OUP user menu

Invasive Spindle Cell Thymomas (WHO Type A)
A Clinicopathologic Correlation of 41 Cases

Cesar A. Moran MD, Neda Kalhor MD, Saul Suster MD
DOI: http://dx.doi.org/10.1309/AJCP7KBP4QQLRLXW 793-798 First published online: 1 November 2010


We report 41 cases of invasive spindle cell thymomas (World Health Organization type A). The patients were 16 women and 25 men between the ages of 38 and 80 years. Clinically, the patients had diverse symptomatology, including chest pain, cough, and dyspnea. None of the patients had a history of myasthenia gravis. According to the Mazaoka surgical staging system, 34 patients had stage II disease, 6 had stage III, and 1 had stage IV. Follow-up information showed that 30 patients were alive after a period ranging from 12 to 96 months; for 8 patients who are alive, the follow-up was less than 12 months; 1 patient died 10 months after initial diagnosis. For 2 patients, no follow-up information was obtained. This study stresses the fact that histologic features do not correlate with invasion or encapsulation because all thymomas, regardless of their histologic type, are capable of invasion.

Key Words:
  • Mediastinum
  • Thymoma
  • Spindle cell
  • World Health Organization

Spindle cell thymomas represent one of the most unusual histologic types in the general spectrum of thymic epithelial neoplasms, namely thymomas. Their actual incidence is difficult to calculate because different series of thymomas have reported different incidences of this particular histologic type. Further complicating the classification is the lack of established criteria for the number of sections that must be examined to make a diagnosis of “pure” spindle cell thymoma (World Health Organization [WHO] type A).135 Actually, in a majority of cases reported, the number of invasive cases has been in the minority. Therefore, many cases in which these particular histologic features are present may be classified as such but without proper sampling of the resected tumor. On the other hand, in cases that are only partially resected or biopsied, the diagnosis of spindle cell thymoma may not necessarily be entirely correct as to their subclassification. If to these facts we add the current concept that spindle cell thymomas are benign tumors, as has been presented in the last version of the WHO classification,36 then, we can assume that these tumors are always encapsulated or, regardless of their staging, are also always benign.

The study reported herein highlights the issue that spindle cell thymomas (WHO type A), just like any other histologic variant of thymoma, have a similar potential to become invasive tumors capable of spreading in or outside of the thoracic cavity. In addition, this study highlights a much more important issue—sampling—and attempts to establish a minimum requirement for the assessment of these tumors, especially if one is to classify these neoplasms using any of the existing classification schemes.

Materials and Methods


We retrieved 41 cases from the files of the Department of Pathology, M.D. Anderson Cancer Center, Houston, TX, and the Department of Pathology, Medical College of Wisconsin, Milwaukee, and from the personal files of two of us (C.A.M. and S.S.) from 1995 to 2009. All clinical information, including follow-up information, was tabulated, and histologic evaluation was performed in all of the cases. In all, the number of cases was more than 400, including biopsy specimens and encapsulated tumors, of which 210 cases of invasive thymoma were selected based on preestablished minimum criteria for inclusion, which are described in the following section.

Case Selection

All of the cases corresponded to surgical resection of the mediastinal tumor. No biopsy material was included in the current cases. In addition, the minimum requirements for inclusion in this study were as follows: (1) A minimum of 5 histologic sections of tumor have to be histologically examined for these neoplasms to be labeled as spindle cell thymoma (WHO type A), as suggested in a previous article.37 (2) All of the tumors have to be invasive. (3) No other histologic pattern is present in any of the sections examined.

Following those parameters, the number of sections evaluated ranged from 5 to 22; of the 41 cases, 22 had 10 sections or more and 19 had 5 to 9 sections. Only 7 cases had fewer than 7 sections to evaluate.

Every section was analyzed, and tumors in which there was another type of histologic feature were subclassified as mixed histologic types. Therefore, the cases herein presented corresponded to the “pure” spindle cell thymoma (WHO type A) histologic type.

Immunohistochemical studies were performed in some cases for diagnostic purposes and did not influence the final diagnosis.


Clinical Features

The patients were 16 women and 25 men between the ages of 38 and 80 years. However, only 2 patients were younger than 50 years. Clinically, all patients had nonspecific symptoms, including chest pain, cough, and dyspnea. None of the patients had a history of myasthenia gravis. Nevertheless, 4 patients had an associated malignancy, which included thyroid, breast, and prostatic carcinoma. Radiologically, the patients showed an anterior mediastinal mass, and all patients underwent resection of the tumor.

Gross Pathology

The tumors were described as solid tumor masses ranging in size from 3 to 18 cm in greatest dimension. The cut surface of the tumors was described as homogeneous, light tan; in some cases, focal cystic areas were described; however, none of the tumors was described as hemorrhagic or necrotic.

Histologic Findings

All 41 cases showed similar morphologic features, namely, the presence of a spindle cell proliferation. However, different growth patterns were observed in some cases. The main morphologic growth pattern was that of a spindle cellular proliferation arranged in short fascicles of plump cells with elongated nuclei and inconspicuous nucleoli. The low-power view showed a pattern that closely mimicked that of hemangiopericytoma because the tumors showed prominent dilated vascular spaces Image 1A. However, in 7 cases, the tumors showed also alternating hypocellular areas with vague myxoid changes Image 1B, while in 5 cases, the tumors showed the so-called micronodular growth pattern. In 4 additional cases, the growth pattern closely resembled that of a neuroendocrine neoplasm because the tumors showed a vague nesting pattern and also prominent rosette-like structures Image 2. In 6 cases, the tumors infiltrated skeletal muscle, pleura, and lung Image 3, while in 1 case, the tumor infiltrated the thyroid gland Image 4. None of the tumors showed necrosis or hemorrhage. In a few cases, rare mitotic figures were present in the epithelial component.


