Robert G. Uzzo, MD
Chairman, Department of Surgery
Professor of Urologic Oncology
G. Willing “Wing” Pepper Chair in Cancer Research
Division of Urologic Oncology
Department of Surgery
Fox Chase Cancer Center-Temple University Health System
Guidelines for the management of the stage I renal mass are continuously evolving and active surveillance has increasingly been the focus of new reports evaluating alternative management strategies in select patients with small tumors. Although excision, via radical or partial nephrectomy, has generally been the rule for localized renal tumors, emerging data suggest that an initial period of observation may be safe for patients with extensive competing risk and small enhancing lesions. This review summarizes the rationale for a period of initial observation in select patients with small renal masses, reviews the cumulative experience describing the natural history of untreated tumors, and emphasizes the relevance of competing risks assessment for clinical decision making and patient counseling at the time of initial diagnosis.
It would seem that the well documented rise in detection of small renal masses (SRMs) due to increased use of cross sectional abdominal imaging, and resulting increase in the number of surgical procedures performed, should bend the mortality curve for patients diagnosed with Renal Cell Carcinoma (RCC). Paradoxically, this may not be true. While the rates of renal surgery have risen simultaneously with increased tumor detection, mortality rates remain largely unchanged; suggesting that the absolute number of detectable lethal lesions has not diminished.1 The implications from these emerging data suggest that a significant proportion of small localized renal masses may be indolent tumors with little chance of tumor progression or cancer related death. As a result, clinicians have begun to reassess contemporary management algorithms for incidentally diagnosed renal tumors in an effort to reduce over treatment of patients in which the risks of surgical resection may outweigh any marginal survival benefit.
Quantitating Competing Risks
When counseling a patient regarding the need for definitive intervention, a number of factors must be considered:
- What is the natural history of the lesion if not treated?
- What are the patient’s competing risks to mortality?
- Do the procedural risks outweigh the potential survival benefit?
Conventional wisdom dictates in most cases that immediate intervention is appropriate for enhancing, localized tumors. While surgical resection of small renal masses in young, healthy patients is currently and will surely remain the accepted standard of care,2 management of localized SRMs in elderly or infirm patients in whom co-morbid medical conditions compete with RCC malignant potential as primary causes of death represents a unique set of challenges. Until recently, it has been difficult to dichotomize a patient’s risk of cancer specific death, which for localized low stage kidney cancer is quite low, versus their competing risks of death from co-morbidities such as cardiovascular or pulmonary disease. There are few randomized controlled trials for localized urologic malignancies to give long term perspective regarding the effectiveness of surgical interventions3but results from the well-publicized EORTC trial 30904 shed important insight on this phenomenon. While designed to compare the efficacy of radical and partial nephrectomy for localized tumors ≤5cm, perhaps the most interesting finding of this non-inferiority trial were the long term mortality results. 117 deaths were documented over a median follow-up of 9.3 years. 10.3% were attributable to RCC versus 89.7% from comorbid conditions.4
A number of other reports quantitating the competing risks of death in patients with RCC demonstrate similar results:
- In a series of 537 patients >75 years of age with renal masses <7 cm and a median follow up of 3.9 years, Lane et al reported the most common cause of death was cardiovascular (29%) compared to cancer progression (4%).5
- Using the Charlson Co-morbidity Index (CCI) to stratify 192 patients with clinically localized RCC, Arrontes et al6 reported that patients with clinically localized RCC and a CCI >2 had significantly reduced overall survival vs patients with a CCI ≤2. No survival differences were observed between CCI stratified groups with locally advanced or metastatic RCC. The authors concluded that patients with CCI scores >2 do not gain a survival advantage from the surgical treatment of localized RCC.
- From Surveillance, Epidemiology, and End Results (SEER) data involving more than 26,000 patients surgically treated for RCC, Hollingsworth et al7 presented 5-year mortality rates for cancer-specific and competing cause mortality (stratified by tumor size and age at presentation). The smallest tumors were associated with the lowest cancer-specific mortality. Competing cause mortality rose with age (28% 5-year competing cause mortality rate in patients ≥70 years).
With dissemination of these results, there has been increased interest as to whether such data can be used to objectively and quantitatively estimate competing risks for patient counseling purposes in the pre-operative setting. An ensuing report proposed a nomogram estimating the risk of kidney cancer death, death from other malignancy, and non-cancer death utilizing select preoperative clinical and demographic variables including age, race, gender, and tumor size in a large cohort of patients with RCC identified from the SEER database.8 However, specific co-morbidity information was not available for inclusion in the analysis, thereby limiting the clinical utility.
