Clinical Impact of Residual Extraretroperitoneal Masses in Patients With Advanced Nonseminomatous Germ Cell Testicular Cancer
Article Outline
Objective
Integration of platinum-based chemotherapy and surgical resection of residual masses is essential in the management of advanced nonseminomatous germ cell tumors (NSGCT). We reviewed our institutional experience in patients undergoing resection of extraretroperitoneal (ERP) residual masses after chemotherapy to assess its impact on cancer progression and survival.
Methods
Between 1989 and 2003, 532 patients with advanced NSGCT underwent postchemotherapy retroperitoneal lymph node dissection (PC-RPLND) with a median follow-up of 41 months. Survival probabilities were estimated by the Kaplan–Meier method. Cox proportional hazards regression analysis was used to determine the prognostic significance of risk factors for progression and survival.
Results
Of 532 patients, 402 (76%) underwent PC-RPLND alone, and 130 (24%) underwent resection of ERP residual disease concurrently or in a staged fashion within 6 weeks. Concordance between retroperitoneal (RP) and ERP sites of disease was 83% in the presence of fibrosis, 42% for teratoma, and 47% for viable NSGCT. Overall, 34% of patients undergoing resection of ERP residual disease had either teratoma or viable disease on final pathology. Five-year probability of freedom from progression was 74% (95% CI 65%, 82%) and disease-specific survival was 84% (95% CI 75%, 89%). On multivariable analysis the histologic findings at the ERP site were significant predictors of disease progression, independent of the RP findings.
Conclusion
Our data suggest that teratoma or viable NSGCT is present in approximately one-third of patients undergoing resection of residual ERP disease. The presence of residual ERP teratoma and viable NSGCT predicts for cancer progression independent of RP histology.
Testicular cancer is the most common cancer in men aged 20-35 years and accounts for approximately 1% of all male malignancies.1 Approximately 30-40% of men with nonseminomatous germ cell tumors (NSGCT) have evidence of metastatic disease at the time of initial presentation.2 Through a multimodal approach to the management of advanced NSGCT incorporating induction chemotherapy followed by postchemotherapy resection of residual disease, cure rates now exceed 80%.3, 4 However, despite advances with platinum-based regimens, up to 35% of patients will have radiographic evidence of extraretroperitoneal (ERP) disease after chemotherapy.5 Because of the varied discordance rates between retroperitoneal (RP) and ERP pathologies ranging from 29-46%6, 7, 8 and limitations in predicting histology, surgical resection of any residual mass identified on postchemotherapy imaging is recommended at our institution. Although outcomes for patients undergoing postchemotherapy retroperitoneal lymph node dissection (PC-RPLND) for residual disease within the RP have been well documented, little is known about the impact residual teratoma or viable germ cell tumor (GCT) within an ERP site has on progression and survival rates.
The purpose of our study was to characterize the clinicopathologic features of patients with advanced NSGCT undergoing surgical resection of ERP residual masses after chemotherapy, and to assess the clinical impact of ERP sites of disease on cancer progression and survival.
Material and Methods
Patients and Clinical Characteristics
After obtaining Institutional Review Board approval, we retrospectively identified 532 consecutive patients with advanced NSGCT who underwent PC-RPLND between 1989 and 2003 from our prospective testis cancer database. All patients underwent induction chemotherapy for metastatic NSGCT, 82 (15%) of whom received additional second-line chemotherapy. Depending on whether there was radiographic evidence of residual ERP disease, patients were categorized according to whether they underwent PC-RPLND alone or in addition to resection of ERP disease. Overall, 130 (24%) patients underwent resection of residual ERP masses within 6 weeks of PC-RPLND.
The following data were collected prospectively: patient demographics (date of birth and age at time of diagnosis), prechemotherapy and preoperative clinical characteristics (serum tumor markers [STM], clinical stage, orchiectomy date and histology, IGCCCG chemotherapy risk classification, use of second-line therapy, computed tomography [CT] RP mass size), perioperative details (estimated blood loss, hospital length of stay, number of lymph nodes resected, and pathologic features of the RPLND specimen), and postoperative data (follow-up imaging and laboratory data, time to progression, and death). Clinical staging was determined using STM levels after initial orchiectomy, CT imaging of the abdomen and pelvis, and either chest CT or chest radiographs. Assessing the burden of metastatic disease was accomplished by measuring the greatest transverse dimension of the largest mass both before chemotherapy and before RPLND. Patients were considered to have elevated STM levels if measurements for α-fetoprotein (AFP) were >15.0 ng/mL or β-human chorionic gonadotropin (β-HCG) levels were >2.2 U/mL and did not decline according to its predicted serum half-life after orchiectomy. Postoperatively, patients were seen monthly for the first year, every second month in year 2, every third month in year 3, every fourth month in year 4, every 6 months in year 5, and annually thereafter.
