Lung cancer is a leading cause of cancer-related mortality,

Original Article Unique Genetic and Survival Characteristics of Invasive Mucinous Adenocarcinoma of the Lung Hyo Sup Shim, MD, PhD,* Mari-Kenudson, MD,* Zongli Zheng, PhD,* Matthew Liebers, BSc,* Yoon Jin Cha, MD, Quan Hoang Ho, BSc,* Maristela Onozato, MD, PhD,* Long Phi Le, MD, PhD,* Rebecca S. Heist, MD, MPH, and A. John Iafrate, PhD* Introduction: Invasive mucinous adenocarcinoma is a unique histologic subtype of lung cancer, and our knowledge of its genetic and clinical characteristics is rapidly evolving. Here, we present next- generation sequencing analysis of nucleotide variant and fusion events along with clinical follow-up in a series of lung mucinous adenocarcinoma. Methods: We collected 72 mucinous adenocarcinomas from the United States and Korea. All had been previously assessed for KRAS and EGFR mutations. For KRAS wild-type cases (n = 30), we performed deep targeted next-generation sequencing for gene fusions and nucleotide variants and correlated survival and other clinical features. Results: As expected, KRAS mutations were the most common alteration found (63% of cases); however, the distribution of nucleotide position alterations was more similar to that observed in gastrointestinal tumors than other lung tumors. Within the KRAS-negative cases, we found numerous potentially targetable gene fusions and mutations, including CD74-NRG1, VAMP2-NRG1, TRIM4-BRAF, TPM3-NTRK1, and EML4-ALK gene fusions and ERBB2, BRAF, and PIK3CA mutations. Unexpectedly, we found only two cases with TP53 mutation, which is much lower than observed in lung adenocarcinomas in general. The overall mutation burden was low in histologically confirmed mucinous adenocarcinomas from the public The Cancer Genome Atlas exome data set, regardless of smoking history, suggesting a link between TP53 status and mutation burden in mucinous tumors. There was no significant difference for recurrence-free survival between stagematched mucinous and nonmucinous adenocarcinomas. It was notable that all recurrence sites were in the lungs for completely resected cases. *Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts; Department of Pathology, Yonsei University College of Medicine, Seoul, Korea; and Department of Thoracic Oncology, Massachusetts General Hospital Cancer Center, Boston, Massachusetts. Disclosure:AJI, LPL, and ZZ have submitted a preliminary patent for the anchored multiplex PCR technology to the US patent office. AJI, LPL, and ZZ are equity holders in ArcherDx, a licensee of the technology. RSH has received honoraria for consulting from Boehringer-Ingelheim and Momenta, unrelated to this project. This study was supported by a faculty research grant of Yonsei University College of Medicine for 2012 and 2013 (6-2012-0043; 6-2013-0016) to HSS, by NIH grant (R21CA161590) to AJI, and by Lungevity Foundation and Upstage Lung Cancer to RSH. Address for correspondence: Department of Pathology, Massachusetts General Hospital, 55 Fruit Street, Jackson 1015A, Boston, MA 02114. E-mail: aiafrate@partners.org DOI: 10.1097/JTO.0000000000000579 ISSN: 1556-0864/15/1008-1156 Conclusions: Our data suggest that mucinous adenocarcinoma is typified by (1) frequent KRAS mutations and a growing list of gene fusions, but rare TP53 mutations, (2) a low mutation burden overall, and (3) a recurrence-free survival similar to stage-matched nonmucinous tumors, with recurrences limited to the lungs. Key Words: Lung, Adenocarcinoma, Mucinous, Mutation, Gene fusion, Targeted therapy. (J Thorac Oncol. 2015;10: 1156 1162) Lung cancer is a leading cause of cancer-related mortality, 1 and adenocarcinoma is its most common histologic type. 2 Invasive mucinous adenocarcinoma, formerly known as mucinous bronchioloalveolar carcinoma, is a distinct variant of adenocarcinoma of the lung, accounting for approximately 5% of lung adenocarcinomas. 