Introduction

In the last decades, the introduction of immune checkpoint inhibitors (ICIs) alone or in combination with vascular endothelial growth factor receptor tyrosine kinase inhibitor (VEGFR-TKI) provided a paradigm shift in the therapeutic landscape of metastatic renal cell carcinoma (mRCC) [1,2,3,4,5,6]. Due to the survival benefit over everolimus observed in the randomized phase III Checkmate 025 trial [7], nivolumab was the first in class ICI approved in 2005 for patients with mRCC previously treated with at least a prior VEGFR-TKI. However, despite the survival advantage achieved with this new therapeutic strategy, mRCC is still a lethal disease accounting for a median overall survival (OS) of 25 months. mRCC is a heterogenous disease reflecting different clinical behaviors spanning from an indolent to a rapidly progressive disease. Similarly, the benefit achieved from nivolumab may vary widely from long-term disease control rate to hyper-progression [8,9,10]. Clinical and biological features underlining long-term response to nivolumab in mRCC are still under investigation [11, 12].

In this analysis of the multicentre retrospective Meet-URO 15 study, we attempt to evaluate the association between clinical characteristics and outcome in long-term to nivolumab among patients with mRCC previously treated with at least a prior VEGFR-TKI.

Materials and methods

Patients and treatments

This retrospective study involved patients with previously treated mRCC, who received at least one cycle of nivolumab between May 2016 and January 2019 across 34 Oncology Centers in Italy. Patients should be at least 18 years old, have a histologically confirmed diagnosis of mRCC and have received at least one completed infusion of nivolumab as a second or further treatment line, as standard clinical practice. Patients’ demographics and clinical characteristics were reported. Nivolumab was initially delivered intravenously at a dose of 3 mg/kg every 2 weeks, and then at a fixed dosage of 240 mg every 2 weeks or 480 mg every 4 weeks, in line with local clinical protocols. The treatment was continued until either disease progression or intolerable toxicity. Written informed consent was obtained from each patient. Ethical sanction for this study was secured from the Ethics Regional Ethical Committee of Liguria, under registration number 068/2019, and the research adhered to the principles of the Declaration of Helsinki.

Outcome assessment and statistical analysis

Tumor assessments were performed every 2–4 months of treatment, according to local clinical practice, or whenever progression was clinically suspected according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 [13]. Progression-free survival (PFS) was defined as the time from treatment initiation to disease progression or death whichever occurred first, while OS was defined as the duration from the beginning of nivolumab to death from any cause or to the final follow-up visit date. In this study, patients remained progression-free for > 24 months while receiving nivolumab were categorized as long-term responders.

Kaplan–Meier method was utilized to estimate both PFS and OS throughout the follow-up period. A χ2 test was applied to compare the distribution of categorical baseline characteristics between long- and short-term responders. Quantitative data were described using median and range, while qualitative data were presented using numbers and percentages. A logistic regression model was employed to assess the influence of each clinical-pathological variable (age, gender, histological type, prior nephrectomy, Karnofsky Performance Status (KPS), International Metastatic RCC Database Consortium (IMDC) score at diagnosis and at start of nivolumab, neutrophil-to-lymphocyte ratio (NLR), sites of metastases, metastatic at diagnosis, line of nivolumab therapy and type of first line therapy) on the long-term response. Significative variables at univariate analysis were included in the multivariate model. Level of statistical significance was set to 0.05. The analyses were conducted using Stata SE version 18.

Results

Patients characteristics

This retrospective analysis included a cohort of 571 patients diagnosed with mRCC treated with nivolumab as second or further line of therapy with a median follow-up of 22.1 months.

