SEARCH

SEARCH BY CITATION

Keywords:

  • kidney;
  • nephrectomy;
  • partial nephrectomy;
  • radical nephrectomy;
  • renal cell carcinoma

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. References

Objectives

To compare the recurrence-free survival of partial nephrectomy and radical nephrectomy in patients with non-metastatic pathological T3a renal cell carcinoma.

Methods

We reviewed the records of 3567 patients who had undergone a nephrectomy for renal cell carcinoma at five institutions in Korea from January 2000 to December 2010. The clinical data of 45 patients with pathological T3a renal cell carcinoma in the partial nephrectomy group were compared with 298 patients with pathological T3a renal cell carcinoma in the radical nephrectomy group. The effects of surgical methods on recurrence-free survival were assessed by a multivariate Cox proportional hazard analysis. All comparisons were repeated in subgroup analysis on 63 clinical T1a patients with tumors ≤4 cm.

Results

During a median 43-month follow-up period, disease recurrence occurred in two patients (4.4%) in the partial nephrectomy group, and 94 patients (31.5%) in the radical nephrectomy group. The results from a multivariate model showed that radical nephrectomy was a significant predictor of recurrence. However, in subgroup analysis that included 63 clinical T1a pathological T3a patients, the recurrence-free survival rates were not significantly different between the two cohorts. The renal function was significantly better preserved in the partial nephrectomy cohort than in the radical nephrectomy cohort.

Conclusions

Partial nephrectomy provides similar recurrence-free survival outcomes compared with radical nephrectomy in patients with clinical T1a pathological T3a renal cell carcinoma. However, there seems to be a higher risk of recurrence for large pathological T3a tumors treated by radical nephrectomy compared with small tumors treated by partial nephrectomy. Thus, large tumors with the same pathological T3a renal cell carcinoma grade could have hidden aggressive features.


Abbreviations & Acronyms
BMI

body mass index

Cr

creatinine

cT1a

clinical T1a

EBL

estimated blood loss

eGFR

estimated glomerular filtration rate

PN

partial nephrectomy

pT3a

pathological T3a

RCC

renal cell carcinoma

RN

radical nephrectomy

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. References

With great improvements in surgical technique, PN has become the standard of care in the management of small renal tumors that are less than 4 cm in size, which are classified as cT1a.[1-3] A recent study provided further evidence of a similarity between PN and RN for oncological control, and showed the superiority of PN for preserving renal function and preventing chronic kidney disease in addition to an associated reduction in cardiovascular morbidity and mortality, and an improvement in overall survival rates.[4, 5]

To achieve oncological results that are equivalent to a RN in patients with clinical T1b tumors, elective PN should be carried out at high-volume centers.[6-8] Nevertheless, elective PN offers better overall survival outcomes than RN after controlling for age, tumor size, stage and comorbidity factors associated with postoperative renal function.[9] Furthermore, a recent report provided further evidence that PN could be carried out for renal tumors greater than 7 cm with acceptable complication rates and oncological outcomes that are comparable with those of RN.[10] However, these studies included only a small proportion of patients who presented with a pT3a renal tumor, and of those patients, it was unknown which surgical method was used; that is, PN or RN. To our knowledge, only one previous European study reported matched cancer-specific survival data in the treatment of pT3a RCC, and no significant difference between PN and RN was reported.[11] Therefore, in the present study, we investigated the recurrence-free survival outcomes of patients with non-metastatic pT3a RCC, and compared the outcomes for small (cT1a) pT3a RCC treated using the PN and RN surgical methods.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. References

After obtaining institutional review board approval from each institution, we reviewed the data of 3567 patients who underwent PN or RN for the treatment of RCC between January 2000 and December 2010 at five institutions, and were followed for more than 1 year postoperatively. Among these patients, 343 men with pT3a were identified. The patients with positive lymph nodes or distant metastases, or a benign renal mass were excluded from the present study.