2 Its histology is unique among the primary lung cancers and is typified by a columnar or goblet cell structure with basally located nuclei and abundant intracytoplasmic mucin. Invasive mucinous adenocarcinoma is well known as having a distinct clinical presentation and genetic profile compared with nonmucinous adenocarcinoma. 2 6 Patients with invasive mucinous adenocarcinoma frequently present with a pneumonia-like pattern and with multifocal and multilobar lesions. 3 There are conflicting data about the relative prognosis of patients with mucinous adenocarcinoma. 7,8 In terms of genetic alterations, invasive mucinous adenocarcinoma shows a strong correlation with KRAS mutations. 4 6 However, comprehensive molecular or clinical studies on invasive mucinous adenocarcinoma have been limited so far because the histology is relatively rare compared with other subtypes. Although a comprehensive molecular profiling of lung adenocarcinomas from The Cancer Genome Atlas (TCGA) has been published, 9 a detailed analysis of this subtype is warranted. Recently, CD74- NRG1 fusions have been discovered in lung mucinous adenocarcinoma, 10 12 showing that these tumors are likely genetically unique. To address the genetic and survival characteristics of invasive mucinous adenocarcinoma of the lung, we performed targeted next-generation sequencing for gene fusions and mutations, analyzed TCGA lung adenocarcinoma data, and investigated clinical features in this subtype when compared with nonmucinous adenocarcinomas. 1156 Journal of Thoracic Oncology Volume 10, Number 8, August 2015

Journal of Thoracic Oncology Volume 10, Number 8, August 2015 Invasive Mucinous Adenocarcinoma of the Lung MATERIALS AND METHODS Study Population We collected a total of 83 mucinous adenocarcinomas from Massachusetts General Hospital, Boston, Massachusetts (n = 35) and Yonsei University Severance Hospital, Seoul, Korea (n = 48) (Supplementary Fig. 1, Supplemental Digital Content, http://links. lww.com/jto/a842) under Institutional Review Board approval. All samples were resected specimens and formalin-fixed and paraffin-embedded. Two pulmonary pathologists (H.S.S. and M.M.-K.) reviewed the slides and confirmed the diagnosis based on the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society (IASLC/ATS/ ERS) classification. 2 All clinical information, such as age, gender, smoking status, and stage, were obtained from medical records. We also collected a total of 269 nonmucinous adenocarcinomas from Massachusetts General Hospital (n = 63) and Yonsei University Severance Hospital (n = 206) as a control group for survival analysis. All patients underwent surgical resection for curative intent. TCGA data for lung adenocarcinomas (n = 230) were obtained from the public on-line cbioportal database. 9,13,14 Histologic review for each case was done using Digital Slide Archive in the cbioportal. SNaPshot Genotyping Multiplexed targeted genotyping was done using the SNaPshot method as previously described. 15,16 The SNaPshot platform from Applied Biosystems (Life Technologies/Applied Biosystems, Foster City, CA) consisted of multiplexed polymerase chain reaction (PCR) and single-base extension reactions that generate fluorescent labeled probes designed to interrogate hot-spot mutation sites. The SNaPshot products were then resolved and analyzed using capillary electrophoresis. Sanger Sequencing for EGFR and KRAS Mutation A representative formalin-fixed, paraffin-embedded block containing at least 50% viable tumor was selected for each sample. After proteinase K digestion, DNA was extracted using a DNeasy DNA isolation kits (Qiagen, Valencia, CA) according to the manufacturer s instructions. Direct DNA sequencing of exons 18 through 21 of the EGFR gene and codons 12 and 13 of the KRAS gene was performed as previously described. 