The incidence of malignant pleural mesothelioma

Operation and Photodynamic Therapy for Pleural Mesothelioma: 6-Year Follow-up Thomas L. Moskal, MD, Thomas J. Dougherty, PhD, John D. Urschel, MD, Joseph G. Antkowiak, MD, Anne-Marie Regal, MD, Deborah L. Driscoll, BA, and Hiroshi Takita, MD Division of Surgical Oncology, Department of Radiation Biology, and Department of Thoracic Oncology, Roswell Park Cancer Institute, Buffalo, New York Background. Conventional therapy for pleural mesothelioma has met with disappointing results. Methods. From 1991 to 1996, 40 patients with malignant pleural mesothelioma were treated with surgical resection followed by immediate intracavitary photodynamic therapy. Results. The series included 9 women and 31 men with a mean age of 60 years. Morbidity and treatment-related mortality rates for the entire series, pleurectomy, and extrapleural pneumonectomy were 45% and 7.5%, 39% and 3.6%, and 71% and 28.6%, respectively. Median survival and the estimated 2-year survival rate for the entire series, stages I and II patients (n 13), and stages III and IV patients (n 24) were 15 months and 23%, 36 months and 61%, and 10 months and 0%, respectively. Multivariate analysis identified stage, length of hospital stay, photodynamic therapy dose, and nodal status as independent prognostic indicators for survival. Conclusions. Surgical intervention and photodynamic therapy offer good survival results in patients with stage I or II pleural mesothelioma. For patients in stage III or IV, better treatment modalities need to be developed. Improvements in early detection and preoperative staging are necessary for proper patient selection for treatment. (Ann Thorac Surg 1998;66:1128 33) 1998 by The Society of Thoracic Surgeons The incidence of malignant pleural mesothelioma (MPM) is 2,000 to 4,000 cases annually in the United States. Much of the interest in MPM stems from the fact that conventional treatment options have only a marginal impact on survival and cure. Surgical intervention followed by chemotherapy and radiation therapy has had the most success, increasing median survival to 11 to 22 months compared with 5 to 7 months with supportive care [1 3]. Because of the lack of efficacious treatments, MPM has become a disease model for the development and implementation of new therapies. Photodynamic therapy (PDT) is a surface-oriented, local intervention relying on photochemical-induced cytotoxicity [4 6]. The efficacy of PDT in the treatment of MPM has been described previously [7 12]. In this review, we update our experience [7,8] in the treatment of MPM with operation and PDT. Material and Methods Patients with a histologic diagnosis of MPM were selected. Pretreatment staging, including computed tomographic (CT) scan of the chest and abdomen, was used to identify patients with disease limited to a hemithorax. Presented at the Thirty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 26 28, 1998. Address reprint requests to Dr Takita, Department of Thoracic Oncology, Roswell Park Cancer Institute, Elm and Carlton St, Buffalo, NY 14263. The photosensitizer Photofrin (porfimer sodium) was obtained from Quadra Logic Technologies Phototherapeutics (Vancouver, BC, Canada). Photofrin, 2 mg/kg, was administered intravenously 2 days before scheduled surgical intervention. Patients were instructed about photosensitivity, a side effect of Photofrin that requires avoidance of cutaneous exposure to bright lights, particularly sunlight, for 4 to 6 weeks. Surgical resection of MPM was performed through a posterolateral thoracotomy using combinations of pleurectomy and lung resection including extrapleural pneumonectomy to remove all gross disease or to debulk the tumor to a depth of less than 5 mm. Parallel incisions were sometimes necessary for bulky diaphragmatic disease. The pericardium and the diaphragm, if resected, were reconstructed with synthetic mesh. Mediastinal lymph nodes were routinely sampled. Intracavitary PDT was carried out intraoperatively. The dose was calculated to achieve light penetration to a depth of greater than 5 mm [8]. The surface area to be treated was determined from CT scan sections of the chest with full scaling based on circumference. The lung was excluded from the calculation because it was moved out of the field during treatment. Light was delivered simultaneously through four to six bulbtype fibers (Quadra Logic Technologies Phototherapeutics) spaced centrally within the anterior and the posterior thoracic cavity. The laser source was a 20-W argon dye laser (Spectra-Physics, San Jose, CA) delivering 630-nm light 1998 by The Society of Thoracic Surgeons 0003-4975/98/$19.00 Published by Elsevier Science Inc PII S0003-4975(98)00799-1

