Iterative Surgical Treatment for Repeated Recurrences After Complete Resection of Thymic Tumors

Transcription

1 GENERAL THORACIC Iterative Surgical Treatment for Repeated Recurrences After Complete Resection of Thymic Tumors Alfonso Fiorelli, MD, PhD, Antonio D Andrilli, MD, PhD, Camilla Vanni, MD, Roberto Cascone, MD, Marco Anile, MD, Daniele Diso, MD, Valentina Tassi, MD, Jacopo Vannucci, MD, Nicola Serra, PhD, Francesco Puma, MD, Erino Angelo Rendina, MD, PhD, Federico Venuta, MD, and Mario Santini, MD Thoracic Surgery Unit, Second University of Naples, Naples; Thoracic Surgery Unit, Sant Andrea Hospital, University of Rome Sapienza, Rome; Thoracic Surgery Unit, University of Rome Sapienza, Rome; Thoracic Surgery Unit, University of Perugia, Perugia; Alma Mater Studiorum, University of Bologna, Bologna; and Department of Radiology, Second University of Naples, Naples, Italy Background. In this study, we evaluated if surgical treatment and iterative operations could confer a survival advantage for patients with recurrent or repeated recurrences of thymoma. Methods. Between 1984 and 2014, 53 of 515 (10%) patients had recurrences after complete thymoma resection. Demographics, stage, treatment, and pathologic findings were statistically analyzed to identify survival prognostic factors. Results. Thirty-eight of 53 (72%) patients underwent resection of recurrent thymoma; 32 (84%) underwent complete resection. Fifteen (28%) patients did not undergo resection; 3 (20%) received chemotherapy alone, 10 (67%) received chemoradiotherapy, and 2 (13%) received supportive care. At univariate analysis, World Health Organization (WHO) AB D B1 histologic types (p < ), R0 resection (p < ), myasthenia gravis (MG) (p [ 0.02), and adjuvant therapy after recurrence (p [ 0.03) were significant prognostic factors. At multivariate analysis, complete resection (p [ ) was the only significant prognostic factor. Among patients with repeated resections, those undergoing complete resection had better survival than did those undergoing incomplete resection or no operative procedure (p [ 0.02). Seven patients are alive and free of disease, with a median survival of 115 months ( months) and 149 months ( months) from the first recurrence and from thymoma resection, respectively. Conclusions. Complete resection is a viable treatment option for selected patients with recurrent thymomas. In cases of technically resectable repeated recurrences, repeated operations should be considered. (Ann Thorac Surg 2017;103:422 31) Ó 2017 by The Society of Thoracic Surgeons Surgical resection is still considered the treatment of choice for thymoma, with significant impact on survival. However, thymoma has an indolent behavior, and recurrence may ensue many years after complete resection (R0). Despite R0 resection of thymoma, recurrence is a factor in long-term survival [1 9]; 27% to 50% of patients have relapses after an R0 resection [1, 5], and the indication for iterative operations is still an object of debate. Actually, only isolated cases of repeated resection for repeated recurrences are reported in most series [2, 3, 5 8], and no report has specifically analyzed the results of this strategy. Therefore in the present study, we reviewed the outcome of patients with recurrence or repeated recurrences (or both) after R0 thymoma resection, supposing that Accepted for publication Aug 11, Presented at the Poster Session of the Fifty-second Annual Meeting of The Society of Thoracic Surgeons, Phoenix, AZ, Jan 23-27, Address correspondence to Dr Fiorelli, Thoracic Surgery Unit, Second University of Naples, Piazza Miraglia, 2, I Naples, Italy; alfonso.fiorelli@unina2.it. surgical resection could also lead to increased long-term survival after repeated recurrences. Patients and Methods Study Design This was a retrospective multicenter study including 4 centers with homogeneous experience in the management of thymic tumors. The institutional review board of each center approved the study design. All patients with thymoma undergoing R0 resection between January 1984 and December 2014 and who then presented with recurrence were eligible. Exclusion criteria were (1) no R0 resection of initial thymoma, (2) initial histologic thymic carcinoma (because of its significantly worse prognosis), (3) no defined histologic type and Masaoka stage of primary thymoma, and (4) Masaoka stage IV thymoma, because it was virtually impossible in these cases to assess if a R0 resection was achieved and therefore to differentiate a recurrence from a progression. The date of the first recurrence was the starting point in the survival analysis, and the date of death or last follow- Ó 2017 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier

2 Ann Thorac Surg FIORELLI ET AL 2017;103: RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS up was the end point. All data were identified from a prospectively maintained surgical database of each participating center and retrospectively analyzed to identify survival prognostic factors. Patients Fifty-three of 515 (10%) patients with recurrence after R0 thymoma resection were analyzed. The medical records were reviewed for demographic information, symptoms, operative procedures, postoperative mortality and morbidity, pathologic features, postoperative therapy, recurrence date, location and treatment, and last follow-up visit or date of death. The flow chart is summarized in Figure 1. Histologic Type and Stage The initial thymoma was classified according to World Health Organization (WHO) classification as types A, AB, B1, B2, and B3. Tumors with mixed histologic type were classified as the type with the highest malignancy, eg, when the tumor had both B2 and B3 components, the tumor was classified as B3. Tumor stage was classified as I, II, and III according to the Masaoka staging classification. Recurrence and Repeated Recurrences Although follow-up varied by institution, it generally included chest and abdominal computed tomography (CT) 423 every 3 months for the first year, every 6 months for the following 3 years, and every year thereafter. From 2006, fluorine-18 fluorodeoxyglucose-positron emission tomography/ct was added to the systemic assessment of patients, and fine-needle aspiration biopsy or surgical biopsy was indicated in all lesions suspected to be recurrences. According to the recommendation of the International Thymic Malignancy Interest Group [10], recurrence was diagnosed when thymoma was strongly suspected clinically regardless of pathologic confirmation by biopsy. The date of the initial thymectomy was the starting point and the date of recurrence was the end point to define the disease-free interval (DFI). Recurrences and repeated recurrences were divided into the following 3 categories according to the International Thymic Malignancy Interest Group classification [10]: (1) local (disease presenting in the bed of the thymus or tissue immediately contiguous with the resected thymoma), (2) regional (intrathoracic disease not immediately contiguous with the thymus or the previously resected thymoma), and (3) distant (extrathoracic recurrence and intraparenchymal pulmonary nodules). Only resectable recurrent lesions underwent operative treatment. Each surgeon at his/her discretion, decided the extension of the operation based on the patient s general GENERAL THORACIC Fig 1. Flow chart of study population.

3 GENERAL THORACIC 424 FIORELLI ET AL Ann Thorac Surg RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS 2017;103: condition and findings in the surgical field. The indication for induction or adjuvant therapy was established according to the multidisciplinary board decision in each participating center. Recurrences were reassessed according to WHO classification, and histologic upgrading was defined as the migration of the histologic WHO classification of initial thymoma to a more aggressive histologic type in the recurrent thymoma. Statistical Analysis Data are presented as number and percentages for categorical variables and as median and interquartile range [IQR] for nonnormally distributed variables. The statistical difference was assessed with the c 2 test for categorical variables and with the Wilcoxon or Kruskal-Wallis test, or both, as appropriate for nonnormally distributed variables. The survival curve was calculated by the Kaplan- Meier method, and the log-rank test was used for univariate analysis of predictive survival factors. In the multivariable analysis (dependent variable ¼ survival), the survival-predicting factors at univariate analysis were entered into the Cox proportional hazard regression model. A p value less than 0.05 was considered significant. MedCalc statistical software, version 12.3 (Ostend, Belgium) was used for analysis. Results Table 1 and Table 2 summarize the characteristics of the study population. A significantly increased relapse rate was found in patients with Masaoka stage III disease (p < ), whereas only a trend was found in patients with B2/B3 subtypes (p ¼ 0.08). The recurrence rate was similar between patients with myasthenia gravis (MG) and patients without MG (p ¼ 0.3). Regional (49%) recurrence was most common when compared with local (24%) and distant (27%) recurrence, despite the difference not being significant (p ¼ 0.05). The mean DFI was 55 months (38-69 months), and it did not show a significant difference among patient groups stratified according to MG (MG, 50 months [41 59 months] versus no MG, 64 months [35 83 months]; p ¼ 0.15), Masaoka stage (stage I, 67 months [20 72 months]; stage II, 54 months [39 67 months]; stage III, 52 [42 69 months]; p ¼.6); WHO classification (stage AB, 65 months [42 74 months]; stage B1, 64 months [55 79 months]; stage B2, 49 months [41 68 months]; stage B3, 40 months [34 50 months]; p ¼ 0.06); and adjuvant therapy (adjuvant therapy, 34 months [17 71 months] versus no adjuvant therapy, 76 months [46 87 months]; p ¼ 0.3). Thirty-eight of 53 (72%) patients were operated on for recurrence. Recurrence was in the mediastinum in 10 patients, was pleural in 20 patients, and was distant in 8 patients. An R0 resection was performed through a thoracotomy in 31 patients (96%) or video-assisted thoracoscopic surgery (VATS) in 1 (4%) patient, whereas in 6 of 38 (16%) patients, an R0 resection was planned according to radiologic findings but was converted to a debulking procedure during the operation (4 VATS procedures and Table 1. Study Population Variable No. or Median Percentage or [IQR] Patients 53 Sex (male) 30 57% Age 48 [29 71] Size of thymoma 6 [3 14] Myasthenia gravis 30 56% Induction therapy 6 11% Chemotherapy 5 9% Radiotherapy 1 2% Surgical approach Sternotomy 32 60% Thoracotomy 21 40% Type of resection Thymectomy alone 45 85% Thymectomy þ lung resection 4 7% Thymectomy þ superior vena cava 3 6% Thymectomy þ atrium 1 2% Masaoka stage I 5 9% II 9 17% III 39 74% WHO classification AB 7 13% B % B % B % Adjuvant therapy Chemotherapy 15 28% Radiotherapy 27 51% IQR ¼ interquartile range; WHO ¼ World Health Organization. 2 thoracotomies) because of the extension of disease. There was no mortality or major postoperative complications. Adjuvant therapy was performed in all 6 patients undergoing incomplete resection and in 26 of 32 (81%) patients undergoing R0 resection. WHO histologic upgrading was found in 14 of 38 (36%) cases. Fifteen of 53 (28%) patients were considered radiologically inoperable because of diffuse intrathoracic (n ¼ 3) or multiple recurrences (n ¼ 12), or both. They received chemotherapy alone (3 of 15 [20%]), chemoradiotherapy (10 of 15 [67%]), or supportive care (2 of 15 [13%]). Repeated Recurrences A second recurrence was seen in 22 of 32 (69%) patients who underwent an R0 resection, with a median DFI of 33 months (17 68 months) from the first recurrence. Eleven of these 22 patients (50%) underwent a repeated resection, which resulted to in an R0 resection in 10 of 11 (91%) of them (Table 3). No death or major complications occurred. Histologic grading changed from B1 to B3 in 1 of 11 (9%) patients undergoing repeated resection. In the remaining 11 of 22 (50%) patients, operative treatment was unfeasible because of the extension of disease (n ¼ 9

4 Ann Thorac Surg FIORELLI ET AL 2017;103: RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS Table 2. Patients With Repeated Recurrences Undergoing Reoperation Variable No. or Mean SD Percentage or [IQR] Patients 53 Disease-free interval (mo) 55 [12 144] Symptoms at recurrence Asymptomatic 33 63% Chest pain 10 19% Myasthenia gravis 19 37% Cough 7 13% Pattern of recurrence Local 13 24% Regional 26 49% Distant (lung/extrathoracic) 14 (11/3) 27% Treatment of recurrence Complete resection 32 60% Incomplete resection 6 11% (debulking) Chemotherapy/radiotherapy 13 25% alone Supportive care alone 2 4% Reoperation 38 72% WHO classification B 1 3% B1 3 9% B % B % C 2 6% WHO upgrading 14 37% AB to B3 3 22% B1 to B2 2 14% B1 to B3 2 14% B2 to B3 5 36% B3 to C 2 14% Adjuvant therapy after resection 32 84% of recurrence Chemotherapy alone 11 35% Radiotherapy alone 2 6% Chemotherapy þ radiotherapy 19 59% IQR ¼ interquartile range; WHO ¼ World Health Organization. of 11 [81%]) or poor clinical condition (n ¼ 2 of 11 [18%]). Chemotherapy (n ¼ 4 of 11 [36%]), radiotherapy (n ¼ 2of 11 [18.5%]), chemoradiotherapy (n ¼ 3 of 11 [27%]), or supportive care (n ¼ 2 of 11 [18.5%]) was carried out in patients not undergoing repeated resection. Eight of 10 patients (80%) undergoing repeated R0 resection had a third recurrence, with a median DFI of 15 months (11 59 months) from the second recurrence. Seven of 8 patients (87%) underwent repeated resection, which resulted in R0 resection in 6 (86%) cases. No death or major complications occurred. Histologic grading changed from B2 to B3 in 1 of 7 (14%) patients undergoing resection for the third recurrence. Five of 6 (83%) patients undergoing R0 resection for a third recurrence had a fourth recurrence, with a median DFI of 19 months (15 22 months). Of these patients, 3 of 5 (60%) had a new R0 resection through a thoracotomy, whereas operative treatment was not performed in the other 2 patients because of the extension of disease. No death or major complications occurred. Survival The median follow-up time was 55 months (38 69 months). Thirty of 53 (57%) patients died because of a recurrent thymoma related cause, whereas the remaining 23 of 53 (43%) patients are alive with stable disease (n ¼ 6 of 23 [26%]) or are disease free (17 of 23 [74%]). The overall 5-year survival rate (YSR) and 10-YSR were 52% and 32%, respectively (Fig 2A). At univariate analysis (Table 4), AB þ B1 histologic types (p ¼ ) (Fig 2B) and R0 resection (p < ) (Fig 2C) were inversely associated with survival, whereas distant recurrence (p ¼ ) (Fig 2D) and MG (p ¼ 0.02) (Fig 2E) were directly associated with survival. Patients with Masaoka stage I (p ¼ 0.07) who were receiving adjuvant therapy after thymoma resection (p ¼ 0.1) and had a DFI greater than 5 years (p ¼ 0.15) had a survival benefit, despite it not being statistically significant. In patients who underwent an R0 resection, adjuvant therapy after recurrence (p ¼ 0.03) (Fig 3F) was a statistically significant prognostic factor, but WHO histologic upgrading (p ¼ 0.3) was not. At multivariate analysis (Table 5), R0 surgical resection was the only significant prognostic factor. Among patients having repeated recurrences, the overall 5-YSR and 10-YSR were 71% and 41%, respectively. Patients who received 1 or more repeated R0 resections (n ¼ 10) had better survival compared with patients with incomplete operations or no operations (n ¼ 12) (Fig 3). Seven patients undergoing repeated resection were alive and disease free, with a median survival of 115 months ( months) from the first recurrence and 149 months ( months) from thymoma resection. Comment 425 Our recurrence rate of 10% for all resected thymomas is in line with most reports [1 9], whereas the DFI (55.9 months) seems to be shorter compared with others [2 4]. The early diagnosis of recurrence resulting from more frequent follow-ups could be a possible explanation. Previous studies [3, 5, 6] showed that the probability of relapse was related to the Masaoka stage and the WHO histologic type; we found a significant increase in relapse rate for Masaoka stage III and a trend for types B2 and B3 thymoma. Pleural relapse was the most common site in our series, as in most other series [1 8]. Interestingly, no mediastinal recurrence was found in patients receiving radiotherapy (n ¼ 27 [51%]) after thymoma resection. Similar findings were reported by Regnard and associates [2]. Despite the fact that the benefit of radiotherapy after resection of early-stage thymoma is controversial [11], theoretically, microscopic residual tumor left in place could be cured by radiotherapy, GENERAL THORACIC

5 GENERAL THORACIC Table 3. Characteristics of Patients With Repeated Recurrence No. Age/ Sex MG Masaoka Stage Initial Thymoma WHO Classification Resection Treatment After Resection Pattern DFI (mo) First Recurrence Resection/ Approach WHO Classification Treatment After Resection 1 45/F Yes III B2 Thymectomyþ SVC RT Lung 70 R0/thoracotomy B2 No 2 36/M Yes II B3 Thymectomy No Lung 16 R0/thoracotomy B3 No 3 55/F No III B3 Thymectomyþ lung RT Lung 16 R0/thoracotomy B3 No 4 47/M No II B2 Thymectomy No Pleura 22 R0/thoracotomy B2 CT 5 57/F Yes I B2 Thymectomy No Pleural þ 25 R0/thoracotomy B2 No mediastinum 6 32/F No II B2 Thymectomy þ SVC RT Pleural 34 R0/thoracotomy B2 No 7 64/M No III B2 Thymectomyþ SVC RT Mediastinum 60 R0/thoracotomy B2 CT 8 45/F No I B1 Thymectomy No Mediastinum 108 R0/thoracotomy B1 CT þ RT 9 31/M No III B1 Thymectomy CT Pleural 144 R0/thoracotomy B2 CT þ RT 10 41/M No II B1 Thymectomy CT Lung þ pleural 12 R0/thoracotomy B2 CT þ RT 11 34/M Yes II B2 Thymectomy No Mediastinum 72 R0/thoracotomy B2 CT 12 36/F No II B2 Thymectomy CT Pleura 36 R0/thoracotomy B3 CT No. Pattern DFI Second Recurrence Resection/ Approach WHO Classification Treatment After Resection Pattern DFI Third Recurrence Approach/ Operation WHO Classification Treatment After Resection 1 Right lung 48 R0/VATS B2 CT þ RT Left lung 96 R0/VATS B2 CT 2 Pleural 37 R0/thoracotomy B3 No Pleural 18 R0/thoracotomy B3 No 3 Pleural 15 R1 resection/vats B3 No Abdomen 23 No surgical... CTþ RT intervention 4 Mediastinum 12 R0/VATS B2 No Pleural 12 R0/thoracotomy B Pleural 20 R0/thoracotomy B2 No Pleural 30 R0/thoracotomy B2 CT Pleural 10 R1/thoracotomy B3 CT 7 Pleural 12 R0/thoracotomy B2 CT Abdomen 12 R0/laparoscopy B2 CT 8 Mediastinum 276 R0/thoracotomy B3 CT Pleural 166 R0/thoracotomy B2 CT Brain 46 No surgical B2 CT þ RT procedure 11 Pleural 89 R0/thoracotomy B2 CT Mediastinum þ 9 R0/thoracotomy B2 CT pleural 12 Right lung 23 R0/VATS B3 CT Lung left 101 R0/VATS B3 CT (Continued) 426 FIORELLI ET AL Ann Thorac Surg RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS 2017;103:422 31

