Difficult Decisions in Thoracic Surgery - part 10 ppt

470 J.R. Sonett many years after it was introduced as a therapeu- tic modality. To help defi ne conclusively the role of thymectomy in nonthymomatous MG, a prospective multi-institutional international trial, approved for funding by the National Institutes of Health (NIH), is planned to randomize patients to thymectomy versus medical treatment begin- ning in 2006. 5 57.1.2. Surgical Approaches to Thymectomy The surgical approaches to thymectomy are varied and refl ect the desire to perform a complete resection weighed against the magnitude and morbidity of the procedure. All approaches enable complete resection of the capsular thymus; what differentiates the approaches are the extent of peri-thymic mediastinal and cervical tissue that are excised. To help understand the different approaches to thymectomy and categorize the extent of resections the Myasthenia Gravis Foun- dation of America (MGFA) has broadly classifi ed varying techniques of resection based on the operative approach and extent of surgical resection (Table 57.1). 6,7 In the ever dynamic surgical fi eld, robotic approaches (T-2 a) as well as bilateral thoracoscopic approaches (T-2 b) are evolv- ing. Overall individual case series have reported data that support the validity and success of all the approaches; however, the lack of prospective, case controlled studies do not provide a signifi - cant level of evidence that one thymectomy technique is superior. 4 Given the lack of defi nitive case controlled and prospective studies, this evidence-based review will highlight selective studies that are reported by established centers in the long-term treatment of MG. All data presented represents level 2 evidence. Additional literature review will examine the failure of thymectomy procedures, morbidity, and results of anatomical studies of the thymic resection. Simple comparison of reported remissions rates and partial remission rates or improvement can be and are misleading when evaluating treatment results. Many patients with MG will improve with time, thus any true refl ection of surgical results should include time after thymectomy. Unfortunately, the majority of the literature does not accommodate for time and are reported as simple crude calculations of remissions (improvement divided by the number of thymic resections). The best method for comparing and understanding results of the literature would be with life table analysis using the Kaplan–Meier method. 8–10 57.1.1.1. Extended Trans-sternal Thymectomy Akira Masaoka 11 of Nagoya University in Japan and Alfred Jaretzki 12 of Columbia University in New York have been amongst the most articulate and persistent leaders in regards to the role extended or complete thymectomy in myasthenia gravis. In 1996, Masaoka and colleagues reported a 20-year review of their experience with extended thymectomy for MG. 11 This procedure involves en bloc resection of the anterior mediastinal fat tissue form phrenic to phrenic laterally and the diaphragm and the thyroid gland caudally and cephalad. All adipose tissues in this region is meticulously resected, including around the bra- chiocephalic veins, thymus, and pericardium. Cervical neck dissection is performed via the sternotomy incision, but aggressive dissection near the recurrent nerves is avoided. In a cohort of 286 patients, remission rates in nonthymomatous MG were 45.8% (5 years), 55.7% (10 years), and 67.2% at 15 years. Similar results have been consistently documented in other series of extended thymectomy. Analysis of multiple pub- lications utilizing extended thymectomy consistently fi nd pathological evidence of thymic tissue within the mediastinal fat out side the capsule of the primary thymus (Table 57.2). 11–16 TABLE 57.1. Myasthenia Gravis Foundation of America (MGFA) thymectomy classification. 6,7 T-1. Transcervical thymectomy a. Basic b. Extended T-2. Videoscopic thymectomy a. VATS b. VATET T-3. Trans-sternal thymectomy a. Standard b. Extended T-4. Transcervical and trans-sternal thymectomy There is insuffi cient evidence to determine which thymectomy technique is superior in the management of myasthenia gravis. 57. Thymectomy for Myasthenia Gravis: Optimal Approach 471 57.1.1.2. Transcervical Thymectomy Basic transcervical thymectomy (T-1a) as an alter- native to trans-sternal thymectomy was introduced on a large scale by Kirschner and colleagues in the late 1960s. 17 However, widespread accep- tance of the procedure only followed the introduc- tion of a more extended and facilitated technique as presented by Cooper: “I do not like to get up and present a paper and look like a blithering idiot by telling people you can take out something through the neck when it is obvious to everybody that it is much easier to take it out through the chest.” 