Surveillance for metastatic disease after definitive treatment for renal cell carcinoma

INTRODUCTION — For patients with renal cell carcinoma (RCC), surgery may be curative when patients present with localized disease. However, many patients who are initially resected eventually relapse, and the prognosis in these cases is often poor. Thus, optimal management requires careful surveillance for recurrent disease in those who have undergone a potentially curative resection.

The timing and patterns of recurrent and/or metastatic disease and specific recommendations for surveillance following the initial treatment of localized RCC will be reviewed here. An overview of the treatment of RCC is presented separately. (See "Prognostic factors in patients with renal cell carcinoma".)

RATIONALE — The advantages of an aggressive surveillance program must be balanced against the financial costs, radiation exposure, and psychological distress that repeated testing can cause the patient. The rationale for surveillance is based upon several observations:

●Diagnosis of recurrence while the metastatic burden is limited may increase the likelihood of a favorable response to systemic treatment. (See "Overview of the treatment of renal cell carcinoma", section on 'Advanced renal cell carcinoma'.)

●Surgical resection of a single or limited number of metastases or an isolated locoregional recurrence can result in long-term survival. (See "Role of surgery in patients with metastatic renal cell carcinoma" and 'Locoregional recurrence' below.)

●Early diagnosis of metastasis may minimize complications even if it does not affect survival (eg, prevention of pathologic fractures due to bone metastases).

●Patients with asymptomatic recurrence detected by surveillance demonstrate improved survival compared with those with symptomatic recurrence [1].

●In addition to concerns about the development of metastatic disease, approximately 3 percent of patients with RCC will develop a metachronous de novo primary renal tumor [2].

The important factors in optimizing a surveillance protocol include the time after treatment when metastasis is most likely, the common sites of involvement, and the diagnostic approaches that are most effective in identifying recurrences.

TIME TO RECURRENCE

Duration of surveillance — The greatest risk of recurrence following resection of RCC is in the first two to three years. As a result, the primary surveillance efforts focus on this period. From years 3 to 5, close follow-up with imaging can become less frequent.

The optimal time to discontinue surveillance is unclear, as there are patients who have local or distant recurrence greater than ten years after treatment [3]. As an example, a series of 1088 patients who recurred after kidney cancer surgery were assessed by current guidelines. There were significant numbers of recurrences missed by stopping surveillance after five years. Extending surveillance to 15 years would capture 95 percent of recurrences; however, it would be significantly more costly, with unproven benefit [4].

Late recurrences — The features of patients with late recurrences were analyzed in a series of 1454 patients who were disease-free a minimum of five years after nephrectomy for localized RCC. With a median postoperative follow-up of 14 years, among those with recurrent disease, 63 patients (4 percent) had late relapse past five years. The estimated recurrence-free survival rates for a renal recurrence at 10 and 15 years after surgery were 97 and 95 percent, respectively. The distant metastasis-free survival rates at 10 and 15 years were 93 and 86 percent, respectively [5].

On multivariate analysis, risk factors associated with late distant metastasis included:

●Tumor size (1 centimeter increase; hazard ratio [HR] 1.07, 95% CI 1.01-1.14)

●Tumor stage (versus T1a)

•pT1b (HR 2.81, 95% CI 1.61-4.91)

•pT2a (HR 4.45, 95% CI 2.34-8.47)

•pT2b (HR 3.38, 95% CI 1.39-8.23)

•pT3a to pT3b, pT3c, pT4 (HR 5.09, 95% CI 2.68-9.67)

●Histology (versus chromophobe/papillary)

•Clear cell, collecting duct, or histology not otherwise specified (HR 3.76, 95% CI 2.12-6.66)

In one cohort from the International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) of 1210 patients with metastatic RCC, late recurrence (>5 years) occurred in 313 patients (26 percent). Predictors of late recurrence included younger age, absence of sarcomatoid features, clear cell histology, and lower Fuhrman grade [6].

Most surveillance protocols now recommend that a follow-up plan of duration greater than five years be individualized at the discretion of the treating clinician.

