Radical surgery for malignant pleural mesothelioma: results and prognosis

doi:10.1510/icvts.2007.166322 Interactive CardioVascular and Thoracic Surgery 7 (2008) 102 106 www.icvts.org Institutional report - Thoracic general Radical surgery for malignant pleural mesothelioma: results and prognosis a,b, a,b c c d Morihito Okada *, Takeshi Mimura, Chiho Ohbayashi, Toshiko Sakuma, Toshinori Soejima, Noriaki Tsubota b a Department of Surgical Oncology, Research Institute for Radiation Biology and Medicine, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima City, Hiroshima, 734-8553, Japan b Department of Thoracic Surgery, Hyogo Cancer Center, Hyogo, Japan c Department of Pathology, Hyogo Cancer Center, Hyogo, Japan d Department of Radiology, Hyogo Cancer Center, Hyogo, Japan Received 3 September 2007; received in revised form 2 November 2007; accepted 5 November 2007 Abstract The role of surgical treatment for malignant pleural mesothelioma (MPM) continues to be controversial. We carried out a retrospective review of the prognosis in patients who had radical surgery for MPM. Of 87 consecutive patients on whom surgical exploration for biopsyproven MPM was performed, 31 patients underwent extrapleural pneumonectomy (EPP) and 34 patients underwent pleurectomyy decortication (PyD). Sixty-five patients having EPP or PyD included 58 men (89%). The median age was 60 years (range 35 78) and the histologic type was epithelial in 48 patients (74%). IMIG staging classification was p-stage I disease in eight patients (12%), p-stage II in 13 (20%), p-stage III in 40 (62%) and p-stage IV in 4 (6%). Operative mortality was 3.2% for EPP and none for PyD. The median and 3-year survivals after EPP were 13 months and 33% whereas those after PyD were 17 months and 24%, respectively. A multivariate analysis demonstrated that older age (Ps0.0467), non-epithelial histology (Ps0.0057) and p-stage III IV disease (Ps0.0019), but not gender, side, surgical procedure, were significant independent negative prognostic factors. Although PyD appears to be acceptable in early stages, we encourage EPP, en bloc resection without entering the pleural cavity with intent for curability, which provides oncologically complete resection of all disease. 2008 Published by European Association for Cardio-Thoracic Surgery. All rights reserved. Keywords: Malignant mesothelioma; Extrapleural pneumonectomy; Pleurectomyydecortication; Survival 1. Introduction Not anything frustrates thoracic surgeons more than to be faced with a malignant pleural mesothelioma (MPM), which is a rare tumor, characterized by locally aggressive behavior, is an unvaryingly fatal disease and has been recalcitrant to various treatments. In general, the median survivals after diagnosis were 8 18 months for the majority because of chemo- and radio-resistance and of the difficulty in performing complete resection; thus specific curative treatments were not offered with a sense of nihilism, and the therapeutic strategies actually performed were not often uniform even in all-inclusive cancer centers w1x. However, ideal therapeutic options should be investigated because the incidence has been continuing to rise in most countries. The significance of surgery continues to be a matter of debate. Sugarbaker and colleagues w2x conducted a major series in surgical treatment by extrapleural pneumonectomy (EPP) for MPM, demonstrating that patients with microscopic negative resection margins, negative lymph *Corresponding author. Tel.: q81-82-257-5869; fax: q81-82-256-7109. E-mail address: morihito1217jp@aol.com (M. Okada). 2008 Published by European Association for Cardio-Thoracic Surgery nodes and epithelial histology had an excellent long-term survival (median and 5-year survivals: 51 months and 46%, respectively). Although EPP has been eagerly adopted for MPM in some medical centers, the superiority over a lung-sparing procedure pleurectomyydecortication (PyD) remains unverified. In addition to the USA and Europe, MPM concerns us seriously in Japan since we are facing an epidemic, which will peak around the year 2020 due to the prevalence of occupational exposure to asbestos and the long latency period. In the present study, we analyzed the surgical results of patients radically performing EPP or PyD for MPM, and evaluated the prognostic factors. This analysis, although small, is an instructive representation of all that is controversial in the management of the patient with the potentially resectable mesothelioma. In addition, we reported our results of recent multimodality treatment including EPP with adjuvant hemithorax conformal radiation for the purpose of clinical trials. 2. Patients and methods Between November 1986 and December 2006, 87 consecutive patients underwent surgical exploration for possi-

