The role of radiation therapy in the management of primary thymic epithelial neoplasms

INTRODUCTION

Therapeutic radiation plays an important role in the management of thymoma and thymic carcinoma. These two tumor types differ substantially in their aggressiveness and prognosis. Despite this, much of the existing literature investigating the use of radiation therapy in the management of these rare tumors considers them together. To the extent that the literature separates the role of radiation in these related disease entities, it will be noted. The most pressing issues in radiotherapy are which thymoma and thymic carcinoma patients need radiation and the best technique to use to deliver radiotherapy.

Since thymoma and thymic carcinoma are rare diseases, there are few randomized trials completed to date assessing the value of any treatment modalities. Notably, however, a collaborative of French hospitals has launched RADIORYTHYMIC, a phase III randomized study investigating the role of post-operative radiation therapy in Masaoka-Koga stage IIB and III thymoma^[1].

Until that data is available, controversy remains regarding which patients can benefit from adjuvant radiation therapy^[2]. Currently, the existing retrospective data and population-based analyses divide radiation therapy into a dichotomous variable (yes/no). The patients included in the analyses span nearly 50 years, during which time radiation therapy has evolved from rudimentary 2-dimensional based planning to conformal 3-dimensional planning to yet more conformal dose painting techniques such as intensity-modulated radiation therapy (IMRT) and proton therapy. If the effect of radiation is small and the natural history of a disease long, as is the case for stage I favorable histology thymoma, then differences in techniques and toxicity may have as much of an impact as whether radiation was given or not.

For thymoma and thymic carcinoma, the target volume is either the post-operative tumor bed and high-risk surgical areas or the intact tumor itself. Given that thymomas and thymic carcinomas arise in the anterior mediastinum, the radiotherapy volume invariably includes critical organs at risk, such as the heart, great vessels, and lungs. In the era that was dominated by 2-dimensional radiation therapy, the fields also encompassed greater proportions of the spinal cord, trachea, and esophagus. Radiation doses to those critical organs can vary substantially depending on the technique used and the era in which radiation therapy was delivered.

LOCALIZED AND RESECTABLE THYMOMA AND THYMIC CARCINOMA

For all localized thymomas and thymic carcinomas, the standard of care is surgery. The ideal surgery is an extirpative surgery removing the entire thymus and tumor en bloc rather than removal of the tumor alone (thymomaectomy). Maximal debulking should be performed, and the high-risk areas should be marked with clips by the surgeon to serve as anatomical landmarks for the planning of adjuvant radiotherapy [Figure 1]. Ideally, negative macroscopic and microscopic margins are achieved (R0 resection). Given the rarity of thymomas, multidisciplinary coordination of care is essential, especially for patients with more advanced-stage diseases^[3].

Figure 1. Depiction of the initial imaging: (A) Coronal CT image; (B) axial CT image; (C) axial MR image; (D) sagittal MR image; (E) PET-CT coronal image; (F) axial PET-CT image; (G) dose Volume Histogram and structure names for the 3D conformal RT plan; (H) dose distribution for the 3D plan axial image at the level of the pulmonary artery. One surgical clip is noted in this image; (I) dose distribution axial slice at the level of the atrium of the heart. (H and I) plan for a 75-year-old man with a stage III thymoma treated with PORT. He had an R0 resection (negative margins), then received 50 Gy in 25 fractions and is NED 2 years later.

If a patient is initially unresectable, options for neoadjuvant chemotherapy or chemoradiation should be explored^[4], although neoadjuvant radiation therapy is less frequently used. Surgery is performed if a patient becomes operable after induction chemotherapy; otherwise, treatment is definitive radiation therapy with or without chemotherapy. For unresectable thymic carcinomas, sequential chemotherapy followed by radiation or concurrent chemoradiation therapy are standard treatment approaches [Figure 2]^[5-8]. Medically inoperable patients are treated with a combination of radiation therapy and chemotherapy depending on their stage and medical comorbidities^[9].

