Abstract
Thymomas are rare tumors of the mediastinum; a limited number of small studies have evaluated the outcomes in these patients. We identified 668 patients with thymoma from the Swedish Cancer Registry, and 2,719 population-based matched controls. We obtained information on autoimmunity from the nationwide inpatient/outpatient hospital discharge Registry. We constructed Kaplan-Meier curves for survival analysis, conditional regression and Cox proportional hazards models to evaluate the association between thymoma and autoimmune diseases, and standardized incidence ratios (SIRs) to evaluate the risk for second cancers following thymoma. Compared with controls, patients with benign or malignant thymoma had a poorer (p <0.001) 5-year (79%, 53% vs. 91%), 10-year (65%, 39% vs. 78%), and 20-year (43%, 18% vs. 55%) overall survival. For thymoma patients, younger age at diagnosis and being diagnosed in recent years were associated with a better survival. Compared with controls, thymoma patients were more likely to have an autoimmune disease at some point during their lives (32.7% vs. 2.4%, respectively, p<0.001), most frequently myasthenia gravis (24.5%), systemic lupus erythematosus (2.4%), and red cell aplasia (1.2%). Thymoma patients had 2-fold excess risk for second cancers compared with the general population, most notably: non-melanoma skin cancer (SIR=10.6, 95% confidence intervals (CI)=6.0–17.3), non-Hodgkin lymphoma (SIR=6.8, 95% CI=3.00–13.0), and cervical (SIR=6.9, 95% CI=1.4–20.1), endocrine (SIR=4.7, 95% CI=1.3–12.0), and prostate cancer (SIR=3.0, 95% CI=1.7–4.8). Despite the improved survival for thymoma patients over time, they have worse survival than controls. Thymoma patients are in need for follow-up to detect and manage autoimmune diseases and cancer.
Keywords: thymoma, autoimmunity, second cancer
INTRODUCTION
Thymomas are rare neoplasms accounting for only 0.2% to 1.5% of all malignancies (1). Despite their rarity, they comprise about 20% of all mediastinal masses and account for up to 50% of anterior mediastinal masses (2;3)
Patients with thymoma are frequently diagnosed with paraneoplastic syndromes, the most common of which is myasthenia gravis (4–6). Multiple studies also suggest that patients diagnosed with thymoma are at an increased risk of developing second primary malignancies, with a reported frequency ranging from 8 to 28% (7–10).
Due to their rarity, most of the literature is based on small case series from single institutions. To our knowledge, only few population-based thymoma studies have been conducted previously (7;9;11).
To improve our knowledge on this rare disease entity of thymic tumors, we have conducted a comprehensive population-based study including all (n=668) patients diagnosed with thymoma in Sweden between 1958 and 2004, and matched population controls (n=2,719). Our study aims were to assess survival patterns as well as risks of autoimmune diseases and second malignancies.
MATERIALS AND METHODS
Population registries and study subjects
From the Nationwide Swedish Cancer Registry, we used the 7th revision of the International Classification of Diseases to identify all patients diagnosed with thymoma (ICD7=195.2) between 1958 and 2004. In the database, thymoma cases were classified by the responsible pathologist either as benign (non-invasive; PAD code 841) or malignant (capsule invasion; PAD code 846). The present version of the database does not include the World Health Organization (WHO) histological classification of thymomas which was released in 1999 (12).
For each thymoma case, up to 4 population-based controls (matched by gender, year of birth and county of residence) were randomly selected from the Swedish Population database. Because this study was part of a larger research effort, persons with a hematopoietic lymphoid malignancy were not eligible for the control selection. All controls had to be alive at the time of diagnosis of the corresponding case. For this study, we excluded cases or controls who had any cancer prior to selection.
Using individuals’ unique national registration number, both cases and controls were linked to the Swedish inpatient/outpatient hospital discharge registry, to obtain information on diagnosis of autoimmune diseases. The Swedish inpatient/outpatient hospital discharge registry started in 1964 and reached 100% coverage of the nationwide hospitalization in 1987. Incorporating outpatient data in the registry started in 2001. Cases and controls were also linked to the Nationwide Cause of Death Registry to identify their vital status, date and cause of death.
