Weissferdt, Annikka MD, FRCPath; Moran, Cesar A. MD
Primary tumors of the anterior mediastinum represent a heterogenous group of neoplasms that can arise from virtually any cell lineage. Apart from benign cystic lesions and hematopoietic tumors, the vast majority of tumors are malignant and derived from various cell types. The most common epithelial tumors in this anatomic location are thymic epithelial neoplasms, namely thymomas and thymic carcinomas. These tumors are characterized by a wide morphologic variability, which not only causes diagnostic challenges but has caused continuing controversy with respect to classification, treatment, and prognosis of these tumors. It is also important to highlight that as thymoma represents one of the most common tumors in this location, it is also the one for which most staging proposals have been put forward. Among these, the schema devised by Masaoka et al1 is one of the oldest and remains the most popular one. This system was originally proposed purely for the staging of thymomas but it has also been frequently applied to other mediastinal tumors such as thymic carcinoma or neuroendocrine carcinomas of the thymus, primarily due to a lack of more specific staging systems for the latter and despite the fact that these tumors are separate clinicopathologic entities with different biological behavior. However, over the years, the applicability of the Masaoka system not only for the purpose of staging nonthymomatous tumors but also thymomas themselves has been questioned and to this end new staging systems for thymic epithelial neoplasms have recently been developed. Another contentious issue pertains to the staging of mediastinal germ cell tumors. Although histologically similar in many respects to their gonadal counterparts, the different anatomic location requires a different staging approach to determine therapy choices and prognosis for these patients. However, no standardized staging exists for this group of tumors and patients often undergo treatment without the benefit of appropriate clinical or pathologic staging.
This review summarizes the advances in the staging of some of the most common anterior mediastinal neoplasms, namely thymoma, thymic carcinoma, neuroendocrine thymic carcinomas, and primary mediastinal germ cell tumors. Particular emphasis is paid to the most recent developments in this field.
STAGING OF THYMOMA
In 1979, Bergh et al2 were the first authors to present a staging schema for thymic epithelial neoplasms investigating a series of 43 thymomas. Their system consisted of 3 different stage groups based on operative findings and histologic examination that were divided according to whether the tumor was encapsulated, had pericapsular growth, or was invasive into surrounding organs (Table 1). Clinical follow-up showed a low recurrence rate and no tumor deaths in stages I and II at 10 years. Aggressive surgical approach was considered the best treatment modality.
One year later, Wilkins and Castleman3 provided a slight modification of the system proposed by Bergh et al2 by incorporating pleural or pericardial invasion into stage II rather than III (Table 1). In their series of 103 patients with thymoma, the 10-year survival rates were 67% for encapsulated tumors and 40% for invasive lesions.
Both of these systems were later criticized for the following aspects: the actual site of intrathoracic metastasis, inadequate description of invasion, too broad stratification of stage III, and lack of representation of lymphogenous or hematogenous metastasis.
A new staging system was then proposed in 1981 by a Japanese group led by Masaoka,1 focusing primarily on the surgical extent of the tumor. On the basis of a description of 96 cases, thymomas were divided into 4 different stage groups according to the macroscopic and microscopic findings (Table 1). The authors were able to show that the use of this system produced a stepwise decline of survival curves in association with the advance of clinical stage demonstrating its usefulness for the prognosis of these tumors. To date, Masaoka’s schema and the subsequent Koga-modified Masaoka staging systems4 (Table 1) remain the most commonly applied schemas for the staging of thymic epithelial neoplasms. However, over the years critical examination of these proposals has highlighted several discrepancies primarily relating to differences in perioperative (clinical) and pathologic examination. If one takes a closer look at the stage definitions, there are several areas of concern. For example, stage II-1 is defined as macroscopic invasion into surrounding fatty tissue or mediastinal pleura at perioperative inspection, even if subsequent histopathologic examination fails to show an invasive process. From a pathologist’s perspective, this makes little sense.5 In a similar manner, stage II-2, defined as microscopic invasion into but not through the capsule, is not a feature of invasive tumors in the pathologic sense.5–7 The next contentious issue relates to stages III and IVa. Here, stage III is defined as “macroscopic invasion into neighboring organs, that is, pericardium, great vessels, or lung,” whereas stage IVa is defined as “pleural or pericardial dissemination.” These 2 definitions are rather ambiguous and tumors that infiltrate the pleura or pericardium can be placed in either one of those categories. However, the stage definitions are not the only points that have raised concern. Over the years, several studies have demonstrated that the system does not provide appreciable prognostic separation between stages I and II8–12 and that tumors invading the mediastinal pleura have a higher risk of recurrence than other stage II tumors.6,13 Another point of concern alludes to the fact that the Masaoka system, although often applied to this purpose, does not lend itself to the staging of tumors that show different biological behavior such as thymic carcinomas or other malignant tumors that commonly show lymph node metastasis.14–16
In 1988, Curran et al17 presented a study of 103 thymoma cases using an adapted staging system based on the ones by Bergh et al2 and Masaoka et al1 (Table 1). The main objective of this study was to evaluate the role of mediastinal irradiation in the treatment of these tumors. By this method, 5-year survival and relapse-free survival rates of 67% and 100% for stage I, 86% and 58% for stage II, and 69% and 53% for stage III were observed, respectively. One of the most significant findings of this study was a lower recurrence rate for stage I compared with other tumor stages and the need for postoperative radiotherapy after total resection of stage II and III tumors.
