Renal Cell Carcinoma

3. EPIDEMIOLOGY, AETIOLOGY AND PATHOLOGY

3.1. Epidemiology

Renal cell carcinoma represents around 3% of all cancers, with the highest incidence occurring in Western countries [6,7]. In 2020, there were an estimated 431,288 new cases of RCC globally, of which 138,611 in Europe [8]. The higher incidence in Europe and North America is hypothesized to be due to a higher prevalence of small renal masses (SRMs) in settings where abdominal imaging is more ubiquitous. In 2020, Lithuania reported the highest overall rate of RCC, followed by Czechia, with estimated age-standardised rates (ASRs) of 14.5/100,000 and 14.42/100,000, respectively. A person living in Czechia has a 2.83% risk of developing RCC [7,8]. Generally, during the last two decades until recently, there has been an annual increase of about 2% in incidence both worldwide and in Europe. In 2022, worldwide mortality from RCC was 179,368 deaths (115,600 men and 63,768 women), with a calculated global ASR rate of 1.8/100,000 [8].

In Europe, overall mortality rates for RCC increased until the early 1990s, with rates generally stabilising or declining thereafter [9]. There has been a decrease in mortality since the 1980s in Scandinavian countries and since the early 1990s in France, Germany, Austria, the Netherlands, and Italy. However, in some European countries (Croatia, Estonia, Greece, Ireland, Slovakia), mortality rates still show an upward trend [6,7].

Renal cell carcinoma is the most common solid lesion within the kidney and accounts for approximately 90% of all kidney malignancies. It comprises different RCC subtypes with specific histopathological and genetic characteristics [10]. There is a 1.5–2.0:1 predominance in men over women with a higher incidence in the older population [7,8,11].

3.2. Aetiology

Established risk factors include lifestyle factors such as smoking (hazard ratio [HR]: 1.23–1.58), obesity (HR: 1.71), BMI (> 35 vs. < 25), and hypertension (HR: 1.70) [7,8,11,12]. 50.2% of patients with RCC are current or former smokers. By histology, the proportions of current or former smokers range from 38% in patients with chromophobe carcinoma (chRCC) to 61.9% in those with collecting duct/medullary carcinoma [13]. In a recent systematic review diabetes was also found to be detrimental [14]. Having a first-degree relative with kidney cancer is also associated with an increased risk of RCC. Moderate alcohol consumption appears to have a protective effect for reasons as yet unknown, while any physical activity level also seems to have some protective effect [7,8,14-18]. A number of other factors have been suggested to be associated with higher or lower risk of RCC, including specific dietary habits and occupational exposure to specific carcinogens, but the literature is inconclusive [11]. The most effective prophylaxis is to avoid cigarette smoking and reduce obesity [7,8,11,12]. Genetic risk factors are known to play a role in the development of RCC (see Section 3.5.6 - Hereditary kidney tumours).

3.3. Screening

Despite a growing interest from both patients and clinicians in RCC screening programmes, there is a relative lack of studies reporting the efficacy, cost-effectiveness, and optimal modality for RCC screening. Urinary dipstick is an inadequate screening tool due to low sensitivity and specificity. No clinically validated urinary or serum biomarkers have as yet been identified. Computed tomography cannot be recommended due to cost, radiation dose and the increased potential for other incidental findings. Ultrasound (US) could be used and has acceptable sensitivity and specificity, although it is tumour size and operator dependant. Major barriers to population screening include the relatively low prevalence of the disease, the potential for false positives and over-diagnosis of slow-growing kidney tumours. Targeting high-risk individuals and/or combining detection of RCC with other routine health screenings may represent pragmatic options to improve the cost-effectiveness and reduce the potential harms of RCC screening [19-21]. Targeting of high-risk patient groups e.g., those with end-stage renal disease (ESRD) which is associated with a 10-fold increased risk of developing RCC may also be a valid approach (see Section 3.5.2) [22]. There is currently no evidence to support primary screening in the general population. However, the panel recommends genetic screening in subgroups of patients with a family history (see Section 5.5).

