Renal cell carcinoma (RCC) is a group of neoplastic diseases that affect the renal parenchyma and represent 2–3% of all cancer diagnoses (1). In 2022, the GLOBOCAN database estimated the global incidence of RCC as 434,419 cases, which were associated with 155,702 deaths (2). RCC is more commonly diagnosed in men (ratio men to women, 2:1), and in the sixth decade of life (3). Up to 70% of patients are incidentally diagnosed with RCC, 50% of them with metastatic disease (4); although localized RCC is curable, in up to 30% of patients this will progress to metastatic disease, which has a median survival time of 6–10 months (5). Among the different types of RCCs, 80% of all diagnoses consist of the clear cell histological variety (ccRCC), which is followed in frequency by the papillary variant (10–15% of cases), the chromophobe cell variant (4–5% of cases), and other molecularly defined phenotypes (<1% of cases) (6).

The high rates of mortality that are observed in RCC have been associated with factors such as the high prevalence of advanced disease at diagnosis (7) and the intrinsic chemoresistance of RCC tumors, which is mostly related to apoptotic resistance, the upregulation of xenobiotic excretion systems (8–10) and the metabolic dependence of such tumors on the Warburg effect (11). In total, <10% of patients with RCC will respond to cytotoxic chemotherapy (12), which makes it necessary to use novel approaches for which little clinical information is available. Current RCC treatments can be grouped into the following categories: cytotoxic chemotherapy, surgery, local therapy, recombinant-cytokine therapy, targeted anti-angiogenic therapy, immunotherapy [immune checkpoint blockade (ICB)], and other therapies that include the use of inhibitors of the mechanistic target of rapamycin pathway (Table I). Despite the broad diversity and availability of RCC treatments, the need for a prognostic tool for use in selecting the treatment of choice and evaluating the clinical responses of patients remains a matter of controversy (13). Treatment selection depends on factors such as the histological variety, cellular grade and clinical stage of the disease, as well as the failure of previous treatments (14–16). In this regard, there are currently several scales that have been validated for risk classification and clinical decision-making in patients with RCC; among these, the Heng score (International Metastatic RCC Database Consortium or IMDC), the MSKCC scale (Memorial Sloan-Kettering Cancer Center), and the general physical status according to ECOG-PS criteria (the Eastern Cooperative Oncology Group performance score) (17–19) stand out (Table II).

In the last decade, the use of monoclonal antibodies for reversing the ‘exhaustion’ state in tumor infiltrating lymphocytes (TILs) has proven to be a valuable treatment for numerous solid tumors and hematologic cancers. In 2015, the ChekMate-025 study demonstrated the efficacy of nivolumab (which was initially approved for treating melanoma and non-small cell lung cancer) in treating patients with RCC (20). In addition to nivolumab [which blocks the programmed cell death protein 1 (PD-1) receptor], four other monoclonal antibodies are currently approved for use in RCC, all of which reverse the inhibition of lymphocyte immune effectors: Pembrolizumab, which is also a PD-1 blocker; ipilimumab, which blocks the cytotoxic T-lymphocyte associated protein 4; and atezolizumab/avelumab, which blocks PD-1 ligand in myeloid and tumor cells and prevents PD-1 inhibitory signals in lymphocytes (Table III) (20–30).

In the following section, the current knowledge of the cellular mechanisms that lead to lymphocyte exhaustion within the highly dynamic tumor-immune microenvironment (TIME) was reviewed. It was concluded with a short description of experimental biomarkers that have been used to predict the patients' responses to ICB therapy.

The TIME comprises a network of interacting elements within the tumor tissue that can be categorized as follows: Cells (tumor, stroma and infiltrating immune cells), small soluble elements (proteins, cytokines, growth factors, metabolites and chemokines), and the extracellular matrix (31). The growth of tumors requires two conditions: Cell transformation, whether genetic or acquired (32) and immune dysfunction (33). This second requirement explains the association between immunodeficiency status and cancer development (34). The clearance of transformed cells from tissues requires the activation of cytotoxic immunity that is achieved through the infiltration of CD8⁺ cytotoxic lymphocytes and natural killer cells into a tumor (35). Antitumor immunity is so efficient that even though transformation and cell damage occur over the course of every life, cancer develops only in a small proportion of patients; the occurrence of cancer is promoted by immune dysfunction.

Tumor cells proliferate at higher rates compared with normal cells. This accelerated proliferation, which is linked to metabolic changes that take place within the tumor bed, results in dysfunctional local immunity and further selection of the best-adapted tumor cells (36–38). Cytotoxic chemotherapy exploits this increased proliferative rate, and cytotoxic treatments are widely used in most neoplastic diseases other than RCC. The peculiarities of RCC include its intrinsic aggressive nature, its increased infiltration with lipids, its high rate of metastasis, and its resistance to cytotoxic chemotherapy (39). This chemo-resistance has numerous causal factors, both intrinsic (or genetic) and acquired. It is important to note that the epithelial renal cells normally have secretion systems for xenobiotics and intrinsic antioxidant mechanisms protecting the cells from toxic damage (40). Given that most RCC subtypes arise from renal epithelia, it is not surprising that such mechanisms are enhanced in renal carcinoma. In addition, renal tumor cells have increased expression of anti-apoptotic proteins Bcl-2, Bcl-_XL, ARC (apoptosis repressor with a caspase recruitment domain) and XIAP (X-linked inhibitor of apoptosis) while pro-apoptotic proteins such as Bim are decreased. Anti-apoptotic and pro-apoptotic proteins are regulated by NF-κB and von Hippel-Lindau (VHL) pathways, respectively (8,10,41).