M. Okada et al. / Interactive CardioVascular and Thoracic Surgery 7 (2008) 102 106 103 ble resection of biopsy-proven MPM. Of the 87 patients, 65 patients (75%) had a radical surgery defined as complete resection which showed no gross residual tumor together with dissection of ipsilateral hilar and mediastinal lymph nodes by one surgical team through EPP in 31 patients (36%) or PyD in 34 patients (39%). Exploration only or palliative partial resection of the tumor was performed in 22 patients (25%). The definite diagnosis of MPM was strictly achieved on immunohistochemistry as well as standard histopathology w3x. All patients had a lung functional test, ventilation-perfusion scan, echocardiography, computed tomography of the chest, brain, and abdomen before resection. They were preoperatively considered having a resectable disease confined to one hemi-thorax and not invading the chest wall, mediastinal organs or abdominal cavity through the diaphragm. Surgical-pathologic staging was performed according to TNM classification by the International Mesothelioma Interest Group (IMIG) w4x. EPP was defined as an en bloc resection of the lung, pleura, pericardium and diaphragm without entering the pleural cavity. In contrast, PyD was defined as total removal of the parietal pleura, visceral pleural, mediastinal pleura, pericardium and diaphragm. Partial or no removal of pericardium or diaphragm was sometimes done for a parietal pleural tumor separable from the pericardium or diaphragm. Mainly, the determination on whether to execute an EPP or a PyD was grounded on the degree of the tumor invasion into the pulmonary parenchyma as well as tolerability for surgical procedure including lung function. Postoperatively all patients had a physical examination every three months, and computed tomography of the chest and abdomen every six months. Biopsy such as needle biopsy was carried out when a patient was suspected of recurrent disease. Survival was calculated by the Kaplan Meier method, and differences in survival were determined by the log-rank analysis. A multivariate analysis of several independent prognostic factors was carried out using Cox s proportional hazards regression model. Zero time was the date of surgery, and the terminal event was death attributable to tumor or non-tumor. Differences were considered to be statistically significant when the P value was -0.05. 3. Results Of 65 patients who had a radical surgery, 7 (11%) were women and 58 (89%) were men. The median age was 60 years (range 35 78 years). Forty-seven patients (72%) had known asbestos exposure. Forty-one patients (63%) had a right-sided lesion and 24 (37%) had a left-sided lesion. The histologic type of the tumor was epithelial in 48 patients (74%), whereas it was mixed in 11 (17%) and sarcomatous in 6 (9%). According to the IMIG staging system, pathological examination demonstrated that 8 patients had stage I disease (12%), 13 patients had stage II disease (20%), 40 patients had stage III disease (62%) and 4 patients had stage IV disease (6%). In the four patients with pathologic stage IV lesion, postoperative pathological examination showed microscopically residual tumors in the chest wall (T4 disease) although intraoperative findings had macroscopically revealed complete resection of the tumor. Regarding 21 patients with nodal involvement, only three patients (14%) had metastasis confined to N1 sites while 18 patients (86%) had N2 disease. The characteristics of the patients according to surgical procedure are shown in Table 1. There were trends toward more epithelial type and more early stage lesion in the PyD group rather than in the EPP group. Major postoperative complications occurred in 15 patients (48%) in EPP group and five patients (15%) in PyD group. One in-hospital operative death secondary to lung infarction occurred in the EPP group. Other complications included supraventricular arrhythmias (eight patients in EPP group and three patients in PyD group), respiratory failure (four patients in EPP group), respiratory infection (one patient in EPP group and two patients in PyD group), bleeding, heart hernia, bronchial stump insufficiency, chylothorax (two patients each in EPP group) as well as heart failure, laryngeal nerve palsy (one patient each in EPP group). Thirty-day operative mortality was 1.5% (1y65 patients) for the entire patient group and was 3.2% (1y31) for the patients who had an EPP. No patients undergoing PyD had operative mortality. Because follow-up length reflected the inclusion of patients who had a poor life expectancy and of patients who died postoperatively, the