Statistical Methods
The outcomes measured in this study included probability of freedom from progression and disease-specific survival estimated using the Kaplan–Meier method. Because all patients underwent PC-RPLND with or without resection of ERP disease, time to progression and/or death was estimated beginning from the date of postchemotherapy surgical intervention. Cox proportional hazards regression analysis was used to determine the prognostic significance of risk factors associated with progression. All statistical analyses were performed using Stata (version 10.0, StataCorp, College Station, TX).
Results
Pretreatment Clinical Parameters
Of the 532 men undergoing PC-RPLND, 130 underwent resection of ERP disease either concurrently with PC-RPLND or in a staged fashion within 6 weeks. Overall, significant differences were seen between groups according to IGCCCG classification and stage of disease preoperatively (Table 1). In patients undergoing PC-RPLND alone, only 9% of patients were poor risk using the IGCCCG classification compared with 40% in patients requiring resection of ERP residual disease (P <.001). Similarly, patients undergoing resection of ERP residual masses were more likely to have received second-line chemotherapy (35% vs 9%, P <.001) and have larger median RP CT mass size both prechemotherapy (5.4 vs 4.0 cm, P = .0002) and postchemotherapy (2.5 vs 1.8 cm, P = .02). The most common site of residual ERP disease was localized in decreasing order to the lung (68%), mediastinum (29%), liver (13%), and supraclavicular region (12%). The vast majority of patients had a solitary site of ERP disease (81%), whereas 2 or more sites of residual disease were seen in 19% of patients.
Table 1. Clinicopathologic characteristics of patients undergoing PC-RPLND with or without resection of ERP sites of residual disease after chemotherapy for metastatic NSGCT
| PC-RPLND Alone | PC-RPLND | Overall | P Value | |
|---|---|---|---|---|
| ERP Resection | ||||
n = 402 (%) | n = 130(%) | n = 532(%) | <.001 | |
| IGCCCG risk⁎ | ||||
| Good | 315(79) | 56(43) | 371(70) | |
| Intermediate | 46(12) | 22(17) | 68(13) | |
| Poor | 37(9) | 52(40) | 89(17) | |
| Secondline chemotherapy | 36(9) | 46(35) | 82(15) | <.001 |
| Median (IQ range) | Median (IQ range) | Median (IQ range) | ||
|---|---|---|---|---|
| Prechemo CT mass size (cm) | 4.0(2.2,6.0) | 5.4(3.0,10.0) | 4.0(2.5,7.0) | .002 |
| Postchemo CT mass size (cm) | 1.8(0.9,3.5) | 2.5(1.5,4.5) | 2.0(1.0,4.0) | .02 |
| Retroperitoneal histology | ||||
| Fibrosis | 198(49) | 65(50) | 263(49) | |
| Teratoma | 160(40) | 50(39) | 210(40) | |
| Viable GCT | 44(11) | 15(11) | 59(11) | |
| ERP histology | ||||
| Fibrosis | — | 86(66) | — | |
| Teratoma | — | 31(24) | — | |
| Viable GCT | — | 13(10) | — |
⁎ Data missing in four patients. |
PC-RPLND and ERP Pathologic Characteristics
When RP histologic findings were compared between patients undergoing PC-RPLND with or without resection of ERP residual disease, identical results were observed (Table 1). The incidence of viable GCT and teratoma at sites of residual ERP disease was 10% and 24%, respectively (Table 1). Concordance rates between the RP and ERP sites of disease were 83% in the presence of RP fibrosis, 42% for teratoma, and 47% for viable GCT (Table 2).
Table 2. Concordance of retroperitoneal and ERP histologic findings reported according to retroperitoneal histology of fibrosis, teratoma, and viable GCT
| RP Histology | ERP Histology | ||
|---|---|---|---|
| Fibrosis | Teratoma | Viable GCT | |
| Fibrosis | 83% | 12% | 5% |
| Teratoma | 52% | 42% | 6% |
| Viable GCT | 40% | 13% | 47% |
Clinical Outcome
For patients undergoing resection of ERP residual disease, the 5-year probability of freedom from progression was 74% (95% CI 65%, 81%) and disease-specific survival was 84% (95% CI 75%, 89%). When outcomes were stratified by histology at ERP sites of disease, the 5-year progression-free probability for fibrosis, teratoma, and viable GCT was 93%, 50%, and 8%, respectively (Fig. 1, P <.0001). On multivariable analysis, the histologic findings at the ERP site were significant predictors of disease progression (hazard ratio for teratoma and viable GCT were 7.7 and 25.0, respectively), independent of the RP findings (Table 3). A similar analysis was conducted excluding patients who received second-line chemotherapy and demonstrated equivalent results. Thus, the histologic finding of teratoma or viable GCT at ERP sites of disease confers an inferior prognosis compared with patients with the histologic finding of teratoma or viable GCT confined to the RP.
