Uncommon Thoracic Tumors

Etiology and Epidemiology

Thymomas are rare in the United States and data from the Surveillance, Epidemiology, and End Results (SEER) program estimate a thymoma incidence of 0.13 to 0.15 per 100,000 people. The typical age of onset of thymoma is between 40 and 60 years with a median age of 52 years and equal gender distribution. Thymomas are even more rare in children, accounting for approximately 15% of anterior mediastinal masses.

Thymic carcinomas are distinct from invasive thymomas, both pathologically and clinically. They account for 5% to 36% of all thymic neoplasms, the wide range in incidences reflecting differences and changes in the pathologic classification of this rare tumor. Patients with thymic carcinoma are typically middle-aged or elderly, and there is a slight male predominance. Thymic carcinoids represent less than 5% of anterior mediastinum lesions and typically affect middle-aged men.

A number of hypotheses surround the genesis of thymic tumors. There is a reported association between Epstein-Barr virus (EBV) infection and tumors of the thymus. As such, thymic diseases are more common in the far East, where the endemic EBV infection rates are high. Defective viral genomes have been isolated in patients with lymphoepithelioma-like thymic carcinoma lending further evidence of a role of viral infection in carcinogenesis. Childhood thymus irradiation has been linked to the development of thymic tumors, and familial cases have been reported, suggesting a possible relationship with cytogenetic anomalies. In an analysis of secondary or concomitant neoplasms in 1495 patients with acute lymphoblastic leukemia enrolled in two consecutive multicenter protocols, an increased risk of solid tumors was identified. Although the histologic characteristics of these second malignant tumors varied, thymoma was among the solid tumors identified in adult survivors of acute lymphoblastic leukemia. Among patients with primary thymic carcinoid tumors, up to 30% had multiple endocrine neoplasia type 1 or 2.

Chromosomal abnormalities and loss of heterozygosity may also play a key role in thymic neoplasia. A distinctive chromosome abnormality involving translocation of fragments of chromosomes 15 and 19 [t(15:19)(q15:p13)] have been identified in children and young adults with thymic carcinoma. Deletion of the short arm of chromosome 6 is also associated with benign thymomas, suggesting that putative tumor suppressor genes involving chromosome 6 may contribute to the pathogenesis.

In a small study of 37 cases, most World Health Organization (WHO) type A thymomas ( Table 52.1 ) did not show any chromosomal aberrations, whereas type B3 thymomas and thymic carcinoma (i.e., type C thymoma) shared some genetic aberrations, including the loss of chromosome 6 and the gain of chromosome 1q. The loss of chromosome 6, on which the human leukocyte antigen locus and some of the tumor suppressor genes have been identified, and the gain of chromosome 1q, to which growth promoter genes have been mapped, may play a role in tumorigenesis and pathogenesis of the paraneoplastic autoimmunity characteristics of thymoma.

Biologic Characteristics and Molecular Biology

Patients with thymoma may have dysregulation of the lymphocyte selection process, associated with abnormal proliferation, autoimmunity, and immunodeficiency. Thymoma-associated autoimmune disease involves an alteration in circulating T-cell subsets.

Factors influencing the different biologic behaviors of thymoma subtypes are poorly understood, although the increase in molecular profiling going forward may provide arguably better links to prognosis than stage alone. Altered TP53 expression may be implicated in the initial stages of tumorigenesis, and increased expression of epidermal growth factor (EGF) and epidermal growth factor receptor (EGFR) may play a role in thymomagenesis. A decreased overall survival (OS) time is predicted by Src tyrosine kinase and TP53 coexpression. It has also been proposed that a deficiency in the autoimmune regulator (AIRE) gene has an oncogenic effect on thymomas. AIRE expression is lacking in approximately 95% of thymomas; therefore, loss of AIRE expression may provide a potential mechanistic link for the well-recognized association of autoimmunity with thymoma along with aneuploidy and expression of muscle autoantigens.

Thymic carcinomas possess overt features of malignancy, similar to those of carcinoma arising in any other organ, with a higher propensity for capsular invasion and metastases than invasive thymomas. Paraneoplastic syndromes occasionally exist with well-differentiated lesions. Most variants of thymic carcinoma are highly lethal, producing frequent metastases to regional lymph nodes, bone, liver, and lung. Thymic carcinoma has been associated with increased expression of epithelial membrane antigen and cytokeratin subtypes. Markers for thymic carcinomas (as compared with thymomas) include somatic KIT gene mutations, CD70, CD5, and CD99, along with negative thyroid transcription factor-1 and a distinct cytokeratin profile. There may also be aberrant epigenetic regulation in these neoplasms, such as that mediated by the TET2 gene. Thymic carcinomas can be distinguished from lung carcinomas by a negative expression of TTF-1, and radiologic typing may provide further genomic/radiomic-based differentiation of thymomas.

