European Journal of Cardio-Thoracic Surgery 49 (2016) 1144 1149 doi:10.1093/ejcts/ezv288 Advance Access publication 30 August 2015 ORIGINAL ARTICLE Cite this article as: Zhao Y, Shi J, Fan L, Hu D, Yang J, Zhao H. Surgical treatment of thymoma: an 11-year experience with 761 patients. Eur J Cardiothorac Surg 2016;49:1144 9. Surgical treatment of thymoma: an 11-year experience with 761 patients Yang Zhao, Jianxin Shi, Limin Fan, Dingzhong Hu, Jun Yang and Heng Zhao* Department of Thoracic Surgery, Shanghai Chest Hospital, Shanghai Jiaotong University, Shanghai, China * Corresponding author. Department of Thoracic Surgery, Shanghai Chest Hospital, No. 241 West Huaihai Road, Shanghai 200030, China. Tel: +86-21-62821990-2908; fax: +86-21-52306656; e-mail: zhaoyang2003@sina.com (H. Zhao). Received 14 January 2015; received in revised form 14 July 2015; accepted 20 July 2015 Abstract OBJECTIVES: Thymomas are rare, and information regarding their surgical outcomes and possible prognostic factors is limited. In this study, we aimed to determine the clinicopathological characteristics of thymoma and estimate independent predictors of both overall and disease-free survival in thymoma patients. METHODS: We carried out a retrospective review of the clinicopathological characteristics and prognostic factors in 761 consecutive patients with pathologically confirmed thymoma treated in Shanghai Chest Hospital between January 2001 and December 2011. Survival was calculated using the Kaplan Meier method and evaluated with log-rank tests. Multivariable analysis was performed using the Cox regression model. RESULTS: Complete follow-up information was available for 544 patients. The overall survival rate was 92.8% at 5 years and 90.5% at 10 years. The 5- and 10-year disease-free survival was 87.9 and 82.1%, respectively. On multiple Cox regression analysis, the Masaoka Koga clinical stage [odds ratio (OR), 2.057; 95% confidence interval (CI), 1.454 2.911; P < 0.01] and sex (OR, 2.244; 95% CI, 1.115 4.519; P = 0.02) were found to be independent predictors of overall survival. The Masaoka Koga clinical stage (OR, 2.127; 95% CI, 1.487 3.042; P < 0.01) and completeness of resection (OR, 2.935; 95% CI, 1.410 6.109; P < 0.01) predicted disease-free survival. CONCLUSIONS: The four-tiered Masaoka Koga clinical stage is the most important prognostic factor, predicting not only overall survival but also disease-free survival after thymoma resection. Completeness of resection predicts disease-free survival, and the World Health Organization histological classification may not have significant prognostic implications. Keywords: Thymoma Classification Prognosis Staging Surgical resection INTRODUCTION Thymomas are uncommon tumours of epithelial origin. They are characterized by remarkable morphological heterogeneity and variable clinical behaviour and have long been a topic of much interest among thoracic surgeons and pathologists. Surgery remains the mainstay of thymoma treatment, and many studies have suggested that the Masaoka staging, completeness of resection and World Health Organization (WHO) classification are important prognostic factors in thymoma. However, because of the relative rarity of this disease, most prior series were limited by inadequate numbers of patients, study periods spanning multiple decades, inadequate statistical analyses or a focus on overall survival only. In addition, the usefulness of the prognostic factors of thymoma in the Asian population has been inadequately assessed. In the present study, we used a comparable, uniform technical approach and retrospectively reviewed a series of 761 thymoma patients treated over the last decade to estimate independent predictors of both overall and disease-free survival. PATIENTS AND METHODS We conducted a retrospective review of all consecutive patients who had pathologically confirmed thymoma and were treated in Shanghai Chest Hospital between January 2001 and December 2011. The patient characteristics and outcomes were abstracted from the medical records. Resection status was determined