Postoperative radiotherapy for patients with completely resected thymoma

A multi-institutional, retrospective review of 103 patients

Authors


Abstract

BACKGROUND

Optimal management of postoperative radiotherapy for patients with completely resected thymoma remains controversial. This study was conducted to assess the efficacy of postoperative mediastinal irradiation in patients with completely resected thymoma.

METHODS

The records of 103 patients with completely resected thymoma who received postoperative mediastinal irradiation during the period between 1979 and 1998 were reviewed. The distribution according to Masaoka stage was Stage I in 17 patients, Stage II in 61 patients, and Stage III in 25 patients. Fifty-two patients were treated with involved field (IF) irradiation, and 51 patients were treated with irradiation of the whole mediastinal field with or without boost (WM irradiation). The total radiation dose to the primary tumor bed was 30–61 grays (Gy), with a median dose of 40 Gy. No patients received chemotherapy during the initial treatment. The median follow-up of the 82 living patients was 112 months (range, 24–244 months).

RESULTS

The 10-year actuarial overall and disease free survival rates for all patients were 81% and 79%, respectively. The 10-year actuarial overall survival rate was 100% for patients with Stage I disease, 90% for patients with Stage II disease, and 48% for patients with Stage III disease. In the analysis, clinical stage alone had a statistically significant impact on both overall survival and disease free survival (P < 0.0001 for both). Recurrent disease was observed in 17 patients, and the pleura was the most frequent site of first recurrence. Of 12 patients who had pleural recurrences, 11 patients had pleural dissemination remote from the initial tumor site. No recurrence was observed in any of the 17 patients with Stage I disease, and 6 of 61 patients (10%) with Stage II disease and 11 of 25 patients (44%) with Stage III disease experienced recurrences. With regard to intrathoracic recurrences, there were no recurrences within the irradiated field in any of the 103 patients, and no dose response correlation was seen in intrathoracic control (incidence of intrathoracic recurrence: 2 of 19 patients in the group that received < 40 Gy, 6 of 45 patients in the group that received 40 Gy, and 7 of 39 patients in the group that received > 40 Gy). With respect to treatment field, mediastinal recurrences were observed in 4 of 52 patients (8%) who were treated with IF irradiation, whereas 0 of 51 patients who were treated with WM irradiation experienced mediastinal recurrences. Pleural-based recurrences were observed both in patients who were treated with IF irradiation (7 of 52 patients) and in patients who were treated with WM irradiation (5 of 51 patients). According to the degree of pathologic tumor invasion, 0 of 71 patients without pleural invasion had pleural-based recurrences (0 of 17 patients with Stage I disease, 0 of 51 patients with Stage II disease, and 0 of 3 patients with Stage III disease), whereas 12 of 32 patients (38%) with pleural invasion had pleural-based recurrences (4 of 10 patients with Stage II disease and 8 of 22 patients with Stage III disease).

CONCLUSIONS

The current results indicated that WM irradiation with a total dose of 40 Gy was effective in preventing mediastinal recurrence for patients with completely resected thymoma. However, in patients with pathologic pleural invasion of the tumor, mediastinal irradiation alone was insufficient to avoid pleural-based recurrence. Cancer 2002;94:1405–13. © 2002 American Cancer Society.

DOI 10.1002/cncr.10373

Thymomas are rare neoplasms that originate from the epithelial cells of the thymus and are the most common tumors of the anterosuperior mediastinum, representing 20–30% of all malignant mediastinal tumors.1–6 These tumors occur with equal gender incidence, mainly in the group age 40–60 years.4 Presentation is usually as an asymptomatic mass on chest radiograph or with associated symptoms, such as myasthenia gravis.7, 8 Clinical staging for patients with thymoma was introduced by Bergh et al. in 19789 and subsequently was modified by Masaoka et al. in 1981.10 Currently, the Masaoka classification has been the most widely used staging system. Malignancy generally is determined by invasiveness of tumor rather than by histologic type.11, 12 However, all thymomas potentially are invasive and should be considered malignant.11, 13 With respect to the prognosis for patients with thymoma, factors that clearly affect treatment outcome are the invasiveness of the tumor4, 7, 8, 10, 14–20 and the extent of surgical resection.10, 11, 14, 18–25 The importance of myasthenia gravis7, 8, 11, 16, 17, 19, 20, 26 and the histologic subclassification3, 4, 7, 12, 17, 19, 20, 27–29 as independent prognostic factors remains controversial.