Ann Thorac Surg MOSKAL ET AL 1998;66:1128 33 PLEURAL MESOTHELIOMA AND PHOTODYNAMIC THERAPY 1129 to activate the Photofrin. The total light dose administered ranged from 20 to 30 J/cm 2. Staging of patients was based on the 1997 American Joint Committee on Cancer staging system. Follow-up visits were at 4-month intervals and included CT scans of the chest and upper abdomen. Recurrence was based on clinical or radiographic changes. After recurrence, patients were eligible for further palliative treatment. Estimated survival distributions with respect to sex, history of cigarette smoking, history of asbestos exposure, side of lesion, roentgenographic findings (pleural effusion or mass), biopsy technique (open or closed), type of operation, completeness of resection, operative morbidity, transfusion history, palliative chemotherapy or radiation therapy, histology, chest wall muscle invasion, stage, T stage, and nodal status were calculated by the method of Kaplan and Meier [13]. Tests of significance for the survival distributions were based on the log-rank test [14]. Cox s proportional hazards model was used to determine the relationship of age, symptom duration, platelet count, operative time, estimated blood loss, PDT dose, PDT time, and length of hospital stay to survival [14]. A multivariate analysis was done using Cox s model and including all of the factors just listed. Results From April 15, 1991, to December 16, 1996, 43 patients entered this phase II study. Three patients did not receive PDT because at the time of operation, unresectable disease penetrating deep into the mediastinum or diaphragm was discovered; they are excluded from this review. The series included 9 women and 31 men, for a male to female ratio of 3.4:1. The median age was 59 years, and the range was 21 to 77 years. Thirty-nine patients were white, and 1 patient was black. The ratio of involvement of the right side of the chest to the left side of the chest was 2.6:1. Thirty-one patients (78%) had a history of asbestos exposure. Twenty-six patients (65%) were cigarette smokers. Presenting symptoms included dyspnea in 20 patients (50%), chest pain in 13 (33%), cough in 9 (23%), chest wall mass in 1 patient (3%), and weight loss in 1 (3%); 6 patients (15%) were asymptomatic. The median duration of symptoms prior to diagnosis was 4 months (range, 1 to 96 months). Chest roentgenograms revealed a pleural effusion in 31 patients (78%), a pleura-based mass in 6 (15%), and pleural effusion with a pleura-based mass in 3 (8%). Diagnosis was achieved by thoracentesis in 1 patient (3%), percutaneous biopsy in 5 patients (13%), open thoracotomy biopsy in 14 (35%), and videoassisted thoracoscopic biopsy in 20 (50%). Preoperative platelet elevation greater than 450 10 3 /ml was seen in 14 patients (35%), and nonspecific elevations in lactate dehydrogenase and alkaline phosphatase were seen in 9 patients (23%). The median operative time, including PDT delivery, was 8.0 hours (range, 5 to 12 hours). Twenty-eight patients had pleurectomy, 7 had extrapleural pneumonectomy, and 5 had combined pleurectomy and lobectomy. The mean estimated blood loss was 2,513 ml, and 35 patients (88%) required blood transfusions. The PDT dose was 20 J/cm 2 for the first 8 patients, 25 J/cm 2 for the next 17 patients, and 30 J/cm 2 for the most recent 15 patients. The median delivery time for the PDT was 2.0 hours (range, 0.7 to 3.7 hours). Three patients died secondary to treatment. Two of them died on postoperative days 12 and 93 of bronchopleural fistula complicating an extrapleural pneumonectomy. The PDT dose was 20 J/cm 2 and 25 J/cm 2 for these patients. One patient died on postoperative day 39 of empyema complicating a pleurectomy. The PDT dose was 25 J/cm 2. In this review, the 30-day mortality rate was 2.5% (1 patient), and the in-hospital treatment related mortality rate was 7.5% (3 patients). The 30-day mortality and in-hospital treatment related mortality rates for pleurectomy, extrapleural pneumonectomy, and pleurectomy with lobectomy were 0% and 3.6%, 14.3% and 28.6%, and 0% and 0%, respectively. Complications