Statistical Analysis
Survival patterns in thymoma cases and their matched controls
We constructed Kaplan-Meier curves, using SPSS16.0 software, to estimate the probability of survival and 95% confidence intervals (CI) at 5-year, 10-year, and 20-year in both thymoma patients and their matched controls. In this analysis, follow up time started at date of thymoma diagnosis for both cases and controls, and ended at death or censoring (emigration or end of census data follow-up, December 31, 2003). The log-rank test was used to compare the survival distribution between cases and controls.
The association between thymoma and autoimmune diseases
For our investigations of autoimmune diseases, we used conditional logistic regression to calculate the odds ratios (ORs) and 95% confidence intervals (CI) of having these diseases before (>1 year), or around (within 1 year before/after) thymoma diagnosis, comparing thymoma cases with their matched controls. All models were adjusted for age and gender. In this analysis, we excluded cases (n=52) who were diagnosed before the start of the inpatient/outpatient hospital discharge registry (1964) and their matched controls (n=171).
We then used Cox regression models to calculate hazard ratios (HRs) and 95% CI of developing autoimmune diseases after thymoma diagnosis/control selection (>1 year) for cases and controls. In these models, age was used as the time matrix and they were all adjusted for gender. Follow up started at age of case diagnosis/control selection, or in 1964 whichever comes first. We reported p-value from univariate Fisher’s exact test when zero observations were noted. We used SAS 9.1 software for these analyses
Second cancers after thymoma diagnosis
To evaluate the risk of second cancer after the diagnosis of thymoma, we calculated standardized incidence ratios (SIR) (observed cancers in the thymoma patients divided by the expected numbers), for all cancers combined and by anatomic site and gender. Expected numbers of events were calculated by applying age- and gender-specific population incidence rates from the Swedish Cancer Registry to the person-years observed among our cases. Thymoma patients were followed-up from date of diagnosis to the development of cancer or censoring at date of death, emigration or end of follow up (December 31, 2004). We evaluated the risk of every cancer site, but present here only those with 2 or more affected cases. For this analysis, we used the SEER*Stat 6.5.2 software.
RESULTS
A total of 668 thymoma patients and 2,719 matched controls were included in this study. We found 64.5% and 35.5% of the thymomas to be benign and malignant, respectively. Most of the thymoma patients were diagnosed after age 40 (82.9%, median age=60 years), and almost half were males (46.6%) (Table 1). This was reflected in a crude incidence of 2.3 and 2.7/1000, 000 person-years for males and females, respectively. Median follow-up of study participant was 10.2 years (interquartile range =14.0 years).
Table 1.
Characteristics of thymoma patients and controls
| Thymoma patients N=668 | Controls N=2,719 | P-value | |
|---|---|---|---|
| N (%) | |||
| Age (Median, range) | 60 (3–91) | 60 (3–93) | |
| ≤40 | 114 (17.1) | 468 (17.2) | 0.87 |
| 41–60 | 242 (36.2) | 956 (35.2) | |
| >60 | 312 (46.7) | 1295 (47.6) | |
| Calendar year of selection | |||
| 1958–1972 | 151 (22.6) | 594 (21.8) | 0.64 |
| 1973–1987 | 212 (31.7) | 915 (33.7) | |
| 1988–2004 | 305 (45.7) | 1210 (44.5) | |
| Gender | |||
| Male | 311 (46.6) | 1299 (47.8) | 0.57 |
