SEARCH

SEARCH BY CITATION

Keywords:

  • surveillance;
  • observation;
  • renal neoplasm;
  • nephrectomy;
  • kidney failure;
  • chronic;
  • glomerular filtration rate;
  • comorbidity

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

BACKGROUND:

Although nephrectomy cures most localized renal cancers, this oncologic benefit may be outweighed by the renal functional costs of such an approach. In this study, the authors examined overall survival in 537 patients who had localized renal tumors ≤7 cm detected at age ≥75 years to investigate whether surgical intervention improved survival compared with active surveillance.

METHODS:

Clinical T1 renal tumors were managed with surveillance (20%), nephron-sparing interventions (53%), or nephrectomy (27%). Cox regression models were constructed based on age, comorbidity, management type, renal function, and other variables.

RESULTS:

The median follow-up was 3.9 years, and death from any cause occurred in 148 patients (28%). The most common cause of death was cardiovascular (29%), and cancer progression was responsible in only 4% of deaths. Kaplan-Meier analysis revealed decreased overall survival for patients who underwent surveillance and nephrectomy relative to nephron-sparing intervention (P = .01); however, surveilled patients were older and had greater comorbidity. In multivariate analysis, significant predictors of overall survival included age (P = .0004) and comorbidity (P < .0001) but not management type (P = .3). Preoperative renal function (P = .006) and comorbidity (P = .005) were predictors of cardiovascular mortality, and nephrectomy was associated with greatest loss of renal function.

CONCLUSIONS:

In patients aged ≥75 years, surgical management of clinically localized renal cortical tumors was not associated with increased survival. Patients died mostly of cardiovascular causes, similar to the general elderly population. Nephrectomy accelerated renal dysfunction, which was associated with cardiovascular mortality. Current paradigms suggest that there is over treatment of localized renal tumors, and further study will be required to evaluate the advisability of various options in patients with limited life expectancy. Cancer 2010. © 2010 American Cancer Society.

Renal cell carcinoma (RCC) is the most lethal of the common genitourinary malignancies. Approximately 25% of patients with RCC present with metastases, and another 25% of patients with localized cancers develop recurrent disease during follow-up. Surgical treatment of suspected renal cancer is the mainstay of therapy, and systemic therapy is reserved for advanced disease.1 Over the last decade, the incidence of RCC has risen by 2.3% per year, disproportionately reflecting an increase in the incidental detection of small renal tumors with noninvasive abdominal imaging.2 Although early detection has increased over the last decade, cancer-specific mortality has not decreased,3 perhaps indicating that increased detection and earlier treatment may not be improving the survival of patients with RCC.

Notably, the greatest increase in the incidence of renal cancer has been observed in the later years of life.4 Thus, as longevity continues to increase in the developed world, an increasing number of older patients are being considered for interventions for incidentally detected renal masses suspected to be RCC. The majority of clinical stage I RCCs will never recur after intervention, and benign pathology is identified in approximately 20% of renal surgeries for enhancing renal tumors.1, 5 In fact, incidental renal tumors have been detected in up to 1.2% of patients at autopsy, the majority of whom died with cancer, not of cancer.6 The effect of diagnosis and management of renal tumors with low predicted oncologic potential on overall survival is largely unknown. Therefore, we examined the overall and cause-specific survival of patients who had a localized renal tumor (≤7 cm) detected at age ≥75 years to investigate whether surgical intervention improved survival compared with active surveillance.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Cohort

Institutional review board approval was obtained for the use of data maintained within our institutional kidney cancer patient registry. Nine hundred seventy-nine patients aged ≥75 years were seen at our institution for a suspected renal mass between January 2000 and December 2006. Patients who did not have a lesion that was suspicious for renal cancer after our review (n = 208 patients) and patients who underwent nephrectomy for other reasons (n = 24 patients) were excluded from the current analysis. For patients who had bilateral tumors, the index tumor was selected based on higher stage and/or greater size. Patients who underwent nephrectomy for clinical stage T2 or greater renal cancer (n = 105 patients) or for upper tract urothelial carcinoma (n = 105 patients) were used as separate cohorts for comparison with the study group.

