The majority of kidney cancer tumors are small renal masses (SRMs). Partial nephrectomy is now established as the preferred treatment modality. In some patients the potential morbidity may outweigh the oncologic risk. Based on active surveillance studies, restriction of radical therapy to patients with aggressive tumors has been proposed. This has spurred renewed interest in development of radiologic and biopsy-based diagnostic techniques that can identify high-risk disease. This article discusses the natural history and pathologic features of SRMs, the evolving role of biopsy, and provides an overview of outcomes of various treatment approaches and current recommendations for management.
Renal cell carcinoma (RCC) accounts for approximately 4% of cancers and 2% of cancer mortality in the United States. In the past, most patients presented with advanced-stage disease and symptoms of gross hematuria, flank pain, and palpable mass. Over the past two decades, renal masses have been increasingly diagnosed as an incidental finding due to the increased use of cross-sectional imaging. At present, more than 60% of renal cancers are discovered in patients undergoing evaluation for unrelated conditions, a fact which has prompted some investigators to term this the “radiologist’s tumor.” Small renal mass (SRM) specifically refers to a contrast-enhancing renal lesion measuring less than or equal to 4 cm which corresponds to the American Joint Committee on Cancer clinical staging designation of T1a. The incidence of SRMs increased twofold between 1993 and 2003, and the percentage of patients who initially presented with tumors less than or equal to 3 cm increased from 33% to 43% over this period. This trend is expected to continue. As a result the SRM has become a frequently encountered clinical dilemma.
Increased rates of renal surgery for T1a tumors have paralleled a rise in detection. It is expected that the shifts in both early detection and increased use of curative-intent surgery will result in improved oncologic and overall survival of patients with RCC. However, cancer-specific mortality has not decreased, with at least one study suggesting that mortality rates have increased for local-stage disease. This rising mortality has called the current paradigm of aggressive surgical therapy for SRMs into question. The observation that early detection and early intervention are not resulting in improved survival rates suggests that nononcologic risk factors weigh heavily on the overall survival of patients treated for SRMs. There has been a trend toward declining health and increasing age of patients presenting with SRMs. Most patients with incidental SRMs are between 60 and 70 years of age; 20% to 30% of patients with SRMs have preexisting chronic kidney disease (CKR); glomerular filtration rate (GFR) less than 60 mL/min per 1.73 m 2 . Many patients have significant comorbidities, including hypertension and cardiovascular disease. These patients are at higher risk for postoperative morbidity and mortality, and are at high risk of development of postoperative CKD.
Radical nephrectomy (RN) had long been held as the mainstay of treatment for RCC, regardless of tumor size. In the 1980s, partial nephrectomy (PN) emerged as an alternative to RN that provided equivalent oncologic control with the added benefit of functional preservation. Over the past two decades, a wealth of data has accrued that attests to the oncologic safety and superior functional outcomes of PN. There has been a steady shift toward use of PN, especially for SRMs. American Urologic Association guidelines published in 2009 specify PN as the reference standard for management of clinical T1 masses given the importance of functional preservation. PN is associated with lower risk of development of CKD compared with RN, but can still result in clinically significant functional decline in high-risk patients. Huang and colleagues reported that 65% of patients had new-onset CKD within 3 years after RN versus 20% in the PN population. In certain patients with SRMs, oncologic risk may be outweighed by competing causes of mortality.
Contemporary data from pathologic studies has revealed that 20% of SRMs are benign, whereas only 20% harbor potentially aggressive features. In this regard, the therapeutic index of RN or PN may be low for the majority of SRMs. By focusing radical therapy on patients with potentially aggressive disease, the net impact of surgery can be optimized. As a result, there has been renewed interest in development of novel radiologic and pathologic techniques that may allow for reliable diagnosis and risk assessment of kidney tumors. In this article, the authors discuss the pathobiology and natural history of SRMs, clinical risk assessment, the role of biopsy, outcomes of current standard treatments, and current recommendations for management based on recent guidelines published by the American Urological Association.