| Female | 357 (53.4) | 1420 (52.2) | |
| Subtypes | |||
| Benign | 431 (65.52) | ||
| Malignant | 237 (35.48) | ||
Patterns of survival
Compared with controls, patients with benign or malignant thymoma had significantly (p<0.001) poorer 5-year (79%, 95% CI=75.1%-82.9% for benign cases; 53%, 46.3%-59.7% for malignant cases vs. 91%, 89.8%-92.2% for controls), 10-year (65%, 60.1%-69.9%; 39%, 32.1%-45.9% vs. 78%, 76.6%-79.8%), and 20-year (43%, 37.1%-48.9%; 18%, 11.3%-24.7% vs. 55%, 52.6%-57.3%) overall survival (Figure 1A). When we compared overall survival rates for benign and malignant thymoma patients diagnosed at three calendar intervals (1958–1972, 1973–1987, and 1988–2004) there was a significant (p<0.01) improvement over time. Further, when we assessed survival patterns by age, we found overall survival to be significantly better for younger patients with benign tumors (p<0.001) but not for those with malignant tumor (p=0.06) (Figure 1B, C). In analyses stratified by sex, we found no survival differences (benign cases: p=0.13; malignant cases: p=0.12) Similar patterns in the stratified analysis were observed when comparing thymoma patients with their matched controls except for a notable decline in the survival gap with increasing age. This was true for both benign and malignant cases. Mean survival time (in years) and 95% CI for cases diagnosed ≤40, compared to those diagnosed after age 60 years and their matched controls are as follows: mean=30.5, 95% CI=26.2–37.7, for benign cases; 15.4, 10.1–20.8, for malignant cases, and 42.7, 41.6–43.8, for controls vs. 10.7, 9.4–11.9, for benign, 7.7, 5.9–9.5, for malignant, and 13.8, 13.2–14.4, for controls). Of note, having an autoimmune disease prior to or at time of thymoma diagnosis was not associated with statistically significant worse survival in thymoma patients (mean survival years=16.6 in cases without vs. 14.0 in cases with an autoimmune disease, p=0.74). Cancer-related mortality was the most commonly reported causes of death in thymoma patients (n=145/300, 48.3% of all deaths); thymoma and its complications accounted for 40% of cancer-related mortality. This was followed by ischemic heart disease (15.7%), and pneumonia or chronic obstructive pulmonary disease (8%).
Figure 1.
(A) Survival curves for benign and malignant thymoma patients and their matched controls. (B) Survival curves for benign (left panel) and malignant (right panel) thymoma patients stratified by age at diagnosis. (C) Survival curves for benign (left panel) and malignant (right panel) thymoma patients stratified by calendar year of diagnosis. Survival curves were compared using the log-rank test; P values are presented.
The association between thymoma and autoimmunity
Compared with controls, thymoma patients were more likely to have an autoimmune disease at some point during their lifetime (32.7% vs. 2.4%, respectively, p<0.001). Among the specific autoimmune diseases, myasthenia gravis (24.5%), systemic lupus erythematosus (SLE) (2.4%), red cell aplasia (1.2%), sarcoidosis (0.9%) and rheumatoid arthritis (RA) (0.7%) were the most frequently reported. The vast majority of myasthenia gravis cases were diagnosed around or after thymoma diagnosis (42.7% and 52.4%, respectively). Of interest, myasthenia gravis was detected up to 38 years, and red cell aplasia up to 6.5 years, following a diagnosis of thymoma. Based on small numbers, SLE, RA, and sarcoidosis occurred primarily after the diagnosis of thymoma (Table 2).
Table 2.
