Epidemiology

Kidney cancer comprises approximately 3% of new cancer cases and 3% of cancer deaths in the United States each year.^1,5 In 2010 there were approximately 58,240 new cases of kidney and renal pelvis cancer diagnosed in the United States and approximately 13,040 deaths attributed to kidney and renal pelvis cancer. Over 90% of kidney cancers among adults are classified as RCC.⁵

RCC is typically diagnosed in the seventh decade of life, with a median age at diagnosis of 65 years; however, it has been observed in children as young as 6 months old. According to data from the Surveillance, Epidemiology and End Results (SEER) program, the overall incidence of RCC in the United States has increased at a rate of 2% to 3% per year since the early 1970s.^6,7 Although the greatest increase in RCC incidence has been among localized tumors, early detection of subclinical tumors cannot fully explain the rising incidence of RCC. There is also an upward trend for more advanced tumors as well as an overall increase in mortality rates.^6,8–10

There is a strong gender preponderance, with incidence rates in men approximately twice that of women.¹ Men account for roughly two thirds of RCC diagnoses and deaths in the United States.¹¹ There is also a notable variability in RCC incidence across racial and ethnic groups. The highest rates are reported for whites and African Americans, whereas the lowest rates are reported for Asian Americans. Between 1975 and 1998, incidence rates among African Americans increased by 4.5% compared with only 2.9% in whites. Globally, the incidence rate of RCC is the highest in North America and Scandinavia and the lowest is in Asia and South America.¹²

Etiology

Cigarette smoking and obesity are two well-established risk factors for RCC, and each may increase the risk of RCC roughly twofold. These factors may account for 40% of all RCCs diagnosed in the United States.^13–20 Although hypertension is often included as a risk factor, this association is considered less certain, given its correlation with both smoking and obesity.^13,16–18 Some reports suggest that hypertension may increase the risk of RCC approximately twofold.^21,22 There may be an inverse association (i.e., protective effect) of RCC development with moderate alcohol consumption and physical activity, as well as a positive association with a history of urinary tract infections.^23–30 There are few consistent associations of RCC and dietary factors.^18,31 Consumption of vegetables may be associated with reduced risk, whereas red meat consumption may slightly increase the risk of RCC^32–34; however, a recent meta-analysis did not support an independent association of red meat consumption and RCC.³⁵ Finally, occupational exposures to asbestos, gasoline, lead, and cadmium, use of Thorotrast, and treatment with phenacetin, as well as a history of end-stage renal disease or acquired cystic kidney disease, have been associated with increased risk of RCC.^36–41

A positive family history has been reported to increase the risk of RCC as much as twofold.⁴² Although genes that are associated with inherited forms of RCC have been identified, there are currently very few published data regarding the role of germline genetic susceptibility and sporadic RCC.^17,43,44 Small case-control studies have demonstrated that individuals carrying polymorphisms in specific genes (particularly phase I enzymes) may be at increased risk of RCC.^45–48 The results of these studies have yet to be validated. Large genome-wide association studies of RCC risk are not available at this time.

Epidemiology

Approximately 5% of urothelial neoplasms occur in the kidneys and ureters.⁴ Data from the SEER program suggest that the age-adjusted annual incidence of renal pelvic and ureteral cancers is 0.73/100,000 person-years.⁴⁹ Similar to RCC, there is a strong male predominance. The incidence increases with age, with the peak age at diagnosis in the sixth and seventh decades of life. Primary tumors of the ureter occur only half as frequently as those of the renal pelvis. The incidence of UUT urothelial cancers has been increasing.^49,50 Disease-specific annual mortality is greater in black than in white individuals and in women than in men.

Etiology

Risk factors and environmental agents that cause bladder cancer appear to play similar roles in the development of upper tract urothelial cancers.^51–55 Patients with carcinomas of the renal pelvis have more than a 30% to 80% risk of developing metachronous or synchronous bladder cancer. The risk of developing UUT malignancy after treatment of superficial bladder cancer is up to 9.8%.⁵⁴ The median time to the development of secondary UUT cancer after bladder cancer diagnosis is 33 months. Contralateral metachronous UUT cancer is relatively uncommon, occurring in 3% to 6% of patients.^53,56

Cigarette smokers have a 2.6- to 7.2-fold increased risk of UUT cancer.^57–59 Chronic use of analgesics that contain phenacetin has been implicated as a risk factor.^59–61 Capillarosclerosis is a pathognomonic change seen in patients with long-standing abuse of compound analgesics.⁶² Ingestion of Chinese herbs containing Aristolochia fangchi, which is used for weight reduction, results in a nephropathy characterized by progressive renal fibrosis and increased risk of urothelial cancer.^63,64 There is a 57- to 100-fold increase in the incidence of UUT cancer in Balkan countries affected by an endemic nephropathy.^65,66 The renal pathology is an interstitial nephritis. Patients with tumors from the endemic region present with higher-grade disease and more solid growth pattern compared with those from nonendemic regions.⁶⁷ The cause of the endemic nephropathy is postulated to be high concentrations of radon and minerals in the drinking water in the area. Occupational exposure to organic chemicals is associated with a higher risk of UUT cancers for workers in the chemical, petrochemical, or plastic industries.⁵⁷ Patients with a history of kidney or ureteral stones have a 2.5-fold increase in risk of renal and ureteral cancers, suggesting that chronic irritation and infection may play a role.⁶⁸ There are reports of kindred with familial urothelial cancers of the UUT.^69–71