To improve on their initial effort, Kutikov et al9 used linked SEER-Medicare data, which affords the ability to derive co-morbidity from patient claims, to identify 6,655 individuals aged 66 years or older with localized RCC. Patients with localized, node negative kidney cancer had a low 3-year (4.7%), 5-year (7.5%) and 10-year (11.9%) probability of cancer specific death but a significantly higher overall risk of death from competing causes with 3-year (10.9%), 5-year (20.1%) and 10-year mortality (44.4%). Although caution must be used when extrapolating outcomes of patients treated surgically to counsel patients in the pre-operative setting, these nomograms provide tools useful in the clinical setting. For example, an 80-year-old African American male with a history of myocardial infarction, moderate renal insufficiency (Charlson score 3) and a 4 cm renal mass is expected to have 5% 5-year mortality from RCC versus 48% from non-kidney cancer causes. However, a 75-year -old white female with no significant comorbidities (CCI 0) and a 7 cm renal mass is predicted to have 13% 5-year mortality and 7.5% mortality from other causes.
A lack of objective trade-offs and guidance information often contributes to physician reluctance to estimate competing risks,10 and to facilitate clinic use in real-time we have recently made a number of these nomograms with a predictive accuracy >70% available on a web based portal (www.cancernomograms.com). With such prognostic information, individualized treatment decisions become more objective and ultimately more informed patient-provider discussions geared towards contextualizing competing risks may help to reduce overtreatment in elderly and/or infirm patients.10
Malignant and Metastatic Potential of the SRM
With growing concern that increasing incidental detection may result in the over diagnosis (and over treatment) of tumors never destined to progress or metastasize, there is considerable interest in quantifying the malignant and metastatic potential of SRMs, of which approximately 15-20% of tumors <4cm are benign, based on pre-treatment radiographic characteristics.11 The most well defined prognostic indicator is tumor size. A number of institutional reports demonstrate that as maximal linear tumor diameter increases, there is increased risk of malignancy,12,13 high-grade disease, clear cell histology, and presence of synchronous metastases.14,15 In the largest study to date, Rothman et al. used SEER data to explore the relationship between primary tumor size at presentation and histopathological features in 19,932 patients with localized renal tumors. They found that for each 1 cm increase in size the probability of a high-grade tumor increased by 13%.16
However, the same authors found that while almost 85% of RCCs <4 cm were low grade, a substantial proportion of lesions >7 cm (70%) were also low-grade lesions. Thus, it appears that localized renal tumors can grow quite large without acquiring a biologically aggressive phenotype. Although available data clearly suggest that a significant proportion of small renal masses tend to be indolent and only small minority present with an aggressive phenotype, size at presentation is also associated with synchronous metastatic disease at presentation. Kunkle et al14 compared 110 patients with biopsy proven synchronous metastatic disease at presentation to 250 controls with classically localized RCC. As expected, tumors associated with synchronous metastases were significantly larger than localized lesions without evidence of distant or regional disease (8.0 versus 4.5 cm), and with each 1 cm increase in tumor size, the odds of finding a synchronous metastases rose by 22%. While this does provide evidence that size does matter, it is important to consider that all aggressive, as well as benign or indolent lesions, start off as small tumors. In the above study, while no patients with tumors 2 cm or smaller presented with biopsy proven metastatic disease, close to 5% of patients presenting with distant metastases occurred in patients with tumors less than 3 cm. A similar study using SEER data confirmed that the risks of synchronous metastasis for lesions <3cm appears to be approximately 5%.17 These findings highlight that the malignant potential for most small lesions is likely low, but illustrate that improved methods of identifying those with increased oncologic risk is essential to best counsel patients presenting with a newly diagnosed SRM considering active surveillance versus definitive surgery or ablation.
Pre Treatment Prediction of Malignant Potential
One of the great challenges in managing the small renal mass remains achieving the ability to match specific tumor biology with an individualized treatment strategy. Currently, use of radiographic characteristics to determine malignant potential is limited, and alternative non-extirpative diagnostic strategies including predictive statistical models and percutaneous biopsy have short-comings as well.
