Prognostic factors in patients with renal cell carcinoma

INTRODUCTION — Renal cell carcinomas (RCCs), which originate within the renal cortex, constitute 80 to 85 percent of primary renal neoplasms. (See "Epidemiology, pathology, and pathogenesis of renal cell carcinoma".)

Surgical resection of localized RCC can be curative, but up to one-third of patients eventually recur. In addition, approximately 15 percent of patients with RCC present with locally advanced or metastatic RCC, for which surgery is noncurative. The natural history of disease for patients with advanced or metastatic RCC can vary widely from a few months to many years depending on the clinical, pathologic, laboratory, and radiographic features of the disease. (See "Systemic therapy of advanced clear cell renal carcinoma".)

The factors affecting prognosis in patients with RCC will be reviewed here. An overview of the approach to treatment is presented separately. (See "Overview of the treatment of renal cell carcinoma".)

ANATOMIC EXTENT OF DISEASE — The 2017 (eighth edition) tumor, node, metastasis (TNM) staging system is used to assess the anatomic extent of disease and define prognostic stage groups (table 1) [1]. The anatomic extent of disease is the most consistent factor that influences prognosis in patients with renal cell carcinoma (RCC (figure 1)). (See "Clinical manifestations, evaluation, and staging of renal cell carcinoma", section on 'TNM staging system'.)

Stage I/II — Patients with stage I RCC have a five-year survival rate over 90 percent in most contemporary series. The survival rate may be slightly lower for patients with stage II disease, with reported five-year survival rates ranging from 75 to 95 percent.

Patients with stage I or II RCC that invades the urinary collecting system appear to have a significantly worse prognosis [2]. In a multivariate analysis of a series of 1124 cases of RCC, the 10-year survival rates for patients with T1 or T2 primary lesions that had invaded the urinary collecting system were 43 and 41 percent, respectively (hazard ratio 3.2, 95% CI 1.4-7.1).

Stage III — The reported five-year survival rate for patients with stage III RCC who undergo nephrectomy ranges from 59 to 70 percent. There are conflicting data about whether extension into the perinephric fat (T3a) alone adversely affects prognosis. Two large studies could not demonstrate a difference when T3a primary tumors were compared with comparably sized T1 and T2 primary tumors [3,4]. By contrast, involvement of the perinephric fat remained a prognostic factor in two other series [5,6]. Among patients with T3a disease, the size of the primary tumor remains a prognostic factor (10-year survival rates of 77, 54, and 46 percent for tumors <4, 4 to 7, and >7 cm, respectively) [5].

Patients with involvement of the renal vein or inferior vena cava are included in the group with stage III RCC. Although some early studies did not identify an adverse impact of renal vein involvement on prognosis [7,8], other reports found that the extent of vena cava invasion and the anatomic location of the tumor thrombus were important prognostic factors [9,10]. The extent of venous involvement is recognized as a prognostic factor in the eighth edition of the American Joint Committee on Cancer (AJCC) TNM system, with T3a tumors having tumor invasion of the renal vein or its branches, T3b tumors having gross involvement of the inferior vena cava below the diaphragm, and T3c lesions having tumor invading the wall of the inferior vena cava or grossly extending into the inferior vena cava above the diaphragm.

In addition, invasion of the urine collecting system also appears to be a prognostic factor in patients with stage III RCC. In a series of 303 cases with stage III disease, multivariate analysis found that patients with urine collecting system invasion had significantly worse disease-specific and overall survival compared with those without invasion (five-year rates 34 versus 59 percent and 30 versus 52 percent, respectively) [11].

Stage IV — Overall survival for patients with stage IV disease has improved in the contemporary era of immunotherapy and targeted therapy. When cytokines were the predominant systemic therapies for these patients, median overall survival was initially a little over one year. In subsequent studies of patients receiving either targeted therapies (such as pazopanib or sunitinib) or checkpoint inhibitor immunotherapy (with nivolumab plus ipilimumab), median overall survival was over two years (approximately 28 months) [12-14] and almost five years (56 months), respectively [15]. (See "Systemic therapy of advanced clear cell renal carcinoma".)

HISTOPATHOLOGY

Tumor type — Whether the tumor subtype (ie, clear cell versus papillary or chromophobe carcinoma) affects prognosis is controversial. A multi-institution study failed to identify a prognostic difference in over 4000 patients when tumor, node, metastasis (TNM) stage, histologic grade, and performance status were considered in a multivariate analysis [16]. In contrast, multivariate analyses of single-institution series from the Mayo Clinic and from Memorial Sloan Kettering Cancer Center (MSKCC) including 3062 and 1668 patients, respectively, both found that patients with clear cell histology had significantly poorer cancer-specific survival [17,18].

