Outcome in excised thymomas: role of prognostic factors and impact of additional malignancies on survival

Abstract

Background and aims:

Although the management of thymomas has been extensively evaluated, the value of prognostic factors in the outcome of these patients remains unclear.

Methods and results:

The medical records of all patients who underwent resection of thymoma between January 1985 and September 2010 at a single thoracic unit were reviewed. Patients were followed up with reference to disease recurrence and development of additional malignancies (AM).

Total thymectomy was performed in all 68 cases. Mean follow-up time was four years. Mean survival was 63.9 months. Mean disease-free interval was 13 months. Factors affecting prognosis were Masaoka staging and WHO histological sub-type. Patients with thymomas had a higher risk of developing AM when compared with a control population of individuals with other tumours (p = 0.0002). Among thymomas, the cortical subtype was associated with a higher risk of AM (p = 0.047) and mortality (p = 0.001).

Conclusions:

This data confirms that Masaoka staging and WHO histologic sub-type are the most important prognostic factors in patients with thymoma. Moreover, thymomas predominantly arising from the thymic cortex are associated with a higher risk of developing other malignancies and with poorer survival. The cortical origin of thymoma could therefore be considered as a significant prognostic factor.

Keywords

Thymoma neoplasms prognosis

Introduction

Thymomas are slow-growing neoplasms arising from thymic epithelium.¹ Recent reports have shown a very low incidence (0.15 per 100,000 person years in the USA²). However, they represent the most common tumour in the anterior mediastinum, accounting for 50% of anterior mediastinal masses in adults.³ The clinical behaviour of thymomas reflects their heterogeneity in terms of biological and histological features. The majority of thymomas are asymptomatic and are often diagnosed incidentally on X-ray or Computed Tomography (CT) Scan. Less commonly, they are associated with immunological diseases such as Myasthenia Gravis (MG).⁴ Most thymomas are encapsulated, but approximately 20% invade surrounding tissues.⁵ Metastases are usually confined to intra-thoracic targets (pleura, pericardium and diaphragm); however, thymomas show a propensity for recurrence even after complete removal.⁶ Histopathologically, thymomas arise exclusively from the epithelial cells of the thymus and have a variable lymphocytic component.⁷ They are currently classified according to the World Health Organisation (WHO) criteria.⁸

The aim of this study is to retrospectively review the experiences of a single centre in the treatment of thymomas over a 25-year period. In addition, the long-term survival and the role of prognostic factors as described in the WHO classification and Masaoka staging system is examined. Finally, the role of additional malignancies (AM) in influencing the long-term outcome and survival of these patients is evaluated.

Material and methods

The medical records of 68 patients who underwent surgical excision of thymomas from January 1985 to September 2010 at University Hospital of Siena, Thoracic Surgery Unit, were reviewed. All pathological specimens underwent histological review according to the recently published World Health Organization (WHO)⁸ classification. Thymomas were classified as follows: type A with spindle/oval shape epithelial cells, type B with plump epithelioid cells (further subdivided in accordance with the extent of lymphocytic infiltrate and the degree of atypia of epithelial cells into B1, B2 and B3 subtypes) and AB thymomas, with mixed features.

Clinical data concerning onset of symptoms, diagnostic procedures and therapeutic approach were collected and are summarized in Table 1. Results of imaging studies were included. Chest X-rays and CT scans were performed for every patient, with a small number of patients also receiving MRI. The severity of MG was assessed according to Osserman’s classification.⁹

Table 1.

Demographic and clinical data.

Thymoma group
Sex	M 36 (52.9%) F 32 (47.1%)
Age (years)	18–80 (59.7 ± 12,2)
Clinical features		No. of cases (%)
Symptoms	Asymptomatic	26 (38)
Dyspnoea	12 (18)
Cough	14 (21)
MG related symptoms	13 (19)
Thoracic pain	9 (13)
SVC syndrome	3 (4)
Co-morbidity	MG	13 (19)
Additional Malignancies	16 (24)
Diagnostic pre-operative procedures	FNAB biopsy	13 (19)
Chamberlain’s biopsy	10 (15)
Surgical approach	Longitudinal median sternotomy	39 (57)
Sternal split	7 (10)
Left VATS plus cervicotomy	10 (15)
Postero-lateral thoracotomy	12 (18)
Adjacent organ resected	Pericardium	4 (6)
SVC	3 (4)
Lung Wedge Resection	7 (10)
Mortality	Peri-operative ARDS	3 (4)
Complications	Respiratory failure	3 (4)
Pneumonia	1 (1)
Lobar atelectasis	1 (1)
Phrenic nerve palsy	1 (1)
Relapse	Stage II thymomas	2 (3)
Stage III thymomas	5 (7)
Adjuvant therapy	CHT	12 (18)
RT	12 (18)

Additional malignancy was defined as any epithelial or mesenchymal malignant neoplasm concomitantly or subsequently detected during the follow-up period.