Therefore, the study population consisted of the 537 patients who had suspected clinical stage T1 renal cancer at initial evaluation. Management in this cohort of patients included radical nephrectomy (n = 146 patients; 27%), nephron-sparing intervention (n = 286 patients; 53%), and active surveillance (n = 105 patients; 20%). The initial treatment modality for patients who underwent nephron-sparing surgery included radiofrequency ablation (n = 26 patients), cryoablation (n = 46 patients), laparoscopic partial nephrectomy (n = 95 patients), and open partial nephrectomy (n = 119 patients); radical nephrectomy was either laparoscopic (n = 121 patients) or open (n = 25 patients).

Management and Follow-Up

The initial management was selected by the patient and treating physician after they considered tumor size; radiographic appearance; overall patient health and life expectancy; available treatment options, including nephrectomy, nephron-sparing intervention (partial nephrectomy or tumor ablation), and active surveillance; and patient and surgeon preference. Comorbidity was evaluated using the Charlson-Romano index.7 Patients who elected active surveillance were followed clinically every 6 months with renal imaging. Follow-up in surgical patients was tailored according to pathologic cancer stage. Cancer recurrence was determined based on clinical and radiographic findings. For each patient, vital status was obtained using the Social Security Death Index (SSDI) with follow-up through April 2008,8 and cause of death information was determined by reviewing the patient's medical records and information obtained from the National Death Index (NDI) with follow-up through December 2005.9 Cancer-specific mortality was attributed to patients with evidence of cancer progression before death and/or the following codes from the International Classification of Diseases 9th and 10th revisions (ICD-9 and ICD-10, respectively) on physician-prepared death certificates (ICD-9 code 189.0, ICD-10 code C64). Cardiovascular deaths included deaths that were attributable to coronary artery disease (ICD-9 codes 410, 411, 413, and 414; ICD-10 codes I20-I25, I46, and I51), congestive heart failure (ICD-9 codes 398, 402, and 428; ICD-10 codes I11 and I50), hypertensive renal disease (ICD-10 codes I12 and I13), cerebrovascular disease (ICD-9 codes 433, 434, and 436; ICD-10 codes I63, I64, and I69), atherosclerosis/peripheral vascular disease (ICD-9 code 440; ICD-10 codes I70, I71, and I73), or as indicated on physician-prepared death certificates, as described previously.10

Statistical Analyses

Wilcoxon/Kruskal-Wallis tests were used to compare nonparametric continuous data, and chi-square/Fisher exact tests were used to compare nominal data according to management type. The Kaplan-Meier method was used to evaluate survival, and differences among management types were tested with the log-rank test. Multivariate analyses evaluating the association between management type and all-cause mortality while adjusting for prespecified clinical characteristics (age, sex, race, Charlson-Romano index, presence of solitary kidney, preoperative glomerular filtration rate [GFR]) were performed using Cox proportional hazard models. Associations are provided as hazard ratios (HRs) and their 95% confidence intervals (CIs). Post-hoc analyses of tumor histology and predicted oncologic potential (classified as benign, indolent, or potentially aggressive, as described previously)5 and interactions between age, comorbidity, and management type were performed in an attempt to identify patients who might benefit from active treatment. All analyses for this study were performed with S-Plus 7.0 (Insightful Corporation, Seattle, Wash) and SAS 9.0 (SAS Institute, Cary, NC). All P values <.05 were considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

