Prognostic Factors in Patients With Non-Small Cell Lung Cancer Univariate Analysis Multivariate Analysis Variable HR (95% CI) P HR (95% CI) P Age<65 y vs 65 y ( ) ( ).006 Women vs men ( ) ( ).074 Stage I/II vs stage III/IV ( ) < ( ) <.001 KIF5B-RET-negative vs KIF5B-RET-positive ( ) ( ).055 Never/light smokers vs smokers ( ).072 a Non-LADC vs LADC ( ).274 Abbreviations: CI, confidence interval; HR, hazard ratio; KIF5B-RET, fusion of the ret proto-oncogene (RET) and kinesin family member 5B (KIF5B) genes; LADC, lung adenocarcinoma. a Smoking status was not calculated as a prognostic factor for multivariate analysis because of a correlation with sex. multiplex RT-PCR based screening and sequencingbased validation. Of these, 4 patients presented with adenocarcinoma, 1 patient presented with squamous cell carcinoma, and 1 patient presented with a low-grade malignant neuroendocrine tumor. Only 2 KIF5B-RET variants, K22;R12 and K15;R12, were identified. No specific clinicopathologic feature was associated significantly with KIF5B-RET fusions. Survival analysis indicated a trend toward worse prognosis for KIF5B-RET fusion-positive patients (P ¼.06). Because there is no standardized method for detecting RET fusions and KIF5B-RET is the most commonly identified gene fusion in NSCLC to date, we used multiplex RT- PCR to screen for known KIF5B-RET fusions in the current study. All KIF5B-RET fusion variants were identified by multiplex RT-PCR using primers for known KIF5B- RET fusions and were validated by direct sequencing. To our knowledge, the Vysis ALK Break-Apart fluorescence in situ hybridization (FISH) Probe Kit (Abbott Molecular, Des Plains, Ill) is the only US Food and Drug Administration-approved test with which to identify ALK rearrangements in patients who have NSCLC. However, the FISH assay has not been used routinely in clinical practice in China because of high costs and difficulty in interpretation, despite its high specificity. In addition, specific variants of fusion genes cannot be distinguished by using a break-apart FISH assay. In addition to FISH, other alternative diagnostic approaches, such as RT-PCR and immunohistochemistry, reportedly are consistent in the detection of gene translocations and are surrogates for FISH. The RT-PCR assay is highly sensitive, cheap, and easily widespread, although, currently, it can detect only known fusion gene variants but not unknown, novel variants. The findings from this study in which we screened for KIF5B-RET fusions in an unselected population with NSCLC indicated that the frequency of KIF5B-RET fusions in Chinese patients with NSCLC was 1.5% (6 of 392 patients). In total, we tested 7 variants for known KIF5B-RET fusions in this study. Subsequently, the variant K22;R12 was identified in 1 patient, and K15;R12 was identified in 5 patients, suggesting that these 2 KIF5B-RET variants may be common in Chinese patients with NSCLC. The 6 KIF5B-RET fusion-positive patients included 4 of 231 patients with adenocarcinoma (1.73%,), 1 of 119 patients with squamous cell carcinoma (0.84%), and 1 patient with a low-grade malignant neuroendocrine tumor, which was confirmed by immunohistochemistry on the basis of resection specimens. Moreover, all KIF5B-RET fusion-positive patients in this study were negative for both EML4-ALK and ROS1 fusions according to the results from multiplex RT-PCR (data not shown), suggesting that KIF5B-RET was a potential oncogenic driver of lung cancer. Our findings basically were consistent with the prevalence of RET fusions in adenocarcinomas reported previously. It is very interesting and important to note that 2 KIF5B-RET fusion-positive patients who had tumors with nonadenocarcinoma histology were identified in our study, suggesting that RET fusions did not exist exclusively in lung adenocarcinoma. Fusion genes typically are present in lung cancer at low frequency, as discussed above. Therefore, identifying the enriched population of fusion genes in lung cancer could contribute to future clinical screening. The clinical characteristics of patients with ALK-rearranged and ROS1-rearranged NSCLC have been reported previously. Several studies revealed that EML4-ALK fusions occurred with significantly greater frequency among younger patients with lung adenocarcinoma who were neversmokers or who had a light smoking history. 2,19-21 Similar to EML4-ALK fusion-positive patients, patients with ROS1 rearrangements tended to be younger never-smokers with adenocarcinoma histology in Asia. 6 However, there was no statistically significant difference in age, sex, histologic type, smoking status, or pathologic stage Cancer April 15,