18 Utilizing a sternal retractor to improve visualization and dissection of the thymus as well as peri- thymic fat, a series of 65 patients were presented with a 52% crude complete remission rate. These remission results have been consistently repeated by other groups, including Defi lippi [50% relative risk (RR)], 19 and Calhoun (44% RR), 20 combined with reports of minimal morbidity and an median length of hospital stay of less than 1.5 days. 21 57.1.1.3. Video-Assisted Thorascopic Surgery Thymectomy, Extended Video-Assisted Thora- scopic Surgery Procedures Video-Assisted Thorascopic Extended Thymectomy More recently, the evolution of videoscopic techniques has enabled excellent visualization and minimally invasive techniques for thymic resection. Early results were initially presented by a con- sortium of minimally invasive centers, describing the technique and safe encouraging initial results. Mack and colleagues 22 described 33 thymectomies (either left or right VATS) performed at three insti- tutions with an 18.6% RR at 23 months follow-up. Yim and colleagues recently presented the most comprehensive experience with VATS thymectomy in 38 patients at a single institution. In this limited study, a crude RR (CRR) of 22% was achieved and a 75% CRR was found as measured by Kaplan– Meier survival curve. 23 In an effort to mimic the approach of the maximal thymectomy as described by Jaretzki, Novellino has described the VATET approach 24 : video-assisted thorascopic extended thymectomy, utilizing a small cervical incision and then bilateral thorascopic approach. In a very well- controlled level 2a series presented by Mantegazza, 159 patients underwent VATET, and at 6 years the CSR by life table analysis was 50.6%. 25 57.1.1.4. Morbidity and Failures Results of the evidence-based review by Gorsenth indicate a remarkably low mortality rate for any of the currently used procedures. 4 Peri-operative mortality rates were found to be higher prior to1970, but after that time reported rates were found to consistently less than 1%. Additionally, with present day techniques of extended trans-sternal thymectomy, particularly with special attention to avoid- ance of injury to the recurrent nerves, morbidity rates for the methods are not signifi cantly different. What is clear is that patients undergoing transcervical and thoracoscopic thymectomy procedures can be discharged earlier and have earlier return to daily activities and function. Importantly, limited but important data document the failure of initial thymectomy secondary to retained thymic tissue missed at initial exploration (Table 57.3). 26–29 TABLE 57.2. Extent of thymic tissue recovered in peri-thymic mediastinal fat tissue. Reference Surgical approach Extracapsular thymic tissue Jaretzki 11 Maximal 50 patients (98%) Masaoka 12 Extended 18 patients (72%) Zielinski 13 Extended 58 patients (56.0%) Ashour 14 Extended 38 patients (39.5%) Scelsci 15 VATET 27 patients (37%) Mineo 16 VATS 31 patients (32%) Abbreviations: VATET, video-assisted thorascopic extended thymectomy; VATS, video-assisted thorascopic surgery. TABLE 57.3 Surgical resection of persistent thymic tissue after initial thymectomy. Pathological thymus Myasthenias Reference No. patients Original procedure found at resection improvement Henze 26 20 Transcervical 20/20 19/20 Masaoka 27 6 Transcervical 6/6 3/6 Miller 28 6 Transcervical (3) 5/6 5/6 Basic trans-sternal (3) Rosenberg 29 13 Transcervical 11/13 6/13 Zielinski 13 21 Transcervical (19) 17/21 Not reported Trans-sternal (2) 472 J.R. Sonett 57.2. Summary of Published Data Unfortunately, it is clear that many answers and approaches to the treatment of MG remain unde- fi ned based on a critical analysis of the data. Although there is no level 1 evidence supporting the role of thymectomy in MG, a preponderance of level 2 evidence supports the role of thymectomy in the treatment paradigm of MG. However, recent NIH support for a randomized trial of medical therapy versus thymectomy in the treatment of MG highlights the uncertainty of the evidence to date. In terms of the different surgical approaches to thymectomy, the literature does not defi nitively support any one particular surgical procedure. This must be interpreted in the context of the preponderance of data being reported as crude data in generally small single- center experiences. These equivocal results must be weighed against clear pathological evidence of extracapsular thymic tissue in the majority of patients and limited but defi ned reports of retained thymic tissue being the cause of some initial surgical failures. Thus some form of complete thymectomy should be the goal of any surgical approach, and this has been shown to be feasible by all the approaches described. 