|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2007.13.6283 on November 12 2007 © 2007 American Society of Clinical Oncology. Improved Clinical Outcome in Recent Years for Men With Metastatic Nonseminomatous Germ Cell Tumors
From the Departments of Urology, Medicine, and Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY Address reprint requests to Brett S. Carver, MD, Department of Urology, Sidney Kimmel Center for Prostate and Urologic Cancers, 353 E 68th St, New York, NY 10021; e-mail: carverb{at}mskcc.org
Purpose The integration of chemotherapy and surgery for metastatic nonseminomatous germ cell tumors (NSGCT) results in survival rates of greater than 80% overall. We evaluated men undergoing postchemotherapy retroperitoneal lymph node dissection (PC-RPLND) for NSGCT to determine associations between year of treatment and clinical outcome. Patients and Methods We evaluated 504 men who underwent PC-RPLND from 1989 to 2002 for NSGCT at our center. Data were obtained from our prospective surgical database and a multivariable logistic regression model was constructed to evaluate variables associated with 15-month relapse in 392 patients with complete data. Results From 1989 to 1997, clinical stage IIa, IIb, IIc, and III NSGCT was seen in 4%, 20%, 23%, and 47% of patients, respectively, compared with 18%, 26%, 11%, and 38%, respectively, from 1998 to 2002 (P < .001). The median prechemotherapy nodal size for 1989 to 1997 and 1998 to 2002 was 5.0 and 3.5 cm, respectively (P < .001). On multivariable analysis, prechemotherapy retroperitoneal nodal size (odds ratio [OR], 1.12; 95% CI, 1.03 to 1.21; P = .005) and presence of visceral metastasis (OR, 2.10; 95% CI, 1.02 to 4.33; P = .04) were significantly associated with 15-month relapse. Men who received a complete RPLND were significantly less likely to experience relapse (OR, 0.22; 95% CI, 0.09 to 0.50; P < .0005). Conclusion In more recent years, men are presenting with less advanced metastatic NSGCT. This stage migration together with effective therapy has resulted in an improved relapse-free survival.
Testicular cancer is the most common malignancy in men age 20 to 35 years and accounts for approximately 1% of all male malignancies.1 It is projected that in the United States, approximately 7,920 new cases of testicular cancer will be diagnosed and 380 men will die as a result of this disease in 2007.2 Through the appropriate use of clinical trials, effective treatment paradigms have been developed for the management of all stages of testicular cancer. The multidisciplinary approach to the management of metastatic nonseminomatous germ cell tumors (NSGCT) has resulted in survival rates of greater than 80% overall.3 During recent years, several noticeable trends have occurred in the presentation of testicular cancer. Although there is considerable geographic variation in the incidence of testicular cancer, with Scandinavia and Switzerland having the highest rates, there has been a worldwide increase in the incidence of testicular cancer.4,5 During a 25-year period the incidence of testicular germ cell tumor (GCT) has nearly doubled. In addition, the appropriate use of serum tumor markers, clinical staging, and risk stratification has had a major impact on the management strategies used in men with GCT.6,7 Patients with persistently elevated serum tumor markers after radical orchiectomy currently are treated with induction chemotherapy, given that these men have occult systemic disease and are not cured by retroperitoneal lymph node dissection (RPLND) alone.6 In 1997, the International Germ Cell Cancer Collaborative Group developed a prognostic model derived from the results of cisplatin-based chemotherapy for the management of metastatic germ cell tumors.7 This model stratifies patients into one of three risk groups based on the probability of achieving a complete response to standard cisplatin-based chemotherapy and has become the cornerstone for developing and evaluating chemotherapy regimens. Finally, an increased appreciation of the clinical entity of late relapse of GCT has emerged in recent years.8-10 Late relapse of GCT is associated with a decreased survival, and despite the appropriate integration of effective cisplatin based chemotherapy and surgery, the incidence of late relapse of GCT seems to be increasing.8-10 In this article, we evaluated men undergoing postchemotherapy (PC) RPLND at our institution to document changes in the clinical presentation of metastatic NSGCT and to determine whether improvement in outcomes is also reflective of better management of these patients.