3.3.1. Summary of evidence and recommendation for epidemiology, aetiology and screening

3.4. Histological diagnosis

Renal cell carcinomas and other renal tumours comprise a broad spectrum of histopathological entities described in the 5^th edition of the World Health Organization (WHO) classification of urogenital tumours published in 2022 [23,24]. The 5^th edition presents standard morphologic diagnostic criteria, combined with immunohistochemistry and relevant molecular tests was significantly revised as compared to the 2016 classification [10]. The global application of next-generation sequencing (NGS) will result in a diagnostic shift from morphology to molecular analyses. Therefore, a molecular-driven renal tumour classification has been introduced in addition to morphology-based renal tumours (Table 3.1). Examples of molecularly-defined epithelial renal tumours include SMARCB1-deficient renal medullary carcinoma, TFEB-rearranged RCC, ALK-rearranged RCC, and elongin C (ELOC)-mutated RCC. The most profound changes in the 2022 WHO classification mainly relate to rare kidney tumours.

There are three main RCC types: clear cell (ccRCC), papillary (pRCC no longer divided into type I and II) and chRCC. The RCC type classification has been confirmed by cytogenetic and genetic analyses [10,25] (LE: 2b). The 5-year OS for non-metastatic (including N1) chromophobe, papillary, clear-cell and collecting duct RCC is 91%, 82%, 81% and 44%, respectively [26]. Sarcomatoid RCC is not a specific subtype, but essentially represents a pattern of de-differentiation associated with adverse outcome and poor cancer-specific survival (CSS), irrespective of the underlying RCC subtype; it should be graded as WHO/ISUP (International Society of Urological Pathology) grade IV. Multilocular cystic renal neoplasm of low malignant potential is a new subtype of cRCC in the 2022 classification. A new group “oncocytic and chromophobe tumours” encompass oncocytoma together with chRCC and other oncocytic tumours. Other oncocytic tumours include tumours that do not strictly fit into either the oncocytoma or chRCC subtypes [23,27].

Histological diagnosis includes, besides RCC type; evaluation of ISUP nuclear grade, sarcomatoid features, vascular invasion, tumour necrosis, and invasion of the collecting system and peri-renal fat, pT, or even pN categories. The four-tiered WHO/ISUP grading system has replaced the Fuhrman grading system [10,23].

Table 3.1 World Health Organization classification of renal tumours 2022 

3.4.1. Clear-cell RCC

Overall, clear-cell RCC (ccRCC) is well circumscribed and a capsule is usually absent. The cut surface is golden-yellow, often with haemorrhage and necrosis. Loss of chromosome 3p and mutation of the von Hippel-Lindau (VHL) gene at chromosome 3p25 are frequently found. The loss of von Hippel-Lindau protein function contributes to tumour initiation, progression, and metastases. The 3p locus harbours additional ccRCC tumour suppressor genes (UTX, JARID1C, SETD2, PBRM1, BAP1) [23]. In general, ccRCC has a worse prognosis compared to pRCC and chRCC, but this difference disappears after adjustment for stage and grade [28,29]. For details about prognosis, see Section 6.3.

3.4.1.1. Multilocular cystic renal neoplasm of low malignant potential (MCNLMP)

Indolent, exclusively cystic, multiloculated renal tumor devoid of any expansile solid growth, with clear cells lining with low grade nuclei. Detection of small solid expansive nodules and tumour necrosis are incompatible with MCNLMP. It represents 0.5–2.5% of all renal tumours and is a benign lesion. There are no reports of progression, metastases or cancer-related death with long-term follow-up [23,24]. Nephron-sparing surgery (NSS) is sufficient, if technically feasible [30].

3.4.2. Papillary RCC

Papillary RCC is the second-most encountered morphotype of RCC accounting for 13–20% of renal epithelial tumours. It is usually circumscribed and characterised by papillary or tubulopapillary architecture, without specific features of other RCCs with papillary architecture [23,24]. Papillary RCC has traditionally been subdivided into two types; Type I and II pRCC [10]. However, in the new 2022 WHO classification, the former pRCC type I is now referred to as “pRCC of classic pattern”. Three additional morphologic patterns of pRCC have been introduced including: a) bi-phasic (alveolo-squamoid) pattern exhibiting mostly solid growth; b) papillary neoplasm with reverse nuclear polarity, previously described as “oncocytic low-grade pRCC”; and c) Warthin-like pRCC that exhibits brisk inflammation mimicking Warthin tumour of the salivary gland.