The demographic and operative data of 45 patients with RCC in the PN group were compared with the data of 298 patients with RCC in the RN group. To evaluate the technical differences and oncological outcome, we compared the operative time, EBL and recurrence-free survival rates. Fuhrman's nuclear grading system was used to evaluate the grade of all of the pT3a RCC tumors.[12] The tumor size was assessed as the maximum diameter of the pathological specimen; tumor stage was revised according to the 2010 TNM classification systems.

The renal function evaluation included a serum creatinine measurement and estimation of the glomerular filtration rate using the Modification of Diet in Renal Disease-2 equation.[13] The creatinine measurements used in the analysis were the preoperative measurements together with those taken at ≥12 months after surgery and at the last follow-up visit. Recurrence was defined as radiologically verified metastasis or local disease recurrence during the study period. The minimal oncological follow up consisted of annual computed tomography or magnetic resonance imaging studies and chest radiography. This occurred once annually during the first 5 years and every 2 years thereafter.

The two groups were compared using the χ2-test or an independent Student's t-test. The survival outcomes and recurrence-free survival were analyzed using the Kaplan–Meier method and the log–rank test. To determine whether the surgical method affected the survival outcomes, Cox proportional hazard regression models were constructed after adjusting for age, BMI, tumor size, tumor location (right vs left), pathological type (clear cell vs non-clear cell types) and nuclear grade. All comparisons were repeated in a subgroup analysis in which a tumor size ≤4 cm (cT1a) confirmed PN clinical efficacy under the same tumor size condition. All statistical tests were carried out using the Statistical Package for the Social Sciences (version 15.0; SPSS, Chicago, IL, USA). All tests were two-sided with a significance level defined as P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. References

Comparison of demographics, operative data and complications

Table 1 lists the demographic and operative data of the enrolled patients stratified by the surgical procedure. Those selected for RN were more likely to have larger tumors than those for PN (7.99 vs 3.50 cm, P < 0.001). In the PN cohort, there were 30 cases (66.7%) with cT1a tumors and 15 cases (33.3%) with cT1b tumors. There was no case in the PN cohort with a tumor size above 7 cm; however, higher clinical stage tumors were included in the RN cohort (P < 0.001). There were no significant differences in age, sex distribution, BMI, tumor laterality, surgical procedures (i.e. open, laparoscopic and robotic) and operative time. The EBL was greater in the RN cohort compared with the PN cohort (582 vs 303 cc, P < 0.001). Additionally, the transfusion rate was greater in the RN cohort compared with the PN cohort (27.2% vs 8.9%, P = 0.003).

Table 1. Demographic characteristics, operative data and pathological outcomes of study patients, stratified according to type of surgery
CharacteristicsPNRNP-value
n45298 
Age (years)57.4 ± 13.060.2 ± 11.80.150
Sex (%)  0.518
Male31 (68.9)219 (73.5) 
Female14 (31.1)79 (26.5) 
BMI (kg/m2)24.8 ± 2.524.2 ± 3.30.138
Surgical procedure  0.811
Open32 (71.1)203 (68.1) 
Laparoscopic6 (13.3)85 (28.6) 
Robotic7 (15.6)10 (3.3) 
Tumor laterality (%)  0.133
Right19 (42.2)162 (54.0) 
Left26 (57.8)136 (45.6) 
Mean tumor size (cm) ± SD (range)3.50 ± 1.55 (0.80–7.00)7.99 ± 3.68 (1.80–26.00)<0.001
Clinical stage (%)  <0.001
T1a (tumor size ≤4 cm)30 (66.7)33 (11.1) 
T1b (4 cm < tumor size ≤7 cm)15 (33.3)127 (42.6) 
T2a (7 cm < tumor size ≤10 cm)NA86 (28.9) 
T2b (10 cm < tumor size)NA42 (14.1) 
T3aNA10 (3.3) 
Reason for pT3a   
Perirenal/sinus fat invasion41 (91.1)265 (88.9)0.248
Vascular invasion4 (8.9)33 (11.1) 
Operative time (min)177.5 ± 64.6190.4 ± 81.10.308
Warm ischemic time (min)26.0 ± 13.3NA 
EBL (mL)302.8 ± 260.6581.9 ± 130.9<0.001
Transfusion (%)4 (8.9)81 (27.2)0.003
Intraoperative complications7 (15.6)43 (14.4)0.842
Postoperative complications  0.844
Prolonged bleeding1 (2.2)13 (4.5) 
Wound problem1 (2.2)5 (1.7) 
Urine leakageNA1 (0.3) 
Prolonged ileus1 (2.2)8 (2.8) 
Other3 (6.6)10 (3.4) 
Pathological diagnosis  0.796
Clear cell36 (80.0)247 (82.9) 
Papillary1 (2.2)7 (2.3) 
Chromophobe5 (11.1)19 (6.4) 
Other3 (6.6)25 (8.4) 
Fuhrman nuclear grade (%)  0.009
12 (4.4)2 (0.7) 
220 (44.4)85 (28.5) 
320 (44.4)149 (50.0) 
43 (6.7)62 (20.8) 
Resection margin positive2 (4.4)NA 