Diagnosis was made pre-operatively by fine-needle aspiration biopsy (FNAB) in 13 cases (24%) and Chamberlain’s biopsy in 10 cases (15%). The histological diagnosis of thymoma was made post-operatively in all other patients.

Histopathological features are summarised in Table 2. Assessment of clinical stage was performed according to the Masaoka classification.

Table 2.

Histo-pathological classification.

WHO histotype
A	AB	B1	B2	B3	Total
Masaoka stage	I	2	5	2	4	–	13
II	7	11	15	6	4	43
III	3	3	–	2	2	10
IV	–	–	–	–	2	2
Total	12	19	17	12	8	68

In all cases, radical surgery was performed to remove all neoplastic tissue and a surrounding margin of mediastinal fat.

Adjuvant treatment, consisting in chemotherapy (ADOC, A-Adriamycin D-Doxarubin O C-Carbo/cis Platinum) and radiation therapy (multiple sessions for a total dose between 50 and 100 Gy), was administered to patients with stages III and IV disease and in cases of recurrent thymomas.

Patients had a clinic review at six weeks post-operatively and were then followed up through surgical archives, the cancer registry and multidisciplinary review by members of the healthcare team.

Statistical analysis was carried out using SPSS software version 16.0 (SPSS, Chicago, IL, USA). The Chi-squared test was used to investigate the association between category variables. The Kaplan–Meier test was used to calculate the survival rate and multivariate analysis through the Cox-regression model allowed us to identify independent prognostic factors. A p-value of less than 0.05 was considered statistically significant.

Results

We observed 36 men (53%) and 32 (47%) women with a median age of 64 (range 18–80). 26 patients (38%) were asymptomatic and were diagnosed following an incidental finding of an anterior mediastinal mass on chest X-ray or CT scan. Fourteen patients (21%) described cough as their only symptom, 13 patients (19%) had symptoms relating to MG, 10 (15%) patients had dyspneoa, 9 (13%) had thoracic pain and 3 (4%) were affected by superior vena cava obstruction. No immunological syndrome other than MG was observed. An additional malignancy was found in 22 patients (32%).

Chest X-ray showed mediastinal enlargement in 51 cases (75%); the mass was associated with a right pleural effusion in four cases (6%) and right diaphragm paralysis in eight cases (12%). CT showed a solid thymic tumour in 46 cases (68%) and cystic tumour in 22 (32%).

Surgery was performed through a total vertical sternotomy in 39 cases (57%), postero-lateral thoracotomy in 12 cases (18%), cervicotomy and left VATS in 10 cases (15%) and sternal split in 7 cases (10%). Total thymectomy including removal of surrounding mediastinal fat was performed in all cases. Four cases (6%) required pericardial resection due to tumour infiltration (6%). In three cases (4%), the involvement of superior vena cava led to its resection and prosthetic replacement with mesh vein graft. Seven (10%) patients underwent wedge resection of the lung due to pulmonary invasion.

No intra-operative mortality was observed. Three patients (4%) died of acute respiratory distress syndrome (ARDS) in the 15 days following surgery. Post-operative complications occurred in seven patients (10%): three patients with MG (4%) developed respiratory failure, one (1%) developed pneumonia and one (1%) experienced left lower lobe atelectasis due to unilateral phrenic nerve paralysis.

Adjuvant chemotherapy (ADOC) and radiotherapy was administrated to 12 patients (18%) corresponding to Masaoka grades III and IV thymomas.

Seven patients (10%) with radical resection had a recurrence. Mean disease free interval was 13 months (SD 9.9, range 5–30) All recurrences were intrathoracic; they were restricted to lungs in three cases (4%), whereas pleuro-pulmonary involvement occurred in four cases (5%). Recurrences were treated by re-resection followed by chemotherapy and radiotherapy.

The mean follow-up time was six years (range 1–13 years). 43 (63%) patients died from progression of the primary disease. The mean survival time was 63.9 months (SD ± 57.9 months); the survival rate was 67% at 36 months and 50% at 60 months (Figure 1).

Figure 1.

Overall survival.