6 Table 3. Continued No. Fourth Recurrence Pattern DFI Resection/Approach WHO Classification Treatment (After Resection) Current Status 1 Pleural 12 R0/thoracotomy B2 CT Alive Pleural 19 R0/thoracotomy B3 CT Alive Death/complication from abdominal 49 metastasis Alive Alive Death/respiratory failure from progressive 70 intrathoracic disease 7 Brain 25 No surgical intervention B3 RT Death/complication of brain metastasis Alive Alive Death/complication from brain metastasis Pleural 17 R0/thoracotomy B2 CT Alive Pleural 21 No surgical intervention B3 RT Death/respiratory failure from progressive intrathoracic disease 161 CT ¼ chemotherapy; DFI ¼ disease-free interval; F ¼ female; M ¼ male; MG ¼ myasthenia gravis; RT ¼ radiotherapy; SVC ¼ superior vena cava; VATS ¼ video-assisted thoracoscopic surgery; WHO ¼ World Health Organization. Survival Ann Thorac Surg FIORELLI ET AL 2017;103: RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS 427 GENERAL THORACIC

7 GENERAL THORACIC 428 FIORELLI ET AL Ann Thorac Surg RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS 2017;103: Fig 2. (A) Overall survival and (B) long-term survival according to World Health Organization (WHO) histologic type. (C) Treatment, (D) pattern of recurrence, (E) myasthenia gravis, and (F) adjuvant therapy. which conversely cannot prevent pleural or distant implantation. In our series, adjuvant chemotherapy or radiotherapy did not prolong the DFI, as was reported by Sandri and colleagues [4]. Furthermore, DFI was not correlated with the WHO subtypes, Masaoka stage, and MG as in other studies [4, 6]. To ensure a complete resection, open surgical procedures were performed in 96% of patients, which is in line with other authors (94.4% 100%) [1 6]. Despite that the surgical approach depended on the site of recurrence, thoracotomy was preferred to repeated sternotomy because of less invasiveness. VATS was used in only 5 cases, probably because of diffuse adhesions related to previous operations and to different indications for this approach in each participating center. Our results confirmed that R0 resection is the most important determinant of long-term survival in recurrent thymoma, with a 5-YSR and 10-YSR of 72% and of 47%, respectively. Hamaji and coworkers [1] reviewed the results of 278 patients treated for recurrent thymoma from 11

8 Ann Thorac Surg FIORELLI ET AL 2017;103: RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS Table 4. Univariate Analysis Variable HR 95% CI p Value Age, y (50 versus >50) Sex (male versus female) Treatment < R0 versus debulking R0 versus no surgical intervention Masaoka stage 0.07 I versus II I versus III WHO classification AB þ B1 versus B AB þ B1 versus B Myasthenia (yes versus no) Recurrence Distant versus local Distant versus regional Adjuvant therapy after initial thymoma resection (yes versus no) DFI (>5 y versus 5 y) Adjuvant therapy versus no adjuvant therapy a WHO upgrading (yes versus no) a a Only patients who underwent R0 resection. CI ¼ confidence interval; DFI ¼ disease-free interval; HR ¼ hazard ratio; WHO ¼ World Health Organization. retrospective studies published between 1991 and Surgical versus nonsurgical treatment conferred a better 5-YSR (70.9% 16.2% versus 49.6% 27.4%, respectively) and 10-YSR (29.6% 21.9% versus 18.4% 26%, respectively). Additional studies published after 2013 also confirmed the positive prognostic impact of R0 resection. Marulli and associates [5] reported a 5-YSR and 10-YSR of 80% and 60%, respectively, after R0 resection of recurrent thymoma. Bae and coworkers [6] reported a 5-YSR and 10-YSR of 91.7% after R0 resection, which was better than that of patients undergoing incomplete resection (44.7% at 5 years) and comparable to that of patients without Fig 3. Survival according to treatment in patients with repeated recurrences. 