Partial nephrectomy was originally indicated for patients with a solitary kidney, renal dysfunction, or bilateral disease. These patients are not good candidates for radical nephrectomy [11]. But many studies have supported the equivalent oncological outcome between radical and partial nephrectomy [12]. In addition, prevention of CKD has been considered as the great advantage of partial over radical nephrectomy as mentioned above [8–10]. Progression of chronic kidney disease is reported to increase the risk of death, cardiovascular events, and hospitalization in a large community-based population [13]. Whether postoperative CKD really induces cardiovascular events or non-RCC-related death has been debated. Huang et al. reported that radical nephrectomy was significantly associated with a 1.4 times higher risk of more cardiovascular events after surgery compared to partial nephrectomy (p < 0.05) [10]. Population-based studies including large numbers of patients show an increased risk of noncancer mortality with radical nephrectomy as compared to partial nephrectomy despite comparable RCC-related mortality between procedures [9, 14, 15]. However, only one randomized EORTC trial comparing postoperative survival between radical and partial nephrectomy shows similar cancer-specific and overall survival [16]. In the original report, there are several points to be criticized. One is a lack of results on renal function after surgery. Thus, this trial is statistically underpowered. Functional outcome was reported thereafter, and partial nephrectomy resulted in higher postoperative renal function than that after radical nephrectomy [17]. The randomized trial did not show any benefit of partial nephrectomy over radical nephrectomy in reducing the risk of overall survival. Nevertheless, renal function was preserved at a higher level after partial nephrectomy.

There may be several other advantages in partial nephrectomy apart from the reduction of the risk of non-RCC-related death. First, about 5 % of patients develop tumors on the contralateral kidney [18]. If the functioning kidney is preserved during the first surgery, the selection of treatment for contralateral tumors can be more flexible. Second, the proportion of benign tumors is reported to be about 20 % in Western countries [19, 20]. Performing radical nephrectomy on these patients with benign disease may constitute overtreatment. Third, preservation of renal function may allow the patients to undergo systemic therapy for other diseases with agents having renal toxicity, such as cisplatin [21].

Taking these results into consideration, the recommendation of the current guideline is partial nephrectomy for patients with stage 1 disease, especially for T1a tumors whose sizes are 4 cm or less [1, 22, 23]. Radical nephrectomy is considered only when partial nephrectomy is not technically feasible due to the anatomical complexity of the tumors.

Partial nephrectomy should be discussed as a standard care even for tumors of T1b, 4.1–7.0 cm [1]. Most studies support the preservation of renal function after partial nephrectomy [24–26]. However, renal function after partial nephrectomy for T1b tumors is slightly lower than that after nephron-sparing surgery for T1a tumors [26]. The advantage in overall survival from partial nephrectomy is even more unclear [26, 27]. Nephron-sparing surgery does have other advantages over radical nephrectomy as mentioned before. Thus, partial nephrectomy also should be considered for T1b tumors when technically feasible.

There has been controversy over which is the better procedure in terms of nephron-sparing surgery—partial nephrectomy or tumor enucleation. In partial nephrectomy, normal kidney margin is attached to the surroundings of the tumors. This concept is based on possible satellite lesions around the tumors. The incidence of satellite lesions has been reported to be 6–15 % within 1 cm from the tumor boundary [28, 29]. Recent studies, however, show no significant correlation between the thickness of normal kidney margin and local recurrence [30], and the incidence of local recurrence is likely to be very low at 1–2 % [31]. In tumor enucleation, tumors are excised without a margin of normal parenchyma [32, 33]. Tumor enucleation has some advantage in the maximum preservation of functional renal parenchyma and a low incidence of major bleeding and collecting system damage, thereby decreasing the incidence of complications such as urinoma and urinary fistula [33]. In addition, oncological outcomes are reported to be comparable between tumor enucleation and partial nephrectomy [34]. The incidence of positive surgical margin was not different between the procedures (0.2 % versus 3.4 %). Thus, either procedure can be considered acceptable.

Figure 6.2 shows the sequential change by year in the utilization of partial nephrectomy at Tokyo Women’s Medical University. Partial nephrectomy has been selected for about 95 % of patients with T1a disease in recent years. Radical nephrectomy for T1a disease is considered only when the patients are over 80 years old, have CKD stage 4 or 5, and are likely to develop end-stage kidney disease soon even after nephron-sparing surgery [35]. The proportion of partial nephrectomy for T1b tumors has increased over the years and reached up to 90 % in the year 2013. Thus, the more the technique improves for partial nephrectomy, the more we can extend the indication of the nephron-sparing procedure.

