Skip to main content
NCBI home page
Canadian Urological Association Journal logoLink to Canadian Urological Association Journal
. 2016 Sep 13;10(9-10):E282–E289. doi: 10.5489/cuaj.3707

Survival after partial and radical nephrectomy for high-risk disease: A propensity-matched comparison

Matthew J Maurice 1, Hui Zhu 1,2, Simon Kim 3, Robert Abouassaly 3,✉
PMCID: PMC5028211  PMID: 27695581

Abstract

Introduction:

Increasingly, partial nephrectomy has been applied to high-risk disease without evidence that its survival benefits can be extrapolated to this entity. We aimed to compare overall survival after partial vs. radical nephrectomy in patients with high-risk renal cell carcinoma.

Methods:

Using the National Cancer Data Base, we identified patients who underwent partial or radical nephrectomy for high-risk disease between 2003 and 2006. High-risk disease was defined as the presence of adverse pathological features within the primary tumour, namely high-grade or unfavourable histology, T3 stage, or both. After matching the partial and radical nephrectomy groups based on propensity scores, 1680, 276, and 76 patients with high-grade or unfavourable histology, T3 stage, or both adverse pathologic features, respectively, were available for analysis. Five-year overall survival was compared after partial vs. radical nephrectomy for each high-risk cohort using the Kaplan-Meier and log rank tests.

Results:

Partial nephrectomy was associated with a statistically significant improvement in five-year overall survival compared to radical nephrectomy for small tumours (median size 3.0 cm; interquartile range 2.1–4.5 cm) with high-grade or unfavourable histology (87% vs. 81%; p<0.01) or with pT3a stage (82% vs. 71%; p<0.01). For patients concomitantly harbouring both adverse pathologic features, no difference in survival was detected (p=0.21).

Conclusions:

Partial nephrectomy is associated with survival benefits in patients with adverse pathologic features, suggesting that renal preservation is not only safe, but also potentially beneficial for high-risk disease. Due to inherent selection bias associated with partial nephrectomy use, prospective validation of these findings is needed.

Introduction

Partial nephrectomy (PN) is the optimal treatment for clinical T1 renal masses.1 This recommendation is supported by numerous retrospective studies, which have demonstrated its superior functional and survival outcomes and at least equivalent cancer control relative to radical nephrectomy (RN).2,3 Accordingly, PN use has increased over the last 10–15 years, with nearly half of all T1 renal tumours now being managed with PN.4,5 Recently, this trend has been associated with increasing PN use for renal cell carcinoma (RCC) with adverse pathologic factors that independently predict cancer-specific death.6–10 The role of PN for this high-risk RCC is ill-defined. Several retrospective studies have investigated the comparative effectiveness of PN vs. RN for high-risk disease, but have failed to detect a survival difference in this setting.11–14 Despite its flaws, the only level 1 evidence comparing PN and RN demonstrated a survival advantage for RN, calling into question the long-held belief that PN is necessarily better than RN for the elective management of small renal masses.15,16 We sought to compare survival outcomes between PN and RN for high-risk disease in a large, nationally representative cohort of American patients in the National Cancer Data Base (NCDB).

Methods

NCDB participant user file

The NCDB was established in 1989 as a joint project of the American Cancer Society and the Commission on Cancer (CoC) of the American College of Surgeons. It is a nationwide, hospital-based cancer registry that captures comprehensive clinical data on approximately 70% of all newly diagnosed malignancies in the U.S. annually. A data use agreement exists between the American College of Surgeons and each of its CoC-accredited hospitals. A detailed description of the NCDB has been published previously.6 We used the NCDB’s 2011 participant user file for kidney and renal pelvis cancers. Institutional review board approval was obtained prior to study initiation.

Study population

We identified 120 926 patients who had been diagnosed with RCC from 2003–2006, based on International Classification of Diseases for Oncology, third edition primary site codes. Diagnoses prior to 2003 were excluded because the Charlson comorbidity index (CCI) variable was not available, and diagnoses after 2006 were excluded to avoid patients with less than five years followup. We restricted the cohort to patients with clinical T1-T2, pathological T1-T3a, non-metastatic RCC who underwent PN or RN (n=30 287). Patients with other primary cancers were excluded to avoid potential confounding (n=5227). Cases diagnosed but not treated at the reporting facility were excluded due to incomplete followup (n=131).