occurred in 18 patients (45%). Fifteen had atrial fibrillation, 11 had sepsis, 10 had respiratory insufficiency requiring tracheostomy, and 3 had bronchopleural fistula after extrapleural pneumonectomy (2 died). Five patients underwent reoperation for spontaneous splenic rupture, diaphragmatic dehiscence, esophageal perforation, empyema (1 died), and diaphragmatic hemorrhage. The morbidity rates for pleurectomy, extrapleural pneumonectomy, and pleurectomy with lobectomy were 39%, 71%, and 40%, respectively. The morbidity rates for PDT doses of 20, 25, and 30 J/cm 2 were 50%, 24%, and 67%, respectively. Pathologic study revealed 25 epithelial (62.5%), 10 biphasic (25%), and 5 sarcomatous (12.5%) subtypes. Fourteen patients (35%) had lymph node involvement, and 17 (43%) had uninvolved lymph nodes; in 9 (23%), lymph node status was unknown. Chest wall muscle invasion was seen in 13 patients (33%). Complete resection of gross disease was accomplished in 16 patients (40%). Postoperative staging was as follows: 12 patients with stage I disease, 10 of whom had complete resection; 1 patient with stage II disease with complete resection; 25 patients with stage III disease, 5 of whom had complete resection and 3 of whom died; and 2 patients with stage IV disease with incomplete resection. Morbidity and mortality rates for patients in stage I or II and patients in stage III or IV were 38% and 0% and 48% and 11%, respectively. The median length of hospital stay, including 2 days before operation for Photofrin infusion, was 14.5 days (range, 9 to 93 days). At the conclusion of this review on February 22, 1997, 28 patients were dead of disease (three were treatment-related deaths), 7 patients were alive with disease, and 5 patients were alive without disease. Seven patients with recurrent disease received palliative chemotherapy, and 7 others with recurrence received palliative radiation therapy. For the survival and prognostic indicator analyses, the 3 patients who died of treatment-related causes were excluded, leaving 37 patients. The median survival and the estimated 2-year survival rates for the entire series (n 37) (Fig 1), stages I and II patients (n 13) (Fig 2), and stages III and IV patients (n 24) (see Fig 2) were 15

1130 MOSKAL ET AL Ann Thorac Surg PLEURAL MESOTHELIOMA AND PHOTODYNAMIC THERAPY 1998;66:1128 33 Fig 1. Kaplan-Meier survival curve for all patients. Fig 3. Kaplan-Meier survival curves for completeness of resection. months and 23%, 36 months and 61%, and 10 months and 0%, respectively. Univariate analysis identified stage (p 0.0001), completeness of resection (p 0.0001) (Fig 3), nodal status (p 0.0002) (Fig 4), T stage (p 0.0003), histology (p 0.02) (Fig 5), preoperative platelet count (p 0.03), chest wall muscle invasion (p 0.04) (Fig 6), and length of hospital stay (p 0.04) as prognostic indicators for survival. Asbestos exposure (p 0.07) and PDT dose (p 0.06) approached significance. Operative time (p 0.15), type of operation (p 0.27), age (p 0.30), sex (p 0.32), history of cigarette smoking (p 0.36), estimated blood loss (p 0.37), biopsy technique (p 0.53), side of lesion (p 0.56), PDT time (p 0.57), further palliative treatment (p 0.59), transfusion history (p 0.77), roentgenographic findings (p 0.82), and morbidity (p 0.85) did not achieve significance. Multivariate analysis identified stage (p 0.0008), length of hospital stay (p 0.0008), PDT dose (p 0.009), and nodal status (p 0.004) as independent prognostic indicators for survival. Asbestos exposure (p 0.05) approached significance. Comment Surgical extirpation is a frequently used treatment of MPM, but in itself, it is not curative [1, 2]. Radiation therapy alone or chemotherapy alone produce little benefit [1, 2]. Therefore, adjunct treatments are being studied to increase survival in these patients. After the initial reports by Pass [9], we have investigated the use of PDT with Photofrin as adjunct treatment after surgical resection in patients with MPM. The intent is to destroy tumor cells remaining postoperatively that may cause tumor regrowth. In this small series, surgical intervention and PDT resulted in median survival of 36 months and overall 2-year survival of 61% in patients with stage I or II Fig 2. Kaplan-Meier survival curves for patients in stages I and II versus patients in stages III and IV. Fig 4. Kaplan-Meier survival curves for nodal status. (Neg negative; Pos positive.)