SITES OF METASTASIS — The most common sites of metastatic disease from RCC are the lungs, bones, liver, renal fossa, and brain, although recurrences may occur elsewhere. Specific imaging studies of each region are necessary to detect recurrence.

While other malignancies utilize whole body imaging with fludeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT), the utility of this modality is limited for RCC, due in part to limited FDG activity in many renal tumors as well as the physiological excretion of FDG into the urine. While the latter in particular limits the role of FDG PET/CT in the initial diagnosis and staging of RCC, the role of FDG PET/CT has also been met with mixed results in the restaging of RCC after nephrectomy [7], and FDG PET/CT is currently not recommended for this purpose based upon current guidelines.

Lung metastases — Lung metastases have been reported in 29 to 54 percent of patients with recurrent RCC and are the most common site of metastatic disease [8-10]. Office assessment of any pulmonary complaints, such as cough or shortness of breath, should occur with each office visit.

All surveillance protocols incorporate periodic imaging of the lungs. Whether CT or plain chest radiograph is optimal remains uncertain, although CT is now being used more routinely. It is not clear that the increased sensitivity of chest CT offers additional benefit compared with plain chest radiograph. Many risk-stratified protocols utilize chest CT for high-risk populations but do mention alternating usage with chest radiography [11,12]. (See 'Surveillance protocols' below.)

Bone metastases — Office visits should assess if there is any bone or joint pain or unexplained fractures. Bone metastases have been observed in 16 to 31 percent of patients with recurrent RCC, and 67 to 91 percent of those with bone metastasis present with localized pain due to their lytic nature [8-10,13-15]. Early identification and treatment of bone metastases can prevent pathologic fractures, minimize pain, and maintain function [16,17]. (See "Radiation therapy for the management of painful bone metastases" and "Role of surgery in patients with metastatic renal cell carcinoma" and "Osteoclast inhibitors for patients with bone metastases from breast, prostate, and other solid tumors" and "Rehabilitative and integrative therapies for pain in patients with cancer".)

Most surveillance strategies to detect bone metastases rely upon the history, physical examination, and measurement of serum alkaline phosphatase. Routine imaging with plain films or bone scan is not recommended, and these tests are reserved to confirm a suspicion of metastasis and to evaluate suspicious findings observed on chest and abdominal scans. (See 'Surveillance protocols' below.)

●Bone scan – The use of a radionuclide bone scan beyond that of history and physical examination is generally restricted to symptomatic patients. This was illustrated in a series of 1357 patients with RCC [18]. Among the 167 cases with bone metastases, 96 percent were symptomatic, with an Eastern Cooperative Oncology Group (ECOG) performance score of 1 or greater (table 1). Among all patients who were asymptomatic (ie, a performance score of 0), only 1.4 percent had bone metastases. In addition, bone scans may be relatively insensitive at detecting the largely lytic metastatic lesions occurring in RCC.

●Alkaline phosphatase – Monitoring the serum alkaline phosphatase is simple and inexpensive, but elevated values are seen in only 40 to 55 percent of patients with documented bone metastasis; in addition, its specificity is poor [14,18]. Elevations in serum alkaline phosphatase may also be due to liver metastases or a paraneoplastic syndrome. (See "Enzymatic measures of hepatic cholestasis (alkaline phosphatase, 5'-nucleotidase, gamma-glutamyl transpeptidase)" and "Clinical manifestations, evaluation, and staging of renal cell carcinoma", section on 'Hepatic dysfunction'.)

●PET/CT – Detection of bone lesions using cross-sectional imaging and bone scans can be challenging. Cross-sectional imaging modalities do not image the axial skeleton, and there are inherent limitations of bone scans due to the osteolytic nature of bone metastases. FDG PET/CT imaging may have a role in detecting bone metastases, with some observational series demonstrating improved sensitivity/accuracy at detecting bone metastases compared with conventional bone scan [19].