Figure 1.
Kaplan–Meier curves demonstrating the probability of freedom from recurrence after PC-RPLND and resection of ERP disease stratified according to RP and ERP histology.
Table 3. Cox proportional hazards regression model evaluating histologic predictors of disease recurrence after chemotherapy and surgery for metastatic NSGCT
| Hazard Ratio | 95% CI | P Value | |
|---|---|---|---|
| RP histology | |||
| Fibrosis | Ref | Ref | |
| Teratoma | 0.6 | 0.3-1.5 | .3 |
| Viable GCT | 1.6 | 0.5-4.7 | .4 |
| Extra-RP histology | |||
| Fibrosis | Ref | Ref | |
| Teratoma | 7.7 | 2.8-21.0 | <.001 |
| Viable GCT | 25.0 | 8.2-76.3 | <.001 |
Comment
The results of this study corroborate the discordance rates of histologic findings between RP and ERP sites of disease previously documented in the literature. In addition, although the clinical outcome after chemotherapy and complete surgical resection of all sites of disease resulted in 5-year probability of freedom from progression of 74%, the histologic finding of teratoma or viable GCT at ERP sites conferred a significantly worse prognosis independent of the RP findings. Thus, the histologic finding of teratoma or viable GCT at ERP sites of disease is associated with inferior survival outcomes compared with those with the histologic findings of teratoma or viable GCT confined to the RP.
Metastatic progression beyond the RP in men with advanced NSGCT occurs in as many as one-third of patients.5, 9 The pattern of spread occurs in a stepwise fashion extending from the RP lymph nodes to the mediastinum and chest via lymphatic channels extending up through the crus of the diaphragm. Further hematogenous spread can be seen to the bones, brain, and liver, among other sites. Fizazi and colleagues reported in an international multicenter study on the distribution of postchemotherapy ERP residual masses10 and found that the lung was the most common site of residual NSGCT (52%), followed by the mediastinum (28%) and cervical lymph nodes (7%). Disease in the liver, bone, and brain was identified in 4%, 1%, and 1% of patients, respectively. Similar findings were seen in the current study, with the 2 most common sites of residual disease occurring in the lung (68%) and mediastinum (29%). In both studies, only one site of residual disease was identified and resected in the vast majority of patients (76% and 81%, respectively).
Attempts have been made to predict pathologic findings at ERP sites using preoperative parameters, such as pre- and postchemotherapy mass size and the presence of RP fibrosis at the time of PC-RPLND. Toner et al reported that 6 of 14 patients with pulmonary lesions <1 cm in size undergoing thoracotomy had either teratoma or viable GCT present, indicating that the size of the metastatic lesion does not predict for benign findings after chemotherapy.11 In addition, in patients undergoing bilateral thoracotomies for multifocal disease, 3 of 8 patients demonstrated discordant histologies within each lung. Histologic discordance after induction chemotherapy between RP and ERP sites of disease has been well documented in the literature, with rates ranging from 25% to 47%.6, 7, 8, 9, 11, 12, 13, 14 In the current study, patients identified with fibrosis in the RP were found to have discordant pathologic findings at the ERP site in 17% of patients (12% teratoma, 5% viable GCT). Discordant rates between the RP and ERP sites were much higher for the presence of teratoma (58%) and viable GCT (53%). Collectively, the ability to accurately predict fibrosis at ERP sites of disease remains limited.