Pathology

In 1997, an international consensus conference on RCC sponsored by the Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC) outlined recommendations for the classification of RCC.⁷⁷ The classification system originally proposed at the Heidelberg conference in 1996 was adopted.⁷⁸ RCC is categorized as clear cell, papillary, chromophobe, or collecting duct subtypes. RCC that does not fall into one of these four groups is classified as “renal cell carcinoma, not otherwise specified.” Granular cell RCC was excluded from the classification because it encompassed oncocytoma (benign kidney tumor) as well as chromophobe and clear cell RCC and was not specific to a single subtype. This classification system recognizes that RCC consists of several histologic subtypes with distinct morphologic and genetic characteristics. Previous reports have demonstrated significant differences in patient outcome based on histologic subtypes.^79–81 Patients with clear cell RCC have a worse prognosis compared with patients with papillary and chromophobe RCC. There is no statistically significant difference in outcome between patients with papillary RCC and those with chromophobe RCC.⁸²

Furthermore, features predictive of a poorer prognosis, including histologic tumor necrosis and sarcomatoid differentiation, have been shown to differ by subtype.^80–83 Histologic tumor necrosis is noted in 20% to 45% of RCC patients, whereas sarcomatoid differentiation occurs in 5% to 15% of patients.^80–82

Fuhrman and colleagues⁸⁴ developed a nuclear grading system based on the size and appearance of nuclei and nucleoli. The Fuhrman nuclear grade is predictive of outcome in patients with RCC.

Algorithms have been developed to predict outcomes for patients with RCC. Using retrospective data from 1801 patients with clear cell RCC, Frank and associates⁸⁵ from the Mayo Clinic developed a predictive model using an SSIGN scoring system that incorporated stage, tumor size, nuclear grade, and histologic tumor necrosis. Ten-year cancer-specific survival can be estimated based on a patient’s SSIGN score. Kattan and coauthors⁸⁶ developed a nomogram to predict 5-year probability of treatment failure using presenting symptoms, histology, tumor size, and stage. Zisman and colleagues⁸⁷ developed a clinical outcome algorithm based on 814 patients treated at the University of California, Los Angeles, using stage, Fuhrman grade, and Eastern Cooperative Oncology Group (ECOG) performance status. This algorithm has been validated by two international multicenter studies.^88,89

Pathways of Spread

RCC may spread by local extension through the renal capsule into the perinephric fat or the adrenal gland or by direct extension through the renal vein to the inferior vena cava (occasionally reaching the right atrium). Regional lymphatic drainage includes renal hilar, paracaval, aortic, and retroperitoneal lymph nodes. RCC may also spread by blood-borne metastases to the lung, soft tissue, bone, and brain or by retrograde venous drainage to the ovary or testis.⁹⁰

Pathology

Transitional cell carcinoma (TCC) accounts for the majority of renal pelvis and ureter neoplasms.⁹¹ Squamous cell carcinoma (SCC) accounts for about 4% of all cases.⁹² Although patients with SCC tend to present with more advanced-stage diseases and worse prognosis, stage for stage, there is no significant difference in prognosis between TCC and SCC. Other cell types such as adenocarcinoma and sarcoma are very rare.

Both stage and tumor grade are important prognostic factors for survival.^93–98 Lymph node metastasis and lymphovascular space invasion are associated with stage and predict for worse outcome.^99,100

Gross tumor architecture (sessile vs. papillary growth pattern) is an independent predictor of tumor relapse and cancer-specific mortality.¹⁰¹ Sessile growth pattern is associated with higher tumor grade, more advanced stage, and nodal metastasis. A history of prior bladder cancer has also been associated with worse disease-specific survival (DSS).¹⁰²

Pathway of Spread

Involvement of multiple areas of the urothelial epithelium by TCC, either synchronously or metachronously, is common in patients with renal pelvis and ureteral cancers. As the tumor progresses, it invades the surrounding tissues with direct extension through the ureteral wall into periureteral tissues or to adjacent renal parenchyma as well as extrarenal tissues. Lymph node metastasis is the most common site of metastasis in UUT cancers. In a multi-institutional series of 1363 patients,¹⁰³ lymph node metastasis was shown to increase with advancing pathologic stage: less than 1% for T0/Ta/Tis, 2% for T1, 8% for T2, 17% for T3, and 46% for T4.

Kondo and coauthors¹⁰⁴ outline the distribution of nodal metastatic sites. For tumors of the right renal pelvis, the primary metastatic sites are the right renal hilar, paracaval, and retrocaval nodes. Tumors of the upper two thirds of the right ureter primarily metastasize to the retrocaval and interaortocaval nodes. Tumors of the left renal pelvis metastasize to the left renal hilar and para-aortic nodes. Tumors of the upper two thirds of the left ureter primarily metastasize to the para-aortic nodes. Tumors of the lower ureter primarily metastasize inferiorly to the aortic bifurcation. The finding of lymph vessel invasion in the tumor is predictive of regional lymph node metastasis.¹⁰⁵ Nodal metastasis is a predictor for distant metastasis and is associated with worse survival.^99,103

Patients with high-grade or advanced-stage disease are at significant risk of developing distant metastasis.^93,106 Common sites of distant failure include lung, liver, and bone.⁹⁶