from the operative notes and the pathology reports. If complete tumour resection was possible, every effort was made to achieve this, including superior vena cava resection and reconstruction, and even pleural pneumonectomy. Incomplete resection was performed in the case of tumour radicality after weighing the benefits and risks of the procedure. In our institute, video-assisted thoracic surgery procedures have been available since August 2008, while robotic surgery was introduced in May 2009. The final pathological staging was performed according to the Masaoka Koga system [1]. The indications for postoperative radiotherapy and chemotherapy were based on the surgeon s subjective assessment of the risk of recurrence (mainly based on the extent of radical extirpation, clinical The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
Y. Zhao et al. / European Journal of Cardio-Thoracic Surgery 1145 stage and clinical appropriateness). Institutional review board approval was granted for this retrospective review along with a waiver of patient consent. Most patients were followed up primarily via chest computed tomography scanning at the outpatient clinic at 3-month intervals for the first year, 6-month intervals for the subsequent year and yearly thereafter. Survival data not available in the medical records were obtained by contacting the patients or their families by telephone or mail. Overall survival was defined as the interval from the date of the treatment to any death or the last follow-up. Disease-free survival was defined as the interval from the date of the treatment to death, any recurrence or the last follow-up. All the outcome measures were defined according to the International Thymic Malignancy Interest Group suggestions [2], and recurrence was diagnosed if proved by biopsy or by unequivocal evidence of tumour masses (newly appearing metastases or local recurrence) with a tendency to grow during further surveillance. In our study, patients with complete lack of any data post-surgery were considered lost follow-up. The last follow-up was finished at the end of October 2013. A total of 761 patients was enrolled, of whom 544 had complete follow-up data. The median duration of follow-up was 58.0 months (range, 22 153 months). Potentially relevant factors in patient characteristics were compared with Student s t-test, Mann Whitney U-test or Fisher s exact test as appropriate. The overall survival and disease-free survival were estimated with the Kaplan Meier method and compared with the log-rank test. Variables found to be significant on univariable analysis (defined as a probability value of <0.15) were included in the multivarable analysis using the Cox proportional hazards model after backward stepwise Wald elimination. A probability value of <0.05 on multivariable analysis was considered significant. We used SPSS software (version 20.0, SPSS Inc., Chicago, IL, USA) for statistical analyses. RESULTS Clinicopathological characteristics All 761 patients, consisting of 390 female patients and 371 male patients, had sufficiently detailed files with complete clinical and pathological data. The demographic characteristics of the study population are shown in Table 1. There was a slight female sex predilection (51.2%), and the median age at diagnosis was 53.0 years (range, 15 83 years). Mixed thymoma was diagnosed in 33.8% patients (257/761), and the histological type was classified as the more advanced one. The median tumour size was 7.0 cm (range, 1 24 cm). There was no correlation between tumour size and the WHO histological classification (P = 0.17), but larger tumours indicated a more advanced clinical stage (P < 0.01). Tumour size also significantly correlated with the completeness of resection (P < 0.01). The WHO histological classification was found to be highly associated with the Masaoka Koga clinical stage (P < 0.01; Table 2). Of the 761 patients, 646 (84.9%) underwent complete resection. Completeness of resection was significantly associated with the WHO histological classification (P < 0.01) and the