Impact of age on clinicopathological outcomes and recurrence-free survival after the surgical management of nonseminomatous germ cell tumour

Authors


Joel Sheinfeld, Sidney Kimmel Center for Prostate and Urologic Cancers, Memorial Sloan-Kettering Cancer Center, 353 East 68th Street, New York, NY 10021, USA. e-mail: sheinfej@mskcc.org.

Abstract

Study Type – Therapy (case series)

Level of Evidence 4

What's known on the subject? and What does the study add?

The effect of advancing age on the clinicopathological outcomes of men with germ cell testicular cancers remains uncertain.

Through the review and comparison of the present large cohort of men with testis cancer, we report on our experience in men aged ≥50 years. Our results showed similar clinical and pathological characteristics, and survival outcomes that compare favourably with those of men aged <50 years.

OBJECTIVE

  • • To determine the impact of age on clinicopathological findings and disease recurrence in men with nonseminomatous germ cell tumour (NSGCT) undergoing retroperitoneal lymph node dissection (RPLND).

PATIENTS AND METHODS

  • • We identified 1246 patients with NSGCT who underwent either primary or post-chemotherapy-RPLND (PC-RPLND) between 1989 and 2006 from our prospective testis cancer database.
  • • Perioperative characteristics were compared among men aged < or ≥50 years.
  • • Multivariable models were used to evaluate the association of age with disease-free survival, controlling for established clinical and pathological features.

RESULTS

  • • Of 514 men undergoing primary and 732 men undergoing PC-RPLND, 12 (2.3%) and 23 (3.1%) were aged ≥50 years, respectively.
  • • There were no significant differences between men aged < or ≥50 years for perioperative clinicopathological characteristics, with the exception of pre-RPLND CT nodal size.
  • • The pathological distributions at primary RPLND were similar in men aged < or ≥50 years. After PC-RPLND, there were no differences in RPLND histology, number of lymph nodes resected, estimated blood loss, hospital stay, or perioperative complication rate.
  • • Age at surgery was not a significant predictor of disease recurrence when subjected to a multivariable analysis.

CONCLUSIONS

  • • Our data suggests that age at RPLND does not predict for disease recurrence and men aged ≥50 years had similar pre- and postoperative characteristics to those aged <50 years.
  • • We conclude that RPLND can be safely performed in men aged ≥50 years and these patients should be offered optimal treatment regimens for NSGCT as directed according to established guidelines.
Abbreviations
SEER

Surveillance, Epidemiology, and End Results

(NS)GCT

(nonseminomatous) germ cell tumour

(PC-)RPLND

(post-chemotherapy) retroperitoneal lymph node dissection

STM

serum tumour marker

IGCCCG

International Germ Cell Consensus Classification Group

EBL

estimated blood loss

LN

lymph node

RFS

recurrence-free survival.

INTRODUCTION

Testicular germ cell cancer is the most common solid malignancy affecting males aged between 15 and 34 years, and evidence suggests the incidence of testicular cancer is increasing worldwide [1,2]. A small but significant percentage of cases occur in men aged >50 years. In 1999, Steele et al. [3] reported 3-year data from the National Cancer Database where the percentage of men with testicular cancer aged >45 years ranged from 13.4% to 17.3%. While lymphomas and seminomas account for most testis tumours in older men, the incidence of mixed germ cell tumours (GCT) has risen nearly two-fold over the last decade [3].

Over the past half century, improvements in the multidisciplinary approach to the treatment of testicular cancers have resulted in cure rates of >80% depending on stage and histology in modern series [1,4–6]. In older men with GCT the prognosis is less clear. Using Surveillance, Epidemiology, and End Results (SEER) data, Spermon et al. [7] in 2002 reported that patients with testicular cancer who are aged >50 years have significantly lower relative survival rates when compared with younger men. Whether these differences in outcome are due to selection bias, suboptimal approaches to care, or differences in tolerability to treatment due to co-morbid medical conditions in the elderly population, these patients face several barriers and biases that may potentially affect their cancer care.

In the present study, we reviewed our institutional experience in a contemporary series of men undergoing retroperitoneal lymph node dissection (RPLND) for nonseminomatous GCTs (NSGCTs). Our goals were to characterise the effect of age on clinicopathological findings and disease-free recurrence rates. Additionally, we sought to determine the morbidity associated with RPLND in men aged >50 years.