Histopathology

In 1999, the WHO published a classification including six subtypes of thymic tumor, based on the relative proportion of epithelial and lymphocytic cells (see Table 52.1 ). This is the most widely accepted histologic classification system, and the WHO cell type is an independent prognostic factor. However, there is no direct correlation between the histopathology of thymomas and their malignant potential.

The histologic diagnosis of thymoma can be difficult, and a variety of systems have been proposed to do so. The WHO classification, developed in 1999 and revised in 2004, is the one most widely accepted. Tumors with true malignant cytologic characteristics are considered to be thymic carcinomas rather than thymomas. Invasive thymomas invade the capsule macroscopically or microscopically, with typically “bland” cytologic characteristics of thymic epithelial cells admixed with mature lymphocytes. The term invasive thymoma should be used instead of malignant thymoma to denote the tumor's predilection for capsular invasion.

Grossly, thymomas are nodular, multilobulated, and firm. They may contain cystic spaces, calcification, or hemorrhage and may be neatly encapsulated, adherent to surrounding structures, or invasive. Thymomas generally have epithelial and lymphatic cells. They are classified as predominantly epithelial, predominantly lymphocytic, mixed lymphoepithelial, or spindle cell type. Morphologically, thymoma cells are rather large and may be round, oval, or spindle shaped with vesicular nuclei and small nucleoli. The cytoplasm is often eosinophilic or amphophilic. Thymic neoplasms arise from epithelial cells ( Fig. 52.1 ). The lymphocytic component is mostly normal-appearing mature lymphocytes. Some of the other microscopic features that may be seen in thymomas include Hassall corpuscles, keratinizing squamous epithelium, rosettes, glands, cysts, papillary structures, and germinal centers. Immunohistochemistry is often helpful in making the diagnosis. Thymomas typically stain positive for thymic epithelial markers, including cytokeratin, thymosin β3 and α1, and epithelial membrane antigen.

Thymic carcinomas exhibit malignant cytologic features, commonly with squamous differentiation. Other subtypes include lymphoepithelioma-like carcinomas, clear cell carcinomas, sarcomatoid carcinomas, adenosquamous carcinomas, mucoepidermoid carcinomas, adenocarcinomas, and basaloid squamous cell carcinomas.

The histologic features of thymic carcinoids are identical to carcinoid tumors in other organs. Unlike thymomas, they are rarely encapsulated. Immunohistochemically, they may stain positively with CAM 5.2, low-molecular-weight cytokeratins, chromogranins, synaptophysin, and leucine-7.

Clinical Features

Most thymic tumors are discovered during myasthenia gravis work-up or incidentally on chest imaging. In fact, up to 70% of thymomas may be associated with paraneoplastic syndromes. Clinical symptoms vary greatly, depending on the size of the tumor and its effect on adjacent structures, but they are usually those of a mediastinal mass producing cough, chest pain, dyspnea, hoarseness, superior vena cava syndrome, and symptoms related to tumor hemorrhage. Patients may also have dysphagia, fever, weight loss, and anorexia.

Some thymomas present with symptomatic paraneoplastic syndromes; the most common is myasthenia gravis, which is seen in approximately 45% of patients. Box 52.1 lists other syndromes. Thymic carcinoids may also be associated with Cushing syndrome, Eaton-Lambert syndrome, syndrome of inappropriate secretion of antidiuretic hormone (SIADH), and hypercalcemia, but the classic carcinoid syndrome is rare. The causes of these syndromes remain obscure; autoantibodies have been demonstrated, albeit mostly in patients with thymoma.

