Renal cell carcinoma: new insights and challenges for a clinician scientist

Abstract

There is a growing recognition of the complex interplay between renal cell cancer (RCC), kidney function, mechanical reduction of nephron mass, and systemic agents targeting the cancer. Earlier detection of RCC and rising life expectancy of cancer survivors places a greater emphasis on preservation of renal function after cancer resection and during systemic therapy. Unique adverse effects associated with RCC drugs not only help reveal cancer pathophysiology but also expand our knowledge of normal cell signaling and metabolism. In this review, we outline our current understanding of RCC biology and treatment, their bidirectional relationship with kidney function, and unmet research needs in this field.

renal cell carcinoma (RCC) is the most common malignancy originating in the kidney and the most lethal genitourinary cancer (110). It has been classically characterized by an asymptomatic disease course, with a late and highly variable presentation portending a poor survival prognosis (43). Although its management in the past involved primarily urologists and oncologists, more recently, changing trends in RCC epidemiology, new therapeutic options, and overall improvement in patients’ life expectancy has drawn in nephrologists in a multidisciplinary approach to this disease and generated interest in the emerging field of onco-nephrology.

Epidemiology and Diagnostic Evaluation

Approximately 65,000 new kidney cancers are diagnosed in the US annually. Clear cell RCC (ccRCC) comprises 70%, papillary and chromophobe RCC 25%, and the more rare tumors of the medullary and collecting system origin 5% of these cancers (Table 1) (31, 96). These diagnoses require surgical resection or percutaneous needle biopsy of the mass and must be viewed in the context of overall prevalence of solitary renal lesions >1 cm estimated to be 12–14% (18). The majority of these (70%) are simple cysts (21) and in most cases do not require further workup or followup.

Table 1.

Histopathological classification of renal cell carcinoma and selected clinical characteristics of its subtypes (see Refs. 16, 68, 93, and 115)

RCC Type	Clear Cell	Papillary	Chromophobe	Collecting Duct
Cell of origin	Proximal tubular	Proximal tubular	Intercalated	Collecting duct
Relative prevalence among RCC	80–90%	10–15%	4–5%	<1%
Cancer-specific 10-yr survival	71%	91%	88%	34%
Prevalence of distant metastases on presentation	15%	3%	4%	32%
Recommended first-line therapy for advanced disease	IL-2 immunotherapy	Anti-VEGF	Anti-VEGF	Cisplatin and gemcitabine

RCC, renal cell carcinoma.

Reduction of glomerular filtration rate (GFR) is associated with progressively higher risk of RCC (20, 75), although a confounding of the shared risk factors, such as smoking, hypertension, diabetes, and obesity, or the altered metabolism and immunology associated with chronic kidney disease (CKD) is possible (15, 19, 60). Incidence of both simple cysts and RCC increases 10- to 100-fold in patients with end-stage renal disease (ESRD) and is proportionate to the dialysis dependence duration (9, 29). In the absence of screening guidelines, this has led to significant variability in screening protocols among nephrology practices. Ultrasonography is frequently employed to that end and can reliably identify simple cysts on the basis of the characteristic triad of findings, such as 1) round shape with smooth walls, 2) absence of echoes within the lesion, and 3) strong signal reflecting off the posterior cyst wall (25). Conversely, complex cysts with multiple thickened septa often containing calcifications and all solid lesions <4 cm enhancing with contrast on computer tomography qualify as small renal masses (SRMs) and warrant a referral to a urologist because of their high (70–80%) malignancy potential (37, 66, 81). The significant remainder of the SRMs is thus benign, which, together with the growing occurrence of incidental masses, has led to implementation of active surveillance with serial imaging as a safe initial approach to managing SRMs, especially in patients with high competing risks of death (1, 11). The same factors led to the emergence of tumor biopsy as an important adjunct to the diagnostic assessment of SRMs in unlikely surgical candidates. Although historically avoided in suspected RCC because of the risk of needle track seeding, more recently, SRM biopsy has been shown to be a safe method of obtaining an accurate diagnosis differentiating between RCC, renal metastasis from a distant malignancy, lymphoma, abscess, and benign cyst (2, 78). It also provides tissue for genomic and molecular marker analyses that can potentially further individualize treatment approach.