Three cohorts were identified from the study population based on the presence of adverse pathologic features (APF): 1) RCC with high-grade or unfavourable histology (n=5150); 2) RCC with pT3a stage (n=1265); and 3) RCC with high-grade or unfavourable histology and pT3a stage (n=491).

Study variables

The NCDB contains data pertaining to patient demographics and comorbidity burden; cancer grade, stage, and histology; first course treatment(s); and overall survival. The study variables included: age, sex, CCI, race, income level, insurance type, county, hospital type, hospital surgical volume, tumour size, pT stage, tumour histology, and tumour grade. Age and tumour size were treated as continuous variables; all other variables were categorical. Missing data for the income, insurance, county, and grade variables were included as unknowns. CCI was calculated based on International Classification of Disease, ninth edition, clinical modification, secondary diagnosis codes and was categorized as 0 (no comorbidities), 1, or >1. Race was categorized as White, African-American, Hispanic, or other. Income levels were estimates based on 2000 U.S. census data that were stratified into annual income quartiles: lowest (<$30 000), lower middle ($30 000–$35 000), middle ($35 000–$46 000), and upper middle (>$46 000). Insurance type included no health insurance, social (Medicare/Medicaid), and private/managed care. County was categorized as urban, metropolitan, or rural according to data from the 2003 U.S. Department of Agriculture Research Service. Using classifications developed by the CoC, hospitals were designated as academic, comprehensive, community, or other hospitals. Surgical volume was calculated based on the hospital’s overall surgical volume (2003–2006). Surgical volume was stratified into tertiles (<8, 8–21, >21 total cases), and hospitals were classified as low-, intermediate-, or high-volume. Pathological T stage included pT1-T3a. RCC histology was categorized as clear-cell, papillary, chromophobe, unfavourable subtype/variant (i.e., collecting duct, medullary, sarcomatoid, or rhabdoid), unclassified RCC, and other. Histologic grade, based on the American Joint Committee on Cancer grading system, was categorized as low-grade (G1/G2) and high-grade (G3/G4).

Study outcomes

The primary outcome was five-year overall survival (OS), which was calculated from the date of surgery to the date of death from all causes or censored at the date of last contact.

Propensity matching

We observed a greater proportion of large, higher-stage, histologically aggressive tumours within the controls (RN) vs. the cases (PN) (Table 1). Using 1:1 propensity score-matching based on the nearest neighbour algorithm, 90% or more of cases were matched to controls, and no additional matches could be made with the given data, confirming the completeness of the match. P values comparing cases to controls were not significant after matching for two out of three cohorts, confirming good matching.

Table 1.

Patient, hospital, and clinical characteristics of three high-risk RCC cohorts by surgery type before propensity matching

