Thymomas are uncommon neoplasms that have generated considerable controversy among pathologists. The following questions can be used to evaluate the evidence supporting current concepts about the pathology of thymomas and the clinical applicability of those concepts.
1. How should thymomas be studied pathologically?
Kalhor and Moran emphasize the need to sample multiple tumor sections from frequently heterogeneous thymomas. The International Thymic Malignancy Interest Group (ITMIG) has recently proposed policies and procedures for the optimal handling of thymomas in the operating room and the pathology laboratory. There is limited high-level evidence evaluating the effects of various practices on the accuracy of pathologic classification and staging of thymomas.
2. Evaluation of microscopic capsular invasion: is it clinically important?
Pathologists routinely evaluate multiple sections of the capsular area of thymomas and upstage lesions with transcapsular invasion to Masaoka stage IIa or Moran et al stage I disease. However, level II evidence demonstrates no significant differences in survival or recurrence rates between Masaoka stage IIa and stage I thymomas.
3. Which is the best current pathologic classification of thymomas? Does “histologic type” correlate with prognosis? Are there benign and malignant thymomas?
In our view, the controversy generated by these questions results from the use of different classification goals. “Splitter” pathologists favor the use of comprehensive classification schemas that stratify lesions into as many recognizable entities as possible in the hope that these entities will correlate now or in the future with prognosis and response to specific therapies. However, comprehensive schemas, such as the World Health Organization (WHO) thymoma classification, are often based on the subjective recognition of somewhat overlapping histopathological features, resulting in the interobserver variability problems discussed by Kalhor and Moran. In contrast, other pathologists are more pragmatic “lumpers” who advocate stratifying thymomas and other lesions only into categories currently associated with significant prognostic differences or selective responses to specific therapies, based on the assumption that there will be less interobserver variability if practitioners have fewer diagnostic choices. Ironically, opinions on this question are somewhat “split,” even among lumpers. Kolhar and Moran argue that there is no difference in the prognosis of patients with A, B1, AB, and B2 thymomas; however, a level II study employing meta-analysis to analyze almost 1000 thymomas from one of our institutions (Cedars-Sinai Medical Center) and others has shown that A, B1, and AB thymomas have 5-year survivals that are significantly different from that of B2 thymomas.[5,6]
How should the difference between splitters and lumpers be bridged? Splitters would probably argue that all that is needed to decrease interobserver variability is to refine and explain in more detail the diagnostic criteria used in the WHO schema. Indeed, an ITMIG-sponsored group led by Professor Alex Marx is currently working on this approach. In our opinion, such studies are very important to the advancement of our understanding of thymomas; however, it remains questionable whether comprehensive thymoma classification schemas developed by expert consensus opinion will be useful in daily clinical practice before their reproducibility and clinical applicability in different practice scenarios have been demonstrated by high-level evidence. Indeed, lumpers would argue that currently there is little downside to using the Suster and Moran classification schema or a simplified WHO schema for patients who are not to be entered into a randomized controlled trial (RCT), since there are relatively few effective therapeutic options for thymoma patients, and since the evidence that the WHO classification helps classify patients into groups that would significantly benefit from particular therapies is somewhat controversial. In thymoma patients who are entered into RCTs, the pathology of their tumors probably needs to be centrally reviewed by the investigators to ensure diagnostic accuracy and consistency.
4. What is the level of interobserver variability using different classification schemas?
A multicenter study of the interobserver variability in the diagnosis of thymomas classified according to the WHO schema by Verghese et al reported moderate agreement levels, with kappa = 0.45. A similar multicenter study by Rieker et al reported good agreement, with kappa = 0.87, for all thymoma categories—but with kappa = 0.49 for the subgroups of B1, B2, and B3 thymomas. Diagnostic variability, as demonstrated by kappa values such as these, can significantly influence the results of studies evaluating prognosis. To our knowledge, there are no studies using kappa statistics to evaluate the Suster and Moran classification of thymomas.
5. Are there benign thymomas, or are all thymomas malignant?
This controversy should have been put to rest by the level III studies discussed by Kalhor and Moran showing examples of thymoma A tumors that have recurred and/or metastasized. Pathologists have long struggled with the selection of the most appropriate terminology to use to designate patients with lesions that seldom metastasize and/or cause death. A discussion of the use of terms such as “borderline lesions,” “low–malignant potential lesions,” and “low-grade malignancies” is beyond the scope of this commentary. In our view, all thymomas should be classified as low-grade malignancies, in view of their proven potential for recurrence or metastasis over a long period.
6. What is the level of evidence supporting various staging systems?
As discussed by Kalhor and Moran, the literature and their own experience includes mostly level III studies evaluating the Masaoka and/or the Moran et al staging systems.[10,11] There is a need for future RCTs that evaluate the prognostic value of various staging systems. In particular, there is a need for high-level evidence that would help in determining which patients with Masaoka stage IIb thymomas would benefit from radiation therapy and which Masaoka stage 3 or 4 thymomas might benefit from the use of neoadjuvant and/or postoperative chemotherapy and radiation therapy.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
1. Marchevsky AM, McKenna RJ, Jr., Gupta R. Thymic epithelial neoplasms: a review of current concepts using an evidence-based pathology approach. Hematol Oncol Clin North Am. 2008;22:543-62.
2. Detterbeck FC, Moran C, Huang J, et al. Which way is up? Policies and procedures for surgeons and pathologists regarding resection specimens of thymic malignancy. J Thorac Oncol. 2011;6(Suppl 3):S1730-S1738.
3. Gupta R, Marchevsky AM, McKenna RJ, et al. Evidence-based pathology and the pathologic evaluation of thymomas: transcapsular invasion is not a significant prognostic feature. Arch Pathol Lab Med. 2008;132:926-30.
4. Marchevsky AM, Wick MR. Evidence based pathology and laboratory medicine. New York: Springer; 2011.
5. Marchevsky AM, Gupta R, McKenna RJ, 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.
6. Marchevsky AM, Gupta R, Casadio C, et al. World Health Organization classification of thymomas provides significant prognostic information for selected stage III patients: evidence from an international thymoma study group. Hum Pathol. 2010;41:1413-21.
7. Verghese ET, den Bakker MA, Campbell A, et al. Interobserver variation in the classification of thymic tumours--a multicentre study using the WHO classification system. Histopathology. 2008;53:218-23.
8. Rieker RJ, Hoegel J, Morresi-Hauf A, et al. Histologic classification of thymic epithelial tumors: comparison of established classification schemes. Int J Cancer. 2002;98:900-6.
9. Marchevsky AM, Gupta R. Interobserver diagnostic variability at "moderate" agreement levels could significantly change the prognostic estimates of clinicopathologic studies: evaluation of the problem using evidence from patients with diffuse lung disease. Ann Diagn Pathol. 2010;14:88-93.
10. Marchevsky AM, Wick MR. Evidence levels for publications in pathology and laboratory medicine. Am J Clin Pathol. 2010;133:366-7.
11. Wick MR, Marchevsky AM. Evidence-based principles in pathology: existing problem areas and the development of "quality" practice patterns. Arch Pathol Lab Med. 2011;135:1398-404.