PATIENTS AND METHODS

After Institutional Review Board approval, we identified 1246 men with NSGCT from our prospective testis cancer database between 1989 and 2006, who underwent either primary or post-chemotherapy-RPLND (PC-RPLND) for the management of NSGCT. All information was treated according to the Health Insurance Portability and Accountability Act and was de-identified before analysis. Men with elevated serum tumour markers (STMs) undergoing primary RPLND were excluded (59 men).

The following data were collected prospectively: demographics (date of birth and age at diagnosis), pre-chemotherapy and preoperative clinical characteristics (STMs, clinical stage, orchidectomy date and histology, International Germ Cell Consensus Classification Group [IGCCCG] chemotherapy risk classification, use of second-line therapy, CT retroperitoneal mass size), perioperative details (age at RPLND, estimated blood loss [EBL], hospital stay, number of lymph nodes [LN] resected, pathological features of the RPLND specimen, and complication data), and postoperative data (follow-up imaging and laboratory data, time to recurrence and death). Clinical staging was determined using STM levels after initial orchidectomy, CT 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 were >15.0 ng/mL or β-HCG levels >2.2 U/mL that did not decline according to its predicted serum half-life after orchidectomy. Generally, primary RPLND was recommended for men with normal STM levels and cross-sectional imaging. Men with elevated STM levels after orchidectomy, evidence of retroperitoneal disease >2 cm or extra-retroperitoneal disease were advised to undergo induction chemotherapy with standard platinum-based regimens followed by PC-RPLND and resection of extra-retroperitoneal residual disease depending upon response. Over the study period, the template used for primary and PC-RPLND varied slightly according to surgeon. However, full infrahilar, bilateral template RPLND was the most widely used. All complications occurring ≤30 days of surgery were recorded and graded according to the Memorial Sloan-Kettering Cancer Center (MSKCC) complication grading system:

  • • Grade I, oral medication/bedside care.
  • • Grade II, i.v. therapy, transfusion.
  • • Grade III, intubation, interventional radiology, or re-operative intervention.
  • • Grade IV, organ resection or chronic disability.
  • • Grade V, death.

Grade III–V complications were considered to be high grade. Postoperatively, men 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. Follow-up surveillance included STM levels, CT cross-sectional imaging of the abdomen and pelvis, and either chest CT or chest radiographs. Men were censored if they had a documented radiographic recurrence or rise in STM levels.

The outcomes measured in the present study included clinicopathological characteristics and recurrence-free survival (RFS) in men aged ≥ 50 years undergoing either primary or PC-RPLND. Perioperative characteristics were compared separately among patients aged < or ≥50 years undergoing primary and PC-RPLND using the Fisher's exact test for categorical variables and Mann–Whitney U-test for continuous variables. Pre- and postoperative multivariable models were used to evaluate the association of age with RFS controlling for established clinical and pathological features (Tables 3 and 4). Additionally, a multivariable model was used to evaluate the association of age as a predictor for the presence of LN positivity at primary RPLND. Actuarial relapse-free survival data were analysed using the Kaplan–Meier method.

RESULTS

Of the 1246 men undergoing surgery, 514 and 732 men underwent primary and PC-RPLND, respectively. Of these, 12 (2.3%) and 23 (3.1%) were aged ≥ 50 years, respectively. The median follow-up for men undergoing primary and PC-RPLND were 36.9 and 57.1 months, respectively.

The clinicopathological features for men aged ≥ 50 years after primary RPLND are listed in Table 1. There were no significant differences between men aged < vs ≥50 years for pre- and postoperative clinical characteristics, with exception of pre-RPLND CT LN size > 2 cm (3.3% vs 3/12, P= 0.008). The pathological distributions of disease at primary RPLND were similar between groups for the absence of disease (64% vs 6/12), teratoma (3% vs 1/12), or viable GCT (33% vs 5/12). On multivariable analysis, assessing for predictors of pN+ disease at primary RPLND, clinical stage and embryonal predominance were highly predictive (Table 2). However, age at RPLND failed to predict for pN+ disease (hazard ratio 1.00; 95% CI 0.98, 1.02; P= 0.86). When assessing the effect of age on RFS in men undergoing primary RPLND, age failed to predict for recurrence in the postoperative model (Table 2). Actuarial relapse-free survival at 24 months for men aged < or ≥50 years was 92% and 91%, respectively (P= 0.41).

Table 1.  Perioperative characteristics of men undergoing primary RPLND according to age
VariableAge P
<50 years≥50 years
  1. LVI, lymphovascular invasion; IQR, interquartile range. *Two-sample Wilcoxon rank-sum (Mann–Whitney U-test).