Among these, 132 patients (23.1%) remained progression-free for > 24 months while receiving nivolumab, and they were categorized as long-term responders. Baseline characteristics according to long-term and short-term responders are detailed in Table 1. Characteristics were well-balanced among the groups in terms of median age, gender distribution, and histology, with clear cell carcinoma being the most prevalent histologic type, accounting for 84.3% of all patients. The median PFS (mPFS) and OS (mOS) for all patients were 7.0 months (95% CI, 5.0–8.0) and 25.0 months (95% CI, 21.0–30.0) respectively (Fig. 1). Long-term responders exhibited a mPFS (mPFS) of 55.0 months (95% CI: 45.0-not reached [NR]), while the median OS was NR (95% CI, 79.0-NR) (Fig. 2). Conversely, short-term responders exhibited a mPFS of 4.4 months (95% CI: 3.9–5.1) with a m OS of 17.0 months (95% CI: 14.0–19.0) (Fig. 3).

Table 1 Patients characteristics
Full size table
Fig. 1
figure 1

Progression-free survival and overall survival and in entire population

Full size image
Fig. 2
figure 2

Progression-free survival and overall survival in long-term responders

Full size image
Fig. 3
figure 3

Progression-free survival and overall survival of short-term responders

Full size image

Long-term response predictors

Among long-term responders, almost 90% of the patients had previous nephrectomy, with a statistically significant predominance when compared to short-term responders (95% vs. 86%; p < 0.01). Analysis based on the IMDC risk group displayed a higher percentage of patients with an intermediate-poor risk status at diagnosis in the short-term response group compared to the long-term response group (69.7% vs. 57.4%; p < 0.01). A NLR ≥ 3.2 was recorded in 30.3% of patients exhibiting a long-term response versus 44.2% of those with short-term responses (p < 0.01).

Furthermore, a higher proportion of patients in the short-term responders group presented bone metastases compared to the long-term responders (39.4% vs. 22.7%; p < 0.01). Notably, no statistically significant differences were noted between the two groups concerning first-line therapy and the number of therapy lines of nivolumab (2nd vs. > 3rd line).

Logistic regression analysis was conducted on the entire cohort of 571 patients to explore the associations between clinical and pathological variables and long-term responses. Factors evaluated as potential risks included age, gender, histological type, prior nephrectomy, KPS, IMDC score at diagnosis and at start of nivolumab, NLR, sites of metastases, metastatic at diagnosis, line of nivolumab therapy and type of first line therapy. The odds ratios (OR) estimated for each variable in both univariable and multivariate analyses are presented in Tables 2 and 3.

Table 2 Univariate analysis of the relationship of clinical-pathological variables with PFS > 24 months
Full size table
Table 3 Multivariate analysis of the relationship of various clinical-pathological variables with PFS > 24 months
Full size table

In the univariate analyses, long-term responders displayed higher odds of having KPS ≥ 80% (OR, 4.10; 95% CI, 1.84–9.11; p < 0.01) (Fig. 1S) and having undergone previous nephrectomy (OR, 3.45; 95% CI, 1.46–8.18; p = 0.01) (Fig. 2S). Conversely, they exhibited lower odds of having bone metastases (OR, 0.45; 95% CI, 0.29–0.71; p < 0.01) (Fig. 3S), an IMDC intermediate-poor status at diagnosis (OR, 0.58; 95% CI, 0.38–0.90; p = 0.01) and at the onset of nivolumab treatment (OR, 0.34; 95% CI, 0.22–0.52; p < 0.01) (Fig. 4S), as well as an NLR ≥ 3.2 (OR, 0.56; 95% CI, 0.37–0.85; p < 0.01) (Fig. 5S) compared to short-term responders.

The multivariable analysis exploring the relationship between clinical-pathological variables and response group is detailed in Table 3. Patients with PFS > 24 months were more likely to have a KPS ≥ 80% (OR, 2.68; 95% CI, 1.18–6.11; p = 0.02) and less likely to have bone metastases (OR, 0.58; 95% CI, 0.36–0.94; p = 0.03), an IMDC intermediate-poor status at the start of nivolumab (OR, 0.44; 95% CI, 0.28–0.68; p < 0.01), and an NLR ≥ 3.2 (OR, 0.59; 95% CI, 0.38–0.92; p = 0.02) compared to patients with PFS ≤ 24 months.