How do we select the approach—open, laparoscopic, or robotic procedure? According to the current guidelines, the standard procedure for partial nephrectomy is open surgery since laparoscopic partial nephrectomy (LPN) is associated with a higher incidence of complications and longer ischemic time (Table 6.1) [1]. However, recent technical innovations—including early unclamping, zero ischemia, and the accumulation of experience with LPN—have resulted in better residual renal function and lowered the incidence of postoperative complications [36, 37]. Based on these results, the current guidelines allow experienced surgeons to perform LPN [22]. In addition, robot-assisted laparoscopic partial nephrectomy (RAPN) has widely spread around the world [38]. The new technology of robotic partial nephrectomy has made suturing and excising easier and more precise compared to the laparoscopic procedure. Meta-analysis showed an equivalent cancer control and safety of RAPN to LPN and the advantage of RAPN in shortening ischemic time over LPN [39]. Thus, RAPN is more feasible and likely to be indicated for more patients with stage 1 than LPN.

Laparoscopic radical nephrectomy has shown a significant benefit over open radical nephrectomy with its reduction in intraoperative bleeding, complication rate, and shorter hospital stay [40]. Laparoscopic radical nephrectomy shows equivalent long-term oncological outcome to open radical nephrectomy [41]. Thus, laparoscopic radical nephrectomy is more preferable for stage 1 disease than open radical nephrectomy (Table 6.1) [22].

The most important thing to be considered is whether the removal of primary renal tumors by cytoreductive nephrectomy (CN) would be a benefit for patients with metastatic disease. The benefit of CN was established from randomized prospective results during the cytokine era when CN followed by interferon-alpha (IFN-α) significantly prolonged patient survival compared to IFN-α treatment without CN (Table 6.3) [86–88].

The role of CN remains to be determined in the era of targeted therapy. Currently, a randomized prospective study (the CARMENA study) is in process that is comparing the results of a group receiving CN followed by sunitinib to another receiving sunitinib without CN. [89] Thus, there is as yet no determinant evidence to support the role of CN in the era of targeted therapy. It is important to note, however, that the retrospective study demonstrates that CN still has an additional benefit in prolonging patient survival compared to systemic therapy with targeted agents alone [90, 91]. Choueiri et al. reported that their CN group showed a longer median overall survival of 19.8 months than that of their no CN group (19.8 versus 9.4 months, p < 0.01) [90]. The results from the International Metastatic Renal Cell Carcinoma Database Consortium show prolongation of overall survival in patients with CN compared to those without CN (20.5 versus 9.5 months, p < 0.001) [91]. The results from our institution are very similar to their report. The combination with CN prolonged overall survival to 17.1 months from the 8.5 months observed in patients treated with targeted therapy alone (p < 0.0001). CN significantly reduced the risk of patient death after adjusting patients’ risk with the criteria proposed by Heng et al. [92]. Thus, cytoreductive nephrectomy still has a role in the era of targeted therapy [22, 23].

It is important to understand which patients are more likely to benefit from CN. Table 6.4 shows possible prognostic factors influencing the survival of those who undergo CN in conjunction with targeted therapy. The retrospective study shows poor risk in the risk classification system proposed by Memorial Sloan Kettering Cancer Center (MSKCC), 70 % or lower in Karnofsky performance score, the advanced age of 75 years or older, and brain metastases as significant prognostic factors [90]. The degree of reduction of tumor burden by CN is also a prognostic factor. [93] Patients show longer survival when CN removes more than 90 % of the tumor burden. These factors are not contraindications for CN, but they should be taken into consideration when deciding whether CN should be performed.

After CN, systemic therapy and/or metastasectomy should be considered. Details regarding this are described in the following section.

Initially, resectability of metastatic lesions (metastasectomy) should be considered. There are no randomized trials examining the survival benefit of metastasectomy. However, the many retrospective studies show longer survival in patients who received metastasectomy than those without [94–96]. Although the grade of recommendation is not high in the current guidelines because of the lack of randomized trials (grade C), metastasectomy is recommended when metastatic lesions are resectable and the patient has a good performance status [22, 23].