Intraoperative complications occurred in 50 (14.4%) of the study patients, with seven (15.6%) cases occurring in the PN group and 43 (14.4%) cases in the RN group. No significant difference was found between the two groups. Postoperative complications occurred in 43 (12.5%) of the study patients, with six (13.3%) cases occurring in the PN group and 37 (12.4%) cases in the RN group. There were no significant differences between the two groups.

Pathological outcomes

On final pathology, 283 RCC (82.5%) showed a clear-cell type, which is the dominant pathological tumor type, and constituted more than 80% of each group (Table 1). The tumors that were removed by RN more likely had a higher grade compared with those removed by PN. The numbers of tumors with Fuhrman nuclear grades of 3 and 4 were 149 (50.0%) and 62 (20.8%), respectively, in the RN group and 20 (44.4%) and three (6.7%), respectively, in the PN group; a significant difference between two cohorts was observed (P = 0.009). In the PN group, a positive resection margin was reported in two patients (4.4%).

Oncological outcomes: Recurrence-free survival

As shown in Table 2, during the median 43-month follow-up period, disease recurrence developed in two patients (4.4%) in the PN group and 94 patients (31.5%) in the RN group (P < 0.001). Two patients with recurrence in the PN group had a negative resection margin. A 5-year recurrence-free survival was estimated at 97% among patients in the PN group and at 55% among patients in the RN group (Fig. 1a; P < 0.001). The overall mortality was one patient (2.2%) in the PN group and 55 patients (18.5%) in the RN group. Cancer-related mortality occurred in 0 patients of the PN group and 38 patients of the RN group. As shown in Table 3, a multivariate Cox proportional hazard analysis showed that RN status (OR 4.175; P = 0.043) was a significant predictor of recurrence during the follow-up period, even when controlling for age, BMI, tumor size, clinical stage (T1a vs others), tumor location (right vs left), Fuhrman nuclear grade (1 and 2 vs 3 and 4) and pathological type (clear cell vs non-clear cell types).

figure

Figure 1. Recurrence-free survival that is stratified by surgical method (i.e. PN vs RN) in (a) patients with pT3a lesions and (b) patients with pT3a lesions and a tumor size ≤4 cm.