Three years later, in 1991, a French group led by Gamondès et al18 investigated 65 thymomas using a staging system that was based on certain surgical and anatomic criteria that were proposed a decade earlier by the French Group d’Etudes des Tumeurs Thymiques (GETT)18 (Table 1). This classification was modeled after 2 major influential factors on outcome: the macroscopic appearance of the tumor and the type of intervention. Although the authors acknowledged that little distinction can be made between tumors in stages III and IV, they argued that both stages are associated with reserved prognosis. The most important observation in this study was that stage I tumors have by far the best prognosis (5 and 10 y survival of 96% and 88%, respectively for stage I, 84% and 56% for stage III, and 73% and 18% for stage IV).
Also in 1991, Yamakawa et al19 introduced a Tumor-Node-Metastasis (TNM) system for thymic epithelial neoplasms. This system was introduced after changes in surgical technique (“extended thymectomy” vs. “simple thymectomy”) led to the discovery that some cases showed lymph node metastasis in the anterior mediastinal adipose tissue. In this system, the T stage was modeled after the Masaoka system1 with newly added N and M categories (Table 2). This system was subsequently applied to a series of 207 thymomas, 13 thymic carcinomas, and 6 thymic carcinoid tumors. Although this system proved of little added value for staging thymomas, it was deemed more useful for thymic carcinomas and thymic carcinoid tumors, based on the fact that these tumors more often show metastatic disease in locally less advanced cases. This system could demonstrate that N and M factors influence survival more than the T stage and it was thought to be an excellent predictor for the prognosis of thymic epithelial neoplasms. More recent studies, however, have demonstrated a significant lack of statistical significance between the stages when applied to cases of thymic carcinoma.15,16
In 1994, Koga et al4 modified the Masaoka system1 after recognizing the controversy caused by some of the definitions used in the latter. The authors adapted the Masaoka system based on the most contentious issue: the fact that microscopic invasion into the capsule was classified as stage II-2, contrary to common pathologic understanding. Therefore, tumors invading into but not through the capsule were classified as stage I and a tumor infiltrating through the capsule and into normal thymic or fatty tissue as stage II-2 (Table 1). However, application of this 4-category system on 73 cases of thymoma revealed no significant differences between stages I and II and between stages III and IV, respectively. Although not very useful for thymoma, which could easily be divided into noninvasive and invasive tumors, the system was thought to be more suitable for thymic carcinoma. The modifications by Koga and colleagues provided a better definition of microscopic transcapsular invasion and also addressed the issue of “adherence” versus “invasion” for stage II. Despite this, several concerns regarding equivocal definitions and assessment remain. Some of these issues include:
* How to categorize tumors that lack a distinct or complete capsule.
* How to define and consolidate macroscopic invasion that is not present microscopically.
* A difference between macroscopic invasion and adherence to the mediastinal pleura or pericardium (stage II-2 vs. stage III).
* The extent of pericardial involvement in stage II-2 and III is ambiguously worded.
* Definition of pericardial invasion remains vague and tumors showing pericardial involvement may be categorized as stage III or IVa.
* How to distinguish separate foci of tumor from hematogenous metastasis.
* As in the original Masaoka system,1 difference of survival remains statistically insignificant.
* The applicability of this system to thymic carcinoma is at best equivocal, if not completely inadequate and inaccurate.