Patient and Clinical Characteristics From 1989 to 2002, a total of 504 men underwent PC-RPLND at our institution after induction or salvage chemotherapy for metastatic NSGCT. Approximately half were treated from 1989 to 1997 (n = 291; 58%) and the other half were treated from 1998 to 2002 (n = 213; 42%). This study was approved by the institutional review board at Memorial Sloan-Kettering Cancer Center. Clinical and pathologic data were obtained from our prospective surgical database and reported for the described time cohorts. Pre- and postchemotherapy nodal size was determined by the transverse diameter of the largest mass on computed tomography imaging. International Germ Cell Cancer Collaborative Group (IGCCCG) risk classification was assigned based on the previously defined criteria. Men were defined as not receiving IGCCCG risk-appropriate chemotherapy if they received a carboplatin induction chemotherapy regimen, received fewer than the standard number of cycles, or received a good-risk chemotherapy regimen for intermediate- or poor-risk disease. Patients were subcategorized as those who were treated with resection of residual mass alone and those undergoing a complete RPLND. Postchemotherapy RPLND histology was reported as fibrosis, teratoma only, and viable germ cell tumor.
Statistical Methods
To evaluate whether the improvement in progression was secondary to a stage migration, we created a multivariable logistic regression model to evaluate variables associated with 15-month relapse. This model included prechemotherapy RP nodal size and year of surgery as continuous variables, and visceral metastasis (present v absent) and IGCCCG risk category (good v intermediate/poor) as categoric variables. Prechemotherapy
Pretreatment Clinical Parameters Table 1 lists the prechemotherapy clinical parameters in patients undergoing PC-RPLND for metastatic NSGCT. The incidence of advanced-stage NSGCT (IIc and III) at initial presentation has declined significantly over time. From 1989 to 1997, clinical stage IIa, IIb, IIc, and III NSGCT was seen at initial presentation in 4%, 20%, 23%, and 47% of patients, respectively, compared with 18%, 26%, 11%, and 38%, respectively from 1998 to 2002 (P < .001). In addition, prechemotherapy nodal size has decreased significantly over time. The median prechemotherapy nodal size for 1989 to 1997 and 1998 to 2002 was 5.0 and 3.5 cm, respectively (P < .001). Despite patients presenting with less advanced metastatic NSGCT during recent years, the proportion of patients with good-, intermediate-, and poor-risk disease has remained stable.
Multidisciplinary Management Table 2 lists the chemotherapeutic and surgical management of our patients. Over time there has been a standardization of chemotherapy regimens, with the majority of good-risk patients (99%; 1998 to 2002) receiving etoposide and cisplatin for four cycles, or bleomycin, etoposide, and cisplatin for three cycles, and intermediate-risk/poor-risk patients (96%; 1998 to 2002) receiving bleomycin, etoposide, and cisplatin for four cycles or high-dose chemotherapy. Although the proportion of patients with good-, intermediate-, and poor-risk disease has remained stable over time, the requirement of second-line chemotherapy has decreased significantly. From 1989 to 1997, 20% of patients required second-line chemotherapy, compared with only 10% from 1998 to 2002 (P = .014). After chemotherapy, men treated from 1998 to 2002 had smaller residual lymph adenopathy compared with men treated between 1989 and 1997 (median postchemotherapy nodal size, 1.6 v 2.0 cm, respectively). However, the relative reduction in nodal size pre- and postchemotherapy was not significantly different. In addition, surgical management has evolved over time, with a significantly greater percentage of men undergoing a bilateral or full modified PC-RPLND.
PC-RPLND Pathologic Characteristics Table 3 lists the histologic findings at the time of PC-RPLND. The incidence of viable GCT has declined significantly from 15% (1989 to 1997) to 6% (1998 to 2002; P = .001). During recent years there has been a significant increase in the incidence of teratoma in the retroperitoneum (34% in 1989 to 1997 v 47% in 1998 to 2002; P = .012). This increase in teratoma has occurred even though the incidence of teratoma in the orchiectomy specimen remained stable over time. The incidence of teratoma in the orchiectomy specimen was 44% (n = 127) from 1989 to 1997% and 43% (n = 91) from 1998 to 2002 (P = .9).