Clinical outcomes in patients with residual masses outside the RP are largely defined by the pathology of the ERP site. Multiple investigators have demonstrated that the presence of residual teratoma or viable GCT after chemotherapy at the ERP site was a significant predictor of adverse survival outcomes.9, 15 In a previous study from our institution evaluating the role of thoracotomy in the management of postchemotherapy residual thoracic masses, McGuire et al reported that 2-year disease-free survival was 96% in men with fibrosis on histologic analysis compared with 82% for teratoma and 25% for viable GCT.9 Indiana University reported similar outcomes after the resection of residual mediastinal disease.15 In addition to an elevated preoperative elevated β-HCG, adverse pathologic characteristics remained an independent predictor for worse disease-specific survival after controlling for several known prognostic variables (P = .006).15 These findings have been corroborated when evaluating survival outcomes in patients with nonpulmonary visceral residual disease. At a median follow-up time of 47 months, Hahn and colleagues published survival rates of 89% for fibrosis and 29% for viable GCT in patients undergoing postchemotherapy hepatic resection for residual NSGCT.16
In our study, patients undergoing resection of ERP residual disease had a 5-year progression-free survival of 74% and disease-specific survival of 83%. Regardless of the site of ERP disease, outcomes in men could be stratified according to the histology of the ERP residual disease. Five-year progression-free probability rates for fibrosis, teratoma and viable GCT in the ERP site of disease were 93%, 50%, and 8%, respectively. Interestingly, outcomes were driven largely by the pathology of the ERP site of disease. On multivariable analysis, RP histology failed to predict for disease recurrence in these patients. Only the presence of teratoma or viable GCT in the ERP site of disease was associated with an increased risk of disease recurrence (HR 7.7, 95% CI 2.8-21.0 and HR 25.0, 95%CI 8.2-76.3, respectively) and conferred a worse prognosis in these patients.
Despite analyzing a consecutive cohort of patients, we acknowledge that this study is limited by its retrospective, observational nature and the inherent biases in patient selection and ascertainment of outcomes data that accompanies them. In addition, our study lacks a comparison group of patients with residual teratoma or viable GCT at an ERP site after chemotherapy to help delineate the benefits and morbidity of surgical resection in this clinical setting of advanced NSGCT.
Conclusions
Our data suggest that teratoma or viable GCT is present in approximately one-third of patients undergoing resection of residual ERP disease. The presence of residual ERP teratoma and viable GCT predicts for cancer progression independent of RP histology.
References
- . Germ cell tumors of the testis . Ann Surg Oncol . 2005;12:871–880
- The National Cancer Data Base report on patterns of care for testicular carcinoma, 1985-1996 . Cancer . 1999;86:2171–2183
- VAB-6: an effective chemotherapy regimen for patients with germ-cell tumors . J Clin Oncol . 1986;4:1493–1499
- . Testicular cancer as a model for a curable neoplasm: the Richard and Hinda Rosenthal Foundation Award Lecture . Cancer Res . 1981;41:3275–3280
- Sequential resection of residual abdominal and thoracic masses after chemotherapy for metastatic non-seminomatous germ cell tumours . Br J Cancer . 1994;70:960–965
- The importance of one-stage median sternotomy and retroperitoneal node dissection in disseminated testicular cancer . Ann Thorac Surg . 1983;36:524–528
- Post-chemotherapy tumor residuals in patients with advanced nonseminomatous testicular cancer (Is it necessary to resect all residual masses?) . J Urol . 1991;145:300–302 [Discussion:302-303]
- Sequential excision of residual thoracic and retroperitoneal masses after chemotherapy for stage III germ cell tumors . Cancer . 1986;57:978–983
- The role of thoracotomy in managing postchemotherapy residual thoracic masses in patients with nonseminomatous germ cell tumours . BJU Int . 2003;91:469–473
- Viable malignant cells after primary chemotherapy for disseminated nonseminomatous germ cell tumors: prognostic factors and role of postsurgery chemotherapy—results from an international study group . J Clin Oncol . 2001;19:2647–2657
- Adjunctive surgery after chemotherapy for nonseminomatous germ cell tumors: recommendations for patient selection . J Clin Oncol . 1990;8:1683–1694
- Simultaneous retroperitoneal, thoracic, and cervical resection of postchemotherapy residual masses in patients with metastatic nonseminomatous germ cell tumors of the testis . J Clin Oncol . 1996;14:1765–1769
- Comparison of histological results from the resection of residual masses at different sites after chemotherapy for metastatic non-seminomatous germ cell tumours . Eur J Cancer . 1997;33:843–847
- Jacobsen NE, Beck SD, Jacobson LE, et al. Is retroperitoneal histology predictive of liver histology at concurrent post-chemotherapy retroperitoneal lymph node dissection and hepatic resection? J Urol. ;184:949-953.
- Mediastinal metastases from testicular nonseminomatous germ cell tumors: patterns of dissemination and predictors of long-term survival with surgery . J Thorac Cardiovasc Surg . 2003;125:913–923
- Hepatic resection of metastatic testicular carcinoma: a further update . Ann Surg Oncol . 1999;6:640–644
PII: S0090-4295(11)02468-X
doi:10.1016/j.urology.2011.09.038
© 2012 Elsevier Inc. All rights reserved.