N50212 
N (%) or n/N   
 Clinical Stage  0.08
  Ia344 (68.5)7/12 
  Ib29 (5.8)0 
  IIa113 (22.5)4/12 
  ≥IIb16 (3.2)1/12 
 Embryonal predominance173 (34.5)5/120.76
 Presence of LVI in orchidectomy specimen306 (61.1)5/120.23
 Pre-RPLND CT size (41 patients with missing data), cm  0.008
  ≤2446 (96.7)9/12 
  >215 (3.3)3/12 
 RPLND Stage  0.53
  pN0319 (63.5)6/12 
  pN187 (17.3)3/12 
  ≥pN296 (19.1)3/12 
Median (IQR):   
 Total LNs29 (19,42)23 (18,32)0.24*
 RPLND EBL, mL200 (150,300)200 (100,350)0.98*
Hospital stay, days7 (6,9)7 (7,8)0.60*
Table 2.  Multivariable analysis for (a) predictors of recurrence and (b) predictors for pN+ disease at primary RPLND
VariablesHazard ratio (95% CI) P
  1. pN+ indicated node positivity; RPLND, retroperitoneal lymph node dissection; CI, confidence interval; LVI, lymphovascular invasion.

(a)
RPLND age1.01 (0.98, 1.04)0.64
Adjuvant chemotherapy:  
 noReferent 
 yes0.13 (0.05, 0.40)<0.001
RPLND stage:  
 pN0Referent 
 ≥pN13.60 (1.96, 6.64)<0.001
(b)
RPLND age1.00 (0.98, 1.02)0.86
Clinical stage3.87 (2.59, 5.79)<0.001
Embryonal predominance2.31 (1.53, 3.49)<0.001
LVI1.30 (0.85, 1.98)0.22

For those undergoing PC-RPLND, there were no significant differences in pre- or postoperative clinical characteristics between men aged < vs ≥50 years (Table 3). Most notably, the burden of disease and risk classification at presentation were similar between older and younger men as determined by clinical stage (P= 0.19), pre-chemotherapy CT mass size (P= 0.87), and IGCCCG risk classification (P= 0.76). Using preoperative CT mass size after induction chemotherapy and need for second-line agents as surrogates for chemoresponsiveness of disease, again there were no differences between groups (Table 3). The pathological distributions in men aged < or ≥50 years showed fibrosis in 51% vs 48%, teratoma in 40% vs 39%, and viable GCT in 9% vs 13%, respectively. Postoperative complications occurring ≤30 days after PC-RPLND were also similar between the groups (13.6% for men aged < 50 vs 4.3% for men aged ≥ 50 years, P= 0.40). Gastrointestinal (C. difficile colitis, prolonged ileus, and small bowel obstruction) and wound-related complications (wound infection) were the most frequently encountered adverse events. Additionally, there were no differences in the rate of high-grade complications for men aged < or ≥50 yearse (2.4% vs 0%, respectively). On multivariable analysis, the effect of age on RFS in men undergoing PC-RPLND was negligible, as age failed to predict for disease recurrence after surgery in both the pre- and postoperative models (Table 4). Actuarial RFS at 24 months after PC-RPLND was 87% for all men. When stratified according to age, i.e. < or ≥50 years, actuarial 24-month relapse-free survival was 87% (se 0.01; 95% CI 0.84, 0.90) and 77% (se 0.1; 95% CI 0.50, 0.91), respectively (P= 0.27).

Table 3.  Perioperative characteristics of men undergoing PC-RPLND according to age
VariableAge P
<50 years≥50 years
  • *

    Two-sample Wilcoxon rank-sum (Mann–Whitney). PC-RPLND indicates post-chemotherapy retroperitoneal lymph node dissection; CT, computed tomography; IGCCCG, International Germ Cell Consensus Classification Group; STM, serum tumor markers; GCT, germ cell tumor; LN, lymph node; EBL, estimated blood loss; IQR, interquartile range.

