Future perspectives for personalized immunotherapy in renal cell carcinoma

KEYWORDS:

• Immunotherapy
• personalized therapy
• renal cell carcinoma
• PD1
• VEGF
• TKI
• PD-L1
• CTLA-4

1. Introduction

Angiogenesis is a hallmark of renal cell carcinoma (RCC) and represents a major target for therapy. RCC has a high expression of antigens resulting in immune cell infiltration associated with primary tumor growth and metastases. Renal cancer cells adopt several strategies to escape immune surveillance. These features provide a strong rational for the use of the immune-checkpoint inhibitors for treating disease. Nivolumab, a monoclonal antibody (mAb) directed against PD-1, has recently demonstrated an impressive long-lasting response (overall response rate [ORR] 25% versus 5%; p < 0.001) and significantly increases the overall survival (OS) of pretreated RCC patients compared to everolimus in the phase III Checkmate-025 study (mOS (median overall survival) 25.0 versus 19.6 months; HR (hazard ratio) 0.73 [98.5% confidence interval [CI], 0.57–0.93]; p = 0.002) [1]. Nivolumab retained the OS benefit regardless of the MSKCC (Memorial Sloan Ketterin Cancer Center) and IMDC (International Metastatic Renal Cell Carcinoma Database Consortium) prognostic risk categories (high benefit in poor-risk patients), prior systemic therapies, number, and the site of metastases [2].

Several questions about treatment with immune-checkpoint inhibitors in RCC patients remain unanswered. Future efforts should be directed at solving these issues to improve patient outcomes.

Regarding the question of whether there is a correlation between ORR and outcomes, patients with progressive disease as best response by month 4 showed improved OS with nivolumab over everolimus despite similar tumor characteristics, suggesting that the potential antitumor activity of nivolumab does not necessarily correlate with tumor response [3].

Assuming that approximately one out of four patients could have a benefit in terms of response to treatment, is it possible to identify in advance this patient? And have we identified biomarkers that can predict patient response?

A meta-analysis of six published studies revealed that a higher level of PD-L1 expression increased the risk of death by 81% (HR: 1.81, 95% CI 1.31–2.49; p <  0.001), representing therefore a negative prognostic factor [4]. PD-L1 expression has demonstrated a higher response rate and clinical benefit from immune-checkpoint inhibitors in melanoma and lung cancer. Regarding RCC, contrary to what was expected, the improved OS with nivolumab was not correlated with PD-L1 expression [1]. However, the results of a phase II trial (IMmotion150) presented at ASCO 2017 Genitourinary Cancers Symposium showed that patients with PD-L1 expression receiving atezolizumab plus bevacizumab as first line had a 36% reduction in the risk of disease worsening or death compared to sunitinib. This finding suggested the possible role of PD-L1 as a predictive response factor.

A genomic study of whole exome and transcriptome (RNAseq) sequencing of RCC samples from patients treated with nivolumab revealed, although in a small cohort (seven metastatic RCC [mRCC] patients), a correlation between the expression of tumor neo-antigens derived from non-synonymous somatic mutations and tumor objective response, assuming a predictive value of treatment response [5].

Of note, gut microbiota – and its altered composition – has been postulated as a potential factor related to immune-checkpoint inhibitor activity. Retrospective analysis showed that 80 mRCC patients treated with anti-PD-1/PD-L1-targeted therapies and broad-spectrum antibiotics up to 1 months prior to the start of immunotherapy had a worsened prognosis, even after multivariable analysis for prognostic risk factors (p < 0.002) [6].

Another important question concerns the ideal duration of the treatment with nivolumab. In the case of radiological disease progression, can we discontinue nivolumab in patients who are clinically benefiting from it or is it conceivable to continue nivolumab beyond progression?

Despite their being evidence showing that immune response activated by immune-checkpoint blockade could maintain itself even when treatment is discontinued (possible role of memory T cells), no conclusive data are available for RCC [7]. Indeed, immunotherapy was continued until there was disease progression in clinical trials exploring immune-checkpoint inhibitors in RCC patients. Escudier et al. showed that a small percentage of asymptomatic patients receiving nivolumab beyond progression could achieve tumor shrinkage and a survival benefit [8]. However, the efficacy of continuing nivolumab beyond progression remains unclear, highlighting the need of clinical trials exploring this strategy.

Another unsolved issue regards the possibility of enhancing the anticancer effect of immune-checkpoint inhibitors by combining immunotherapy with other therapies.