7 between the KIF5B-RET fusion-positive and fusion-negative groups in our current study (P >.05). Although all KIF5B-RET fusion-positive patients were never-smokers, statistical analysis did not indicate a significant difference in smoking status, possibly because of the small number of KIF5B-RET fusion-positive patients and the high proportion of never smokers and light smokers (63.8%) in our patients. KIF5B-RET fusions appeared to be common in nonsmokers; however, we cannot draw this hasty conclusion on the basis of the low frequency of KIF5B-RET fusions and limited RT-PCR based screening. Our study screened for RET fusions in unselected patients with NSCLC and subsequently identified 1 patient with squamous cell carcinoma and 1 patient with a low-grade malignant neuroendocrine tumor. The confirmed histology images from immunohistochemical analysis are provided in Figure 2. The frequency of KIF5B-RET fusions was numerically higher in patients with adenocarcinoma than in patients with nonadenocarcinoma, although the difference did not reach statistical significance (P >.05). Some scholars have used new-generation sequencing to reveal genomic alternations in squamous cell lung cancer. Govindan et al reported the comprehensive genomic characterization of squamous cell lung cancer in 178 previously untreated patients who had stage I through III squamous cell lung cancer. 22 In a study conducted by The Cancer Genome Atlas (TCGA), several gene rearrangements were detected involving phosphatase and tensin homolog (PTEN), retinoblastoma 1 (RB1), NOTCH1, neurofibromin 1 (NF1), and cyclin-dependent kinase inhibitor 2A (CDKN2A); however, RET fusions still were not identified in their study. We believe that 3 factors may explain the difference in findings. First, there may be a potential ethnic difference in RET fusions between patients with NSCLC among Asians and Non- Asians like the ethnic difference in EGFR mutations. Second, different screening methods also may be a factor, because the sensitivity of multiplex RT-PCR is relatively greater than high-throughput sequencing. Finally, potential causal factors cannot be excluded based on the low frequency of RET fusions in our study. However, given the low response rate to EGFR TKIs in patients with squamous cell carcinoma of only 4% 23 and the limited efficacy of chemotherapy, the first identification of KIF5B-RET in squamous cell carcinoma, for which there has been no effective therapy, may provide a guide for clinical applications despite the very low frequency of RET fusions in squamous cell carcinoma. In addition, RET gene abnormalities are closely associated with the pathogenesis of neuroendocrine tumors ; and, among these, thyroid cancer is the most typical. Also, lung neuroendocrine tumors account for approximately 20% of lung cancers and are classified as high-grade (approximately 15% of lung cancers), intermediate-grade, and low-grade neuroendocrine carcinoma (accounting for 1%-2% of lung cancers). 30,31 One KIF5B-RET fusion-positive patient in our current study was diagnosed with low-grade neuroendocrine carcinoma (typical carcinoids); and, again, our findings demonstrated a close correlation between RET gene abnormality and neuroendocrine tumors. Moreover, reports indicate that patients with advanced typical carcinoids often have an unsatisfactory response to platinum-based doublet chemotherapy 27,32,33 and limited treatment options. Therefore, the identification of RET fusions in lung neuroendocrine tumors is very important for treatment decision-making in these patients. Most patients in our current study had completely resected, early stage disease, so the survival data were very beneficial in determining the effect of RET fusions on prognosis. A previous study demonstrated that patients who are positive for gene fusions, including ALK, ROS1, and RET fusions, may have a better prognosis than patients without gene fusions. 