Myasthenia gravis is an autoimmune neuromuscular junction disorder characterized by the presence of antiacetylcholine receptor antibodies, which cause an acetylcholine receptor deficiency at the motor endplate. The disease is characterized by rapid exhaustion of voluntary muscular contractions, with a slow return to the normal state. Myasthenia gravis is common with thymoma, but rare in thymic carcinoma. Death in patients with thymoma and myasthenia gravis is commonly caused by complications of myasthenia gravis, whereas in patients without myasthenia gravis, death often is attributed to local progression of tumor. Older series reported a poor prognosis associated with myasthenia gravis, but several modern series have failed to confirm this observation. Myasthenia gravis may even confer a survival advantage because neuromuscular symptoms may lead to earlier discovery of localized disease. After thymectomy, patients with myasthenia gravis have attenuation of symptomatic severity over time, but not necessarily complete resolution.

The predominant pattern of spread of thymomas is by direct invasion into adjacent organs. The degree of encapsulation and the invasion of adjacent tissues define prognosis for these tumors, rather than the histologic appearance. However, approximately 50% of cases in surgical series are noninvasive. Thymomas may metastasize as implants on pleural surfaces or pulmonary nodules, but rarely to extrathoracic areas. When thymomas disseminate, the most common site is the pleural cavity, where they form plaques, malignant pleural effusions, and diaphragmatic masses. Invasion into the superior vena cava, brachiocephalic vein, lung, and pericardium may also be observed.

Thymic carcinomas invade locally and often involve the pleura and mediastinal nodes. Up to 30% of thymic carcinomas and carcinoids are metastatic to regional lymph nodes and distant sites at diagnosis. Thymic carcinomas have also been noted to spread to nonregional nodal regions, such as the neck and axilla. Distant metastases to the lungs, liver, and bone can also occur in 30% to 40% of cases, and they may be seen in up to 70% of patients within 8 years of initial diagnosis. Brain metastases are altogether rare, but have been documented.

Prognosis

The two most important prognostic factors for thymoma are invasiveness (stage) and completeness of surgical resection. Capsular invasion is commonly used as the basis for designation as benign or malignant. Tumor size and the presence of symptoms could also have prognostic value. Patients with complete or radical excision have significantly improved survival over those with subtotal resection or biopsy only. Although almost all noninvasive thymomas can be totally resected, the ability to achieve a complete resection in invasive cases varies from 58% to 73%.

The 5-year and 10-year survival rates for well-encapsulated thymomas without invasion are more than 90%, and for invasive thymomas, the rates range from 30% to 70%. The 5-year survival rates according to Masaoka stage are 83% to 100% for stage I disease, 86% to 98% for stage II, 68% to 89% for stage III, and 50% to 71% for stage IV disease. The approximate 10-year survival rates are 80%, 78%, 47%, and 30%; the respective 15-year rates are 78%, 73%, 30%, and 8% for stages I to IV. Table 52.2 provides a summary of treatment results of the aforementioned thymoma studies, as well as additional investigations.

A retrospective study of 324 patients found that patients with WHO types A, AB, and B1 had a 100% disease-specific survival rate without RT and therefore did not benefit from adjuvant RT. There was no survival difference for cell types B2 and B3 with or without adjuvant RT. Rieker et al. noted that survival rates of patients with types A, AB, B1, and B2 were better than for type B3, and that type C had the lowest OS. The prognostic value of the subtype of type B thymomas has been examined in one series that found no differences in recurrence or survival rates among the three subtypes of type B, but all patients who experienced recurrence had stage III disease, indicating an association with Masaoka stage.

Although the degree of tumor invasiveness is strongly related to stage and prognosis, no data support the prognostic significance of the histologic findings, independent of tumor stage. The historical classification proposed by Marino and Muller-Hermelink categorized thymoma as cortical, mixed, and medullary types. Thymomas arising from the epithelial cells of the cortex were classified as cortical thymomas, and those arising from the medullary spindle cells were classified as medullary thymomas. Of note, thymic carcinomas were categorized as a separate entity.

As with thymomas, total resection and stage at presentation are important prognostic factors for thymic carcinoma. Extensive lymph node dissection (> 10 nodes) is needed to accurately stage patients, and disease-free survival (DFS) can be as high as 90% in patients with N0 with extensive nodal dissection, whereas patients with node-positive disease can have DFS on the order of 33%.

A nine-gene assay has been developed that dichotomizes thymomas into high- and low-risk of metastasis, with 10-year metastasis-free survival of 77% and 26%, respectively. This product is available commercially.

Diagnostic Work-Up for Thymic Tumors

The history and physical examination for a suspected thymoma should focus on signs and symptoms suggestive of myasthenia gravis, such as fatigue, diplopia, ptosis, and dysarthria. Constitutional symptoms such as fever, chills, and weight loss may suggest a mediastinal lymphoma. Routine screening blood work and chemistry testing may give clues to the presence of associated syndromes. In the case of suspected Cushing syndrome, a dexamethasone suppression test and urinary cortisol level should be obtained, although some carcinoids can be suppressed, which can make diagnosis difficult. The differential diagnosis, in addition to thymic lesions, includes germ cell tumors, lymphomas, and thyroid proliferative disorders. Serum alpha-fetoprotein and beta-human chorionic gonadotropin levels should be obtained in young men if a nonseminomatous germ cell tumor is suspected. Exclusion of metastases from extrathymic primary tumors is additionally important.

The chest radiograph ( Fig. 52.2 ) and CT scan with contrast defines the characteristics of a mediastinal lesion, as well as its relation to or invasion of other mediastinal structures. Thymic tumors are usually homogeneous, well-demarcated lesions with a round or lobulated shape, and occasionally have calcifications ( Fig. 52.3 ). The presence of fat planes between the tumor and adjacent structures suggests localized disease. Pleural involvement may be seen in advanced disease. The radiographic presence of both an anterior compartment mass as well as “drop” metastases to the pleura is highly suggestive of the diagnosis. Thymic carcinomas often contain calcifications, cysts, or necrosis on imaging. A nomogram has been developed by Marom et al. using CT characteristics to predict the Masaoka stage ( Fig. 52.4 ), which can also be utilized to predict tumoral response to neoadjuvant therapy. In the absence of symptoms and signs, extensive radiographic imaging is unnecessary. Magnetic resonance imaging (MRI) has not been shown superior to CT scanning.

The role of positron-emission tomography (PET) has not been well established, as FDG (fluorodeoxyglucose) uptake in thymomas is variable. However, PET/CT can help distinguish thymomas from thymic carcinomas, although conflicting data exist regarding the capacity to predict WHO grade. A report of 51 patients found that FDG uptake was higher in patients with thymic carcinoma than thymoma, and type B3 thymomas had higher uptake than lower histologies (A through B2). Similar results were seen in 47 patients from Italy, where maximum standardized uptake value (SUVmax) and ratio of SUVmax to tumor size correlated with WHO grade and Masaoka stage. However, a series of 58 patients from Japan found that, although SUVmax can differentiate between thymic carcinoma and thymoma, there was no difference between the low- and high-risk thymoma groups. High uptake of FDG also appears to correlate with the degree of tumor invasiveness, and may be helpful in identifying nodal and distant metastasis.

Depending on the particular institution, patients presenting with a thymic mass may need histologic diagnosis before definitive therapy. CT- or ultrasound-guided fine-needle aspiration biopsy can establish the diagnosis preoperatively with a sensitivity and specificity of 87% to 90% and 88% to 100%, respectively. When larger tumor samples are required to distinguish between lymphoma and lymphoid-predominant thymoma, core-needle biopsy provides sufficient specimens with an overall sensitivity of 96% and specificity of 100%. Bronchoscopy, video-assisted thoracoscopic surgery, mediastinoscopy, or anterior thoracoscopy may help yield the diagnosis before resection, especially if enlarged lymph nodes are present. The potential risk of capsule rupture, leading to spillage and seeding of tumor cells during biopsy, has been debated and remains unsettled.

Staging

Bergh et al. introduced the first clinical staging system for thymoma in 1978. Their staging system was subsequently modified by Masaoka et al. in 1981 and is the most widely accepted ( Table 52.3 ). It is largely based on pathologic findings at time of surgery.

A separate, simplified staging paradigm was proposed by Suster and Moran. Stage I lesions are localized and encapsulated; stage II lesions are locally invasive; and stage III lesions have nodal, visceral, or distant metastasis. Although a tumor-node-metastasis (TNM) staging system was proposed by Tsuchiya et al., there were no TNM classifications for thymic neoplasms until the AJCC 8 ^th edition ( Table 52.4 ). This recent framework may better help to predict and stratify outcomes.

Treatment

Radiation Therapy for Thymoma

Although the role of adjuvant irradiation for invasive thymomas has never been tested in a randomized fashion, RT is an effective adjuvant therapy for invasive thymomas, and most retrospective studies have reported improvements in local tumor control and survival rates after adjuvant irradiation, along with a corresponding shift in patterns of failure. Patients with completely resected stage I thymoma should not routinely receive postoperative RT as the recurrence rate is approximately 0 to 2%.

Controversy exists regarding whether adjuvant RT should be used for completely resected stage II thymomas. The rationale for offering adjuvant RT is that local recurrence rates may approach 30% in certain series and that RT can spare patients from repeat thoracotomy to salvage recurrence. Kondo and Monden reported a multi-institutional retrospective study of 1320 patients with stage II or III thymic epithelial tumors and found no significant difference in recurrence rates with surgery alone versus adjuvant RT. This observation may reflect the fact that most patients underwent complete resection (100% with stage II disease and 85% with stage III disease); as has been pointed out by the study authors, the recurrence rates they found were lower than in previous reports. On the other hand, a recent analysis from the International Thymic Malignancies Interest Group (n = 1263) showed an increase in both 5- and 10-year overall survival with the addition of adjuvant radiotherapy for both stage II and III disease.

Additional “meta-analyses” and database reviews have provided conflicting results as to the value of adjuvant radiotherapy for stage II thymoma. Two reviews limited to completely resected stage II thymoma found no reduction in recurrence after adjuvant RT, whereas other studies (albeit with mixed stages included) suggest RT may improve survival. A SEER study of 901 patients showed a 10% absolute improvement in 5-year OS rates, but no statistical difference in cause-specific survival rates in patients with stages II to III disease. Two studies from national Japanese databases demonstrated no differences in outcomes with stages II and/or III disease, yet two investigations from the United States National Cancer Database demonstrated increased OS with adjuvant RT for stage II/III cases.

Although the impact of adjuvant radiotherapy on late toxicity has not been extensively studied, a SEER database analysis of 1334 patients treated between 1973 and 2005 did not find an increase in long-term cardiac mortality rates or rates of secondary malignant tumors for patients treated with RT compared with patients treated with surgery alone.

Patients with gross fibrous adhesions of the tumor to the pleura at the time of surgery may be at particularly high risk for local failure following surgery alone. Haniuda et found that patients with fibrous adhesions to the mediastinal pleura without microscopic invasion benefited most from postoperative RT. Although postoperative RT may decrease local recurrence, as expected, it does not decrease the incidence of subsequent pleural dissemination that may occur outside of the radiation field, which is the most common site of failure after radiation.

Locally Advanced Thymoma

For patients with stage III to IV disease, there is greater consensus regarding the use of adjuvant RT. Urgesi et al. reported no in-field recurrences in a study of 33 patients with completely resected stage III thymoma treated with postoperative irradiation. Curran et al. reported a 53% 5-year actuarial mediastinal relapse rate in patients with stages II/III disease after surgery alone, compared with 0 after total resection and irradiation and 21% after subtotal resection or biopsy and irradiation. Similarly, in a study of 70 patients with Masaoka stages III to IV thymoma, the relapse rate for patients receiving postoperative radiation therapy was reduced from 50% to 20%, and most disease (80%) recurred outside of the irradiated field.

Neoadjuvant RT has been advocated for nonresectable or marginally resectable thymomas. Several small studies assessing preoperative RT for extensive disease found a decrease in tumor burden at the time of surgery, with response rates as high as 80%, and described a theoretical decrease in the potential for tumor seeding during surgery. Onuki et al. found that a dose of 12 Gy to 20 Gy given to 21 patients with stage III thymomas resulted in a 76% response rate, with more aggressive WHO histologic subtypes having a less robust volume reduction. Preoperative chemoradiation has also been tested in locally advanced thymic tumors with favorable results. Korst et al. found that in a group of 22 patients with 71% Masaoka stage III/IV disease with aggressive histology (62% B3 thymoma or thymic carcinoma), two cycles of cisplatin and etoposide concurrent with 45 Gy led to a 77% R0 resection rate and 14% R1 resection rate. These series demonstrated that preoperative irradiation facilitated total or subtotal resection of the invasive thymoma mass by reducing the tumor volume.

Definitive RT has been used in nonsurgical candidates or patients with nonresectable advanced disease. Overall, definitive RT achieves 65% local control and 5-year survival rates of 40% to 50%. In a report by Marks et al., tumor was controlled in all nine cases treated with RT with a median follow-up of 5.5 years. Another report observed tumor control in 8 of 11 patients with malignant thymoma with a minimal follow-up of 2 years. Five-year survival rates of 53% to 87% and 10-year survival rates of 44% have been reported for these patients. Urgesi et al. reported the use of radiation therapy alone in 21 patients with intrathoracic recurrences of thymoma. The 7-year survival rate of 70% was similar between those treated with definitive RT and surgery with adjuvant therapy. However, the retrospective nature of these studies, small number of patients, differing amounts of clinical disease, and variations in radiation doses and techniques are significant confounding variables.

Chemotherapy for Thymoma

In general, thymomas are chemosensitive tumors, but chemotherapy is mostly reserved for locally advanced or metastatic disease. No randomized trials have compared different chemotherapeutic agents. Anthracycline- and/or platinum-based regimens have commonly been used, with response rates of more than 50% with the application of cisplatin-based combination chemotherapy. Commonly employed combinations include cisplatin, doxorubicin, and cyclophosphamide (PAC), with reported overall responses in excess of 70%, along with cisplatin, doxorubicin, vincristine, and cyclophosphamide (ADOC) with response rates approximating 90% in one study with a 43% complete response rate.

Similar results with combination PAC have been reported in Taiwan; all responders had overexpression of the topoisomerase 2α gene, but patients with no detectable expression progressed with treatment. Trials of cis/carboplatin with amirubicin or paclitaxel have also been reported, with noted efficacy in thymic carcinomas.

Targeted Agents for Thymoma

Because thymic tumors harbor a number of molecular aberrations, this has encouraged the exploration of targeted biologic agents. Although EGFR overexpression is common in thymic tumors, complete and durable responses to EGFR-directed therapies are uncommon. Although patients with thymic carcinomas also have higher serum vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) than controls, low response rates have been observed with bevacizumab and erlotinib. Similarly, KIT overexpression led to testing imatinib, which showed no substantial activity. A Phase II trial of sunitinib in thymic neoplasms that progressed on prior chemotherapy displayed disease control in 91% of patients with thymic carcinoma and 81% of those with thymoma. Inhibitors of the mammalian target of rapamycin (mTOR), such as everolimus, have been shown in a Phase II study to have a disease control rate of 78% in thymic carcinomas and 94% in thymomas. Epigenetic modulation has also shown activity in thymic neoplasia in a phase I/II study of the histone deacetylase inhibitor belinostat and the anti-IGF-1 antibody cixutumumab for recurrent/refractory disease have also shown antineoplastic activity.

Combined Modality Therapy for Thymoma

The role of chemotherapy after resection is controversial and has not been well established, but some evidence indicates that it can be effective with acceptable toxicity in selected cases. Typically the same drugs used for nonresectable or advanced disease are employed. There is also a suggestion that neoadjuvant chemotherapy is feasible and may improve the resectability of locally advanced thymomas.

A Phase II study of 22 patients with nonresectable thymoma treated with induction PAC resection, adjuvant irradiation, and consolidative chemotherapy found a 77% response rate, a 95% 5-year OS rate, and a 79% 7-year OS rate. A trial from the Japanese Clinical Oncology Group reported on 23 patients with nonresectable stage III thymoma treated with weekly dose-dense cisplatin, vincristine, doxorubicin, and etoposide and found no deaths from toxicity, but only 57% were able to complete chemotherapy as planned. Thirteen of the patients were able to undergo resection, with nine R0 resections. With a 62% response rate, efficacy of the dose-dense regimen was no better than with conventional chemotherapy. Additional small studies suggest neoadjuvant chemotherapy may improve resectability of locally advanced invasive thymoma. Chemotherapy followed by radiotherapy has also been studied without surgery in locally advanced disease with a median survival of 93 months in one trial using PAC chemotherapy.

Treatment of Thymic Carcinoma

Although the optimal treatment of thymic carcinoma is unclear, surgical extirpation remains the cornerstone of therapy and, in most published studies, surgery has been followed by adjuvant RT. Because many series are gathered over decades, it has been difficult or impossible to control for the changes in pretreatment tumor imaging, surgical techniques, and RT planning/delivery, all of which may contribute to the inconsistent results in the literature. A prescriptive dose range has yet to be identified, with most studies using 40 Gy to 70 Gy at 1.8 Gy to 2 Gy per fraction; in general, doses of 45 Gy to 55 Gy are used in the adjuvant setting. It is generally recommended not to exceed definitive doses of 60 Gy.

In a series of 26 patients treated with surgery and adjuvant irradiation, Hsu et al. observed a 77% 5-year OS rate, with respective 82% and 66% survival rates for completely resected and subtotally resected cohorts. With a median dose of 60 Gy (range, 40 Gy to70 Gy), an excellent 5-year local control rate of 91% was observed. For a cohort of 40 patients receiving definitive or adjuvant RT, Ogawa et al. reported an absence of local recurrence for those with complete resection and radiation doses greater than 50 Gy. Kondo and Monden reported the largest retrospective comparative study and found no statistically significant survival benefit from the addition of adjuvant RT to surgical resection, although no definitive conclusions could be made owing to subgroup sample size limitations and the retrospective nature of the study.

Although local control rates are increased with irradiation, a survival benefit remains to be demonstrated. For patients with any question of clinical resectability, neoadjuvant platinum-based chemotherapy is a reasonable treatment consideration. For patients with early stage cancer with completed resected disease, there is some thought that adjuvant radiation is not indicated. Table 52.5 summarizes some of the treatment results of thymic carcinoma, including publications not discussed in the text.

Treatment of Thymic Carcinoid

Complete surgical resection is the preferred method of treatment, although recurrence and distant metastases are common. A SEER analysis of 160 patients found that the median survival time was 79 months in patients who had surgical resection compared with 26 months in those who did not have surgery. Patients who received RT had a worse OS rate, but these patients were also more likely to have advanced disease. In a study of 40 patients, the recurrence rate was 64%, even though 35 of the 40 patients had a complete resection, and the 5-year survival rate was 84%. Despite a lack of conclusive evidence, incomplete resections followed by irradiation or chemotherapy (or both) may improve local control rates. However, distant metastases occur in approximately 30% of patients. The long-term prognosis is poor, with an overall 5-year survival rate of less than 30%.

Techniques of Irradiation

For optimal radiation treatment planning, a 4D CT scan with intravenous contrast should be performed to account for respiratory motion. Clips placed at the time of surgery denoting the extent of resection in completely resected tumors or outlining regions of residual disease and preoperative imaging are crucial to delineate the postoperative radiation volume. Preoperative imaging should be fused with the 4D CT scan used for radiation planning to aid with contouring the preoperative gross tumor volume (GTV). An internal clinical target volume (iCTV) should be generated encompassing the tumor bed and areas of suspected subclinical disease based on the maximal intensity projection and typically, the iCTV will extend 1 to 2 cm from the preoperative GTV. Prophylactic nodal irradiation of regions, such as the uninvolved mediastinal and supraclavicular nodal, is not warranted. The margin for the planning target volume (PTV) for setup error depends on institution, but generally is a 0.5 to 1 cm expansion from the iCTV when using daily image guidance.

Historically, 2D- and 3D-conformal radiation therapy has been used for the treatment of thymic neoplasms. Conventional port arrangements using a number of field arrangements such as two opposed anteroposterior (AP/PA) ports (weighted 2 : 1 or 3 : 2), wedged-pairs, and AP/PA with a posterior off-cord oblique have been used. However, 2D and 3D techniques expose large amounts of uninvolved normal tissues, such as the heart and lungs, to high doses of radiation. As many patients with thymic malignancies are expected to be cured, every effort should be made to employ advances in technology to achieve conformity to reduce the likelihood of late radiation complications.

Intensity-modulated radiation (IMRT) with motion management should be considered to improve conformity and allow better sparing of normal critical structures. Fig. 52.5 shows an IMRT plan, compared with other modalities and techniques in Fig. 52.6 .

Adjuvant doses of 45 Gy to 55 Gy given by conventional fractionation have been used effectively in most cases. For patients with microscopic or gross residual disease, definitive doses of 60 Gy are likely required to achieve tumor control. With the increasing availability of proton radiation therapy, there is a potential for an improved therapeutic ratio in radiation therapy for thymomas by taking advantage of the proton dose distribution. Multiple studies have displayed the dosimetric superiority of proton therapy over intensity-modulated radiotherapy in decreasing doses to the lungs and heart, which may be correlated with cardiac events and risk of secondary malignancy.

Respecting normal tissue tolerances is paramount to minimizing risk of acute and late radiation toxicity. The maximal dose to the spinal cord should be less than 45 Gy. Complex beam arrangements and arc therapy should be encouraged to maximize degrees of freedom to reduce high and intermediate dose delivered to the heart, lungs, and esophagus. The lung V20 should be kept below 30% to 35%, the heart V40 less than 30%, and maximum esophageal dose less than 60 Gy when using IMRT.