During the follow-up period, 22 patients (24%) experienced an additional tumour. Patients with thymomas demonstrated a statistically significant higher risk of developing an AM when compared with patients affected by primary cancer other than thymoma (p = 0.0002). Moreover, association between thymic tumours and AM was significantly related to thymic neoplasm histotype: with cortical subtypes having a higher risk of AM when compared with the medullary type (p = 0.0474) (Table 3).

Table 3.

Thymoma and additional malignancies.

Without AM (%)	With AM (%)	Total (%)
Thymoma	46 (68)	22 (32)	68 (100)	p = 0.0002
CC group	103 (90)	11 (10)	114 (100)
TC group	115 (94)	8 (6)	123 (100)
HL/NHL group	100 (93)	8 (7)	108 (100)	p > 0.05
A thymomas	11 (92)	1 (8)	12 (100)	p = 0.047
B1, B2, B3, and AB thymomas	35 (62)	21 (38)	56 (100)
Total	46 (68)	22 (32)	68 (100)

Note: Comparison between Thymoma group and Control Groups. Relationship between histotype and additional malignancies.

CC: Colorectal carcinoma; TC: thyroid carcinoma; HL/NHL: Hodgkin and non-Hodgkin lymphomas; AM: additional malignancy.

There was no statistically significant associations between the development of AM and tumour stage, gender, administration of radiation and/or chemotherapy.

In univariate analyses, the clinical stage and the WHO histological subtype showed a significant impact on overall survival (p < 0.05). The development of AM was found to significantly worsen the survival of patients with thymoma on log rank test (p = 0.001) (Figure 2). Multivariate analysis confirmed that the only factors with prognostic significance in thymomas were Masaoka’s staging and WHO’s histological sub-type; stages III and IV tumours and B2/B3 histotypes were associated with shorter survival than stage I/II tumours and A/AB/B1 thymomas (p = 0.02 and p = 0.03, respectively) (Table 4, Figures 3 and 4).

Figure 2.

Impact of additional malignancies (AM) on survival. Log rank test.

Figure 3.

Survival curve of group A-AB-B1 vs B2-B3.

Figure 4.

Survival curve of stages I and II vs stages III and IV.

Table 4.

Statistical analysis. Impact on survival of prognostic factors.

Statistical tests	Fixed variable	p value
Univariate analysis	WHO	0.007
Univariate analysis	Stage	0.017
Multivariate analysis	WHO	0.03
Multivariate analisys	Stage	0.02
Log rank test	Additional malignancy	0.0001

Discussion

Thymomas represent a wide spectrum of neoplasms with different biological behaviours and prognosis.

In one-third of patients, these tumours are asymptomatic and are diagnosed as incidentally following routine chest radiological investigations.¹⁰ In another third of patients, thymomas are detected because of local symptoms due to infiltration or compression of neighbouring organs. The remaining patients exhibit features of an immunologic tumour-related syndrome, especially MG. Controversy still exists about the effect of MG on the survival of patients with thymomas. According to research by Kohman and Rios,^11,12 there was no evidence to support MG being associated with worse outcomes. This may be because MG could potentially facilitate earlier tumour diagnosis as symptoms alert the clinician to the possibility of underlying thymic neoplasia. However, Appelqvist et al. suggests that when associated with MG, thymic tumours have a worse prognosis.¹³

A medistinal mass on chest X-ray may represent a diagnosis of thymoma. However, the differential diagnosis includes a wide range of neoplasms, e.g. lymphoma, thyroid and parathyroid tumours, germ cell neoplasms and benign diseases such as thymic hyperplasia, pericardial, thymic, bronchial and oesophageal cysts, retrosternal goitre, lymphadenitis and vascular aneurysm.

CT allows more accurate characterisation of the mass and its relationship to adjacent structures. Anecdotal evidence suggests that MRI has an important role in the assessment of mediastinal masses; however, more studies are required in this area.¹⁴ In those cases where imaging has proved equivocal tissue diagnosis by FNAB and mediastinal biopsy can be performed. In our series, the aforementioned invasive procedures were performed in 18 cases (26%); in only one case of direct resection, post-operative specimen examination revealed the presence of a germ cell tumour.