Ann Thorac Surg MOSKAL ET AL 1998;66:1128 33 PLEURAL MESOTHELIOMA AND PHOTODYNAMIC THERAPY 1131 Fig 5. Kaplan-Meier survival curves for histology. Fig 6. Kaplan-Meier survival curves for chest wall muscle invasion. disease and poor survival in patients with stage III or IV disease. Better outcomes were observed in patients with lower stage, negative nodal status, lower T stage, complete resection, epithelial histology, platelet count less then 450 10 3 /mm 3, absence of chest wall invasion, short hospital stay, higher PDT dose, and no history of asbestos exposure. Treatment had no impact on curative outcomes, with few patients surviving long term. Sugarbaker and associates [3] reported a median survival of 22 months and an overall 2-year survival rate of 50% in 57 stage I patients undergoing extrapleural pneumonectomy followed by chemotherapy and radiation therapy. Rusch and Venkatraman [15] obtained a median survival of 35 months and an overall 2-year survival rate of approximately 60% in 16 stage I patients also treated surgically followed by chemotherapy or irradiation. Pass and coworkers [16] at the National Cancer Institute recently reported that a phase III trial of surgical intervention with immunochemotherapy and randomization to PDT resulted in a median survival of 14 months for both groups. However, this study was terminated early; it accrued only 48 of the required 88 patients, and survival of the few patients with early stage I (n 4) and stage II (n 4) disease was not reported. A major challenge in MPM is effective preoperative staging. In our series, CT scans rarely correctly predicted the extent of the disease. On preoperative staging with CT scan and chest roentgenograms, all 43 patients in our series appeared to have disease limited to one hemithorax. On pathologic staging, only 13 (33%) of 40 patients had early stage I or II disease. Because stage is an important predictor of outcome (see Fig 2), it is critical to be able to stage correctly preoperatively. If surgical intervention and PDT benefit stages I and II patients only, then one third of patients in this series benefited maximally from therapy based on the selection criteria used. At present, we are exploring the use of positron emission tomographic scans to determine if they are more reliable predictors of disease extent than the CT scans. Magnetic resonance imaging has not been useful in the staging of our patients. Thoracoscopy and mediastinoscopy certainly remain as more invasive alternative staging techniques. As complete resection is an important prognostic factor, every attempt was made to remove all gross tumor using combinations of pleurectomy, extrapleural pneumonectomy, and lobectomy. Complete resection was achieved in only 16 (40%) of 40 patients, most of whom were in the earlier stages of disease. Complete resection was possible in 11 (85%) of 13 patients with stage I or II disease. In patients in whom complete resection is not possible, PDT may still be effective in removing residual disease, as was demonstrated in preclinical studies [17] where effective eradication of human mesothelioma xenografts in nude mice was found. It should be noted, however, that much higher light doses were used in these animal studies (120 to 180 J/cm 2 ) than could be achieved in the larger volume of the thoracic cavity in humans. In fact, we may have reached the maximum safe light dose in humans at 30 J/cm 2, as morbidity clearly increases from 20 J/cm 2 to 30 J/cm 2, albeit with an indication of increased survival. When coupled with PDT therapy, operative morbidity affected 45% of patients. Overall mortality was acceptable. The photosensitizer, although relatively preferentially taken up by tumor cells, is also absorbed by normal surrounding tissue, and this may account for the high morbidity in our series. Morbidity increased with higher PDT dose. However, higher PDT dose was also associated with a better outcome. Morbidity and mortality increased significantly for patients undergoing extrapleural pneumonectomy in our series, and for the most part, these patients have not benefited from the combined treatment. The rate of bronchopleural fistula was 43% compared with the 1% to 4% typically described in the literature for pneumonectomy. In fact, the addition of PDT appears to increase morbidity without prolonging survival in this group of patients. Treatment modifications may include flap coverage of the bronchial stump if