HG/UH (N=5150) T3 (N=1265) HG/UH and T3 (N=491)
RN (n=4216) PN (n=934) RN (n=1125) PN (n=140) RN (n=452) PN (n=39)
Variable n % n % p value n % n % p value n % n % p value
Diagnosis year 0.02 0.26 0.36
  2003 775 18.4 144 15.4 260 23.1 22 15.7 95 21.0 6 15.4
  2004 1009 23.9 205 21.9 249 22.1 34 24.3 99 21.9 10 25.6
  2005 1130 26.8 255 27.3 281 25.0 40 28.6 114 25.2 14 35.9
  2006 1302 44.7 330 54.6 335 42. 44 45.8 144 46.8 9 30.0
Age (years) 0.01 0.10 0.04
  Median (IQR) 61 (52–71) 61 (51–69) 63 (54–74) 63 (54–70) 64 (55–75) 59 (52–66)
Sex 0.02 0.28 0.32
  Female 1539 36.5 303 32.4 397 35.3 43 30.7 151 33.4 10 25.6
  Male 2677 63.5 631 67.6 728 64.7 97 69.3 301 66.6 29 74.4
Charlson comorbidity index 0.68 0.12 0.97
  0 3017 71.6 665 71.2 771 68.5 84 60.0 307 67.9 26 66.7
  1 861 20.4 200 21.4 264 23.5 41 29.3 104 23.0 9 23.1
  >1 338 8.0 69 7.4 90 8.0 15 10.7 41 9.1 4 10.3
Race 0.66 0.61 0.83
  White 3356 79.6 757 81.0 930 82.7 118 84.3 376 83.2 34 87.2
  African-American 455 10.8 89 9.5 84 7.5 11 7.9 34 7.5 3 7.7
  Hispanic 237 5.6 49 5.2 71 6.3 5 3.6 27 6.0 1 2.6
  Other 168 4.2 39 4.4 40 3.7 6 4.5 15 3.4 1 2.6
Income level 0.62 0.55 0.11
  Lowest 542 12.9 122 13.1 135 12.0 19 13.6 54 11.9 9 23.1
  Lower-middle 738 17.5 162 17.3 206 18.3 21 15.0 91 20.1 3 7.7
  Middle 1089 25.8 220 23.6 311 27.6 33 23.6 133 29.4 9 23.1
  Upper-middle 1608 38.1 371 39.7 413 36.7 60 42.9 151 33.4 15 38.5
  Unknown 239 5.7 59 6.3 60 5.3 7 5.0 23 5.1 3 7.7
County 0.70 0.16 0.90
  Urban 614 14.6 131 14.0 192 17.1 19 13.6 73 16.2 8 20.5
  Metropolitan 3256 77.2 720 77.1 834 74.1 115 82.1 341 75.4 28 71.8
  Rural 78 1.9 15 1.6 26 2.3 2 1.4 10 2.2 1 2.6
  Unknown 268 6.4 68 7.3 73 6.5 4 2.9 28 6.2 2 5.1
Hospital type <0.01 <0.01 0.74
  Academic 1566 37.1 513 54.9 441 39.2 75 53.6 188 41.6 19 48.7
  Comprehensive 2296 54.5 373 39.9 576 51.2 60 42.9 229 50.7 18 46.2
  Community 300 7.1 42 4.5 94 8.4 4 2.9 28 6.2 2 5.1
  Other 54 1.3 6 0.6 14 1.2 1 0.7 7 1.5 0 0.0
Hospital surgical volume <0.01 <0.01 0.67
  Low 1394 33.1 210 22.5 420 37.3 36 25.7 172 38.1 12 30.8
  Intermediate 1476 35.0 281 30.1 382 34.0 46 32.9 156 34.5 15 38.5
  High 1346 31.9 443 47.4 323 28.7 58 41.4 124 27.4 12 30.8
Tumour size (cm) <0.01 <0.01 <0.01
  Median (IQR) 5.5 (3.8–7.8) 3.0 (2.2–4.0) 6.5 (4.6–8.5) 3.0 (2.0–4.0) 7.0 (5.0–9.5) 3.0 (2.3–4.5)
Histology <0.01 <0.01 <0.01
  Clear cell 1392 33.0 300 32.1 457 40.6 35 25.0 174 38.5 3 7.7
  Papillary 389 9.2 179 19.2 91 8.1 32 22.9 35 7.7 10 25.6
  Chromophobe 241 5.7 73 7.8 54 4.8 9 6.4 21 4.6 4 10.3
  Aggressive type 88 2.1 6 0.6 33 2.9 0 0.0 22 4.9 0 0.0
  Unclassified RCC 1808 42.9 335 35.9 435 38.7 59 42.1 168 37.2 19 48.7
  Other 298 7.1 41 4.4 55 4.9 5 3.6 32 7.1 3 7.7
Clinical T stage <0.01 <0.01 <0.01
  T1 2877 68.2 894 95.7 599 53.2 130 92.9 219 48.5 34 87.2
  T2 1339 31.8 40 4.3 526 46.8 10 7.1 233 51.6 5 12.8
Pathological T stage <0.01 - -
  T1 2661 63.1 859 92.0 - - - -
  T2 1103 26.2 36 3.9 - - - -
  T3a 452 10.7 39 4.2 1125 100.0 140 100.0 452 100.0 39 100.0
Grade - 0.01 -
  Low-grade - - 517 46.0 83 59.3 - -
  High-grade 4216 100.0 934 100.0 452 40.2 39 27.9 452 100.0 39 100.0
  Unknown/Not applicable - - 156 13.9 18 12.9 - -
Followup (months) 0.15 0.01 0.18
  Median (IQR) 67 (43–82) 69 (49–81) 65 (35–81) 70 (55–82) 60 (26–76) 64 (43–80)
Open in a new tab

HG/UH: high-grade or unfavourable hstology; IQR: interquartile range; PN: partial nephrectomy; RCC: renal cell carcinoma; RN: radical nephrectomy; T3: pathological T3a stage.