Percutaneous biopsy. Procedural risks, sampling error, and skepticism regarding the clinical relevance of the data gained for clinical decision making until recently have largely obviated the utility of percutaneous biopsy except in specific circumstances, such as suspicion for lymphoma, renal abscess, or metastatic disease from another primary malignancy. However, there have been advances in technique, most notably the use of 18 gauge core needles and improved immunohistochemical characterization, and contemporary studies have reported more than 90% accuracy in differentiating malignant from benign histologies as well as low risks of minor complications (5%). Despite these improvements, percutaneous biopsy is often insufficient to determine or underestimates tumor grade which is arguably the most important prognostic indicator for clinical decision making.18
Although the role of biopsy in the treatment algorithm for the SRM is evolving and has been incorporated into a number of active surveillance protocols, 19, 20it has not yet been accepted as standard practice in the treatment planning for young or otherwise healthy candidates for excision. In our practice, renal mass biopsy is preferred in patients with absolute or relative indications for active surveillance (such as significant co-morbidity, poor underlying renal function, secondary malignancy) or specific circumstances, such as synchronous bilateral lesions or a solitary kidney.21 The elusive goal remains identification of molecular biomarkers obtained from percutaneous biopsy specimens predictive of an aggressive RCC phenotype. While biomarker discovery remains an area of intense interest which is beyond the scope of this review, until these markers are available alternative prognostic tools are required for risk stratification in patients with small renal masses.
Imaging techniques. The gold standard for evaluation of a renal mass remains CT or MRI—to assess characteristics, bilateral renal flow and function, and radiographic staging data. These imaging techniques can differentiate most renal cystic lesions from solid masses but cannot reliably distinguish between benign and malignant solid tumors, histologic subtypes, or indolent versus aggressive disease.11 Although also beyond the scope of this review, there are emerging histology specific imaging modalities currently under evaluation that may play a more definitive role in the future.22
Clinical Nomograms. Models to predict disease recurrence following definitive therapy are useful for patient reassurance and to help guide surveillance imaging intensity but they have limited application for counseling in the preoperative setting. Unfortunately, early algorithms designed to predict malignant or biologic potential of small renal masses using pre-treatment characteristics also fall short due to poor predictive accuracy. For example, a nomogram published by Lane et al23 incorporating gender, tumor size, and smoking history to predict malignant versus benign disease had only modest accuracy (Concordance index [CI] of 0.64). Similarly, Jeldres et al developed a nomogram to estimate risk of high grade disease based on age at diagnosis, gender, tumor size, and symptom classification, but their predictive model was only accurate in 58.3% of cases.24 Other nomograms have been proposed to assess the probability of death from co-morbid conditions compared to death from kidney cancer. (Figure 1)
However, encouraging data have emerged from more recent reports suggesting a relationship may exist between renal mass anatomic characteristics and pathologic features. The Nephrometry Score was recently introduced as a means of quantifying salient tumor anatomy to provide a reliably reproducible renal mass nomenclature and facilitate meaningful comparison of case mix 25 In a 2011 report Kutikov et al. evaluated the relationship between nephrometric characteristics and pathologic characteristics at the time of surgical excision, finding that total nephrectomy score and all individual anatomic descriptors significantly differed between tumor histology groups except for anterior/posterior(A) designation.26 Using institutional data, the authors developed a clinical nomogram integrating nephrometric variables with patient age and gender with the highest predictive accuracy for malignant histology (AUC 0.76) and high-grade features (AUC 0.73) to date. Further, these accuracy rates, particularly for tumor grade, compare favorably with results of percutaneous core biopsy, adding a valuable tool for counseling in patients that are poor surgical candidates.
The Natural History of SRMs Under a Period of Observation
Other than reluctance to recommend against surgical resection when technically possible, one of the biggest hurdles in the acceptance of active surveillance as a viable management strategy is the poor understanding of the natural history of untreated renal tumors. The current cumulative experience consists of small retrospective institutional accounts of tumors under a period of observation for unspecified intervals.
In 2006, Chawla et al. performed a meta-analysis of 9 single institutional series (286 lesions), which reported that the majority of small enhancing renal masses grow at a slow rate with short term follow up.27 We recently updated this report with a pooled analysis of 259 patients (284 SRMs), with a mean maximum tumor diameter of 2.4cm and a mean linear growth rate (LGR) of 0.3cm/year with intermediate (median 3 year) follow up.28 These reports confirm that the majority of small renal tumors grow at a slow rate, but offer little in terms of prognostic characteristics that may indicate the need for early versus delayed intervention.