The characteristics of patients who underwent nephrectomy, nephron-sparing intervention, and active surveillance for suspected clinical stage T1 renal cancer are listed in Table 1. Significant differences in almost every pretreatment variable indicated that treatment type was not used evenly in these patients. For example, the patients who underwent active surveillance were older and had more comorbid conditions than patients in the active treatment groups. The median follow-up was 3.9 years, and 148 patients (28%) died during this period. Cancer recurrence was detected in 34 patients (6.3%), including 22 patients with deaths attributable to kidney cancer, 4 patients who remained alive with distant metastases, 2 patients who had metachronous localized renal cortical tumors, and 6 patients with suspicion of persistence after tumor ablation. Metachronous renal cortical tumors were identified after initial laparoscopic partial nephrectomy in a contralateral and in bilateral kidneys in 1 patient each and were managed with surveillance. Cancer persistence after tumor ablation was managed with repeat radiofrequency ablation (n = 4 patients), repeat cryoablation (n = 1 patient), or surveillance (n = 1 patient).

Table 1. Characteristics of Patients Aged ≥75 Years With Clinical T1 Renal Cancer (≤7 cm) According to Type of Management
CharacteristicNo. of Patients (%)P
All Patients, n=537Active Surveillance, n=105Nephron-Sparing Intervention, n=286Radical Nephrectomy, n=146
  • IQR indicates interquartile range; GFR, glomerular filtration rate; CKD, chronic kidney disease; RCC, renal cell carcinoma.

  • a

    Kruskal-Wallis test.

  • b

    Chi-square test.

  • c

    The initial GFR was unknown for 8 patients, and the ultimate GFR was unknown for 15 patients.

  • d

    Defined as any new renal parenchymal lesion (excluding index tumors that were managed with active surveillance).

Median age [IQR], y79 [76-82]81 [78-86]78 [76-80]79 [77-83]<.0001a
Men340 (63)58 (55)200 (70)82 (56).003b
Race    <.0001b
  Caucasian470 (88)77 (73)259 (91)134 (92) 
  African-American50 (9.3)24 (23)20 (7)6 (4.1) 
  Other17 (3.2)4 (3.8)7 (2)6 (4.1) 
Median Charlson comorbidity index [IQR]1 [0-2]2 [1-3]1 [0-2]1 [0-2]<.0001a
  Score ≥2224 (42)68 (65)111 (39)47 (32)<.0001b
Hypertension322 (60)65 (62)183 (64)74 (51).026b
Median initial GFR [IQR], ml/min/1.73m2c62 [47-75]54 [41-67]63 [46-75]64 [55-76].0008a
  CKD stage ≥3236 (45)53 (54)126 (44)57 (39).07b
Median radiographic size [IQR], cm3.3 [2.4-4.4]2.3 [1.6-3.3]3.0 [2.3-3.9]4.6 [3.5-5.8]<.0001a
  Clinical T1b159 (30)15 (14)58 (20)86 (59)<.001b
Solitary kidney59 (11)9 (9)46 (16)4 (3).0001b
Bilateral renal involvement49 (9)5 (5)35 (12)9 (6).026b
Postoperative data     
 Histology    <.0001b
  Conventional RCC207 (39)2 (1.9)117 (41)88 (60) 
  Other cancer109 (20)0 (0)80 (28)31 (21) 
  Benign112 (21)4 (3.8)79 (28)27 (18) 
  Unknown109 (20)99 (94)10 (3.5)0 (0) 
 Predicted oncologic potential    <.0001b
 Potentially aggressive207 (39)0 (0)33 (12)58 (40) 
  Indolent109 (20)2 (2)171 (60)61 (42) 
  Benign112 (21)4 (4)79 (28)27 (18) 
  Unknown109 (20)99 (94)3 (1)0 (0) 
 Mortality from any cause148 (28)41 (39)66 (23)41 (28).007b
  Cardiovascular43 (8)13 (12)17 (5.9)13 (8.9).10
  Cancer-specific22 (4.1)6 (5.7)9 (3.1)7 (4.8).14
 Cancer recurrence34 (6.3)6 (5.7)20 (7)8 (5.5).98b
  Death attributed to kidney cancer22 (4.1)6 (5.7)9 (3.1)7 (4.8) 
  Alive with distant metastasis4 (0.7)0 (0)3 (1.0)1 (0.7) 
  Alive with local recurrence only8 (1.5)0 (0)d8 (2.8)0 (0) 
 Median follow-up in living patients [IQR], y3.9 [2.7-5.5]3.7 [2.4-5.1]3.9 [2.7-5.6]4.4 [2.8-5.5].36a
 Median ultimate GFR [IQR], mL/min/1.73m246 [34-62]51 [36-67]48 [35-63]41 [29-51]<.0001a
  CKD stage ≥3375 (72)56 (59)194 (69)125 (86)<.0001b
 Percentage loss in renal function [IQR]21 [5-37]0 [0-19]18 [5-31]34 [25-47]<.0001a