57.3. Personal View and Clinical Practice I strongly believe that the evidence to date supports the role of thymectomy in the treatment of MG. This recommendation and practice is bol- stered by the modern day ability to perform the procedure with a very low morbidity and mortality, thus fulfi lling the basic surgical tenant of risk versus benefi t. Given that recommendation and practice, I clearly understand the limits of the data to date, and would support the randomized trial of thymectomy versus medical therapy. But, as with any trial, I would have to bow to some of my biases, and would be reluctant to enter patients into the trial who present with signifi cant respiratory failure. In terms of surgical approach, my bias is toward some type of maximal or extended thymectomy. I believe this can be accomplished best by sternotomy or by bilateral VATS with possible cervical exploration. However, this practice paradigm must be viewed with the understanding that the published results to date do not clearly support any one particular approach and transcervical and unilateral VATS resection are used by many accomplished thoracic surgeons. In the fi nal analysis, the onus is on the thoracic surgical community to investigate the potential surgical benefi t of thymectomy in MG. This benefi t, if proven, will allow us to proceed with further studies to best defi ne the appropriate and perhaps best approaches to resection as well as refi ne indications in terms of symptoms and timing of surgery. I thus would encourage and support the impending trial of thymectomy versus medical therapy in the treatment of MG. References 1. Jaretzki A, Steinglass KM, Sonett JR. Thymectomy in the management of myasthenia gravis. Semin Neurol 2004;24:49–62. 2. Blacock A, Mason MF, Morgan HJ, Riven SS. Myathenias gravis and tumors of the thymic region: report of a case in which the tumor was removed. Ann Surg 1939;110:544–561. 3. Clagett OT, Eaton LM. Surgical treatment of myathenias gravis. J Thorac Surg 1947;16:62–80. 4. Gronseth SG, Barohn RJ. Practice parameter: thymectomy for autoimmune myasthenia gravis (an evidence-based review). Neurology 2000:55: 7–15. 5. Wolfe GI, Kaminski HJ, Jaretzki A III, Swan A, Newsom-Davis J. Development of a thymecotmy trial in nonthymomatous myasthenia gravis patients receiving immunosuppressve therapy. Ann N Y Acad Sci 2003;998:473–480. 6. Jartzki A III, Barohn RJ, Ernstoff RN, et al. Myasthenia gravis: recommendations for clinical research standards. Neurology 2000;55:16–23. 7. MG Task Force. Recommendations for Clinical Research Standards. 2002. Available from: http:// www.myasthenia.org/clinical/research/Clinical_ Research_Standards.htm 8. Masaoka A, Extended trans-sternal thymectomy for myasthenia gravis. Chest Silla Clin Na Am 2001;11:369–387. 9. Jaretzki A III. Thymectomy for myasthenia gravis: an analysis of the controversies regarding technique and results. Neurology 1997;48(suppl 5): S52–S63. 10. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–481. 57. Thymectomy for Myasthenia Gravis: Optimal Approach 473 11. Masaoka A, Nagaoka Y, Kotake Y. Distribution of thymic tissue at the anterior mediastinum. Current procedures in thymectomy. J Thorac Cardiovasc Surg 1975;70:747–754. 12. Jaretzki III, Wolff M. “Maximal” thymectomy for myasthenia gravis. Surgical anatomy and operative technique. J Thorac Cardiovasc Surg 1988;96: 711–716. 13. Zielinski M, Kusdsal J, Szlubowski A, Soja J. Comparison of late results of basic transsternal and extended thymectomies in the treatment of myasthenia gravis. Ann Thorac Surg 2004;78: 253–258. 14. Ashour M. Prevalance of ectopic thymic tissue in myasthenia gravis and its clinical signifi cance. J Thorac Cardiovasc Surg 1995;109:632–635. 15. Scelsi R, Ferro T, Novellino L, et al. Detection and morphology of thymic remnants after video- assisted thorcoscopic extended thymectomy (VATET) in patients with myasthenia gravis. Int Surg 1996;81:14–17. 16. Mineo CT, Pompeo E, Lerut T, Bernardi G, Coose- mans W, Nofroni I. Thoracoscopic thymectomy in autoimmune myastheni: results of left-sided approach. Ann Thorac Surg 2000;69:1537–1541. 17. Krischner PA, Osserman KE, Kark AE. Studies in myasthenias gravis. JAMA 1969;209:906–991. 18. Cooper JD, Al-Jilaihawa AN, Pearson FG, Hum- phrey JG, Humphrey HE. An improved technique to facilitate transcervical thymectomy for myathe- nia gravis. Ann Thorac Surg 1988:45:242–247. 19. DeFilippi VJ, Richman DP, Ferguson MK. Trans- cervical thymectomy for myasthenia gravis. Ann Thorac Surg 1994:57:194–197. 20. Calhoun RF, Ritter JH, Guthrie TJ, et al. Results of transcervical thymectomy for myasthenia gravis in 100 consecutive patients. Ann Surg 1999; 230:555. 21. Ferguson MF. Transcervical thymectomy. Semin Thorac Cardiovasc Surg 1999;11:59–64. 22. Mack MJ, Landreneau RJ, Yim AP, Hazelrigg SR, Scruggs GR. Results of video-assisted thymectomy in patients with myasthenia gravis. J Thorac Cardiovasc Surg 1996;112:1352–1360. 23. Manalulu A, Lee TW, Wan I, Law CY, Chang C, Garzon JC, Yim AP. Video-assisted thoracic surgery thymectomy for nonthymomatous myasthenia gravis. Chest 2005;128:3454–3460. 24. Novellino L, Longoni M, Spinelli L, Andretta M, Cozzi M, Faillace G. Extended thymectomy without sternotomy performed by cervicotomy and thoracoscopic technique in the treatment of myasthenia gravis. Int Surg 1994;79:1378–1381. 25. Mantegazza R, Fulvio B, Bernasconi P, et al. Video- assisted thoracoscopic extended thymectomy and extended transsternal thymectomy (T-3b) in nonthymomatous myasthenia gravis patients: remission after 6 years of follow-up. J Neurol Sci 2003; 212:31–36. 26. Henze A, Biderfeld P, Chrisensson B, Matell G, Pirsanen R. Failing transcervical thymectomy in myasthenis gravis. An evaluation of transternal re-exploration. Scand J Thorac Cardiovasc Surg 1984;18:235–238. 