Clinical Manifestations

RCC may present as an incidental finding on imaging or may demonstrate multiple presenting signs and symptoms caused by either local extension or metastatic disease. Because of these multiple symptoms, RCC has been called the “internist’s tumor.” Symptoms resulting from local tumor extension include hematuria, abdominal pain, and a flank mass. However, this “classic triad” of symptoms occurs in only about 10% of cases.⁹⁰ Symptoms caused by metastatic disease include fever, weight loss, and night sweats. Two percent of male patients present with a varicocele, usually left sided, as a result of impaired drainage of the testicular vein.¹³¹

Paraneoplastic syndromes occur in about 30% of patients with RCC. Symptoms include hypertension, hypercalcemia, pyrexia, and hepatic dysfunction.^132,133

Patient Evaluation

The widespread use of CT, MRI, and ultrasonography to evaluate the upper abdomen has significantly increased the incidental finding of RCC. Currently, 25% to 40% of diagnoses are made after the incidental detection of a renal mass.^134,135 These tumors usually are smaller and, therefore, are more likely to be surgically resectable.^136,137

The diagnosis of RCC is usually based on clinical and radiologic findings and pathologically confirmed at the time of nephrectomy. The staging evaluation includes a history and physical examination, complete blood cell count, serum chemistries, determination of lactate dehydrogenase level, urinalysis, chest radiography, and CT or MRI of the abdomen and pelvis (Table 54-1). If clinically indicated, bone scintigraphy and brain MRI may be considered. CT can predict tumor extent preoperatively in 90% of cases.¹³⁸ MRI has been demonstrated to be superior to CT in determining inferior vena cava tumor extent.¹³⁹ Magnetic resonance angiography can be used to evaluate vasculature preoperatively, especially when nephron-sparing surgery is being considered for patients with bilateral cancers.

TABLE 54-1 Diagnostic Evaluation/Algorithm for Kidney Carcinoma

Staging

Two staging systems are used to define disease extent in RCC. Historically, Robson’s modification of the Flocks and Kadesky system was used.¹⁴⁰ More recently, the AJCC Staging and End Results Reporting Classification has been used (Table 54-2).¹⁴¹ This system appears to be superior to the Robson system because it delineates more clearly local tumor extent and quantifies lymph node involvement.⁷⁴

TABLE 54-2 American Joint Committee on Cancer Staging Classification System for Kidney and Ureteral Carcinoma, 7th Edition (2010)

* Laterality does not affect the N classification.

From Edge S, Byrd D, Compton C: AJCC Cancer Staging Manual. New York, Springer-Verlag, 2010.

Clinical Manifestations

The most common presenting symptom is gross or microscopic hematuria, which occurs in about 75% to 80% of patients. Flank pain is present in 27% to 35%.^94,142 The pain may mimic that of ureteral calculus. Urinary symptoms such as frequency and dysuria are reported in 25% to 50% of patients. A palpable mass may be found in up to 10% of patients and is usually a hydronephrotic kidney. Constitutional symptoms such as malaise and weight loss are frequently seen in patients with extensive disease or metastases.

Patient Evaluation

Patients are evaluated with urine cytology, CT urography, and ureteroscopy. Cytology of voided urine detects 35% to 59% of UUT urothelial carcinomas and is more useful in higher-grade cancers.⁹⁴ Urine specimens obtained by ureteral catheterization, or by washings of the upper tract, improve the positive yield of the study. Upper tract urine cytology is positive in up to 70% of patients.¹⁴³ The positive predictive value of renal pelvis washing for high-grade cancer is 93% but is 43% for low-grade cancer.¹⁴⁴ Radiologic findings of these tumors include a radiolucent filling defect or obstructive hydronephrosis. Ureteroscopy or percutaneous nephroscopy is performed if exfoliative cytology or radiologic studies show suspicious findings. Ureteroscopic biopsy provides the tissue diagnosis in up to 89% of cases.¹⁴⁵

Additional studies include complete blood cell count and chemistries and chest radiography or chest CT. CT of the abdomen and pelvis is useful to assess the local extent of disease and intra-abdominal metastasis, although it is not a sensitive or specific test for nodal staging.¹⁴⁶ If there is any concern about the function of the contralateral kidney, a renal scan should be performed before nephroureterectomy.

Staging

The current TNM staging system is listed in Table 54-2. Another staging system that is often used is the Jewett-Strong classification, which was originally used for bladder cancer: stage 0, limited to the mucosa; stage A, submucosal infiltration; stage B, invasion into but not through the muscle wall; stage C, invasion through the wall; stage D, lymphatic or distant metastasis.¹⁴⁷ Tumor grade has a strong association with disease stage (Table 54-3). Both stage and tumor grade are important prognostic factors for survival. Table 54-4 illustrates the impact of stage and grade on the overall survival.

Surgery

Complete surgical resection is the only possible curative treatment for RCC. A radical nephrectomy typically includes removal of the ipsilateral kidney with perinephric tissues including Gerota’s fascia and the ipsilateral adrenal gland. Survival results for treatment with standard open radical nephrectomy^{140,148–153} are delineated in Table 54-5. In a report of 643 patients,¹⁵³ the 5-year DSS was 91% for those with stage I disease, 74% for those with stage II disease, 67% for those with stage III disease, and 32% for those with stage IV disease based on the 1997 TMN staging criteria. In a Mayo Clinic series of 1547 patients,¹⁵⁴ the 10-year cause-specific survival (CSS) was 91% for T1, 70% for T2, 53% for T3a, 42% for T3b, and 43% for T3c tumors.