Liver metastases — Metastases to the liver have been reported in between 8 and 30 percent of patients with recurrent RCC [8-10,13,18]. In approximately 90 percent of cases, such metastases are detected because of symptoms or abnormal laboratory values [9,13]. Consequently, all surveillance regimens utilize periodic history, physical examination, and liver function tests to assess for symptoms, hepatomegaly, and laboratory abnormalities. Abdominal CT or magnetic resonance imaging (MRI) is the best tool to detect liver metastases, which may be present even in the absence of symptoms or abnormal liver function tests [9,13].

Early detection of liver metastases allows carefully selected patients to undergo local therapies such as microwave/radiofrequency ablation or metastasectomy, although the evidence supporting the clinical benefit of such an approach is anecdotal [20,21]. (See 'Surveillance protocols' below and "Role of surgery in patients with metastatic renal cell carcinoma", section on 'Liver metastases'.)

Locoregional recurrence — Locoregional recurrence can result from metastatic disease to the ipsilateral adrenal gland, regional lymph nodes, perirenal fat or Gerota's fascia, invasion into the psoas muscle, or due to tumor spillage. Local recurrence is more common in patients with locally advanced tumors and those with sarcomatoid histology [9].

Although the rate of isolated local recurrence following radical nephrectomy varies with patient selection, such relapses are uncommon. As examples:

●In a series of over 1700 patients from the Mayo Clinic who underwent radical nephrectomy for localized (T1-T3N0M0) RCC (table 2), isolated recurrence in the renal fossa was identified in less than 2 percent [22]. Of these 30 patients with isolated local recurrence, 60 percent were symptomatic and 40 percent were found on routine follow-up evaluation.

●In a subgroup analysis of a randomized trial (ASSURE), among 300 patients with intermediate-/high-risk nonmetastatic disease with local recurrences, 22 percent (66 patients) experienced recurrences in the kidney/renal fossa. However, in the overall study population of 1943 patients, this represented an isolated local recurrence rate of only 3 percent [23]. Further details on the ASSURE trial are discussed separately. (See "Overview of the treatment of renal cell carcinoma", section on 'Adjuvant therapy for locoregional disease'.)

Early diagnosis of patients with isolated local recurrence is important because surgical resection of such relapses may improve outcome by preventing progression to disseminated disease [22,24,25]. Retrospective series have evaluated the outcome following such resection, although this presumably represents a highly selected population. As an example, in 102 patients who underwent resection of an isolated retroperitoneal recurrence at MD Anderson Cancer Center, metastatic progression was observed in 59 percent with a median follow-up of 32 months [25]. Approximately 30 percent of patients remained recurrence-free at 10 years of follow-up. The median cancer-specific survival with resection of a local recurrence was more than five years.

For soft tissue recurrences away from the great vessels, percutaneous ablation is possible. However, as the bowel often migrates into the renal fossa, hydrodissection of the bowel is necessary to avoid inadvertent damage to adjacent organs.

Limited data are available for resection of local recurrence after partial nephrectomy; however, these tumors are often smaller and lower stage, and thus are likely to have similar or improved outcomes with surgery.

Although the incidence of isolated local recurrence is low, abdominal CT is recommended to simultaneously evaluate for liver metastasis and local recurrence. While abdominal ultrasound is included in some surveillance algorithms, it may be insufficient for identification of locoregional recurrences. (See 'Surveillance protocols' below.)

Renal recurrences — In the presence of advanced disease, a new renal lesion may represent either an independent lesion or a de novo recurrence. Without distant disease, the majority of lesions are independent tumors with the exception of tumors after partial nephrectomy, which recur in the surgical bed. Metachronous renal tumors may be more common than previously thought in some populations. While the 10- and 20-year cumulative incidence is relatively low (1.5 and 3.1 percent, respectively), specific factors such as age <40, male gender, being from a Black population, and papillary histology are significant risk factors [26].

Brain metastases — Brain metastases occur in 2 to 10 percent of patients with recurrent RCC [8-10,13] and are symptomatic in ≥80 percent of cases [15,22,27,28]. Affected patients can present with headaches, altered behavior, seizures, or focal neurologic symptoms. (See "Epidemiology, clinical manifestations, and diagnosis of brain metastases", section on 'Clinical manifestations'.)