Basic Biology

Traditional classification of RCC is based on histomorphology of tumors, separating them into the three main types: clear cell (ccRCC), papillary (pRCC), and chromophobe (chRCC). Recent advances in the genomic profiling of RCC yielded further differentiation of tumor types, allowing for better understanding of their biology and greater specificity of treatment targets.

ccRCC

The most common type of renal cancer, ccRCC, originates from the proximal tubular cells. Genetic aberrations in ccRCC are broadly divided into those affecting the von Hippel-Lindau (VHL) gene, epigenetic regulators/chromatin remodeling genes, and disrupting phosphatidylinositol 3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) signaling.

Most cases are sporadic, and only 2–3% of ccRCC are accounted for by hereditary diseases, including VHL autosomal dominant syndrome, characterized by a germline mutation of the VHL tumor suppressor gene on chromosome 3p25 (73). Conversely, mutations or silencing of the same VHL gene are associated with >80% of sporadic ccRCC (71, 85). The best known function of VHL protein (pVHL) is polyubiquitination of several proteins, including hypoxia-inducible factor (HIF)-1 and -2, thereby marking them for degradation in the presence of normal tissue oxygenation (79). Under hypoxic conditions, HIF activates transcription of hypoxia-inducible genes, resulting in increased production of erythropoietin, vascular endothelial growth factor (VEGF), PDGFβ, and transforming growth factor-α (TGF-α) and several glycolytic enzymes (62, 76, 107). Failure of pVHL to reign in the activity of HIF in normoxia has crucial implications for the clinical behavior and metabolic features of ccRCC. Among them are erythropoietin-dependent erythrocytosis, marked angiogenesis, and cell proliferation mediated by TGFα and VEGF (62, 113) as well as aerobic glycolysis (Warburg effect) and immunosuppressive effects of increased tryptophan catabolism (125). The nature of the interactions between HIF and cellular metabolism is likely bidirectional, as shown by the inhibitory effects of the gluconeogenic enzyme fructose-1,6-bisphosphatase on HIF function (69). Although some studies showed an association between such metabolic abnormalities and tumor aggressiveness, metabolomic analyses have not yet become a part of clinical practice (10).

The second-most common type of mutation involves several epigenetic regulators also localized on 3p25 chromosome, including Polybromo 1 (PBRM1), BRCA-associated protein-1 (BAP1), and SET domain containing 1 (SETD2) genes involved in regulation of chromatin maintenance and remodeling. Generally, hypermethylation of promoter sites is associated with the tumors carrying the above mutations (with the exception of SETD1) and higher tumor grade, but their exact role in ccRCC pathogenesis is not well understood (26, 30, 45, 97, 122).

Genetic aberration of mTOR pathway proteins (PTEN, Akt, PIK3CA, and mTOR) involved in cell proliferation signaling have been identified in ≤28% of RCC patients (12). Deficiency in PTEN in particular is associated with more aggressive tumors (108). Together with the abnormal expression of focal adhesion kinase modulator of mTOR pathway, these genetic defects are present in >50% of ccRCC patients (12). As a part of the Cancer Genome Atlas project, less frequent mutations of genes involved in cell metabolism and division have been described, with some of them, such as p53 overexpression, associated with a worsening prognosis (12).