Masaoka Koga clinical stage (P < 0.01). Myasthenia gravis (MG) was present in 164 of the 761 patients (21.6%). MG was not significantly correlated with sex (P = 0.72) or age (P = 0.09). The tumour size was significantly smaller in MG patients than in patients without MG (P < 0.01). In addition, MG was Table 1: Clinical features of 761 patients who underwent thymoma treatment Variable Number Frequency Age (years) Mean ± SD 51.7 ± 12.1 Gender Male 371 48.8% Female 390 51.2% Symptom Yes 431 56.6% No 330 43.4% Size (cm) Mean ± SD 7.3 ± 3.1 WHO histological type A 38 5.0% AB 173 22.7% B1 89 11.7% B2 218 28.6% B3 243 31.9% Masaoka Koga stage I 380 49.9% II 176 23.1% III 156 20.5% IV 49 6.4% Myasthenia gravis Yes 164 21.6% No 597 78.4% Pure red cell aplasia Yes 16 2.1% No 745 97.9% Surgical procedures Sternotomy 485 63.7% Thoracotomy 205 26.9% Video-assisted approach 64 8.4% Cervical approach 5 0.7% Robotic approach 2 0.3% Resection R0 646 84.9% R1 38 5.0% R2 68 8.9% Biopsy 9 1.2% WHO: World Health Organization. Table 2: The relationship between Masaoka Koga clinical stage and World Health Organization histological type WHO type Masaoka Koga stage Total Stage I Stage II Stage III Stage IV A 27 9 0 2 38 AB 124 47 2 0 173 B1 54 19 15 1 89 B2 112 63 32 11 218 B3 63 38 107 35 243 Total 380 176 156 49 761 WHO: World Health Organization. particularly frequent in patients with type B2 or B3 thymoma (P < 0.01). However, MG was not correlated with the Masaoka Koga clinical stage (P = 0.59) or completeness of resection (P = 0.75). THORACIC
1146 Y. Zhao et al. / European Journal of Cardio-Thoracic Surgery Among the 112 MG patients with follow-up data, 43 patients (38.4%) achieved complete remission of MG, 52 (46.4%) experienced an improvement in MG symptoms, 8 (7.1%) exhibited no change in MG symptoms and 9 patients (8.0%) showed a deterioration in MG symptoms after the operation. In addition, 9 patients developed an MG crisis, and 25 patients without MG developed MG after the operation. The rate of remission or improvement of MG after thymectomy had no relationship with the WHO histological classification or Masaoka Koga clinical stage (P = 0.63 and P = 0.34, respectively). Of the 761 patients, 16 (2.1%) had pure red cell aplasia. The presence of pure red cell aplasia was not correlated with WHO histological classification (P = 0.46), Masaoka Koga clinical stage (P = 0.27) or completeness of resection (P =0.74). Follow-up information was available for 544 patients (71.5%). After operation, 240 patients (44.1%) underwent single-modality treatment with radiotherapy, while 14 patients (2.6%) underwent chemotherapy alone. In addition, 47 patients (8.6%) received a combination of both modalities and 243 patients (44.7%) received no adjuvant treatment. Adjuvant therapy was significantly associated with WHO histological classification (P < 0.01), Masaoka Koga clinical stage (P < 0.01) and completeness of resection (P < 0.01). Secondary neoplasia developed in 1.8% of our thymoma patients. Its occurrence had no relationship with the age (P = 0.36), sex (P = 0.60), tumour size (P = 0.06), histological type (P = 0.06), clinical stage (P = 0.27), completeness of resection (P = 0.34), MG (P = 0.96), pure red cell aplasia (P = 0.59) or adjuvant radiotherapy (P = 0.27). Univariable survival analysis The overall survival rate was 92.8% at 5 years and 90.5% at 10 years. The disease-free survival rate was 87.9% at 5 years and 82.1% at 10 years. The median disease-free survival of patients with complete resection, macroscopically complete resection and incomplete resection was 142.1, 67.0 and 63.0 months, respectively. The mean survival time was 140.7 ± 1.8 months, and the mean time to post-operative recurrence after the initial surgery was 131.9 ± 2.3 months. Sixty-eight patients (12.5%) experienced progression after the operation. The median time to recurrence in these patients was 54.0 months (range, 1 152 months). Pure local recurrence was evident in 11 patients (16.2%), and regional relapse occurred in 26 patients (38.2%). In addition, 31 patients (45.6%) developed distant metastasis, affecting the lung (21), bone (6), liver (4), brain (2), skin (2) and adrenal glands (1). Although most recurrences occurred in patients with type B3 (41/68, 60.3%), the incidence of relapse had no relationship with the WHO histological subtype (P = 0.70) or Masaoka Koga clinical stage (P = 0.96). At the conclusion of the study, 459 patients (84.4%) were alive with no evidence of the disease and 46 were alive with the disease (8.5%). In addition, 22 patients had died of the disease (4.0%) and 17 patients (3.1%) had died of disease-unrelated causes. The significant predictors of overall survival and disease-free survival as calculated by univariable analysis are listed in Table 3. Multiple Cox regression analysis The Masaoka Koga clinical stage [odds ratio (OR), 2.057; 95% confidence interval (CI), 1.454 2.911; P < 0.01; Fig. 1] and sex (OR, 2.244; 95% CI, 1.115 4.519; P = 0.02; Fig. 2) significantly affected overall survival (Table 4). The Masaoka Koga clinical stage (OR, 2.127; 95% CI, 1.487 3.042; P < 0.01; Fig. 3) and completeness of resection (OR, 2.935; 95% CI, 1.410 6.109; P < 0.01; Fig. 4) were predictors of disease-free survival. Surprisingly, the WHO histological classification failed to be an independent predictive determinant of both overall survival and disease-free survival, as analysed by the Cox regression model. DISCUSSION The objective of this study was to characterize the clinicopathological features of patients with thymomas and identify independent predictors of survival. Our study is the largest single-institution study on thymomas thus far, and the principal findings of the analysis of the clinical data gathered in this study are as follows: (i) the Masaoka Koga clinical stage is the primary determinant of outcomes and effectively predicts both overall survival and diseasefree survival; (ii) the completeness of resection predicts disease-free survival; (iii) surprisingly, the WHO histological classification does not have significant prognostic implications for predicting the outcomes of thymoma patients; and (iv) female sex is associated with increased overall survival. The four-tiered Masaoka staging system was proposed in 1981 (based on 93 patients) [3], and a modification was suggested by Koga et al. in 1994 [1]. This anatomical classification assesses the degree of tumour invasiveness, including macro- and microscopic invasion of the capsule and surrounding structures. Consistent with previous reports [4 6], the Masaoka Koga clinical stage was an excellent predictor of prognosis in our study populations. Our study reiterates that the current clinical staging system plays an important role in deciding therapeutic strategy and estimating prognosis. We think that future research should focus on stratifying patients for stage-specific therapies. The WHO classification of thymomas was first introduced in 1999 and revised in 2004. Thymomas are now stratified into five entities (types A, AB, B1, B2 and B3) on the basis of the morphology of epithelial cells and the lymphocyte-to-epithelial cell ratio [7]. This classification is the subject of intense discussion on whether morphological features are a reliable prognostic factor and reflect oncologic behaviour. Furthermore, this classification is to some extent arbitrary, because thymomas often show morphological heterogeneity within the same tumour. Our results clearly demonstrated that neither overall survival nor disease-free survival can be predicted by the current WHO histological classification. In other words, no particular WHO type of thymoma would help clinicians to stratify patients into various therapeutic modalities. The likely explanation is that there was a striking correlation between the WHO classification and the Masaoka Koga clinical stage (P < 0.01). Our multivariate Cox regression analysis indicated that the factor WHO histological type was merely associated with the more important prognostic factor Masaoka Koga clinical stage and did not have an independent effect on survival. Similar to our findings, the more recent report that all thymomas have malignant potential seems to contradict the rationale of the current histological classification [8 11]. Chalabreysse et al. [12] reported the case of a patient with a type A thymoma that recurred 9 years after surgery; this patient succumbed to the disease despite radiotherapy and chemotherapy. Moran et al. [8] reported the case of 41 patients with invasive spindle cell thymomas (type A), and found that 6 were in stage III,