Risk of selected autoimmune diseases in thymoma patients (compared with controls)
| Before thymoma dx (>1 year prior) | Around thymoma dx (1 year before/after) | After thymoma dx (>1 year after) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Cases | Controls | RR (95% CI)* | Cases | Controls | RR (95% CI)* | Cases | Controls | RR (95% CI)** | |
| N (%) | N (%) | N (%) | |||||||
| Myasthenia Gravis | 8 (1.2) | 0 (0) | α (<0.001)‡ | 70 (10.3) | 1 (0.04) | 324 (44.91-α)† | 83 (12.6) | 0 (0) | α (p<0.001) ‡ |
| Red Cell Aplasia | 0 (0) | 0 (0) | N/A | 4 (0.6) | 0 (0) | α (p=0.002) ‡ | 4 (0.6) | 0 (0) | α (p=0.001) ‡ |
| Systemic Lupus Erythematosus | 2 (0.3) | 1 (0.04) | 8.09 (0.73–89.41) | 2 (0.3) | 0 (0) | α (p=0.04)‡ | 12 (1.9) | 2 (0.1) | 32.8 (7.3–146.6)† |
| Rheumatoid Arthritis | 1 (0.2) | 1 (0.04) | 4.21 (0.26–67.54) | 2 (0.3) | 0 (0) | α (p=0.04)‡ | 2 (0.3) | 0 (0) | α (p=0.04)‡ |
| Sarcoidosis | 0 (0) | 1 (0.04) | 0 | 4 (0.6) | 0 (0) | α (p=0.002) | 2 (0.3) | 4 (0.2) | 3.1 (0.6–16.9) |
Relative risks, represent odds ratios, and 95% confidence intervals calculations accounted for the matching of cases and controls, and adjusted for age and gender
Relative risks, represent hazard ratios, and 95% confidence intervals adjusted for gender, and age was used as the time matrix
p≤0.05
Fisher’s exact test
Second cancers after thymoma diagnosis
Six hundred thirty-eight thymoma patients with any follow-up data contributed 5,585.19 person-year of follow up; 101 (15.8%) developed a subsequent cancer. This reflects a second cancer incidence of 21.1 and 15.8 per 1,000 person-years for males and females, respectively. Compared with the Swedish population, thymoma patients had more than two-fold increase in the risk of developing any cancer, with no overall significant gender difference (SIR=2.9 in males, vs. 2.3in females). This excess risk was mainly driven by non-melanoma skin cancer (SIR=10.6, 95% CI=6.0–17.3), non-Hodgkin lymphoma (SIR=6.3, 95% CI=3.0–13.0), cervix cancer (SIR=6.9, 95% CI=1.4–20.1), endocrine cancer (SIR=4.7, 95% CI=1.3–12.0) and prostate cancer (SIR=3.0, 95% CI=1.7–4.8) Overall, no gender differences in cancer risk were noticed (SIRs=2.9 in males vs. 2.3 in females) (Table 3).
Table 3.
Standardized incidence ratios of subsequent cancers following thymoma diagnosis by anatomic site and sex
| Overall N=638 | Male N=295 | Female N=343 | |||||
|---|---|---|---|---|---|---|---|
| Observed | Expected | SIR (95% CI) | Observed | SIR (95% CI) | Observed | SIR (95% CI) | |
| All Sites | 101 | 38.8 | 2.6 (2.0–3.2)* | 50 | 2.9 (2.2–3.9)* | 51 | 2.3 (1.7–3.0)* |
| Gastro-intestinal | 11 | 6.0 | 1.8 (0.9–3.3) | 8 | 2.8 (1.22–5.5)* | 3 | 0.9 (0.2–2.7) |
| Stomach | 2 | 0.8 | 2.5 (0–9.2) | 2 | 5 (0.5–16.5) | 0 | 0 (0–10.7) |
| Colon | 5 | 2.6 | 1.9 (1.0–4.5) | 3 | 3.0 (0.59–8.3) | 2 | 1.3 (0.2–4.8) |
| Rectum and Anus | 4 | 1.5 | 2.6 (1.0–6.5) | 3 | 4 (0.8–11.3) | 1 | 1.3 (0.3–7.1) |
| Pancreas | 2 | 0.8 | 2.5 (0–8.9) | 0 | 0 (0–10.8) | 2 | 4.3 (0.5–15.5) |
| Respiratory tract | 6 | 3.0 | 2.0 (0.7–4.3) | 2 | 1.3 (0.2–4.7) | 4 | 2.6 (0.7–6.7) |
| Lung/Bronchus | 5 | 2.9 | 1.7 (1.0–4.1) | 2 | 1.0 (0.2–5.2) | 3 | 2.0 (0.4–5.9) |
| Urinary system | 3 | 2.5 | 1.2 (0.2–3.5) | 2 | 1.3 (0.1–4.6) | 1 | 1.0 (0.03–5.9) |
| Other than kidneys† | 3 | 1.6 | 1.9 (0–5.5) | 2 | 2.0 (0.2–6.7) | 1 | 1.9 (0.05–10.7) |
| Melanoma skin | 2 | 1.5 | 1.3 (0–4.7) | 0 | 0 (0–5.7) | 2 | 2.3 (0.3–8.2) |
| Non-melanoma skin# | 16 | 1.5 | 10.6 (6–17.3)* | 7 | 8.8 (3.5–18.2)* | 9 | 12.7 (5.8–24.1)* |
| Endocrine glands## | 4 | 0.8 | 4.7 (1.3–12.0)* | 3 | 13.0 (2.7–38.1)* | 1 | 1.4 (0.04–8.1) |