Papillary RCC

This second-most common RCC is also of proximal tubule origin but much less studied (13). Type 1 papillary RCC (pRCC) is characterized by MET proto-oncogene-activating mutations occurring in germ cells in the hereditary form of the disease and 10–20% of cases of somatic mutations of the sporadic (and most common) form of the disease (106). Additionally, trisomy 7 has been implicated in MET amplification in some cases of pRCC (65). Type 2 pRCC is associated with activation of the NRF2/antioxidant response element pathway due to the increased oxidative stress. Mutations in the fumarate hydratase (FH) gene in this disorder are associated with hereditary leiomyomatosis and renal cell carcinoma (HLRCC) (71, 116). Mutations in epigenetic modifier genes PBRM1, BAP1, and SETD2 have been described with this tumor type but seem to be less prevalent compared with ccRCC (65). Type 2 pRCC portends a considerably worse prognosis compared with type 1 pRCC (3, 86).

Chromophobe RCC

This rare cancer arising from the collecting duct and resembling benign oncocytoma comprises <5% of RCC and is commonly associated with whole chromosome losses and germline mutations in the folliculin gene (FLCN) found in the autosomal dominant Birt-Hogg-Dube syndrome (74, 117). The most common mutation in sporadic cases of chRCC involves downregulation of p53 tumor suppressor signaling and loss of function of PTEN, leading to PI3K-driven cell proliferation. ChRCC is more common in younger females and is typically the least aggressive of all RCC types, unless it is characterized by sarcomatous transformation (27).

Management

As more RCC are diagnosed at earlier stages, resection of smaller solitary tumors offers a cure for most patients (121). Similarly, advances in systemic therapies improve life expectancy in patients with advanced metastatic disease. Together, these positive trends highlight the importance of preserved renal function for RCC patients’ survival.

Surgical Strategies

Complete surgical tumor resection remains the standard of care for all localized RCC, whereas cytoreductive surgery improves the survival of patients with metastatic disease undergoing immunotherapy (83) and possibly molecularly targeted therapies (17, 23). Cryo- and radiofrequency ablation are associated with fewer perioperative complication profiles but higher cancer recurrence rates and are usually reserved for patients unfit for surgery (44, 57, 98). Radical and partial nephrectomies (RN and PN) seem to offer similar overall survival on the basis of a single randomized controlled trial (EORTC) (121) and the largest meta-analysis to date that included over 180,000 patients from 107 studies (98). Some cohort studies, however, demonstrated a more favorable renal functional prognosis associated with PN (49, 70, 126), although the absolute volume of resected parenchyma may not necessarily correlate with renal outcomes (42). In a meta-analysis from 2012, PN was associated with a 61% risk reduction for the incident severe CKD (61).

Postsurgical Sequelae

The overall prevalence of CKD stage 3 or higher is ∼50% after tumor resection (46) and is associated with cardiovascular complications, including death (124). Although significant proportion of RCC patients have CKD at baseline, nephrectomy-induced decline in GFR has been well described and attributed to the nephron loss. In recognition of this surgical CKD significance, there has been an increase in PN rates over the past decade. Despite this trend, RN remains indispensable for patients with larger tumors, and its effects are yet to be fully understood. In fact, a recent study demonstrated that larger tumor volume is associated with better postoperative renal function in patients undergoing RN (128). This finding suggests possible but unconfirmed significant compensatory hyperfiltration by the contralateral kidney preceding the actual RN. However, a similar phenomenon has been observed in living kidney donors (58), raising concerns about the validity of MDRD or CKD-EPI formulas for eGFR in staging the CKD in the settings of surgically induced nephron loss. There have been several attempts to employ three-dimensional image reconstruction to correlate tumor and renal volumes with eGFR (72). One interesting approach to estimating change in renal function due to tumor resection involves correcting eGFR by functional renal volume (eGFR/FRV), which emerged as an independent predictor of long-term postoperative renal function. However, the eGFR/FRV index remains to be validated in terms of long-term consequences of CKD defined in this particular fashion (48).