27. Masaoka A, Monden Y, Seike Y, Tanioka T, Kago- tani K. Reoperation after transcervical thymectomy for myasthenias gravis. Neurology 1982;32: 83–85. 28. Miller RG, Filler-Katz A, Kiprov D, Roan R. Repeat thymectomy in chronic refractory myasthenia gravis. Neurology 1991;41:923–924. 29. Rosenberg M, Jauregui WO, De Vewga M, Herrera MR, Roncoroni AJ. Recurrence of thymic hyper- plasia after thymectomy in myasthenia gravis. Its importance as a cause of failure of surgical treatment. Am J Med 1983;74:78–82. 474 58 Management of Residual Disease after Therapy for Mediastinal Germ Cell Tumor and Normal Serum Markers Luis J. Herrera and Garrett L. Walsh cult, and the indication and timing for surgery is tailored to the individual patient and tumor biology. Given the rarity of this disease, the literature consists of retrospective series accumulated over several decades in selected high-volume centers. Due to the lack of controlled trials, defi n- itive recommendations for the management of mediastinal germ cell tumors are based on these small case series only. Furthermore, patient diversity in terms of the extent of disease makes cohort studies or controlled trials diffi cult. This chapter focuses on the management of PMNSGCT, with a focus on the role of surgery for the treatment of residual disease after chemotherapy with normalization of serum tumor markers, based on the best available evidence to date. Other histological types of germ cell tumors often occur in the mediastinum, including teratoma, seminoma, and metastatic gonadal germ cell tumor. This chapter primarily focuses on the management of the primary tumors of the mediastinum of nonseminomatous histology. 58.1. Clinical Evidence: Surgical Management of Primary Mediastinal Nonseminomatous Germ Cell Tumors Primary mediastinal nonseminomatous germ cell tumors are the most malignant subgroup of germ cell tumors, with poor prognosis despite aggressive therapy. PMNSGCT are classifi ed as Primary mediastinal nonseminomatous germ cell tumors (PMNGCT) are rare, representing less than 6% of all germ cell tumors (GCT) and 10% to 20% of all anterior mediastinal masses. 1,2 These tumors can be biologically aggressive, with regional involvement of adjacent structures and a high metastatic potential. The biology of extragonadal GCT is often different than their gonadal counterparts, despite having similar histological features (Table 58.1). 3,4 Due to the aggressive behavior of these tumors, a multimodality approach is the most effective treatment strategy. Controversy still exists regarding the optimal chemotherapy regimen and the timing and indications for surgical intervention. One complex feature of PMNSGCT is the unpredictability of tumor response to induction treatment when based solely on radiographic evaluation and serum tumor marker analysis. In resected specimens after chemotherapy, tumors may exhibit extensive necrosis, teratoma, persistent malignant cells, or malignant transformation, regardless of the serum tumor marker status and the radiographic tumor response in imaging studies. 5–8 Signifi cant advances have occurred in the treatment of germ cell tumors over the past 30 years using multimodality therapy, with high chemotherapy response rates and dramatic improvement in long-term survivors. In most cases, surgical resection of residual disease still plays an important role in the overall management of these patients. The decision to resect residual disease after chemotherapy can be diffi - 58. Management of Residual Disease after Therapy for Mediastinal Germ Cell Tumor and Normal Serum Markers 475 poor prognosis germ cell tumors by the Interna- tional Germ Cell Cancer Collaborative Group consensus classifi cation based solely on the mediastinal location and regardless of any other variable. 9 After confi rmation of the diagnosis with serum tumor markers and, if possible with tumor biopsy, chemotherapy is the fi rst-line treatment modality for these malignancies. Initial surgical resection or debulking of anterior mediastinal NSGCT are not indicted because it rarely achieves complete resection due to the infi ltrative nature of these tumors. This will also have the negative conse- quence of delaying the initiation of chemotherapy. Cisplatin-based chemotherapy is standard induction therapy. First-line therapy usually consists of a combination of cisplatin with etoposide and bleomycin (BEP). 2 The response rates after chemotherapy for PMNSGCT are much lower than for the testicular malignant germ cell tumors. Serum tumor markers (STM) consist of α fetopro- tein (AFP), β-human chorionic gonadotropin (β- HCG), and lactate dehydrogenase (LDH). They are elevated in up to 90% of patients with PMNSGCT. 10 Normalization of STM after chemotherapy occurs in approximately 45% to 90% of patients, with other patients demonstrating persistently elevated tumor markers and persistent disease in the mediastinum. 6,11 Normalization of STM is not necessarily associated with a complete radiographic resolution of the mediastinal mass because persistent viable tumor, residual teratoma, or necrosis can still be present in the mediastinum after induction therapy with marker stabilization or normalization. 