While only a small proportion (Figures 2 and 3) of these lesions grow rapidly, to date no conclusive associations between tumor characteristics such as size at presentation, solid/cystic appearance, or tumor multifocality and linear growth rate have been demonstrated.29 Further, positive growth rate is not indicative of malignant phenotype, as studies have revealed that benign lesions such as oncocytomas demonstrate positive growth rates over time as well.30As a result, once the decision has been made to implement an active surveillance protocol frequent re-imaging at defined short-term time intervals is mandatory to accurately characterize individual tumor growth kinetics.
In reviewing the available literature, it has become apparent that approximately 20-25% of tumors under observation exhibit no net growth over time.31 While no difference has been exhibited in terms of malignant pathology in lesions that grow and do not grow at the time of surgical resection, to date no lesion exhibiting zero growth over time has progressed to metastatic disease with intermediate duration follow up.28 As a result, tumors exhibiting slow or zero growth kinetics may represent an ideal population for prolonged active surveillance.
Perhaps the most critical piece of information for informed patient counseling is the risk of disease progression or development of metastatic disease during a period of observation. While the available literature suggests that this risk is very low, the cumulative active surveillance experience has not matured, and it is clear that there are some lesions currently being observed under active surveillance protocols that may represent high risk malignancies with aggressive malignant potential. In a recent review, Smaldone et al. identified 880 patients (936 SRMs) from 18 studies, of these lesions only 18 patients (2%) progressed to metastases with a mean follow up of 40.2 months. Tumors that progressed were more likely to be larger (4.1 versus 2.3cm in maximum diameter) with a faster linear growth rate (0.8 versus 0.3cm/year).28
From reviewing these data, a number of observations were made that are useful in contextualizing the risks of active surveillance for patient counseling.
- The vast majority of localized renal tumors exhibit slow radiographic growth while under an initial period of observation.
- A positive growth rate is not always indicative of malignant histology
- Progression to metastatic disease in patients with small renal masses under surveillance appears to be uncommon (2%) and poorly documented in the literature.
- Tumors that progressed were more likely to be large, with positive rapid growth kinetics, high grade pathology, or clear cell histology
- No lesion <3cm or demonstrating zero order kinetics over time progressed to metastatic disease
The impact of a delay in treatment on definitive management strategy
Another commonly asked question is…”if the tumor grows will it impact how I am treated?”. Close to 50% of patients who are initially managed with active surveillance go on to definitive intervention, due to rapid growth kinetics, change in performance status, or patient preferences).28 Recent reports suggest that these patients that proceed to treatment have similar size tumors at baseline compared to patients managed only expectantly, but demonstrate more rapid growth kinetics.32 However, a study describing a large tertiary active surveillance experience reported that 76% of the patients progressing to definitive treatment underwent nephron-sparing approaches and 60% were treated in a minimally invasive approach.33These findings indicate that in the vast majority of patients, an initial period of observation should not impact their future management.
Criteria for surveillance
In our practice, each patient that presents with a newly diagnosed small mass is categorized by absolute, relative and elective indication for active surveillance.
Absolute: when surgical intervention poses an immediate and unacceptable risk due to comorbidity or performance status
Relative: significant but not over-riding medical co-morbidity or potential need for renal replacement therapy
Elective: patients who are low risk surgical candidates but decide against definitive intervention
Stratification of patients according to risk status is crucial when evaluating a patient with a SRM to determine if active surveillance may be the appropriate treatment strategy. For each individual case, efforts are made to con- textualize patient competing risks of mortality, life expectancy, performance status, perioperative adverse events following surgical resection, and risks of disease progression prior to determining a final treatment plan.
Active Surveillance SRM Imaging Protocols
To date, there is no consensus or defined active surveillance imaging protocol that is universally adhered to across institutions. It is our preference to monitor SRM growth at short 3-4 month time intervals during the first year following diagnosis, and then increase the duration between imaging studies as the growth kinetics become better characterized. We also to prefer to change the imaging modality to magnetic resonance imaging, or in select cases renal ultrasound, as early as possible. In this manner it is possible to still vigilantly monitor for disease progression but over time minimize the risks of secondary malignancy inherent to increased radiation exposure.
Features that are fairly consistent among protocols include:
- Initial review of all previous imaging studies to document size at presentation and establish a baseline measurement.