Y. Zhao et al. / European Journal of Cardio-Thoracic Surgery 1147 Table 3: Univariable analysis of various potential prognostic factors for overall and disease-free survival in 544 thymoma patients Variables Number of patients Overall survival Disease-free survival Mean ± SD survival time (months) P-value a Median ± SD survival time (months) P-value a Gender Male 282 136.3 ± 2.8 0.010 b 130.6 ± 3.3 0.402 Female 262 145.3 ± 2.1 133.2 ± 3.3 Age (years) <51.7 240 140.4 ± 2.8 0.935 125.6 ± 3.7 0.005 b 51.7 304 140.7 ± 2.4 137.6 ± 2.8 Symptoms Yes 296 136.9 ± 2.7 0.019 b 130.0 ± 3.2 0.411 No 248 146.1 ± 1.8 134.5 ± 3.2 Tumour size (cm) <7.3 330 141.1 ± 2.2 0.095 133.5 ± 2.7 0.096 7.3 214 137.3 ± 3.1 127.5 ± 3.8 WHO classification A 33 122.1 ± 5.5 0.004 b 124.8 ± 3.2 <0.001 b AB 125 149.6 ± 1.7 150.7 ± 1.3 B1 65 131.7 ± 3.0 118.2 ± 5.8 B2 142 143.1 ± 2.9 133.0 ± 4.7 B3 179 130.5 ± 4.0 116.8 ± 4.6 Masaoka Koga stage I 276 147.9 ± 1.5 <0.001 b 146.6 ± 1.7 <0.001 b II 127 137.8 ± 3.6 139.3 ± 3.6 III 104 135.6 ± 4.4 104.4 ± 6.7 IV 37 91.7 ± 9.4 53.1 ± 9.5 Complete resection Yes 461 144.0 ± 1.7 <0.001 b 136.4 ± 2.2 <0.001 b No 82 117.7 ± 6.4 73.4 ± 10.6 Radiotherapy Yes 287 138.4 ± 2.5 0.100 123.7 ± 3.4 <0.001 b No 257 127.2 ± 2.0 127.0 ± 1.9 Chemotherapy Yes 61 113.6 ± 7.5 <0.001 b 94.2 ± 8.7 <0.001 b No 483 143.4 ± 1.7 136.1 ± 2.3 Cyst degradation Yes 112 139.6 ± 3.0 0.390 133.3 ± 4.1 0.163 No 432 140.0 ± 2.1 130.5 ± 2.7 Myasthenia gravis Yes 112 133.2 ± 5.0 0.039 b 132.7 ± 4.9 0.926 No 432 142.7 ± 1.8 131.7 ± 2.6 Pure red cell aplasia Yes 15 116.8 ± 22.1 0.278 c 0.185 No 529 141.0 ± 1.8 c Relapse type Local recurrence 11 32.3 ± 9.7 0.644 Regional recurrence 26 35.4 ± 5.4 Distant recurrence 31 29.3 ± 4.7 THORACIC WHO: World Health Organization. a Kaplan Meier method, log-rank test. b Statistically significant. c All data are censored. and 1 was in stage IV. Nonaka and Rosai [13] reported 13 cases of type A thymoma displaying atypical features, such as increased mitotic activity, mild-to-moderate nuclear atypia and/or small scattered necrotic foci. They suggested that type A thymomas may exhibit a more variable cytological appearance than generally believed and could be subdivided the way type B tumours have been (A1, A2 and A3). In our study, 38 patients (5.0%) had type A thymoma. Among them, nine patients were in stage II, and two were in stage IV. One patient diagnosed with a type A thymoma electroscopically developed multiple metastases to the brain, lung and bone. Furthermore, the survival curve of patients with type A thymomas was distinctive, with an intermediate location between the curves for type B3 thymomas and other thymoma types. This finding is similar to that reported by Rieker et al. [14]. We all emphasize that type A thymoma is not always benign with no risk of recurrence or metastases, and some of these tumours can behave aggressively. The prognostic determinants of thymoma have been the subject of lively discussion over the past 20 years. In the present study, female sex was identified as a significant favourable predictor of overall survival in our multivariate Cox regression analysis (OR, 2.244; 95% CI, 1.115 4.519; P = 0.02). The reason for this is unclear. Perhaps there are unidentified differences in the biology of thymomas between men and women, or some other variables