Clinical Outcome Overall, there has been a significant improvement in relapse rates in recent years. To evaluate whether the improvement in disease-free survival was secondary to improved treatment over time or was the result of a stage migration, a pretreatment multivariable model was constructed. Of the 504 patients in our sample, 72 had insufficient follow-up for our analysis of 15-month relapse and were therefore excluded from this analysis. All 72 excluded patients were observed postoperatively; 34 patients (47%) were observed for more than 6 months. Approximately one half of these patients (n = 37) underwent surgery from 1989 to 1997 and the other half (n = 35) underwent surgery from 1998 to 2003. Complete data for multivariable analyses were available for 392 of 432 patients with sufficient follow-up. Among the 432 patients with sufficient follow-up, 66 patients (15%) experienced relapse within 15 months of surgery. The majority of relapses (52 of 66; 79%) occurred from 1989 to 1997. On univariate analysis, year of surgery was significantly associated with 15-month relapse (odds ratio [OR], 0.88 per year; 95% CI, 0.83 to 0.95; P < .0005; Fig 1). On multivariable analysis, the OR for year of surgery was 0.96, although this was not statistically significant (P = .3) with adjustment for clinical characteristics (Table 4). Prechemotherapy RP nodal size (OR, 1.13; 95% CI, 1.05 to 1.22; P = .001) was significantly associated with 15-month relapse; presence of visceral metastasis (OR, 1.93; 95% CI, 0.94 to 3.94; P = .07) and intermediate/poor IGCCCG risk (OR, 1.93; 95% CI, 0.94 to 3.99; P = .08) predicted poorer survival but this association did not reach statistical significance (Table 4). As a sensitivity analysis, we conducted additional multivariable analyses using a time point of 12 and 18 months from surgery for the binary outcome variable. For both time points, the results did not change substantially. For example, the OR for year of surgery was 0.94 (P = .17) for the end point of 12-month relapse and 0.94 (P = .13) for the end point of 18-month relapse. Prechemotherapy RP nodal size was significantly associated with both 12- and 18-month relapse (both P < .01). Therefore, the decline in relapse rates over time is significantly attributable to improved pretreatment clinical characteristics.
To formally examine treatment characteristics, we examined relapse rates by whether or not a man received IGCCCG risk-appropriate chemotherapy and whether or not a man underwent a complete PC-RPLND. Only 14% of men who received IGCCCG risk-appropriate chemotherapy experienced relapse at 15 months, compared with 26% of men managed with inappropriate chemotherapy (P = .07, Fisher's exact test). Similarly, 12% of men who underwent a complete PC-RPLND experienced relapse at 15 months, compared with 54% of men who had resection of residual mass alone (P < .0005, Fisher's exact test). To evaluate this further, we constructed a multivariable model incorporating the treatment variables (appropriate induction chemotherapy as dictated by IGCCCG risk criteria and the performance of a complete RPLND) in place of year of surgery (Table 4). Although both treatment variables (IGCCCG risk-appropriate chemotherapy and complete RPLND) were associated with an improved freedom from progression, only complete RPLND was statistically significant. In addition, increasing prechemotherapy nodal size and the presence of visceral metastases remained predictors of disease recurrence, highlighting the contribution of the stage migration over recent years in the improved clinical outcome observed. From these results, we analyzed whether the improvement in progression has occurred across all postchemotherapy retroperitoneal histology classifications. To formally test whether the change in progression over time varied according to retroperitoneal histology, we performed an interaction analysis: an interaction between two variables exists if the effect of one variable depends on the level of the other variable. If the interaction term between year of surgery and retroperitoneal histology (fibrosis v teratoma v viable GCT) is statistically significant, then the change in relapse is affected by whether a patient has worse characteristics. Treating fibrosis as the reference category, the interaction term between year of surgery and viable GCT was statistically significant (P = .011). The predicted probability of 15-month relapse by year of surgery is shown in Figure 2 separately for patients with and without persistent viable GCT. We grouped patients with fibrosis and teratoma together in Figure 2 because their results were comparable. Patients with teratoma or fibrosis consistently improved throughout the study period; for example, the predicted 15-month relapse probability decreased from 12% to 6% from 1995 to 2000. In contrast, patients with persistent viable GCT did not improve. It seems that the results for these patients actually get worse; however, the change is well within the 95% CIs and so the risk should be viewed as essentially flat. However, it is plausible that during recent years, patients with persistent viable GCT despite IGCCCG risk-appropriate chemotherapy have biologically more aggressive and chemotherapy-refractory disease compared with patients in the past with viable GCT after substandard chemotherapy regimens, who were initially undertreated and therefore, were more responsive to salvage chemotherapy.