| Breast | 0 | 9 | 1.3 (0.6–2.4) | ||||
| Cervix | N.A. | 3 | 6.9 (1.4–20.1)* | ||||
| Ovary/fallopian tube | N.A. | 3 | 3.2 (0.7–9.4) | ||||
| Prostate | 17 | 3.0 (1.7–4.8)* | N.A. | ||||
| Non-Hodgkin lymphoma | 7 | 1.11 | 6.3 (3.00–13.0)* | 2 | 3.8 (0.5–13.8) | 5 | 8.5 (2.8–19.9)* |
| Lymphocytic lymphoma | 6 | 1.02 | 5.9 (2–12.8)* | 1 | 2.1 (0.05–11.7) | 5 | 9.2 (3.0–21.6)* |
| Lymphocytic Leukemia | 2 | 0.4 | 5.0 (1–18.0) | 1 | 4.7 (0.12–26.1) | 1 | 5.3 (0.1–29.5) |
p<0.05
Other urinary organs includes urinary bladder, ureters, urethra, others, multiple parts, and not otherwise specified
Does not include basal cell carcinoma
Endocrine glands include thyroid, parathyroid, thymus, pituitary, adrenal, Beta cells of the pancreas, and others or unspecified
N.A. = not applicable.
DISCUSSION
In this large comprehensive population-based study, we found that patients diagnosed with thymoma had worse survival than their age- and gender- matched controls, and were at high risk of developing autoimmune diseases and second cancers. Previously, only a few population-based studies have evaluated different long-term patterns of the disease (7;9;11). Using the U.S. NCI SEER database, Engels et al. identified 849 thymoma cases diagnosed between 1973–1998 and estimated demographic incidence patterns, and risk of second cancers in a subset of 733 patients (7). Further, based on the SEER database, Travis et al. identified 815 thymoma patients diagnosed 1973–2000 (9), and found that second cancers occurred in 81 patients (Observed-to-expected [O/E] ratio = 1.54; 95% CI= 1.22–1.91). Using the Netherlands National Pathological Archives and the Netherlands Cancer Registry, de Jong et al. identified 537 thymomas and reported on incidence and diagnostic procedures as well as survival patterns in a subset of 232 cases between 1994–2003 (11). To our knowledge, our study is the first large comprehensive population-based study using matched controls to evaluate survival patterns and autoimmunity in patients with thymoma. In addition, we evaluated the risk of second malignancies in those patients using the general Swedish population estimates.
Based on 431 benign and 237 malignant thymoma cases, we found that thymoma patients have a significantly worse overall survival compared with matched controls. Thymoma or second malignancies were the most commonly reported cause of death in thymoma patients. Our findings are consistent with data reported by de Jong et al. from 232 cases, who were histologically classified according to the WHO classification, and derived from two nationwide Dutch databases (11). We found that survival patterns improved over time which, to a large degree, likely reflects improvements in disease-specific diagnostic approaches and therapeutic intervention (13), as well as general improvements in supportive care. This could also reflect the improved survival in the general population over time. To take that into account, we compared survival patterns among thymoma patients and their matched-controls at three different calendar-period intervals (<1973, 1973–1987, 1988–2004), and noticed the same previously-observed improvement over time; but, in the recent years, malignant thymoma patients still die, on average, 5.2 years earlier than controls compared with 1.1 for benign cases. Importantly, although age is associated with a poorer survival among thymoma patients, compared with controls, younger (≤40 years) patients had the largest excess mortality reflected in 8.4% for benign cases and 37.9% for malignant cases lower 10-year overall survival versus 3.5% and 15.6% for older patients (>60 years).