8 After chemotherapy, the stage of the disease is reassessed with repeat imaging and STM. Patients may either have: (1) complete radiologic and serologic response; (2) complete serologic response but with a residual mediastinal tumor; (3) growth of the tumor with normalization of STM; or (4) growth of tumor with persistently elevated markers. Surgery is felt to play an important role in groups 2 and 3, but perhaps is less warranted in groups 1 and 4. Surgery can be an adjunct to chemotherapy to achieve a complete response and it can also evaluate the nature and viability of residual masses in order to guide further therapy. In addition, the resection of residual teratomatous elements halts tumor growth and minimizes possible future complications related to growing teratoma syndrome with tumor compression or invasion of vital structures. Because of the rarity of these tumors, no controlled or randomized clinical trials are available and perhaps will never be performed. The literature regarding PMNSGCT consists of case series reviewed retrospectively over decades (Table 58.2). Nevertheless, important points can be gathered from the available literature in order to base clinical decisions. Based on the reported literature, the ideal candidate with PMNSGCT for surgical resection has normalization of STM after fi rst-line chemotherapy, has a residual and resectable mediastinal mass on imaging, has no evidence of extramediastinal metastatic disease, and has good performance and physiologic status (Figure 58.1). Nevertheless, many patients evaluated for surgery after fi rst-line chemotherapy do not fulfi ll these criteria but may still benefi t from surgical resection. Several factors must be considered prior to surgery after the completion of fi rst-line chemotherapy: (1) the radiographic response to chemotherapy; (2) the level of serum tumor markers; (3) the presence of extramediastinal metastatic disease; (4) the extent and resectability of the residual tumor; and (5) the physiological reserve of the patient and estimated morbidity of the planned operation. TABLE 58.1. Pathological classification of primary mediastinal germ cell tumors. Teratomatous tumors Benign Mature teratomas (well differentiated, mature elements; benign) Immature teratomas (immature mesenchymal or neuroepithelial tissue) Malignant Teratoma with additional malignant components (germ cell elements, epithelial cancer, sarcoma) Nonteratomatous tumors Seminomas Nonseminomatous Yolk sac tumors Embryonal carcinomas Choriocarcinomas Mixed nonseminomatous and seminomatous tumor Source: Modified from Moran et al. 4 476 L.J. Herrera and G.L. Walsh Anterior mediastinal mass STM+/– core biopsy PMNSGCT First line platinum based chemotherapy STM: normal STM: elevated, trend down STM: elevated, trend up STM: elevated, no improvement Second line chemotherapy Imaging: CR Observe Recurrence Unresectable disease Salvage chemotherapy Necrosis Teratoma Resection Malignant transformation Viable NSGCT Salvage chemotherapy Unresectable, progressive disease Resectable mediastinal disease Observe Adjuvant chemotherapy Imaging: PR Imaging: SD or PD Growing teratoma FIGURE 58.1. Algorithm for management of PMNSGCT. TABLE 58.2. Summary of published series of PMNSGCT treated with chemotherapy followed by surgery of residual disease. References PMNSGCT (n) Year Patients resected n (%) Preoperative NL STM Overall survival Level of evidence Schneider 15 47 1987–2002 47 (100%) 21 (45%) 3 year, 30% 4 Takeda 19 8 1986–2000 7 (87%) 7 (100%) 5 year, 43% 4 Bokemeyer 11 287 1975–1996 145 (49%) 124 (45%) 5 year, 45% 4 Vuky 7 49 1979–1999 32 (65%) 19 (59%) 2 year, 40% 4 Ganjoo 5 75 1983–1997 62 (82%) 44 (70%) 5 year, 48% 4 Walsh 6 20 1993–1998 11 (55%) 10 (91%) 2 year, 68% 4 Kesler 8 92 1981–1998 79 (86%) 50 (63%) 5 year, 56% 4 Bacha 21 14 1979–1995 6 (43%) 8 (57%) 5 year, 48% 4 Hidalgo 20 27 1978–1995 6 (22%) na 5 year, 31% 4 Lemarie 22 64 1983–1990 22 (49%) na 2 year, 53% 4 Gerl 16 12 1981–1994 12 (100%) na 5 year, 56% 4 Wright 10 28 1976–1988 16 (57%) 22 (78%) 5 year, 57% 4 Abbreviations: na, not available; NL STM, normalization of serum tumor markers; PMNSGCT, primary mediastinal nonseminomatous germ cell tumor; Rec, level of recommendation. 58. Management of Residual Disease after Therapy for Mediastinal Germ Cell Tumor and Normal Serum Markers 477 58.1.1. Radiographic Response to Chemotherapy After completion of chemotherapy, repeat imaging is obtained to reassess the extent of residual disease. In radiographic complete response with no residual tumor, observation alone is indicated. Patients with a partial response and residual resectable tumor can then be considered for surgery, particularly if STM have normalized. For patients with stable disease or disease progression that does not appear to be completely resectable, consideration to further chemotherapy is warranted. 58.1.2. Level of Serum Tumor Markers The impact of STM levels at the time of surgical intervention for patients with PMNSGCT who have had fi rst-line induction chemotherapy has not been well studied, but several case series have illustrated important points in the management of this disease. Vuky and colleagues from Memorial Sloan- Kettering Cancer Center published a retrospective study of 32 patients with PMNSGCT who underwent surgical resection over a 20-year period. 7 After induction chemotherapy, normalization of STM occurred in 19 of the 32 patients (59%), but having an elevated STM level at the time of surgery did not exclude patients for resection. Patients with normal STM had less residual viable tumor (56% vs. 77%). However, in patients with persistently elevated STM, a complete surgical resection was achieved in 10 patients (77%). There was a trend towards decreased survival in patients with increasing STM at the time of surgery compared with patients with STM normalization (p = 0.09). Similarly, in our study at M.D. Anderson Cancer Center, all patients resected postchemotherapy had normalization of STM, and the one patient with persistent elevation had rapid progression of disease postoperatively. 6,7 In another study, Kesler and colleagues reported a retrospective review of 92 patients with PMNSGCT, 79 of whom underwent surgery after platinum-based chemotherapy over a 16- year period. 8 Levels of STM normalized in 50 of the 79 patients (63%), with those patients who had normal levels at the time of resection having decreased incidence of viable NSGCT in the resected specimen when compared with patients with elevated STM (18% vs. 52%). On multivariate analysis, a signifi cantly elevated AFP level (>1000 ng/mL) after fi rst-line chemotherapy showed an associated relative risk of death of 6.5 [95% confi dence interval (95% CI), 1.3–33.2; p = 0.03), however, AFP levels less than 1000 ng/mL had no apparent signifi cant impact on survival. It is unclear from the reviewed literature what is the optimal timing and role of surgery in a patient who has persistent elevation of STM after fi rst-line chemotherapy. Several factors must be considered: (1) the absolute level and trend of STM elevation; (2) the resectability of the residual tumor; (3) the radiographic response; and (4) the feasibility of further chemotherapy cycles or alternate agents. It is important to consider that the outcome of patients treated with salvage chemotherapy due to residual disease after fi rst-line therapy is poor, with long-term survival attain- able in less than 7% of patients. 12 Such dismal results would favor surgical resection of residual tumor in selected patients, despite persistently elevated STM. 58.1.2. Impact of Extramediastinal Metastatic Disease Patients with PMNSGCT often present with metastatic disease outside the mediastinum. As many as 15% to 65% of patients can have distant disease in the liver, bone, spine, brain, and lungs. 6,7,13,14 Intuitively, it would seem that patients with metastatic disease would fare much worse than patients with isolated medi astinal masses, but this has been varably described. In a study by Ganjoo and colleagues, of the 75 patients with PMNSGCT, 19 (25%) had visceral metastasis at the time of pre- sentation. 5 Five-year disease-free survival was 37% for patients with metastatic disease versus 55% for patients without metastases (p = 0.042). Trends towards decreased survival in patients with metastatic disease has been reported in other 478 L.J. Herrera and G.L. Walsh studies, but statistically signi fi cant differences have not been consistently found. 6–8,10,15 Patients with elevated STM after induction therapy who also present with extramediastinal disease present a particular challenge for surgeons. If the mediastinal disease is the most feasible site to resect or if it is causing local compression symptoms, it is reasonable to proceed with resection of the mediastinal tumor to assess tumor viability and guide further therapy for the other extramediastinal lesions. All disease that is amenable to resection, including lung metastases, should be resected subse- quently or concomitantly. In cases of widespread metastatic disease, surgery is at times indicated as a means for tissue procurement to establish the histology of residual disease in order to guide further therapy. The most accessible or the most symptomatic site of disease is surgically resected. If possible, an aggressive approach with resection of metastatic sites is performed if the estimated morbidity is acceptable. 58.1.4. Resectability and Extent of Resection Surgical resection of mediastinal germ cell tumors can be challenging. These tumors tend to develop an intense desmoplastic reaction, obscur- ing all tissue planes, and making safe dissection around vascular structures, lung, and cardiac chambers diffi cult. If the disease is completely resectable, en bloc resection of the mass and any invaded structures is performed, including resection of vascular structures, a phrenic nerve, lung, partial cardiac chambers, and chest wall. At times, en bloc resection of these tumors is not feasible due to encircling of both phrenic nerves, or involvement of multiple mediastinal structures. In some cases, bisecting the tumor allows safer access to the thoracic great vessels for better vascular control and delineation of the anatomy. Some authors recommend four quadrant epicen- ter biopsies with frozen section evaluation, and if no viable tumor is present, near total endole- sional resection with preservation of lung, phrenic nerves, and vascular structures is performed. 