median follow-up time for all 65 patients was nine months. When overall survival of the 31 patients who had an EPP was compared with that of the 34 patients who had a PyD (Fig. 1), the difference was not statistically significant (Ps0.9220). The median and 3-year survivals after EPP were 13 months and 33%, whereas those after PyD were 17 months and 24%, respectively. Because of the small number of patients with mixed histology or sarcomatoid histology, they were classified together for survival analysis and compared to patients with pure epithelial histology (Fig. 2). The median and 3-year survivals with epithelial histology were 18 months and 33% while those with non-epithelial histology were 8 months and 8%, respectively. There was a greatly significant difference in survival in favor of patients who had a pure epithelial histology (Ps0.0048). The overall survivals in accordance with pathologic stage are shown in Fig. 3. When individually compared between four stage categories, there was no difference in survival Table 1 Characteristics of the patients according to the type of surgical procedure performed Extrapleural pneumonectomy Pleurectomy decortication Number 31 cases 34 cases Sex Maleyfemale 27y4 31y3 Age (years) Range 35 72 37 78 Median 60 60 Side Leftyright 14y17 10y24 Histology Epiymixysar 19y9y3 29y2y3 IMIG p-stage IyIIyIIIyIV 0y5y24y2 8y8y16y2 Epi, epithelial type; mix, mixed; sar, sarcomatoid; IMIG, International Mesothelioma Interest Group.

104 M. Okada et al. / Interactive CardioVascular and Thoracic Surgery 7 (2008) 102 106 Fig. 1. Overall survival analyzed according to surgical procedure. There was no significant difference between patients with extrapleural pneumonectomy vs. pleurectomyydecortication (Ps0.9220). Fig. 3. Overall survival analyzed according to pathologic stage. Stage had a high impact on prognosis (stage I vs. II, Ps0.8927; stage II vs. III, Ps0.0615; stage III vs. IV, Ps0.0012). Fig. 2. Overall survival analyzed according to histologic type. Patients who had epithelial histology had a significantly better prognosis (Ps0.0048). for stage I vs. stage II (Ps0.8927), but a marginal difference for stage II vs. stage III (Ps0.0615) and a significant difference for stage III vs. stage IV (Ps0.0012). The small numbers of patients with stage I (ns8) or stage II (ns13) could make the comparison less trustworthy. When overall survivals according to T status were examined, there was significant difference from the median survival of 23 months for T1-2 tumors and of eight months for T3 4 tumors (Ps0.0037). Regarding overall survival classified by N status, the median survival for N0 tumors was 17 months while that for N1-2 tumors was 12 months (Ps0.2016). In a multivariate analysis with various prognostic factors which included gender, age, tumor side, histology, pathologic stage, and surgical procedure, older age (Ps0.0467), non-epithelial histology (Ps0.0057) and pathologic stage III IV disease (Ps0.0019) had a significantly negative impact on survival. Thus, a younger patient with a stage I II epithelial tumor should be a good candidate for radical surgery (Table 2). Lastly, we showed our results of multimodality treatment prospectively performed in recent times (Table 3). Since September 2004, EPP with adjuvant conformal radiotherapy of 54 Gy to ipsilateral hemi-thorax was performed in four patients. The difficulty in providing adjuvant chemotherapy following EPP led us to carry out trimodality therapeutic strategy with neo-adjuvant setting of chemotherapy. Therefore, since May 2006, five patients were treated with two 2 cycles of cisplatin (80 mgym ) and gemcitabine (1000 2 mgym ) followed by EPP and adjuvant conformal hemithorax radiation of 54 Gy. All the patients underwent surgery within six weeks of completion of neo-adjuvant chemotherapy and radiotherapy was initiated within 11 weeks postoperatively. All patients but one were male, and age range was 51 71 years. There were six epithelial type and three mixed type. Interestingly, although all the patients had been preoperatively diagnosed as clinical stage I II disease (N0 disease), pathologic examination postoperatively demonstrated N2 disease in five patients (56%). Table 2 Multivariate analysis of prognostic factors (ns65) Factors Unfavorable Favorable Risk ratio 95% CI P-value Gender Female Male 1.849 0.752 4.546 0.1892 Age (years) )60 F60 1.907 1.009 3.601 0.0467 Side Right Left 1.382 0.688 2.775 0.3626 Histology Non-epi Epi 2.922 1.366 6.249 0.0057 Pathologic stage IIIqIV IqII 2.792 1.459 5.343 0.0019 Procedure PyL EPP 1.498 0.806 2.786 0.2011 CI, confidence interval; Epi, epithelial type; PyL, pleurectomyydecortication; EPP, extrapleural pneumonectomy.