Genetic changes of pRCC include trisomies and tetrasomies of chromosomes 7 and 17 and loss of Y chromosome. Mesenchymal-epithelial Transition (MET) gene mutations are more frequent in low-grade pRCC.

The typical histology of classical pattern pRCC, formally type I pRCC, (narrow papillae without any binding, and only microcapillaries in papillae) explains its typical clinical signs. Narrow papillae without any binding and a tough pseudo-capsule explain the ideal rounded shape (Pascal’s law) and fragility (specimens have a “minced meat” structure). Tumour growth causes necrotisation of papillae, which is a source of hyperosmotic proteins that cause subsequent “growth” of the tumour, fluid inside the tumour, and only a serpiginous, contrast-enhancing margin. Stagnation in the microcapillaries explains the minimal post-contrast attenuation on CT. Classical pattern pRCC can imitate a pathologically changed cyst (Bosniak IIF or III). The typical signs of classical pattern pRCC are an ochre colour, frequently exophytic, extra-renal growth and low grade. A risk of renal tumour biopsy tract seeding exists (12.5%), probably due to the fragility of the tumour papillae [31].

3.4.3. Chromophobe RCC

Chromophobe RCC is now grouped in “oncocytic and chromophobe tumours”. Most chRCCs are discovered incidentally in asymptomatic patients [23,24]. Overall, chRCC presents as a pale tan, relatively homogenous and tough, well-demarcated mass without a capsule. Most tumours are sporadic. Rare hereditary forms include Birt-Hogg-Dubé (BHD) syndrome with mutations in folliculin and Cowden syndrome with mutations in PTEN (see Section 3.5.6 for further information) [23,24]. Chromophobe RCC cannot be graded by the WHO/ISUP (formerly Fuhrman) grading system because of its innate nuclear atypia. An alternative grading system has been proposed, but has yet to be validated [23,24]. Loss of chromosomes Y, 1, 2, 6, 10, 13, 17 and 21 are typical genetic changes [23,24]. The prognosis is relatively good, with high 5-year recurrence-free survival (RFS), and 10-year CSS [32]. The five- and 10-year RFS rates were 94.3% and 89.2%, respectively. Recurrent disease developed in 5.7% of patients, and 76.5% presented with distant metastases with 54% of metastatic disease diagnoses involving a single organ, most commonly bone. Recurrence and death after surgically resected chRCC is rare. For completely excised lesions < pT2a without coagulative necrosis or sarcomatoid features, the prognosis is excellent [33].

3.5. Other renal tumours

Other renal tumours constitute the remaining renal cortical tumours. These include a variety of uncommon, sporadic, and familial carcinomas/tumours, some only recently described, as well as a group of unclassified carcinomas. A summary of these tumours is provided in Table 3.1, but some clinically relevant tumours and extremely rare entities are mentioned below.

3.5.1. Renal medullary carcinoma (SMARCB1-deficient renal medullary carcinoma)

Renal medullary carcinoma (RMC) (referred to as SMARCB1-deficient renal medullary carcinoma in the 2022 WHO Classification) is a very rare tumour, comprising < 0.5% of all RCCs [34], predominantly diagnosed in young adults of African ancestry (median age 28 years) with sickle haemoglobinopathies (including sickle cell trait). It has a male predominance of 2:1. It is mainly centrally located with ill-defined borders. Renal medullary carcinoma is one of the most aggressive RCCs [35,36] and most patients (~67%) will present with metastatic disease [35,37]. Even patients who present with seemingly localised disease may develop unequivocal metastases shortly (within weeks) after diagnosis (for treatment see Chapter 7). Apart from the RMC described above, some patients present with identical tumours without haemoglobinopathy. Such tumours have been described as “unclassified RCC with medullary phenotype” [23].