17,18 Anchored Multiplex PCR and Next- Generation Sequencing To detect gene fusions and mutations, we used a gene enrichment method, anchored multiplex PCR, to perform next-generation sequencing using MiSeq (Illumina, San Diego, CA) platform as previously described in detail. 19 Total nucleic acid containing total RNA and genomic DNA were extracted from formalin-fixed paraffin-embedded tissue, using the Agencourt FormaPure Kit (Beckman Coulter, Indianapolis, IN). We used at least 50 ng of total nucleic acid for fusion analysis and 200 ng of genomic DNA for mutation analysis. The genes covered in each primer panel are shown in Supplementary Table 1 (Supplemental Digital Content, http:// links.lww.com/jto/a842). Immunohistochemisty Formalin-fixed and paraffin-embedded tissues were sectioned with a thickness of 4 μm and stained with antibody for p53 (mouse monoclonal, clone DO-7, ready-to-use, Leica Biosystems, United Kingdom) using Leica Bond 6 automated stainer according to the manufacturer s protocol. p53 immunohistochemistry was considered to be abnormal when the expression was present in 50% or greater of the tumor cells or was completely negative. Statistical Analysis Relationships between clinicopathologic parameters were evaluated using the chi-square test. Student s t test was used to compare means between two independent groups. We planned a survival comparison study of mucinous cases and nonmucinous cases (controls) with four controls per case. Prior data indicated that the overall 5-year survival rate among controls is 60%. If the survival rate among mucinous cases is 80%, we needed to study at least 52 patients with mucinous adenocarcinoma and 208 control patients to be able to reject the null hypothesis that the survival rates for case and controls are equal with probability (power) 0.8. This sample size was calculated by Power and Sample Size Calculations (Version 3.1.2; Vanderbilt University, Nashville, TN). The disease-free survival and overall survival were evaluated using the Kaplan Meier method, and statistical differences in survival times were determined using the logrank test. Data analysis was conducted using SPSS v.17 (SPSS, Chicago, IL) or Prism 6 (GraphPad Software, San Diego, CA). Significance was defined as p value less than 0.05. RESULTS We identified 83 cases of mucinous adenocarcinoma, including 81 invasive mucinous adenocarcinomas, one mucinous adenocarcinoma in situ, and one minimally invasive mucinous adenocarcinoma (all defined using the IASLC/ ATS/ERS criteria) and 269 nonmucinous cases from the case records of the Massachusetts General Hospital and the Yonsei University Severance Hospital (detailed information according to the institutions in Supplementary Table 2, Supplemental Digital Content, http://links.lww.com/jto/a842). There were no significant differences between patients with mucinous adenocarcinoma and those with nonmucinous adenocarcinoma with respect to age, sex, smoking status, or stage (Table 1). Of 83 patients with mucinous adenocarcinoma, KRAS genotyping was previously done for 72 cases. For KRAS wild-type cases (n = 30), we performed targeted deep sequencing for gene fusions and mutations using the anchored multiplex PCR method. We found driver mutations and fusions in 16 of the 30 cases (Fig. 1), including three cases with KRAS mutation not found with the prior less-sensitive methods (Supplementary Tables 3 and 4, Supplemental Digital Content, http://links.lww. com/jto/a842). There was no significant difference between KRAS mutated (n = 45) and KRAS wild-type (n = 27) groups in terms of age, gender, smoking status, and stage (Table 2). We identified gene fusions in nine cases, including four with a CD74-NRG1 fusion, two with an EML4-ALK fusion, one each with a VAMP2-NRG1 fusion, a TRIM4-BRAF fusion, and a TPM3-NTRK1 fusion. We confirmed novel fusions using 1157