Removal of the adrenal gland is not required for RCC involving the middle or lower poles. However, upper pole cancers may necessitate adrenal gland removal with nephrectomy depending on the size, pathologic type, and extent of invasion of the renal tumor. Preoperative CT has demonstrated 99.6% specificity and a 94.4% negative predictive value in predicting adrenal gland involvement.155,156 In a review of 511 radical nephrectomies with adrenal gland removal,¹⁵⁵ the incidence of renal cancer involvement with the adrenal gland was 5.7% and T1-2 tumors had an incidence of adrenal involvement of 0.6%.

A radical nephrectomy can include resection of hilar lymph nodes and occasionally includes regional lymph node dissection. Although regional lymph node dissection provides additional prognostic information, it has not been proven to prolong survival.⁷⁴ The European Organization for Research and Treatment of Cancer (EORTC) conducted a phase III trial evaluating the role of lymph node dissection.¹⁵⁷ Seven hundred seventy-two patients were randomized to a radical nephrectomy with or without a complete lymphadenectomy. All patients were clinically node negative based on preoperative CT. The incidence of positive lymph nodes was 4%. After a median follow-up of 12.6 years there were no differences in overall survival (OS), time to progression of disease, progression-free survival (PFS), or complications. One criticism of this trial is that 70% of patients were T1 or T2 category, suggesting a low risk of lymph node involvement. It is difficult to conclude whether patients with locally advanced tumors would benefit from a lymph node dissection. Researchers at the Mayo Clinic reported a review of 955 patients who underwent a lymph node dissection as part of a radical nephrectomy.¹⁵⁸ High-grade tumor, presence of sarcomatoid features, tumor size greater than or equal to 10 cm, T3 category, and histologic tissue necrosis were significant predictors of lymph node-positive disease. The presence of these high-risk factors can help to determine which patients may benefit from an extensive lymph node dissection. Pantuck and colleagues¹⁵⁹ retrospectively analyzed 900 patients with RCC and concluded that a lymph node dissection offers no survival advantage in clinically node negative patients but is associated with improved survival and improved response to immunotherapy in node-positive patients.

Partial nephrectomy, or nephron-sparing surgery, has been performed for small tumors. The initial indications for this surgery were presence of bilateral tumors or a cancer involving an anatomic or functional solitary kidney. Local recurrence rate was low (<6%), and this procedure has subsequently been performed in patients with small tumors (<4 cm).¹⁶⁰ The incidence of local recurrence in these “electively” treated patients is 3% or less. Recent data from several institutions^161,162 suggest larger tumors (>7 cm) can also be considered for nephron-sparing surgery. Patients undergoing nephron-sparing surgery have a decreased cumulative incidence of chronic renal insufficiency compared with patients who undergo radical nephrectomy.^163,164

Laparoscopic radical nephrectomy and laparoscopic partial nephrectomy have become a commonly used surgical option for patients with RCC. Both pure and hand-assisted laparoscopic techniques for radical nephrectomy and nephron-sparing surgery achieve oncologic outcomes comparable to the standard open approach.^165–169 Laparoscopic approaches have demonstrated advantages regarding blood loss, length of hospital stay, pain medication usage, cosmetics, and patient recovery compared with the standard open approach of radical nephrectomy.^165,166

Ablation through radiofrequency heating and cryosurgical freezing are two frequently utilized techniques to obliterate suspicious small renal lesions.^170,171 Employment of these techniques through an open, laparoscopic, or percutaneous approach has been demonstrated.^170–172 A review of radiofrequency ablation and cryosurgical freezing with a laparoscopic or percutaneous approach¹⁷³ revealed that laparoscopic ablation using either technique had a 0% to 15% relapse rate with a follow-up of 5 years. The percutaneous approach for ablation also demonstrated similar recurrence rates, but follow-up was only 2 years. The overall complication rate was 5%.

In summary, radical nephrectomy continues to be the first line of therapy for apparent localized renal carcinoma. Less invasive techniques, such as laparoscopic radical nephrectomy, are becoming more popular. Nephron-sparing surgery is increasingly performed for patients with normal contralateral kidneys and for small to medium-sized tumors.

Adjuvant Irradiation

Although the use of postoperative irradiation has been evaluated both retrospectively and prospectively, its role remains controversial174,175–178 (Table 54-6). In a nonrandomized comparison of surgery with or without postoperative irradiation, Rafla¹⁷⁴ found a 5-year OS advantage for 94 patients undergoing postoperative irradiation versus 96 patients treated by nephrectomy alone (56% vs. 37%). These data were confirmed in an expanded updated analysis. The 5-year OS was 38% (40/105) for patients who received postoperative irradiation versus 18% (24/135) for those undergoing nephrectomy alone.¹⁷⁹

In a pseudo-randomized trial of 100 patients (randomization by odd vs. even date of birth), Finney¹⁷⁵ found that incidence of distant metastases, local recurrence, and survival were not affected by the addition of postoperative radiation therapy (55 Gy in 25 fractions in 5 weeks). These results are difficult to interpret because randomization was not blinded or stratified by risk factors.

A small prospective, randomized study of postoperative radiation therapy was conducted by the Copenhagen Renal Cancer Study Group.¹⁷⁶ Patients were randomized to nephrectomy alone (33 patients) versus nephrectomy and postoperative radiation therapy (32 patients). Postoperative irradiation consisted of 50 Gy in 20 fractions. The 5-year OS was 62% for nephrectomy alone versus 38% for nephrectomy and postoperative radiotherapy. Forty-four percent of the patients treated with radiation therapy had significant complications involving the stomach, duodenum, or liver that contributed to the death of 19% of these patients.