Although surgery remains the treatment of first choice for patients with thymoma, radiotherapy is used widely in patients with invasive thymoma and in some patients with noninvasive thymoma. The role of postoperative radiotherapy after incomplete resection or biopsy for invasive thymoma has been well established.7, 15, 24, 29–31 Conversely, there is no clear consensus regarding the optimal adjuvant treatment of patients with thymoma after they undergo complete resection. In patients with completely resected, Masaoka Stage I (noninvasive) tumors, routine postoperative irradiation has not been recommended because of the low incidence of recurrence even without postoperative irradiation.7, 16, 32 In contrast, in patients with completely resected, Stage II and III (invasive) tumors, postoperative mediastinal irradiation has been recommended routinely to prevent local recurrence.7, 13, 16, 17, 33–35 However, a controversy exists concerning the optimal postoperative radiotherapy for tumors at these stages, especially with regard to the total dose and treatment field.32, 36 Some authors have reported low recurrence rates with the use of postoperative mediastinal irradiation alone in patients with Stage II and III tumors and have pointed out the efficacy of this treatment method.7, 15, 17, 36 Others have found that postoperative mediastinal irradiation may be insufficient, because recurrences of pleural dissemination sometimes occurred remote from the initial tumor site even in patients with Stage II tumors.32, 37 Furthermore, the optimal dose of postoperative radiotherapy in patients who undergo complete resection has remained unclear.

In this study, we reviewed a retrospective, multi-institutional series of 103 patients with completely resected thymoma who received postoperative mediastinal irradiation. The efficacy of adjuvant mediastinal irradiation was assessed for patients with completely resected thymoma.

MATERIALS AND METHODS

Patient Characteristics

A retrospective review of medical records from 1979 to 1998 identified 103 patients with documented, completely resected thymoma who were treated with postoperative mediastinal irradiation at the Department of Radiology, University of the Ryukyus Hospital, National Okinawa Hospital, Chiba University Hospital, Yamanashi Medical University Hospital, Asahikawa Medical University Hospital, and International Medical Center of Japan. Patients who were treated with entire hemithorax irradiation were excluded from this study. Patient ages ranged from 6 years to 79 years (median, 52 years); 58 patients were male, and 45 patients were female. On presentation, there was an asymptomatic mass on chest radiograph in 63 patients; myasthenia gravis in 31 patients; symptoms of mediastinal compression, such as chest pain and dyspnea, in 7 patients, erythroblastopenia in 1 patient; and symptoms of both mediastinal compression and myasthenia gravis in 1 patient.

Pathologic Classification

All pathologic specimens and operative reports were reviewed, and the diagnosis of thymoma was based on the pathologic classification of Rosai and Levine, i.e., neoplasms of thymic epithelial cells regardless of the presence of a lymphoid component.1 This definition was modified later by the same authors, such that thymic tumors containing cells with the cytologic aspect of malignancy were classified separately as thymic carcinoma2: Therefore, patients with such carcinomas, who have a very different clinical outcome,4, 18 were excluded from this study. The tumors were classified into lymphocytic type tumors (n = 29 patients), mixed lymphoepithelial type tumors (n = 41 patients), epithelial type tumors (n = 31 patients), and spindle cell type tumors (n = 2 patients), as described by Bernatz et al.26

Clinical Staging

Clinical staging was based on the surgical and pathologic criteria described by Masaoka et al.10 The criteria used in this study were as follows: Stage I, macroscopically completely encapsulated and no microscopic capsular invasion; Stage II, 1) macroscopic invasion into surrounding fatty tissue or mediastinal pleura and 2) microscopic invasion into the capsule; Stage III, macroscopic invasion into neighboring organs (pericardium, great vessels, or lung); Stage IVa, pleural or pericardial dissemination of thymoma; and Stage IVb, lymphogenous or hematogenous metastases.