Blocking the VEGF (Vascular Endothelial Growth Factor) axis could not only impair angiogenesis, but could also enhance immunity against cancer cells by mediating tumor cell escape from the host immune system, affecting dendritic cell maturation from precursors, and reducing the expression of major histocompatibility complex class II and the ability to take up soluble antigens [9]. Interestingly, a significant relationship between VHL(Von Hippel-Lindau) gene mutations, HIF-2a (Hypoxia-inducible factors 2 alpha) stabilization, and PD-L1 expression in ccRCC (clear cell Renal Cell Carcinoma) reinforced the hypothesis of a close interplay between angiogenesis and the PD-1 pathway [10]. Furthermore, sunitinib could diminish immunosuppressive myeloid-derived suppressor cells, leading to an increase in effector T-cell function and T-cell interferon-gamma production, and a decline in regulatory T-cell number [11,12]. Atezolizumab, when combined with bevacizumab, enhances antigen-specific intra-tumoral CD8(+) T-cells migration [13]. The association between anti-angiogenic and immune-checkpoint action is an attractive strategy that is currently under investigation in several phase III trials (Table 1).

Table 1. Phase III trials of immunotherapy alone or in combination with tyrosine kinases inhibitors in metastatic RCC.

Study name	Experimental arm	Comparator arm	Key inclusion criteria	N	Primary end point	Estimated completion data
NCT02684006 Javelin Renal 101	Avelumab + axitinib	Sunitinib	Untreated patients with metastatic/locally advanced RCC with clear cell component	583	PFS	September 2020
NCT02420821 IMmotion151	Atezolizumab + bevacizumab	Sunitinib	Untreated patients with metastatic/locally advanced RCC with clear cell component	915	OS PFS	July 2020
NCT03141177 CheckMate 9ER	(A) nivolumab + cabozantinib (B) nivolumab + ipilimumab + cabozantinib	Sunitinib	Untreated patients with metastatic/locally advanced RCC with clear cell component	1014	PFS	June 2022
NCT02231749 CheckMate214	Nivolumab + ipilimumab	Sunitinib	Untreated patients with metastatic/locally advanced RCC with clear cell component	1070	OS PFS ORR	September 2019
NCT02853331 KEYNOTE-426	Pembrolizumab + axitinib	Sunitinib	Untreated patients with metastatic/locally advanced RCC with clear cell component	840	OS PFS	January 2020

RCC: renal cell carcinoma; OS: overall survival; PFS: progression-free survival; ORR: overall response rate.

A phase I open-label study of nivolumab combined with sunitinib or pazopanib showed promising antitumor activity (ORR 52% with nivolumab plus sunitinib, and 45% with nivolumab plus pazopanib) with durable responses, but at the cost of a considerable toxicity profile [14]. The combination of atezolizumab/bevacizumab (IMmotion 150) showed promising efficacy compared to sunitinib in the PD-L1-positive subgroup (mPFS (median progression free survival) 14.7 vs. 7.8 months; HR 0.64, 95% CI 0.38–1.08). Data from phase I clinical studies presented at the 2017 ESMO meeting showed that the association between axitinib and avelumab (NCT02493751) as well as the combination between cabozantinib and nivolumab (NCT02496208) had a manageable safety profile with an interesting antitumor activity, justifying the planning of randomized clinical trials exploring these associations.

Another interesting strategy concerns the possibility of dual immune-checkpoint inhibition, by simultaneously targeting PD-1 and CTLA-4. Encouraging clinical activity and acceptable safety have been reported with the combination of nivolumab and ipilimumab in mRCC in a phase I trial [15], supporting the planning of phase III clinical trials exploring this association.

Recently, a phase III trial (IMPRINT) failed to demonstrate a survival benefit with the combination of sunitinib and IMA901 (a multi-peptide cancer vaccine) compared to sunitinib monotherapy [16]. Of note, sunitinib was associated with a decrease in monocytes after the first treatment cycle and a reduction of IMA901-specific T-cell response. These findings suggest that the interaction between anti-angiogenic therapy and immune response is far from being defined and further studies are required.

The role of immune-checkpoint inhibitors is under evaluation in both adjuvant and neoadjuvant settings (Table 2). In particular, the high levels of circulating tumor antigens due to the presence of the primary tumor could enhance the immune response following PD-1/PD-L1 inhibition, suggesting a potential activity of these agents as a preoperative treatment.

Table 2. Phase III clinical trials of immunotherapy in adjuvant and neoadjuvant setting.

Study name	Experimental arm	Comparator arm	Key inclusion criteria	N	Primary end point	Estimated completion data
NCT03142334 KEYNOTE-564	Pembrolizumab	Placebo	Patients who have undergone nephrectomy and have - intermediate (pT2G4 or pT3) - high risk (pT4 or N+), or M1(NED) RCC with clear cell component.	950	DFS	November 2022
NCT03138512 CheckMate 914	Nivolumab + ipilimumab	Placebo	Patients who have undergone nephrectomy(pT2aG3-4, pT2b, pT3, pT4 or N+) with clear cell component RCC	800	DFS	June 2023
NCT03024996 IMmotion010	Atezolizumab	Placebo	Pathologically confirmed RCC with a component of either clear cell histology or sarcomatoid histology that has not been previously treated in the adjuvant or neoadjuvant setting and classified as being at high risk of RCC recurrence	664	DFS	September 2023
NCT03055013 PROSPER	Nivolumab 2 cycles + nephrectomy + nivolumab 6 cycles	Nephrectomy alone	Patients with newly diagnosed higher risk RCC of any histology including sarcomatoid or (if preoperative biopsy was uninformative) – ‘unknown’ histology	766	DFS	July 2022

RCC: renal cell carcinoma; DFS: disease-free survival; M1 NED: participants who present not only with the primary kidney tumor but also solid, isolated, soft tissue metastases that can be completely resected at the time of nephrectomy.