Statistical analyses

Differences between the PN and RN groups were assessed by univariate analysis using the Mann-Whitney U-test for continuous variables and the chi-squared test for categorical variables. The PN and RN groups were compared in terms of OS using the Kaplan-Meier method and log-rank test. In order to adjust for covariates, which remained unbalanced after propensity-matching, the association between surgery type (PN or RN) and OS was analyzed using a multivariable Cox proportional hazards model. Statistical tests were performed using SAS® University Edition (SAS Institute Inc., Cary, NC, U.S.). P values <0.05 were considered statistically significant.

Results

In the final study cohorts, there were 1680, 276, and 76 patients with high-grade or unfavourable histology (HG/UH), pT3a stage (T3), or both APF, respectively (Table 2). The HG/ UH and T3 cohorts were well matched. In the combined HG/UH and T3 cohort, elderly and male patients were over-represented among PN cases.

Table 2.

Patient, hospital, and clinical characteristics of three high-risk RCC cohorts by surgery type after propensity matching

HG/UH (n=1680) T3 (n=276) HG/UH and T3 (n=76)
RN PN RN PN RN PN
Variable n % n % p value n % n % p value n % n % p value
Diagnosis year 0.76 0.09 0.38
  2003 143 17.0 134 16.0 39 28.3 22 15.9 11 28.9 6 15.8
  2004 199 23.7 194 23.1 27 19.6 33 23.9 8 21.1 10 26.3
  2005 228 27.1 222 26.4 37 26.8 39 28.3 8 21.1 13 34.2
  2006 270 47.4 290 52.7 35 34.0 44 46.8 11 40.7 9 31.0
Age (years) 0.19 0.09 0.04
  Median (IQR) 61 (52–70) 61 (52–70) 63 (54–72) 65 (54–74) 64 (54–73) 70 (59–76)
Sex 0.47 0.52 0.02
  Female 292 34.8 278 33.1 48 34.8 43 31.2 19 50.0 9 23.7
  Male 548 65.2 562 66.9 90 65.2 95 68.8 19 50.0 29 76.3
Charlson comorbidity index 0.49 0.54 0.87
  0 599 71.3 600 71.4 90 65.2 84 60.9 23 60.5 25 65.8
  1 167 19.9 178 21.2 38 27.5 39 28.3 11 28.9 9 23.7
  >1 74 8.8 62 7.4 10 7.2 15 10.9 4 10.5 4 10.5
Race 0.98 0.94 0.53
  White 673 80.1 671 79.9 113 81.9 116 84.1 29 76.3 33 86.8
  African-American 82 9.8 85 10.1 14 10.1 11 8.0 4 10.5 3 7.9
  Hispanic 50 6.0 47 5.6 5 3.6 5 3.6 4 10.5 1 2.6
  Other 35 4.3 37 4.6 6 4.5 6 4.5 1 2.7 1 2.7
Income level 0.13 0.98 0.71
  Lowest 83 9.9 116 13.8 18 13.0 19 13.8 7 18.4 9 23.7
  Lower-middle 158 18.8 153 18.2 25 18.1 21 15.2 7 18.4 3 7.9
  Middle 216 25.7 204 24.3 31 22.5 32 23.2 7 18.4 8 21.1
  Upper-middle 336 40.0 313 37.3 57 41.3 59 42.8 15 39.5 15 39.5
  Unknown 47 5.6 54 6.4 7 5.1 7 5.1 2 5.3 3 7.9
County 0.80 0.33 0.64
  Urban 116 13.8 126 15.0 20 14.5 19 13.8 4 10.5 8 21.1
  Metropolitan 648 77.1 641 76.3 110 79.7 113 81.9 30 78.9 27 71.1
  Rural 17 2.0 13 1.5 0 0.0 2 1.4 1 2.6 1 2.6
  Unknown 59 7.0 60 7.1 8 5.8 4 2.9 3 7.9 2 5.3
Hospital type 0.82 0.56 0.68
  Academic 447 53.2 432 51.4 71 51.4 75 54.3 15 39.5 18 47.4
  Comprehensive 350 41.7 360 42.9 57 41.3 58 42.0 19 50.0 18 47.4
  Community 39 4.6 42 5.0 9 6.5 4 2.9 3 7.9 2 5.3
  Other 4 0.5 6 0.7 1 0.7 1 0.7 1 2.6 0 0.0
Hospital surgical volume 0.36 0.51 0.71
  Low 180 21.4 203 24.2 27 19.6 35 25.4 12 31.6 12 31.6
  Intermediate 268 31.9 267 31.8 49 35.5 45 32.6 12 31.6 15 39.5
  High 392 46.7 370 44.0 62 44.9 58 42.0 14 36.8 11 28.9
Tumour size (cm) 0.64 0.51 0.48
  Median (IQR) 3.0 (2.4–4.3) 3.0 (2.4–4.5) 3.0 (2.1–4.5) 3.0 (2.1–4.5) 3.4 (2.5–4.5) 3.7 (2.9–4.5)
Histology 0.71 0.53 0.98
  Clear cell 268 31.9 279 33.2 42 30.4 35 25.4 3 7.9 3 7.9
  Papillary 145 17.3 133 15.8 35 25.4 31 22.5 10 26.3 10 26.3
  Chromophobe 59 7.0 65 7.7 10 7.3 9 6.5 4 10.5 3 7.9
  Aggressive type 3 0.4 6 0.7 0 0 0 0.0% 0 0.0
  Unclassified RCC 298 35.5 317 37.7 42 30.4 58 42.0 16 42.1 19 50.0
  Other 67 8.0 40 4.8 9 6.5 5 3.6 5 13.2 3 7.9
Clinical T stage 0.03 0.02 0.53
  T1 778 92.6 800 95.2 115 83.3 128 92.8 31 81.6 33 86.8
  T2 62 7.4 40 4.8 23 16.7 10 7.3 7 18.4 5 13.2
Pathological T stage 0.24 - -
  T1 757 90.1 765 91.1 0 0.0 0 0.0 0 0.0 0 0.0
  T2 50 6.0 36 4.3 0 0.0 0 0.0 0 0.0 0 0.0
  T3a 33 3.9 39 4.6 138 100.0 138 100.0 38 100.0 38 100.0
Grade - 0.51 -
  Low-grade 0 0.0 0 0.0 84 60.9 81 58.7 0 0.0 0 0.0
  High-grade 840 100.0 840 100.0 42 30.4 39 28.3 38 100.0 38 100.0
  Unknown/Not applicable 0 0.0 0 0.0 12 8.7 18 13.0 0 0.0 0 0.0
Followup (months) 0.57 0.09 0.34
  Median (IQR) 69 (48–83) 70 (48–83) 68 (47–82) 63 (41–83) 61 (32–81) 50 (22–82)
Open in a new tab