According to the Mazaoka surgical staging system, 34 cases were stage II, 6 cases were stage III, and 1 case was stage IV. In the latter case, the tumor invaded the thyroid gland.


Clinical follow-up information obtained for 30 patients ranging from 12 to 96 months showed that all patients were alive without recurrence or metastatic disease. In addition, in 8 patients who are alive, the follow-up obtained was less than 12 months. One patient died 10 months after initial diagnosis, and for 2 additional patients, no follow-up information was obtained.


During the last decade, much controversy has been generated about the proper classification of thymomas and whether histologic subtyping can predict prognosis over clinical staging. In addition, publications in which specific types of thymomas have been regarded as “benign” tumors with no potential for recurrence have cast some controversy, at least in the understanding of these tumors. What is of great interest is that the studies supporting the idea that histology is the most important predictor of prognosis have not dealt with the most important issue—sampling. We suggest that any classification scheme that does not address the issue of a specific number of sections to be examined to properly classify a thymoma is faulty. Thus far, little light has been shed on establishing minimum criteria for the number of sections that must be histologically evaluated or the percentage of a specific growth pattern that must be seen before classifying a tumor into any of the known histologic schemes available.

Image 1

A, Low-power view of a spindle cell thymoma with a hemangiopericytic-like growth pattern (H&E, ×4). B, Spindle cell thymoma showing a hypocellular area with vague myxoid changes (H&E, ×10).

It is interesting that all studies published in the literature on the topic of thymoma have ignored the specifics of the material examined, have been vague about the materials and methods used, or simply have included cases based on biopsy material.136 Unfortunately, this latter aspect has, in some cases, been used in the formulation of specific schemes in the subclassification of thymomas.38 In addition, regardless of the material used, in the vast majority of series of thymomas, the incidence of spindle cell tumors has been, for the most part, minimal, and when present, the larger number of cases has been encapsulated spindle cell tumors. Nevertheless, it is important to highlight that even in cases in which the number of spindle cell tumors has been a small proportion of the entire population of thymomas, some of the spindle cell thymomas have been described as invasive tumors, just the same as the most common histologic variants, leading, in some circumstance, to a fatal outcome. Thus, little can be gained from reviewing previous literature regarding the minimum criteria for the subclassification of thymomas, and, if we are to follow the recommendations of the WHO, then any tumor that is subclassified as spindle cell morphology should be deemed benign from the outset. Some authors have gone even further when addressing the issue of spindle cell thymomas by stating “medullary thymomas are benign tumors with no risk of recurrence and adjuvant therapy appears unnecessary even when invasive,”7 even though the number of cases to support such a notion was scant. However, other authors have taken a different approach by stating that all thymomas have malignant potential and have included spindle cell thymomas in the moderately differentiated tumors,23 or they have simply called all invasive tumors “malignant thymomas.”24

Image 2

Spindle cell thymoma showing rosette-like structures mimicking a neuroendocrine neoplasm (H&E, ×20).

Image 3

A, Spindle cell thymoma infiltrating skeletal muscle (H&E, ×10). B, Spindle cell thymoma infiltrating lung parenchyma (H&E, ×10).

Image 4

Spindle cell thymoma infiltrating thyroid gland. Note the presence of residual thyroid follicles (arrows) (H&E, ×10).

Based on our case preselection, we evaluated 210 invasive thymomas, in which a strict criterion for inclusion was used, namely the minimum requirement of evaluating 5 histologic sections after surgical resection. By using this method, we consider that a tumor is not only most likely to be properly subclassified but also we can determine with a higher degree of certainty the true incidence of invasive spindle cell thymoma. In addition, 2 other possibilities are highly important to rule out, one being spindle cell thymic carcinoma and the other, a combined spindle cell thymoma and thymic carcinoma.39,40 Both of these unusual histologic types can best be excluded only by proper sampling of the tumor in question. In our experience and following strict criteria, the number of cases of invasive spindle cell thymoma represents approximately 19.5% of the total cases of invasive thymomas. A similar percentage of spindle cell thymomas (17%) was also reported by Hofmann et al29; however, that series of cases included encapsulated and invasive tumors, and no detailed information was provided as to the number of sections examined to determine specific subclassifications of the thymomas reported. It is interesting that we also noted that this specific histologic type (spindle cell) shows similar patterns of invasion, just like other more common histologic types. For instance, the great majority of cases were stage II in which the tumor breached the capsule into the surrounding adipose tissue, while in other cases, the tumor invaded the pleura and lung, and, in some unusual cases, the tumor spread outside of the thoracic cavity into neck structures such as the thyroid gland. Similar patterns of invasion have also been described for thymomas in general and not exclusively for any particular group or a very specific histologic type.

The study reported herein highlights a few important issues: (1) Invasive spindle cell thymomas exist and represent approximately 19% of all invasive thymomas. (2) The invasive nature of spindle cell thymomas is similar to that already described for other subtypes. (3) The subclassification of these tumors can be determined only after complete surgical resection of the tumor. (4) A minimum of 5 histologic sections of tumor or, in cases of small tumors (<5 cm), an evaluation of the entire tumor must be performed for proper classification. (5) It is improper to attempt any subclassification or prediction of tumor behavior of thymomas based on biopsy material. (6) Surgical staging at the time of diagnosis remains the best way to predict prognosis in thymomas. Last, regarding treatment for these tumors, it is too premature and risky to state that spindle cell thymomas, even when they are invasive, should not be treated. As we have presented, this histologic variant also has the potential to spread within and outside of the thoracic cavity. The issue of treatment remains a clinical judgment based on the spread of the tumor at the time of diagnosis4143 and should not be based on the histologic subtyping of thymoma.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
View Abstract