Brain metastases are accompanied by other sites of metastatic disease in most cases [27-29]. As an example, in a single-institution series of 138 patients with brain metastases, 95 percent had extracranial sites of metastasis when brain metastases were diagnosed [29]. In a separate series of 158 patients, 91 percent had additional sites of disease [30]. Despite historically poor outcome, carefully selected patients with good performance statuses and brain metastases can demonstrate long-term survival with aggressive therapy [29]. With improvements in local and systemic therapy, these patients have experienced significantly improved outcomes [30].

Screening asymptomatic patients in the absence of other documented distant disease is not recommended given the low incidence of brain metastasis, high incidence of symptomatic presentation, and frequent association with extracranial sites of metastasis [8-10,13]. However, if recurrence at other sites is detected, brain imaging may be considered as it could identify small asymptomatic lesions that could be amenable to radiation. MRI and CT are both useful for evaluation of symptomatic patients or at the time of extracranial recurrence. Currently, there is no specific preference per major guidelines for MRI versus CT. Although MRI has greater sensitivity for small asymptomatic brain lesions, in the setting of a symptomatic patient that may require an emergent evaluation and treatment, CT can be obtained more expeditiously in most centers.

For individuals being considered for immunotherapy or antiangiogenic therapy, brain imaging is useful to document the absence of disease because of the risk of treatment-associated toxicity or hemorrhage.

The approach to treating brain metastases from RCC is discussed separately. (See "Overview of the treatment of brain metastases".)

Impact of histology — The patterns of metastatic spread have been analyzed based upon the histologic subtype of RCC [31]. In this analysis, clear cell tumors preferentially spread to the lungs, while chromophobe tumors more frequently metastasized to the liver. Based on differential sites of metastasis and time to metastasis, histology-specific protocols have been recommended with differential chest and abdominal imaging surveillance intervals [32]. While further refining risk stratification, these protocols have not been widely adopted. (See 'Surveillance protocols' below.)

SURVEILLANCE PROTOCOLS — The two most widely used surveillance protocols following definitive treatment for RCC are those from the American Urologic Association (AUA) and the National Comprehensive Cancer Network (NCCN). Both of these surveillance protocols utilize a risk-stratified approach, whereas first-generation models did not stratify by stage or tumor grade [33].

AUA guidelines — The AUA recommends specific guidelines for surveillance after surgical management of RCC [34]. These guidelines were developed based on a systematic review of the literature:

Stage I or T1N0/X (partial or radical nephrectomy)

●History and physical examination at months 6, 12, 24, and 36 and further follow-up at the discretion of the treating clinician.

●Serum blood urea nitrogen (BUN) or creatinine and urine analysis are recommended and other tests as clinically indicated.

●Abdominal imaging:

•After partial nephrectomy – Baseline abdominal computed tomography (CT)/magnetic resonance imaging (MRI) at month 6 and then abdominal CT/MRI/or ultrasound (US) at 12, 24, and 36 months.

•After radical nephrectomy – Baseline abdominal CT, MRI at month 6 then imaging as clinically indicated.

●Chest imaging – Chest radiograph or CT annually for three years, then as clinically indicated.

●Central nervous system (CNS) imaging, pelvic imaging, and bone imaging as clinically indicated.

PT2-4N0/X or pTanyN1

●History and physical examination every six months until five years and further follow-up at the discretion of the treating clinician.

●BUN or creatinine and urinalysis; other tests as clinically indicated.

●Abdominal imaging – CT/MRI recommended at month 6, after which CT/MRI/US use acceptable for abdominal imaging every six months until five years, with further follow-up at the discretion of the treating clinician.

●Chest imaging – Chest CT initially at month 6 followed by chest radiograph or CT every six months until five years, and further follow-up at the discretion of the treating clinician.

●CNS imaging, pelvic imaging, and bone imaging as clinically indicated.