Non-clear cell RCC (nccRCC) represents another promising field of research for immunotherapy. As usual, evidence for these heterogeneous and underrepresented containers of malignancies is limited.

The pattern of PD-L1 expression in nccRCC seems to be broadly heterogeneous (PD-L1 positivity in about 10% of tumor cells and more than half of tumor-infiltrating mononuclear cells) and related to RCC histology (mainly Xp11.2 translocation RCC and collecting duct carcinomas) [17]. Moreover, analogous to ccRCC, tumor cell PD-L1 positivity has been associated with aggressive clinical–pathological features and shorter OS.

A recently reported retrospective multicenter study of 40 metastatic nccRCC patients treated with PD-1/PD-L1 inhibitors (75% monotherapy, 25% a combination of PD-1/PD-L1 with an anti-VEGFR (vascular endothelial growth factors receptor) or anti-CTLA-4 therapy) revealed interesting activity with these agents (ORR 10% with monotherapy, 18% in the overall population; 6-month OS rate of 81%), further supporting the rationale for exploring the role of PD-1/PD-L1 inhibitors also in non-clear cell histologies [18].

A phase I study of atezolizumab enrolled 70 RCC patients, 10% (7 patients) of whom presented non-clear cell histology. The small cohort of nccRCC patients experienced no objective responses per RECIST (Response Evaluation Criteria in Solid Tumor) (versus 15% in the ccRCC population); however, one patient had a durable response according to immune response criteria. Of note, atezolizumab revealed impressive activity (ORR of 33%) in six patients with sarcomatoid features, suggesting more aggressive de-differentiated tumors as a potential search field [19]. In agreement with this observation, a case report of a 34-year-old male with papillary RCC with sarcomatoid differentiation showed a dramatic response to nivolumab [20]. The presence of a sarcomatoid differentiation has been associated with a higher percentage of PD-L1 expression (50% compared to only 3% in ccRCC without sarcomatoid features) [21], as a result of PBMR1 mutations that induce increased PD-L1 expression in RCC cells [22]. Therefore, RCC tumors with sarcomatoid features may represent a particular subgroup that benefits from anti-PD-1/PD-L1-targeted agents.

The activity of several anti-PD-1/PD-L1-targeted therapies is being prospectively evaluated in nccRCC patients: nivolumab (NCT02596035), pembrolizumab (NCT02853344/KEYNOTE-427 study), and atezolizumab plus bevacizumab (NCT02724878).

2. Conclusion

Immunotherapy represents a promising novel therapeutic weapon for RCC. Many efforts have to be addressed to understand the intimate mechanism of action of immunotherapy, in order to perfect its management (patients selection, treatment sequence, therapy duration, and response interpretation) and to further enhance its effects (combinations with other agents, identification of predictive biomarkers, etc.).

3. Expert opinion

Immunotherapy with immune-checkpoint inhibitors represents an important therapeutic option for ccRCC patients, given the favorable ratio between benefits (possibility of prolonging survival) and treatment-related side effects (excellent tolerability profile).

The survival advantage of immune-checkpoint inhibitors excites both clinicians and patients. Moreover, the prolongation of survival without any advantage in progression-free survival (PFS) could lead to a higher percentage of patients receiving a third (or more) line of therapy after progression to immune-checkpoint inhibitors. The problem is how to choose subsequent treatments. Does immunotherapy have impact on subsequent therapies? Probably, the rational management of available drugs (which differ in terms of mechanism of action, toxicity, and efficacy) needs to be based on the clinical therapeutic goal.

VEGF-targeted therapies (i.e. sunitinib, pazopanib, axitinib, and sorafenib), which prolong PFS without a significant survival benefit, could be a valid option if the PFS gain goes along with a prolonged control of symptoms or with a delay in their appearance. Theoretically, the ideal drug should improve response rate, PFS, and OS. But this result can be achieved in different ways. The response, and therefore the disease control, can be reached early; or alternatively it may be delayed or associated with an initial pseudo-progression – as observed with immunotherapy (in this case we will not have any PFS benefit, albeit with a potential OS gain). In the case of a symptomatic patient, with high disease burden, it will be crucial to get a rapid and prolonged response, so as to control the disease and reduce symptoms. Conversely, the therapeutic aim for an asymptomatic patient, with low disease volume that affects non-vital organs, can be reasonably directed at prolonging the OS even at the price of an initial pseudo-progression (Figure 1).

Figure 1. Proposed future treatment algorithm of metastatic Renal Cell Carcinoma.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This manuscript has not been funded.