In addition to RN, preoperative hypertension, diabetes, reduced GFR, and albuminuria are the recognized risk factors for postoperative CKD (28, 48, 49). They are present frequently in RCC patients and are potentially modifiable. Albuminuria in particular has been given much attention in recent nephrology literature and has been established as a strong and independent predictor of adverse cardiovascular outcomes and mortality (40). Importantly, albuminuria is a manifestation of glomerular injury in many systemic and kidney diseases, including membranous and membranoproliferative glomerulonephritis (GN), focal segmental glomerulosclerosis (FSGS), IgA nephropathy, minimal change disease, amyloidosis, and crescentic GN, that are found in 16–43% of patients undergoing nephrectomies (6, 7, 47, 105) and may prove helpful in managing postoperative CKD. However, albuminuria is largely overlooked in both clinical urology practice and research, with most studies on proteinuria relying on semiquantitative methodology (48, 64). Moreover, only a minority of nonneoplastic tissue samples are formally examined (4, 105), squandering an opportunity for the safe establishment of a formal histopathological diagnosis for many patients that would benefit from a disease-specific or disease-modifying therapy immediately or later in life. Because CKD is a well-established risk factor for cardiovascular disease, all-cause mortality, and ESRD (38, 41), the 50%, 5-yr mortality of which exceeds that of RCC itself (22, 94), efforts of urologists, nephrologists, and pathologists should be coordinated and concentrated on minimizing the risks of GFR decline in the perioperative period in RCC patients.

Systemic Therapies

Systemic treatment is indicated for patients with unresectable or metastatic RCC (Fig. 1). IL-2 immunotherapy is historically the first-line treatment option intended to direct the immune system against the tumor. Because of its prohibitively high rate of severe toxicities, IL-2 treatment is currently offered only to a small proportion of exceptionally healthy RCC patients, in whom it may produce a durable response (63). Antiangiogenic drugs are the mainstay for the remaining patient majority and those whose disease progresses after immunotherapy, whereas immune checkpoint inhibitors may be offered as part of a clinical trial. Failure of antiangiogenic drugs in patients who were not exposed to immunotherapy warrants initiation of immune checkpoint inhibitors. Otherwise, a different antiangiogenic agent can be offered. Mechanistic target of rapamycin (mTOR) inhibitors may be considered as a third-line therapy or for patients with non-ccRCC.

Fig. 1.

Mechanisms of action of commonly used systemic agents in treatment of renal cell carcinoma (RCC). Multiple strategies have been developed for the treatment of advanced RCC. High-dose IL-2 is still the first-line treatment for the minority of patients with intact organ function. If available, protocol treatment with nivolumab is another preferred strategy aimed at activating an immune response against RCC. It can also be used for patients with progression of disease on molecularly targeted therapy that includes vascular endothelial growth factor (VEGF) antibody, bevacizumab, and multiple small-molecule tyrosine kinase inhibitors (TKI) active against VEGF receptor and, in the case of sorafenib, against members of the mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinases (ERK) that promote endothelial cell proliferation and angiogenesis. Mammalian target of rapamycin (mTOR) inhibitors, including everolimus and temsirolimus, disrupt growth factor signaling via the phosphatidylinositol-3-kinase (PI3K)/protein kinase B (Akt)/mTOR pathway regulating cell proliferation. At present, these agents are recommended for patients not responding to first-line anti-VEGF therapy.

The heterogeneic nature of RCC poses significant challenges in evaluating patient treatment responses. Recently expanding knowledge of ccRCC pathophysiology in particular continues to produce novel therapeutic agents tailored to the specific biological pathways described below.