Four more recent retrospective studies evaluated the role of postoperative irradiation in patients at high risk for local recurrence after surgery for RCC, using multiple treatment fields and CT-based planning. Stein and associates¹⁷⁷ evaluated 147 patients with localized RCC who underwent a nephrectomy. Fifty-six patients received postoperative irradiation with a median dose of 46 Gy. With a median follow-up of 19 months, there was no statistical difference in OS. However, there were more local recurrences in nonirradiated T3 patients (37% in nonirradiated patients vs. 11% in irradiated patients). In a follow-up descriptive study of patients treated with radiation after a nephrectomy, Gez and coauthors¹⁸⁰ reported no benefit of postoperative radiation, even for T3 patients.

Kao and colleagues¹⁷⁸ described 12 patients treated with nephrectomy alone and 12 patients treated with nephrectomy and postoperative irradiation, with a median dose of 45 Gy. After a median follow-up of 5 years, no difference in DFS was noted. However, the local failure rate was 30% for the nephrectomy-alone patients and 0% for those who had postoperative radiation therapy. In another retrospective study, Ulutin and colleagues¹⁸¹ found no difference in OS but a significant difference in DFS in favor of radiation therapy for 26 patients treated with postoperative irradiation compared with 14 patients treated with nephrectomy alone.

Rabinovitch and associates¹⁸² performed a patterns-of-failure analysis by CT on patients with localized RCC who underwent a radical or partial nephrectomy without adjuvant therapy. The risk of local failure after a gross resection was approximately 5%. Positive surgical margins and positive lymph nodes had a significant impact on local failure, increasing the rate of local failure to 21%. The risk of distant metastasis after gross resection was 26%. Lymph node involvement and renal vein involvement were significant prognosticators of distant metastases. Approximately half of patients who experienced a local relapse also experienced distant metastases concurrently or as the first site of relapse. Therefore the risk of micrometastatic disease at the time of surgery must be weighed against the potential benefit of local irradiation.

In conclusion, the role of routine postoperative irradiation in the management of RCC is not established in patients with localized disease after gross total resection. However, the risk of local failure is higher in patients with positive margins and lymph node involvement, and postoperative irradiation may be considered.

Adjuvant Chemotherapy, Immunotherapy, and Molecularly Targeted Treatment

Adjuvant therapy is given to patients with high risk of systemic disease relapse after curative surgery. Three separate clinical trials evaluating adjuvant interferon-alfa after nephrectomy^183–185 failed to demonstrate a survival benefit. The Cytokine Working Group had conducted a high-dose intravenous bolus interleukin-2 (IL-2) trial versus observation in postnephrectomy patients,¹⁸⁶ and it did not show any survival benefit. A different approach was studied by Jocham and colleagues¹⁸⁷ using autologous tumor cell vaccine. Five hundred and fifty-eight patients with resected pT2-3b, pN0-3, M0 RCC were randomly assigned to vaccine compared with observation. At 4.5 years, PFS was 77% with the vaccine and 68% with observation. Its effect on OS would require longer follow-up. The role of targeted agents in the adjuvant setting is actively being investigated. One study in progress is the ECOG phase III study ECOG 2805, or the ASSURE study (Adjuvant Sorafenib or Sunitinib for Unfavorable risk REnal carcinoma).

Stereotactic Body Radiotherapy

The relative radioresistance of RCC and the previous experience with improvement in local control for patients treated with stereotactic radiosurgery for renal cell brain metastases have led to growing interests in the treatment of primary RCC with high-dose per fraction stereotactic body radiotherapy (SBRT). Beitler and colleagues¹⁸⁸ reported on nine patients with nonmetastatic primary RCC who refused a nephrectomy. The majority of patients received 40 Gy in five fractions over 15 days. With a median follow-up of 26.7 months, four of nine patients survived. Only one patient had a local failure, which consisted of a new primary tumor in the unirradiated portion of the kidney. All surviving patients had an initial tumor size less than 3.4 cm and no clinical evidence of nodal disease, renal vein, or inferior vena caval extension or penetration of Gerota’s fascia. Wersall and associates¹⁸⁹ treated eight patients with an inoperable primary or local inoperable recurrence. The median follow-up surpassed 58 months and only one of the patients experienced a local failure.

Svedman and colleagues¹⁹⁰ conducted a single-institution phase II trial to evaluate the efficacy and safety of SBRT for RCC. Thirty patients with inoperable primary or metastatic RCC received treatment to 82 lesions. The most common dose fractionations were 32 to 40 Gy in four or five fractions of 8 Gy, 30 to 40 Gy in three or four fractions of 10 Gy, and 30 to 45 Gy in two or three fractions of 15 Gy. With a median follow-up time of 52 months for living patients and 22 months for deceased patients, local control was 98%. Sixteen of 28 patients experienced side effects, 96% of which were grade I or II.

In summary, early results of SBRT for primary RCC suggest good local control rates in patients unable to undergo a nephrectomy. However, additional studies are needed to confirm the role of SBRT in the management of RCC.

Surgery

Nephroureterectomy with removal of the bladder cuff is the standard treatment for patients with UUT cancer. Locoregional relapses occur in 9% to 15% of patients with low-grade, low-stage disease and in 30% to 50% of those with high-grade, high-stage disease. The removal of the bladder cuff is recommended because ureteral stump recurrence is noted in 11% to 55% of patients if the stump is left in place.147,191,192 Patients with T2-4 disease should have lymphadenectomy. Nodal status is a significant prognostic factor and may be used to guide the decision for adjuvant treatment. For selected patients with distal ureteral cancers, distal ureterectomy and lymphadenectomy with reimplantation of ureter may be an option.