1148 Y. Zhao et al. / European Journal of Cardio-Thoracic Surgery Table 4: Multiple Cox regression analysis of overall and disease-free survival in 544 thymoma patients a Variable OR (95% CI) P-value b Overall survival Gender 2.244 (1.115 4.519) 0.024 c Symptom 1.619 (0.740 3.544) 0.228 Tumour size 1.393 (0.718 2.705) 0.327 WHO classification 1.173 (0.837 1.643) 0.354 Masaoka Koga stage 2.057 (1.454 2.911) <0.001 c Complete resection 1.093 (0.437 2.734) 0.849 Radiotherapy 0.709 (0.332 1.516) 0.375 Chemotherapy 2.044 (0.971 4.304) 0.060 Myasthenia gravis 1.933 (0.991 3.772) 0.053 Disease-free survival Age 0.649 (0.392 1.076) 0.094 Tumour size 0.918 (0.556 1.517) 0.738 WHO classification 1.107 (0.834 1.470) 0.482 Masaoka Koga stage 2.127 (1.487 3.042) <0.001 c Completeness of resection 2.935 (1.410 6.109) 0.004 c Radiotherapy 1.484 (0.789 2.790) 0.220 Chemotherapy 1.169 (0.662 2.064) 0.591 Figure 1: Kaplan Meier analysis of overall survival in thymoma patients stratified by Masaoka Koga clinical stage. P-values were obtained using the log-rank test. CI: confidence interval; OR: odds ratio; WHO: World Health Organization. a Variables with a probability value of <0.15 on univariable analysis. b Cox proportional hazards regression analysis. c Statistically significant. Figure 2: Kaplan Meier analysis of overall survival in thymoma patients divided according to sex. P-values were obtained using the log-rank test. affecting the outcome have not yet been assessed. We also noticed that the sex distribution was not balanced between patients with follow-up data and those who were lost to follow-up (P < 0.01), which complicated our conclusion on this issue. Nevertheless, this result may have a potential impact on clinical decision-making, such as subjecting female patients to more aggressive adjuvant therapy. Similar to the findings of Bae et al. (41/305, 13.4%) [15], 68 of our patients (12.5%) experienced progression after the operation. Of note, 31 patients (5.7%) developed distant metastases, primarily in the lungs. This is consistent with the finding that the lung is the most common site of distant haematogenous metastases in Figure 3: Kaplan Meier analysis of disease-free survival in thymoma patients stratified by the Masaoka Koga clinical stage. P-values were obtained using the log-rank test. thymoma patients [16, 17]. In our series, the frequency of death from causes unrelated to the thymoma was similar to the rate of thymoma-associated deaths (3.1 and 4.0%, respectively), indicating the indolent nature of thymomas. Contrary to previous studies [15], we did not find that patients with local recurrence had better survival than those with regional or distant recurrence (P = 0.64).
Y. Zhao et al. / European Journal of Cardio-Thoracic Surgery 1149 may not have significant prognostic implications. To advance our knowledge of these fascinating tumours, global collaboration guided by the International Thymic Malignancy Interest Group should be more focused. Conflict of interest: none declared. REFERENCES Figure 4: Kaplan Meier analysis of disease-free survival in thymoma patients stratified by completeness of resection. P-values were obtained using the log-rank test. The phenomenon of second primary malignancies is well described in thymoma patients, with an 18-fold increased incidence [18]. In our cohorts, 1.8% thymoma patients developed a second neoplasm, which is comparable to the incidence rate of 1.3% from the SEER database [19]. It is unsurprising that such an association exists, given the role of the thymus in immune function, but the precise mechanism is unclear. Our data also excluded the possibility of adjuvant radiotherapy as a predisposing factor (P = 0.27), in line with other reports [18]. Our study is limited by the small number of cases, which precluded the acquisition of informative data on interesting issues, such as whether any subset of malignancies is associated with an increased risk. The study results must be interpreted with caution, given the inherent biases associated with a retrospective study design and the experience of a single centre. Owing to the primary selection bias in treatment allocation, our study cannot define the role of adjuvant treatment. The strengths of our study are the one-decade time span, the use of more contemporary clinical practices and a comparatively uniform