N70923 
N (%):   
 Clinical stage (4 patients missing data):  0.19
  Ia/b52 (7.3)3 (13.4) 
  Is12 (1.7)1 (4.3) 
  IIa73 (10.3)0 
  IIb170 (24.0)5 (21.7) 
  IIc128 (18.1)8 (34.8) 
  III270 (38.1)6 (26.1) 
 Pre-chemotherapy CT size (88 patients with missing data), cm:  0.87
  <2164 (26.2)4 (22.2) 
  2–5264 (42.2)9 (50) 
  >5198 (31.6)5 (27.8) 
 Pre-chemotherapy elevated STMs580 (81.8)20 (87.0)0.78
 IGCCCG risk category (9 patients missing data):  0.76
  Good484 (69.0)14 (63.6) 
  Intermediate104 (14.8)4 (18.2) 
  Poor113 (16.1)4 (18.2) 
 Pre-RPLND CT size (23 patients with data missing), cm:  0.53
  <2390 (56.8)10 (45.5) 
  2–5172 (25.0)7 (31.8) 
  5125 (18.2)5 (22.7) 
 Second-line chemotherapy85 (12.0)3 (13.0)0.75
 Elevated Pre-RPLND STMs98 (13.9)4 (17.4)0.55
 RPLND histology (1 patient missing data):  0.82
  Fibrosis363 (51.3)11 (47.8) 
  Teratoma263 (37.1)9 (39.1) 
  Viable GCT63 (8.9)3 (13.0) 
  Malignant transformation19 (2.7)0 
Median (IQR):   
 Total LNs26 (15,38)34 (14,47)0.29*
 RPLND EBL, mL300 (200,600)300 (200,1500)0.36*
 Hospital stay, days7 (8,10)7 (8.5,10)0.51*
Table 4.  Multivariable analysis for (a) pre-operative and (b) post-operative predictors of recurrence after PC-RPLND
VariablesHazard ratio (95% CI)P
  1. PC-RPLND indicates post-chemotherapy retroperitoneal lymph node dissection; CI, confidence interval; STM, serum tumor markers; IGCCCG, International Germ Cell Consensus Classification Group; CT, computed tomography; GCT, germ cell tumor; TMT, teratoma malignant transformation.

(a)
RPLND age1.01 (0.99, 1.04)0.29
Elevated STMs1.05 (0.61, 1.80)0.86
IGCCCG risk category:  
 GoodReferent
 Intermediate/poor1.30 (0.76, 2.23)0.34
Pre-RPLND CT size, cm:  
 <2Referent 
 2–51.42 (0.78, 2.60)0.25
 >53.53 (2.02, 6.17)<0.001
Clinical stage III2.41 (1.42, 4.10)0.001
Second-line chemotherapy1.75 (0.61, 2.95)0.04
(b)
RPLND age1.01 (0.99, 1.03)0.51
RPLND pathology:  
 Fibrosis/necrosisReferent 
 Teratoma1.68 (1.03, 2.73)<0.04
 Viable GCT/TMT4.50 (2.69, 7.52)<0.001

DISCUSSION

The data from the present study showed that men aged ≥ 50 years had similar pre- and postoperative characteristics as those aged < 50 years. Additionally, age at RPLND in men being treated for NSGCT does not predict for disease recurrence. These findings suggest that while obstacles may exist in the care of men aged > 50 years, the treatment approach to managing NSGCT should still rely upon the basic principles applied to younger men. The data suggests that RPLND can be safely performed in men aged ≥ 50 years and these men should be offered optimal treatment regimens for NSGCT as directed according to established guidelines.

Although predominantly a malignancy of younger men, the incidence of testicular cancer in men aged > 45 years increased from 13.4% in 1985/1986 to 17.3% in 1995/1996 [3]. Over that 10-year period, the incidence of NSGCT in men aged > 45 years nearly doubled, representing 6.4% of all cases of NSGCT reported in the National Cancer Data Base from 1995 to 1996. The population of men aged ≥ 50 years represents a heterogeneous group of men who present several diagnostic and therapeutic challenges related to the presence of co-morbid medical conditions and alterations in baseline organ function. Due to this, these men face several barriers and biases that serve as hindrances to the detection, treatment and recovery of cancer [8]. Misconceptions about the ability of older patients to tolerate chemotherapeutic and/or surgical treatments due to age-dependent alterations in physiology and overall health are held by both physicians and patients alike. These generalisations have been implicated in elderly patients receiving less aggressive or suboptimal treatment based upon chronological age rather than physiological age [9,10]. Nevertheless, these assumptions fail to consider the heterogeneity of the geriatric population and the reality that elderly patients have been shown to tolerate cancer treatment equally well as their middle-aged counterparts [11,12].

The effect of advancing age on the tolerability of men undergoing surgical treatment for both early and advanced staged NSGCT has been implicated to adversely affect clinical outcomes [7]. Spermon et al. [7] reported that stage and morphology adjusted survival rates were significantly lower among men aged > 50 years, despite the similarities in clinical stage of disease at presentation. Using the SEER cancer registry, the 10-year relative survival in men with metastasized NSGCT was 76.9% vs 57.0% for men aged < and >50 years, respectively. The authors suggest these differences may be due in part to inadequate treatment vs difference in tolerability to different therapeutic methods amongst the different age groups.