Overall Survival

Kaplan-Meier curves demonstrate that overall survival differed between the 3 groups (Fig. 1). The unadjusted Kaplan-Meier estimates of overall survival at 5 years for patients with clinical stage T1 renal cancer was 72% (95% CI, 63%-80%) with nephrectomy, 76% (95% CI, 69%-81%) with a nephron-sparing intervention, and 58% (95% CI, 46%-69%) with active surveillance. For patients with clinical T1a disease (confined; ≤4 cm) and clinical T1b disease (confined; 4.1-7.0 cm) disease, the Kaplan-Meier estimate of overall survival at 5 years was 74% (95% CI, 68%-79%) and 66% (95% CI, 56%-74%), respectively. In comparison, the Kaplan-Meier estimate of overall survival at 5 years for patients who had more advanced renal cancer (clinical stage ≥T2) and upper tract urothelial carcinoma was 51% (95% CI, 40%-61%) and 42% (95% CI, 31%-52%), respectively.

thumbnail image

Figure 1. These are Kaplan-Meier estimates of overall survival for patients aged ≥75 years who had localized renal tumors and who were managed with radical nephrectomy, nephron-sparing intervention, or active surveillance.

Download figure to PowerPoint

To evaluate the effect of management type on overall survival while controlling for potential confounding covariates within the 3 treatment groups, multivariate Cox proportional hazard analysis was performed (Table 2). Comorbidity was a significant predictor of mortality, with an HR of 1.33 (95% CI, 1.20-1.48) for every unit increase in the Charlson-Romano index (P < .0001). Age also was a significant predictor of mortality, with an HR of 1.34 (95% CI, 1.11-1.60) for every 5-year increase (P = .002). Although they were associated significantly with overall survival in univariate analysis, neither management type (P = .22), tumor size (P = .16), nor preoperative GFR (P = .68) was associated with overall survival after adjusting for age, comorbidity, and the other variables.

Table 2. Cox Proportional Hazards Model of All-Cause Mortality
VariableUnivariate AnalysisMultivariate Analysis
HR (95% CI)PHR (95% CI)P
  • HR indicates hazard ratio; CI, confidence interval; GFR, glomerular filtration rate.

  • a

    P values for multivariate pair-wise comparisons: radical nephrectomy versus surveillance, P = .28; nephron-sparing intervention versus surveillance, P = .08.

Management type .0021 .22a
 Radical nephrectomy0.64 (0.42-0.99) 0.75 (0.45-1.26) 
 Nephron-sparing intervention0.50 (0.34-0.73) 0.67 (0.42-1.05) 
 Active surveillance1.00 1.00 
Age, per 5-y increase1.36 (1.16-1.60).00021.34 (1.11-1.60).002
Men0.94 (0.67-1.30).690.86 (0.61-1.22).40
Race .25 .64
 Caucasian1.99 (0.63-6.24) 1.65 (0.52-5.26) 
 African American2.68 (0.78-9.22) 1.83 (0.52-6.50) 
 Other1.00 1.00 
Charlson comorbidity index, per 1-unit increase1.35 (1.23-1.48)<.00011.33 (1.20-1.48)<.0001
Clinical size, per 1-cm increase1.13 (1.02-1.25).021.09 (0.97-1.22).16
Solitary kidney1.10 (0.70-1.76).681.16 (0.69-1.93).58
Bilateral renal involvement0.86 (0.49-1.52).611.12 (0.61-2.04).72
Initial GFR, per 10 mL/min/1.73m2 decrease1.11 (1.03-1.20).00791.02 (0.93-1.11).68