1132 MOSKAL ET AL Ann Thorac Surg PLEURAL MESOTHELIOMA AND PHOTODYNAMIC THERAPY 1998;66:1128 33 pneumonectomy is considered in future patients. On the other hand, patients in stage I or II not undergoing pneumonectomy appear to have increased survival (61% survival rate at 2 years [see Fig 2]) compared with patients receiving only surgical resection [1, 2]. Proper light dosimetry is also difficult to achieve in the thoracic cavity because its multiple contours and angles, particularly near the diaphragm, make light application uneven. This aspect of treatment certainly needs improvement. Improvements in the photosensitizer itself are being considered; these include the use of agents absorbing at longer wavelengths to achieve better tissue penetration by the activating light and also the use of agents with decreased photosensitive side effects to reduce patient discomfort and inconvenience. In spite of these drawbacks to our current approach, we are encouraged by what appears to be unexpected long-term survival in a select group of patients. Resections are often piecemeal and require a dedicated surgical and pathology staff to guarantee proper tissue and nodal analysis for accurate pathologic staging. In our review using the 1997 American Joint Committee on Cancer staging guidelines, only 1 patient had stage II disease and 2 patients had stage IV disease. This suggested that our survival results should be analyzed in a stages I and II group versus a stages III and IV group. A new staging system proposed by the International Mesothelioma Interest Group in 1995 takes new T and N stage prognostic factors into account [18]. If applied to our series, 9 of the 40 patients would change stage. The median survival and the estimated 2-year survival rates for stage I (n 7), II (n 6), III (n 20), and IV (n 4) by the International Mesothelioma Interest Group criteria would be 36 months and 67%, 22 months and 27%, 10 months and 8%, and 10 months and 0%, respectively (p 0.01). The median survival and the estimated 2-year survival rate for the International Mesothelioma Interest Group stages I and II and stages III and IV are 36 months and 51% and 10 months and 6%, respectively (p 0.001). These are similar to survivals using American Joint Committee on Cancer staging. The natural history of MPM reveals that two thirds of patients will have local recurrence and one half to three quarters will have recurrence at distant sites [19 21]. Distant recurrence occurs at many sites including liver, adrenal gland, kidney, pancreas, contralateral lung, bone, thyroid, and brain. Clearly, local and systemic control issues must be addressed in the treatment of MPM. Operation and PDT are both local therapies and may explain our good outcomes in early-stage disease. Treatment failure patterns were not consistently documented with postmortem examination in our patient population. Nevertheless, treatment of MPM with operation and PDT should benefit from the addition of systemic therapy. Chemotherapy has been used in multimodality treatment of MPM with success [3, 15, 22]. Other novel therapies such as tumor vaccines or gene therapy may find application as well [23, 24]. Besides attempting to affect cure rates and long-term survival, such local and systemic treatment combinations may also have an impact on patients with later-stage disease. Conventional treatment of MPM consists of surgical intervention, radiation therapy, and chemotherapy. As long as results are disappointing, new treatments will continue to be developed and applied to MPM. Operation and PDT offer good survival for patients with stage I or II MPM. For stage III or IV patients, better treatment modalities need to be developed. Improvements in early detection and preoperative staging are necessary for proper patient selection for treatment. The addition of systemic therapy to this regimen may have an impact on curative treatment, long-term survival, and treatment of later-stage disease. The final decision on optimal treatment for MPM will come from carefully designed phase III trials. References 1. Pass H, Pogrebniak H. Malignant pleural mesothelioma. Curr Probl Surg 1993;30:923 1012. 2. Aisner J. Current approach to malignant mesothelioma of the pleura. Chest 1995;107:332S 44S. 3. Sugarbaker D, Garcia J, Richards W, et al. Extrapleural pneumonectomy in the multimodality therapy for malignant pleural mesothelioma: results in 120 consecutive patients. Ann Surg 1996;224:288 96. 4. Dougherty T, Marcus S. Photodynamic therapy. Eur J Cancer 1992;28A:1734 42. 