Download figure to PowerPoint

Table 2. Oncological outcomes data
CharacteristicsPNRNP-value
343 Total patients with pT3a
n45298 
Mean tumor size (cm) ± SD (range)3.50 ± 1.55 (0.80–7.00)7.99 ± 3.68 (1.80–26.00)<0.001
Fuhrman nuclear grade (%)  0.009
12 (4.4)2 (0.7) 
220 (44.4)85 (28.5) 
320 (44.4)149 (50.0) 
43 (6.7)62 (20.8) 
Median follow up (months) ± SD44.0 ± 31.143.0 ± 34.40.982
Mean follow up (months)35.9135.79 
Overall mortality (%)1 (2.2)55 (18.5)<0.001
Cancer specific mortality (%)038 (12.8) 
Recurrence (%)2 (4.4)94 (31.5)<0.001
Median duration between operation to recurrence (months)3.06.5 ± 18.60.492
63 Subgroup analysis in patient with pT3a and tumor size ≤4 cm (cT1a and pT3a)
n3033 
Mean tumor size (cm) ± SD (range)2.75 ± 0.86 (0.80–4.00)3.37 ± 0.74 (1.80–4.00)0.004
Fuhrman nuclear grade (%)  0.600
11 (3.3)1 (3.0) 
211 (36.7)13 (39.4) 
315 (50.0)16 (48.5) 
43 (10.0)3 (9.1) 
Median follow up (months) ± SD35.0 ± 33.230.5 ± 35.30.732
Mean follow up (months)40.9340.31 
Overall mortality (%)1 (3.3)3 (9.1)<0.001
Cancer-specific mortality (%)02 (6.1) 
Recurrence (%)1 (3.3)4 (12.1)0.265
Median duration between operation to recurrence (months)3.039.0 ± 30.80.005
142 Subgroup analysis in patient with pT3a and tumor size 4 to 7 cm (cT1b and pT3a)
n15127 
Mean tumor size (cm) ± SD (range)5.52 ± 0.95 (4.10–7.00)5.72 ± 0.88 (4.10–7.00)0.492
Fuhrman nuclear grade (%)  0.003
11 (6.7)1 (0.8) 
29 (60.0)32 (25.2) 
35 (33.3)69 (54.3) 
4025 (19.7) 
Median follow up (months) ± SD15.0 ± 24.325.50 ± 34.60.130
Mean follow up (months)23.7336.76 
Overall mortality (%)015 (11.8)<0.001
Cancer-specific mortality (%)07 (5.5)<0.001
Recurrence (%)1 (6.7)25 (19.7)<0.001
Median duration between operation to recurrence (month)3.08.5 ± 19.30.004
Table 3. Multivariate Cox hazard model predicting recurrence after nephrectomy among patients in pT3a renal tumors
 OR95% CIP-value
343 Total patients with pT3a   
Multivariate analysis   
Age0.9070.980–1.0180.907
BMI0.8810.814–0.9540.002
Tumor size1.1251.066–1.189<0.001
Clinical stage1.6580.553–4.9730.367
Tumor location (right vs left)0.8640.558–1.3370.512
Fuhrman grade1.9071.114–3.2640.019
Pathological type (clear vs non-clear)0.6400.356–1.1490.135
Operative method (RNx vs PNx)6.1751.078–13.3730.043
Subgroup analysis in 63 patients with pT3a and tumor size ≤4 cm (cT1a and pT3a)   
Multivariate analysis   
Age0.9660.913–1.0220.225
BMI0.8410.654–1.0810.175
Tumor size1.4870.613–3.6070.380
Tumor location (right vs left)0.9340.246–3.5460.920
Fuhrman grade1.4620.380–5.6330.581
Pathological type (clear vs non-clear)0.8440.167–4.2710.838
Operative method (RNx vs PNx)3.9310.408–7.8950.236

However, in a subgroup analysis that included 63 pT3a patients with a small renal mass (cT1a; tumor size ≤4 cm), recurrence was observed in one patient (3.3%) in the 30-patient PN group and in four patients (12.1%) in the 33-patient RN group. There was no significant difference between the two cohorts (Table 2; P = 0.265). In the small renal mass, cT1a and pT3a subgroup analysis, the 5-year recurrence-free survival was an estimated 96% in the PN group and 86% in the RN group. There was no significant difference between the two groups (Fig. 1b; P = 0.306). The subgroup multivariate Cox proportional hazard analysis showed that the operative method was no longer a significant predictor of RCC recurrence among patients with a small renal mass (≤4 cm) pT3a RCC (OR 3.931; P = 0.236) after controlling for age, BMI, tumor size, tumor location, Fuhrman grade and pathological type (Table 3).