Shim et al. Journal of Thoracic Oncology Volume 10, Number 8, August 2015 TABLE 1. Clinical Characteristics of All Patients Enrolled in This Study TABLE 2. Clinical Characteristics of Patients with Mucinous Adenocarcinoma according to KRAS Status Histology Mucinous (n = 83) Nonmucinous (n = 269) p Factors Total (n = 72) KRAS Mutant KRAS Wild (n = 45) a (n = 27) a p Mean age (range) 62.2 (36 90) 62.7 (34 85) 0.743 Gender Male 36 (43.4) 130 (48.3%) 0.429 Female 47 (56.6) 139 (51.7%) Smoking status Never 39 (47.0) 137 (50.9) 0.530 Ever 44 (53.0) 132 (49.1) Stage I 55 (66.3) 192 (71.4) 0.063 II 19 (22.9) 51 (19.0) III 7 (8.4) 26 (9.7) IV 2 (2.4) 0 (0.0) Mean age (range) 63.4 (37 90) 63.3 (37 90) 63.6 (41 84) 0.129 Sex Male 30 (41.7) 18 (40.0) 12 (44.4) 0.711 Female 42 (58.3) 27 (60.0) 15 (55.6) Smoking status Never 33 (45.8) 21 (46.7) 12 (44.4) 0.855 Ever 39 (54.2) 24 (53.3) 15 (55.6) Stage I 45 (63.4) 30 (66.7) 15 (57.7) 0.889 II 17 (23.9) 10 (22.2) 7 (26.9) III 7 (9.9) 4 (8.9) 3 (11.5) IV 2 (2.8) 1 (2.2) 1 (3.8) a The KRAS status was determined by next-generation sequencing. reverse-transcriptase PCR (data not shown). The CD74-NRG1 fusion and VAMP2-NRG1 fusion seem to be generated by interchromosomal translocation (Supplementary Figs. 2 and 3, Supplemental Digital Content, http://links.lww.com/jto/a842) and retain an intact epidermal growth factor-like domain. The TRIM4-BRAF fusion and TPM3-NTRK1 fusion seem to be generated by intrachromosomal tandem duplication and paracentric inversion, respectively (Supplementary Fig. 4, Supplemental Digital Content, http://links.lww.com/jto/a842). Both BRAF and NTRK1 retain an intact kinase domain. There were no significant difference in gene fusion status according to smoking history (p = 0.194), although six out of nine fusion-positive cases were identified in never smokers. There was no statistical difference in molecular profile between the U.S. and Korean patients (Supplementary Table 5, Supplemental Digital Content, http:// links.lww.com/jto/a842). We also identified mutations in eight of the cases, including three cases with KRAS mutation. Additional mutations included two with ERBB2 mutation (exon 20 AYVM insertion), and one each with BRAF mutation (V600E), PIK3CA mutation (E542K), and TP53 mutation (H197D) (Fig. 1). Of 30 cases tested by next-generation sequencing, apart from TP53, these mutations and fusions were mutually exclusive. The distribution of mutations revealed that 63% of all 72 cases had KRAS mutations, which as expected were the most common variant observed (Fig. 2A). NRG1 fusions were second most common, accounting for 7%. When all fusion cases were combined, they accounted for 13% of all cases in our cohort. The distribution of KRAS amino acid changes in the mucinous tumors was examined and was found to be distinct from the KRAS mutation profile in lung adenocarcinoma in general as well as the profile of TCGA lung adenocarcinoma cases. G12D (42%) and G12V (31%) were the most common variants observed in mucinous tumors (Fig. 2B), whereas G12C was the most common in lung adenocarcinoma. 20 Interestingly, this pattern of KRAS mutations more resembles the mutational pattern seen in colorectal or pancreatobiliary tumors. 20 A total of 50 cases were examined for TP53 mutation by next-generation sequencing, revealing only two TP53 mutations (Fig. 1; Supplementary Fig. 5, Supplemental Digital Content, http://links.lww.com/jto/a842). Because this rate of TP53 mutation was surprisingly low, we randomly analyzed a subset of tumors with immunohistochemistry for p53 and found 10 TP53 wild-type cases to be weak and focal staining compared with a positive control (TP53 mutated case) showing the expected diffuse and strong staining (Supplementary Fig. 6, Supplemental Digital Content, http://links.lww.com/ JTO/A842). We attempted to confirm the low TP53 mutation rate in mucinous tumors in another data set and identified 12 FIGURE 1. Mutation distribution for 30 mucinous adenocarcinomas that underwent NGS. All 30 cases were previously determined to be KRAS wild type by targeted methodology. NGS, next generation sequencing. 1158