Discussion

As the therapeutic landscape of advanced RCC has changed due to the development and reimbursement of new treatment combinations [1,2,3,4,5,6], a deeper understanding of clinical characteristics and baseline laboratory features affecting clinical outcomes is needed and may help clinicians in the treatment decision making. Several nomograms for mRCC were developed to better predict prognosis and are mostly based on clinical factors laboratory parameters [14, 15]. These models have many limitations since they have been designed before the approval of ICI and do not take into account prognostic factors such as age, site of metastases, number and duration of previous treatments and inflammatory scores [16].

The phase 3 CheckMate 025 trial showed longer median OS with nivolumab (25 months) compared with everolimus (19.6 months) in previously treated patients with advanced RCC [7]. This benefit was sustained across all the subgroups, including Memorial Sloan Kettering Cancer Center (MSKCC) and IMDC risk groups, number and sites of metastases, age < 65 and ≥ 65 years, number, and duration of prior therapies15. The safety and efficacy observed in the CheckMate 025 trial were consistent with those reported in real-world setting major series, showing a good correspondence from the results in clinical trials and those in clinical practice [15, 17]. As such, further investigations of clinical predictive factors that could more accurately define the outcome of advanced RCC in the current treatment landscape from clinical practice remains a clinical need.

The multicentre retrospective Meet-URO 15 study [18] explored the prognostic role of baseline peripheral blood inflammatory indices and clinical factors in advanced RCC patients receiving nivolumab as second or further line to develop a prognostic score that could better predict survival outcome and overcome the limitations of previous analyses in short series with nomograms [14, 15]. Inflammatory indexes as neutrophil–lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), platelet-to-lymphocyte ratio (PLR), systemic inflammation index (SII), and systemic inflammation response index (SIRI) have been recently developed and confirmed for outcome prediction in pretreated mRCC [16, 19, 20]. The MeetURO-Score included priori and recent biomarkers showing a prognostic impact for NLR, IMDC score, and bone metastases identifying five different prognostic groups: group 1 (mOS not reached), group 2 (mOS 43.9 months), group 3 (mOS 22.4 months), group 4 (mOS 10.3 months) and group 5 (mOS 3.2 months).

However, although nivolumab provided a survival benefit in pretreated advanced RCC patients [21], usually only a small proportion of them achieves a long-term benefit [8].

In the present analysis of Meet-URO 15 we attempted to define the clinical characteristics that correlate with longer response to nivolumab. At multivariable analysis we found out that patients with PFS > 24 months were more likely to have a KPS ≥ 80% and less likely to have bone metastases, an IMDC intermediate-poor status, and an NLR ≥ 3.2.

Our data are like that reported in a previous long-term response study of sunitinib and pazopanib with accordance to the general characteristics of mRCC patients [22,23,24].

Age < 65 years, previous nephrectomy, absence of bone or lung metastases and favorable MSKCC risk status were the factors associated with long-term responses in mRCC patients receiving TKI as first line therapy [22,23,24]. Accordingly with previous studies, NLR is significantly associated with poorer OS and PFS, and lower rates of response and clinical benefit, after ICI therapy across multiple cancer types [25]. We acknowledge several limitations of the study including the retrospective design, and the numbers of previous treatment received. However, we believe that our study provides the rationale for prospectively exploring the presented putative biomarkers of prolonged response to ICI. Given the lack of validated biomarkers, the identification of prognostic and predictive clinical and biochemical features would allow to identify patients that could most benefit from immunotherapy. Moreover, these data are extremely punctual since several ongoing phase III clinical trials are exploring the efficacy of ICI combinations in patients previously progressed to first line ICI-based therapy [26, 27].

Conclusion

In this large retrospective Meet-URO 15 analysis, we identified, among patients with mRCC suitable for nivolumab treatment, the prognostic role of clinical factors and inflammatory indices that may predict long response to nivolumab in real life setting. Future perspectives include the external validation of these findings in the International multicenter real-world REGistry for patients with metastatic renAL cell carcinoma—Meet-URO 33 study (REGAL study) [28].