Our study provides evidence, in agreement with previous reports (4;14–17), of associations between thymoma and autoimmune diseases, particularly myasthenia gravis, red cell aplasia, SLE and RA. Our ability to evaluate these associations before, around, and after the development of thymoma provides evidence of associations that cannot be completely explained on the basis of paraneoplastic phenomena. In fact, the vast majority of autoimmune diseases were reported after thymoma diagnosis, reflected in the most pronounced risk estimates around and after (i.e., not before) thymoma diagnosis, when comparing risks of autoimmunity to those among matched controls. Interestingly, myasthenia gravis appeared to be a risk factor and a long-term consequence of thymoma; there was one thymoma case who was diagnosed with myasthenia gravis 38 years after thymoma diagnosis. The development of autoimmune disorders in patients with thymoma is thought to be related to alterations in T-cell development within the thymus (4;18–20).
We found a greater than 2-fold increase in the risk of developing a subsequent cancer in thymoma patients. This is in agreement with the United States population-based studies that reported an increase in the risk of subsequent cancers after thymoma (7;9). This observation might be explained by the immune system alteration in association with thymectomy. In support of that hypothesis, we observed significantly elevated risks of non-melanoma skin cancer, non-Hodgkin lymphoma, cancers of the cervix, the same cancers associated with severe immunosuppression disorders such as acquired immunodeficiency syndrome (AIDS) (21). In addition, we observed excess risk of endocrine, prostate, and gastrointestinal tract cancers (only in males). Similar findings were reported by Engels et al. for non-Hodgkin lymphoma, and gastrointestinal cancer, with the exception that they did not evaluate gender differences (7), and did not evaluate non-melanoma skin or cervical cancer. It is important to note that our finding of excess risk of non-melanoma skin, cervical and prostate cancer might be indicative of a detection bias, in which thymoma cases are under closer medical observation than the general population, and thus more likely to be diagnosed with a screening-related cancer.
Our study has several strengths, including the large sample size, the use of the nationwide Swedish cancer registry that allowed us to identify all patients with thymoma in Sweden 1958–2004 as well as the record-linkage with the Swedish inpatient/outpatient hospital discharge registry, and the population-based design that ensures generalizability of the results. However, the study was limited by the lack of detailed clinical and therapeutic information for the thymoma patients. As stated in the methods section, because this study was part of larger research effort, persons with a hematopoietic lymphoid malignancy were not eligible for control selection. To avoid biases due to this selection criterion, we used SIRs, based on population rates, to assess risk of second cancers. Any bias in the survival estimates due to this control selection should be small due to the relative rarity of these malignancies in the population. In addition, our study was limited by the lack of WHO pathological classification of thymoma patients. Several studies have demonstrated that the WHO classification of thymoma predicts patients’ outcome (22–25). The recent implementation of this classification and the rarity of thymic tumors make it challenging to currently assess the long-term outcomes after thymoma by WHO histological subtype. To our knowledge, only one prior population-based study has assessed survival in 232 thymoma cases diagnosed between 1994 and 2003 using this classification (11). Clearly, more studies are needed to evaluate survival and other outcomes by thymoma pathological subtypes based on the new classification.
In conclusion, although thymomas are considered to be relatively indolent solid tumors, survival is significantly worse compared with controls. Our study suggests evidence of improved management of thymoma over time, and confirms the necessity for follow-up and assessment of thymoma patients for the development of autoimmune diseases and second cancers.
Acknowledgments
This research was supported by grants from the Swedish Cancer Society, Stockholm County Council, the Karolinska Institutet Foundations, and the Intramural Research Program of the National Cancer Institute, USA. The authors thank Ms. Shiva Ayobi, the national board of health and welfare, Stockholm, Sweden, Ms. Emily Steplowski and Mr. Joe Barker, Information Management Services, Silver Spring, MD, for important efforts in the development of this database, and Dr. Mark H. Greene, Clinical Genetics Branch chief, for his valuable input throughout the study.