M. Okada et al. / Interactive CardioVascular and Thoracic Surgery 7 (2008) 102 106 105 Table 3 Summary of patients who prospectively had multimodality therapy Age gender Side Histology ctnm ptnm Recurrence Prognosis weppqrtx 55 male Right Epi ct1bn0m0 pt3n2m0 ( ) 38 m alive 66 male Right Epi ct2n0m0 pt2n2m0 (q) 13 m dead (tumor) 63 male Left Epi ct1bn0m0 pt2n2m0 ( ) 27 m alive 58 male Right Mix ct1bn0m0 pt2n0m0 ( ) 24 m alive wctqeppqrtx 62 male Right Epi ct1bn0m0 pt2n0m0 ( ) 19 m alive 61 male Right Mix ct2n0m0 pt3n0m0 ( ) 11 m alive 58 male Left Epi ct2n0m0 pt4n2m0 ( ) 13 m alive 51 male Left Mix ct1bn0m0 pt2n2m0 ( ) 6 m alive 71 female Left Epi ct1bn0m0 pt2n0m0 ( ) 6 m alive 2 2 CT, neoadjuvant systemic chemotherapy (two courses of GEM 1000 mgym qcddp 80 mgym ); EPP, extrapleural pneumonectomy; RT, adjuvant 54 Gy hemithorax radiation; Epi, epithelial type; Mix, mixed type. All patients have completed the prescribed chemotherapy, surgery, and radiotherapy without grade four toxicity or treatment-related fetal complication. All patients but one who received EPP with adjuvant hemi-thorax radiotherapy and died from the tumor 13 months after treatment, are alive without any recurrence. This multimodality therapy is feasible although our result has to be translated judiciously due to the relatively short follow-up length and small number of patients. 4. Discussion The multivariate analysis of this study showed that stage I and II disease had a much more favorable survival than more advanced disease. The results including other series w2, 5x implied a survival advantage for patients with early stage disease undergoing radically surgical resection although a lead time must be always considered. Thus, our data would dismiss our despair over treatment of MPM and support early diagnosis and radical surgery. We, however, must exercise a caution for the potential of confusion between improving prognosis by treatment and by beginning the clock sooner. Our analysis also suggested that epithelial histology had a better prognosis than a nonepithelial one. As the same outcomes were shown in other reports w2, 5x, it should be appropriate to employ the factor as a stratification in clinical trials. Even now, the criteria for selecting patients for radical surgery remain a matter of debate. In this series, we included patients if the thoracic surgeon deemed a macroscopically complete resection possible. Historically EPP had met with a high operative mortality in the order of 10 20%, although it has been reported to be approximately 5% in highly specialized centers w2, 5x. Another surgical procedure PyD, which allows radical resection of the earlystage tumor while preserving the lung parenchyma, is technically difficult but it is generally less morbid and better tolerated than EPP w6 8x. In the present study, the operative mortality was 3.2% for the patients who had an EPP while none of the patients who had a PyD died perioperatively. This series demonstrated that the type of surgery between EPP and PyD never affects the postoperative prognosis with long follow-up, supporting data of the Sloan-Kettering Cancer Center w5x. In our series, it should be noted that relatively early stage disease and epithelial histology were somewhat predominant in patients having PyD. In general, the surgical procedure is inherently stage dependent and potentially driven by surgeon bias. The most crucial matter is which procedure the surgeon should choose if heyshe is confronted with a patient of equal cardiopulmonary demographics that tolerate either EPP or PyD. We favor an en bloc resection of the tumor without entering the pleural cavity by EPP. The surgeon should do what is required to achieve the purpose of the surgery, which is the grade of curability for us, while