5. Pass H, Delaney T. Photodynamic therapy. In: Devita V, Hellman S, Rosenberg S, eds. Cancer: principles and practice of oncology. Philadelphia: Lippincott-Raven, 1993:2687 700. 6. Keller S. Photodynamic therapy: biology and clinical application. Chest Surg Clin North Am 1995;5:121 37. 7. Takita H, Mang TS, Loewen GM, et al. Operation and intracavitary photodynamic therapy for malignant pleural mesothelioma: a phase II study. Ann Thorac Surg 1994;58: 995 8. 8. Takita H, Dougherty T. Intracavitary photodynamic therapy for malignant pleural mesothelioma. Semin Surg Oncol 1995;11:368 71. 9. Pass H, Delaney T, Tochner Z, et al. Intrapleural photodynamic therapy: results of a phase I trial. Ann Surg Oncol 1994;1:28 37. 10. Pass H, Donington J. Use of photodynamic therapy for the management of pleural malignancies. Semin Surg Oncol 1995;11:360 7. 11. Ris H, Altermatt H, Nachbur B, et al. Intraoperative photodynamic therapy with m-tetrahydroxyphenylchlorin for chest malignancies. Lasers Surg Med 1996;18:39 45. 12. Baas P, Murrer L, Zoetmulder F, et al. Photodynamic therapy as adjuvant therapy in surgically treated pleural malignancies. Br J Cancer 1997;76:819 26. 13. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457 86. 14. Cox D. Regression models and life tables. J R Stat Soc 1972;34:187 220. 15. Rusch V, Venkatraman E. The importance of surgical staging in the treatment of malignant pleural mesothelioma. J Thorac Cardiovasc Surg 1996;111:815 26. 16. Pass H, Temeck B, Kranda K, et al. Phase III randomized trial of surgery with or without intraoperative photodynamic therapy and postoperative immunochemotherapy for malignant pleural mesothelioma. Ann Surg Oncol 1997;4:628 33. 17. Feins RH, Hilf R, Ross H, Scott G. Photodynamic therapy for human malignant mesothelioma in the nude mouse. J Surg Res 1990;49:311 4. 18. Rusch V. A proposed new international TNM staging system for malignant pleural mesothelioma. Chest 1995;108:1122 8.

Ann Thorac Surg MOSKAL ET AL 1998;66:1128 33 PLEURAL MESOTHELIOMA AND PHOTODYNAMIC THERAPY 1133 19. Ruffie P, Feld R, Minkin Y, et al. Diffuse malignant mesothelioma of the pleura in Ontario and Quebec: a retrospective study of 332 patients. J Clin Oncol 1989;7:1157 68. 20. Baldini EH, Recht A, Strauss GM, et al. Patterns of failure after trimodality therapy for malignant pleural mesothelioma. Ann Thorac Surg 1997;63:334 8. 21. King J, Tucker J, Wong S. Mesothelioma: a study of 22 cases. South Med J 1997;90:199 205. 22. Ong S, Vogelzang N. Chemotherapy in malignant pleural mesothelioma: a review. J Clin Oncol 1996;14:1007 17. 23. Upham J, Garlepp M, Musk A, Robinson B. Malignant mesothelioma: new insights into tumor biology and immunology as a basis for new treatment approaches. Thorax 1995;50:887 93. 24. Albelda S. Gene therapy for lung cancer and mesothelioma. Chest 1997;111:144S 9S. DISCUSSION DR HARVEY I. PASS (Detroit, MI): This report represents the combined efforts of surgeons and bench-work investigators from the Roswell Park Cancer Institute, the cancer center that took the lead in developing clinical photodynamic therapy (PDT). Doctor Moskal and associates are to be congratulated for this contribution. The National Cancer Institute in Bethesda, MD, was also investigating this modality, and after a series of phase I and II trials, we recently published the results of a prospectively randomized trial comparing operation, PDT, and immunochemotherapy with operation and postoperative immunochemotherapy. All patients had resection of disease to a standard residual thickness of no more than 5 mm by extrapleural pneumonectomy or by pleurectomy decortication along with mediastinal nodal dissection. Despite no differences in operative stage, including preoperative volume of disease and lymph node status, the median survival of the 23 patients who did not receive PDT compared with the 22 who did was equivalent at 14 months. This compares quite favorably with the 15-month median survival in this series from the Roswell group, but it also implies that PDT did not have an impact on survival. Moreover, there were no differences in the time to progression or recurrence or progression-free survival whether PDT was used or not. Eight of the 48 patients in our series were in stage I or stage II, and half of them received PDT. These patients had an 87% 2-year survival rate compared with the 61% 2-year survival rate of the Roswell stage I or II patients. The stage I or II patients in our