Pathologic features of SRMs
The kidney can give rise to a variety of neoplasms, all of which can be found in SRMs. Benign tumors include fat-poor angiomyolipoma, oncocytoma, metanephric adenoma, cystic nephroma, and a variety other nonmalignant histologies. Malignant tumors consist predominantly of RCC, but non-RCC tumors such as transitional cell carcinomas, sarcomas, and metastatic tumors are also occasionally encountered. RCC consists of a heterogenous group of tumors with varied aggressiveness and tumor architecture. Clear cell (conventional) RCC comprises 70% of tumors, whereas papillary (chromophil) and chromophobe subtypes comprise 15% and 5%, respectively. Of the common RCC subtypes, clear cell and papillary type II tumors have the highest risk for metastatic progression. Other RCC subtypes such as collecting duct carcinoma and medullary carcinoma can be particularly aggressive, although fortunately rare.
Histologic analysis of kidney tumors serves as the definitive basis for prognosis and clinical management. Table 1 outlines several contemporary pathologic studies that have looked specifically at pathologic features of SRMs. These studies represent the current clinical picture of SRM presentation, as the majority of tumors (60%–70%) in these studies were both incidentally detected and asymptomatic. Data from these studies confirm that 20% of SRMs are benign, that 20% are potentially aggressive RCC, and that 60% are potentially indolent RCC. These studies defined potential aggressiveness based on presence of high-risk pathologic features including Fuhrman nuclear grade greater than or equal to 3; presence of type II papillary, sarcomatoid or collecting duct histology; presence of invasion into the renal sinus capsule or vasculature (ie, Stage ≥T3a); or presence of metastasis at the time of diagnosis. Retrospective studies have shown clear associations between these high-risk pathologic features and advanced tumor stage and metastasis.
| Author | N | Percent Benign | Percent Indolent | Percent Aggressive | Percent Metastatic | Aggressiveness Criteria |
|---|---|---|---|---|---|---|
| Frank (2003) | 947 | 23% | 64% | 13% | NA | FG ≥3 |
| Remzi (2006) | 287 | 20% | 58% | 22% | 5% | Stage ≥T3a or metastatic |
| Schlomer (2006) | 206 | 23% | 52% | 25% | NA | FG ≥3 |
| Pahernik (2007) | 663 | 17% | 70% | 13% | 3% | FG ≥3, stage ≥T3a or metastatic |
| Lane (2007) | 862 | 20% | 56% | 24% | NA | FG ≥3 or stage ≥T3a |
| Mean Values | 20% | 60% | 20% | 4% |
Although histologic phenotype and Fuhrman nuclear grade is relied on to predict tumor behavior, it must be recognized that these prognostic features are limited. The assumption is that the clinical behavior of two clear cell Fuhrman grade 3 tumors will be similar based on shared pathologic features and information obtained from retrospective clinical outcomes studies. However, a recent study by Dalgliesh and colleagues illustrates the potential inaccuracy of phenotypic features to predict tumor behavior. The investigators conducted a large-scale genetic sequencing analysis in 101 clear cell RCC tumors, and found substantial variability in gene expression and mutation patterns. Although these tumors appear the same histologically, their clinical courses can be highly variable. This and other studies suggest that discerning genetic and molecular analysis along with clinical correlation will be required to more accurately predict tumor behavior.
Natural history of SRMs
The second series of important studies relating to the pathobiology of SRMs has focused on active surveillance (AS) of solid stage T1 lesions. Table 2 summarizes data from several larger contemporary AS studies. These data refer specifically to solid-enhancing SRMs that do not demonstrate evidence of aggressiveness at the time of study initiation. Most studies have followed patients for only 2 to 3 years, although emerging studies include follow-up as long as 5 years. In general, it was observed that SRMs grow slowly with an average diameter increase of 3 mm/year. In these studies, the incidence of de novo metastatic progression during the observation period was low at 0% to 5%.