More recently, a new TNM staging system was presented to replace Masaoka’s system as the standard.20 In 2005, Bedini et al20 examined a total of 123 cases of thymic epithelial tumors using the so-called Istituto Nazionale Tumori (INT) system including 106 cases of thymoma and 17 thymic carcinomas. This system uses 11 definitions and stratifies cases into 3 stage groups depending on the extent of the disease (locally restricted vs. locally advanced vs. systemic disease). In addition, the completeness of surgical resection is assessed based on the suggestion that debulking procedures may be potentially curative (Table 2). The authors concluded that in contrast to Masaoka’s schema, the INT system seems applicable to all subtypes of the WHO nomenclature and has high prognostic distinctiveness. In addition, the authors concluded that the different stages offer distinct treatment options: surgery alone for limited disease, multimodality treatment for locally advanced disease, and primary chemotherapy plus radiotherapy as palliative or adjuvant treatment in cases of systemic disease. Eventually, however, such elaborate categorization was deemed too complicated to be practical enough21 and has not been adapted to a great extent.
The deficiencies of the preceding staging schemas have led to the proposal of a novel staging system by Moran et al.5 This system is exclusively designed for application to thymomas. In this study, only those cases derived from thymectomy were included (excluding any biopsy-only material) and in addition, a minimum of 5 histologic tumor slides were necessary to allow for proper subclassification of the tumors. On the basis of that study, no statistically significant correlation was obtained using any of the subtypes proposed by the WHO, leaving staging as the most important parameter to predict outcome. On the basis of the data derived from this material, a 4-tier staging system was proposed (Table 3; Figs. 1 A–F). When the proposed schema was applied to 250 cases of thymoma, statistically significant correlation was observed. This system offers several advantages over previous ones:
* Represents the largest and most representative series addressing the issue of thymoma staging.
* Contrary to its predecessors, this system provides a clear definition of encapsulated versus invasive thymomas, as well as providing clarification of the terms “pleural” and “pericardial involvement.”
* The issue of the so-called “drop metastasis” is addressed and defined as discontinuous tumor extension to the diaphragm.
* The concept of stage 0 is introduced to emphasize the similarity of these tumors with in situ malignancies of other organ sites, in which complete resection of the lesion will likely result in a complete cure.
* Results of the statistical analysis derived from these data demonstrated improved overall survival and rates of recurrence for patients with stage 0 and I tumors as opposed to patients with stage II and III tumors, indicating that the former group of patients may benefit from complete surgical resection alone whereas the latter group likely requires additional therapies.
* Seems to be more practical and reproducible based on clear definitions facilitating application to daily practice.
* Clearly demonstrates that staging is the single most important parameter to correlate with prognosis, and that the WHO schema does not provide significant information regarding clinical course.
STAGING OF THYMIC CARCINOMA
Traditionally, thymic carcinomas have always been staged according to the same criteria as for thymomas, despite the fact that these tumors represent a different clinicopathologic entity and follow a different biological course. This is largely due to the fact that thymic carcinomas are even rarer than thymomas preventing more comprehensive assessment of these tumors. Although in the past concerns have been raised that the majority of the proposed staging systems for thymoma or thymic epithelial neoplasms cannot reliably predict outcome for thymic carcinoma, only few studies have addressed this issue and investigated exclusively thymic carcinomas for the purpose of staging.
The first introduction of a staging system exclusively for thymic carcinoma was introduced by Tsuchiya et al15 in 1994 after recognizing that thymomas and thymic carcinomas display different biological behavior with different patterns of spread. On the basis of the higher incidence of metastasis among thymic carcinomas, the authors adapted the TNM system proposed by Yamakawa et al19 according to the biological characteristics of thymic carcinoma (Table 2). In this study, 16 primary thymic carcinomas were evaluated. Unfortunately, a statistically significant difference between the different stages was not achieved, a fact attributed to the small number of cases. A number of other concerns relate to the practicability of this system. Similar to several of the thymoma staging systems, the stage division between stages T3 and T4 is ill defined and may cause confusion. A more contentious issue, however, involves the lymph node categories. The presence of 4 different N categories seems rather elaborate for a type of tumor that is not universally excised in any standardized manner. In addition, formal lymph node dissection is not commonly undertaken.16 In addition, the 16 cases of thymic carcinoma in Tsuchiya’s study only encompassed 3 different tumor types (squamous cell carcinoma, undifferentiated carcinoma, and basaloid carcinoma) despite the fact that thymic carcinoma displays a great morphologic spectrum thereby restricting the validity of the study to a degree.