HG/UH: high-grade or unfavorable histology; IQR: interquartile range; PN: partial nephrectomy; RCC: renal cell carcinoma; RN: radical nephrectomy; T3: pathological T3a stage.

Median followup in months was 69 (interquartile range [IQR]) 48–83), 68 (IQR 47–82), and 61 (IQR 32–81) for the HG/UH, T3, and combined cohorts, respectively. Within the cohorts, followup did not differ significantly between the PN and RN groups (p>0.05).

RCC with HG/UH

At the time of last contact, 350 (21%) patients had died, including 8.0% (135/1680) treated with PN and 13% (215/1680) treated with RN. The five-year OS for the total cohort was 84% (Fig. 1). The five-year OS was 87% for PN and 81% for RN (p<0.01; Fig. 2A). PN was associated with a 34% decreased hazard of death (hazard ratio [HR] 0.66, 95% confidence interval [CI] 0.53–0.82; p<0.01; Table 3).

Fig. 1.

Fig. 1.

Open in a new tab

Kaplan-Meier survival curves for RCC with HG/UH (blue), T3 (green), or HG/UH and T3 (red) after radical or partial nephrectomy. HG/UH: high-grade or unfavourable histology; RCC: renal cell carcinoma; T3: pathological T3a stage.

Fig. 2.

Fig. 2.