| Author (Year) | N | Mean cm Tumor Size | Mean cm/y Growth Rate | Mean Months Follow-up | Metastatic Progression (%) |
|---|---|---|---|---|---|
| Bosniak et al, (1995) | 40 | 1.73 | 0.36 | 39.0 | 0 |
| Kato et al, (2004) | 18 | 2.0 | 0.42 | 26.9 | 0 |
| Lamb et al, (2004) | 36 | 7.2 | 0.39 | 27.7 | 1 (2.8%) |
| Volpe et al, (2005) | 32 | 2.48 | 0.10 | 38.9 | 0 |
| Wehle et al, (2004) | 29 | 1.83 | 0.12 | 32.0 | 0 |
| Abouassaly et al, (2008) | 110 | 2.5 a | 0.26 | 24.0 a | 0 |
| Kouba et al, (2007) | 46 | 2.92 | 0.70 | 35.8 | 0 |
| Kunkle et al, (2007) | 106 | 2.0 a | 0.19 a | 29.0 a | 1 (1.1%) |
| Youssif et al, (2007) | 41 | 2.2 | 0.21 | 47.6 | 2 (5.7%) |
| Crispen et al, (2008) | 173 | 2.45 | 0.285 | 31.0 | 2 (1.3%) |
| Rosales et al, (2010) | 223 | 2.8 a | 0.34 a | 35.0 a | 4 (1.9%) |
| Mean values | 2.7 | 0.31 | 33.0 | 1.2% |
Contemporary AS studies have provided important data; however, they have significant limitations. Patient selection bias is present as those selected for these studies are older, less healthy, and have tumors that are considered low risk upon initiation of observation. In this regard, AS study data may pertain only to this low-risk subset of SRMs. There are no objective parameters that define clinically significant tumor growth or change in appearance. “Surveillance failure,” defined as conversion from observation to intervention, is not necessarily indicative of disease progression because treatment initiation is based on subjective indications such as patient emotional burden, surgeon anxiety, and economic factors. Another limitation is that pathologic information is not available for a significant percentage of patients in these studies. Most patients do not undergo initial biopsy when recruited into AS and the only patients with post-surveillance pathology data are those that underwent treatment. This precludes prospective correlation of pathologic features with clinical behavior. Although current AS studies are not ideal, they do provide a loose framework for the general clinical behavior profile for select SRMs. It should be emphasized that these data may not be generalized to all SRMs.
Natural history of SRMs
The second series of important studies relating to the pathobiology of SRMs has focused on active surveillance (AS) of solid stage T1 lesions. Table 2 summarizes data from several larger contemporary AS studies. These data refer specifically to solid-enhancing SRMs that do not demonstrate evidence of aggressiveness at the time of study initiation. Most studies have followed patients for only 2 to 3 years, although emerging studies include follow-up as long as 5 years. In general, it was observed that SRMs grow slowly with an average diameter increase of 3 mm/year. In these studies, the incidence of de novo metastatic progression during the observation period was low at 0% to 5%.
| Author (Year) | N | Mean cm Tumor Size | Mean cm/y Growth Rate | Mean Months Follow-up | Metastatic Progression (%) |
|---|---|---|---|---|---|
| Bosniak et al, (1995) | 40 | 1.73 | 0.36 | 39.0 | 0 |
| Kato et al, (2004) | 18 | 2.0 | 0.42 | 26.9 | 0 |
| Lamb et al, (2004) | 36 | 7.2 | 0.39 | 27.7 | 1 (2.8%) |
| Volpe et al, (2005) | 32 | 2.48 | 0.10 | 38.9 | 0 |
| Wehle et al, (2004) | 29 | 1.83 | 0.12 | 32.0 | 0 |
| Abouassaly et al, (2008) | 110 | 2.5 a | 0.26 | 24.0 a | 0 |
| Kouba et al, (2007) | 46 | 2.92 | 0.70 | 35.8 | 0 |
| Kunkle et al, (2007) | 106 | 2.0 a | 0.19 a | 29.0 a | 1 (1.1%) |
| Youssif et al, (2007) | 41 | 2.2 | 0.21 | 47.6 | 2 (5.7%) |
| Crispen et al, (2008) | 173 | 2.45 | 0.285 | 31.0 | 2 (1.3%) |
| Rosales et al, (2010) | 223 | 2.8 a | 0.34 a | 35.0 a | 4 (1.9%) |
| Mean values | 2.7 | 0.31 | 33.0 | 1.2% |
Contemporary AS studies have provided important data; however, they have significant limitations. Patient selection bias is present as those selected for these studies are older, less healthy, and have tumors that are considered low risk upon initiation of observation. In this regard, AS study data may pertain only to this low-risk subset of SRMs. There are no objective parameters that define clinically significant tumor growth or change in appearance. “Surveillance failure,” defined as conversion from observation to intervention, is not necessarily indicative of disease progression because treatment initiation is based on subjective indications such as patient emotional burden, surgeon anxiety, and economic factors. Another limitation is that pathologic information is not available for a significant percentage of patients in these studies. Most patients do not undergo initial biopsy when recruited into AS and the only patients with post-surveillance pathology data are those that underwent treatment. This precludes prospective correlation of pathologic features with clinical behavior. Although current AS studies are not ideal, they do provide a loose framework for the general clinical behavior profile for select SRMs. It should be emphasized that these data may not be generalized to all SRMs.