The necessity of a separate staging system for thymic carcinomas was recently reconfirmed in a study evaluating a large series of thymic carcinomas.16 The realization that neither the Masaoka et al1 nor the Tsuchiya et al15 systems lend themselves to the staging of this tumor type, have led to the proposal of a novel staging system by Weissferdt and Moran22 drawing from their experience with 33 of these tumors including 9 tumor subtypes (Table 4; Figs. 2A–C). The basis for this staging system was the fact that lymph node involvement is one of the most important prognostic factors in thymic carcinoma, hence a TNM system was chosen for this purpose. However, contrary to most other TNM systems for malignant tumors, this system divides tumors into 3 instead of 4 stage categories. This is based on the observation that statistically significant survival differences can only be achieved when combining 2 of the 4 stage categories of the Masaoka1 and Tsuchiya systems15 and condensing them into 3. Another important difference in that proposal is the classification of the lymph node status. Lymph node involvement is not a common feature of thymoma and therefore not included in the Masaoka system.1 In thymic carcinomas, however, lymph node involvement is present in up to 40% of cases16 and the inclusion of lymph node status is therefore mandatory for these tumors. In contrast, lymph node dissection during thymectomy is neither performed routinely nor in a standardized manner. Contrary to Tsuchiya et al15 who used 4 different lymph node stages requiring knowledge of the exact origin of the sampled nodes, Weissferdt and Moran22 suggest a more basic approach, placing patients into N0 (lymph node negative) or N1 (lymph node positive) strata offering a more practical approach to lymph node staging. Furthermore, the Weissferdt and Moran approach provides better correlation with clinical outcome by providing statistically significant differences for all stages.
STAGING OF PRIMARY NEUROENDOCRINE CARCINOMAS OF THE THYMUS
The classification and staging of primary thymic neuroendocrine carcinomas is another contentious issue lacking a uniform approach. With approximately 400 cases reported in the English literature, these neoplasms are rare and therapeutic strategies are not well defined. Comparison of the results of different studies has been hindered by the application of different classification systems using different cutoff values to assess mitotic activity and the presence of necrosis resulting in inconsistent terminology. Some of the larger series investigating these tumors have shown that these tumors are more aggressive than other similar foregut tumors, which in turn has generated much discussion as to what are the most important prognostic factors. Although some authors have shown that histologic grade has a major impact on survival,23–26 others found staging, Ki67 index and completeness of resection more significant in this context.27–29
As the nature of thymic neuroendocrine carcinomas is more akin to thymic carcinoma than thymoma reflected by its tendency for lymph node metastasis, a tentative TNM-based staging system was proposed by Fukai et al27 (Table 2). However, the authors failed to show any prognostic value of the T or N classification and stated that distant metastasis is the most important prognostic factor. Hence, most investigators still use the Masaoka1 or Koga-modified Masaoka4 staging systems for the staging of these tumors. This approach was recently endorsed by the International Thymic Malignancy Interest Group that proposed the use of these systems not only for thymoma and thymic carcinoma but also for neuroendocrine thymic carcinoma and other rare thymic tumors.30
In contrast, we believe that the strong similarity of the biological behavior of thymic neuroendocrine carcinoma and thymic carcinoma justifies application of the same staging approach. On the basis of our experience with the staging of thymic carcinoma and our experience over the years with thymic neuroendocrine carcinomas,23 we have little doubt that the proposed Weissferdt-Moran staging system for thymic carcinomas22; Table 4) should not also be applicable to thymic neuroendocrine carcinomas. The difference between these 2 groups of tumors is not in the manner of spread but rather their biological behavior, prognosis, and possible alternative treatment.