Open in a new tab

Kaplan-Meier survival curves for PN (blue) and RN (red) for RCC with adverse pathologic features, HG/UH (A), T3 (B), or HG/UH and T3 (C). HG/UH: high-grade or unfavourable histology; PN: partial nephrectomy; RCC: renal cell carcinoma; RN: radical nephrectomy; T3: pathological T3a stage.

Table 3.

Cox proportional hazards for mortality after surgery for high-risk RCC

Variable HR 95% CI p value
A. HG/UH
  Surgery (reference = radical nephrectomy) <0.01
  Partial nephrectomy 0.66 0.53–0.82
B. T3
  Surgery (reference = radical nephrectomy) 0.01
  Partial nephrectomy 0.54 0.34–0.87
C. HG/UH and T3
  Surgery (reference = radical nephrectomy) 0.28
  Partial nephrectomy 0.66 0.30–1.42
  Age 1.03 1.00–1.07 0.07
  Sex (reference = female) 0.67
    Male 0.84 0.38–1.86
Open in a new tab

CI: confidence interval; HR: hazard ratio; RCC: renal cell carcinoma.

RCC with T3

At the time of last contact, 75 (27%) patients had died, including 9.8% (27/275) treated with PN and 17% (48/275) treated with RN. The five-year OS for the total cohort was 77% (Fig. 1). The five-year OS was 82% for PN and 71% for RN (p<0.01; Fig. 2B). PN was associated with a 46% decreased hazard of death (HR 0.54, 95% CI 0.34–0.87; p=0.01; Table 3).

RCC with HG/UH and T3

At the time of last contact, 29 (38%) patients had died, including 16% (12/76) treated with PN and 22% (17/76) treated with RN. The five-year OS for the total cohort was 65% (Fig. 1). OS was not significantly different between approaches (p=0.21; Fig. 2C). After adjusting for unbalanced covariates, namely age and sex, on Cox multivariable analysis, the results were unchanged (p=0.28; Table 3).

Discussion

High-risk RCC harbouring one or more APF has been treated increasingly with PN over the last decade, despite a lack of evidence of the safety of PN in this setting.6 Our results demonstrate superior or at least equivalent survival for patients with high-risk RCC treated with PN. For RCC with HG/UH or T3, PN was an independent significant predictor of improved OS, after adjusting for age, comorbidity burden, tumour size, and other risk factors for all-cause mortality. For RCC with both APF, the OS curve for PN was better than that for RN, but this difference was not statistically significant, likely due to insufficient power (n=76).

We used propensity-matching to address the inherent differences between patients undergoing PN and RN and to adjust for patient factors which may affect OS, including age and comorbidity. Optimal matching necessitated the exclusion of larger tumours, which had been treated mostly with RN; therefore, the median tumour size of the analyzed cohorts was only 3.0–3.7 cm (IQR 2.1–4.5). For this reason, although both PN and RN are being used to treat larger tumours, our study only addresses the comparative effectiveness of PN and RN for high-risk, clinical T1 tumours. Irrespective of RCC risk, numerous observational studies have demonstrated improved OS with PN.17–20 In a contemporary meta-analysis comparing PN and RN outcomes, PN was associated with a 19% reduction in all-cause mortality.2 A recent instrumental variable analysis, which adjusts for selection bias and confounders — both known and unknown — in observational studies, arrived at similar results, with a 16% reduction in all-cause mortality attributed to PN.21 This survival benefit is attributed to improved renal and cardiovascular functional outcomes with PN.3,17,18

Four observational studies, using data from 1976–2008, have examined the outcomes of PN for high-risk disease.11–14 These studies all showed comparable oncologic outcomes for PN and RN. Only two studies examined OS, finding no difference between PN and RN; however, these studies were underpowered.12,13 Within the largest published study, a population-based study using the Surveillance, Epidemiology, and End Results (SEER) database, Hansen et al failed to detect statistically significant cancer-specific mortality differences between PN and RN for large, high-grade, or T3 tumours. This study is consistent with our findings, suggesting that PN offers at least equivalent survival relative to RN for high-risk disease.14 However, due to its lack of information on baseline comorbidity burden, this study was unable to assess the effect and potential benefit of PN on OS.