Antiangiogenic Agents

The angiogenic response in RCC tissue characterized by “functional hypoxia” induced by VHL gene aberrations made this tumor a prime target for the first intravenous antiangiogenic drug, bevacizumab, that was approved for use in RCC in combination with immunomodulator interferon-α in 2009 (36, 102). This regimen is reported to be associated with the overall survival (OS) of ≤23 mo (34). Since then, several small molecule tyrosine kinase inhibitors with anti-VEGF activity and comparable efficacy have been developed for oral administration (35, 91, 101, 112). An extensively studied representative of this group, sunitinib, exerts its antiangiogenic and antiproliferative effects by inhibiting VEGF (1 and 2) and PDGF (α and β) receptors and is a common first-line agent for patients with prior exposure or contraindications to immunotherapy and is reported to prolong OS to 26 mo (90). Another recently approved first-line agent, pazopanib, has similar receptor targets and efficacy but somewhat better side effect profile compared with sunitinib (89). Axitinib is thought to be a more potent TKI with a wider range of receptor targets (VEGF1-3) compared with suntinib. Although it is a second-line agent, its dose adjustability can be potentially advantageous for balancing the incidence of side effects and maximizing clinical outcomes in select patients (101, 103). A distinguishing feature of another second-line agent, sorafenib, is its additional activity against Raf kinases that can be important for ≤50% of RCC patients with constitutive activation of the BRAF pathway (120). Both sorafenib and axitinib are associated with similar progression-free survival (53), but neither has been directly compared with sunitinib or pazopanib.

Among the side effects associated with antiangiogenic therapy, hypertension and proteinuria are frequently the reason for referral to a nephrologist. They have been well described, but the relationship between the two remains poorly understood (55, 77). Limited data from human kidney tissue and murine experimental models suggest the crucial role of podocyte-produced VEGF (33). Its selective inhibition results in the loss of glomerular endothelium fenestration, which promotes thrombotic microangiopathy and resultant hypertension (32). Additionally, hypertension is promoted by inactivation of VEGF-mediated endothelial nitric oxide synthase and prostacycline downregulation (104, 109). Unlike in rodents, proteinuria and hypertension in humans are much less uniform; both appear to be dose dependent and often diminish with dose reduction, treatment break, or addition of an antihypertensive agent, which is often an angiotensin-converting enzyme inhibitor (ACEi) (67, 103, 127). In some reports, incidents of anti-VEGF-induced hypertension have been described as markers or treatment responses and appear to be associated with prolonged patient survival; however, this relationship has been inconsistent (51, 84, 100). The choice of an antihypertensive agent has been addressed in several retrospective studies uniformly reporting better survival in patients treated with ACEi or angiotensin receptor blockers (ARB), highlighting the plausible role of angiotensin II in tumorigenesis (54, 59, 80). This effect held true for patients starting these agents before and after anti-VEGF treatment, although a significant proportion of the former group did not have the actual preexisting diagnosis of hypertension (80). Because significant creatinine elevation frequently leads to cancer therapy suspension or discontinuation, the optimal use of ACEi and ARB remains to be elucidated in the settings of RCC therapy frequently preceded by cytoreductive nephrectomy.

mTOR Inhibitors

Intravenous temsirolimus and oral everolimus are two derivatives of rapamycin most commonly used in patients with advanced metastatic RCC. Both agents form a complex with an intracellular protein FKPB-12, which blocks mTOR signaling. Disruption of the mTOR activation leads to G1 phase cell cycle arrest and HIF-1, HIF-2, and VEGF level reduction attenuating angiogenesis (82). These agents are commonly used for high-risk patients with nonclear cell RCC, confirmed PI3K pathway mutations, or those with disease refractory to VEGF-targeting drugs (50, 52, 88). Although generally well tolerated and described to have antifibrotic effects in renal tissue (8), mTOR inhibitors have been reported to be associated with AKI in renal transplant population (5) and cause significant glomerulosclerosis and proteinuria in animal models of reduced nephron volume (123). A possible explanation of these discordant treatment effects may be related to the dose of the drug and the predominant cell type (mesangial, endothelial, or epithelial) injured before mTOR therapy initiation (118).