References
- 1.Schmidt-Wolf IG, Rockstroh JK, Schuller H, Hirner A, Grohe C, Muller-Hermelink HK, Huhn D. Malignant thymoma: current status of classification and multimodality treatment. Ann Hematol. 2003 Feb;82(2):69–76. doi: 10.1007/s00277-002-0597-6. [DOI] [PubMed] [Google Scholar]
- 2.Detterbeck FC, Parsons AM. Thymic tumors. Ann Thorac Surg. 2004 May;77(5):1860–9. doi: 10.1016/j.athoracsur.2003.10.001. [DOI] [PubMed] [Google Scholar]
- 3.Shields TW. In: Primary lesions of the mediastinum and their investigation and treatment. 4. Shields TW, editor. Baltimore: Williams and Wilkins Publishers; 1994. pp. 1724–96. [Google Scholar]
- 4.Evoli A, Minicuci GM, Vitaliani R, Battaglia A, Della MG, Lauriola L, Fattorossi A. Paraneoplastic diseases associated with thymoma. J Neurol. 2007 Jun;254(6):756–62. doi: 10.1007/s00415-006-0429-z. [DOI] [PubMed] [Google Scholar]
- 5.Kondo K, Monden Y. Thymoma and myasthenia gravis: a clinical study of 1,089 patients from Japan. Ann Thorac Surg. 2005;79(1):219–24. doi: 10.1016/j.athoracsur.2004.06.090. [DOI] [PubMed] [Google Scholar]
- 6.Monden Y, Uyama T, Taniki T, Hashimoto J, Fujii Y, Nakahara K, Kawashima Y, Masaoka A. The characteristics of thymoma with myasthenia gravis: a 28-year experience. J Surg Oncol. 1988 Jul;38(3):151–4. doi: 10.1002/jso.2930380305. [DOI] [PubMed] [Google Scholar]
- 7.Engels EA, Pfeiffer RM. Malignant thymoma in the United States: demographic patterns in incidence and associations with subsequent malignancies. Int J Cancer. 2003 Jul 1;105(4):546–51. doi: 10.1002/ijc.11099. [DOI] [PubMed] [Google Scholar]
- 8.Pan CC, Chen PC, Wang LS, Chi KH, Chiang H. Thymoma is associated with an increased risk of second malignancy. Cancer. 2001;92(9):2406–11. doi: 10.1002/1097-0142(20011101)92:9<2406::aid-cncr1589>3.0.co;2-7. [DOI] [PubMed] [Google Scholar]
- 9.Travis LB, Boice JD, Jr, Travis WD. Second primary cancers after thymoma. Int J Cancer. 2003;107(10):868–70. doi: 10.1002/ijc.11480. [DOI] [PubMed] [Google Scholar]
- 10.Welsh JS, Wilkins KB, Green R, Bulkley G, Askin F, ener-West M, Howard SP. Association between thymoma and second neoplasms. JAMA. 2000 Mar 1;283(9):1142–3. doi: 10.1001/jama.283.9.1142. [DOI] [PubMed] [Google Scholar]
- 11.de Jong WK, Blaauwgeers JL, Schaapveld M, Timens W, Klinkenberg TJ, Groen HJ. Thymic epithelial tumours: a population-based study of the incidence, diagnostic procedures and therapy. Eur J Cancer. 2008;44(1):123–30. doi: 10.1016/j.ejca.2007.11.004. [DOI] [PubMed] [Google Scholar]
- 12.Rosai J, Sobin LH World Health Organization. Histological typing of tumors of th thymus. 2. Berlin: Springer Verlag; 1999. [Google Scholar]
- 13.Tomaszek S, Wigle DA, Keshavjee S, Fischer S. Thymomas: review of current clinical practice. Ann Thorac Surg. 2009 Jun;87(6):1973–80. doi: 10.1016/j.athoracsur.2008.12.095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nishioka R, Nakajima S, Morimoto Y, Suzuki H, Nakamura H, Suzuki M. T-cell acute lymphoblastic leukemia with transient pure red cell aplasia associated with myasthenia gravis and invasive thymoma. Intern Med. 1995 Feb;34(2):127–30. doi: 10.2169/internalmedicine.34.127. [DOI] [PubMed] [Google Scholar]