Journal of Thoracic Oncology Volume 10, Number 8, August 2015 Invasive Mucinous Adenocarcinoma of the Lung FIGURE 2. Pie charts showing the fraction of mucinous adenocarcinomas that harbor the indicated drivers (A) and the fraction of subtypes of KRAS mutations (B). invasive mucinous adenocarcinomas out of 230 lung cancers from the public TCGA lung adenocarcinoma Digital Slide Archive (Supplementary Table 6, Supplemental Digital Content, http://links.lww.com/jto/a842). Of the 12 cases, nine cases (75%) harbored KRAS mutation, one case had an ALK fusion, and one case had a RET fusion (Supplementary Table 6, Supplemental Digital Content, http://links.lww.com/ JTO/A842), whereas only one case (8%) harbored TP53 mutation (R273C). Concurrently, TP53 mutations were present in 104 of the 218 remaining TCGA cases (48%), confirming the significant difference in TP53 mutations between mucinous and nonmucinous tumors (p = 0.007). Interestingly, we observed in the TCGA data set that the overall mutational burden was lower in mucinous tumors versus nonmucinous tumors (p = 0.0099; Fig. 3A). TP53-mutant cases showed a higher mutation count, suggesting that the low mutation burden in mucinous tumors may be associated with a lack of TP53 mutations (Figs. 3B and 4). Because never smokers also harbored a lower mutation burden (Fig. 3C), possibly due to the reduced exposure to chronic DNA damage, we performed multivariate analysis which revealed that TP53 status likely contributed to the low mutation burden in mucinous tumors (Supplementary Table 7, Supplemental Digital Content, http:// links.lww.com/jto/a842). The smoking status, although an independent factor for the mutation burden, was not a significant confounding factor for the low mutation burden observed in mucinous tumors (Supplementary Table 7, Supplemental Digital Content, http://links.lww.com/jto/a842). Survival analysis was performed in our cohort patients with clinical data available and limited to those who FIGURE 3. Scatter dot plots for total mutation count between two groups defined by mucinous histology (A), TP53 mutation (B), and smoking status (C) in TCGA lung adenocarcinoma. TCGA, The Cancer Genome Atlas. 1159

Shim et al. Journal of Thoracic Oncology Volume 10, Number 8, August 2015 FIGURE 4. Distribution of mutation counts in TCGA lung adenocarcinoma. Mucinous adenocarcinomas (red) and TP53-mutant cases (blue) are indicated. EGFRmutant cases (enriched in nonsmokers), KRAS-mutant cases (enriched in smokers), and ALK/ROS1/RET fusions (enriched in nonsmokers) are included as a validation of the data set. TCGA, The Cancer Genome Atlas. underwent complete surgical resection for stage I to IIIA disease (79 patients diagnosed as invasive mucinous adenocarcinoma compared with 269 control patients diagnosed as invasive nonmucinous adenocarcinoma; Supplementary Fig. 1, Supplemental Digital Content, http://links.lww.com/jto/ A842). As expected, there was clear stratification in the cohort between stages I, II, and III for overall (all-cause mortality) survival and recurrence-free survival (Supplementary Fig. 7, Supplemental Digital Content, http://links.lww.com/jto/ A842; p < 0.001). When comparing patients with mucinous versus nonmucinous adenocarcinoma, there was no statistically significant difference in overall survival when combining all stages (Fig. 5A; p = 0.499). The overall 5-year survival rates for patients with mucinous and nonmucinous adenocarcinoma were 71.7% and 67.2%, respectively, and there was no statistically significant difference. Because we did not have access to disease-specific mortality data, we instead performed recurrence-free survival analysis that showed a tendency for the mucinous cohort associating with better recurrence-free survival, although it did not reach statistical significance (Fig. 5B; p = 0.313). When examining stage-specific survival (stage