Renal functional outcomes

As shown in Table 4, there was no significant difference in the preoperative serum creatinine and eGFR levels between the PN and RN groups. However, patients who underwent a PN were more likely to have better renal function compared with patients who underwent a RN. At the 12-month postoperative period, the serum creatinine (1.11 vs 1.32 mg/dL, P = 0.001) and eGFR (70.2 vs 57.9 mL/min, P = 0.001) levels were significantly favorable in the PN group compared with the RN group among the whole cohort. During the median 44-month follow-up period, the mean creatinine (1.07 vs 1.37 mg/dL, P = 0.001) and eGFR (75.4 vs 59.8 mL/min, P < 0.001) levels measured at the last follow-up visit were significantly favorable in the PN group compared with the RN group. The superiority of PN with respect to preserving renal function was observed in the subgroup analysis that included the pT3a small renal mass group. The renal functional parameters, which were checked 12 months postoperatively and at the last follow-up period, showed that patients in the PN group had more favorable renal functional outcomes than the RN group among the pT3 and tumor size ≤4 cm subgroup.

Table 4. Renal functional outcomes data
CharacteristicsPNRNP-value
343 Total patients with pT3a
n45298 
Mean preoperative Cr (mg/dL)1.21 ± 1.361.10 ± 0.550.336
Mean preoperative eGFR (mL/min)73.9 ± 23.377.8 ± 24.70.332
Mean postoperative 12 months Cr (mg/dL)1.11 ± 0.311.32 ± 0.300.001
Mean postoperative 12 months eGFR (mL/min)70.2 ± 17.057.9 ± 13.90.001
Last checked Cr (mg/dL)1.07 ± 0.321.37 ± 0.870.001
Last checked eGFR (mL/min)75.4 ± 21.859.8 ± 21.7<0.001
Subgroup analysis in 63 patients with pT3a and tumor size ≤4 cm (cT1a and pT3a)
n3033 
Mean preoperative Cr (mg/dL)1.07 ± 0.191.35 ± 1.180.201
Mean preoperative eGFR (mL/min)73.9 ± 14.668.7 ± 22.90.320
Mean postoperative 12 months Cr (mg/dL)1.13 ± 0.291.37 ± 0.410.041
Mean postoperative 12 months eGFR (mL/min)69.0 ± 15.457.5 ± 21.20.058
Last checked Cr (mg/dL)1.12 ± 0.331.44 ± 0.660.032
Last checked eGFR (mL/min)72.4 ± 17.857.8 ± 20.30.006

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. References

In the current study, we observed that patients with pT3a RCC after RN had a higher odds ratio of recurrence compared with pT3a RCC patients treated with PN; however, in a subgroup analysis of patients who presented with small renal mass ≤4 cm (cT1a), there was no significant difference in oncological outcomes. Although the pathological grouping was the same for non-metastatic pT3a RCC, the RN cohort, which included more large-sized tumors, had a significantly higher recurrence rate than the PN cohort. However, in the cases with traditional indication for PN, the operative method was not a significant predictor of RCC recurrence among men who were found to have pT3a tumors after undergoing a nephrectomy. Furthermore, renal function was significantly preserved in patients after a PN compared with patients receiving a RN.

The indications for PN have broadened over the past decade.[14-22] The use of PN in patients with larger tumors has led to incidental diagnoses of perinephric fat invasion.[6] This finding might concern many urologists who are confronted with such a pathology report. Some may question whether a RN would have resulted in better cancer control. Unfortunately, there are no clear data addressing such cancer control outcomes in patients with pT3a stage tumors who were treated with a PN.[14, 19, 23, 24]

Currently, most studies examining PN in the treatment of large tumors reveal the oncologic safety compared with subjects after a RN. Weight et al. reported that PN offers cancer control that is equivalent to a RN among patients with renal tumors greater than 4–7 cm.[9] In their recent study, Becker et al. reported that patients who underwent elective PN in renal tumors ≥7 cm had acceptable oncological outcomes that were comparable with RN outcomes.[10] Although these studies were encouraging to urologists who carry out PN in large renal masses, the final pathological data after PN showed that the extension out of the organ was a concerning factor. Furthermore, the aforementioned PN series with large renal masses had a small number of patients with organ-extensive RCC; therefore, further investigation is required.