STAGING OF PRIMARY MEDIASTINAL GERM CELL TUMORS
Primary mediastinal germ cell tumors are another set of neoplasms that not infrequently involve the mediastinum.31 In fact, 10% to 20% of all tumors in this region can be attributed to primary germ cell tumors31 and clinical and pathologic staging is very important not only for the prognosis but also for the choice of therapy. In 1997, Moran and Suster31 examined 322 cases of primary mediastinal germ cell tumors and proposed a novel staging system specifically for mediastinal germ cell tumors based on the clinical outcome of their cases (Table 5; Figs. 3A–D). Their recommendation was to treat tumors exclusively confined to the mediastinum without infiltration of adjacent structures (stage I) conservatively, with surgery alone or with surgery and an added modality based on the histology of the tumor. Lesions of advanced stage (II or III) that showed infiltration into adjacent structures or intrathoracic metastases would require more aggressive treatment with curative intent, whereas palliative treatment was the choice in tumors with extrathoracic metastasis. In their study, this staging approach correlated well with the clinical outcome of the patients. The only other recommendation for the staging of primary mediastinal germ cell tumors originates from the WHO32 who proposed using a modification of the AJCC TNM staging of soft tissue tumors for this purpose. Needless to say, this proposal is groundless as the authors of that publication failed to conduct any specific study with the target of staging mediastinal germ cell tumors. It is important to note that in the Moran-Suster staging schema, the authors were not only able to stratify patients by tumor infiltration into adjacent anatomic structures but were also able to analyze the different histologic types that more often present in advanced stages of the disease. Needless to say, the large size of their sampling has not been surpassed over the last 15 years, and very likely will not be surpassed in the foreseeable future.
Continuing advances in oncology increasingly require not only accurate histopathologic tumor assessment but also appropriate staging information to ensure the best possible treatment for patients. Because of their rarity and heterogeneity, primary mediastinal neoplasms have always posed a problem not only in terms of classification but also when it comes to the staging of these tumors. Among them, thymic epithelial neoplasms are those that have attracted the most attention over the last few decades and numerous proposals as to their classification and staging protocols have been put forward. Although the Masaoka staging system for thymomas1 has been most commonly used not only for the staging of the former but also a range of other mediastinal neoplasms, it has been recently demonstrated that this system is suboptimal not only for mediastinal neoplasms other than thymoma but also for thymoma itself.5,16 On the basis of this, 2 novel and distinct staging protocols for thymoma and thymic carcinoma have recently been proposed.5,22 These systems use clear stage definitions, demonstrate statistically significant correlation with prognosis, and are practical and reproducible not only for pathologists but also for other specialists involved in the care of these patients. It is expected that the applicability of the staging schemas herein presented, will lead to better stratification of patients who may benefit from adjuvant treatment and in turn will improve the overall understanding of these tumors.
In contrast, neuroendocrine carcinomas of thymic origin have only sporadically been assessed for staging purposes. This seems to be mainly due to a paucity of large enough series of these tumors and the often associated lack of sufficient clinical follow-up information. Although the clinical outcome of mediastinal germ cell tumors seems to be corresponding best with the staging concept devised by Moran and Suster,31 thymic neuroendocrine carcinomas are best staged using the approach proposed for thymic carcinomas22 as these tumor groups display similar manner of spread.
1. Masaoka A, Monden Y, Nakahara K, et al. Follow-up study of thymomas with special reference to their clinical stages. Cancer. 1981;48:2485–2492
2. Bergh NP, Gatzinsky P, Larsson S, et al. Tumors of the thymus and thymic region: I. Clinicopathological studies on thymomas. Ann Thorac Surg. 1978;25:91–98
3. Wilkins EW Jr, Castleman B. Thymoma: a continuing survey at the Massachusetts General Hospital. Ann Thorac Surg. 1979;28:252–256
4. Koga K, Matsuno Y, Noguchi M, 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;44:359–367