I only), there were also no significant differences in all-cause overall (p = 0.627) and recurrence-free survival (p = 0.159) between the two cohorts (Supplementary Fig. 8A, B, Supplemental Digital Content, http://links.lww.com/jto/ A842), although once again patients with mucinous tumor showed a tendency to have better recurrence-free survival. In patients with mucinous adenocarcinoma, there were no differences in prognosis according to the status of KRAS mutations or gene fusions (p = 0.552 and 0.848, respectively). Of 79 patients with the mucinous type, 14 patients (17.7%) had disease recurrence that was limited to the lungs. No patients with recurrent disease had extrapulmonary recurrence. By contrast, of 192 stage I patients with the nonmucinous type, 38 patients (19.8%) had recurrence of disease; the recurrence sites for 13 patients (34.2%) were limited to the lungs, but 25 patients (65.8%) presented with at least one extrapulmonary site of metastasis. DISCUSSION In this study, we have shown that invasive mucinous adenocarcinoma of the lung is genetically and clinically distinct. The tumors have a low mutation burden and most often have a single identifiable driver mutation. KRAS mutations were the FIGURE 5. Survival curves comparing mucinous and nonmucinous cases for overall (A) and recurrence-free (B) survival in stage I IIIA. 1160

Journal of Thoracic Oncology Volume 10, Number 8, August 2015 Invasive Mucinous Adenocarcinoma of the Lung most frequent driver in our mucinous cohort (63%), as have been previously described. 3,6 In terms of specific KRAS mutations, G12C is most common in lung adenocarcinoma in general, but G12D and G12V were most common in our cohort. G12D and G12V are most common in colorectal and pancreatobiliary carcinomas, 20 suggesting that invasive mucinous adenocarcinoma of lung may have more in common with pancreatobiliary and intestinal tract cancers. Interestingly, pancreatic adenocarcinomas metastatic to the lung often exhibit mucinous lepidic pattern and can be difficult to differentiate from primary invasive mucinous adenocarcinomas of the lung for the pathologist. 21 Following KRAS mutations, unique gene fusions were the next most common drivers. Three groups have recently reported CD74-NRG1 fusions in invasive mucinous adenocarcinoma of the lung. 10 12 CD74-NRG1 fusion protein activates ERBB3 and the PI3K-AKT signaling pathway through ERBB2 and ERBB3 dimerization. 10 Thus, patients with CD74-NRG1 rearranged lung cancers may be treated by blocking ligand-receptor interactions or by inhibition of ERBB receptors. We identified another NRG1 fusion with a novel partner, VAMP2-NRG1 fusion. Vesicle-associated membrane protein is involved in the docking and fusion of synaptic vesicles. The epidermal growth factor like domain of NRG1 was preserved in this fusion protein. We also identified a TRIM4-BRAF fusion. TRIM24-BRAF fusion has been recently described in invasive mucinous adenocarcinoma of the lung 11 ; however, to our knowledge, TRIM4 is a novel partner for BRAF in all cancers. Previous studies indicate that patients with cancer harboring BRAF fusion can be treated with RAF or MEK inhibitors. 22,23 We could not find CD74- NRG1 or other BRAF fusions in another cohort of 192 cases, suggesting their specificity for invasive mucinous adenocarcinoma (data not shown). The MPRIP-NTRK1 and CD74-NTRK1 fusions have previously been identified in lung adenocarcinoma. 24 The TPM3-NTRK1 fusion was reported in papillary thyroid carcinoma and colorectal adenocarcinoma, but the case in our cohort describes the first TPM3-NTRK1 fusion in lung cancer. 25,26 Given that the prior success of treatment with kinase inhibitors in patients with cancers harboring ALK and ROS1 fusions, NTRK1 fusions can also be a therapeutic target to selective NTRK1 inhibitors. 