Several papers^15–17 assessed the role of radical resection in establishing long-term survival. The cornerstone of thymoma surgery is the complete removal of thymus and surrounding fatty tissue, and it has been considered as mandatory in all cases. The majority of recently published studies suggest the use of surgery alone for the treatment of stage I and II thymomas.¹⁵ In the advanced III and IV stages, even though many researchers agree about the importance of a multidisciplinary approach, the prognosis is mainly dependant on the surgeon’s ability to remove as much neoplastic tissues as possible.^16,17 In our series, surgery was performed with radical intent in all cases. Fifty-six (82%) patients with stages I and II disease were successfully treated and showed a mean survival time of 68.5 months (SD ± 59.2 months); the survival rate was 72% at 36 months and 55% at 60 months. Only seven patients (10%) with stages III and IV had intra-thoracic recurrence. This group of patients, after radical resection followed by adjuvant treatment, showed excellent long-term survival which was comparable with that of patients treated by radical resection without recurrence. The type of surgical treatment was not found to be an independent prognostic factor and was not significant in multivariate analysis.

Chemotherapy was initially performed only in inoperable patients with thymoma.¹⁸ After positive results in this group of patients, it was also used as adjuvant or neo-adjuvant treatment. Although single agents have shown to have favourable outcomes in recurrent, metastatic and locally advanced disease, combination regimens had better response rates. To date, the best results were obtained using PAC and ADOC regimens, with a higher rate of early remission for ADOC.¹⁹ Non-invasive thymomas are currently considered as a localized disease with a very low potential for distant metastases, whereas invasive thymomas when treated only by radical surgery develop distant metastasis in 5% of cases and relapse in 20%.²⁰ Therefore, in the present series, stages III and IV and recurrent thymomas underwent ADOC regimen.

The role of radiation therapy has never been evaluated in prospective randomised studies. Nevertheless, many retrospective studies have shown significant improvement in disease control and survival in patients treated with adjuvant radiation therapy.²¹

Neo-adjuvant chemotherapy or radiotherapy has recently been used with encouraging results. Even though neo-adjuvant multimodal protocols may improve resectability or at least make resection easier, it is not clear whether these treatments affect overall survival or are really effective after complete tumour resection.²²

In our series, the adjuvant protocol included ADOC and 50–100 Gy radiation therapy in patients with Masaoka stages III and IV and in recurrent thymomas.

The survival rate after five years was 34% in stages III and IV thymomas, 27% in recurrent thymomas. Even though adjuvant chemotherapy was well tolerated with a trend towards improved patient survival, neither therapy showed significant benefit on multivariate analysis.

Literature suggests that the incidence of AM in patients with thymomas ranges between 3 and 27%.²³ In our series, the rate was higher at 32% (22 patients). Our previous research²⁴ shows that thymoma patients have a significantly greater risk of developing AM compared with patients affected by other forms of malignancy. In particular, among thymoma’s histologic subtypes, prevalently cortical tumours represent a significantly higher risk of AM.

Our results confirm the prognostic value of Masaoka staging system in long-term outcomes,^25,26 suggesting longer survival in patients with stages I and II disease. Fifty-six patients with stages I and II disease were successfully treated (i.e. no evidence of local recurrence at time of follow-up) and showed significantly better survival in univariate and multivariate analysis when compared with patients with stages III and IV disease.

Only a few studies^25–29 have evaluated the clinical and prognostic relevance of WHO classification. In these studies, the neoplastic transformation of thymic epithelial cells was considered as a biological continuum from the benign clinical aspect of A, AB and B1 types to the more aggressive behaviour of B2 and B3 tumours. As a consequence, B2 and B3 thymomas showed poor survival and high recurrence rate, whereas A, AB and B1 thymomas are currently considered to be benign tumours with low invasive and metastatic potential. Our series demonstrated that the WHO histologic type was, together with clinical stage, the only significant prognostic parameter in univariate and multivariate analysis in terms of survival. In particular, patients with A, AB and B1 thymomas demonstrated longer survival.

In this paper, the impact of AM on overall survival has been evaluated. Patients with a diagnosis of AM demonstrated a significantly poorer outcome in terms of survival (p = 0.001), underlining the importance of cancer surveillance in this group of patients.

In conclusion, this study re-iterates the results of existing studies which demonstrated that the clinical stage and WHO histologic sub-type are the most important prognostic factors in patients with diagnosis of thymoma. In particular, epithelial tumours prevalently arising from the thymus cortex were associated with a higher risk of developing additional malignancy and shorter survival. The cortical origin of thymoma could therefore be considered as another significant prognostic element. Prospective studies are however required in order to examine this further.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Levine

Rosai

. Thymic hyperplasia and neoplasia: a review of current concepts. Human Pathol 1978; 9: 495–515.

