Renal cell carcinoma (RCC) accounts for around 2% of all adult malignancies [1]. It comprises different histological subtypes with different molecular characteristics, biological behavior, and response to systemic therapies [2]. Clear cell RCC (ccRCC) is the most frequent histologic subtype representing around 70–80% of all RCC [3]. At diagnosis almost 30% of cases are metastatic and around 30% will recur after surgery, and will probably require systemic therapy [4, 5].

The treatment landscape of advanced ccRCC has significantly changed during the last decade with the incorporation of immune checkpoint inhibitors (ICI) to the treatment armamentarium of advanced ccRCC. Combinations of ICI with either another ICI or an anti-angiogenic (AA), have significantly improved survival outcomes, becoming the new standard of care in first-line setting [6–14].

Despite this, there is still a notable proportion of patients that will be refractory to ICI while another subset will eventually develop resistance to ICI. Additionally, ICI are associated with non-negligible toxicities. In this context, research efforts are directed towards identifying reliable biomarkers that can help determine the best treatment strategy for our patients (i.e., ICI-ICI, ICI-AA, or even ICI or AA monotherapy).

To date, tissue-based approaches have failed to identify reliable biomarkers that are predictive of response to ICI [15]. Immunohistochemical biomarkers, such as PD-L1 or tumor infiltrating cytotoxic T cells, have been extensively studied but have failed to demonstrate a predictive role in ccRCC. Gene expression signatures developed within ICI-AA combination trials have not been validated when assessed in patients treated with dual ICI in the Checkmate 214 trial [16–18].

Circulating biomarkers emerge as an attractive alternative to tissue-based biomarkers. Blood-based biomarkers allow repeated evaluations at different time points, are easily accessible and less invasive for patients, and may be able to overcome the heterogeneity associated with biopsies. In addition, they may provide more information about the host’s immune status and could be useful in determining a patient’s potential for developing effective tumor immunity.

This review provides an overview of the different blood-based candidate biomarkers currently under study for predicting response to ICI in advanced RCC (aRCC).

The effectiveness of ICI therapy is intricately linked to the host’s capacity to initiate an anti-tumor immune response. Consequently, the immune profile of the host theoretically impacts the efficacy of ICI. Effector immune cells, including cytotoxic T cells and natural killer (NK) cells, are essential for the effectiveness of ICIs, as they directly attack tumor cells when checkpoint pathways are inhibited. Additionally, immune cells express checkpoint molecules such as PD-1 and CTLA-4, and the levels and activation status of these molecules can impact their function and response to ICIs [19, 20]. Regulatory T cells (Tregs) can suppress immune responses and limit the effectiveness of ICIs by inhibiting effector T cells, with high levels of Tregs in tumors counteracting checkpoint blockade effects [21]. Antigen-presenting cells (APCs) like dendritic cells and macrophages are crucial for presenting antigens to T cells and initiating immune responses, thus influencing T cell activation and the overall immune response to ICIs [22, 23]. Furthermore, cytokines produced by immune cells can modulate the immune response; for instance, interferon gamma (IFN-γ) from activated T cells can enhance PD-L1 expression in tumor cells, affecting the efficacy of ICI [24]. Tumor-associated macrophages (TAMs) can exhibit pro-inflammatory (M1-like) or anti-inflammatory (M2-like) phenotypes, with the balance between these phenotypes affecting the tumor microenvironment (TME) and the response to ICIs [25]. Additionally, interactions between immune cells and endothelial or stromal cells in the TME can influence tumor vascularization and the immune milieu, further affecting ICI efficacy [26].