24 Nakaoku et al. 11 identified a ERBB4 fusion in their cohort, but none were identified in our cohort. Thus, this may be due to low frequency of ERBB4 fusion. We found that TP53 mutations are rare in invasive mucinous adenocarcinomas, with a TP53 mutation identified in only two out of the 50 cases examined. We hypothesize that the lack of TP53 mutations may explain in part why invasive mucinous adenocarcinomas (even those arising in smokers) have a lower mutational burden than nonmucinous tumors. Lung adenocarcinomas in general have one of the highest mutational burdens of any tumor types, likely due to chronic DNA damage resulting from smoking. Our data suggest that not smoking but rather the TP53 status may be the main contributor to the mutation burden in invasive mucinous adenocarcinomas that seem to be genetically simple with one principal driver. If there is indeed dependence on one driver, one might expect these tumors to respond well to agents that target the driver. We have anecdotal evidence of complete and durable responses to crizotinib from two invasive mucinous adenocarcinomas. This includes one with an ALK fusion and a second with an ROS1 fusion. Survival data for invasive mucinous adenocarcinoma have been limited due to its low incidence, and the results of the few published reports have been conflicting. 7,8 The most recent reports of lung adenocarcinomas classified in accordance with the IASLC/ATS/ERS classification scheme have excluded invasive mucinous adenocarcinomas from survival analysis because their number is limited and the IASLC/ATS/ ERS classification recommended that patients with variant subtypes be separated from those with conventional invasive adenocarcinoma. 8,27 Yoshizawa et al. 7 reported recurrence-free survival of 514 stage I cases including 13 invasive mucinous adenocarcinomas. They classified invasive mucinous adenocarcinoma in the high grade group based on a high incidence of recurrence. In our study, however, there was no significant difference in recurrence-free survival between invasive mucinous adenocarcinoma and the others, even though mucinous adenocarcinoma showed a tendency for better recurrencefree survival. Interestingly, all recurrences were limited to the lungs without extrapulmonary metastases in our invasive mucinous adenocarcinoma cohort. This finding suggests that invasive mucinous adenocarcinomas may not be aggressive tumors. CONCLUSION In conclusion, our data suggest that invasive mucinous adenocarcinoma is typified by (1) frequent KRAS mutations and a growing list of gene fusions, but rare TP53 mutations, (2) a low mutation burden overall, and (3) a recurrence-free survival that is at least as long as nonmucinous tumors, with recurrences limited to the lungs. Invasive mucinous adenocarcinoma of the lung is a disease that, due to its apparently limited number of associated driver mutations, may be expected to show durable responses to targeted agents. REFERENCES 1. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9 29. 2. Travis WD, Brambilla E, Noguchi M, et al. International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol 2011;6:244 285. 3. Casali C, Rossi G, Marchioni A, et al. A single institution-based retrospective study of surgically treated bronchioloalveolar adenocarcinoma of the lung: clinicopathologic analysis, molecular features, and possible pitfalls in routine practice. J Thorac Oncol 2010;5:830 836. 4. Kadota K, Yeh YC, D Angelo SP, et al. Associations between mutations and histologic patterns of mucin in lung adenocarcinoma: invasive mucinous pattern and extracellular mucin are associated with KRAS mutation. Am J Surg Pathol 2014;38:1118 1127. 5. Marchetti A, Buttitta F, Pellegrini S, et al. Bronchioloalveolar lung carcinomas: K-ras mutations are constant events in the mucinous subtype. J Pathol 1996;179:254 259. 6. Finberg KE, Sequist LV, Joshi VA, et al. Mucinous differentiation correlates with absence of EGFR mutation and presence of KRAS mutation in lung adenocarcinomas with bronchioloalveolar features. J Mol Diagn 2007;9:320 326. 7. Yoshizawa A, Motoi N, Riely GJ, et al. Impact of proposed IASLC/ATS/ ERS classification of lung adenocarcinoma: prognostic subgroups and 1161