Engels

Pfeiffer

. Malignant thymoma in the United States: demographic patterns in incidence and association with subsequent malignancies. Int J Cancer 2003; 105: 546–551.

Detterbeck

Pearson

. Thymic tumours. Ann Thoracic Surg 2004; 77: 1860–1869.

Arriagada

Gerard-Merchant

Tubiana

. Radiation therapy in the management of malignant thymic tumours. Acta Radiol Oncol 1981; 20: 167–172.

Bergh

Gatzinsky

Larsson

. Tumours of the thymus and thymic region. Clinicopathological studies on thymomas. Ann Thoracic Surg 1979; 28: 252–256.

Gray

Gutowski

III

WT . Thymoma. A clinicopathologic study of 54 cases. Am J Surg Pathol 1979; 3: 235–249.

Lauriola

Maggiano

Marino

. Human thymoma: immunologic characteristics of the lymphocytic component. Cancer 1981; 48: 1992–1995.

Rosai

. World Health Organization histological typing of tumours of the thymus, 2nd ed. New York, NY: Springer-Verlag, 1999.

Osserman

Genkins

. Studies in myasthenia gravis: review of a twenty years experience in over 1200 patients. Mt Sinai J Med 1971; 38: 497–537.

10.

Loehrer

Sr . Current approaches to the treatment of thymoma. Ann Med 1999; 31: 73–79.

11.

Kohman

. Controversies in the management of malignant thymoma. Chest 1997; 112: 296–300.

12.

Ríos

Torres

Galindo

. Prognostic factors in thymic epithelial neoplasms. Eur J Cardiothoracic Surg 2002; 21: 307–313.

13.

Appelqvist

Kostiainen

Franssila

. Treatment and prognosis of thymoma: a review of 25 cases. J Surg Oncol 1982; 20: 265–268.

14.

Camera

Brunetti

Romano

. Morphological imaging of thymic disorders. Ann Med 1999; 31: 57–62.

15.

Mangi

Wright

Allan

. Adjuvant radiation therapy for stage II thymoma. Ann Thoracic Surg 2002; 74: 1033–1037.

16.

Regnard

Magdeleinat

Dromer

. Prognostic factors and long-term results after thymoma resection: a series of 307 patients. J Thorac Cardiovasc Surg 1996; 112: 376–384.

17.

Rea

Marulli

Girardi

. Long-term survival and prognostic factors in thymic epithelial tumours. Eur J Cardiothorac Surg 2004; 26: 412–418.

18.

Boston

. Chemotherapy of invasive thymoma. Cancer 1976; 38: 49–52.

19.

Hernandez-Ilizaliturri

Tan

Cipolla

. Multimodality therapy for thymic carcinoma (TCA): results of a 30 year single institution experience. Am J Clin Oncol 2004; 27: 68–72.

20.

Jackson

Ball

. Post-operative radiotherapy in invasive thymoma. Radiother Oncol 1991; 21: 77–82.

21.

Ohara

Tatsuraki

Fuji

. Radioresponse of thymomas verified with histologic response. Acta Oncol 1998; 37: 471–474.

22.

Venuta

Rendina

Pescarmona

. Multimodality treatment of thymoma: a prospective study. Ann Thorac Surg 1997; 64: 1585–1591.

23.

Pan

Chen

Wang

. Thymoma is associated with an increased risk of second malignancy. Cancer 2001; 92: 2406–2411.

24.

Granato

Ambrosio

Spina

. Patients with thymomas have an increased risk of developing additional malignancies: lack of immunological surveillance? Histopathology 2012; 60: 437–442.

25.

Okumura

Ohta

Tateyama

. The World Health Organization histologic classification system reflects the oncologic behaviour of thymoma: a clinical study of 273 patients. Cancer 2002; 94: 624–632.

26.

Chen

Marx

Wen-Hu

. New WHO histologic classification predicts prognosis of thymic epithelial tumours: a clinicopathologic study of 200 thymoma cases from China. Cancer 2002; 95: 420–429.

27.

Rieker

Hoegel

Horresi-Hauf

. Histologic classification of thymic epithelial tumors: comparison of established classification schemes. Int J Cancer 2002; 98: 900–906.

28.

Chalabreysse

Roy

Cordier

. Correlation of the WHO schema for the classification of thymic epithelial neoplasms with prognosis: a retrospective study of 90 tumours. Am J Surg Pathol 2002; 26: 1605–1611.

29.

Marchevsky

Gupta

McKenna

. Evidence-based pathology and pathologic evaluation of thymomas. Cancer 2008; 112: 2780–2788.