429 recurrence (90.7% and 86.5% at 5 and 10 years, respectively). Sandri and colleagues [4] found a 5-YSR and 10-YSR of 82.4% and 65.4%, respectively, after R0 resection of recurrent thymoma. Compared with our series, Haniuda and associates [12] obtained a lower 5-YSR and 10-YSR (47% and 35%, respectively) after resection of recurrent thymoma. However, a R0 resection was performed in only 4 of 15 (27%) patients compared with 80% in our series, and this could explain the worse survival results. In addition to R0 resection, previous studies [1, 9] found that patterns of recurrence, DFI, and WHO classification were significant survival prognostic factors for recurrent thymoma. Despite that a negative prognostic trend was seen in our patients with distant recurrence, this was not significant, which is in agreement with data reported by Regnard and coworkers [2] and Sandri and associates [4]. This may be because of the small number of patients or the more aggressive surgical approach used for metastatic recurrence. For management of pleural relapses (n ¼ 20), we performed multimodal treatment including complete pleurectomy and adjuvant chemotherapy. If comparing this approach with the treatment of the resectable pulmonary metastases (n ¼ 6), the lung resection could probably ensure a higher probability of a complete resection than would pleurectomy [13]. Therefore, Sandri and colleagues [4] supposed that pulmonary metastases could be classified as locoregional relapses and, accordingly, not be equivalent regarding treatment and prognosis with other hematogenous recurrences. The possibility of achieving complete resection when treating pleural recurrence is still an object of controversy. Some authors suggested a more extensive procedure than pleurectomy alone extrapleural pneumonectomy [14]. In our series, WHO classification and histologic WHO upgrading were not significant prognostic factors, which was in line with other authors findings [3, 5]. In previous literature, it was postulated that epithelial cells in thymoma relapsed more frequently than did the lymphocytic component [15]. Pescarmona and colleagues [16] found a depletion or loss of lymphoid cells in 5 of 9 cases of recurrent thymoma, compared with primary thymoma, and they ascribed it to the histologic progression of the disease. However, some observations seem to deny this hypothesis. Bae and colleagues [6] found that lymphoid depletion could be caused by high doses of steroid therapy independent of disease progression. Vladislav and colleagues [15] speculated that lymphoid-rich stroma in thymoma was probably a function of cytokine secretion and therefore an epiphenomenon rather than a sign of true progression. Ciccone and Rendina [17] supposed that only a minority component within the cortically differentiated primary tumor relapsed. Therefore, the current WHO histologic classification of thymoma is heavily biased toward the character of this epiphenomenon rather than the character of the neoplastic cells. In fact, recurrences are reported even after resection of type A tumor, despite that this is generally considered a benign tumor. A more precise staging system based on new biomarkers could GENERAL THORACIC

9 GENERAL THORACIC 430 FIORELLI ET AL Ann Thorac Surg RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS 2017;103: Table 5. Multivariate Analysis Variable p Value HR 95% CI Complete resection (yes versus no) Myasthenia gravis (yes versus no) Who classification (B3 þ C versus others) Site of recurrence (distant versus others) Adjuvant therapy after thymoma (yes versus no) CI ¼ confidence interval; HR ¼ hazard ratio. help to improve the prognostic value of the WHO classification. Most data available on debulking operations [18], including ours, showed no benefit; therefore this strategy should be reserved for very selected cases, with alternative treatments being preferred for the others. The role of adjuvant therapy after resection of recurrent thymoma is not completely defined. In the meta-analysis of Hamaji and coworkers [1], the incidence of adjuvant chemotherapy and radiotherapy was low (20.4% 32.7% and 20.9% 32.8%, respectively). However, because 27% to 50% of patients undergoing R0 resection had relapses, this could suggest that a truly curative resection seems unlikely, and in theory adjuvant therapy could improve local and systemic control of disease. Supporting this hypothesis, in our series and that of Bae and colleagues [6], 81% and 70% of patients, respectively, underwent adjuvant therapy after R0 resection of recurrent thymoma. Because many series are collected over multiple decades, the strategy for adjuvant therapy is based on the surgeon s or oncologist s preference rather than on a standardized protocol, and this may explain the different results between various centers. Adjuvant therapy was not a significant prognostic factor in our series or in other series [3 6]; however, the small amount of data and the nonuniform strategies make the evaluation of its effectiveness difficult. Finally, for the first time we reviewed a consecutive series of patients with repeated recurrences and found that repeated R0 resection could lead to a survival advantage over