Shim et al. Journal of Thoracic Oncology Volume 10, Number 8, August 2015 implications for further revision of staging based on analysis of 514 stage I cases. Mod Pathol 2011;24:653 664. 8. Warth A, Muley T, Meister M, et al. The novel histologic International Association for the Study of Lung Cancer/American Thoracic Society/ European Respiratory Society classification system of lung adenocarcinoma is a stage-independent predictor of survival. J Clin Oncol 2012;30:1438 1446. 9. Cancer Genome Atlas Research Network. Comprehensive molecular profiling of lung adenocarcinoma. Nature 2014;511:543 550. 10. Fernandez-Cuesta L, Plenker D, Osada H, et al. CD74-NRG1 fusions in lung adenocarcinoma. Cancer Discov 2014;4:415 422. 11. Nakaoku T, Tsuta K, Ichikawa H, et al. Druggable oncogene fusions in invasive mucinous lung adenocarcinoma. Clin Cancer Res 2014;20:3087 3093. 12. Gow CH, Wu SG, Chang YL, Shih JY. Multidriver mutation analysis in pulmonary mucinous adenocarcinoma in Taiwan: identification of a rare CD74-NRG1 translocation case. Med Oncol 2014;31:34. 13. Gao J, Aksoy BA, Dogrusoz U, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cbioportal. Sci Signal 2013;6:pl1. 14. Cerami E, Gao J, Dogrusoz U, et al. The cbio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401 404. 15. Dias-Santagata D, Akhavanfard S, David SS, et al. Rapid targeted mutational analysis of human tumours: a clinical platform to guide personalized cancer medicine. EMBO Mol Med 2010;2:146 158. 16. Sequist LV, Heist RS, Shaw AT, et al. Implementing multiplexed genotyping of non-small-cell lung cancers into routine clinical practice. Ann Oncol 2011;22:2616 2624. 17. Han SW, Kim TY, Hwang PG, et al. Predictive and prognostic impact of epidermal growth factor receptor mutation in non-small-cell lung cancer patients treated with gefitinib. J Clin Oncol 2005;23:2493 2501. 18. Pao W, Wang TY, Riely GJ, et al. KRAS mutations and primary resistance of lung adenocarcinomas to gefitinib or erlotinib. PLoS Med 2005;2:e17. 19. Zheng Z, Liebers M, Zhelyazkova B, et al. Anchored multiplex PCR for targeted next-generation sequencing. Nat Med 2014;20:1479 1484. 20. Vasan N, Boyer JL, Herbst RS. A RAS renaissance: emerging targeted therapies for KRAS-mutated non-small cell lung cancer. Clin Cancer Res 2014;20:3921 3930. 21. Krasinskas AM, Chiosea SI, Pal T, Dacic S. KRAS mutational analysis and immunohistochemical studies can help distinguish pancreatic metastases from primary lung adenocarcinomas. Mod Pathol 2014;27:262 270. 22. Palanisamy N, Ateeq B, Kalyana-Sundaram S, et al. Rearrangements of the RAF kinase pathway in prostate cancer, gastric cancer and melanoma. Nat Med 2010;16:793 798. 23. Hutchinson KE, Lipson D, Stephens PJ, et al. BRAF fusions define a distinct molecular subset of melanomas with potential sensitivity to MEK inhibition. Clin Cancer Res 2013;19:6696 6702. 24. Vaishnavi A, Capelletti M, Le AT, et al. Oncogenic and drug-sensitive NTRK1 rearrangements in lung cancer. Nat Med 2013;19:1469 1472. 25. Butti MG, Bongarzone I, Ferraresi G, Mondellini P, Borrello MG, Pierotti MA. A sequence analysis of the genomic regions involved in the rearrangements between TPM3 and NTRK1 genes producing TRK oncogenes in papillary thyroid carcinomas. Genomics 1995;28:15 24. 26. Ardini E, Bosotti R, Borgia AL, et al. The TPM3-NTRK1 rearrangement is a recurring event in colorectal carcinoma and is associated with tumor sensitivity to TRKA kinase inhibition. Mol Oncol 2014;8:1495 1507. 27. Hung JJ, Yeh YC, Jeng WJ, et al. Predictive value of the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society classification of lung adenocarcinoma in tumor recurrence and patient survival. J Clin Oncol 2014;32:2357 2364. 1162