considering the evaluation of the patient s cardiopulmonary function. As a single modality, surgery or radiation for MPM has individually been disappointing. Addition of aggressive radiotherapy following removal of the whole lung by EPP can be effective for local control of pathology because inadequate local control has been the main cause of death. A phase II trial from Sloan-Kettering of postoperative radiation in high doses indicated a reduction in the local failure rate to 6% without apparent impact on prognosis w9x. Caution should be taken for high doses of radiotherapy after EPP because of severe complications such as pneumonitis, pericarditis and mediastinitis. Whether high doses of conformal or intensity-modulated radiotherapy are feasible and effective for local control following EPP is unknown, but promising. The difficulty in providing aggressive adjuvant chemotherapy after EPP led us to perform a pilot study with neo-adjuvant setting of chemotherapy. While phase II studies showed that the response rate of MPM to cisplatin and gemcitabine has been reported to range between 16% and 47% w10, 11x, a phase III study showed that the response to cisplatin and pemetrexed has been reported to be 41% in advanced MPM w12x. Weare now using the latter combination in a neo-adjuvant approach. Our experience in the nine patients having multimodality therapy showed no in-hospital mortality and perioperative severe complications, providing the basis of a prospective clinical trial which will start in Japan. Radical surgery and hemi-thorax radiotherapy in high doses following neo-adjuvant systemic chemotherapy including cisplatin is a demanding therapeutic strategy and should be reserved

106 M. Okada et al. / Interactive CardioVascular and Thoracic Surgery 7 (2008) 102 106 for dedicated centers with an acceptable mortality and morbidity. References w1x Treasure T, Sedrakyan A. Pleural mesothelioma: little evidence, still time to do trials. Lancet 2004;364:1183 1185. w2x Sugarbaker DJ, Flores RM, Jaklitsch MT, Richards WG, Strauss GM, Corson JM, Decamp MM Jr, Swanson SJ, Bueno R, Lukanich JM, Baldini EH, Mentzer SJ. Resection margins, extrapleural nodal status, and cell type determine postoperative long-term survival in trimodality therapy of malignant pleural mesothelioma: results in 183 patients. J Thorac Cardiovasc Surg 1999;117:54 63. w3x Mimura T, Ito A, Sakuma T, Ohbayashi C, Yoshimura M, Tsubota N, Okita Y, Okada M. Novel marker D2-40, combined with calretinin, CEA and TTF-1: an optimal set of immunodiagnostic markers for pleural mesothelioma. Cancer 2007;109:933 938. w4x Rusch VW. A proposed new international TNM staging system for malignant pleural mesothelioma. From the International Mesothelioma Interest Group. Chest 1995;108:1122 1128. w5x Rusch VW, Venkatraman ES. Important prognostic factors in patients with malignant pleural mesothelioma, managed surgically. Ann Thorac Surg 1999;68:1799 1804. w6x Lee TT, Everett DL, Shu HK, Jahan TM, Roach M 3rd, Speight JL, Cameron RB, Phillips TL, Chan A, Jablons DM. Radical pleurectomyy decortication and intraoperative radiotherapy followed by conformal radiation with or without chemotherapy for malignant pleural mesothelioma. J Thorac Cardiovasc Surg 2002;124:1183 1189. w7x Moskal TL, Dougherty TJ, Urschel JD, Antkowiak JG, Regal AM, Driscoll DL, Takita H. Operation and photodynamic therapy for pleural mesothelioma: 6-year follow-up. Ann Thorac Surg 1998;66:1128 1133. w8x Schouwink H, Rutgers ET, van der Sijp J, Oppelaar H, van Zandwijk N, Burgers S, Stewart FA, Zoetmulder F, Baas P. Intraoperative photodynamic therapy after pleuropneumonectomy in patients with malignant pleural mesothelioma: dose finding and toxicity results. Chest 2001;120:1167 1174. w9x Rusch VW, Rosenzweig K, Venkatraman