no surgical treatment or debulking. Although it is largely accepted that redo operations, when feasible, should be attempted in selected patients with repeated recurrences, actually validated therapeutic strategies have not been established, and chemotherapy is still considered the treatment of choice [1]. Our study included 10 patients undergoing R0 resections for repeated recurrences. Reoperations were carried out without significant morbidity and mortality. Good performance status of patients and appropriate surgical approaches may have allowed this result. In fact, we avoided repeated sternotomies because of potential related morbidity, especially in the irradiated field. Seven patients are actually alive and disease free, with a median survival of 149 months after thymoma resection. Similarly, Bae and colleagues [6] found a second recurrence after R0 resection of recurrent thymoma in 3 of 15 patients (20%). Of these patients, two-thirds underwent repeated R0 resections and were alive and disease free. Regnard and associates [2] reported 3 of 19 (16%) patients with repeated recurrences after R0 resection. One of them had a repeated R0 resection and was still alive and disease free at 46 months. Yano and associates [7] reported that 6 of 12 patients underwent a repeated resection of recurrence and 5 of 6 were alive. Margaritora and coworkers [3] reported a second recurrence in 4 of 22 (18%) patients undergoing R0 resection of recurrent thymoma; a total of 11 repeated resections were performed and 1 patient was alive and disease free 21 years after the initial diagnosis. In the series of Bott and colleagues [8] and Marulli and coworkers [5], 55% and 21.9% of patients, respectively, underwent multiple resections for repeated recurrences, but no data regarding the outcome of these subgroups were available. These data confirm that the natural history of thymoma is unpredictable. Lesions may progress slowly and can recur up to decades after operations and reoperations. Because of the lack of specific nonsurgical or pathologic biomarkers that could predict the time of repeated recurrences, it is crucial to perform long-term surveillance (>10 years) after the last resection to allow early diagnosis of the repeated recurrences and to perform a complete resection. This could explain the observation that in our series of patients with repeated recurrences, an incomplete resection was performed in only 2 patients. Improvement in radiologic findings over the long followup period probably helped us to reduce the rate of incomplete resection. Study Limitations Because of the retrospective nature of this study, patients undergoing surgical treatment had limited disease and better performance status than did nonsurgical patients, and this could have affected survival. In addition, different follow-up and therapeutic regimens, resulting from the long duration of patient inclusion and the multicenter nature of the study represented further significant limitations. Furthermore, we did not provide the number of recurrent lesions, despite several authors [3, 5, 15] finding it to be a significant prognostic factor. However, Bae and associates [6] reported that complete resection is more important as a prognostic determinant than the number of recurrent lesions. Conclusions In patients with recurrent thymoma, R0 resection, when feasible, should be strongly considered, because it is

10 Ann Thorac Surg FIORELLI ET AL 2017;103: RE-RECURRENCE AFTER RESECTION OF THYMIC TUMORS associated with good long-term outcomes. Long-term surveillance and careful follow-up are essential for early detection of recurrence to increase the chances of R0 resection. Iterative operations should be reserved for those patients in good clinical condition and with technically resectable repeated recurrence. References 1. Hamaji M, Ali SO, Burt BM. A meta analysis of surgical versus nonsurgical management of recurrent thymoma. Ann Thorac Surg 2014;98: Regnard JF, Zinzindohoue F, Magdeleinat P, et al. Results of repeated resection for recurrent thymomas. Ann Thorac Surg 1997;64: Margaritora S, Cesario A, Cusumano G, et al. Single-centre 40-year results of redo operation for recurrent thymomas. Eur J Cardiothorac Surg 2011;40: Sandri A, Cusumano G, Lococo F, et al. Long-term results after treatment for recurrent thymoma: a multicenter analysis. J Thorac Oncol 2014;9: Marulli G, Margaritora S, Lucchi M, et al. Surgical treatment of recurrent thymoma: is it worthwhile? Eur J Cardiothorac Surg 2016;49: Bae MK, Byun CS, Lee CY, et al. Clinical outcomes and prognosis of recurrent thymoma management. J Thorac Oncol 2012;7: Yano M, Sasaki H, Moriyama S, et