Laparoscopic nephroureterectomy has been shown to achieve similar disease control in many patients, with decrease in hospital stay and shorter recovery time, as compared with open nephroureterectomy.^193–197 In a report of 115 patients,¹⁹⁵ port-site metastasis occurred in one patient (0.9%). Minimally invasive surgery with robot-assisted ureterectomy and ureteral reconstruction appears to be safe and feasible in some patients, but the oncologic effectiveness of this procedure needs to be confirmed.¹⁹⁸

There are conflicting reports96,97,199,200,201 about the efficacy of more conservative nephron-sparing surgery or partial ureterectomy, which often can be done endoscopically. One study reported up to 62% recurrence, and careful follow-up is imperative for these patients. The selection criteria usually include low-grade and low-stage disease, superficial disease, small lesion size (<1.0 to 1.5 cm), absence of multifocal disease, solitary kidney, bilateral disease, or major medical comorbidities. These patients would need to be followed very closely so that nephroureterectomy can be offered if recurrence is detected.

Primary Irradiation

Primary irradiation is rarely used as initial treatment. The few published reports using primary irradiation are mostly case reports or small series. Some patients with gross residual disease after nephroureterectomy or who have recurrent disease are treated with external beam irradiation. In a series with 19 patients with unresectable disease,¹⁹¹ the median survival was 11 months for the 11 patients treated with radiation therapy versus 4 months for those who did not receive radiation. Two patients were disease free after receiving 45 Gy and 50.4 Gy, respectively, for gross residual disease after surgery. Batata and coauthors¹⁴⁷ reported one of eight patients had long-term disease control after EBRT for gross residual disease after surgery, using doses of 10 Gy in 5 fractions to 60 Gy in 35 fractions.

Although combined chemotherapy and radiotherapy has been used successfully as primary treatment for bladder cancer, there are little data for such an approach for UUT cancers. Combined chemoradiation would seem reasonable for patients who are not surgical candidates. However, many patients with medical comorbidities that make them ineligible for surgery may also be poor candidates for cisplatin-based chemotherapy. Studies evaluating combined chemoradiation are warranted.

Adjuvant Irradiation

The risk of local relapse correlates with the stage of disease and tumor grade and can be up to 45% despite radical surgery. In a study by Cozad and colleagues,¹⁹¹ the 5-year local control rate was 90% for grade 1 and 2 tumors and 41% for grade 3 and 4 tumors, respectively. For stage I and II disease, the 5-year local control rate was 83%, compared with 52% for stage III disease. About half of the local relapses presented initially as isolated site of relapse. Distant metastasis occurred in 19% of patients with stage I and II disease and in 53% with stage III disease. Distant metastasis was the predominant site of initial relapse for stage III disease. The risk of lymph node metastasis also increases with disease stage and tumor grade.¹⁰³ The pattern of failure demonstrates that patients with high-stage or high-grade disease have significant risks of both distant and local relapse, and many of them may have locoregional failure as the initial site of relapse.

Few studies have been published to evaluate the role of adjuvant irradiation in the management of UUT cancers.147,202,203 In a study by Cozad and colleagues,²⁰² 26 patients with T3 or T4 N0/+ disease had radical surgery, 9 of whom received adjuvant irradiation to a median dose of 50 Gy. Local failure occurred in 9 of 17 without and 1 of 9 with adjuvant irradiation (p = .07). The effect of adjuvant irradiation was seen only in high-grade tumors, because no local relapse developed in patients with low-grade disease regardless of whether adjuvant therapy was given. Metastasis developed in 4 of 9 and 8 of 17 patients with and without radiation therapy, respectively. Five-year OS was 44% with, and 24% without, adjuvant irradiation, respectively (p = .23). In a study by Brookland and Richter,²⁰³ adjuvant irradiation was given to 11 of 23 patients who had tumor grade 3 or 4, or pathologic stage C or D. The median dose was 50 Gy to the ureteropelvic bed, with or without inclusion of the para-aortic nodes. For the nonirradiated group, the median survival was 26 months and the local relapse rate was 45%, compared with 35 months and 11%, respectively, for those who received adjuvant irradiation. Other studies have questioned the benefits of adjuvant irradiation. In a series of 138 patients,⁹³ 45 received postoperative irradiation with a median dose of 50 Gy. The 5-year OS was 21% for those who received postoperative irradiation, compared with 33% for those who did not. However, patients who received postoperative irradiation had significantly higher T category and tumor grade. Maulard-Durdux and colleagues¹⁰⁶ reported the use of 45 Gy of adjuvant irradiation in 26 patients and concluded that local control and OS were similar to those of other surgical series, with distant metastasis as the predominant site of relapse.

Czito and colleagues²⁰⁴ reported a series of 31 patients with T3/4 or N+ disease that was treated with adjuvant irradiation to a median dose of 46.9 Gy. Adjuvant MVAC chemotherapy (methotrexate, vinblastine, doxorubicin [Adriamycin], cisplatin) with concurrent cisplatin was given to 9 patients. Five-year OS, DSS, locoregional control, and metastasis-free survival rates were 39%, 52%, 67%, and 48%, respectively. Compared with irradiation alone, the use of concurrent cisplatin improved the OS (67% vs. 27%) and DSS (76% vs. 41%). This study suggests that the addition of concurrent cisplatin to adjuvant radiotherapy improves the survival of patients with locally advanced disease.