The distribution of patients according to stage was as follows: 17 patients had Stage I tumors, 61 patients had Stage II tumors, and 25 patients had Stage III tumors. Of 17 patients with Stage I tumors, 10 patients had significant peritumoral adherences, as assessed by intraoperative analysis of the surgeon, and the remaining 7 patients were without peritumoral adherences. In all patients, the degree of pathologic tumor invasion was assessed, such as capsule, mediastinal fat, pleura, and neighboring organs. This was based on pathologic microscopic invasion to the surrounding tissues or neighboring organs.

Radiotherapy

Radiotherapy was administered with a 60Co teletherapy unit or with a 6-MV or 10-MV linear accelerator (60Co teletherapy unit; n = 6 patients). The mediastinal irradiation fields were classified into two patterns: 52 patients were treated with involved field (IF) irradiation that covered the primary tumor bed with margins of about 1–2 cm, and 51 patients were treated with irradiation of the whole mediastinal field with or without boost (WM irradiation), including the primary tumor bed, with the upper margin at the thoracic inlet and the lower margin at the diaphragmatic crurae. Patients were treated with anteroposterior opposed fields with the spinal cord dose limited to 45 grays (Gy) in all patients (up to 41.4 Gy for the whole mediastinal field). Two anterior, wedged portals or off-cord, oblique, opposed portals were used to boost the anterior mediastinum to higher doses. The supraclavicular fossa was included in the irradiated field in 2 of 103 patients (2%). The total dose to the primary tumor was 30–61 Gy (median, 40 Gy). Daily fraction sizes of 1.8–2.0 Gy 5 days per week were used the most. However, one patient age 6 years at the time of treatment for thymoma received a total dose of 40 Gy, namely, anteroposterior opposed fields (20.8 Gy) and two anterior wedged fields (19.2 Gy), with a daily fraction size of 1.6 Gy to reduce the chance of later complications.

Chemotherapy

No patients received chemotherapy during the initial treatment.

Statistical Analysis

To assess the prognostic factors for overall and disease free survival, the data were analyzed with respect to age (< 50 years or ≥ 50 years), gender (male or female), pathologic classification (lymphocytic, mixed lymphoepithelial, epithelial, or spindle cell), mediastinal compression (present or absent), myasthenia gravis (present or absent), Masaoka stage (Stage I, Stage II, or Stage III), total dose (< 40 Gy, 40 Gy, or > 40 Gy), and treatment field (IF or WM). The median follow-up for the 82 living patients was 112 months (range, 24–244 months), and no patient was lost to follow-up. Overall and disease free survival rates, intrathoracic control rates, and pleural recurrence free rates were calculated actuarially according to the Kaplan–Meier method38 and were measured from the day of surgery. Intrathoracic failure was defined as the appearance of a tumor at an intrathoracic lesion, such as the mediastinum, pleura, pericardium, or lung. Differences between groups were estimated using the log-rank test.39 A probability level (P) of 0.05 was chosen for statistical significance. Statistical analysis was performed using the SPSS software package (version 6.1; SPSS, Inc., Chicago, IL).

RESULTS

Survival

Twenty-one of 103 patients (23%) died during the period of this analysis. Fourteen patients (14%) died of thymoma, and the remaining 7 patients died without any sign of clinical recurrence (due to myasthenia gravis in 2 patients, renal failure in 1 patient, lung carcinoma in 1 patient, hepatocellular carcinoma in 1 patient, rectal carcinoma in 1 patient, and glioblastoma in 1 patient). The 10-year actuarial overall and disease free survival rates for all patients were 84% and 81%, respectively. Figure 1 presents the actuarial overall survival curves according to Masaoka stage. The 10-year actuarial overall survival rates were 100% for patients with Stage I disease, 90% for patients with Stage II disease, and 48% for patients with Stage III disease. In the univariate analysis, clinical stage alone had a statistically significant impact on overall and disease free survival (p < 0.0001 for both). No significant differences in survival were seen with respect to other factors (Table 1).

Figure 1.

Actuarial overall survival rates according to Masaoka stage.

Table 1. Univariate Analysis of Various Potential Prognostic Factors for Survival in Patients with Completely Resected Thymoma Followed by Postoperative Radiotherapy
FactorP value
Overall survivalDisease free survival
  1. RT: radiation therapy; n.s: not significant (P ≥ 0.05).