- 15.Barritt AW, Bacsi AM, Tschuchnigg M, Sharpe D, Kiernan MC. Myasthenia gravis seven years after removal of an invasive thymoma. J Clin Neurosci. 2009 Jul;16(7):966–8. doi: 10.1016/j.jocn.2008.09.016. [DOI] [PubMed] [Google Scholar]
- 16.Thomas CR, Wright CD, Loehrer PJ. Thymoma: state of the art. J Clin Oncol. 1999 Jul;17(7):2280–9. doi: 10.1200/JCO.1999.17.7.2280. [DOI] [PubMed] [Google Scholar]
- 17.Bjorkholm M, Holm G, Mellstedt H, Carlberger G, Nisell J. Membrane bound IgG on erythroblasts in pure red cell aplasia following thymectomy. A case report. Scand J Haematol. 1976 Nov;17(5):341–6. doi: 10.1111/j.1600-0609.1976.tb01444.x. [DOI] [PubMed] [Google Scholar]
- 18.Ströbel P, Rosenwald A, Beyersdorf N, Kerkau T, Elert O, Murumägi A, Sillanpää N, Peterson P, Hummel V, Rieckmann P, Burek C, Schalke B, et al. Selective loss of regulatory T cells in thymomas. Ann Neurol. 2004;56(6):901–4. doi: 10.1002/ana.20340. [DOI] [PubMed] [Google Scholar]
- 19.Okumura M, Fujii Y, Shiono H, Inoue M, Minami M, Utsumi T, Kadota Y, Sawa Y. Immunological function of thymoma and pathogenesis of paraneoplastic myasthenia gravis. Gen Thorac Cardiovasc Surg. 2008 Apr;56(4):143–50. doi: 10.1007/s11748-007-0185-8. [DOI] [PubMed] [Google Scholar]
- 20.Bjorkholm M, Holm G, Johansson B, Mellstedt H. T-Lymphocyte deficiency following adult thymectomin man. Scand J Haematol. 1975 May;14(3):210–5. [PubMed] [Google Scholar]
- 21.Engels EA, Biggar RJ, Hall HI, Cross H, Crutchfield A, Finch JL, Grigg R, Hylton T, Pawlish KS, McNeel TS, Goedert JJ. Cancer risk in people infected with human immunodeficiency virus in the United States. Int J Cancer. 2008 Jul 1;123(1):187–94. doi: 10.1002/ijc.23487. [DOI] [PubMed] [Google Scholar]
- 22.Chen G, Marx A, Wen-Hu C, Yong J, Puppe B, Stroebel P, Mueller-Hermelink HK. New WHO histologic classification predicts prognosis of thymic epithelial tumors: a clinicopathologic study of 200 thymoma cases from China. Cancer. 2002 Jul 15;95(2):420–9. doi: 10.1002/cncr.10665. [DOI] [PubMed] [Google Scholar]
- 23.Kondo K, Yoshizawa K, Tsuyuguchi M, Kimura S, Sumitomo M, Morita J, Miyoshi T, Sakiyama S, Mukai K, Monden Y. WHO histologic classification is a prognostic indicator in thymoma. Ann Thorac Surg. 2004 Apr;77(4):1183–8. doi: 10.1016/j.athoracsur.2003.07.042. [DOI] [PubMed] [Google Scholar]
- 24.Okumura M, Ohta M, Tateyama H, Nakagawa K, Matsumura A, Maeda H, Tada H, Eimoto T, Matsuda H, Masaoka A. The World Health Organization histologic classification system reflects the oncologic behavior of thymoma: a clinical study of 273 patients. Cancer. 2002 Feb 1;94(3):624–32. doi: 10.1002/cncr.10226. [DOI] [PubMed] [Google Scholar]
- 25.Rena O, Papalia E, Maggi G, Oliaro A, Ruffini E, Filosso P, Mancuso M, Novero D, Casadio C. World Health Organization histologic classification: an independent prognostic factor in resected thymomas. Lung Cancer. 2005 Oct;50(1):59–66. doi: 10.1016/j.lungcan.2005.05.009. [DOI] [PubMed] [Google Scholar]