NCCN guidelines — The National Comprehensive Cancer Network (NCCN) guidelines include the following risk-stratified recommendations [11]:

Stage I

●History and physical examination annually and basic laboratory studies annually as indicated. Additional imaging of brain, bone, or pelvis based upon clinical indications.

●Chest imaging (either CT or plain radiograph) annually for five years or longer based on clinical judgment.

●Abdominal imaging with CT, MRI (preferred), or US within the first year after surgery. This should be repeated at years 2 and 3 or longer based on clinical judgment.

●More intensive follow-up schedule can be offered to patients with high-risk features (eg, sarcomatoid, high grade, or positive margins).

Stage II/III

●History and physical examination every three to six months for the first three years and then annually through year 5. Additional imaging of brain, bone, or pelvis based upon symptoms and clinical indications.

●Comprehensive laboratory studies every six months for the first two years and then annually through year 5. Longer evaluation may be offered based on clinical judgment.

●Chest CT every three to six months through year 3 and then annually through year 5. Longer evaluation may be offered based on clinical judgment.

●Abdominal imaging with CT, MRI (preferred), or US at three- to six-month intervals for first three years and then annually through year 5. Longer evaluation may be offered based on clinical judgment.

Nomograms — A wide variety of factors are known to be independent prognostic indicators of tumor progression and influence cancer-specific survival. Prognostic models incorporate factors related to tumor anatomy, histology, and clinical presentation. Many prognostic models are available that can fairly accurately stratify patients more likely to recur [34-39]. Some of these, such as the University of California, Los Angeles (UCLA) Integrated Staging System [35,40] or the Mayo Clinic Leibovich prognostic model [36,41], have been used for clinical trial selection. However, many of these predictive models perform poorly in high-risk cohorts, as illustrated from one study using data from the ASSURE trial [42]. (See "Prognostic factors in patients with renal cell carcinoma", section on 'Clinical factors'.)

Biomarkers and the future of surveillance — Over the past 10 years, multiple biomarkers have been linked to increased risk of recurrence after nephrectomy. As these are validated, such biomarkers may enable refinements in surveillance protocols following definitive treatment of RCC.

Both UCLA and Mayo groups have developed predictive models that incorporate panels of biomarkers with clinicopathologic information [35,36]. The addition of these biomarkers appears to improve the predictive capability of these models. Neither model has been externally or prospectively validated, but this concept probably represents the future of prognostication.

Challenges to the adoption of these biomarker-based systems include the variability of tissue preparation and storage, and in interpretation of immunohistochemical scoring [37]. For these reasons, a cytogenetic approach may be more easily adopted as many centers have facilities capable of performing this technique. Cytogenetic evaluation of renal tumors has been routinely performed at UCLA, and assessment of chromosomal alterations can aid in prognostication [38]. (See "Prognostic factors in patients with renal cell carcinoma" and "Prognostic factors in patients with renal cell carcinoma", section on 'Molecular markers'.)

With the expansion of genomic characterization, there is some hope that next-generation sequencing could assist with prognostication. Specific mutations in chromatin remodeling genes such as BAP1 may lead to a worse prognosis [39,43]. Challenges exist in using genomic sequencing to determine mutational status due to significant intra-tumor heterogeneity. A biopsy of a single site may not adequately assess a specific gene alteration, as over half of mutations in driver alterations are often not shared throughout the tumor.

SURVEILLANCE AFTER THERMOABLATION — Thermoablation is an alternative to partial nephrectomy in selected patients with clinical T1 RCC [12]. (See "Diagnostic approach, differential diagnosis, and management of a small renal mass" and "Radiofrequency ablation, cryoablation, and other ablative techniques for renal cell carcinoma".)