E, Leon L, Raben A, Harrison L, Bains MS, Downey RJ, Ginsberg RJ. A phase II trial of surgical resection and adjuvant high-dose hemithoracic radiation for malignant pleural mesothelioma. J Thorac Cardiovasc Surg 2001;122:788 795. w10x Nowak AK, Byrne MJ, Williamson R, Ryan G, Segal A, Fielding D, Mitchell P, Musk AW, Robinson BW. A multicentre phase II study of cisplatin and gemcitabine for malignant mesothelioma. Br J Cancer 2002;87:491 496. w11x van Haarst JM, Baas P, Manegold C, Schouwink JH, Burgers JA, de Bruin HG, Mooi WJ, van Klaveren RJ, de Jonge MJ, van Meerbeeck JP. Multicentre phase II study of cisplatin and gemcitabine for malignant pleural mesothelioma. Br J Cancer 2002;86:342 345. w12x Vogelzang NJ, Rusthoven JJ, Symanowski J, Denham C, Kaukel E, Ruffie P, Gatzemeier U, Boyer M, Emri S, Manegold C, Niyikiza C, Paoletti P. Phase III study of pemetrexed in combination with cisplatin vs. cisplatin alone in patients with malignant pleural mesothelioma. J Clin Oncol 2003;21:2636 2644. ecomment: Crucial Japanese evidence of surgery for malignant pleural mesothelioma Author: Noriyoshi Sawabata, Dokkyo Medical University, 880 kitakobayashi Mibu Tochigi 321-0293, Japan doi:10.1510/icvts.2007.166322a Recently, the number of patients with malignant pleural mesothelioma (MPM) has been increasing in Japan. Thus, it is crucial to establish an effective modality for MPM. A phase II trial of multi-modality for MPM will start using Japanese patients. The paper in this journal w1x is the first report showing the feasibility of surgery for MPM using a large number of Japanese patients, offering a referral evidence of the Japanese prospective study. Reference w1x Okada M, Mimura T, Ohbayashi C, Sakuma T, Soejima T, Tsubota N. Radical surgery for malignant pleural mesothelioma: results and prognosis. Interact CardioVasc Thorac Surg 2008;7:102 106. ecomment: EPP and P/D: which to choose? Author: Seiki Hasegawa, Hyogo College of Medicine, 1-1 Mukogawacho, Nishinomiya 663-8501, Japan doi:10.1510/icvts.2007.166322b I read with great interest the article by Okada and others w1x. I believe that this article represents the state of the art of surgical treatment for malignant plural mesothelioma (MPM) in Japan, not only because this is Japan s largest series from a single center, but also the above team is famous for close collaboration of MPM experts. In the present study, there was no significant difference of survival between extrapleural pneumonectomy (EPP) and pleurectomy/decortication (P/D). Okada interprets that the result reflects relatively early stage disease and predominant epithelial histology in P/D group. However, a similar result has been reported from Memorial Sloan-Kettering Cancer Center Group w2x, where more than half of the patients were belonging to unclassified histology and staging information was missing in almost half of the patients. The above facts lead to a hypothesis that EPP and P/D might be equal in view of postoperative median survival. The possible mechanism is that severity of EPP may offset its curative power. If so, EPP should be indicated only for patients who seek for cure, not for extending survival with MPM. It is still unsolved whether P/D is just a limited surgery for patients with poor surgical durability or not. I am interested in the authors current policy for P/D. References w1x Okada M, Mimura T, Ohbayashi C, Sakuma T, Soejima T, Tsubota N. Radical surgery for malignant pleural mesothelioma: results and prognosis. Interact CardioVasc Thorac Surg 2008;7:102 106. w2x Flores RM, Zakowski M, Venkatraman E, Krug L, Rosenzweig K, Dycoco J, Lee C, Yeoh C, Bains M, Rusch V. Prognostic factors in the treatment of malignant pleural mesothelioma at a large tertiary referral center. J Thorac Oncol 2007;2:957 965.