3.5.2. Carcinoma associated with end-stage renal disease; acquired cystic disease-associated RCC

Cystic degenerative changes (acquired cystic kidney disease [ACKD]) and a higher incidence of RCC, are typical features of ESRD. Renal cell carcinomas of native end-stage kidneys are found in approximately 4% of patients with ESRD. Their lifetime risk of developing RCCs is at least ten times higher than in the general population. Compared with sporadic RCCs, RCCs associated with ESRD are generally multicentric and bilateral, found in younger patients (mostly male), and are less aggressive. Whether the relatively indolent outcome of tumours in ESRD is due to the mode of diagnosis or a specific ACKD-related molecular pathway still has to be determined. Although the histological spectrum of ESRD tumours is like that of sporadic RCC; pRCC occur relatively more frequently [27,38]. A specific subtype of RCC occurring only in end-stage kidneys has been described as “acquired cystic disease-associated RCC” (ACD-RCC). Tumours present exclusively in patients with ACKD, usually after long-term dialysis. The vast majority occur in men. Tumours are often multiple and bilateral and, in most cases, have an indolent clinical behaviour; although, aggressive courses have been documented [23].

3.5.3. Papillary adenoma

These tumours have a papillary or tubular architecture of low nuclear grade and may be up to 15 mm in diameter, or smaller, according to the 2022 WHO classification [23].

3.5.4. Renal oncocytoma

Oncocytoma is a benign tumour representing 3–7% of all solid renal tumours and its incidence increases to 18% when tumours < 4 cm are considered [10,39]. The diagnostic accuracy of imaging modalities (CT, magnetic resonance imaging [MRI]) in renal oncocytoma is limited and histopathology remains the only reliable diagnostic modality [10,39]. However, the new imaging technology ^99mTc-sestamibi (SestaMIBI, MIBI) SPECT/CT has shown promising initial results for the differentiation between benign and low-grade RCC [40]. Standard treatment for renal oncocytoma is similar to that of other renal tumours; surgical excision by partial- or radical nephrectomy (RN) with subsequent histopathological verification. However, due to the inability of modern imaging techniques to differentiate benign from malignant renal masses, there is a renewed interest in renal mass biopsy (RMB) prior to surgical intervention. Accuracy of the biopsy and management of advanced/progressing oncocytomas need to be considered in this context since oncocytic renal neoplasms diagnosed by RMB at histological examination after surgery showed oncocytoma in only 64.6% of cases. The remainder of the tumours were mainly chRCC (18.7% including 6.3% hybrid oncocytic/chromophobe tumours which have now been grouped histologically with chRCC) [23,24], other RCCs (12.5%), and other benign lesions (4.2%) [41]. The 2022 WHO classification strictly excludes that a definitive diagnosis of oncocytoma be done on a needle-core biopsy. The majority of oncocytomas slowly progress in size with an annual growth rate < 14 mm [42-44]. Preliminary data show that active surveillance (AS) may be a safe option to manage oncocytoma in appropriately selected patients. Potential triggers to change management of patients on AS are not well defined [45,46].

3.5.5. Other oncocytic tumours of the kidney

Other oncocytic tumours of the kidney are a heterogeneous group of oncocytic tumours not classifiable as oncocytoma, chRCC, or other tumour types with eosinophilic features. These tumours are typically indolent, so it is important to distinguish such low-grade tumours from the high-grade unclassified RCCs that typically behave aggressively. In the setting of Birt–Hogg–Dubé syndrome (see Section 3.5.6), tumours with such intermediate features (hybrid oncocytic tumours) also exist, typically being multifocal and bilateral. As this is a heterogeneous tumour group, it is likely that new subtypes of renal neoplasia will emerge. There are already two emerging entities: eosinophilic vacuolated tumour (EVT) and low-grade oncocytic tumour (LOT) [23].