Different research groups have investigated how peripheral immune cell populations influence the response to ICI therapy across different solid tumors, including ccRCC. Infiltrating cytotoxic T cells, as primary protagonists and targets of ICI therapy, have been extensively studied; however, the results in aRCC patients remains controversial [16]. Increasing evidence highlights the role of other immune cells in the antitumor response, such as B lymphocytes [27]. Following the encouraging results of different studies supporting the role of B cells within tertiary lymphoid structures (TLS) in fostering antitumor responses across different solid tumors treated with ICI, fresh blood immune-monitoring of advanced ccRCC patients receiving nivolumab within the NIVOREN study revealed that pre-existing high levels of unswitched memory B cells (CD19⁺CD27⁺IgD⁺IgM⁺) were associated with improved clinical outcomes (n = 44) [overall survival (OS) HR = 0.08, P = 0.002 and progression-free survival (PFS) HR = 0.54, P = 0.048] [28] (Table 1). This B cell subset has the ability to reinitiate B cells response but also initiate a germ center reaction upon repeated antigenic stimulation [29, 30]. Interestingly, unswitched memory B cells also correlated with circulating T follicular helper (Tfh) cells, which are known to enhance B cell maturation, stimulate the expansion of T CD8⁺ lymphocytes, and with the presence of TLS. These results contrast with those reported by a pan-tumor study (n = 78), which included RCC patients, showing that high pretreatment levels of circulating B cells were negatively associated with response to ICI (P < 0.001) [31] (Table 1). The results of this study also suggest that an increased B cells frequency could identify patients at risk of progression after an initial response to ICI [31]. Finally, a small pan-tumor study, including 7 RCC tumors among the 45 different solid tumors analyzed, found that patients with increased frequency of naive B cells were more likely to benefit from ICI [disease control rate (DCR) odds ratio (OR) = 12.31, P = 0.039], while those with increased frequency of switched memory B cells were associated with resistance to ICI (DCR OR = 0.06, P = 0.025) [32] (Figure 1).

To understand the importance of circulating PD-L1⁺ T cells, a study investigated their relative changes at the third cycle of treatment (atezolizumab plus bevacizumab) compared to baseline levels prior to therapy [33] (Table 1). During treatment, all patients experienced a decrease in the percentage of CD8⁺PD-L1⁺ T cells. Patients with therapeutic resistance experienced a greater decrease in the percentage of CD8⁺PD-L1⁺ T cells compared to those who responded to therapy. This greater decrease in CD8⁺PD-L1⁺ T cells seemed to correlate with resistance. However, a similar trend was not clearly observed with the relative change in CD4⁺PD-L1⁺ T cells on-treatment. A larger decrease of PD-L1⁺ T cells on-treatment was associated with worse PFS and OS for both CD4⁺PD-L1⁺ T cells and CD8⁺PD-L1⁺ T cells.

Another small study, including 45 patients with RCC, evaluated whether the baseline diversity of the T-cell receptor β-chain (TCRB) was associated with prognosis and how the baseline and dynamic RCC tumor burdens affected the T‐cell repertoire [34]. Higher TCRB diversity was shown to be connected with an increased lymphocyte‐to‐neutrophil ratio, especially indicating elevated naive T cells. In addition, high baseline TCRB diversity in stage IV patients was associated with improved OS (HR = 0.195, P = 0.037).

In summary, although evidence on the value of circulating immune cell populations is limited, recent findings from small cohort studies provide promising results that warrant further investigation in larger prospective studies.

Blood-based inflammatory markers could, in theory, indicate the host’s proinflammatory status and systemic immune response to cancer-related inflammation. Systemic inflammation in aRCC is associated with a poor prognosis [35, 36]. Different inflammatory routine blood parameters have been associated with worse prognosis and resistance to ICI in a variety of solid tumors, including RCC. Some of these parameters (neutrophil count, platelets, or LDH) are well-known prognostic factors in RCC and are included in established prognostic scores used in daily clinical practice, such as the IMDC (International mRCC Database Consortium) or MSKCC (Memorial Sloan Kettering Cancer Center) scores [35, 36].

The neutrophil-to-lymphocyte ratio (NLR) is one of the most studied inflammatory markers in cancer. Reflecting the balance between systemic inflammation and adaptive immune response, it has been associated with poor prognosis across different solid tumors, including RCC. This marker has been extensively studied in aRCC, with a median NLR cut-off varying between 2.5 and 5 [37, 38] (Table 2). The most robust data on the prognostic role of NLR in RCC come from a systematic review and meta-analysis by Shao et al. [38] including 6,461 RCC patients. High NLR was associated with a poor prognosis in both localized and aRCC. In addition, elevated baseline NLR in ICI-treated aRCC patients was also associated with worse OS [38]. Recent data from the JAVELIN Renal 101 trial showed a significant association between high baseline levels of NLR (cut-off: median NLR = 2.8) and inferior clinical outcomes in both the avelumab plus axitinib (low vs. high NLR OS HR, 0.51; 95% CI, 0.300–0.871) and the sunitinib arms (low vs. high NLR OS HR, 0.30; 95% CI, 0.174–0.511), continuing to support the prognostic role of NLR in patients with aRCC [39]. Moreover, Rebuzzi et al. [40, 41] developed an improved version of the IMDC score, the “MeetURO score” by including the NLR index and the presence of bone metastases, which enabled them to split RCC patients treated with ICI into five prognostic groups. Additionally, on-therapy NLR variation has also been associated with clinical outcomes to ICI. In a retrospective study including advanced non-small cell lung cancer (NSCLC) (n = 75) and RCC (n = 86) patients, any NLR increase at week 6 was associated with worse outcomes, compared to NLR decrease [42]. Similarly, Young et al. [43] found that NLR ≥ 3 after 12 weeks of ICI-based first-line therapy was associated with worse outcomes (17.5 months vs. 40.3 months, P < 0.001), and normalization of NLR in patients with baseline elevation was associated with superior OS (40.3 months vs. 14.7 months, P = 0.004) [43]. Monitoring NLR over time could help guide treatment intensification or de-escalation strategies; however, validation of these findings in prospective studies is still needed.