- Imaging should be performed at 3-6 month intervals following diagnosis, increasing the interval to 6-12 months once growth kinetics have been established.
- Serial imaging should use a consistent radiographic modality (CT, MRI, or US)
- The most reliable method of reporting renal lesion growth is measurement of maximum tumor diameter at the same axial (cross-sectional) cuts to reduce variability
- Additional methods of measuring tumor growth include estimated tumor volume (ETV), tumor doubling time, or the relative proportional change in tumor diameter or volume per year
If active surveillance is selected, prior to initiating the protocol the patient and physician need to agree upon defined triggers for intervention. These vary but include a rapid linear growth rate, appearance of new lesions or evidence of disease progression, or the onset of new clinical symptoms. Serial observation mandates rigorous patient compliance, and patients at risk to be lost to follow up, or those that have difficulty traveling to appointments may not be appropriate candidates as they are at risk to miss early detection of disease progression during the window that it may most effectively be treated.
Growing concern regarding the over diagnosis and treatment of patients with indolent incidentally detected tumors has led to the adoption of active surveillance as an alternative management strategy in select patients with small localized renal tumors. Further, as evidence accumulates that partial nephrectomy may not afford as great a survival advantages as initially expected compared to radical nephrectomy,34 an initial period of active surveillance has become a more attractive proposition in the elderly and infirmed.
Although the cumulative experience is limited, the literature suggests that tumors managed expectantly grow slowly, with low intermediate term risk of progression to metastatic disease. Currently, clinical tools to predict malignant tumor behavior are in their infancy, and in patients with tumors under a period of observation, tumor growth kinetics on serial radiographic imaging may be the most effective mechanism to guide the need for definitive intervention. Although fraught with accrual challenges, prospective randomized trials are sorely needed to clearly delineate risks of disease progression as well as define optimal surveillance regimens. Until the natural history of untreated renal tumors becomes more clearly defined or improved means of matching treatment to individual tumor biopsy is achieved, active surveillance for SRMs must remain restricted to select patients with extensive competing risks to mortality, limited life expectancy, or those refusing conventional treatment such as surgical resection or ablation.
1. Hollingsworth JM, Miller DC, Daignault S, et al. Rising incidence of small renal masses: a need to reassess treatment effect. J Natl Cancer Inst. 98(18),1331-4 (2006).
2. Campbell SC, Novick AC, Belldegrun A, et al. Guideline for management of the clinical T1 renal mass. J Urol. 182(4),1271-9 (2009).
3. Corcoran, A., Smaldone, M.C., Kutikov, A., et al. Addressing the evidence gap: Randomized controlled trials for the surgical management of localized genitourinary malignancies (in press, Urol Oncol).
4. Van Poppel H, Da Pozzo L, Albrecht W et al. A prospective, randomised EORTC intergroup phase 3 study comparing the oncologic outcome of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol. 2011 Apr;59(4):543-552.
5. Lane BR, Abouassaly R, Gao T, et al. Active treatment of localized renal tumors may not impact overall survival in patients aged 75 years or older. Cancer. 2010;116:3119-3126.
6. Arrontes DS, Acenero MJF, Gonzalez JIG, et al. Survival analysis of clear cell renal carcinoma according to the Charlson comorbidity index. J Urol. 2008;179:857-861.
7. Hollingsworth JM, Miller DC, Daignault S, et al. Five-year survival after surgical treatment for kidney cancer:a population-based competing risk analysis. Cancer. 2007;109:1763-1768.
8. Kutikov A, Egleston BL, Wong YN, et al. Evaluating overall survival and competing risks of death in patients with localized renal cell carcinoma using a comprehensive nomogram. J Clin Oncol. 2010;28:311-317.
9. Kutikov A, Egleston BL, Canter D, et al. Competing risks of death in patients with localized renal cell carcinoma: a comorbidity based model. J Urol. 2012;188:2077-2083.
10. Lamont EB, Christakis NA. Complexities in prognostication in advanced cancer: “to help them live their lives the way they want to.” JAMA. 2003:290:98.
11. Kutikov A, Fossett LK, Ramchandani P, et al. Incidence of benign pathologic findings at partial nephrectomy for solitary renal mass presumed to be renal cell carcinoma on preoperative imaging. Urol. 68(4),737-40 (2006).
