INTRODUCTION
Renal cell carcinoma (RCC) represents 2.2% (n=431,288) of all newly diagnosed cancers, with a mortality rate of 1.8% (n=179,368) according to 2020 GLOBOCAN data [1]. Over the past 2 decades, surgical approaches to RCC have evolved [2]. Compared to radical nephrectomy (RN), partial nephrectomy (PN) has demonstrated benefits in preserving renal function [3,4]. The principle of nephron preservation is thought to play a role in patient prognosis, prompting the increased use of PN [5-9]. Consensus indicates that PN should be the treatment of choice for T1-stage tumors when feasible [2]. However, the optimal approach for T2 tumors is still under debate. Although RN remains the gold standard for tumors of stage T2 or higher, several recent reviews have suggested that PN may represent a viable alternative for T2 tumors [3,10-12].
In South Korea, the Korean Renal Cell Carcinoma (KORCC) registry has been established as a multicenter RCC database [13]. This study aimed to leverage the KORCC database to aid in determining whether to perform PN or RN in patients with stage T2 RCC. By utilizing data from the KORCC registry for clinical T2 (cT2) RCC cases, the study retrospectively compared outcomes between PN and RN.
MATERIALS AND METHODS
The study was conducted in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines. The research focused on adult patients with cT2 RCC, identified from 8 tertiary hospitals in the KORCC database between 2003 and 2023. Specifically, patients with a postoperative pathologic confirmation of RCC were included. The exclusion criteria encompassed patients with a solitary kidney, bilateral renal tumors, pathologic stage N1 (pN1), or pathologic stage M1 (pM1). The patient cohort was subsequently divided into PN and RN arms based on the surgical method employed.
1. Variable Outcomes
Baseline characteristics such as age, sex, body mass index (BMI), history of diabetes mellitus (DM), history of hypertension (HTN), smoking history, American Society of Anesthesiologists physical status classification, serum creatinine level, C-reactive protein (CRP) level, neutrophillymphocyte ratio (NLR), Modification of Diet in Renal Disease Study equation-estimated glomerular filtration rate (eGFR), radiologic tumor size, RENAL nephrometry score (radius, exophytic/endophytic, nearness to collecting system or sinus, anterior/posterior, and location relative to polar lines), symptoms at diagnosis, tumor laterality, and cT stage were assessed. Perioperative outcomes, including operative time, estimated blood loss (EBL), ischemic time, intraoperative complications, and postoperative complications, were evaluated. Ischemic time was assessed exclusively for the PN arm. Postoperative complications were compared in terms of both overall and—more specifically— urologic complications. Within each category, comparisons were also made for Clavien-Dindo classification grade 3 or higher complications. Kidney function was evaluated at 1, 3, and 6 months postoperatively. The presence of de novo chronic kidney disease (CKD) was assessed at 6 months. De novo CKD is defined as the new postoperative occurrence of CKD, as evidenced by eGFR declining to 60 mL/min/1.73 m2 or lower. Recurrence, all-cause mortality, and RCC-specific mortality were evaluated in the comparison of survival outcomes. Additionally, 5-year survival rates for each of these outcomes were determined. Pathologic outcomes, including tumor size, pathological T (pT) stage, histologic type, proportion of Fuhrman grade 3/4, presence of sarcomatoid differentiation, necrosis, lymphovascular invasion, capsular invasion, and margin positivity, were evaluated. As a subanalysis, hazard ratios (HRs) for factors influencing overall survival (OS) and recurrence-free survival (RFS) were examined.
2. Statistical Analysis
Propensity score matching (PSM) was performed using the “MatchIt” R package with a 1:3 matching ratio and the ‘optimal’ method. The variables used for matching were the eGFR and RENAL score risk group. The independent t-test and chi-square test were employed to compare the PN and RN arms. Kaplan-Meier (KM) curves were used to analyze OS, cancer-specific survival (CSS), and RFS. Cox regression univariable and multivariable analyses were conducted to identify risk factors associated with OS and RFS. The variables included in the Cox regression analysis were age, sex, BMI, history of DM, history of HTN, smoking history, CRP level, NLR level, PN operative method, pT stage, Furhman grade 3/4, sarcomatoid differentiation, necrosis, capsular invasion, and lymphovascular invasion. All statistical analyses were performed using IBM SPSS Statistics ver. 29.0 (IBM Co., Armonk, NY, USA) and R Statistics ver. 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria). A p-value of less than 0.05 was considered to indicate statistical significance.