18 Conversely, the survival analysis of 384 patients in our study demonstrated that OS for KIF5B-RET fusion-negative patients with NSCLC was better than OS for KIF5B-RET fusion-positive patients (median OS, 52.6 months vs 21.0 months), with borderline statistical significance (P ¼.06). Univariate analysis suggested that other 3 prognostic factors age 65 years (P ¼.011), being a man (P ¼.023), and high pathologic stage (P <.001) were associated significantly with poor survival. In multivariate analysis, RET fusion status did not reach statistical significance (hazard ratio, 2.398; 95% CI, ; P ¼.055). However, we must mention several potential limitations of statistical analysis in the current study. First, only known KIF5B- RET fusion variants were detected using the multiplex RT-PCR assay. This difference in OS between the KIF5B-RET fusion-positive and fusion-negative groups could not reflect an association of RET fusion status with the survival of patients with NSCLC. It is very possible that some patients who had a good prognosis but harbored other RET fusion gene variants were missed. Second, the proportion of patients with stage IV disease in KIF5B-RET fusion-positive group obviously was higher than the proportion of patients in the KIF5-RET fusionnegative group (33.3% vs 7.7%). The survival analysis was not adjusted for stage because of the small number of KIF5B-RET fusion-positive patients. Therefore, the 1492 Cancer April 15, 2013

8 KIF5B-RET Fusions in NSCLC/Cai et al factors mentioned above, including a limited RT-PCRbased assay, the small number of fusion-positive patients, and the significant imbalance in disease stage between patients in the 2 groups, may have a potential impact on the statistical evaluations in this study. Given the striking efficacy of ALK-targeted and ROS1-targeted therapy observed in patients with NSCLC, we expect that RET inhibitors may produce the same satisfactory results in RET fusion-positive patients who have NSCLC. RET inhibitors reportedly are active in vitro. Lipson et al have demonstrated that sunitinib, sorafenib, and vandetanib (Zactima; AstraZeneca Pharmaceuticals, London, United Kingdom) effectively inhibited RETpositive lung cancer cells in vitro. 8 Kohno et al observed that vandetanib could suppress the phosphorylation of KIF5B-RET kinase. 7 Vandetanib selectively targets RET, vascular endothelial growth factor receptor, and EGFR and has been approved for advanced thyroid cancer. Further studies of RET inhibitors in NSCLC are ongoing. Therefore, the preclinical study of RET fusions in Chinese patients with NSCLC is very important to the future application of RET inhibitors in clinical practice. In conclusion, to our knowledge, the current study is the first to provide new insights into a better understanding of the role of KIF5B-RET fusions in Chinese patients with NSCLC. The results indicate that adenocarcinoma is not an exclusive histologic type harboring RET fusions, and RET fusions also may exist in squamous cell carcinomas and low-grade lung neuroendocrine tumors. Further study on the impact of RET inhibition on the survival of Chinese patients with NSCLC who harbor RET fusions is ongoing. FUNDING SOURCES This study was supported by grants from the National Natural Science Foundation of China (grant ) and the Key Project of the Science and Technology Commission of Shanghai Municipality (grant 11JC ). CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures. REFERENCES 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin. 2011;61: Soda M, Choi YL, Enomoto M, et al. Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer. Nature. 2007;448: Boland JM, Erdogan S, Vasmatzis G, et al. Anaplastic lymphoma kinase immunoreactivity correlates with ALK gene rearrangement and transcriptional up-regulation in non-small cell lung carcinomas. Hum Pathol. 2009;40: 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;115: Kwak EL, Bang YJ, Camidge DR, et al. Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer. N Engl J Med. 2010;363: Bergethon K, Shaw AT, Ou SH, et al. ROS1 rearrangements define a unique molecular class of lung cancers. J Clin Oncol. 