Some less common variants, including collecting duct carcinomas, renal medullary carcinomas, and any histology with sarcomatoid or rhabdoid features, are considered more aggressive and are associated with a shorter survival [19]. (See "Epidemiology, pathology, and pathogenesis of renal cell carcinoma", section on 'Pathology'.)

Tumor grade — Histologic grade is an independent factor correlating with survival [20-25]. Multiple systems are used to grade renal cell carcinoma (RCC), of which Fuhrman grade is the most widely used [23-25]. In one report, the five-year survival rates based upon tumor grade were 89, 65, and 46 percent for tumors of histologic grade 1, 2, and 3 to 4, respectively [20].

The International Society of Urological Pathology (ISUP) Consensus Conference tumor grading system is also available, in addition to recommendations for other prognostic factors related to renal tumors [26]. In this system, ISUP grade 1 tumors were defined as having inconspicuous/absent nucleoli at ×400 magnification; for ISUP grade 2 tumors, nucleoli should be distinctly visible at ×400, though invisible at ×100 magnification; and for ISUP grade 3 tumors, nucleoli should be visible at ×100 magnification. ISUP grade 4 tumors are the ones showing extreme nuclear pleomorphism, clumping of chromatin, or sarcomatoid/rhabdoid dedifferentiation [26].

Tumor necrosis — Histologic coagulative tumor necrosis is an independent predictor of outcome for clear cell and chromophobe RCC and should be routinely reported. It is also part of several integrated staging systems, such as the Stage, Size, Grade, and Necrosis (SSIGN) score [27,28].

CLINICAL FACTORS — In addition to the anatomic extent of disease, clinical factors can influence survival. Adverse prognostic signs include a poor performance status, the presence of symptoms and/or paraneoplastic syndromes (eg, anemia, hypercalcemia, thrombocytosis, fever, weight loss), and obesity [21,29-34]. Although younger patients (ie, 20 to 40 years old) are more likely to be symptomatic at presentation, their outcome may be slightly better due to a lower incidence of nodal involvement [35].

Multiple models have been developed to integrate the information from anatomic staging with histopathology and clinical prognostic parameters [36-40]. Nevertheless, validation studies show a low predictive ability of these models, with several only marginally outperforming tumor, node, metastasis (TNM) staging [41].

A validated prognostic model has been the University of California, Los Angeles (UCLA) integrated staging system (UISS) (calculator 1) [21,36,42,43]. The UISS incorporates the Eastern Cooperative Oncology Group (ECOG) performance status (table 2) and Fuhrman histologic grade (1 through 4 [24]) into the TNM anatomic staging system. Using these variables, patients are characterized into distinct prognostic categories (low-, intermediate-, and high-risk) for disease recurrence postnephrectomy [36]. The value of this system has subsequently been validated [21,42].

MOLECULAR MARKERS — Although none of these factors has a clinical application for patient care, some markers have shown promise as prognostic markers in patients with clear cell RCC. Examples of markers that are potentially associated with a worse prognosis for patients with clear cell RCC include:

●Human B7 homolog 1 (B7H1) and 4 (B7H4) expression [44].

●Low levels of carbonic anhydrase IX (CAIX) [45].

●High levels of the proliferation marker Ki-67 [45].

●Higher levels of hypoxia-inducible factor (HIF)-1 alpha expression [46]; although at least one other study has suggested that patients with tumors that express HIF-1 alpha have a better prognosis than those with tumors that only express HIF-2 alpha [47]. (See "Molecular biology and pathogenesis of von Hippel-Lindau disease", section on 'Hypoxia-inducible factor 1 and 2'.)

●Expression of the U3 small nucleolar ribonucleoprotein (IMP3) [48-50]; which may extend to papillary and chromophobe tumors as well as clear cell RCCs [50].

●Deletion of chromosome 9p [51-54].

●Mutations of tumor suppressor genes on chromosome 3p21, including mutations of breast cancer type 1 (BRCA1)-associated protein 1 (BAP1) and SET domain containing 2 (SETD2) [55]. In contrast, mutations involving polybromo-1 (PBRM1) had a more favorable prognosis [56], although in one study, the presence of both a PBRM1 and a BAP1 mutation conferred the worst prognosis [56].