8 If at all possible, every effort should be made to preserve lung parenchyma because many of these patients have limited pulmonary reserve secondary to bleomycin toxicity. 58.1.5. Physiological Reserve and Estimated Morbidity A careful physiological evaluation is performed in these patients, who, although young, can have signifi cant compromise in their respiratory function due to chemotherapy-related toxicity. Complete pulmonary function testing including ventilation/ perfusion scans and evaluation of diffusion capacity (DLCO) is necessary. The risk of the planned operation is assessed based on the patient’s performance status, comorbidities, and functional reserve. These patients often develop a persistent postoperative sinus tachycardia that is not related to their volume status, hemoglobin, or pain level which may take several days to resolve. 58.1.6. Prognosis and Impact of Postresection Tumor Histology One of the most interesting aspects of the biology of PMNSGCT is the diversity of histological features and the capacity for cellular transformation after chemotherapy. It has been shown that the histology of the residual mediastinal mass is an important predictor of survival and disease recurrence. The histology in the pathology of the resected masses may reveal necrosis (24%–27%), residual teratoma only (35%–45%), viable NSGCT (10%–26%), or malignant transformation to carcinoma or sarcoma (5%–10%). 5,6,8 Patients with necrosis have an excellent survival (mean, 139 months), compared to an intermediate survival of patients with teratoma (mean, 111 months), and the decreased but still acceptable survival of patients with residual malignant NSGCT (mean, 52 months). Malignant transformation into sarcoma has the worst prognosis with few patients alive past 57 months (Figure 58.2). 8 Current recommendations support the addition of adjuvant chemotherapy for patients with residual viable tumor in the resected specimen consisting of at least two cycles of chemotherapy. The fi nding of malignant transformation to an epithelial histology or to a sarcoma warrants a change in chemotherapy regimens. 58. Management of Residual Disease after Therapy for Mediastinal Germ Cell Tumor and Normal Serum Markers 479 58.2. Current Evidence-Based Management of Primary Mediastinal Nonseminomatous Germ Cell Tumors Overall, PMSCGCT have a poor prognosis when compared with testicular NSGCT; however, important advances have been made in the management of these aggressive malignancies. Due to the rarity of these tumors, few centers have accumulated signifi cant experience with this disease, and prospective trials are not available to gener- ate clear recommendations for treatment. With multimodality therapy, including resection of residual masses after chemotherapy, 5-year survival rates of 30% to 57% can be achieved (Table 58.2). 5–8,11,15–23 Defi nite improvements have been made over the last two decades with the addition of cisplatin-based chemotherapy and surgical resection of residual disease, with much higher rates of long- term survivors. Due to limited number of cases, the basis for current practice is derived from small case series reported to date (Figure 58.1). 58.2.1. Surgical Resection of Residual Tumor after Completion of Initial Chemotherapy Once initial chemotherapy is completed, evaluation of response is performed. There is enough literature available to support the role of surgical resection of residual mediastinal disease after induction therapy; however, the level of evidence is low due to retrospective studies of small number of patients in several series accumulated over many years. Normalization of STM is indicative of a good response and it seems clear that if the disease is resectable, surgery should be performed in physiologically fi t patients with isolated mediastinal tumors (level of evidence 4; recommendation grade C). Patients with STM levels that decrease, but remain elevated, after initial chemotherapy display a trend of decreased survival after resection but some authors still recommend resection due to the low specifi city of STM elevation and the poor results of salvage or second line chemotherapy (level of evidence 4 to 5; insuffi cient data to make a recommendation). 7,8 0 100 200 0.0 0.2 0.4 0.6 Cumulative Survival 0.8 1.0 Necrosis (n = 19) p < 0.01 vs Rest Teratoma (n = 28) p < 0.01 vs GCT & Sarc Carcinoma (n = 3) Persistent GCT (n = 24) Sarcoma (n = 5) Months 2 2 12 18 8 6 FIGURE 58.2. Kaplan–Meier survival curve based on postoperative pathological category. Numbers represent the patients at risk for death. (Reprinted from Kesler KA, Rieger KM, Ganjoo KN, et al. Primary mediastinal nonseminomatous germ cell tumors: the influence of postchemotherapy pathology on long- term survival after surgery. J Thorac Cardiovasc Surg 1999;118:692–700, with permission from Elsevier.) Normalization of serum tumor markers is indicative of a good response to systemic therapy; if the residual disease is resectable, surgery should be performed in physiologically fi t patients with isolated mediastinal tumors (level of evidence 4; recommendation grade C). Patients with serum tumor marker levels that remain elevated after initial chemotherapy display a trend of decreased survival after resection; resection may be appropriate due to the poor results of salvage or second-line chemotherapy (level of evidence 4 to 5; insuffi - cient data to make a recommendation). [...]