Both radiofrequency ablation and cryoablation can be performed through an open, laparoscopic, or percutaneous approach, and it appears that both procedures have similar oncologic outcomes. The American Urological Association (AUA) guidelines suggest that for those patients who are candidates for ablation, a percutaneous approach is preferred, when feasible, to minimize the morbidity of other approaches [12]. Compared with partial nephrectomy, there is a 5- to 20-fold increased risk for local recurrence, while the risk for developing distant metastases appears to be similar [44,45]. There may be a role for risk stratification, as increased surveillance may be required for larger lesions and less surveillance may be required for lesions of indolent potential, such as a fat-poor angiomyolipoma.

With the higher rate of local recurrence, it is crucial to monitor the ablated kidney with either contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI). National Comprehensive Cancer Network (NCCN) guidelines recommend such imaging within three to six months following ablation, followed by annual imaging for up to five years or longer as clinically indicated. However, given the high risk of local recurrence, it is prudent to offer postablation imaging on a more frequent schedule. Our approach is to perform CT or MRI imaging at three to six, 12, 18, and 24 months after ablation, and yearly thereafter. The proposed abdominal surveillance scheme is similar to the one for the intermediate- and high-risk groups of patients treated with nephrectomy. Chest imaging can be performed similar to that for other patients with similar disease characteristics managed with surgical resection. Recurrent tumors can often be successfully retreated with ablation. (See "Radiofrequency ablation, cryoablation, and other ablative techniques for renal cell carcinoma", section on 'Surveillance after ablation'.)

A repeat renal mass biopsy should be considered with new renal enhancement after ablative therapy but may be performed at the time of repeat ablation [46].

SURVEILLANCE AFTER ADJUVANT THERAPY — Some patients with renal cell carcinoma (RCC) who are treated with surgical resection and are at increased risk for recurrence may receive adjuvant systemic therapy (such as pembrolizumab). Surveillance performed during and following adjuvant therapy should follow the stage and risk categorization of the initial surgical tumor staging.

The approach to adjuvant therapy in resected RCC (including patient selection and available agents) is discussed separately. (See "Overview of the treatment of renal cell carcinoma", section on 'Adjuvant therapy for locoregional disease' and "Overview of the treatment of renal cell carcinoma", section on 'Adjuvant therapy after metastasectomy'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cancer of the kidney and ureters".)

SUMMARY AND RECOMMENDATIONS

●Rationale – The rationale for surveillance following resection is based upon indirect evidence from patients who are treated for metastatic disease. There are no randomized trials demonstrating increased survival from specific surveillance practices following definitive therapy. (See 'Rationale' above.)

●Surveillance protocols – Following complete resection of renal cell carcinoma (RCC), patients should be monitored for evidence of local recurrence, distant metastasis, or a second primary RCC. Contemporary recommendations for surveillance incorporate information from the tumor, node, metastasis (TNM) staging system along with other key prognostic factors. (See 'Surveillance protocols' above.)

•Surveillance protocols focus on the sites most frequently involved with metastases, including the lungs, bones, liver, and renal fossa, and match testing with the period of greatest risk of recurrence. These surveillance protocols must be adjusted for the needs of the individual patient based upon clinical judgment and experience. (See 'Sites of metastasis' above and 'Time to recurrence' above and 'Surveillance protocols' above.)

•Guidelines from the American Urological Association (AUA) and the National Comprehensive Cancer Network (NCCN) are similarly efficient in detecting both locoregional and distant metastases and provide a useful tool for patient management. Both of these guidelines incorporate an estimate of the risk of recurrence based upon extent of disease. (See 'AUA guidelines' above and 'NCCN guidelines' above.)

●Surveillance after thermoablation – Patients whose tumor has been managed with thermoablation (eg, cryotherapy, radiofrequency ablation) should have surveillance abdominal imaging performed on a schedule similar to that for patients with intermediate- or high-risk disease treated with nephrectomy, because of the increased risk of local recurrence. (See 'Surveillance after thermoablation' above.)

●Surveillance after adjuvant therapy – For patients treated with resected RCC followed by adjuvant systemic therapy, surveillance performed during and following adjuvant therapy should follow the stage and risk categorization of the initial surgical tumor staging. (See 'Surveillance after adjuvant therapy' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Tobias Klatte, MD, who contributed to earlier versions of this topic review.