5. Moran CA, Walsh G, Suster S, et al. Thymomas II: a clinicopathologic correlation of 250 cases with a proposed staging system with emphasis on pathologic assessment. Am J Clin Pathol. 2012;137:451–461
6. Shimosato Y. Controversies surrounding the subclassification of thymoma. Cancer. 1994;74:542–544
7. Shimasato Y, Mukai KRosai J. Tumors of the mediastinum. Atlas of Tumor Pathology, 3rd series. 1997 Washington DC Armed Forces Institute of Pathology:40–182 fascicle 21
8. Regnard JF, Magdeleinat P, Dromer C, et al. Prognostic factors and long-term results after thymoma resection: a series of 307 patients. J Thorac Cardiovasc Surg. 1996;112:376–384
9. Begg CB, Cramer LD, Venkatraman ES, et al. Comparing tumour staging and grading systems: a case study and a review of the issues, using thymoma as a model. Stat Med. 2000;19:1997–2014
10. Okumura M, Ohta M, Tateyama H, et al. The World Health Organization histologic classification system reflects the oncologic behavior of thymoma: a clinical study of 273 patients. Cancer. 2002;94:624–632
11. Quintanilla-Martinez L, Wilkins EW Jr, Choi N, et al. Thymoma. Histologic subclassification is an independent prognostic factor. Cancer. 1994;74:606–617
12. Kondo K, Monden Y. Therapy for thymic epithelial tumors: a clinical study of 1,320 patients from Japan. Ann Thorac Surg. 2003;76:878–884
13. Haniuda M, Morimoto M, Nishimura H, et al. Adjuvant radiotherapy after complete resection of thymoma. Ann Thorac Surg. 1992;54:311–315
14. Blumberg D, Burt ME, Bains MS, et al. Thymic carcinoma: current staging does not predict prognosis. J Thorac Cardiovasc Surg. 1998;115:303–308
15. Tsuchiya R, Koga K, Matsuno Y, et al. Thymic carcinoma: proposal for pathological TNM and staging. Pathol Int. 1994;44:505–512
16. Weissferdt A, Moran CA. Thymic carcinoma, part 1: a clinicopathologic and immunohistochemical study of 65 cases. Am J Clin Pathol. 2012;138:103–114
17. Curran WJ Jr, Kornstein MJ, Brooks JJ, et al. Invasive thymoma: the role of mediastinal irradiation following complete or incomplete surgical resection. J Clin Oncol. 1988;6:1722–1727
18. Gamondès JP, Balawi A, Greenland T, et al. Seventeen years of surgical treatment of thymoma: factors influencing survival. Eur J Cardiothorac Surg. 1991;5:124–131
19. Yamakawa Y, Masaoka A, Hashimoto T, et al. A tentative tumor-node-metastasis classification of thymoma. Cancer. 1991;68:1984–1987
20. Bedini AV, Andreani SM, Tavecchio L, et al. Proposal of a novel system for the staging of thymic epithelial tumors. Ann Thorac Surg. 2005;80:1994–2000
21. Masaoka A. Staging system of thymoma. J Thorac Oncol. 2010;5(suppl 4):S304–S312
22. Weissferdt A, Moran CA. Thymic carcinoma, part 2: a clinicopathologic correlation of 33 cases with a proposed staging system. Am J Clin Pathol. 2012;138:115–121
23. Moran CA, Suster S. Neuroendocrine carcinomas (carcinoid tumor) of the thymus. A clinicopathologic analysis of 80 cases. Am J Clin Pathol. 2000;114:100–110
24. Gaur P, Leary C, Yao JC. Thymic neuroendocrine tumors: a SEER database analysis of 160 patients. Ann Surg. 2010;251:1117–1121
25. Cardillo G, Treggiari S, Paul MA, et al. Primary neuroendocrine tumours of the thymus: a clinicopathologic and prognostic study in 19 patients. Eur J Cardiothorac Surg. 2010;37:814–818
26. Cardillo G, Rea F, Lucchi M, et al. Primary neuroendocrine tumors of the thymus: a multicenter experience of 35 patients. Ann Thorac Surg. 2012;94:241–245
27. Fukai I, Masaoka A, Fujii Y, et al. Thymic neuroendocrine tumor (thymic carcinoid): a clinicopathologic study in 15 patients. Ann Thorac Surg. 1999;67:208–211
28. Gal AA, Kornstein MJ, Cohen C, et al. Neuroendocrine tumors of the thymus: a clinicopathological and prognostic study. Ann Thorac Surg. 2001;72:1179–1182
29. Tiffet O, Nicholson AG, Ladas G, et al. Clinicopathologic study of 12 neuroendocrine tumors arising in the thymus. Chest. 2003;124:141–146
30. Detterbeck FC, Nicholson AG, Kondo K, et al. The Masaoka-Koga stage classification for thymic malignancies: clarification and definition of terms. J Thorac Oncol. 2011;6:S1710–S1716
31. Moran CA, Suster S. Primary germ cell tumors of the mediastinum: I. Analysis of 322 cases with special emphasis on teratomatous lesions and a proposal for histopathologic classification and clinical staging. Cancer. 1997;80:681–690
32. Wick MR, Perlman EJ, Orazi A, et al.Travis WD, Brambilla E, Muller-Hermelink HK, et al. Germ cell tumors of the mediastinum. World Health Organization Classification of Tumours Pathology and Genetics Tumours of the Lung, Pleura, Thymus, and Heart. 2004 Lyon IARC press:198–201
© 2013 Lippincott Williams & Wilkins, Inc.