Adjuvant Chemotherapy

Urothelial cancer of the bladder has been shown to be sensitive to chemotherapy, with improvement in OS when preoperative chemotherapy is given. However, there are very few studies evaluating the use of adjuvant chemotherapy for UUT cancers. Kwak and colleagues²⁰⁵ reported a series in which 32 patients received four or more cycles of cisplatin-based adjuvant chemotherapy while 11 did not. Most of these patients had T3 cancer. There was a significant difference in OS and DFS. Disease relapses occurred in 38% of those who had adjuvant chemotherapy, compared with 64% for those who did not. The 3-year OS was 81% versus 36%. In another small series of 46 patients, most of whom had pT3N0 disease, 24 were given adjuvant MVAC chemotherapy while 22 were observed.²⁰⁶ Owing to side effects, only 9 patients were able to complete the three planned cycles of treatment. There was a significant improvement in 5-year intravesical recurrence-free survival in the chemotherapy group compared with the nonchemotherapy group (84% vs. 39%). However, there was no difference in OS.

In an international collaborative database that included 542 high-risk patients,²⁰⁷ among which 121 received adjuvant chemotherapy, there was no difference in median OS and DSS between the chemotherapy and nonchemotherapy group. However, the percentage of patients with high-grade and high-stage disease in the chemotherapy group was significantly more than in the nonchemotherapy group. Large prospective studies are needed to define the role of adjuvant chemotherapy for UUT cancer.

Intraoperative Irradiation Containing Approaches for Locally Advanced Disease

In addition to neoadjuvant EBRT, an intraoperative irradiation supplement with electrons (IOERT) or high-dose-rate brachytherapy (HDR-IORT) may be considered as an additive component of treatment in patients with locally advanced primary or recurrent disease.

Solitary Metastases

Treatment of the solitary metastatic site has been evaluated in patients with both synchronous and metachronous solitary metastasis. Of 18 patients who presented with a solitary metastasis at diagnosis at the Mayo Clinic,²²³ 4 patients (22%) who underwent surgical treatment of the primary and metastatic sites survived longer than 2 years. Patients with metachronous solitary metastasis have a more favorable outlook. Of 26 patients treated with surgery or radiation therapy, 18 (69%) lived more than 2 years, and 5-year OS from the time of nephrectomy was 50%. Six of the 26 patients (23%) lived more than 5 years after removal of the solitary metastatic lesion.

Kjaer²²⁴ evaluated 25 patients with RCC and solitary metastasis. Twelve patients had a focus in the bone, and the remaining 13 patients had metastasis in soft tissue, including the lung, thyroid, flank, and epididymis. Radiation therapy was used primarily to treat the bone lesions. For the 13 patients with soft tissue metastasis, surgery was done in 7 patients and radiation therapy in 6. The median survival of this select group of patients was 4.3 years. The 5- and 10-year OS were 36% and 16%, respectively. Females had 5- and 10-year OS of 76% and 38%, respectively, versus 18% and 12% for males (p = .05).

Kavolius and coauthors²²⁵ retrospectively evaluated 278 patients initially treated with a curative nephrectomy who developed recurrent and/or metastatic RCC. The 5-year OS for 141 patients treated with a curative metastectomy for their first relapse was 44% as compared with 14% for patients who underwent noncurative surgery and 11% for patients treated nonsurgically. On multivariate analysis, factors associated with improved OS included a solitary site of first relapse, curative resection of first metastasis, and a long disease-free interval.

In summary, patients presenting with a metachronous solitary metastasis after a long disease-free interval may experience a prolonged survival compared with patients with more extensive metastases and may warrant more aggressive treatment to the initial metastatic site.

Bone and Soft Tissue Metastases

Metastases to the bone (solitary or multiple) may occur in 25% to 50% of all patients with RCC. Althausen and associates²²⁶ evaluated 38 patients with osseous metastases secondary to RCC. Patients were treated with resection with or without allograft implantation, radiation therapy alone, or a combination of treatments. The survival for the entire group was 90% at 6 months, 84% at 1 year, 55% at 5 years, and 39% at 10 years. Presentation without metastasis, long disease-free interval between nephrectomy and first metastases, appendicular skeletal location, and solitary metastasis were associated with longer survival. The authors concluded that patients who have an appendicular (preferably solitary) metastasis and who have a long duration from diagnosis of disease to metastasis should be treated with complex orthopedic surgical procedures if necessary. Postoperative irradiation should be considered in selected cases.

The results of irradiation alone in the treatment of bone metastases have been evaluated. Halperin and Harisiadis²²⁷ evaluated the results of 36 sites irradiated for bone pain. The pain responded in 77% of the treated sites, and the majority of the sites received a time-dose fractionation (TDF) equivalent dose ranging from 45 to 85. No relation was observed between the TDF equivalent dose and the probability of response. Twenty-four of the 28 sites (86%) were partially or completely free of pain for the remainder of the patients’ lives.

Wilson and colleagues²²⁸ reported partial and symptomatic response rates to palliative irradiation, with pain reduction of 67% and 73%, respectively. A dose response based on the biologic effective dose could not be established.