series were among the patients with low-bulk disease on presentation, patients who would be expected to have prolonged survival independent of the operative procedure performed or perhaps in the absence of an operative procedure, if one thinks that this represents only lead-time bias. My questions for Dr Moskal involve when to use PDT in this population, how to reach conclusions about the benefit of PDT, and what the risks of operation for mesothelioma are. Doctor Moskal, your manuscript implies that the patients who may have the greatest benefit from PDT are those in the early stage of disease, as they seemed to do better than historical controls. How did you reach this conclusion given the 61% 2-year survival rate of stage I or II mesotheliomas reported by Rusch and the 70% 2-year survival rate of the node-negative epithelial patients reported by Dr Sugarbaker? In your manuscript, you alluded to a higher rate of complications with higher doses of PDT? However, the two deaths in the 7 patients having pneumonectomy occurred after lower doses. Is this higher than average mortality after pleuropneumonectomy really related to the PDT? How did you define your maximal dose? The manuscript mentions all gross disease and complete resection, and you convincingly showed that the degree of tumor burden reduction is important prognostically. What were your criteria for resection? That is, how did you standardize this from patient to patient? Did you verify the standardization with early postoperative computed tomographic scans? Finally, what technical changes in PDT are necessary to have a more convincing effect on survival and local recurrence in these patients, particularly given that this group of patients has such difficult geometry for PDT? DR MOSKAL: Review of the literature clearly shows that the natural history of pleural mesothelioma is that it is both a local process and a systemic process. Sixty-six percent of patients after treatment will have local recurrence, and 50% to 75% will have recurrence at distant sites. The treatment of surgical intervention and PDT discussed in this phase II study consists of purely local therapy. The only patients who had good survival were those in stage I or II. Patients in stage III or IV had poor survival, and there were no long-term survivors or curative outcomes. An important area of future work with this regimen will be the addition of systemic therapy, such as chemotherapy or tumor vaccine, to attempt to affect distant disease and ultimately survival. Photodynamic therapy has proved to be a good local therapy for a variety of malignancies including skin, esophageal, lung and bladder cancers. Phase III trials with the regimen of operation, PDT, and systemic therapy will have to be accomplished to draw any true conclusions about the benefit of this treatment. Our dose for PDT was titrated from our initial phase I trials. Most of our dosimetric work comes from preclinical preliminary work done by Dr Tom Dougherty in the biophysics department where the goal was to achieve penetrations of 5 mm by PDT into the tissue to be treated. Our 7 patients who underwent pneumonectomy had increased morbidity and mortality. Compared with pneumonectomy plus other treatments in the literature, our rate of three bronchopleural fistulas in these 7 patients is extremely high. This leads us to believe that the PDT may be partly responsible for this morbidity. We are strongly considering the use of prophylactic latissimus dorsi muscle flaps when pneumonectomy is combined with this therapy. Our selection criteria for treatment were preoperative computed tomographic scans showing disease limited to one hemithorax in physiologically fit individuals. After treatment, follow-up computed tomographic scans were performed at 4 month intervals. With these selection criteria, only 13 of 40 patients ultimately had stage I or II disease. If we were going to limit this therapy to stages I and II patients, clearly better selection criteria are necessary. Our institution has had mixed results with magnetic resonance imaging and positron emission tomography in this role. Finally, the thoracic cavity is very difficult to treat with PDT because of the multiple contours and angles. The dose application is naturally uneven in such a cavity. An effective way to achieve a more exact application may ultimately require measuring the absorption of the PDT in the tissue being treated to generate an attenuation coefficient for dose control.