Recently, the survival benefits of PN have been called into question. The only prospective, randomized, controlled trial to address the comparative effectiveness of PN and RN in the elective setting did not find a significant improvement in OS with PN.15 In fact, based on the initial intention-to-treat analysis in which 8–10% of the patients had high-risk disease, i.e., tumours >5 cm or ≥pT3 stage, PN resulted in inferior OS outcomes compared to RN. Despites its flaws, including poor accrual and 15% crossover between groups, this study represents the only level 1 evidence to date. The discrepancy between these results and those of prior observational studies may be explained by the effect of potential bias and confounders on OS comparisons between PN and RN in the elective setting.16,22

Our study has limitations. Due to its retrospective nature, our study is susceptible to bias. Although we adjusted for inherent biases in treatment selection using propensity-based matching, we were unable to adjust for unmeasured confounders not included in the NCDB, particularly preoperative renal function and unknown confounders, which may have biased our results. Furthermore, since cause-specific survival is not available in the NCDB, we relied on OS, which is a limitation. However, we adjusted for age and CCI in order to address the risk of non-cancer. Although this is the largest study to compare OS between PN and RN for high-risk disease, a larger dataset would have helped. Lastly, despite the use of advanced statistical methods and a large representative national cohort, our study does not substitute for a prospective, randomized trial. In the absence of a trial, our study represents best available evidence on the topic.

Conclusion

PN offers at least equivalent and possibly improved survival relative to RN in RCC patients with small (<5 cm), high-risk tumours. Due to the retrospective nature of the study, the OS benefits of PN for high-risk disease may be influenced by bias.

Footnotes

Competing interests: The authors report no competing personal or financial interests.

Disclaimer: The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methodology employed, or the conclusions drawn from these data by the investigator.

This paper has been peer-reviewed.