3. Operative Methods
The surgical procedures included open, laparoscopic, hand-assisted laparoscopic, and robot-assisted laparoscopic methods. The open approach encompassed both retroperitoneal and transperitoneal techniques. Conversely, the laparoscopic and hand-assisted laparoscopic procedures were exclusively performed using the transperitoneal route, with 3 to 4 ports. Similarly, robot-assisted laparoscopic procedures were conducted transperitoneally, utilizing 4 ports along with an additional 1 or 2 assistant ports. All PNs were performed under warm ischemia, employing either bulldog or Satinsky clamps.
RESULTS
A total of 740 patients were included in this study. Prior to PSM, the PN arm comprised 53 patients, while the RN arm included 687 patients. Baseline characteristics before PSM are presented in Supplementary Table 1. The PN arm demonstrated a significantly higher eGFR compared to the RN arm (92.0 mL/min/1.73 m2 vs. 77.0 mL/min/1.73 m2, p<0.001), along with a significantly lower RENAL nephrometry score (9.0 vs. 10.1, p<0.001). Additionally, the PN arm had a substantially higher percentage of clinical T2a (cT2a) cases (86.3% vs. 73.2%, p=0.041). After PSM, the PN arm consisted of 44 patients, while the RN arm included 132. Baseline characteristics following PSM are detailed in Table 1. Following PSM, no significant differences were noted between the PN and RN arms, with the exception of laterality (68.2% vs. 48.5%, p=0.023) (Table 1).
1. Perioperative Outcomes
Table 2 delineates the perioperative outcomes for each arm. The operative time was longer for the PN arm (169.0 minutes vs. 150.4 minutes, p=0.042). No significant difference was found in the type of operation; however, the use of the robotic method was more prevalent in the PN arm (79.5% vs. 44.7%). The intraoperative complication rate was lower in the PN arm (2.3% vs. 12.9%, p=0.044), whereas the postoperative complication rate was higher in that arm (22.7% vs. 9.8%, p=0.028). Furthermore, complications of Clavien-Dindo classification grade III or higher were significantly more frequent in the PN arm (11.4% vs. 0.08%, p<0.001). Urological complications of Clavien-Dindo classification grade III or higher were also more prevalent among the arm receiving PN (6.8% vs. 0%, p=0.015). Complication rates for open and minimally invasive surgical procedures are available in Supplementary Table 2.
At follow-up, the PN arm displayed significantly better renal function. The incidence of de novo CKD in the sixth postoperative month was 3.1% in the PN compared to 37.6% in the RN arm. This indicates that the incidence in the RN arm was over 12 times higher than that in the PN arm (p<0.001).
2. Pathology
Pathologic outcomes are presented in Table 3. A significant difference was noted in the distribution of pT stages between the 2 arms (p=0.006), with PN displaying a higher proportion of pT1 (45.5% vs. 21.2%). Other microscopic findings, including positive margins (0.0% vs. 0.8%, p>0.999), Fuhrman grade 3/4 (69.8% vs. 77.6%, p=0.302), sarcomatoid differentiation (7.0% vs. 11.3%, p=0.420), necrosis (25.0% vs. 39.3%, p=0.088), lymphovascular invasion (2.3% vs. 6.8%, p=0.259), and capsular invasion (30.2% vs. 37.5%, p=0.389), were not significantly different between the arms.
3. Oncologic Outcomes
The KM curves revealed no significant differences in OS, CSS, or RFS between the 2 arms (p=0.192, p=0.289, and p=0.471, respectively) (Fig. 1). In the 5-year survival analysis, the RN arm exhibited a 5-year OS rate of 94.4%, compared to a 100% rate in the PN arm. In comparison, the RN arm displayed a 5-year CSS rate of 96.3%, while the PN arm demonstrated a rate of 100%. Regarding RFS, the RN arm had a rate of 80.8%, and the PN arm had a rate of 93.1%.
4. Subanalysis
Following Cox regression analysis for OS, no significant risk factors for OS were identified, perhaps due to an insufficient number of events for analysis. The results of the HR analysis for RFS in cT2 RCC are presented in Table 4. Univariable Cox regression analysis revealed significant associations for CRP levels (HR, 1.128; 95% confidence interval [CI], 1.047-1.216), necrosis (HR, 2.572; 95% CI, 1.150-9.688), capsular invasion (HR, 2.674; 95% CI, 1.172- 6.102), and lymphovascular invasion (HR, 5.047; 95% CI, 1.133-22.470). PN did not significantly impact RFS (HR, 0.675; 95% CI, 0.230-1.979). However, in the multivariable Cox regression analysis, CRP levels remained a significant factor (HR, 1.127; 95% CI, 1.045-1.215).