SRM risk assessment
Radiologic imaging is the primary means by which kidney tumors are clinically evaluated. Cross-sectional imaging with CT scan or MRI allows for anatomic analysis of the kidney, tumor, vasculature, and adjacent structures. A three-phase study with noncontrast, arterial, and delayed excretory phases is ideal. Any lesion with enhancement of greater than or equal to 15 Hounsfield units (HU) on CT scan imaging is considered suspicious for RCC, whereas renal lesions demonstrating less than 10 HU of enhancement are predominantly benign. MRI with administration of gadolinium contrast is an acceptable alternative for evaluation of kidney tumors in patients with iodine contrast allergy and, at some institutions, MRI is the preferred routine imaging modality. Gadolinium should not be administered to patients with stage 4 CKD (ie, GFR ≤30 mL/min/1.73m 2 ) due to the risk for nephrogenic systemic fibrosis, a severe and potentially irreversible condition. Once it is established that the mass is a solid, contrast-enhancing, kidney tumor, it is assumed RCC until proven otherwise by pathologic analysis.
Demographic Risk Features
Patient demographic factors can provide insight into the clinical behavior of SRMs. Patients with multifocal small kidney tumors may have heritable syndromes such as von Hippel-Lindau (VHL), the familial form of clear cell RCC. Management strategy in patients with VHL and most familial syndromes of RCC is critical given the high incidence of de novo tumor growth and risk for CKD that can be associated with multiple surgical interventions, thus observation of tumors less than 3 cm is accepted practice. For sporadic RCC, age and gender appear to correlate with likelihood of malignancy. Younger women with contrast-enhancing solid SRMs have a benign tumor rate of 40% to 50%, which is much higher than for males. DeRoche and colleagues reported the rate of benign tumors in females at 27% compared with 15% for males. The rate of malignancy in this study for tumors less than or equal to 3.5 cm was both age and gender-dependent. SRM malignancy rates in patients less than 68 years of age were 87% for males and 69% for females. Younger patients of African descent with sickle cell trait are a special consideration because they have a predilection for medullary RCC. In this demographic, aggressive management of a SRM is advisable. Although demographic and clinical correlates may associate with certain tumors types, they are by no means diagnostic. For any SRM the likelihood of RCC is high, and these tumors are best managed surgically.
Tumor Size
Tumor size has been shown to correlate with malignancy and high nuclear grade. Two large studies have reported that for every 1 cm increase in tumor diameter there is a 16% to 17% increase in the odds of malignancy. In these studies, 38% to 46% of tumors measuring less than or equal to 1 cm were benign compared with 6% to 7% of tumors greater than or equal to 7 cm. The risk of high nuclear grade (ie, Fuhrman nuclear grade ≥3) disease also increases in relation to tumor size. Among clear cell RCC, each 1 cm increase in tumor size increased the odds of high grade disease by 32%. Variability in these data reinforce that all tumors should be assumed to be malignant regardless of size until proven otherwise on pathologic analysis.