Data from The Cancer Genome Atlas (TCGA) have elucidated potential molecular prognostic signatures using discovery (n = 193) and validation (n = 253) datasets. Main results suggested that worse survival is associated with upregulation of the fatty acid synthesis genes, acetyl-CoA carboxylase alpha (ACACA) and fatty acid synthase (FASN), as well as upregulation of multiple genes involved in the pentose phosphate pathway. On the other hand, better survival was associated with upregulation of adenosine monophosphate-activated kinase (AMPK) and multiple genes involved in the Krebs cycle and the mammalian (mechanistic) target of rapamycin (mTOR) pathway [57].

These data enabled the identification of four prognostic signatures for RCC, which appear to represent the metabolic states of the tumors and their variable use of key pathways and metabolites (figure 2A-C).

A gene expression panel that included 16 genes was developed to predict the risk of recurrence in a series of 942 patients who had undergone radical nephrectomy for stage I to III clear cell RCC [58]. This information was used to develop a recurrence score, which was then validated in a series of 626 patients. On multivariate analysis, this score was independently associated with an increased risk of tumor recurrence. Although this recurrence score provides additional information regarding the risk of recurrence, additional research to develop effective adjuvant therapies will be required if this panel is to be used to influence patient management.

In addition, germline genetic polymorphisms might affect the risk of recurrence in patients with localized RCC. In one large series, patients with MET polymorphism rs11762213 had an increased risk of recurrence after nephrectomy [59], and these results were validated in an independent cohort from TCGA [60].

PROGNOSTIC FACTORS IN STAGE IV DISEASE — Although the prognosis for patients with recurrent or metastatic renal cell carcinoma (RCC) has historically been poor (figure 1), many studies performed in the era before effective therapy documented specific clinical features that were associated with longer survival; these factors remain pertinent in the targeted therapy era, reflecting the natural history of the disease [61]. One of the most commonly used prognostication systems is the one developed in the cytokines era by the Memorial Sloan Kettering Cancer Center (MSKCC) group, which integrates five adverse clinical and laboratory factors in previously untreated patients. Only three adverse factors are used in patients that were previously treated.

Contemporary trials in patients treated with agents targeting the vascular endothelial growth factor (VEGF) pathway have been analyzed to define prognostic factors relevant in the targeted therapy era [62-65]. The International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) compared baseline characteristics in 645 patients who were treated with several VEGF-targeted agents [65].

The six factors associated with poorer survival on multivariate analysis included (table 3):

●Karnofsky Performance Status (KPS) <80 (table 4)

●Time from original diagnosis to initiation of targeted therapy <1 year

●Hemoglobin less than the lower limit of normal

●Serum calcium greater than the upper limit of normal

●Neutrophil count greater than the upper limit of normal

●Platelet count greater than the upper limit of normal

These factors were subsequently validated in another cohort of 849 patients from the IMDC database [66]. This database was used to generate a similar model that can be applied to the outcome of second-line therapy following resistance to VEGF-targeted therapy [67], as well as to patients with non-clear cell RCC [68].

The relevance of the IMDC prognostic criteria in the era of frontline combination immunotherapy remains to be established. In the absence of alternative immunotherapy-based prognostic criteria, these criteria continue to be used in clinical trials to risk-stratify patients and, to some extent, by providers and clinical guidelines to direct therapy. Further details on such trials are discussed separately. (See "Systemic therapy of advanced clear cell renal carcinoma", section on 'Risk stratification'.)

Unlike other models, the IMDC model utilizes the same baseline prognostic factors for previously treated and untreated patients. The IMDC model was shown to improve prognostication compared with other prognostic models, such as the Cleveland Clinic Foundation (CCF) model [69], the International Kidney Cancer Working Group (IKCWG) model [70], the French model [71], and the MSKCC model [72].

SUMMARY

●Prognosis for localized renal cell carcinoma – For patients presenting with localized renal cell carcinoma (RCC), the tumor, node, metastasis (TNM) staging system provides the primary prognostic information (figure 1 and table 1). (See 'Anatomic extent of disease' above.)

●Additional prognostic factors – Additional prognostic information can be provided by incorporating parameters such as performance status and histologic grade. These have been integrated with anatomic stage in the University of California, Los Angeles (UCLA) Integrated Staging System (UISS) (calculator 1). Ongoing research studies indicate that molecular markers may also be useful, though not commonly used in clinical practice. (See 'Clinical factors' above and 'Molecular markers' above.)

●Prognostic factors for metastatic disease – For patients with metastatic disease, multiple clinical parameters have been identified that are associated with prognosis. These are integrated into the International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) model (table 3), which has been useful in predicting outcomes in the contemporary era of therapy. (See 'Prognostic factors in stage IV disease' above.)