The lung immune prognostic index (LIPI), defined by pretreatment levels of derived NLR (dNLR) and LDH, was initially developed and assessed in lung cancer patients treated with ICI [44–46]. Subsequent studies have confirmed its association with clinical outcomes to ICI across other tumor types [44–50]. Until recently, the only evidence of the role of LIPI in aRCC relied on a multi-tumor retrospective study including NSCLC, melanoma and RCC (145, 25%) patients treated with ICI [47] (Table 2). In this study, LIPI stratification was associated with OS in aRCC patients treated with ICI (P < 0.005) [47]. The most robust data to date on the value of LIPI in aRCC come from a recent study evaluating its impact in three different prospective studies [NIVOREN study: nivolumab cohort; TORAVA trial: vascular-endothelial growth factor (VEGF)/VEGFR targeted therapy; and Checkmate 214 trial: nivolumab plus ipilimumab (nivo-ipi) vs. sunitinib] [48]. Initial results showed an association of LIPI stratification with worse outcomes in aRCC treated with nivolumab (LIPI-good 30.1 vs. 13.8 months in the LIPI intermediate/poor; HR, 0.47), but no associations with clinical outcomes were found in those treated with VEGF/VEGFR therapy, suggesting a potential predictive role for LIPI. However, this was not confirmed in the Checkmate 214 trial, where LIPI stratification was associated with worse survival outcomes irrespective of therapy type, whether nivolumab plus ipilimumab or sunitinib (nivo-ipi: LIPI good vs. intermediate/poor: HR, 0.55; P < 0.001; sunitinib: LIPI good vs. int/poor: HR, 0.38; P < 0.001) [51] (Figure 1). Thus, in contrast to NSCLC, LIPI appears to have more of a prognostic rather than a predictive value for response in aRCC.

Although their role in the immune antitumor response is less known, pre-clinical studies suggest tumor-associated eosinophilia may enhance antitumor response by promoting CD8⁺ T cell infiltration [52]. In the study by Simon et al. [53], higher levels of circulating eosinophils were associated with improved response to ICI, and eosinophils from ICI-treated patients were enriched for IFN-γ response signatures, which are known to be associated with benefit from ICI therapy. Based on these data, Zahoor et al. [54] conducted a retrospective study (n = 90) of aRCC patients treated with nivolumab. They found that patients with higher baseline eosinophil counts were associated with a lower risk of progression (HR, 0.54; P = 0.042). In contrast, the retrospective study (n = 65) conducted by Herrman et al. [55] failed to show a significant association between baseline circulating eosinophil counts and outcomes to ICI but found that patients experiencing an increase in eosinophils at six weeks of treatment were associated with an improved response to ICI. In the same line, Tucker et al. [56] showed that aRCC patients receiving nivolumab plus ipilimumab with lower baseline levels of neutrophil-to-eosinophil ratio (NER) presented improved clinical outcomes compared to those with higher NER at baseline (OS HR 0.31, P < 0.01) [56]. In addition, the same team evaluated NER on-treatment changes and found that patients with a decreased NER at week 6 of treatment were associated with improved clinical outcomes in double ICI-treated aRCC patients (HR: 0.67, P-value: 0.002) [57] (Figure 1).

Other scores, such as the platelet-to-lymphocyte ratio (PLR) and the systemic immune-inflammation (SII) index, have also been evaluated in the context of aRCC treated with ICI. High PLR has been correlated with worse outcomes to ICI across different cancer types [58]. Similarly, Iinuma et al. [59] reported an association between high PLR and worse outcomes in aRCC patients receiving nivolumab plus ipilimumab (1-year PFS, 75.5% for low PLR vs. 49.7% for high PLR; P = 0.034). In this study, the SII was also associated with worse PFS, which is in line with previous studies across different cancer types (P = 0.023) [60].