treatment strategy. This is the largest sample size of thymoma patients in a single-institution study ever, so the results of multivariable analysis were more reliable. Furthermore, we calculated both both overall and disease-free survival, because the former is insufficiently sensitive to serve as an appropriate outcome measure and the latter is a more clinically relevant end point. In conclusion, our results reiterate that the four-tiered Masaoka Koga clinical staging system is the most important prognostic factor, predicting not only overall survival but also disease-free survival after thymoma resection. The completeness of resection predicts disease-free survival, and the WHO histological classification [1] Koga K, Matsuno Y, Noguchi M, Mukai K, Asamural H, Goya T et al. A review of 79 thymomas: modification of staging system and reappraisal of conventional division into invasive and non-invasive thymoma. Pathol Int 1994;44:359 67. [2] Huang J, Detterbeck FC, Wang Z, Loehrer PJ Sr. Standard outcome measures for thymic malignancies. J Thorac Oncol 2011;6:S1691 7. [3] Masaoka A, Monden Y, Nakahara K, Tanioka T. Follow-up study of thymomas with special reference to their clinical stages. Cancer 1981;48: 2485 92. [4] Kondo K, Monden Y. Therapy for thymic epithelial tumors: a clinical study of 1,320 patients from Japan. Ann Thorac Surg 2003;76:878 85. [5] Kondo K, Yoshizawa K, Tsuyuguchi M, Kimura S, Sumitomo M, Morita J et al. WHO histologic classification is a prognostic indicator in thymoma. Ann Thorac Surg 2004;77:1183 8. [6] Lowenhaupt E. Tumors of the thymus in relation to the thymic epithelial anlage. Cancer 1948;1:547 63. [7] Moller-Hermelink H, Engel P, Kuo T. Pathology and genetics of tumours of the lung, pleura, thymus, and heart. In: Travis WD (ed). World Health Organization Classification of Tumours. Lyon: IARC Press, 2004, 344. [8] Moran CA, Kalhor N, Suster S. Invasive spindle cell thymomas (WHO type A): a clinicopathologic correlation of 41 cases. Am J Clin Pathol 2010;134: 793 8. [9] Kim DJ, Yang WI, Choi SS, Kim KD, Chung KY. Prognostic and clinical relevance of the World Health Organization schema for the classification of thymic epithelial tumors: a clinicopathologic study of 108 patients and literature review. Chest 2005;127:755 61. [10] Detterbeck FC. Clinical value of the WHO classification system of thymoma. Ann Thorac Surg 2006;81:2328 34. [11] Marchevsky AM, Gupta R, McKenna RJ, Wick M, Moran C, Zakowski MF et al. Evidence-based pathology and the pathologic evaluation of thymomas: the World Health Organization classification can be simplified into only 3 categories other than thymic carcinoma. Cancer 2008;112:2780 8. [12] Chalabreysse L, Roy P, Cordier JF, Loire R, Gamondes JP, Thivolet-Bejui F et al. Correlation of the WHO schema for the classification of thymic epithelial neoplasms with prognosis: a retrospective study of 90 tumors. Am J Surg Pathol 2002;26:1605 11. [13] Nonaka D, Rosai J. Is there a spectrum of cytologic atypia in type A thymomas analogous to that seen in type B thymomas? A pilot study of 13 cases. Am J Surg Pathol 2012;36:889 94. [14] Rieker RJ, Hoegel J, Morresi-Hauf A, Hofmann WJ, Blaeker H, Penzel R et al. Histologic classification of thymic epithelial tumors: comparison of established classification schemes. Int J Cancer 2002;98:900 6. [15] Bae MK, Byun CS, Lee CY, Lee JG, Park IK, Kim DJ et al. Clinical outcomes and prognosis of recurrent thymoma management. J Thorac Oncol 2012; 7:1304 14. [16] Ahmad U, Huang J. Current readings: the most influential and recent studies involving surgical management of thymoma. Semin Thoracic Cardiovasc Surg 2013;25:144 9. [17] Bott MJ, Wang H, Travis W, Riely GJ, Bains M, Downey R et al. Management and outcomes of relapse after treatment for thymoma and thymic carcinoma. Ann Thorac Surg 2011;92:1984 92. [18] Weksler B, Nason KS, Mackey D, Gallagher A, Pennathur A. Thymomas and extrathymic cancers. Ann Thorac Surg 2012;93:884 9. [19] Engels EA. Epidemiology of thymoma and associated malignancies. J Thorac Oncol 2010;5:S260 5. THORACIC