Figure 2. Patient who presented at age 74 with an unresectable thymic squamous cell carcinoma with disease in his left supraclavicular lymph nodes, his right level 2 and 4 lymph nodes and in the thymus [(A) PET-CT at diagnosis]. He was treated with definitive chemoradiation to a dose of 66 Gy in 33 fractions [(B) dose-volume histogram with structure set for D plan; (C) axial image of dose distribution at the level of the supraclavicular lymph nodes; (D) axial image of dose distribution at the level of the sternomanubrial joint; (E) axial image of dose distribution at the level of the pulmonary artery; (F) sagittal image of dose distribution] with weekly carboplatin and Taxol, followed by 2 additional cycles of carboplatin and Taxol.

In the curative setting, there are three main scenarios for the use of radiation therapy: post-operative radiation therapy (PORT), definitive radiation therapy, and neoadjuvant radiation therapy. In the non-curative setting, radiation therapy can be used for palliation of symptoms or to treat oligoprogressive disease^[10]. By far, the most common scenario is post-operative radiation therapy. Definitive radiation therapy is reserved for medically inoperable patients and those patients who are unresectable despite neoadjuvant chemotherapy. A small minority of initially unresectable patients are treated with both chemotherapy and radiation therapy either sequentially or concurrently.

POST-OPERATIVE RADIATION THERAPY

Given the long natural history of thymomas with favorable 10-year survival rates, the most important question remains which patients with thymoma benefit from post-operative radiation therapy. Such a definitive answer remains unanswered and unanswerable with current data where known prognostic factors such as WHO histologic subtype and margin status (RO vs. R1 vs. R2) are not always accounted for in the larger databases that span generations, and there are inherent limitations of large databases and retrospective analysis in a rare disease, as well as inherent biases in who is referred to receive adjuvant radiotherapy in nonrandomized reports. Only one small underpowered randomized trial of 29 total patients with the very early-stage disease has been reported in the literature^[11].

One retrospective analysis stands out from the rest, as the database and publication were developed by the International Thymic Malignancy Interest Group (ITMIG). ITMIG strives to coordinate the world’s efforts in the investigation and treatment of thymic epithelial tumors. They have published standardized radiation therapy definitions and reporting guidelines as well as standardized outcome measures^[12,13]. The use of these resources in the creation of databases and the collection of data could address many of the limitations in the currently available datasets. The analysis assessed patients with Masaoka stage II and III thymomas who underwent a complete R0 resection. Sixty-nine percent of patients had stage II disease, and seventy percent of tumors were WHO B1, B2, or B3. Fifty-five percent of patients received PORT. PORT was associated with improved 5- and 10- year overall survival (OS) whether the group was analyzed together or separated by stage (OS benefit P = 0.021 for stage II and P = 0.0005 for stage III), with 5-year OS 95% with PORT vs. 90% without and 10-year OS 86% with PORT vs. 79% without, P = 0.002^[14].

In the last decade, there have been at least six analyses of the Surveillance, Epidemiology, and End Results (SEER) database^[15-20]. SEER databases are limited in their ability to definitively answer the question of PORT in thymomas for many reasons. Most important among them are the fact that the stage is not clearly captured in the SEER database and must be inferred from other variables and the reason that PORT was chosen to be or not be administered is also not captured. Table 1 describes results from SEER analyses completed since 2010. These analyses provide conflicting data regarding which patients with thymoma and thymic carcinoma benefit from PORT.

Similar analyses have been performed using the American College of Surgeons National Cancer Data Base, shown in Table 2^[21-23]. These databases have the advantage of capturing the first course of treatment, including chemotherapy, surgery, radiation, and more detailed pathologic characteristics including margin status. These studies show a benefit to PORT in various populations, but like the SEER analyses, they provide conflicting evidence on the benefits of PORT.

For completeness, Table 3 details the conclusions of systematic reviews and meta-analyses performed in the last 5 years^[24-28]. These are of more limited help in answering which patients can benefit most from PORT given the heterogeneity of the data included, different populations, and conflicting results. One systematic review used the data collected to inform the formal expert consensus using a modified Delphi approach and is not included in the table^[9]. Several larger multi-institutional databases have been compiled to determine the benefit of PORT [Table 4]^[14,29,30]. The patient populations in each study varied, and taken together, there is no clear consensus on the benefit of PORT. Similarly, several single institutional trials have demonstrated a benefit to PORT in sub-populations of thymoma patients [Table 5]^[31-35].

Taken together, several general themes emerge. Patients benefit from PORT with positive margins, more advanced stage, and more aggressive histologic subtypes, especially when multiple adverse features are present. These findings are seen in most, but not all, of the series.

NOMOGRAMS TO PREDICT RECURRENCE RISK AND BENEFIT OF POST-OPERATIVE RADIATION THERAPY

Given the heterogeneity of the existing data, two nomograms have been developed to aid decision making for the use of PORT. Neither is particularly easy to use, as both require the use of a specific chart with varying points assigned to different prognostic factors^[36,37]. One from Korea incorporates age, sex, T stage, N stage, M stage, and histologic subtype. The second one, developed for thymic carcinomas, incorporates serum lactose dehydrogenase (LDH), dichotomous variables of complete resection and great vessel invasion, T stage (T1/2 vs. T3/4), Masaoka stage (I/II vs. III/IV), and histologic grade (Low/intermediate vs. high).

RECURRENCE PATTERNS

Recurrences after initial treatment have been described. In a study of 53 patients with Masaoka stage II-IV thymic carcinoma and thymic neuroendocrine carcinoma patients, 25 recurrences were noted^[38]. Forty-four of these had initial R0 resections. The vast majority were in the pleura, out of the radiation field, followed by the lung parenchyma and lymph nodes. Radiation therapy for isolated pleural metastases has been described with limited toxicity and excellent local control and survival benefits. Higher radiation doses (50-52 Gy at 2 Gy per fraction) were correlated with better results^[39,40]. Lower doses tended to be used with larger recurrences to respect normal tissue constraints. More recently, a report using stereotactic body radiation therapy (SBRT) to treat thymoma has been reported and was associated with excellent rates of local control [Figure 3]^[41]. Extrapolating from other diseases that originate or metastasize to the pleural, SBRT could be a promising emerging modality for managing pleural disease in thymic patients with oligoprogression^[42].

Figure 3. The patient in Figure 2 recurred with an isolated deposit anterior to the heart 1.5. years after initial therapy that was just outside of the previously irradiated field. This recurrence is depicted on PET/CT in (A) sagittal and (B) axial. He was treated with 50 Gy in 5 fractions using a stereotactic body radiation therapy approach in (C) dose-volume histogram; (D) dose distribution in the axial plane; and (E) dose distribution in the sagittal plane, and he remains disease-free 5 years after his initial presentation and 3.5 years after treatment of his recurrence.

GUIDELINES

Given the contradictory evidence presented above, several large medical societies have developed guidelines for the treatment of resectable thymoma [Table 6] and thymic carcinoma [Table 7] by stage^[5-9]. Given the relatively recent introduction of the TNM staging, both Masaoka and AJCC staging are included in the tables^[43].

RADIATION THERAPY SIMULATION AND PLANNING

Once it has been determined that a patient needs radiation therapy, a radiation simulation or planning session should be performed. Radiation should be started within 3 months of completion of surgery, although it can be later if a patient is receiving adjuvant chemotherapy. Clearance from the thoracic surgeon before starting adjuvant radiation therapy, especially if a patient has had a median sternotomy, is essential.

If possible, patients should be immobilized in the supine position with their arms above their head. This position allows for multiple beam angles or arc radiotherapy with the intent of achieving a highly conformal radiation treatment plan. Once the position is established, a CT scan should be performed. If needed, intravenous contrast can be used to accentuate vascular anatomy and identify the area of residual disease or better define gross disease in patients who undergo an R2 rection or are inoperable, respectively. Tumor motion or tumor bed (depending on treatment indication) should be assessed with a 4-dimensional CT scan. If that technology is not available or if a patient’s breathing pattern is not reproducible, a slow helical CT or CT scan performed at end inspiration and end-expiration can determine target motion.

Once the CT simulation is complete, adjunct images may be fused to the treatment planning CT, such as 18-FDG PET-CT, diagnostic CT, and/or MR images. This is true whether patients are being simulated in the pre-operative, definitive, post-operative, or palliative settings. If a patient is treated in the pre-operative, definitive, or palliative setting, the gross disease should be contoured and labeled as the gross tumor volume (GTV). The clinical target volume (CTV) represents an expansion of the GTV to include microscopic extension of disease. In the post-operative setting, the CTV includes the pre-operative tumor volume, the tumor bed (including resected involved lymph node areas and pleural deposits if relevant), and any surgical clips. The internal target volume (ITV) encompasses the CTV and adds a margin derived from the motion analysis from the 4-D scan (or other methods). The planning target volume (PTV) adds a margin to account for set-up uncertainty. No benefit has been derived from elective nodal irradiation^[44], and its use is not recommended.

Margins are added from one tumor volume to another depending on whether motion assessment has been performed and whether daily imaging will be performed. Typical margins range from 0.3 to 1.5 cm depending on the technology used and the volume being expanded.

Conformal planning techniques are considered standard and include 3-D conformal radiation therapy (3DCRT) and intensity-modulated radiation therapy (IMRT). IMRT is more conformal and more complicated than 3DCRT and requires rigorous quality assurance mechanisms for safe delivery. There is the potential for advanced technologies, such as proton therapy, especially when delivered with pencil beam scanning, to further improve the therapeutic ratio and reduce the risks of late radiation-induced cardiac events and secondary malignancies [Figure 4]^[45-51]. Careful motion management and daily image guidance should be employed when delivering advanced modalities like IMRT and proton therapy^[52].

Figure 4. Advanced Radiotherapy Modality Comparison. IMRT and proton therapy plans for a 32-year-old with stage IIA thymoma, WHO type B2 s/p thymectomy showing an 11.2 cm tumor with close margins (< 1 mm) and invasion into the mediastinal fat treated with adjuvant radiation therapy to 50.4/1.8 Gy. Radiation dose distribution and beam arrangements for proton (left) and VMAT IMRT (right) comparison plans depicting dose above 20% of the prescription (A) and 50% of the prescription (B). Dose-volume histogram (C) and key organ at risk tabular summary (D) between proton and photon plans.

Given the rarity of thymomas and thymic carcinomas, no randomized radiation dose-finding studies have been performed. Generally accepted radiation doses vary by margin status and the presence of gross residual disease. In the curative setting, radiation is delivered 5 days per week, with fraction sizes of 1.8-2 Gy per day. For completely resected, macroscopic and macroscopic margin negative (R0) patients, adjuvant radiation dose is 45-50.4 Gy. In patients with a microscopic positive margin (R1), the dose ranges from 50-54 Gy. In patients treated without surgery or with gross residual disease (R2), the dose ranges from 60-70 Gy. Typical dose constraints for critical organs at risk are enumerated in Table 8. Given that patients with thymomas have favorable long-term survival rates, keeping doses to critical structures as low as possible is essential.

Hemi-thoracic radiation for advanced thymoma or thymic carcinoma, or tumor spillage has been reported^[53]. Pleural control rates would be expected to be higher with hemi-thoracic radiation, but this has not been shown to be statistically significant in the limited data employing its use to date. Overall survival and local control rates are no better than for those not receiving hemi-thoracic RT. Hemi-thoracic RT is associated with increased high-grade toxicity and is technically challenging, and it should only be considered in very limited scenarios for well-select patients and in centers with experience using this technique.

CONCLUSION

Radiation therapy can be used to treat thymic epithelial cancers in the pre-operative, definitive, and post-operative settings. By far, the most common indication for radiation therapy is in the post-operative setting (PORT). There are limited randomized data to date guiding decision making and radiation recommendations. Several themes emerge from the review of the existing literature. First, it is likely that radiation therapy is most beneficial in terms of improved overall survival for patients with positive margins after surgery, more advanced stage, and more aggressive histologic subtypes. Second, advances in radiation therapy delivery may make the benefit of PORT more dramatic, as techniques like 3D-conformal radiation therapy, IMRT, and proton therapy delivered with motion management and daily image guidance can be associated with less long-term morbidity and mortality, thus improving the therapeutic ratio of radiotherapy delivery. Third, it is clear that multidisciplinary management should be employed for all patients with thymic neoplasms, and that further studies on the optimal multi-modality approach to these rare tumors are warranted.