References

  • 1.Campbell SC, Novick AC, Belldegrun A, et al. Guideline for management of the clinical T1 renal mass. J Urol. 2009;182:1271–9. doi: 10.1016/j.juro.2009.07.004. [DOI] [PubMed] [Google Scholar]
  • 2.Kim SP, Murad MH, Thompson RH, et al. Comparative effectiveness for survival and renal function of partial and radical nephrectomy for localized renal tumours: A systematic review and meta-analysis. J Urol. 2012;188:51–7. doi: 10.1016/j.juro.2012.03.006. [DOI] [PubMed] [Google Scholar]
  • 3.Weight CJ, Larson BT, Fergany AF, et al. Nephrectomy induced chronic renal insufficiency is associated with increased risk of cardiovascular death and death from any cause in patients with localized cT1b renal masses. J Urol. 2010;183:1317–23. doi: 10.1016/j.juro.2009.12.030. [DOI] [PubMed] [Google Scholar]
  • 4.Sun M, Abdollah F, Bianchi M, et al. Treatment management of small renal masses in the 21st century: A paradigm shift. Ann Surg Oncol. 2012;19:2380–7. doi: 10.1245/s10434-012-2247-0. [DOI] [PubMed] [Google Scholar]
  • 5.Liss MA, Wang S, Palazzi K, et al. Evaluation of national trends in the utilization of partial nephrectomy in relation to the publication of the American Urologic Association guidelines for the management of clinical T1 renal masses. BMC Urol. 2014;14:101. doi: 10.1186/1471-2490-14-101. Evaluation of national trends in the utilization of partial nephrectomy in relation to the publication of the American Urologic Association guidelines for the management of clinical T1 renal masses. BMC Urol 2014;14:101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Maurice MJ, Zhu H, Kim SP, et al. Increased use of partial nephrectomy to treat high-risk disease. BJU Int. 2016;117:E75–86. doi: 10.1111/bju.13262. [DOI] [PubMed] [Google Scholar]
  • 7.Ficarra V, Righetti R, Pilloni S, et al. Prognostic factors in patients with renal cell carcinoma: Retrospective analysis of 675 cases. Eur Urol. 2002;41:190–8. doi: 10.1016/S0302-2838(01)00027-6. [DOI] [PubMed] [Google Scholar]
  • 8.Bretheau D, Lechevallier E, de Fromont M, et al. Prognostic value of nuclear grade of renal cell carcinoma. Cancer. 1995;76:2543–9. doi: 10.1002/1097-0142(19951215)76:12&#x0003c;2543::AIDCNCR2820761221&#x0003e;3.0.CO;2-S. [DOI] [PubMed] [Google Scholar]
  • 9.Keegan KA, Schupp CW, Chamie K, et al. Histopathology of surgically treated renal cell carcinoma: Survival differences by subtype and stage. J Urol. 2012;188:391–7. doi: 10.1016/j.juro.2012.04.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Delahunt B, Cheville JC, Martignoni G, et al. The International Society of Urological Pathology (ISUP) grading system for renal cell carcinoma and other prognostic parameters. Am J Surg Pathol. 2013;37:1490–504. doi: 10.1097/PAS.0b013e318299f0fb. [DOI] [PubMed] [Google Scholar]
  • 11.Jeldres C, Patard JJ, Capitanio U, et al. Partial vs. radical nephrectomy in patients with adverse clinical or pathologic characteristics. Urology. 2009;73:1300–5. doi: 10.1016/j.urology.2008.08.492. [DOI] [PubMed] [Google Scholar]
  • 12.Breau RH, Crispen PL, Jimenez RE, et al. Outcome of stage T2 or greater renal cell cancer treated with partial nephrectomy. J Urol. 2010;183:903–8. doi: 10.1016/j.juro.2009.11.037. [DOI] [PubMed] [Google Scholar]
  • 13.Weight CJ, Lythgoe C, Unnikrishnan R, et al. Partial nephrectomy does not compromise survival in patients with pathologic upstaging to pT2/pT3 or high-grade renal tumours compared with radical nephrectomy. Urology. 2011;77:1142–6. doi: 10.1016/j.urology.2010.11.058. [DOI] [PubMed] [Google Scholar]
  • 14.Hansen J, Sun M, Bianchi M, et al. Assessment of cancer control outcomes in patients with high-risk renal cell carcinoma treated with partial nephrectomy. Urology. 2012;80:347–53. doi: 10.1016/j.urology.2012.04.043. [DOI] [PubMed] [Google Scholar]
  • 15.Van Poppel H, Da Pozzo L, Albrecht W, et al. A prospective, randomized EORTC intergroup phase 3 study comparing the oncologic outcome of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol. 2011;59:543–52. doi: 10.1016/j.eururo.2010.12.013. [DOI] [PubMed] [Google Scholar]
  • 16.Tobert CM, Riedinger CB, Lane BR. Do we know (or just believe) that partial nephrectomy leads to better survival than radical nephrectomy for renal cancer? World J Urol. 2014;32:573. doi: 10.1007/s00345-014-1275-8. [DOI] [PubMed] [Google Scholar]
  • 17.Miller DC, Schonlau M, Litwin MS, et al. Renal and cardiovascular morbidity after partial or radical nephrectomy. Cancer. 2008;112:511–20. doi: 10.1002/cncr.23218. [DOI] [PubMed] [Google Scholar]
  • 18.Huang WC, Elkin EB, Levey AS, et al. Partial nephrectomy vs. radical nephrectomy in patients with small renal tumours — is there a difference in mortality and cardiovascular outcomes? J Urol. 2009;181:55–62. doi: 10.1016/j.juro.2008.09.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Zini L, Perrotte P, Capitanio U, et al. Radical vs. partial nephrectomy: Effect on overall and non-cancer mortality. Cancer. 2009;115:1465. doi: 10.1002/cncr.24035. [DOI] [PubMed] [Google Scholar]
  • 20.Weight CJ, Larson BT, Gao T, et al. Elective partial nephrectomy in patients with clinical T1b renal tumours is associated with improved overall survival. Urology. 2010;76:631–7. doi: 10.1016/j.urology.2009.11.087. [DOI] [PubMed] [Google Scholar]
  • 21.Tan HJ, Norton EC, Ye Z, et al. Long-term survival following partial vs. radical nephrectomy among older patients with early-stage kidney cancer. JAMA. 2012;307:1629–35. doi: 10.1001/jama.2012.475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Shuch B, Hanley J, Lai J, et al. Overall survival advantage with partial nephrectomy: A bias of observational data? Cancer. 2013;119:2981–9. doi: 10.1002/cncr.28141. [DOI] [PubMed] [Google Scholar]

Articles from Canadian Urological Association Journal are provided here courtesy of Canadian Urological Association

RESOURCES