In the present study, men of advanced age undergoing RPLND for both early and advanced staged disease had remarkably similar preoperative clinical characteristics for burden of disease at presentation compared with younger men. It has been reported that elderly men participate in fewer cancer detection behaviours and therefore are more likely to present when their disease is at a more advanced state [13]. In a Turkish study, Inci et al. [14] subsequently published data showing that differences in survival between men aged < and ≥50 years were due to a higher proportion of the older cohort presenting with metastatic disease (85%) than younger men (47%, P= 0.002). In the present cohort of surgically treated men, the proportion of men that presented with metastatic disease was similar for those aged < 50 years (63.7%) or ≥50 years (60.0%). These finding were corroborated in the National Cancer Data Base of both as the distribution of men aged ≥45 years presenting with advanced staged NSGCT from 1995 to1996 was identical to the entire cohort of men mostly aged < 45 years (34.2% vs 34.7%, respectively) [3]. In men undergoing PC-RPLND, both the burden of residual disease after induction chemotherapy and the pathological incidence of viable disease at surgery in men aged ≥ 50 years were similar to those of younger men. Interestingly, there was no statistical difference in relapse-free survival between groups when stratified by age, i.e. < or ≥50 years (P= 0.27). This implies that for those men undergoing post-chemotherapy resection, the chemosensitivity of disease appears to be unaffected by age. While the true number of men presenting with a more advanced stage or chemo-refractory disease that never progress to surgery is unknown, our institutional experience suggests that this number is negligible. We can state that those selected for surgery represented a similar distribution of stage at presentation and showed similar responses to therapy between age groups. This implies that either intolerance to chemotherapy initially or a suboptimal therapeutic intent may contribute to the worse outcomes seen in the other series.

The second finding noteworthy of discussion showed that the surrogates for perioperative morbidity including operative endpoints of EBL, hospital stay and early postoperative complication rates were the same for men aged < or ≥50 years. While it has been suggested that a dose reduction to chemotherapy occurs in the elderly patients as a result of either toxicity to treatment or for fear of treatment-related toxicity, the percentage of men who received fewer than the intended number of chemotherapy rounds was identical (4.3%) between men aged < or ≥50 years who subsequently underwent PC-RPLND in the present series. Additionally, Inci et al. [14] found no statistical differences in the chemotherapy related complication rates between men aged < 40 years compared with those aged > 50 years (28.9% vs 18.2%, respectively; P > 0.05). We would contend that those who tolerate chemotherapy in general are ‘fit’ enough to proceed to surgery and therefore may indirectly serve as a ‘litmus test’ by selecting for older men who are likely to tolerate the stress of a post-chemotherapy resection.

Lastly, age appears not to adversely affect clinical outcomes in men for RFS after both primary and PC-RPLND. While these results are contrary to the findings of Inci et al. [14] and Spermon et al. [7], this is explainable by differences in management of these men. While Spermon et al. did not state the number of men with advanced stage NSGCT that were treated with post-chemotherapy surgery, the study by Inci et al. reported that post-chemotherapy surgery was carried out only in partial responders, representing 12 of 50 men (24%) receiving chemotherapy for metastatic GCT. Additionally, the surgical technique applied to these men involved resection of radiologically apparent residual masses and macroscopically suspicious neighbouring LNs. It is our approach in these men receiving chemotherapy for NSGCT to routinely undergo a PC-RPLND using a bilateral template, with or without nerve-sparing, in addition to resection of any visceral or extra-retroperitoneal disease identified on post-chemotherapy imaging. Nevertheless, the results of the present study show that when men diagnosed with NSGCT are managed aggressively with a multimodal approach using platinum-based chemotherapy regimens for metastatic disease and surgery for both early and late staged disease according to established guidelines, outcomes can be optimized for men independent of chronological age. The implications of these findings suggest that men who are determined to be ‘fit’ candidates for surgical resection should be managed according to the strict guidelines applied to younger men. Although men who present with testis cancer represent a spectrum of disease, age does not appear to adversely alter the biology of disease presented and these men should be approach with the same convictions applied to younger men to optimise treatment outcomes.

In conclusion, the present data suggests that age at RPLND does not predict for disease recurrence and men aged ≥ 50 years showed similar pre- and postoperative characteristics in all measured outcomes as those aged < 50 years. We conclude that RPLND can be safely performed in men aged ≥ 50 years and these men should be offered optimal treatment regimens for NSGCT as directed according to established guidelines.

CONFLICT OF INTEREST

None declared.

Ancillary