In the current series, only 13 patients underwent a renal mass biopsy (2.3%) before initial management, reflecting current practice patterns.11 To investigate the impact of information that potentially could be obtained from renal mass biopsy, we added histologic tumor type (clear cell RCC, other cancer, or benign) and predicted oncologic activity (potentially aggressive, indolent, or benign) based on adding final surgical pathology to the model. Neither was associated significantly with overall survival after accounting for all other variables (P = .75 and P = .61, respectively).

To investigate the hypothesis that intervention might benefit patients who had the longest predicted longevity, we evaluated potential interactions between age, comorbidity, and management type. No interactions were identified between age and management type or between age and comorbidity, indicating that these effects were constant across the range of potential values. An interaction between comorbidity and management type was present (P = .036), suggesting a benefit to nephron-sparing intervention only for patients who had minimal comorbidity, but the concordance index for a model incorporating the interaction term (0.684) was no more robust than the base model (0.684).

Cardiovascular Mortality

Cardiovascular events were the leading cause of death in this patient population, as would be expected if the patients did not have a malignancy, accounting for 29% of all deaths and 42% of deaths from identifiable causes. Kaplan-Meier curves demonstrated that the cumulative incidence of cardiovascular mortality exceeded that of cancer-specific mortality in each treatment group (Fig. 2). Univariate and step-wise multivariate analyses identified initial GFR (HR, 1.25; 95% CI, 1.07-1.45; P = .006) and Charlson score (HR, 1.3; 95% CI, 1.1-1.6; P = .005), but not management type (P = .28), as significant predictors of cardiovascular death. The Kaplan-Meier estimate of the incidence of cardiovascular death at 5 years was 19% (95% CI, 11%-28%) for patients with stage III or greater chronic kidney disease (CKD) before management compared with 7.4% (95% CI, 2.8%-12%) for patients who initially had normal renal function (P < .0001).

thumbnail image

Figure 2. The cumulative incidence of cardiovascular (CV) or cancer-specific mortality is illustrated for patients aged ≥75 years who had localized renal tumors managed by radical nephrectomy (RN), nephron-sparing intervention (NSI), or active surveillance (AS). RCC indicates renal cell cancer.

Download figure to PowerPoint

Chronic kidney disease (GFR <60 mL/minute/1.73m2) was present in 45% of patients before management and in 72% of patients after management (Table 1). A greater median reduction in GFR occurred after nephrectomy (34%) relative to nephron-sparing intervention (18%) or surveillance (0%), resulting in a much larger proportion of patients with newly diagnosed CKD after nephrectomy (47%) than after nephron-sparing intervention (25%) or surveillance (5%). The Kaplan-Meier estimate of the incidence of cardiovascular death at 5 years was 15% (95% CI, 9%-21%) for patients with stage III or greater CKD and 6% (95% CI, 1%-11%) for those without stage III or greater CKD after management (P = .04).