3.5.6. Hereditary kidney tumours

Five to eight percent of RCCs are hereditary; to date there are ten hereditary RCC syndromes associated with specific germline mutations, RCC histology, and comorbidities. Hereditary RCC syndromes are often suggested by family history, age of onset and presence of other lesions typical for the respective syndromes. Median age for hereditary RCC is 37 years; 70% of hereditary RCC tumours are found in the lowest decile (age 46 years or younger) of all RCC tumours [47]. Hereditary kidney tumours are found in the following entities: VHL syndrome; hereditary pRCC; Birt-Hogg-Dubé syndrome; Fumarate hydratase-deficient RCC (FHD-RCC), previously called hereditary leiomyomatosis and RCC (HLRCC); tuberous sclerosis; germline succinate dehydrogenase (SDH) mutation; non-polyposis colorectal cancer syndrome; hyperparathyroidism-jaw tumour syndrome; phosphatase and tensin homolog (PTEN) hamartoma syndrome (PHTS); constitutional chromosome 3 translocation; familial non-syndromic ccRCC and BAP1-associated RCC [48]. Renal medullary carcinoma can be included because of its association with hereditary haemoglobinopathies [49-52].

Patients with hereditary kidney cancer syndromes may require repeated surgical intervention [53,54]. In most hereditary RCCs nephron-sparing approaches are recommended. The exceptions are FHD-RCC and SDH syndromes for which immediate surgical intervention is recommended due to the aggressive nature of these tumours. For other hereditary syndromes such as VHL, surveillance is recommended until the largest tumour reaches 3 cm in diameter; this to limit the number of repeat interventions [55,56]. Active surveillance for VHL, SDH and FHD-RCC should, in individual patients, follow the size, growth rate and location of the tumours, rather than applying a standardised follow-up interval. Regular screening for both renal and extra-renal lesions should follow international guidelines for these syndromes [56]. Multidisciplinary and co-ordinated care should be offered, where appropriate [57]. In FHD-RCC, renal screening in relatives has shown benefit in detecting early-stage RCCs [58], with HLRCC RCCs appearing to have unique molecular profiles.

Although not hereditary, somatic fusion translocations of TFE3 and TFEB may affect 15% of patients with RCC younger than 45 years and 20–45% of children and young adults diagnosed with RCC [59].

A recent phase II trial demonstrated clinical activity of an oral HIF-2α (hypoxia-inducible factor) inhibitor MK-6482 (belzutifan) in VHL patients [60]. Additional information on treatment of VHL can be found in Section 7.4.4.5.1.

3.5.7. Classical angiomyolipoma

Classical angiomyolipoma (AML)/PEComa of the kidney is a benign mesenchymal tumour, which can occur sporadically or as part of tuberous sclerosis complex [61]. Overall prevalence is 0.44%, with 0.6% in female and 0.3% in male populations. Only 5% of these patients present with multiple AMLs [62]. Angiomyolipoma belongs to a family of so-called PEComas (perivascular epithelioid cell tumours), characterised by the proliferation of perivascular epithelioid cells. Some PEComas can behave aggressively and even metastasize, while classic AMLs are completely benign [10,49,63]. Ultrasound, CT, and MRI often lead to the diagnosis of AMLs due to the presence of adipose tissue; however, in fat-poor AML, diagnostic imaging cannot reliably identify these lesions. Percutaneous biopsy is rarely useful. Renal tumours that cannot be clearly identified as benign during the initial diagnostic work-up should be treated according to the recommendations provided for the treatment of RCC. In tuberous sclerosis, AML can be found in enlarged lymph nodes (LNs), which does not represent metastatic spread but a multicentric spread of AMLs. In rare cases, an extension of a non-malignant thrombus into the renal vein or inferior vena cava can be found, associated with an angiotrophic-type growth of AML. Epithelioid AML, a very rare variant of AML, consists of at least 80% epithelioid cells and with mean age of onset of 50 years (range 30–80 years), without gender predilection [49,63]. Epithelioid AMLs are potentially malignant with a variable proportion of cases with aggressive behaviour [64]. Criteria to predict the biological behaviour in epithelioid AML were proposed by the WHO 2022 [23,24]. Angiomyolipoma, in general, has a slow and consistent growth rate, and minimal morbidity [65]. Subtypes of AML are oncocytic AML and AML with epithelial cysts [23].