... results from an international analysis J Clin Oncol 2002;20:1864–1873 12 Saxman SB, Nichols CR, Einhorn LH Salvage chemotherapy in patients with extragonadal nonseminomatous germ cell tumors: the Indiana University experience J Clin Oncol 1994;12:1390– 1393 13 Fizazi K, Culine S, Droz JP, et al Primary mediastinal nonseminomatous germ cell tumors: results of modern therapy including cisplatin-based chemotherapy... listed in the table Although no cases of malignant degeneration have been reported, two complications resulted in mortality The first involved acute hemorrhage into an extrapericardial cyst in an 84-year-old man, acutely compressing the heart into the left hemithorax,12 while the second involved the sudden asystolic death of a 44-year-old man immediately following an exercise stress test Postmortem examination... extragonadal germ-cell tumors – a Southwest Oncology Group Study Cancer 1993;71:2631–2638 19 Takeda S, Miyoshi S, Ohta M, Minami M, Masaoka A, Matsuda H Primary germ cell tumors in the mediastinum – a 50-year experience at a single Japanese institution Cancer 2003;97:367–376 20 Hidalgo M, PazAres L, Rivera F, et al Mediastinal non-seminomatous germ cell tumours (MNSGCT) treated with cisplatin-based combination... tomography imaging is invaluable in planning the operative approach, and in determining which pleural cavity to enter Unlike percutaneous drainage or open subxiphoid drainage, the transthoracic approach requires general anesthesia, and is facilitated by double-lumen intubation Patients 59 Management of Malignant Pericardial Effusions 485 not able to tolerate general anesthesia and singlelung ventilation,... adjunctive surgery in the management of patients with nonseminoma arising from the mediastinum J Clin Oncol 2001;19:682–688 8 Kesler KA, Rieger KM, Ganjoo KN, et al Primary mediastinal nonseminomatous germ cell tumors: the in uence of postchemotherapy pathology on long-term survival after surgery J Thorac Cardiovasc Surg 1999;118:692–700 9 International Germ Cell Consensus Classification: a prognostic factor-based... disease, and the presence of a large (8.5 cm), in amed, intrapericardial cyst in ltrating into the wall of the heart in the region of the conduction system.14 Further examination of Table 60.1 reveals several interesting observations concerning lifethreatening complications of pericardial cysts First, these severe complications may occur at any age, including children, young and middleaged adults, and... disadvantage is the possibility of aseptic necrosis or infection requiring removal of the bone 61.2 Venous Thoracic Outlet Syndrome Venous TOS (VTOS) is subclavian vein obstruction with or without thrombosis The pathology is compression of the subclavian vein at the point where the vein crosses over the first rib to join the innominate vein At this point, the vein is surrounded medially by the costoclavicular... Measurements at baseline and during supine exercise were within normal limits and similar to a cohort of control individuals with normal chests However, during upright exercise, the cardiac index fell significantly below that seen in the normal group When these measurements were repeated 4 months following repair of the chest-wall deformity, the cardiac index during upright exercise had increased 38% compared... or in ammation As stated previously, pericardial cysts normally contain clear, transudative fluid In contrast, in nearly all cases listed in Table 60.1, the cysts 490 R.J Korst TABLE 60.1 Life-threatening complications of pericardial cysts Patient age (years) Gender Cyst size Cyst type Complication/etiology 8 10 12 15 17 21 29 M M F M F M M 7.5 cm na na 8 cm 15 cm 7 cm 5 cm intra extra intra intra intra... Cisplatin-based chemotherapy of primary extragonadal germ cell tumors – a single institution experience Cancer 1996;77:526–532 17 Nichols CR, Saxman S, Williams SD, et al Primary mediastinal nonseminomatous germ cell tumors: a modern single institution experience Cancer 1990;65:1641–1646 18 Bukowski RM, Wolf M, Kulander BG, Montie J, Crawford ED, Blumenstein B alternating combination chemotherapy in patients . conse- quence of delaying the initiation of chemotherapy. Cisplatin-based chemotherapy is standard induction therapy. First-line therapy usually consists of a combination of cisplatin with. unanswered questions remain. Multi-institutional trials may be needed in order to developed a more standardized staging system and better defi ne the role and timing of surgery, in particular for those. with cisplatin-based combination chemotherapy. Ann Oncol 1997;8:555–559. 21. Bacha EA, Chapelier AR, Macchiarini P, Fadel E, Dartevelle PG. Surgery for invasive primary mediastinal tumors.