The data suggest that during recent years there has been a stage migration for men with metastatic NSGCT, with an increasing number of men in recent years undergoing initial treatment with less advanced metastatic disease. Our data demonstrate that this stage migration together with effective therapy has resulted in an improved disease-free survival. In addition, these data emphasize the importance of complete PC-RPLND for the management of metastatic NSGCT, given that men undergoing resection of residual retroperitoneal masses alone had a significantly inferior disease-free survival. In a review of the Surveillance, Epidemiology, and End Results program, investigators demonstrated that despite an increasing incidence of testicular cancer in recent years, a greater percentage of white men had localized GCT at the time of diagnosis.11 From 1973 to 1977, only 55% of men were had stage I GCT compared with 73% in the 1998 to 2001 time period. This may be attributed to a heightened patient/physician awareness leading to earlier diagnosis. In addition, a similar increase in localized testicular cancer has been observed in the United Kingdom.12 Furthermore, investigators have demonstrated that there has been a significant decrease in primary tumor size in recent years, declining from a mean of 4 cm (1984 to 1995) to 2.5 cm (1999 to 2002).13 In our current study of men undergoing PC-RPLND, there has been a significant decrease in the incidence of men initially presenting with cIII disease as well as a significant reduction in retroperitoneal tumor volume at diagnosis. The median prechemotherapy retroperitoneal nodal size has decreased significantly from 5.0 cm (1989 to 1992) to 3.5 cm (1998 to 2002), and the incidence of metastatic disease outside the retroperitoneum at time of initial presentation has declined from 47% (1989 to 1997) to 38% (1998 to 2002). This stage migration is possibly the result of improved patient and physician awareness of testicular cancer. Recent studies have demonstrated that there has been an improvement in the delay of testicular cancer diagnosis. In Leeds (United Kingdom), the median time that men took from the onset of symptoms until diagnosis decreased from 5 weeks before 1985 to 2 weeks in recent years.14 The stage migration may also reflect the evolution of clinical criteria for patient selection for initial treatment with induction chemotherapy. We have shown previously that the percentage of patients with positive serum tumor markers and those with clinical stage IIB NSGCT have been referred for induction chemotherapy rather than primary RPLND in recent years.15 We have also demonstrated that the multidisciplinary management for metastatic NSGCT has evolved over time. With the implementation of the IGCCCG risk classification in 1997, standard chemotherapy regimens for the management of metastatic NSGCT have been established. Before this, various risk stratification protocols existed that often led to both undertreatment and overtreatment with chemotherapeutic agents.16-18 Attempts to compare and contrast these previous classifications have demonstrated their varying efficiency in separating patients into good and poor prognostic groups.19,20 In addition, a significantly greater percentage of men are undergoing a complete PC-RPLND rather than resection of residual mass alone. Previous studies have demonstrated the therapeutic importance of adequate retroperitoneal control for both the treatment of metastatic NSGCT and the prevention of late relapse of GCT.9,21,22 Even though IGCCCG risk (assigned at the time of induction chemotherapy) remained stable over time, there has been a significant reduction in the incidence of persistent viable GCT in the retroperitoneum. The incidence of viable GCT at PC-RPLND has decreased significantly