In some cases, larger AMLs can cause local pain. The main complication of AMLs is spontaneous bleeding in the retroperitoneum or into the collecting system, which can be life threatening. Bleeding is caused by spontaneous rupture of the tumour. Little is known about the risk factors for bleeding, but it is believed to increase with tumour size and may be related to the angiogenic component of the tumour that includes irregular blood vessels [65]. The major risk factors for bleeding are tumour size, grade of the angiogenic component, and the presence of tuberous sclerosis.

3.5.7.1. Treatment of angiomyolipoma

Active surveillance is the most appropriate option for most AMLs (48%). In a group of patients on AS, only 11% of AMLs showed growth, and spontaneous bleeding was reported in 2%, resulting in active treatment in 5% of patients [65,66] (LE: 3). The association between AML size and the risk of bleeding remains unclear and the traditionally used 4-cm cut-off should not per se trigger active treatment [65]. When surgery is indicated, NSS is the preferred option, if technically feasible. Main disadvantages of less invasive selective arterial embolisation (SAE) are more recurrences and a need for secondary treatment (0.85% for surgery vs. 31% for SAE). For thermal ablation only limited data are available, and this option is used less frequently [65].

Active treatment (SAE, surgery or ablation) should be instigated in case of persistent pain, ruptured AML (acute or repeated bleeding) or in case of very large AMLs. Specific patient circumstances may influence the choice to offer active treatment; such as patients at high risk of abdominal trauma, females of childbearing age or patients in whom follow-up or access to emergency care may be inadequate. Selective arterial embolisation is an option in case of life-threatening AML bleeding.

In patients diagnosed with tuberous sclerosis, size reduction of often bilateral AMLs can be induced by inhibiting the mTOR pathway using everolimus, as demonstrated in RCTs [67,68]. In a small phase II trial
(n = 20), efficacy of everolimus was demonstrated in sporadic AML as well. A 25% or greater reduction in tumour volume at four and six months was demonstrated in 55.6% and 71.4% of patients, respectively. However, 20% of patients were withdrawn due to toxicities and 40% self-withdrew from the study due to side effects [69].

Table 3.2: Other renal cortical tumours and recommendations for treatment (strength rating: weak)

3.5.8. Cystic renal tumours

Cystic renal lesions are classified according to the Bosniak classification (see Section 5.2.5). Bosniak I and II cysts are benign lesions which do not require follow-up [73]. Bosniak IV cysts are mostly (83%) malignant tumours with pseudo-cystic changes only [74]. Bosniak IIF and III cysts remain challenging for clinicians. The differentiation of benign and malignant tumour in categories IIF/III is based on imaging, mostly CT, with an increasing role of MRI and contrast-enhanced US (CEUS). Computed tomography shows poor sensitivity (36%) and specificity (76%; κ [kappa coefficient] = 0.11) compared with 71% sensitivity and 91% specificity (κ = 0.64) for MRI and 100% sensitivity and 97% specificity for CEUS (κ = 0.95) [75]. Surgical and radiological cohorts pooled estimates show a prevalence of malignancy of 0.51 (0.44–0.58) in Bosniak III and 0.89 (0.83–0.92) in Bosniak IV cysts, respectively. In a systematic review, less than 1% of stable Bosniak IIF cysts showed malignancy during follow-up. Twelve percent of Bosniak IIF cysts had to be reclassified to Bosniak III/IV during radiological follow-up, with 85% of these showing malignancy, which is comparable to the malignancy rates of Bosniak IV cysts [73]. The updated Bosniak classification strengthens the classification and includes also
MRI [76] and even CEUS diagnostic criteria [77].

The most common histological types for Bosniak III cysts is ccRCC with pseudo-cystic changes and low malignant potential [78,79]; multilocular cystic renal neoplasm of low malignant potential [MCRNLMP], see Section 3.4.1.1; classical pattern pRCC (very low malignant potential); benign multilocular cyst; benign group of mixed epithelial and stromal renal tumour (mixed epithelial and stromal tumour of the kidney and adult cystic nephroma); and other rare entities. Surgery in Bosniak III cysts will result in over-treatment in 49% of the tumours which are lesions with a low malignant potential. In view of the excellent outcome of these patients in general, a surveillance approach is an alternative to surgical treatment [73,76,80,81].

3.6. Summary of evidence and recommendations for the management of other renal tumours