When considering enrollment of a patient into an AS protocol it is generally agreed that smaller tumors are “safer” to observe than larger ones. One group has described surveillance outcomes in patients with T1a versus T1b-T2 masses. Failure of surveillance was defined as progression to metastatic disease or conversion from surveillance to delayed treatment. Of 41 patients with pT1a tumors, only one failed surveillance (2%), and none progressed to metastasis. In contrast, of 42 patients with T1b-T2 tumors, 14% failed surveillance and 6% progressed to metastasis disease. No study has yet been conducted to assess the risk differential within the 0 to 4 cm size range, but it is likely that similar trends would be observed. Most urologic oncologists would entertain the idea of surveillance for masses measuring 1 to 2 cm in appropriate scenarios and would elect to proceed to treatment once these masses reached the 3 to 4 cm range.
Tumor Growth Rate
Tumors that demonstrate rapid growth on serial imaging studies are assumed clinically aggressive. Bosniak and colleagues examined 40 cT1a incidental kidney tumors over a period of approximately 3 years and found the median tumor diameter growth rate to be 0.36 cm per year. No patients with a growth rate less than or equal to 0.35 cm/y exhibited metastatic progression within the 3 year study period. Similar studies have shown equivalent data with mean tumor growth rates ranging from 0.2 to 0.4 cm per year and no progression to metastatic disease in slow growing tumors. Although slow growth rate was associated with decreased metastatic rates, it did not correlate with malignancy rates. Kunkle and colleagues followed 106 enhancing masses for a minimum of 12 months and found that the proportion of malignant tumors was equivalent in tumors with and without growth, 89% versus 83% ( P = .56), respectively. Kato and colleagues found similar malignancy rates in slow and fast growing tumors, but did find a difference in the proportion of high Fuhrman nuclear grade greater than or equal to 3 cm in the latter group. Among masses that were malignant, high-grade tumors had a faster growth rate (0.93 cm/y) compared with low-grade tumors (0.28–0.37 cm/y). As more data emerge, a greater understanding of the relationship between tumor growth rate and metastatic potential will develop.
Tumor Enhancement and Morphology
Anatomic features that may signify aggressive tumor behavior include tumor morphology and contrast-enhancement characteristics. An invasive or infiltrative appearance is strongly associated with high-grade or sarcomatoid differentiation and poor prognosis. A homogeneous appearance suggests preservation of tissue organization, uniform histology, and ordered vascularity. Heterogenous enhancement suggests poor differentiation, loss of organization, central necrosis, and aberrant vascularization. Zhang and colleagues examined 193 tumors and found that 79% to 88% of tumors with heterogeneous appearance had clear cell RCC pathology. In contrast papillary and chromophobe tumors tended to have more homogenous appearance. These features can enhance tumor evaluation, but are too nonspecific to be used in isolation to guide clinical management.
Much attention has been paid to diagnostic potential of the degree of contrast enhancement. The basis for this is the observation that tumors of specific histology have different vascular organization. Contrast enhancement tends to be lowest in papillary RCC compared with other solid tumors. Alshumrani and colleagues examined 46 cT1 tumors and found contrast enhancement in clear cell RCC, oncocytoma, and papillary RCC tumors to be 65 HU, 80 HU, and 16 HU, respectively. Similarly, Sun and colleagues examined dynamic contrast-enhanced MRIs in 122 patients. Clear cell RCC had the greatest signal-intensity change (205.6% in corticomedullary phase and 247.1% in nephrographic phase), which was significantly higher than in papillary RCC (32.1% and 96.6%, respectively; P <.001). They reported 93% sensitivity and 96% specificity to distinguish between papillary and clear cell RCC using these parameters. The importance of accurately detecting these features is diminished because nearly half of papillary RCC tumors behave as aggressively as clear cell RCC. Type II papillary tumors exhibit 5 year cancer specific mortality rates double that of type I papillary RCC. Although tumor enhancement and complexity may suggest a specific tumor type or behavior, radiologic appearance is nonspecific and should be interpreted with caution.
Related posts:
Renal Cell Carcinoma
mTOR Inhibitors in Advanced Renal Cell Carcinoma
Future Directions in Renal Cell Carcinoma: 2011 and Beyond
The Role of Surgery in the Management of Early-Stage Renal Cancer
mTOR Inhibitors in Advanced Renal Cell Carcinoma
Systemic Therapy for Metastatic Non–Clear-Cell Renal Cell Carcinoma: Recent Progress and Future Directions
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