from 15% (1989 to 1997) to 6% (1998 to 2002). Although this may be due in part to the standardization of effective chemotherapy regimens that has occurred in recent years, the reduced volume of metastatic disease associated with the stage migration likely contributes to these results. In addition, the histologic finding of teratoma in the retroperitoneum has increased significantly. This may be explained in part by the reduction in viable GCT; approximately 40% of patients with viable GCT will have concomitant chemotherapy-refractory teratomatous elements. We have demonstrated that in recent years there has been a reduction in disease progression for men undergoing PC-RPLND for metastatic NSGCT. Based on our multivariable model, this improvement appears in part to be secondary to a stage migration, given that men in more recent years are presenting with a lower volume of disease before treatment. Importantly, we have demonstrated that improvements in the surgical management of men with metastatic NSGCT, which have occurred in recent years, also contribute in part to the improved outcome observed, emphasizing the importance of complete PC-RPLND. Our data suggest that the greatest improvement in clinical outcome has occurred in those with the histologic finding of fibrosis or teratoma at PC-RPLND. Thus the retroperitoneal histologic finding of fibrosis or teratoma is not a surrogate end point; these patients benefit the most from stage migration and appropriate multidisciplinary management. Therefore, the converse is also true: in the presence of inadequate chemotherapy or surgery, these patients are at a significant risk for unnecessary recurrences.
The author(s) indicated no potential conflicts of interest.
Conception and design: Brett S. Carver, Angel M. Serio, Dean Bajorin, Robert J. Motzer, George J. Bosl, Andrew J. Vickers, Joel Sheinfeld Administrative support: Dean Bajorin, Robert J. Motzer, George J. Bosl, Joel Sheinfeld Provision of study materials or patients: Brett S. Carver, Jason Stasi, Joel Sheinfeld Collection and assembly of data: Brett S. Carver, Angel M. Serio, Jason Stasi, Andrew J. Vickers, Joel Sheinfeld Data analysis and interpretation: Brett S. Carver, Angel M. Serio, Dean Bajorin, Robert J. Motzer, George J. Bosl, Andrew J. Vickers, Joel Sheinfeld Manuscript writing: Brett S. Carver, Angel M. Serio Final approval of manuscript: Brett S. Carver, Angel M. Serio, Dean Bajorin, Robert J. Motzer, Jason Stasi, George J. Bosl, Andrew J. Vickers, Joel Sheinfeld
published online ahead of print at www.jco.org on November 12, 2007. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Carver BS, Sheinfeld J: Germ cell tumors of the testis. Ann Surg Oncol 12:871-880, 2005[CrossRef][Medline] 2. American Cancer Society: Cancer Facts and Figures 2007. http://www.cancer.org 3. Sheinfeld J, Herr HW: Role of surgery in management of germ cell tumor. Semin Oncol 25:203-209, 1998[Medline] 4. McKiernan J, Goluboff E, Liberson G, et al: Rising risk of testicular cancer by birth cohort in the United States from 1973 to 1995. J Urol 162:361-363, 1999[CrossRef][Medline] 5. Purdue MP, Devesa SS, Sigurdson AJ, et al: International patterns and trends in testis cancer incidence. Int J Cancer 115:822-827, 2005[CrossRef][Medline] 6. Rabbani F, Sheinfeld J, Farivar-Mohseni H, et al: Low-volume nodal metastases detected at retroperitoneal lymphadenectomy for testicular cancer: Pattern and prognostic factors for relapse. J Clin Oncol 19:2020-2025, 2001 7. IGCCCG: International Germ Cell Consensus Classification: A prognostic factor-based staging system for metastatic germ cell cancers. J Clin Oncol 15:594-603, 1997 8. Carver BS, Motzer RJ, Kondagunta GV, et al: Late relapse of testicular germ cell tumors. Urol Oncol 23:441-445, 2005[Medline] 9. Baniel J, Foster