A total of up to 2/3 of patients with RCC have mutations in VHL gene (42). Among its numerous functions, VHL targets proteins for proteasomal degradation, including the pro-apoptotic protein Bim, whose levels are increased in patients with RCC (41). Conversely, the increase in anti-apoptotic pathways has also been implicated in chemoresistance, specifically in the increased expression of anti-apoptotic proteins Bcl-2, ARC and XIAP (9). Concomitant to these mechanisms, renal cancer cells are highly dependent on aerobic glycolysis (Warburg effect), a phenotype associated to a lack of response after ICB therapy (43), and a recent putative target for enhancing combined treatments for RCC (11).

In addition, although the presence of TILs is a favorable prognostic factor for most cancers, such infiltrative cells are associated with poor outcomes in RCC (44,45). Consequently, a hostile TIME develops that leads to the selection of tumors that are resistant to hypoxia, acidosis and the low availability of nutrients. In addition, a hostile TIME negatively affects lymphocyte-dependent antitumor immunity (Fig. 1 and Table IV) (31,46–51). Previous studies suggested that dysbiosis is an important element to be considered when evaluating either the responses of tumors to ICB or the general prognosis of patients with neoplastic diseases (52,53). The molecular basis for lymphocyte dysfunction involves several mechanisms that are associated with peripheral tolerance, namely anergy, suppression and exhaustion (54). Immune cell exhaustion is reversed by ICB, which blocks interactions between the inhibitory receptors of lymphocytes and their ligands (55). The PD-1/programmed death-ligand 1 (PD-L1) axis is the most studied of these interactions within TILs (56,57).

Even though numerous patients achieve complete clinical response to ICB therapy, a significant proportion of patients show only a partial response or no response at all, which is occasionally accompanied by signs of systemic toxicity (58). This situation explains why the search for prognostic markers for ICB in patients with RCC is a highly active area of research. Most prognostic markers for RCC can be divided into those addressing disease progression and those addressing responses to treatments, in particular immunotherapy and directed therapies. Thus, predicting the ICB responsiveness will impact in selecting the patients who will benefit the most and have the lowest probability of developing adverse events after receiving immunotherapy. There are currently no standardized and validated methods for properly assessing the risk/benefit ratio of using ICB therapy. Therefore, the goal of the present review was to synthesize the reported findings of the studies that have focused on this important topic.

ICB therapy is associated with an average mortality rate of 1%, with death mostly caused by immune-related adverse events (58). After the reporting of successful results from the CheckMate-025 study, and given the low predictability and consistency of ICB responsiveness (59), interest has been increasing in the search for reliable markers of ICB responsiveness. Among the biomarkers that have been studied, three stand out for their consistency: the level of C-reactive protein (CRP), the number of TILs and the basal expression levels of exhaustion markers in tumor tissue. CRP is an acute phase reactant that is frequently used for evaluating systemic inflammation, making it a logical putative marker for ICB responsiveness given the close association between inflammation and cancer. Previous studies have indicated that low basal CRP levels or low normalized levels of CRP measured after patients received their first ICB doses were predictive of their ICB responsiveness (60–63). Studies investigating TILs found that a patient's prognosis is associated with the nature of his or her infiltrating cells; infiltration with inflammatory leukocytes was associated with the best responses to ICB therapy (64–66). In peripheral blood, a recent study observed that increased numbers of circulating eosinophils were associated with an improved response to ICB, which is an intriguing result given the regulatory role that eosinophils play in systemic inflammation (67). Interestingly, the basal expression level of PD-L1 has not been consistently predictive of ICB responsiveness (68–70), whereas the basal expression level of T cell immunoglobulin and mucin-domain containing-3 (TIM-3) appears to be (71).

Several additional markers have been studied for predicting the response to ICB in patients with RCC. For example, longer responses to ICB have been associated with decreased levels of circulating tumor DNA, increased levels of chemokine CXCL14, increased levels of circulating miR-22 and miR-24, and the presence of immunogenic transcriptional signatures (72–75). In this context, several markers have been associated with poor responses after ICB therapy (increased levels of interleukin-8) or predictive of the development of immune-related adverse events (decreased levels of miR-146a) (76,77). All the studies described in this section are summarized in Table V.

From a technical perspective, the evaluation of leukocytes' exhaustion markers require the performance of histopathology after the direct sampling of tissue, which is not feasible for all diagnoses (78). In this context, the studies assessing the suitability of peripheral biomarkers for predicting ICB responsiveness are encouraging (79,80).

Several limitations currently prevent the formal use of markers to predict ICB responsiveness. First, the inconsistency of reported outcomes is mostly related to heterogeneity in the target population and the low number of patients that have been studied. In addition, a lack of standardization in the use of methods and reagents is complicating the replication of pioneering studies. Furthermore, the relationship between the expression of such potential biomarkers and the underlying mechanisms of resistance to ICB is not fully understood, making it difficult not only to predict ICB responsiveness but also to know the clinical safety of using ICB. Finally, the intrinsic resistance of RCCs to chemotherapy and the complex combinations required for treatment make it difficult to determine a logical approach for studying the expression of putative cancer biomarkers. In the future, by integrating multi-omics data and machine-learning approaches, it should be possible to create more accurate prediction models that will help in the identification of novel biomarkers for ICB responsiveness.