Cancer-Specific Survival

The cumulative incidence of cancer-specific mortality was substantially less than that for cardiovascular mortality in patients who had localized renal tumors (Fig. 2). In contrast, cancer-specific mortality was the major cause of death in patients who had more advanced renal cancer or upper-tract urothelial carcinoma, accounting for 51% and 46% of deaths, respectively. At 5 years, the cumulative incidence of cancer-specific mortality was 9.3% (95% CI, 1.1%-18%) with nephrectomy, 4% (95% CI, 0.7-7.3%) with nephron-sparing intervention, and 5.8% (95% CI, 0.8%-11%) with active surveillance (P = .33), likely reflecting selection bias and more aggressive treatment for larger, more aggressive tumors. Indeed, the radical nephrectomy group had tumors with greater oncologic potential: 53% had a >10% likelihood of metastasis at 12 years based on preoperative parameters.12 This proportion was significantly greater than the 8.6% and 14% of patients in the surveillance and nephron-sparing groups, respectively (P < .001). In addition, potentially aggressive pathologic features were present in 40% of radical nephrectomy specimens but in only 12% of specimens from the nephron-sparing group (P < .001).5 In univariate analysis, several factors predicted for cancer-specific death in patients who had localized renal tumors, including age, sex, tumor size, and predicted oncologic activity. By using step-wise selection of variables, cancer-specific mortality retained significant associations with increasing age (P = .01), being a man (P = .01), and increasing tumor size (P = .045), but not with management type (P = .68).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

The management of localized kidney cancer has undergone tremendous change with the advent of nephron-sparing and minimally invasive interventions. Active surveillance represents the least invasive strategy of all of these approaches and has been applied selectively at our institution in patients with limited life expectancy for renal tumors with low predicted malignant potential.13 Against these advances in the management of kidney cancer are long-standing practices of treating patients with genitourinary cancer more aggressively, including older and sicker patients.1, 14 In general terms, aggressive management is appropriate when it can be curative and if the cancer is likely to reduce the life expectancy of a given individual, but it may be unwarranted for an indolent tumor with little to no chance of dissemination. For the population of patients with localized renal cancer, mortality continues to increase despite earlier detection and aggressive treatment.15 In the current study, we observed that active treatment was not associated with improved overall survival relative to active surveillance in a heterogeneous population of patients aged ≥75 years who had clinically localized renal cancer after adjusting for potential confounders. In part, this likely reflects selection biases; however, the low malignant potential of many of these tumors, advanced age, and prevalence of comorbidities contributed to these findings. Reflecting this, the most common cause of death was cardiovascular in this cohort, as would be expected for elderly individuals who had no renal malignancy. The findings that 86% of patients had renal dysfunction after radical nephrectomy and that renal dysfunction was associated with a higher rate of cardiovascular death emphasize the potential downside of over treatment in this population. Current research indicates that localized renal tumors are over treated, and our data suggest that active surveillance is a reasonable strategy and is greatly underused in the elderly population.

There are several limitations to the current study, including its retrospective nature, selection biases, reliance on date and cause of death information from publicly accessible databases (SSDI and NDI), and lack of pathologic confirmation of malignancy in most patients who underwent surveillance. In addition, the small number of cancer-specific deaths during the follow-up interval (median, 3.9 years) prevented a definitive determination of treatment effect on cause-specific survival. In the absence of a randomized clinical trial, our current findings are viewed best as hypothesis-generating. Nevertheless, our results build on observations in several recent publications and increase the validity of a growing concern regarding the over treatment of small renal cortical tumors. Thompson et al first reported that radical nephrectomy was associated with decreased survival in patients who had a single renal cortical tumor ≤4 cm compared with partial nephrectomy, suggesting that this effect might be related to a loss of renal function that outweighed any oncologic benefit.16 Pettus et al subsequently reported that moderate and severe renal dysfunction before surgery were associated with decreased overall survival in patients undergoing partial or radical nephrectomy.17 Hollingsworth et al, in a population-based competing risk analysis, demonstrated that the competing-cause mortality for elderly patients (aged ≥70 years) was 28% and concluded that active surveillance might be a reasonable strategy for some tumors.15 Most recently, using data from the Surveillance, Epidemiology, and End Results cancer registry linked with Medicare claims, Huang et al demonstrated that radical nephrectomy was associated with an increased risk of death and cardiovascular events after surgery.18