Agen.s (0.43)n.s (0.76)
Gendern.s (0.75)n.s (0.73)
Pathologic classificationn.s (0.14)n.s (0.42)
Presence of mediastinal compressionn.s (0.45)n.s (0.57)
Myasthenia gravisn.s (0.52)n.s (0.62)
Masaoka stage< 0.0001< 0.0001
Total dose of RTn.s (0.13)n.s (0.10)
Treatment field of RTn.s (0.49)n.s (0.38)

Recurrence Patterns

Recurrence of thymoma was observed in 17 of 103 patients (17%). Table 2 shows the incidence and site of first recurrence according to Masaoka stage. None of the patients with Stage I disease experienced a recurrence, regardless of adherence of peritumoral fat. In patients with Stage II disease, 6 of 61 patients (10%) experienced a recurrence, and 11 of 25 patients (44%) with Stage III disease experienced a recurrence. The pleura was the most frequent site of first recurrence, and 11 of 12 patients who had pleural recurrences showed pleural dissemination remote from the primary tumor site. Eight patients had recurrences of ipsilateral pleural dissemination only at the first recurrence, and three patients had recurrences not only of ipsilateral pleural dissemination but also at other sites simultaneously at the first recurrence (mediastinum, pericardium, and supraclavicular fossa, respectively). Ten of 11 patients died of thymoma, and the remaining 1 patient has been alive with disease despite aggressive treatment.

Table 2. Incidence and Sites of Recurrence in Patients with Completely Resected Thymoma Followed by Mediastinal Irradiation According to Masaoka Stage
StageNo. of patientsNo. of recurrences (%)Site of first recurrence (%)a
MediastinumPleuraPericardiumSFDM
  • SF: supra-clavicular fossa; DM: distant metastases.

  • a

    In patients with two sites of first recurrence, both sites were scored.

I17000000
II616 (10)2 (3)4 (7)01 (2)0
III2511 (44)2 (8)8 (32)1 (4)2 (8)3 (12)
Total10317 (17)4 (4)12 (12)1 (1)3 (3)3 (3)

Intrathoracic Recurrence Patterns According to the Mode of Radiation Therapy

Intrathoracic recurrence patterns were analyzed with respect to the mode of radiation therapy (irradiated dose and irradiated field). Table 3 shows the incidence and sites of intrathoracic recurrence according to the irradiated dose. There were no in-field recurrences, and no dose-response relation was seen for intrathoracic control (incidence of intrathoracic recurrence: < 40 Gy group, 2 of 19 patients; 40 Gy group, 6 of 45 patients; > 40 Gy group, 7 of 39 patients). Table 4 presents the incidence and sites of intrathoracic recurrence according to the irradiated field. Mediastinal recurrences were observed in 4 of 36 patients (11%) who were treated with IF irradiation, whereas 0 of 51 patients who were treated with WM irradiation had mediastinal recurrences. Among four patients who had mediastinal recurrences, marginal recurrences were observed in two patients, and out-field recurrences were observed in the other two patients. Pleural-based recurrences were observed both in patients who were treated with IF irradiation (7 of 52 patients) and with WM irradiation (5 of 51 patients).

Table 3. Incidence of Intrathoracic Recurrence in Patients with Completely Resected Thymoma Followed by Mediastinal Irradiation According to the Total Dose
Total dose (Gy)No. of patients (Stage I patients)No. of IT recurrencesSite of first recurrence (%)a
In fieldMarginal and out field
MediastinumPleuraPericardium
  • Gy: grays; IT: intrathoracic.

  • a

    In patients with two sites of first recurrence, both sites were scored.

< 4039 (1)701 (3)7 (18)1 (3%)
4045 (9)602 (5)4 (9)0
> 4019 (7)201 (5)1 (5)0
Total103 (17)1504 (4)12 (12)1 (1)
Table 4. Incidence of First Recurrence in Patients with Completely Resected Thymoma Followed by Mediastinal Irradiation According to the Irradiated Field
Irradiated fieldNo. of patients (Stage I patients)No. of IT recurrencesSite of first recurrence (%)a
In fieldMarginal and out field
MediastinumPleuraPericardium
  • L: localized field; WM: whole mediastinum field; B: boost field; IT: intrathoracic.

  • a

    In patients with two sites of first recurrence, both sites were scored.

L52 (12)1004 (8)7 (13)1 (2)
WM with or without B51 (5)5005 (10)0
Total103 (17)1504 (4)12 (12)1 (1)

Incidence of Pleural-Based Recurrence According to the Degree of Pathologic Tumor Invasion

Table 5 indicates the incidence of pleural-based recurrence according to the degree of pathologic tumor invasion. None of 71 patients without pleural invasion had pleural-based recurrence (Stage I, 0 of 17 patients; Stage II, 0 of 51 patients; Stage III, 0 of 3 patients). Conversely, patients with pleural invasion had pleural-based recurrences (Stage II, 4 of 10 patients; Stage III, 8 of 22 patients). Figure 2 shows the pleural recurrence free rates according to the presence of pathologic pleural invasion. The 10-year actuarial pleural recurrence free rates for patients with and without pleural invasion were 63% and 100%, respectively. There was a statistically significant difference in pleural recurrence free rates between the two groups (P < 0.0001).

Table 5. Incidence of Pleural Recurrence in Patients with Completely Resected Thymoma Followed by Mediastinal Irradiation According to the Degree of Pathologic Tumor Invasion
Masaoka stageDegree of tumor invasionIncidence (%)
  1. NO: neighboring organs, (e.g., pericardium, great vessels, or lung).

INone0 of 17
IICapsule0 of 22
IIMediastinal fat0 of 29
IIPleura4 of 10 (40)
IIINO with pleural invasion8 of 22 (36)
IIINO without pleural invasion0 of 3
Total12 of 103 (12)
Figure 2.

Actuarial pleural recurrence free rates according to the presence of pathologic pleural invasion.

Treatment-Related Side Effects or Complications

Eight of 103 patients (8%) developed symptomatic radiation pneumonitis that required steroids (7 patients who received WM irradiation and 1 patient who received IF irradiation). The radiation doses to the primary tumor bed were 30.0 Gy, 32.0 Gy, 36.0 Gy, 40.0 Gy, 55.8 Gy, 56.0 Gy, 56.0 Gy, and 60.0 Gy, respectively. One patient (1%) developed pericarditis that required conservative treatment. One patient who was given radiation therapy at the age of 6 years is now age 19 years and is in good health without treatment-related complications. No patients died due to side effects or complications resulting from treatment.

DISCUSSION

There is a general consensus that surgery is the treatment of first choice for patients with thymoma, and the extent of surgery is a significant factor for local control and survival.14, 15, 21 Therefore, many authors have recommended complete resection for these patients whenever possible.5, 18, 22–24, 40 However, even in patients who undergo macroscopically complete resection, these tumors often recur in the thoracic region, such as the mediastinum or pleural cavity,7, 10, 13, 24, 36, 37 and a poor prognosis has been observed in patients who experienced disease recurrence.7, 18 The incidence of intrathoracic recurrence for patients with Stage II and III tumors who do not receive adjuvant therapy has been approximately 28–40%.7, 13, 24, 40 Therefore, there is general agreement that postoperative radiotherapy should be used in the treatment of patients with Stage II and Stage III tumors.7, 13, 16, 17, 33–35

Conversely, for patients with completely resected Stage I tumors, the role of postoperative radiotherapy has been less clear. For patients with Stage I thymoma, most studies have reported no recurrences or very few recurrences after patients undergo surgery without receiving any adjuvant therapies.7, 16, 32 Many authors have stated that there is no rationale for the use of postoperative radiotherapy for patients with Stage I tumors.7, 35 In contrast, Regnard et al. reported that, among 135 patients with Stage I disease, 5 of 26 patients (19%) with peritumoral adherence experienced recurrences, whereas 0 of 109 patients without peritumoral adherence experienced recurrences.21 The recurrence rate was significantly higher among patients with peritumoral adherence compared with patients who had no peritumoral adherence (P < 0.001). Pollack et al. observed 2 recurrences in 11 patients with Stage I tumors and advocated postoperative radiotherapy for patients who underwent complete resection for Stage I thymoma when the tumor is large and adherent to pleura or pericardium.18 In the current study, there were 10 patients with adherence of peritumoral fat who received radiation therapy, and none of them experienced a tumor recurrence. Cowen et al. also found no failures among irradiated patients with Stage I disease who had peritumoral adherence.22 Because there were no failures among our irradiated patients with Stage I disease who had peritumoral adherence, we also advocate the use of postoperative radiotherapy in Stage I patients with peritumoral adherences.

Although postoperative irradiation of invasive thymoma has been recommended commonly, the optimal radiation fields and doses remain uncertain.5, 19, 22, 24, 32, 34–37, 41 The dose response in patients who receive irradiation for thymoma has not been established clearly. With regard to the total dose, most authors recommend a total dose of 40–50 Gy in daily fractions of 1.8–2.0 Gy after patients undergo complete resection.18, 24, 36, 37, 41, 42 Jackson and Ball found that, in 14 patients who underwent macroscopic complete resection, there was no dose-response relation to local control within the range of 40–45 Gy.24 In our series, no in-field recurrence was observed, and no dose-response relation was seen for the intrathoracic control (incidence of intrathoracic recurrence: < 40 Gy group, 2 of 19 patients; 40 Gy group, 6 of 45 patients; > 40 Gy group, 7 of 39 patients). A total dose of 40 Gy appears to be appropriate in patients who undergo complete resection.

With regard to treatment field, mediastinal recurrences were observed in 4 of 52 patients (8%) who were treated with IF irradiation, whereas 0 of 51 patients who were treated with WM irradiation experienced a mediastinal recurrence. Among four patients who had mediastinal recurrences, marginal recurrences were observed in two patients, and out-field recurrences were observed in the other two patients. These results indicate that IF irradiation with a 1–2 cm margin carried some risk of mediastinal recurrence, whereas WM irradiation was effective in preventing mediastinal recurrences in patients with completely resected thymoma. WM irradiation also can cover the lower mediastinum, which may be at risk for gravitational metastases. Several authors also have stated that it is advantageous that the treatment field include the whole mediastinum.11, 34

Some authors have recommended the inclusion of supraclavicular fossa in the radiotherapy field.11, 43 Chahinian et al. found that 2 of 11 patients (18%) with invasive or metastatic thymoma experienced disease recurrence at the supraclavicular fossa.43 In contrast, other authors have questioned the need of radiation therapy to the supraclavicular fossa.24, 44 Jackson and Ball found that 2 of 28 patients (7%) with invasive thymoma recurred at the supraclavicular fossa, and in neither patient was it the sole site of recurrence.24 Our results indicate that only 3 of 101 patients (3%) without irradiation to the supraclavicular fossa experienced tumor recurrences. Prophylactic irradiation of the supraclavicular fossa does not appear to confer any therapeutic advantage to patients who undergo complete resection.

Pleural-based recurrences were observed both in patients who were treated with IF irradiation (7 of 52 patients) and in 5 of 51 patients who were treated with WM irradiation. Several authors also have indicated that pleural-based recurrences were found even in patients who received mediastinal irradiation.32, 37, 45, 46 These results indicate that mediastinal irradiation was effective in preventing mediastinal (local) recurrences, although it did not control pleural-based recurrences.

The most common pattern of failure in patients who undergo complete resection for invasive thymoma after they receive mediastinal irradiation reportedly is pleural dissemination.10, 32, 45, 46 In the current study, the pleura was also the most frequent site of first recurrence, and 11 of 12 patients who had pleural recurrences showed pleural dissemination remote from the primary tumor site. However, it may be difficult to predict the risk of pleural-based recurrence by Masaoka stage alone, because recurrences sometimes occur even in patients with Stage II tumors.32, 37 In the current study, analysis of pleural-based recurrences according to the degree of pathologic tumor invasion showed that, in patients without pleural invasion, no recurrences (0 of 71 patients) were seen after mediastinal irradiation (Stage I, 0 of 17 patients; Stage II, 0 of 51 patients; Stage III, 0 of 3 patients). Several reports also have indicated that, with the use of mediastinal irradiation, no recurrences were observed in patients without pleural invasion.32, 45 These results suggest that invasive thymomas without pleural invasion may be localized disease with a lesser risk of pleural dissemination and that mediastinal irradiation would be effective in preventing recurrence.

Conversely, among patients with pathologic pleural invasion of the tumor, 12 of 32 patients (38%) had pleural-based recurrences (4 of 10 patients with Stage II disease and 8 of 22 patients with Stage III disease). Haniuda et al. reported that, of 80 patients with thymoma who underwent complete resection, 12 of 13 patients (92%) developed a recurrence of pleural dissemination, and, among patients with Stage II and III pleural invasion, 6 of 15 recurrences (40%) occurred even with mediastinal irradiation (3 of 5 patients with Stage II disease and 3 of 10 patients with Stage III disease.32 These results suggest that some patients with pathologic pleural invasion of the tumor already had latent microscopic pleural dissemination at the time they underwent surgery, even though they were classified with Stage II disease. In such patients with pleural invasion, mediastinal irradiation may have only a limited effect. Not merely localized field irradiation but even whole mediastinal irradiation may be insufficient to prevent pleural-based recurrence, because it can cover only a part of the pleura.

To reduce the risk of pleural-based recurrence, some authors have advocated the use of prophylactic pleural and pulmonary irradiation.19, 21, 37 Uematsu et al. compared irradiation of the entire hemithorax plus the mediastinum (Group 1) with mediastinal irradiation only (Group 2) in patients with Masaoka Stage II and III disease.37 In both treatment groups, the median dose applied was 40 Gy. That study found local control rates of 60% in Group 2 (12 of 20 patients) and 96% in Group 1 (22 of 23 patients). The authors concluded that, except in elderly patients, entire hemithorax irradiation after patients undergo complete resection appears to be safe and feasible and can reduce intrathoracic recurrence rates. Because systemic chemotherapy reportedly is capable of producing durable remissions in patients with advanced or metastatic thymoma,46, 47 several authors have concluded that chemotherapy may deserve evaluation as adjuvant therapy for patients with invasive thymoma who undergo complete resection.14, 32, 36

Conversely, several authors have indicated that the incidence of pleural-based recurrence seemed too low to recommend routine extensive treatment of the pleura.7, 17, 19, 36 Nakahara et al. noted a 95% 15-year survival rate in 35 patients with Stage III thymoma who underwent complete tumor resection and who were given routine mediastinal irradiation.17 Curran et al. reported that, among 43 patients with Stage II and III tumors, the intrathoracic recurrence rate after they underwent complete resection and postoperative mediastinal irradiation was only 5%.7 Urgesi et al. indicated that, among patients with Stage III thymoma, only 5 of 59 patients (8.5%) experienced tumor recurrences outside the irradiated field.19 These results are conflicting and are likely a function of selection bias, a common problem when comparing retrospective data sets.35 To clarify the need of prophylactic treatment for patients with pleural-based recurrences, properly randomized trials are highly warranted. In this study, the pathologic pleural invasion of the tumor was the risk factor for pleural-based recurrence. We believe that additional treatment should be given to patients with pathologic pleural invasion of the tumor. A correlation of computed tomography scanning and magnetic resonance imaging with the surgical and pathologic findings also may provide additional data and may merit further investigation.48

In the current study, 8 of 103 patients (8%) developed symptomatic radiation pneumonitis that required steroids, and 1 patient (1%) developed pericarditis that required conservative treatment. However, no patient died due to side effects or complications resulting from treatment. Several reports also indicated that some patients developed radiation pneumonitis or pericarditis.15, 18, 36, 37 Cowen et al. indicated that 7 of 149 patients (5%) developed Grade 3 or 4 (World Health Organization grading) pericarditis (3 patients) or lung fibrosis (4 patients), although no patient died due to side effects or complications resulting from treatment.22 These results indicate that postoperative mediastinal irradiation may carry some risk of side effects; however, it can be administered without fatal complications.

From these views and findings, we concluded that WM irradiation with a total dose of 40 Gy was effective in preventing mediastinal recurrences in patients with completely resected thymoma, although IF irradiation carried some risk of mediastinal recurrence. Analysis of the mode of recurrence revealed that pathologic pleural invasion of the tumor was predictive of pleural-based recurrence for patients with completely resected thymoma. In patients without pleural invasion, the tumor may be localized disease with a lesser risk of pleural dissemination, and postoperative mediastinal irradiation would be sufficient to prevent recurrence. Conversely, in patients with pathologic pleural invasion of the tumor, there is some possibility of microscopic pleural dissemination at or before the time of surgery. Therefore, additional treatment, such as prophylactic pleural and pulmonary irradiation or chemotherapy, should be administered to prevent pleural-based recurrence.

Ancillary