RS, Gonin R, et al: Late relapse of testicular cancer. J Clin Oncol 13:1170-1176, 1995[Abstract] 10. George DW, Foster RS, Hromas RA, et al: Update on late relapse of germ cell tumor: A clinical and molecular analysis. J Clin Oncol 21:113-122, 2003 11. McGlynn KA, Devesa SS, Graubard BI, et al: Increasing incidence of testicular germ cell tumors among black men in the United States. J Clin Oncol 23:5757-5761, 2005 12. Powles TB, Bhardwa J, Shamash J, et al: The changing presentation of germ cell tumours of the testis between 1983 and 2002. BJU Int 95:1197-1200, 2005[CrossRef][Medline] 13. Bhardwa JM, Powles T, Berney D, et al: Assessing the size and stage of testicular germ cell tumours: 1984-2003. BJU Int 96:819-821, 2005[CrossRef][Medline] 14. Vasudev NS, Joffe JK, Cooke C, et al: Delay in the diagnosis of testicular tumours: Changes over the past 18 years. Br J Gen Pract 54:595-597, 2004[Medline] 15. Stephenson AJ, Bosl GJ, Motzer RJ, et al: Retroperitoneal lymph node dissection for nonseminomatous germ cell testicular cancer: Impact of patient selection factors on outcome. J Clin Oncol 23:2781-2788, 2005 16. Bosl GJ, Geller NL, Cirrincione C, et al: Multivariate analysis of prognostic variables in patients with metastatic testicular cancer. Cancer Res 43:3403-3407, 1983 17. Medical Research Council: Prognostic factors in advanced non-seminomatous germ-cell testicular tumours: Results of a multicentre study—Report from the Medical Research Council Working Party on Testicular Tumours. Lancet 1:8-11, 1985[Medline] 18. Stoter G, Sylvester R, Sleijfer DT, et al: Multivariate analysis of prognostic variables in patients with disseminated non-seminomatous testicular cancer: Results from an EORTC multi-institutional phase III study. Int J Androl 10:239-246, 1987[Medline] 19. Bajorin D, Katz A, Chan E, et al: Comparison of criteria for assigning germ cell tumor patients to "good risk" and "poor risk" studies. J Clin Oncol 6:786-792, 1988 20. Hitchins RN, Newlands ES, Smith DB, et al: Long-term outcome in patients with germ cell tumours treated with POMB/ACE chemotherapy: Comparison of commonly used classification systems of good and poor prognosis. Br J Cancer 59:236-242, 1989[Medline] 21. McKiernan JM, Motzer RJ, Bajorin DF, et al: Reoperative retroperitoneal surgery for nonseminomatous germ cell tumor: Clinical presentation, patterns of recurrence, and outcome. Urology 62:732-736, 2003[CrossRef][Medline] 22. Muramaki M, Hara I, Miyake H, et al: Clinical outcome of retroperitoneal lymph node dissection after induction chemotherapy for metastatic non-seminomatous germ cell tumors. Int J Urol 11:763-767, 2004[CrossRef][Medline] Submitted July 20, 2007; accepted September 12, 2007.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
x months after PC-RPLND and were considered to have experienced treatment success if they were relapse-free at x months after PC-RPLND; patients with less than x months of follow-up and without a relapse were deemed to have insufficient follow-up and were excluded from this analysis. In choosing a time for x, we wanted to select a time point such that few patients would be excluded due to insufficient follow-up, and also such that few patients experienced relapse after the time point. We chose x = 15 months for our main analysis because the majority of relapses (66 of 84) occurred by this time and only 72 patients (14%) had insufficient follow-up; to determine whether our choice of time point influenced our findings, we performed sensitivity analyses for x = 12 and x = 18 months. 
-fetoprotein and human chorionic gonadotropin levels were not entered in the multivariable model because IGCCCG risk category is based on these levels. Statistical analyses were conducted using Stata 9.0 (Stata Corp, College Station, TX). P values 