Soluble factors, including cytokines and chemokines, play a critical role in anti-tumor immunity [61]. Cytokines and chemokines are released by both adaptive and innate immune cells, stromal cells, and tumor cells, and can be measured at both tissue and systemic levels. Certain soluble factors, such as IL-6, IL-8, and VEGF, are involved in carcinogenesis, myeloid inflammation, and promote immunosuppression, while others, such as the chemokine CXCL13 and the B-cell activating factor (BAFF), are involved in B cell activation, survival, and differentiation [62–64].

The idea of measuring tumor-derived blood biomarkers is not new. For years, and in several cancer types, we have used different tumor protein markers such as CA 15.3, CA 125, prostate-specific antigen (PSA), or carcinoembryonic antigen (CEA) to detect disease recurrence, progression, and response to therapy [93]. However, these tumor markers lack specificity and are not able to predict response to specific treatment types and thus guide treatment selection. During the last decade, new tumor-derived biomarkers such as circulating tumor cells (CTCs), circulating tumor deoxyribonucleic (DNA; ctDNA) and micro-ribonucleic acids (miRNAs) have been developed. This is what we commonly refer to as liquid biopsy, and it has the potential to help us detect cancer, guide treatment selection, assess real-time tumor response to therapy, and identify resistant clones [94].

The tryptophan-kynurenine-aryl hydrocarbon receptor (Trp-Kyn-AhR) pathway contributes to immunosuppression in T cell inflammed tumors [131]. Kyn results from Trp catabolism by indoleamine 2,3-dioxygenase (IDO) or trytpophan 2,3-dioxygenase (TDO) [132]. Trp degradation and depletion contribute to tumor evasion, while increasing Kyn metabolites which contribute to immunosuppression and cancer progression [132]. Despite promising results in previous preclinical studies showing that suppression of this pathway could enhance ICI efficacy, the combination of a selective IDO1 inhibitor and pembrolizumab in unselected melanoma patients failed to improve outcomes in a phase 3 randomized study [133, 134]. Recently, Li and colleagues [135] conducted a comprehensive study of the Trp-Kyn pathway in melanoma and RCC patients treated with nivolumab. In this study, treatment with PD-1 blockade induced Trp/Kyn conversion. More importantly, the increase of the Kyn/Trp ratio during treatment was a predictor of survival in both cohorts of melanoma and RCC, and it was further validated in a larger independent study comparing nivolumab vs. everolimus in pretreated RCC patients. At week 4, the Kyn/Trp increase was significantly associated with worse OS in the nivolumab arm but not in the everolimus arm (Figure 1). These results suggest that serum Kyn/Trp monitoring could help identify which patients are more likely to benefit from IDO and PD-1 inhibition and deserve further study.

Current research on circulating biomarkers for predicting response to ICI treatment in RCC is promising, although it is still in its early stages (Figure 1). In addition, circulating biomarkers have not been extensively studied in the context of ICI combination therapy in aRCC, the standard of care in first-line setting. However, these biomarkers, encompassing immune cell populations, soluble factors, ctDNA, CTCs, and metabolomic profiles, hold significant potential. Compared to tissue biopsies, they are less invasive, potentially more comprehensive, and may enable real-time monitoring. The analysis of immune cell populations and circulating soluble factor seems to provide valuable insights into the immune landscape and real-time monitoring that could help identify patients who are resistant or responsive to ICI. Meanwhile circulating DNA and CTCs could help guide treatment selection and de-escalation/intensification strategies, by idenfitying patients at higher risk of recurrence and by discerning good from poor responders. Circulating biomarkers hold the potential to help us identify which patients will benefit most from double ICI therapy or ICI-AA therapy.

To harness these potentials, it is essential to standardize detection methods and establish consistent cut-offs, and validate these biomarkers in large, randomized clinical trials. Moreover, achieving meaningful predictive accuracy will likely require the integration of multiple biomarkers and multiomics techniques. Studies like the pragmatic European CARE1 trial, which evaluates double ICI against ICI-AA therapy, are incorporating integrated circulating biomarkers studies that could provide new insights. Integration of biomarker studies into clinical trials will be essential for advancing these biomarkers towards clinical practice.