DISCUSSION
This study compared the outcomes of PN and RN for the treatment of cT2 RCC. Survival analysis using KM curves for OS, CSS, and RFS revealed no significant differences between arms. Subanalysis demonstrated that the use of PN did not significantly influence RFS; however, RFS was significantly associated with CRP levels. Urologic complications, including acute kidney injury and urine leakage, contributed to a greater incidence of Clavien-Dindo classification grade III or higher complications in the PN arm. Urine leakage was classified as a Clavien-Dindo classification grade III complication due to the need for additional intervention, specifically the placement of a ureteral stent. However, the PN arm included only 1 case of de novo CKD at 6 months postoperatively, in contrast to its occurrence in nearly onethird of the patients in the RN arm.
In a systematic review and meta-analysis, Huang et al. [14] revealed that compared to RN, PN—a more technically demanding surgery—necessitates a longer operative time and results in greater intraoperative blood loss. In our study, the operative time was significantly longer for PN than RN; however, EBL did not differ significantly between the arms. Despite the absence of a significant difference in EBL, the higher frequency of complications underscores the technically challenging nature of the procedure.
Several studies have examined survival outcomes for cT2 RCC. Bertolo et al. [15] reported an all-cause mortality rate of 3.3%, a cancer-specific mortality rate of 0.8%, and a recurrence rate of 10% after robot-assisted laparoscopic PN for cT2a RCC. Klett et al. [11] found no significant difference in OS, CSS, or metastasis-free survival between RN and PN. Similarly, a multicenter study by Kopp et al. [16] on cT2 RCC revealed no significant difference in OS, CSS, or progressionfree survival. An investigation of cT2 RCC conducted by Alanee et al. indicated that PN and RN yielded comparable results in terms of CSS [12]. Amparore et al. [17] compared the 5-year survival of cT2 RCC between PN and RN and found no significant difference in all-cause or cancer-specific mortality. However, some research suggests that PN may be associated with superior OS outcomes [18]. After reviewing several studies on cT2 RCC, we conclude that the oncologic outcomes of PN and RN are similar. In our study, KM curves also displayed no significant differences between PN and RN in OS, CSS, or RFS.
The results of this study suggest a lower complication rate than those reported in previous research [15,19-21]. PN carries an inherently higher risk of complications compared to RN because it involves breaching the collecting system, potentially leading to the development of pseudoaneurysms after renorrhaphy. De novo CKD has been shown to occur in approximately half of cases following RN, whereas the incidence after PN is more variable and generally lower [3,22]. In our study, de novo CKD occurred in approximately one-third of the patients after RN, while only 1 case arose following PN. CKD is associated with reduced survival in patients with cancer, as noted by Lee et al. [23].
According to O’Brian et al. [24], elevated CRP levels are indicative of poor prognosis in urological cancers. Hu et al. [25] reported that a higher CRP concentration was associated with comparatively poor OS and CSS in cases of RCC. Additionally, elevated CRP levels have been associated with worse CSS and progression-free survival in patients with localized RCC [25]. Wu et al. [26] found that both OS and CSS were significantly associated with pooled HR, suggesting a connection between increased CRP and reduced survival in RCC. In our study, we confirmed an association between CRP level and RFS.
This study is notable for its utilization of data from multiple tertiary hospitals in Korea. However, the multicenter nature of the data may have introduced variations in the techniques employed for PN. The accuracy of renal function tracking was also compromised by incomplete follow-up data. Additionally, several biases are inherent to the retrospective nature of the study and the reliance on surgical data from tertiary hospitals. The exclusion of pN1 and pM1 cases further restricted the comparison of surgical outcomes in advanced RCC and limited the survival analysis. Preoperatively determining nodal status is challenging, which is another inherent limitation of retrospective study designs. The methods used, such as open surgery, handassisted laparoscopic surgery, pure laparoscopic surgery, and robotic-assisted laparoscopic surgery, are varied and may not reflect current surgical trends. We also attempted a Cox regression analysis for OS, but it did not yield meaningful results, presumably due to an insufficient number of events; thus, it was difficult to analyze the HRs for OS. Finally, the PSM process resulted in the removal of many cases, which likely contributed to the challenges regarding Cox analysis. Nevertheless, the use of PSM to address confounding variables and balance the data is another strength of the present study.
CONCLUSIONS
In cT2 RCC, PN demonstrated a lower incidence of de novo CKD compared to RN, with no significant differences in survival outcomes. These findings suggest that PN could represent a viable alternative to RN for select patients with cT2 RCC, offering comparable survival outcomes while reducing the risk of de novo CKD. Additional research and prospective studies are required to validate these findings in advanced RCC.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
Print