Prognostic factors have been examined to help predict poor prognosis after metastasectomy. Table 6.5 shows the prognostic factors and risk stratification reported by recent representative studies. Naito et al. collected data from 556 patients and report four risk factors that predict poor outcome: grade 3 at the primary tumor, brain metastases, C-reactive protein level higher than 1.0 mg/dl, and incomplete resection [95]. The overall survival of favorable risk patients who meet 0 or 1 factor is 105.6 months, whereas that of patients with 3 or more factors is only 10.3 months. Tosco et al. analyzed the results from 109 patients and report the following five predictive factors: T stage 3 or higher, Fuhrman grade 3 or higher, disease-free interval of 12 months or shorter, non-pulmonary metastases, and multiple site metastases [96]. The overall survival of patients with 0 or 1 factors is more than 108 months, but for those with four or more factors is only 13 months. Alt el al. included 887 patients in their analysis and reported risk factors for poor prognosis after metastasectomy, such as incomplete resection, ECOG-PS 1 or higher, non-pulmonary metastases, and synchronous metastases to nephrectomy [94]. They did not perform risk stratification, but they emphasized complete resection as an important factor for selecting patients for metastasectomy. Eggener et al. reported that patients with poor risk by the MSKCC risk classification show minimum impact from metastasectomy [97]. Although there are no clear criteria for indicating metastasectomy, these factors need to be considered before performing one.

We need to keep in mind that these results are based on data from patients treated during the cytokine era. It remains to be determined whether metastasectomy is still beneficial to prolong patient survival during the targeted therapy era since there have been few reports supporting the benefit of metastasectomy. In our institute, 53 patients underwent metastasectomy after 2008 when the first agent of molecular-targeted therapy was introduced in Japan. Complete resection was performed in 43 patients and incomplete resection in ten. One hundred twenty-one patients were treated with systemic therapy alone. Figure 6.4 shows overall survival of patients with metastatic RCC after treatment for metastatic disease. Four-year overall survival of patients with complete resection is 82.2 %. This is significantly higher than that for those with incomplete resection (46.8 %) or no metastasectomy (19.4 %) (p < 0.001). These results are similar to those reported by Alt et al. who show incomplete resection for patients with multiple metastases provided longer survival than that of patients with no metastasectomy [94]. Thus, metastasectomy is likely to be beneficial even in the era of targeted therapy. This is also supported by results from Naito et al. who report that patients receiving metastasectomy and targeted therapy showed prolonged survival as compared to those not receiving targeted therapy (132.2 versus 74.1 months) [95]. Collective results also support the use of metastasectomy when we treat patients with metastatic RCC. However, when patients meet the conditions related to poor prognosis after metastasectomy or when the metastatic lesions are not considered resectable, systemic therapy should be the first choice.

Systemic therapy is mainstream for the treatment of patients with metastatic RCC. If patients are unlikely to benefit from metastasectomy, systemic therapy is initiated. Stratification of patients with the risk classification system proposed by Memorial Sloan Kettering Cancer Center (MSKCC) is necessary before starting treatment (Table 6.6). The MSKCC risk classification consists of five factors to predict poor prognosis, including time from initial RCC diagnosis to the start of systemic therapy of less than 1 year, low Karnofsky performance status (less than 80 %), hemoglobin of less than the lower limit of normal, corrected serum calcium higher than the upper limit of normal, and lactate dehydrogenase higher than 1.5 times the upper limit of normal [98]. This classification system is based on results from patients treated with interferon-alpha, but is still useful in today’s era of targeted therapy. Heng et al. proposed another risk classification system for patients treated with targeted therapy [92]. This classification consists of the following six variables: time from diagnosis to treatment of less than 1 year, Karnofsky performance status less than 80 %, hemoglobin less than the lower limit of normal, corrected calcium greater than the upper limit of normal, neutrophils greater than the upper limit of normal, and platelets greater than the upper limit of normal. If patients meet no risk factor, they are designated as favorable risk, one to two factors as intermediate risk, and three or more as poor risk. Both classification systems are effective in stratifying patients.

Another factor to be determined before starting systemic therapy is histological subtype. If prior nephrectomy has not been performed at the time of systemic therapy and if cytoreductive nephrectomy is not indicated because of poor patient condition, tumor biopsy should be considered. Histological subtype influences the choice of agents.

Table 6.7 shows the currently accepted algorithm based on evidence from clinical trials [99]. For treatment of naïve patients with clear cell histology and good/intermediate risk, standard treatment includes sunitinib, bevacizumab plus interferon-alpha, and pazopanib [100–103]. The use of cytokines, including high-dose interleukin-2 or sorafenib, is an alternative treatment option. Thus, these second options can be used when patients are unfit for the standard option. When patients are at poor risk, temsirolimus is recommended as standard care [104]. Sunitinib can be used as an alternative.