Immune Checkpoint Inhibitors

Monoclonal antibodies against programmed cell death 1 (PD-1) protein have become an important option in the treatment of advanced RCC. PD-1 prevents activation of T cells by binding its target ligands PD-L1 and PD-L2, which are expressed on antigen-presenting cells; therefore, PD-1 inhibition is thought to augment the immune response against tumor cells (95). Compared with everolimus, intravenous nivolumab showed an improved overall survival of RCC patients, who had previously been treated with anti-VEGF drugs, and was approved by the FDA for RCC in 2015 (87). Unique toxicities of these agents termed immune-related adverse effects include colitis, hepatitis, interstitial nephritis, and inflammatory disorders of other organ systems that may limit therapy duration (92). In cases of biopsy-confirmed acute interstitial nephritis, renal dysfunction seems to readily resolve with immunotherapy discontinuation and short courses of glucocorticoids (24). With a rapid increase in the number of available therapeutic agents, multiple studies have been launched to compare the efficacy of combinations of drugs from different classes and their sequential use.

Conclusion and Future Directions

Significant stage migration of RCC at presentation due to earlier detection of smaller renal masses and greater life expectancy with novel treatment strategies associated with unique adverse effects offer fertile ground for research aimed at improving patient survival and understanding the molecular biology of RCC (Fig. 2). A significant proportion of RCC patients with metastatic disease at presentation calls for reliable screening biomarkers. Increased levels of markers of programmed cell death or altered angiogenesis such as cytokeratins or (PHD3) antibodies, respectively, have been detected in the serum of RCC patients (111, 114). Multiple biomarkers such as osteopontin, VEGF, and interleukins have been tested individually and in combinations to predict disease progression and response to different therapies with variable degrees of success (56, 99, 119). Metabolomic analysis of serum lipids and amino acids yielded promising preliminary results for RCC identification (39, 130). However, when applied to patients with resectable tumors, metabolomic analyses appear to be greatly impacted by the changes in body metabolism arising from the surgically reduced nephron volume (129). The underutilized opportunity to analyze nonneoplastic tissue removed during nephrectomy offers a prospect of discovering tissue biomarkers of postoperative CKD progression and adverse effects from systemic therapies. For example, discovery of immunoglobulin A nephropathy at the time of resection may inform oncologists against using everolimus in such patients with active glomerular inflammation that can be worsened by mTOR inhibition. Incident hypertension and proteinuria observed with anti-VEGF agents have not been adequately studied, and although their development often triggers treatment modifications, the rationale for treatment adjustment is not guided by the knowledge of long-term outcomes. In particular, the choice between ACEi and ARB for treatment of hypertension is currently overlooked but may be important because of the ARB’s preferential inhibition of angiotensin receptor (ATR) 2, which, unlike ATR1, mediates antiangiogenic effects (14). Common use of dipstick urinalysis in RCC patients is inadequate for accurate quantification of urine protein and may grossly underestimate the prevalence of thrombotic microangiopathy, its role in the development of anti-VEGF-induced hypertension, and the appropriate supportive therapy. The recently characterized genetic profiles of different histological types of RCC frequently overlap and may lead to reclassification of RCC while offering a more individualized approach to treatment strategy and establishing new treatment targets.

Fig. 2.

Research questions for future investigations in the areas of renal cell cancer diagnosis and treatment. CKD, chronic kidney disease; GFR, glomerular filtration rate; mTOR, mammalian target of rapamycin; RCC, renal cell carcinoma; VEGF, vascular endothelial growth factor.

The complexity of RCC calls for a collaborative effort of basic scientists and medical specialists directed at advancing and consolidating our knowledge of this tumor.

GRANTS

We acknowledge funding from National Institutes of Health/NCI Cancer Center Support Grant P30-CA-00874 and the Byrne Research Fund.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.S. and E.A.J. prepared figures; R.S. drafted manuscript; R.S. and E.A.J. approved final version of manuscript; E.A.J. edited and revised manuscript.