Jeldes et al. reported the results from a 13-center multi-institutional study of 861 patients, who had pT3a lesions, where a RN series was compared with PN surgical procedures.[11] In their study, just 72 patients (8.4%) underwent PN; the tumors that were resected by RN were significantly larger and had more high-grade features. In their subgroup analysis, with a matched cohort study, they compared 30 PN series with 63 RN series. After matching the series, the 5-year disease-specific survival rate was 81.9% in the pT3a PN series and 90.1% in the pT3a RN series. Although there was no significant difference in the 5-year disease-specific survival, PN was associated with a 2.5-fold higher rate of cancer-specific mortality compared with RN (P = 0.9). In multivariate analysis among the unmatched 861 patients, the surgical method, which was either PN or RN, was not a significant factor influencing the cancer-specific mortality rate. As in the study by Jeldes et al., there was no significant difference in the recurrence-free survival rate between the PN and RN surgical methods in the present results among cT1a and pT3a tumors.[11] Subgroup analyses of the present results also showed that the surgical method was not a significant predictor for the outcome after controlling for other factors.

However, in contrast to the report by Jeldes et al., we observed that the operative method between PN and RN was a significant predictor using a multivariate Cox proportional hazard model after controlling other variable factors.[11] This difference might indicate that tumor size is an important factor in circumstances of pre-existing, extensive organ disease. In fact, oncological outcomes significantly decreased with increasing size in all tumors with pathologically confirmed RCC.[25] Therefore, similar to pT3a tumors, RCC that presents with a large tumor size might have a greater probability of worse oncological outcomes, such as micro-metastasis, than smaller tumors classified as being at the same pathological stage. Using existing technology, micro-metastasis detection is unfeasible to apply clinically; therefore, additional diagnostic tools for early micro-metastasis detection should be investigated.

Despite its large multi-institutional nature and sample size, the lack of pT3a pathological characteristics represents a significant limitation in the current study. The particularly small number of cT1b and cT2 RCC cases in the PN cohort should be validated in large-scale future studies. Additionally, detailed data on recurrence location, which was local or metastatic, were not included in the analysis because of the retrospective nature. A surgical selection bias was also present in this study. The clinically large renal tumors were mainly in the RN cohort; therefore, the pT3a RN cohort had a higher recurrence rate than the pT3a PN cohort in our analysis. This result might have been due to the tumor size disparity. The multivariate analysis was not adjusted for differences in surgical approach (i.e. open, laparoscopic and robotic) because of the known equivalence of outcomes associated with these surgical approaches reported by previous studies.[11, 26-28] In our multivariate analysis, tumor size, Fuhrman grade and BMI were also significant factors for predicting disease recurrence. The tumor size, as reported by Crispen et al., was an important factor for oncological outcomes despite adjusting for the pathological stage.[25] The Fuhrman nuclear grade and BMI could also be unfavorable factors for RCC outcomes; however, further studies are necessary for the accurate evaluation of these factors in pT3a RCC.[3, 29, 30] Despite these limitations, the present study provides important results regarding the equivalence of PN oncological outcomes, and the superiority of using PN in preserving renal function compared with RN among pT3a RCC patients with a small renal mass (≤4 cm). The results from the present study also suggest that larger tumor size might be associated with an increased potential to show aggressive features compared with smaller tumors, even if the T3a pathological stage is the same.

Based on the present findings, PN has similar recurrence-free survival outcomes compared with RN among patients with only cT1a and pT3a RCC. Furthermore, the PN group experienced superior renal function compared with the RN group. However, pT3a RCC with large tumor size treated by RN had a significantly higher disease recurrence outcome than pT3a RCC with relatively small tumor size treated by PN. An additional large-scale, prospective evaluation is required to ensure that oncological safety is warranted.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. References