al. Number of recurrent lesions is a prognostic factor in recurrent thymoma. Interact Cardiovasc Thorac Surg 2011;13: Bott MJ, Wang H, Travis W, et al. Management and outcomes of relapse after treatment for thymoma and thymic carcinoma. Ann Thorac Surg 2011;92: Dai J, Song N, Yang Y, et al. Is it valuable and safe to perform reoperation for recurrent thymoma? Interact Cardiovasc Thorac Surg 2015;21: Huang J, Detterbeck FC, Wang Z, et al. Standard outcome measures for thymic malignancies. J Thorac Oncol 2010;5: Haniuda M, Miyazawa M, Yoshida K, et al. Is postoperative radiotherapy for thymoma effective? Ann Surg 1996;224: Haniuda M, Kondo R, Numanami H, et al. Recurrence of thymoma: clinicopathological features, re-operation, and outcome. J Surg Oncol 2001;78: Lucchi M, Basolo F, Mussi A. Surgical treatment of pleural recurrence from thymoma. Eur J Cardiothorac Surg 2008;33: Ishikawa Y, Matsuguma H, Nakahara R, et al. Multimodality therapy for patients with invasive thymoma disseminated into the pleural cavity: the potential role of extrapleural pneumonectomy. Ann Thorac Surg 2009;88: Vladislav T, Jain RK, Alvarez R, et al. Extrathoracic metastases of thymic origin: a review of 35 cases. Mod Pathol 2012;25: Pescarmona E, Rendina EA, Venuta F, et al. Recurrent thymoma: evidence for histological progression. Histopathology 1995;27: Ciccone AM, Rendina EA. Treatment of recurrent thymic tumors. Semin Thorac Cardiovasc Surg 2005;17: Attaran S, Acharya M, Anderson JR, et al. Does surgical debulking for advanced stages of thymoma improve survival? Interact Cardiovasc Thorac Surg 2012;15: GENERAL THORACIC ABTS Announcement for Maintenance of Certification The American Board of Thoracic Surgery s Maintenance of Certification program was adopted 9 years ago. Since that time, there has been a continuous evaluation in the Board s thinking about the overall process, based upon internal discussions and input from our diplomates. These inputs resulted in our decision to migrate from a purely knowledge-based multiple choice exam, utilizing a Pearson Testing Center to a Mastery Learning Process, using a SESATS format. Diplomates, enrolled in this year s (2017) 10-year MOC process, will fulfill their Part III requirement by completion of a home or office-based secure learning exam, following the instructions on the ABTS website. In brief, you will be directed to a secure website. The only special computer hardware needed will be a camera for your home or office computer (most laptops now come with a built-in camera). Once logged in, you will be asked to verify your identity by holding up your driver s license with your picture next to your face. You will be visually monitored for the time you are logged onto the website. There are 100 SESATS-based questions that focus on your specialty designation (Adult Cardiac, General Thoracic, Cardiothoracic, and Congenital), that you will need to work through as instructed. The exam will now be modular and tailored to your practice for example, if your practice is 100% adult cardiac, you will only have adult cardiac and critical care questions. You will have 15 hours with as many as 10 logins to complete the 100 questions during the months of September and October For those diplomates who have used SESATS in the past, the process of working through the questions is the same. For those who are not familiar with SESATS, it might be beneficial to purchase and download SESATS and work through the specialty specific module. This preparation will give you familiarity with the process. While SESATS may be helpful preparation, it is not required. The goal of this exam is to provide a learning opportunity using judgement and decision making as well as knowledge. The Board and MOC Committee believe that reading the critique is key to the learning process using SESATS. The remote proctoring system will allow us to verify the pace of completion and thus limit the passing grades to those who earnestly participate in the process. The Board sincerely hopes that this focus on life-long learning is viewed favorably by our diplomates. The Board will continue with this new strategy and refine SESATS as we go forward to assure that new standards of care are communicated to members of the ABTS community as part of the MOC process. There will be a brief survey following the last SESATS question which needs to be completed to officially finish the process. Everyone at the ABTS thanks you for embracing the primary principle of MOC life-long learning, which is consistent with our obligation to the public trust. Ó 2017 by The Society of Thoracic Surgeons Ann Thorac Surg 2017;103: /$36.00 Published by Elsevier