In a prospective phase II trial, Lee and associates²²⁹ evaluated the effect of radiation therapy on symptoms in 31 patients with symptomatic bone or soft tissue RCC metastases. With a median follow-up of 4.3 months, 83% of patients experienced site-specific pain relief with a median duration of site-specific pain response of 3 months. Global assessment of quality of life was limited owing to progression of disease outside the fields of treatment.

Stereotactic body radiotherapy (SBRT) or stereotactic radiosurgery (SRS) has been used for treatment of metastatic RCC to the spine. Gerszten and colleagues described 48 patients who underwent SRS to 60 spinal lesions.²³⁰ A single fraction with a median dose of 20 Gy was given. Eighty-nine percent of patients experienced improvement in pain. Six of 48 patients eventually required surgical intervention because of progression of disease. Nguyen and co-workers²³¹ reviewed 48 patients who underwent SBRT to 55 spinal lesions. Doses included 24 Gy in one fraction, 27 Gy in three fractions, and 30 Gy in five fractions. Forty-four percent and 52% of patients were pain free at 1 month and 12 months, respectively, compared with 23% before treatment. No grade 3 or grade 4 neurologic toxicity occurred.

Brain Metastases

Brain metastases are diagnosed in approximately 10% of patients with metastatic RCC.²³² Wronski and associates²³² evaluated the results of whole-brain radiation therapy in a retrospective study of 119 patients treated at M.D. Anderson Cancer Center. The median radiation dose was 30 Gy (range, 18 to 56 Gy). The median OS from the time of diagnosis of the brain metastases was 4.4 months. Patients with multiple brain tumors had a median survival of 3.0 months, compared with 4.4 months for those with a single brain metastasis (p = .043). The cause of death was neurologic in 90 of the 119 patients (76%). Patients with brain metastases diagnosed synchronously with the renal primary tumor (24 patients) had a median survival of 3.4 months compared with 3.2 months for the 95 patients diagnosed metachronously. Favorable prognostic factors included solitary brain metastasis, lack of other distant metastases at diagnosis, and tumor diameter less than 2 cm.

In view of the poor outcome of patients after whole-brain irradiation, more aggressive treatments have been evaluated in selected patients with favorable prognostic features. Patients who could undergo surgical resection of brain metastases from RCC had a 12-month median survival.²³³ Several retrospective studies^{234,235,236–239} have reported local tumor control rates of 88% to 96% in patients treated with radiosurgery alone, or in combination with whole-brain irradiation. Brown and colleagues²³⁵ reported that the addition of whole-brain irradiation to SRS improved local control and decreased distant brain failure. The Radiation Therapy Oncology Group recursive partitioning analysis (RPA) class strongly influenced the survival rate.^234,240

Chemotherapy, Immunotherapy, and Molecularly Targeted Treatment

Treatment options for patients with metastatic RCC include chemotherapy, immunotherapy, hormonal therapy, and, more recently, molecularly targeted treatment. Generally, RCC is not sensitive to chemotherapy, with a response rate of about 5.6%.²⁴¹ One regimen worth mentioning is the combination of gemcitabine and continuous 5-fluorouracil, which achieved a response rate of 17% as the second-line treatment in a phase II study.²⁴² In non–clear cell RCC, the response rates to chemotherapy appear to be better, and response rates of more than 10% to 15% have been reported using carboplatin/paclitaxel, cisplatin/gemcitabine, and doxorubicin/gemcitabine (for collecting duct or sarcomatoid histology).²⁴³ Current focus of systemic treatment is primarily on three different pathways: immunotherapy, vascular endothelial growth factor (VEGF) inhibition, and mammalian target of rapamycin (mTOR) inhibition. (A more extensive version of this section on immunotherapy of RCC can be found on the Expert Consult website.)

Treatment-Related Toxicity

CT-aided radiation therapy treatment planning (three-dimensional conformal or IMRT) is essential to reduce the risk of treatment-related toxicity. The Copenhagen Renal Study Group randomized 32 patients to receive postoperative irradiation after surgery.¹⁷⁶ Twenty-seven of the 32 patients randomized completed the course of postoperative radiation therapy to a dose of 50 Gy in 20 fractions. Significant complications to the stomach, duodenum, or liver occurred in 12 of the 27 patients and contributed to the death of 5 patients (19%). The increased fraction size of 2.5 Gy may have partially contributed to the high complication rate. In studies utilizing modern CT-based planning, the risk of late complications was less than 5%.^177,178

Potential Clinical Trials

Future clinical trials should evaluate both local and systemic treatment modalities. Patients who have positive surgical margins or positive lymph nodes are at increased risk for local failure (with or without abdominal seeding). The use of CT-based treatment planning (three-dimensional conformal and IMRT) and conventional fractionation of 1.8 to 2 Gy per day to a total of 45 to 50 Gy have a low risk of treatment-related complications and may potentially improve local control and survival. A prospective randomized trial comparing observation versus postoperative irradiation (with or without concomitant systemic therapy) would help to define the benefits of postoperative irradiation.

Patients who have positive lymph nodes and renal vein involvement are at increased risk for distant metastases. Future trials should evaluate the use of adjuvant systemic treatment including molecular targeted therapy for these patients.

Patients with locally advanced disease appear to benefit from aggressive treatment, including preoperative radiation therapy, surgical debulking, and IORT. Future trials can evaluate the use of radiation sensitizers and combining systemic agents with local treatment modalities to improve local control and reduce distant spread of disease. For patients with metastatic disease, future clinical investigation should continue to focus on improving survival and quality of life with molecular targeted therapy and other systemic agents while minimizing the toxicity of treatment.