12. Frank I, Blute MI, Chevelle JC, et al. Solid renal tumors: an analysis of pathological features related to tumor size. J Urol. 2003;170:2217-2220.
13. Thompson RH, Kurta JM, Kaag M, et al. Tumor size is associated with malignant potential in renal cell carcinoma cases. J Urol. 2009;181:2033-2036.
14. Kunkle DA, Crispen PL, Li T, et al. Tumor size predicts synchronous metastatic renal cell carcinoma: implications for surveillance of small renal masses. J Urol. 2007;177:1692-1696.
15. Nyugen MM, Gill IS. Effect of renal cancer size on the prevalence of metastasis at diagnosis and mortality. J Urol. 2009;181:1020-1027.
16. Rothman J, Egleston B, Wong YN, Iffrig K, Lebovitch S, Uzzo RG. Histopathological characteristics of localized renal cell carcinoma correlate with tumor size: a SEER analysis. J Urol. 2009;181:29-33.
17. Nguyen MM, Gill IS. Effect of renal cancer size on the prevalence of metastasis at diagnosis and mortality. J Urol. 2009;181:1020-1027.
18. Blumenfeld AJ, Guru K, Fuchs GJ,et al. Percutaneous biopsy of renal cell carcinoma underestimates nuclear grade. Urol. 2010;76:610-613.
19. Jewett MA, Mattar K, Basiuk J, et al.Active surveillance of small renal masses: progression patterns of early stage kidney cancer. Eur Urol. 2011;60:39-44.
20. Halverson SJ, Kunju LP, Bhalla R, Accuracy of determining small renal mass management with risk stratified biopsies: confirmation by final pathology. J Urol. 2013 Feb;189(2):441-6.
21. Rothman J, Crispen PI, Wong YN, et al. Pathologic concordance of sporadic synchronous bilateral renal masses. Urol. 2008;72:138-142.
22. (Divgi CR, Uzzo RG, Gatsonis C,et al. Positron Emission Tomography/Computed Tomography Identification of Clear Cell Renal Cell Carcinoma: Results From the REDECT Trial. J Clin Oncol. 2013;3:187-94.
23. Lane BR, Babineau D, Kattan MW, et al. A preoperative prognostic nomogram for solid enhancing renal tumors 7 cm or less amenable to partial nephrectomy. J Urol. 2007;178:429-434.
24. Jeldres C, Sun M, Liberman D, et al. Can renal mass biopsy assessment of tumor grade be safely substituted for by a predictive model? J Urol. 2009;182:2585-2589.
25. Kutikov A, Uzzo RG. The R.E.N.A.L. nephrometry score: a comprehensive standardized system for quantitating renal tumor size, location and depth. J Urol. 2009;18:844-853.
26. Kutikov A, Smaldone MC, Egleston BL, et al. Anatomic features of enhancing renal masses predict malignant and high-grade pathology: a preoperative nomogram using the RENAL Nephrometry score. Eur Urol. 2011;60:241-248.
27. Chawla SN, Crispen PL, Hanlon AL, et al. The natural history of observed enhancing renal masses: meta-analysis and review of the world literature. J Urol. 2006;175:425-31.
28. Smaldone MC, Kutikov A, Egleston B, et al. Small renal masses progressing to metastases under active surveillance: a systematic review and pooled analysis. Cancer. 2012;1184-1197.
29. Crispen PL, Wong YN, Greenberg RE, et al.Predicting growth of solid renal masses under active surveillance. Urol Oncol. 2008;26:555-9.
30. Kawaguchi S, Fernandes KA, Finelli A, et al. Most renal oncocytomas appear to grow: observations of tumor kinetics with active surveillance. J Urol. 2011;186:1218-1222.
31. Kunkle DA, Crispen PL, Chen DY, et al.Enhancing renal masses with zero net growth during active surveillance. J Urol. 2007;177:849-53.
32. Crispen PL, Viterbo R, Boorjian SA, et al. Natural history, growth kinetics, and outcomes of untreated clinically localized renal tumors under active surveillance. Cancer. 2009;115:2844-2852.
33.Crispen PL, Viterbo R, Fox EB, et al. Delayed intervention of sporadic renal masses undergoing active surveillance. Cancer. 2008;112:1051-1057.
34. Smaldone MC, Egleston B, Uzzo RG, et al. Does partial nephrectomy result in a durable overall survival benefit in the medicare population? J Urol. 2012;188:2089-94. KCJ