2012;30: Kohno T, Ichikawa H, Totoki Y, et al. KIF5B-RET fusions in lung adenocarcinoma. Nat Med. 2012;18: Lipson D, Capelletti M, Yelensky R, et al. Identification of new ALK and RET gene fusions from colorectal and lung cancer biopsies. Nat Med. 2012;18: Li F, Feng Y, Fang R, et al. Identification of RET gene fusion by exon array analyses in pan-negative lung cancer from never smokers. Cell Res. 2012;22: Ju YS, Lee WC, Shin JY, et al. A transforming KIF5B and RET gene fusion in lung adenocarcinoma revealed from whole-genome and transcriptome sequencing. Genome Res. 2012;22: Davies KD, Le AT, Theodoro MF, et al. Identifying and targeting ROS1 gene fusions in non-small cell lung cancer. Clin Cancer Res. 2012;18: Shaw AT, Camidge DR, Engelman JA, et al. Clinical activity of crizotinib in advanced non-small cell lung cancer (NSCLC) harboring ROS1 gene rearrangement [abstract]. J Clin Oncol. 2012(30S). Abstract Baloh RH, Enomoto H, Johnson EM Jr, Milbrandt J. The GDNF family ligands and receptors implications for neural development. Curr Opin Neurobiol. 2000;10: Grieco M, Santoro M, Berlingieri MT, et al. PTC is a novel rearranged form of the ret proto-oncogene and is frequently detected in vivo in human thyroid papillary carcinomas. Cell. 1990; 60: Fugazzola L, Pilotti S, Pinchera A, et al. Oncogenic rearrangements of the RET proto-oncogene in papillary thyroid carcinomas from children exposed to the Chernobyl nuclear accident. Cancer Res. 1995;55: Jhiang SM, Sagartz JE, Tong Q, et al. Targeted expression of the ret/ptc1 oncogene induces papillary thyroid carcinomas. Endocrinology. 1996;137: Murakumo Y, Jijiwa M, Asai N, Ichihara M, Takahashi M. RET and neuroendocrine tumors. Pituitary. 2006;9: Takeuchi K, Soda M, Togashi Y, et al. RET, ROS1 and ALK fusions in lung cancer; Nat Med. 2012;18: Shaw AT, Yeap BY, Mino-Kenudson M, et al. Clinical features and outcome of patients with non-small-cell lung cancer who harbor EML4-ALK. J Clin Oncol. 2009;27: 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;17: Koivunen JP, Mermel C, Zejnullahu K, et al. EML4-ALK fusion gene and efficacy of an ALK kinase inhibitor in lung cancer. Clin Cancer Res. 2008;14: Govindan R, Hammerman PS, Hayes DN, et al. Comprehensive genomic characterization of squamous cell carcinoma of the lung [abstract]. J Clin Oncol. 2012(30S). Abstract Comis RL. The current situation: erlotinib (Tarceva) and gefitinib (Iressa) in non-small cell lung cancer. Oncologist. 2005;10: Mulligan LM, Kwok JB, Healey CS, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature. 1993;363: Donis-Keller H, Dou S, Chi D, et al. Mutations in the RET protooncogene are associated with MEN 2A and FMTC. Hum Mol Genet. 1993;2: Hofstra RM, Landsvater RM, Ceccherini I, et al. A mutation in the RET proto-oncogene associated with multiple endocrine neoplasia type 2B and sporadic medullary thyroid carcinoma. Nature. 1994;367: Oberg K, Hellman P, Kwekkeboom D, Jelic S. Neuroendocrine bronchial and thymic tumours: ESMO clinical practice guidelines Cancer April 15,

9 for diagnosis, treatment and follow-up. Ann Oncol. 2010;21(suppl 5): Govindan R, Page N, Morgensztern D, et al. Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the Surveillance, Epidemiology, and End Results database. J Clin Oncol. 2006;24: Navada S, Lai P, Schwartz A, Kalemkerian G. Temporal trends in small cell lung cancer: analysis of the national Surveillance, Epidemiology, and End Results (SEER) database [abstract]. J Clin Oncol. 2006;24(18S). Abstract Travis W, Brambilla E, Muller-Hermelink H, eds. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IARC Press; Bajetta E, Catena L, Procopio G, et al. Is the new WHO classification of neuroendocrine tumours useful for selecting an appropriate treatment? Ann Oncol. 2005;16: Gustafsson BI, Kidd M, Chan A, Malfertheiner MV, Modlin IM. Bronchopulmonary neuroendocrine tumors. Cancer. 2008; 113: Moertel CG, Kvols LK, O Connell MJ, Rubin J. Treatment of neuroendocrine carcinomas with combined etoposide and cisplatin. Evidence of major therapeutic activity in the anaplastic variants of these neoplasms. Cancer. 1991;68: Cancer April 15, 2013