At our institution, some form of nephron-sparing intervention is recommended almost uniformly for tumors that are amenable to such an approach. It has been demonstrated that both open and laparoscopic partial nephrectomy produce superior functional outcomes and oncologic outcomes that are at least equivalent to the outcomes produced after radical nephrectomy for selected tumors.16, 19-22 Percutaneous tumor ablation and active surveillance are considered for patients who may be poor surgical candidates. Although it is not quite as effective as partial nephrectomy, approximately 80% of patients are disease-free after thermal ablation, and this remains a valid nephron-sparing intervention for this age group.23 The expectation based on this clinical practice pattern is that patients who undergo active surveillance would have greater mortality during follow-up, and Figure 1 confirms this suspicion. It is noteworthy, however, that overall survival was not associated with the type of management after controlling for significant confounders, such as patient age and comorbidity. An increase of either 5 years in age or 1 unit on the Charlson-Romano index was associated with a 1.3-fold increase in the hazard of death of from any cause.

The lack of treatment effect in patients with localized kidney cancer underscores the biologic indolence of many of these tumors. Although 1 of the often quoted weaknesses of the active surveillance literature is the lack of pathologic confirmation, the lack of impact of histologic subtype or predicted oncologic activity on overall survival suggests that this limitation is less important than generally perceived. Recent data suggest that, in elderly men, up to 40% of suspected renal cancers that are amenable to nephron-sparing surgery are benign, and at least an additional 40% are likely to follow an indolent course.5 In the current study, only 12% of tumors that were treated with nephron-sparing intervention and 40% of those that were removed by nephrectomy had potentially aggressive pathologic features. Supporting this predicted tumor biology is the detection of distant metastases in only 26 of 537 patients, resulting in a cancer-specific mortality rate of only 4.1% in this series. Determining the oncologic potential of small renal tumors, including the use of renal mass biopsy, might provide information that would enable physicians and patients to select less invasive treatments, such as surveillance or tumor ablation, and to reserve surgical intervention for the small proportion of patients who have potentially aggressive tumors.11 The value of renal mass biopsy and/or radiographic testing that incorporates molecular diagnostic tools, such as those that detect carbonic anhydrase IX expression (which is limited to the more aggressive, clear cell variant of RCC) in this patient population warrants further investigation.11, 24

The lack of treatment effect in the current study also may be related to off-setting effects with oncologic benefit and renal-functional detriment to surgical approaches. Our data confirmed earlier data associating decreased renal function with an increased risk of cardiovascular death,10 which is the leading cause of mortality in elderly patients. Pre-existing CKD was present in 39% of elderly patients before nephrectomy, and renal functional loss after radical nephrectomy contributed to the development of CKD in the majority of patients in whom it did not exist before nephrectomy. Highlighting this finding, 86% of patients had CKD after nephrectomy. Any potential oncologic benefit from radical nephrectomy probably was over shadowed by the cost of this approach in terms of renal function and cardiovascular morbidity. For those reasons, the potential benefit of nephron-sparing intervention in elderly patients at lowest risk for cardiovascular mortality and with greatest life expectancy warrants further investigation, because these are the patients who might derive oncologic benefit from treatment.

In this retrospective evaluation of a heterogeneous population of patients aged ≥75 years, interventions for clinical T1 renal cancers were not associated with an improvement in overall survival. Increasing age and comorbidity were associated with death from any cause, and the most common cause of death was cardiovascular. Renal dysfunction was present in 86% of patients after radical nephrectomy and was a significant predictor of cardiovascular mortality. We conclude that the management of clinically localized, suspected RCC should be individualized based on predicted life expectancy, and strong consideration should be given to active surveillance. Further studies will be required to evaluate the advisability of nephron-sparing approaches in elderly patients who have minimal comorbidities and the role of renal mass biopsy for refining patient selection.

CONFLICT OF INTEREST DISCLOSURES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Dr. Lane receives support from the American Urologic Association Foundation as a Research Scholar. Dr. Gill owns a stock option in Hansen Medical and has acted as a consultant for EDAP.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES