Purpose:

Inflammation indexes and body mass index (BMI) are easily evaluated, predict survival, and are potentially modifiable. We evaluated the potential association of inflammatory indexes and BMI with the clinical outcome of patients with renal cell carcinoma (RCC) undergoing immune checkpoint inhibitor therapy.

Experimental Design:

A prospective cohort of patients with metastatic RCC treated with nivolumab enrolled in the Italian Expanded Access Program from July 2015 through April 2016 was examined. Reference measures of inflammation were identified for neutrophil-to-lymphocyte ratio (NLR) </≥ 3, systemic immune inflammation index (SII) </≥ 1,375, and platelet-to-lymphocyte ratio (PLR) </≥ 232. Patients were classified as high BMI (≥25 kg/m2) versus normal BMI (<25 kg/m2).

Results:

Among 313 evaluable patients, 235 (75.1%) were male, and median age was 65 years (range, 40–84 years), with 105 (33.69%) ≥70 years. In univariate analysis, age, performance status, BMI, SII, NLR, and PLR were able to predict outcome. In multivariate analyses, SII ≥1,375, BMI <25 kg/m2, and age ≥70 years independently predicted overall survival [OS; HR = 2.96, 95% confidence interval (CI), 2.05–4.27; HR = 1.59, 95% CI, 1.10–2.30; and HR = 1.65, 95% CI, 1.07–2.55, respectively). A patient with both SII ≥1,375 and BMI <25 kg/m2 was estimated to have much worse OS (HR, 3.37; 95% CI, 2.29–4.95; P <0.0001) than a patient with neither or only one risk factor. SII changes at 3 months predicted OS (P <0.0001).

Conclusions:

Normal BMI combined with inflammation tripled the risk of death, suggesting that these biomarkers are critical prognostic factors for OS in patients with RCC treated with nivolumab.

Translational Relevance

Inflammation indexes associated with body mass index (BMI) are risk factors associated with survival of patients with metastatic renal cell carcinoma (RCC) treated with VEGF-targeted therapies. We investigated whether lower BMI and higher inflammatory indexes combining neutrophils (N), platelets (P), and lymphocytes (L), as NLR (N/L), PLR (P/L), and SII (N × P/L) are associated with overall survival (OS) in metastatic RCC treated with nivolumab. In a cohort of 313 patients with metastatic RCC, age younger than 70 years, elevated SII, and normal BMI were independently associated with poor OS. Because BMI and SII are commonly collected and potentially modifiable, they have a high potential for clinical use in determining prognosis and may help guide interventions to optimize survival with nivolumab in metastatic RCC.

Nivolumab is a mAb that targets programmed death-1 receptor (PD-1). It has been approved as monotherapy for the treatment of patients with advanced renal cell carcinoma (RCC) progressing after prior antiangiogenic therapy (1). Biomarkers predicting clinical outcome with nivolumab would allow early identification of nonresponders and timely use of other therapies.

Two novel prognostic indicators are receiving increasing attention across many cancer types and RCC in particular. These are body mass index (BMI) and an elevated neutrophil-to-lymphocyte ratio (NLR, a measure of systemic inflammation). Obesity is an established risk factor for RCC and is associated with a better prognosis in these patients (2). The potential correlation between high BMI and more favorable treatment outcome to VEGF-targeted therapy in metastatic RCC has also been shown (3, 4). The role of blood cell count indexes as possible biomarkers of response and efficacy has been recently investigated. NLR is the most tested prognostic index and has been associated with prognosis in several tumors including RCC treated with antiangiogenic agents (5–10). Thrombocytosis and lymphopenia have also been associated with prognosis in metastatic RCC treated with first-line VEGF-targeted treatment (11, 12). Systemic immune inflammation index (SII) combines these three parameters: neutrophils, platelets, and lymphocytes, and has been previously significantly associated with prognosis in several tumor types including metastatic RCC (13–15).

Both inflammation and BMI may be able to identify patients with poor prognosis to VEGF-targeted therapies, but the relationship between these factors and their independent associations with survival are not well studied. Recent trials underscore the importance of systemic inflammation as a driver of muscle degradation in patients with far advanced disease, but other studies have suggested that similar processes may occur in patients with localized disease. Elevated inflammatory indexes as NLR were present in nearly half of patients and were associated with decreased BMI as predictive of poorer survival even in patients with nonmetastatic colorectal cancer (16, 17). In RCC, the “obesity paradox” of longer survival in patients with high BMI has been explained by altered fatty acid pathways in obese relative to normal weight patients with marked downregulation of fatty acid synthase, which could have an impact in immune inflammatory function with potential impact on cancer survival (4, 18, 19).

To our knowledge, no prior study has examined the combined associations of the host systemic inflammatory response and BMI with RCC survival. Moreover, no prior study has evaluated the potential association of BMI and inflammatory indexes with clinical outcome of patients undergoing immune checkpoint inhibitor therapy. Accordingly, we conducted an analysis to determine whether pretreatment inflammatory indexes and BMI could predict clinical outcome to nivolumab using data from patients with metastatic RCC enrolled in the Italian Expanded Access Program (EAP). Subsequently, we assessed the independent and combined associations of these two prognostic indicators with survival.

Study cohort

The purpose of the EAP was to provide nivolumab to patients with RCC who had progressed on VEGF-targeted treatment while the drug was evaluated by the European Medicines Agency. Key inclusion criteria were histologically confirmed metastatic RCC, Eastern Cooperative Oncology Group (ECOG) performance status ≤2, at least one prior VEGF-targeted therapy in the advanced setting, and presence of asymptomatic brain metastases. Patients with active or suspected autoimmune disease were excluded. Treatment consisted of nivolumab 3 mg/kg every 2 weeks until unacceptable toxicity, disease progression, or patient choice. The study protocol was reviewed and approved by local Ethics Committees and in accordance with the precepts established by the Declaration of Helsinki. Written informed consent was obtained from all study participants. All data were prospectively collected on electronic patient files.

From July 2015 to April 2016, 389 patients were treated with at least one dose of nivolumab in the EAP in Italy. Of these, 313 (80.5%) had baseline complete blood counts necessary for the inflammatory indexes and all clinical data available and were considered fully evaluable for this post hoc analysis.

Post hoc analysis variable definitions

For this post hoc analysis, three inflammatory indexes were determined on the basis of baseline values of neutrophils (N), lymphocytes (L), and/or platelets (P) in patients with metastatic RCC included in the EAP: SII defined as P × N/L, NLR defined as N/L, platelet–lymphocyte ratio (PLR) defined as P/L. X-tile 3.6.1 software was used to determine the most appropriate discriminatory cut-off values for SII, NLR, and PLR for the analysis (13). The cutoffs were selected as the values that maximize differences between overall survival (OS) in the two groups identified (below and above the cutoff). We selected the height and weight obtained closest to beginning of nivolumab treatment to calculate the BMI defined as the ratio of weight divided by the height squared. Patients were classified into BMI groups defined by the World Health Organization: underweight (BMI < 18.5 kg/m2), normal weight (BMI 18.5–<25 kg/m2), overweight (BMI 25–<30 kg/m2), and obese (BMI ≥ 30 kg/m2) and condensed to high BMI (≥25 kg/m2) versus normal BMI (<25 kg/m2).

Statistical analysis

Data were presented by absolute frequency and percentage for categorical variables and by median and range for continuous variables (e.g., age). Association between categorical variables was assessed using the χ2 test. Differences were considered statistically significant when P < 0.05. The primary clinical outcome was OS. OS was defined as the time period between nivolumab initiation and the date of death, or it was censored on the day of the last follow-up visit. Objective response rate (ORR) was defined as the proportion of patients who achieved a complete or partial response in all evaluated patients. Disease control rate (DCR) was defined as the proportion of patients who achieved a complete or partial response or stable disease in all evaluated patients. The Kaplan–Meier method was used to estimate OS and the log-rank test was used to assess differences between survival. Cox proportional hazards regression model was used to estimate HRs and their 95% confidence intervals (CI). A multivariate analysis was carried out considering only factors significant at the univariate analysis and based on a stepwise forward procedure with enter and remove limits of 0.05 and 0.10, respectively.

The impact of change on survival outcomes was evaluated by the landmark analysis at 3 months. For this analysis, patients with early disease progression/death or patients lost to follow-up before the landmark time were excluded. All statistical analyses were conducted by an experienced biostatistician with SPSS Statistical software, version 21.0.

Patient characteristics

A total of 313 cases were considered for this analysis, with a median age before starting nivolumab of 65 years (range 40–84), 235 (75.1%) male. The histotype was clear cell RCC in 280 (89.5%) cases. Only 69 (22.1%) patients received nivolumab as second-line therapy after a VEGF-targeted treatment, most cases received nivolumab as third-line (n = 107, 34.3%) or further line of therapy (n = 136, 43.6%). Baseline clinical characteristics are shown in Table 1. An optimal cut-off point for the SII of 1,375 × 109 stratified these patients into high (≥1,375, n = 96, 30.7%) and low SII (<1,375, n = 217, 69.3%) groups, cutoff for NLR was 3 and categorized as high (≥3, n = 195, 62.3%) and low NLR (<3, n = 118, 37.7%), for PLR was 232 with high (≥232, n = 108, 34.5%) and low PLR (<232, n = 205, 65.5%). Normal BMI < 25 kg/m2 was reported in 50.2% of patients, high BMI ≥ 25 kg/m2 in 49.8%. The percentage of cases with BMI < or ≥25 kg/m2 was independent from the line of therapy.

Table 1.

Patient characteristics at baseline (N = 313)

CharacteristicsN (%)
Age in years, median (range) 65 (40–84) 
≥70 years, n (%) 105 (33.6%) 
Gender, male/female 235/78 
ECOG Performance status 
 0 143 (47.2%) 
 1 146 (48.2%) 
 2 14 (4.6%) 
 NA 10 
IMDC Prognostic group 
 Favorable (0 risk factors) 53 (19.0%) 
 Intermediate (1–2 risk factors) 194 (69.8%) 
 Poor (≥3 risk factors) 31 (11.2%) 
 NA 35 
 Nephrectomy 295 (94.3%) 
Histology 
 Clear cell 280 (89.5%) 
 Nonclear cell 28 (8.9%) 
 Undifferentiated/unknown 5 (1.6%) 
Metastasis site 
 Lung 230 (73.5%) 
 Lymph node 193 (61.7%) 
 Bone 154 (49.2%) 
 Liver 106 (33.9%) 
 Brain 30 (9.6%) 
Number of prior therapy lines 
 1 69 (22.1%) 
 2 107 (34.3%) 
 ≥3 136 (43.6%) 
 NA 
Hemoglobin level (mg/dL) 
 ≥12 139 (44.4%) 
 <12 174 (55.6%) 
BMI 
 <18.5 7 (2.4%) 
 Normal, 18.5–<25 140 (47.8%) 
 Overweight, 25–<30 111 (37.9%) 
 Obese, ≥30 35 (11.9%) 
 NA 20 
CharacteristicsN (%)
Age in years, median (range) 65 (40–84) 
≥70 years, n (%) 105 (33.6%) 
Gender, male/female 235/78 
ECOG Performance status 
 0 143 (47.2%) 
 1 146 (48.2%) 
 2 14 (4.6%) 
 NA 10 
IMDC Prognostic group 
 Favorable (0 risk factors) 53 (19.0%) 
 Intermediate (1–2 risk factors) 194 (69.8%) 
 Poor (≥3 risk factors) 31 (11.2%) 
 NA 35 
 Nephrectomy 295 (94.3%) 
Histology 
 Clear cell 280 (89.5%) 
 Nonclear cell 28 (8.9%) 
 Undifferentiated/unknown 5 (1.6%) 
Metastasis site 
 Lung 230 (73.5%) 
 Lymph node 193 (61.7%) 
 Bone 154 (49.2%) 
 Liver 106 (33.9%) 
 Brain 30 (9.6%) 
Number of prior therapy lines 
 1 69 (22.1%) 
 2 107 (34.3%) 
 ≥3 136 (43.6%) 
 NA 
Hemoglobin level (mg/dL) 
 ≥12 139 (44.4%) 
 <12 174 (55.6%) 
BMI 
 <18.5 7 (2.4%) 
 Normal, 18.5–<25 140 (47.8%) 
 Overweight, 25–<30 111 (37.9%) 
 Obese, ≥30 35 (11.9%) 
 NA 20 

Abbreviation: NA, not available.

Systemic inflammation, BMI, and response

In the overall population of 313 cases, an ORR was reported in 70 of 286 (24.5%) assessable patients, including complete response in 1 (0.4%) case, partial response in 69 (24.1%), stable disease was reported in 106 cases (37.1%) and progressive disease in 110 (38.4%), whereas in the remaining 27 cases (8.6%) the tumor response was not assessed, in the majority of cases because of clinical deterioration before the first CT scan evaluation. Lower ORRs were associated with higher values of inflammatory indexes at baseline, and these results were consistent across all markers of systemic inflammation.

The ORR for patients with SII < 1,375 was 29% (60/207 cases), whereas those with SII ≥ 1,375 was only 13% (10/79 cases; P = 0.004). The ORR for patients with NLR < 3 was 27% (31/114 cases), whereas those with NLR ≥ 3 was 23% (39/172 cases; P = 0.38). The ORR for patients with PLR < 232 was 29% (56/196 cases), whereas those with PLR ≥ 232 was 16% (14/90 cases; P = 0.02). The ORR in patients with normal BMI < 25 was 24% (32/132 evaluable cases), whereas those with high BMI ≥ 25 was 27% (37/136 evaluable cases; P = 0.58). The DCR for patients with SII < 1,375 was 68% (141/207 cases), whereas those with SII ≥ 1,375 was only 41% (32/79 cases; P < 0.0001). The DCR for patients with NLR < 3 was 64% (73/114 cases), whereas those with NLR ≥ 3 was 60% (103/172 cases; P = 0.50). The DCR for patients with PLR < 232 was 68% (133/196 cases), whereas those with PLR ≥ 232 was 35% (32/90 cases; P < 0.0001). The DCR in patients with normal BMI < 25 kg/m2 was 58% (76/132 evaluable cases), whereas those with high BMI ≥ 25 kg/m2 was 67% (91/136 evaluable cases; P = 0.13).

Baseline high inflammatory indexes were associated with International mRCC Database Consortium (IMDC) prognostic factors, but no with histotypes and metastatic sites (Table 2).

Table 2.

Association between baseline high inflammatory indexes and IMDC prognostic factors, histotypes, and metastatic sites

Pts., NSII > 1,375, PNLR > 3, PPLR > 232, P
IMDC Prognostic score 
 Favorable (0 risk factors) 53 1 (2%) 21 (40%) 8 (15%) 
 Intermediate (1–2 risk factors) 194 58 (30%) 124 (64%) 68 (35%) 
 Poor (≥3 risk factors) 31 26 (84%) 27 (87%) 21 (68%) 
  P < 0.0001 P = 0.0005 P < 0.0001 
Bone metastases 
 No 159 43 (27%) 96 (60%) 52 (33%) 
 Yes 154 52 (34%) 100 (65%) 55 (36%) 
  P = 0.18 P = 0.36 P = 0.58 
Brain metastases 
 No 283 81 (29%) 179 (63%) 92 (32%) 
 Yes 30 14 (47%) 17 (57%) 15 (50%) 
  P = 0.04 P = 0.52 P = 0.05 
Lung metastases 
 No 83 22 (26%) 53 (64%) 25 (30%) 
 Yes 230 73 (32%) 143 (62%) 82 (36%) 
  P = 0.31 P = 0.75 P = 0.32 
Liver metastases 
 No 207 60 (29%) 131 (63%) 73 (35%) 
 Yes 106 35 (33%) 65 (61%) 34 (32%) 
  P = 0.47 P = 0.73 P = 0.60 
Lymph node metastases 
 No 120 31 (26%) 75 (62%) 35 (29%) 
 Yes 193 64 (33%) 121 (63%) 72 (37%) 
  P = 0.19 P = 0.86 P = 0.15 
Clear cell histotype 
 No 33 11 (33%) 18 (54%) 14 (42%) 
 Yes 280 84 (30%) 178 (64%) 93 (33%) 
  P = 0.72 P = 0.26 P = 0.30 
Pts., NSII > 1,375, PNLR > 3, PPLR > 232, P
IMDC Prognostic score 
 Favorable (0 risk factors) 53 1 (2%) 21 (40%) 8 (15%) 
 Intermediate (1–2 risk factors) 194 58 (30%) 124 (64%) 68 (35%) 
 Poor (≥3 risk factors) 31 26 (84%) 27 (87%) 21 (68%) 
  P < 0.0001 P = 0.0005 P < 0.0001 
Bone metastases 
 No 159 43 (27%) 96 (60%) 52 (33%) 
 Yes 154 52 (34%) 100 (65%) 55 (36%) 
  P = 0.18 P = 0.36 P = 0.58 
Brain metastases 
 No 283 81 (29%) 179 (63%) 92 (32%) 
 Yes 30 14 (47%) 17 (57%) 15 (50%) 
  P = 0.04 P = 0.52 P = 0.05 
Lung metastases 
 No 83 22 (26%) 53 (64%) 25 (30%) 
 Yes 230 73 (32%) 143 (62%) 82 (36%) 
  P = 0.31 P = 0.75 P = 0.32 
Liver metastases 
 No 207 60 (29%) 131 (63%) 73 (35%) 
 Yes 106 35 (33%) 65 (61%) 34 (32%) 
  P = 0.47 P = 0.73 P = 0.60 
Lymph node metastases 
 No 120 31 (26%) 75 (62%) 35 (29%) 
 Yes 193 64 (33%) 121 (63%) 72 (37%) 
  P = 0.19 P = 0.86 P = 0.15 
Clear cell histotype 
 No 33 11 (33%) 18 (54%) 14 (42%) 
 Yes 280 84 (30%) 178 (64%) 93 (33%) 
  P = 0.72 P = 0.26 P = 0.30 

Abbreviation: pts, patients.

Survival analysis

Univariate and multivariate Cox proportional hazards regression analyses were performed to assess the associations between factors of interest and OS. In univariate analysis, the following clinical variables predicted OS: age, performance status and BMI; whereas among the blood cell count variables, neutrophil and platelets, but not lymphocytes predicted outcome; in addition, all three blood cell count indexes (SII, NLR, and PLR) were able to predict outcome (Table 2). As observed in the Kaplan–Meier curves, patients with normal BMI, those with SII ≥ 1,375, and <70-year-old patients had the worst survival (Fig. 1A–C). Moreover, we analyzed BMI as a continuous variable and correlated with OS (HR, 0.93; 95% CI, 0.88–0.99; P = 0.02), showing that an increase of the weight is directly correlated with better prognosis.

Figure 1.

Kaplan–Meier estimate of OS of patient according to BMI (A), SII (B), and age (C).

Figure 1.

Kaplan–Meier estimate of OS of patient according to BMI (A), SII (B), and age (C).

Close modal

In multivariate analyses, SII ≥ 1,375, BMI < 25 kg/m2, and age ≥ 70 years independently predicted OS (HR = 2.96, 95% CI, 2.05–4.27; HR = 1.59, 95% CI, 1.10–2.30; and HR = 1.65; 95% CI, 1.07–2.55, respectively; Table 3). Under the model of independent effects, a patient with both SII ≥ 1,375 and BMI < 25 kg/m2 was estimated to have much worse OS (HR, 3.37; 95% CI, 2.29–4.95; P < 0.0001) than a patient with neither or only one risk factor.

Table 3.

Univariate and multivariate analyses of the association between baseline characteristics and OS

Univariate analysis HR (95% CI), PMultivariate analysis HR (95% CI), P
Age (<70 vs. ≥70) 1.66 (1.12–2.44), 0.01 1.62 (1.05–2.50), 0.03 
Gender (male vs. female) 1.34 (0.88–2.03), 0.17 — 
ECOG PS 
 (1 vs. 0) 1.41 (0.99–2.02), 0.06 — 
 (2 vs. 0) 2.28 (1.03–5.03), 0.04  
IMDC Prognostic groups 
 (1–2 vs. 0) 1.92 (1.06–3.47), 0.03 — 
 (≥3 vs. 0) 4.25 (2.11–8.56), <0.0001  
Clear cell (no vs. yes) 1.23 (0.62–2.43), 0.55 — 
Bone metastases (yes vs. no) 1.06 (0.74–1.53), 0.73 — 
Brain metastases (yes vs. no) 1.16 (0.64–2.11), 0.62 — 
Lung metastases (yes vs. no) 1.14 (0.74–1.75), 0.55 — 
Liver metastases (yes vs. no) 1.05 (0.72–1.53), 0.81 — 
Lymph node metastases (yes vs. no) 1.01 (0.70–1.46), 0.96 — 
Hb baseline (<12 vs. ≥12 mg/dL) 1.67 (1.05–2.67), 0.03 — 
Neutrophils (≥8,000 vs. <8,000) 2.62 (1.73–3.99), <0.0001 – 
Lymphocytes (≤1,000 vs. >1,000) 1.30 (0.87–1.94), 0.20 — 
Platelets (≥400,000 vs. <400,000) 2.48 (1.67–3.70), <0.0001 — 
SII (≥1,375 vs. <1,375) 3.35 (2.38–4.73), <0.0001 2.99 (2.07–4.31), <0.0001 
NLR (≥3 vs. < 3) 2.23 (1.52–3.28), <0.0001 — 
PLR (≥232 vs. <232) 2.37 (1.69–3.35), <0.0001 — 
BMI (<25 vs. ≥25) 1.50 (1.05–2.15), 0.02 1.58 (1.09–2.28), 0.01 
Univariate analysis HR (95% CI), PMultivariate analysis HR (95% CI), P
Age (<70 vs. ≥70) 1.66 (1.12–2.44), 0.01 1.62 (1.05–2.50), 0.03 
Gender (male vs. female) 1.34 (0.88–2.03), 0.17 — 
ECOG PS 
 (1 vs. 0) 1.41 (0.99–2.02), 0.06 — 
 (2 vs. 0) 2.28 (1.03–5.03), 0.04  
IMDC Prognostic groups 
 (1–2 vs. 0) 1.92 (1.06–3.47), 0.03 — 
 (≥3 vs. 0) 4.25 (2.11–8.56), <0.0001  
Clear cell (no vs. yes) 1.23 (0.62–2.43), 0.55 — 
Bone metastases (yes vs. no) 1.06 (0.74–1.53), 0.73 — 
Brain metastases (yes vs. no) 1.16 (0.64–2.11), 0.62 — 
Lung metastases (yes vs. no) 1.14 (0.74–1.75), 0.55 — 
Liver metastases (yes vs. no) 1.05 (0.72–1.53), 0.81 — 
Lymph node metastases (yes vs. no) 1.01 (0.70–1.46), 0.96 — 
Hb baseline (<12 vs. ≥12 mg/dL) 1.67 (1.05–2.67), 0.03 — 
Neutrophils (≥8,000 vs. <8,000) 2.62 (1.73–3.99), <0.0001 – 
Lymphocytes (≤1,000 vs. >1,000) 1.30 (0.87–1.94), 0.20 — 
Platelets (≥400,000 vs. <400,000) 2.48 (1.67–3.70), <0.0001 — 
SII (≥1,375 vs. <1,375) 3.35 (2.38–4.73), <0.0001 2.99 (2.07–4.31), <0.0001 
NLR (≥3 vs. < 3) 2.23 (1.52–3.28), <0.0001 — 
PLR (≥232 vs. <232) 2.37 (1.69–3.35), <0.0001 — 
BMI (<25 vs. ≥25) 1.50 (1.05–2.15), 0.02 1.58 (1.09–2.28), 0.01 

Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); Hb, hemoglobin.

When the SII groups (<1,375 or ≥1,375) were analyzed according to the BMI (<25 or ≥25), the four groups were as follows: (i) SII < 1,375 and BMI < 25 kg/m2 in 94 (32.3%) cases, (ii) SII < 1,375 and BMI ≥ 25 kg/m2 in 111 (38.1%), (iii) SII ≥ 1,375 and BMI ≥ 25 kg/m2 in 35 (12.0%), and (iv) SII ≥ 1,375 and BMI < 25 kg/m2 in 51 (17.5%). Figure 2 shows the OS according to these four groups.

Figure 2.

Kaplan–Meier estimate of OS according to the four groups identified using SII and BMI at baseline.

Figure 2.

Kaplan–Meier estimate of OS according to the four groups identified using SII and BMI at baseline.

Close modal

The change of the three inflammatory indexes over time was evaluable in 260 (83.1%) of 313 cases at one time point of the follow-up after 2–3 months. Patients with baseline SII < 1,375 with follow-up SII ≥ 1,375 were estimated to have a worse OS than those with follow-up SII < 1,375, P < 0.0001, as well as patients with baseline SII ≥ 1,375 with follow-up SII ≥ 1,375 were estimated to have a worse OS than those with follow-up SII < 1,375, P = 0.05. Patients with baseline NLR < 3 with follow-up NLR ≥ 3 were estimated to have a worse OS than those with follow-up NLR < 3, P = 0.02, as well as patients with baseline NLR ≥ 3 with follow-up NLR ≥ 3 were estimated to have a worse OS than those with follow-up NLR < 3, P = 0.002. Patients with baseline PLR < 232 with follow-up PLR ≥ 232 were estimated to have a worse OS than those with follow-up PLR < 232, P = 0.001, as well as patients with baseline PLR ≥ 232 with follow-up PLR ≥ 232 were estimated to have a worse OS than those with follow-up PLR <232, P = 0.03. Table 4 shows the progression-free survival (PFS) and OS according to SII changes.

Table 4.

Change in immune inflammation index and clinical outcome

SII GroupsaMedian PFS (95% CI)HR (95% CI)P1-Year OSHR (95% CI)P
Low-Low 138 7.7 (6.2–9.2)   86%  
Low-High 49 5.7 (4.0–7.4) 1.32 (0.91–1.94) 0.14 48% 3.4 (2.1–5.8) <0.0001 
High-Low 28 4.6 (2.5–6.7) 1.43 (0.89–2.29) 0.14 53% 3.2 (1.7–5.9) <0.0001 
High-High 43 3.0 (2.6–3.4) 2.36 (1.61–3.44) <0.0001 34% 6.1 (3.6–10.2) <0.0001 
SII GroupsaMedian PFS (95% CI)HR (95% CI)P1-Year OSHR (95% CI)P
Low-Low 138 7.7 (6.2–9.2)   86%  
Low-High 49 5.7 (4.0–7.4) 1.32 (0.91–1.94) 0.14 48% 3.4 (2.1–5.8) <0.0001 
High-Low 28 4.6 (2.5–6.7) 1.43 (0.89–2.29) 0.14 53% 3.2 (1.7–5.9) <0.0001 
High-High 43 3.0 (2.6–3.4) 2.36 (1.61–3.44) <0.0001 34% 6.1 (3.6–10.2) <0.0001 

NOTE: For this analysis, patients with early disease progression/death or patients lost to follow-up before the landmark time were excluded.

aFour groups of SII changes: (i) low-low (baseline SII < 1,375 and follow-up SII < 1,375); (ii) low-high (baseline SII < 1,375 and follow-up SII ≥ 1,375); (iii) high-low (baseline SII ≥ 1,375 and follow-up SII < 1,375); and (iv) high-high (baseline SII ≥ 1,375 and follow-up SII ≥ 1,375).

The efficacy of immune checkpoint inhibitors in metastatic RCC varies greatly across individual patients and among different histotypes. Whereas PD-L1 expression on the primary tumor is now accepted as a potential biomarker indicating the use of these mAbs as a first-line approach, no substantial results have been reported with clinical and molecular biomarkers for nivolumab used as second or greater line of treatment in metastatic RCC (1). We previously conducted an analysis of the Italian RCC EAP, which showed that efficacy and safety of nivolumab for the treatment of metastatic RCC were comparable to phase III results; OS, PFS, and ORR in our cohort were similar to the observations in the CheckMate 025 studies (1, 20).

In this analysis, we showed that for patients with metastatic RCC treated with nivolumab, higher SII was independently associated with worse outcomes in terms of reduced ORR, DCR, and shorter OS, whereas a relative increase of BMI was associated with better OS, and the combination of these two factors (high SII with normal BMI) nearly tripled the risk of death, suggesting that these biomarkers are prognostic factors for OS in patients with metastatic RCC treated with nivolumab. We also showed that high levels of inflammatory parameters at 2- or 3-month follow-up were associated with OS, in particular SII changes showed a stronger impact on OS (Table 4). Further studies should investigate SII changes over time associated with changes on the radiological response and the clinical outcome of these patients.

Interestingly, the results seen at univariate analysis confirmed the pretreatment impact on OS of neutrophils and platelets, but not lymphocytes, as already demonstrated in the IMDC prognostic classification (12, 21). As expected, high inflammatory indexes were associated with IMDC prognostic factors (Table 2); however, the IMDC prognostic score did not remain significant at multivariate analysis (Table 3). Among the three derived indexes NLR, PLR, and SII, only SII remained significant at multivariate analysis, as a consequence of using both neutrophil and platelet levels. Taken together these data confirm the prognostic role of platelets and neutrophils and suggest the combined index SII as the strongest inflammatory index for metastatic RCC. Our findings are consistent with and build upon previous reports evaluating NLR in patients with RCC treated with nivolumab (22–24), but suggest that SII appears superior to NLR as an inflammatory, readily available, prognostic marker in patients with metastatic RCC treated with immune checkpoint inhibitors in metastatic RCC, and warrants further larger prospective validation.

BMI has been already shown to be a robust biomarker for outcomes in RCC both in the preoperatory localized disease and in the metastatic VEGF–targeted treated population (2–4). In our experience, BMI was associated with OS but not with either ORR (P = 0.58) or DCR (P = 0.13); as such it may be speculated that BMI represents a simple biomarker for “better RCC”. However, improved OS in obese patients with cancer compared with those with normal BMI, has been recently observed also in patients with melanoma treated with immune checkpoint inhibitors (25). In addition, these clinical finding are corroborated by preclinical data indicating that obesity increases T-cell aging resulting in higher PD-1 expression and dysfunction, which is driven by leptin signaling, this effect leaves tumors notably more sensitive to checkpoint blockade (26).

Patient age and gender are arising as potential prognostic/predictive factors for immunotherapy response. In melanoma, older patients’ ≥60 years responded more efficiently to anti-PD-1 agents, and the ORR to these drugs increased with age (27). In RCC, elderly patients’ ≥70 years were associated with increased ORR of nearly 30% higher than the nearly 20% reported in younger patients (820). Results of a meta-analysis of 20 randomized clinical trials with immune checkpoint inhibitors in solid tumors suggest that males derive more benefit than females, even if only the CheckMate 025 trial included patients with RCC and the HR was unadjusted for other confounding factors (28). Moreover, a recent large retrospective analysis of the IMDC has shown that gender has no impact on the efficacy of nivolumab (29). Results of the multivariate analysis of this study, confirmed the positive prognostic impact of older age (≥70 years) of patients with RCC, whereas it did not show any impact based on gender (Table 3), suggesting that the conclusions of the above-mentioned meta-analysis may not be generalizable to the setting of RCC. Median values of BMI in healthy people is approximately 24.5 kg/m2 for men and 21.5 kg/m2 for women (30), thus a possible explanation of the marginal advantage of male gender on the efficacy of nivolumab in the CheckMate 025 trial could be also due to the higher BMI in males than females (1, 28). However, in the recent retrospective study in patients with metastatic melanoma, obesity, compared with normal BMI, has been associated with improved PFS and OS in patients treated with immunotherapy (pembrolizumab or nivolumab or atezolizumab, total of 331 cases), and this association has been mainly observed in male patients (255). Future research should consider BMI exploring the effectiveness of immunotherapies in men and women.

An unmet need is the development of biomarkers of clinical outcome to immune checkpoint inhibitors to identify patients who are likely to respond and obtain clinical benefit from such treatments. This aspect is particularly relevant for tumors with low response rates, such as metastatic RCC treated with nivolumab which has response rates ≤25%–30%, but long-term survival of more than 4 years in 20%–25% of cases (31, 32).

To our knowledge, this is the first study to examine the relationship between biomarkers of systemic inflammation and BMI in RCC, and the only study to examine whether inflammatory indexes and BMI are independently associated with OS in metastatic RCC. As in any retrospective analyses, in our study several potential biases should be considered. In addition, other limitations are the use of BMI as the only morphometric parameter, because it does not provide information on muscle mass and fat repartition, and lacks metabolic biomarkers. Finally, lack of a validation cohort is a concern, however our data refer to a late use of nivolumab, mostly in third or fourth line (Table 1), whereas the current use in mainly in second-line and in the first-line with the combination of nivolumab plus ipilimumab (1, 33). Therefore, our findings should be considered only hypothesis generating and require additional validation in larger series with immune checkpoint inhibitors in earlier therapeutic settings.

In conclusion, SII, BMI, and older age of patients were independent prognostic factors in metastatic RCC treated with nivolumab. SII and BMI are routinely collected in clinical practice and thus deserve a potential role for use as prognostic indices, if these findings are confirmed by further studies. We also found that the cooccurrence of normal BMI and high inflammatory indexes identified patients with a more than 3-fold risk of mortality compared with patients with neither condition. A better understanding of biological mechanisms may help to guide interventions to optimize survival outcomes with nivolumab in metastatic RCC. Further studies are needed to clarify whether reducing systemic inflammation or possibly increasing low BMI with supportive care can enhance OS outcomes exploring the effectiveness of nivolumab in men and women. The mechanisms that are responsible for these findings will potentially be of great interest.

U. De Giorgi is a consultant/advisory board member for Bristol-Myers Squibb, Ipsen, Pfizer, Novartis, Merck, Astellas, Janssen, Sanofi, and Bayer. R. Sabbatini is a consultant/advisory board member for Bristol-Myers Squibb, Ipsen, Pfizer, Novartis, Astellas, Janssen, and Sanofi. U. Basso is a consultant/advisory board member for Janssen, Incyte, and MSD. P. Bidoli is a consultant/advisory board member for Eli Lilly, Bristol-Myers Squibb, and Boheringer. P. Marchetti reports receiving speakers bureau honoraria from Bristol-Myers Squibb and MSD, and is a consultant/advisory board member for Bristol-Myers Squibb, MSD, and Roche. C. Verusio reports receiving speakers bureau honoraria from Edra Edition. C. Sternberg reports receiving speakers bureau honoraria from Bristol-Myers Squibb, Novartis, Pfizer, and Ipsen, and is a consultant/advisory board member for Eisai, Pfizer, Ipsen, Bristol-Myers Squibb, Roche, Bayer, and MSD. No potential conflicts of interest were disclosed by the other authors.

The financial sponsor of the trial had no role in the design or conduct of the trial, data collection or analysis, and preparation of the article.

Conception and design: U. De Giorgi, G. Procopio, P. Bidoli, P. Marchetti, G. Cartenì, C.N. Sternberg

Development of methodology: U. De Giorgi, D. Giannarelli, L. Crinò, G. Fornarini, G. Cartenì, C.N. Sternberg

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): U. De Giorgi, R. Sabbatini, A. Bearz, S. Buti, U. Basso, M. Mitterer, C. Ortega, P. Bidoli, F. Ferraù, L. Crinò, A. Frassoldati, P. Marchetti, E. Mini, C. Verusio, G. Cartenì, C. Caserta, C.N. Sternberg

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): U. De Giorgi, D. Giannarelli, P. Bidoli, L. Crinò, A. Frassoldati

Writing, review, and/or revision of the manuscript: U. De Giorgi, D. Giannarelli, R. Sabbatini, S. Buti, U. Basso, M. Mitterer, P. Bidoli, A. Frassoldati, P. Marchetti, E. Mini, A. Scoppola, C. Caserta, C.N. Sternberg

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): U. De Giorgi, D. Giannarelli

Study supervision: U. De Giorgi, G. Procopio, G. Cartenì, C.N. Sternberg

We would like to thank the contributions of the study teams from the various sites in this trial [Italian Renal Cell Cancer Expanded Access Program (RCC EAP) group]. The authors thank the patients, their caregivers, and families who participated in this study. The authors gratefully acknowledge financial support by Bristol-Myers Squibb for data collection of the Italian RCC EAP.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Motzer
RJ
,
Escudier
B
,
McDermott
DF
,
George
S
,
Hammers
HJ
,
Srinivas
S
, et al
Nivolumab versus everolimus in advanced renal-cell carcinoma
.
N Engl J Med
2015
;
373
:
1803
13
.
2.
Choi
Y
,
Park
B
,
Jeong
BC
,
Seo
SI
,
Jeon
SS
,
Choi
HY
, et al
Body mass index and survival in patients with renal cell carcinoma: a clinical-based cohort and meta-analysis
.
Int J Cancer
2013
;
132
:
625
34
.
3.
Steffens
S
,
Grünwald
V
,
Ringe
KI
,
Seidel
C
,
Eggers
H
,
Schrader
M
, et al
Does obesity influence the prognosis of metastatic renal cell carcinoma in patients treated with vascular endothelial growth factor-targeted therapy?
Oncologist
2011
;
16
:
1565
71
.
4.
Albiges
L
,
Hakimi
AA
,
Xie
W
,
McKay
RR
,
Simantov
R
,
Lin
X
, et al
Body mass index and metastatic renal cell carcinoma: clinical and biological correlations
.
J Clin Oncol
2016
;
34
:
3655
63
.
5.
Templeton
AJ
,
McNamara
MG
,
Šeruga
B
,
Vera-Badillo
FE
,
Aneja
P
,
Ocaña
A
, et al
Prognostic role of neutrophil-to-lymphocyte ratio in solid tumors: a systematic review and meta-analysis
.
J Natl Cancer Inst
2014
;
106
:
dju124
.
6.
Keizman
D
,
Gottfried
M
,
Ish-Shalom
M
,
Maimon
N
,
Peer
A
,
Neumann
A
, et al
Pretreatment neutrophil-to-lymphocyte ratio in metastatic castration-resistant prostate cancer patients treated with ketoconazole: association with outcome and predictive nomogram
.
Oncologist
2012
;
17
:
1508
14
.
7.
Rossi
L
,
Santoni
M
,
Crabb
SJ
,
Scarpi
E
,
Burattini
L
,
Chau
C
, et al
High neutrophil-to-lymphocyte ratio persistent during first-line chemotherapy predicts poor clinical outcome in patients with advanced urothelial cancer
.
Ann Surg Oncol
2015
;
22
:
1377
84
.
8.
Passardi
A
,
Scarpi
E
,
Cavanna
L
,
Dall'Agata
M
,
Tassinari
D
,
Leo
S
, et al
Inflammatory indexes as predictors of prognosis and bevacizumab efficacy in patients with metastatic colorectal cancer
.
Oncotarget
2016
;
7
:
33210
9
.
9.
Keizman
D
,
Ish-Shalom
M
,
Huang
P
,
Eisenberger
MA
,
Pili
R
,
Hammers
H
, et al
The association of pre-treatment neutrophil to lymphocyte ratio with response rate, progression free survival and overall survival of patients treated with sunitinib for metastatic renal cell carcinoma
.
Eur J Cancer
2012
;
48
:
202
8
.
10.
Santoni
M
,
De Giorgi
U
,
Iacovelli
R
,
Conti
A
,
Burattini
L
,
Rossi
L
, et al
Pre-treatment neutrophil-to-lymphocyte ratio may be associated with the outcome in patients treated with everolimus for metastatic renal cell carcinoma
.
Br J Cancer
2013
;
109
:
1755
9
.
11.
De Giorgi
U
,
Rihawi
K
,
Aieta
M
,
Lo Re
G
,
Sava
T
,
Masini
C
, et al
Lymphopenia and clinical outcome of elderly patients treated with sunitinib for metastatic renal cell cancer
.
J Geriatr Oncol
2014
;
5
:
156
63
.
12.
Heng
DY
,
Xie
W
,
Regan
MM
,
Harshman
LC
,
Bjarnason
GA
,
Vaishampayan
UN
, et al
External validation and comparison with other models of the International Metastatic Renal-Cell Carcinoma Database Consortium prognostic model: a population-based study
.
Lancet Oncol
2013
;
14
:
141
8
.
13.
Hu
B
,
Yang
XR
,
Xu
Y
,
Sun
YF
,
Sun
C
,
Guo
W
, et al
Systemic immune inflammation index predicts prognosis of patients after curative resection for hepatocellular carcinoma
.
Clin Cancer Res
2014
;
20
:
6212
22
.
14.
Lolli
C
,
Caffo
O
,
Scarpi
E
,
Aieta
M
,
Conteduca
V
,
Maines
F
, et al
Systemic immune-inflammation index predicts the clinical outcome in patients with mCRPC treated with abiraterone
.
Front Pharmacol
2016
;
7
:
376
.
15.
Lolli
C
,
Basso
U
,
Derosa
L
,
Scarpi
E
,
Sava
T
,
Santoni
M
, et al
Systemic immune-inflammation index predicts the clinical outcome in patients with metastatic renal cell cancer treated with sunitinib
.
Oncotarget
2016
;
7
:
54564
71
.
16.
Kroenke
CH
,
Neugebauer
R
,
Meyerhardt
J
,
Prado
CM
,
Weltzien
E
,
Kwan
ML
, et al
Analysis of body mass index and mortality in patients with colorectal cancer using causal diagrams
.
JAMA Oncol
2016
;
2
:
1137
45
.
17.
Cespedes Feliciano
EM
,
Kroenke
CH
,
Meyerhardt
JA
,
Prado
CM
,
Bradshaw
PT
,
Kwan
ML
, et al
Association of systemic inflammation and sarcopenia with survival in nonmetastatic colorectal cancer: results from the C SCANS study
.
JAMA Oncol
2017
;
3
:
e172319
.
18.
Ho
PC
,
Liu
PS
. 
Metabolic communication in tumors: a new layer of immunoregulation for immune evasion
.
J Immunother Cancer
2016
;
4
:
4
.
19.
Biswas
SK
. 
Metabolic reprogramming of immune cells in cancer progression
.
Immunity
2015
;
43
:
435
49
.
20.
De Giorgi
U
,
Cartenì
G
,
Giannarelli
D
,
Basso
U
,
Galli
L
,
Cortesi
E
, et al
Safety and efficacy of nivolumab for metastatic renal cell carcinoma: real-world results from an expanded access program
.
BJU Int
2019
;
123
:
98
105
.
21.
Ko
JJ
,
Xie
W
,
Kroeger
N
,
Lee
JL
,
Rini
BI
,
Knox
JJ
, et al
The International Metastatic Renal Cell Carcinoma Database Consortium model as a prognostic tool in patients with metastatic renal cell carcinoma previously treated with first-line targeted therapy: a population-based study
.
Lancet Oncol
2015
;
16
:
293
300
.
22.
Templeton
AJ
,
Knox
JJ
,
Lin
X
,
Simantov
R
,
Xie
W
,
Lawrence
N
, et al
Change in neutrophil-to-lymphocyte ratio in response to targeted therapy for metastatic renal cell carcinoma as a prognosticator and biomarker of efficacy
.
Eur Urol
2016
;
70
:
358
64
.
23.
Bilen
MA
,
Dutcher
GMA
,
Liu
Y
,
Ravindranathan
D
,
Kissick
HT
,
Carthon
BC
, et al
Association between pretreatment neutrophil-to-lymphocyte ratio and outcome of patients with metastatic renal-cell carcinoma treated with nivolumab
.
Clin Genitourin Cancer
2018
;
16
:
e563
e75
.
24.
Lalani
AA
,
Xie
W
,
Martini
DJ
,
Steinharter
JA
,
Norton
CK
,
Krajewski
KM
, et al
Change in neutrophil-to-lymphocyte ratio (NLR) in response to immune checkpoint blockade for metastatic renal cell carcinoma
.
J Immunother Cancer
2018
;
6
:
5
.
25.
Kugel
CH
 III
,
Douglass
SM
,
Webster
MR
,
Kaur
A
,
Liu
Q
,
Yin
X
, et al
Age correlates with response to anti-PD1, reflecting age-related differences in intratumoral effector and regulatory T-cell populations
.
Clin Cancer Res
2018
;
24
:
5347
56
.
26.
McQuade
JL
,
Daniel
CR
,
Hess
KR
,
Mak
C
,
Wang
DY
,
Rai
RR
, et al
Association of body-mass index and outcomes in patients with metastatic melanoma treated with targeted therapy, immunotherapy, or chemotherapy: a retrospective, multicohort analysis
.
Lancet Oncol
2018
;
19
:
310
22
.
27.
Wang
Z
,
Aguilar
EG
,
Luna
JI
,
Dunai
C
,
Khuat
LT
,
Le
CT
, et al
Paradoxical effects of obesity on T cell function during tumor progression and PD-1 checkpoint blockade
.
Nat Med
2019
;
25
:
141
51
.
28.
Conforti
F
,
Pala
L
,
Bagnardi
V
,
De Pas
T
,
Martinetti
M
,
Viale
G
, et al
Cancer immunotherapy efficacy and patients' sex: a systematic review and meta-analysis
.
Lancet Oncol
2018
;
19
:
737
46
.
29.
Graham
J
,
Abdel-Rahman
O
,
Choueiri
TK
,
Heng
DYC
. 
Re: Fabio Conforti, Laura Pala, Vincenzo Bagnardi, et al. Cancer immunotherapy efficacy and patients' sex: a systematic review and meta-analysis. Lancet Oncol 2018;19:737–46: Outcomes of metastatic renal cell carcinoma by gender: contrasting results from the International mRCC Database Consortium
.
Eur Urol
2018
;
74
:
e139
e140
.
30.
Flegal
KM
. 
Body mass index of healthy men compared with healthy women in the United States
.
Int J Obes
2006
;
30
:
374
9
.
31.
Motzer
RJ
,
Rini
BI
,
McDermott
DF
,
Redman
BG
,
Kuzel
TM
,
Harrison
MR
, et al
Nivolumab for metastatic renal cell carcinoma: results of a randomized phase II trial
.
J Clin Oncol
2015
;
33
:
1430
7
.
32.
McDermott
DF
,
Drake
CG
,
Sznol
M
,
Choueiri
TK
,
Powderly
JD
,
Smith
DC
, et al
Survival, durable response, and long-term safety in patients with previously treated advanced renal cell carcinoma receiving nivolumab
.
J Clin Oncol
2015
;
33
:
2013
20
.
33.
Motzer
RJ
,
Tannir
NM
,
McDermott
DF
,
Arén Frontera
O
,
Melichar
B
,
Choueiri
TK
, et al
Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma
.
N Engl J Med
2018
;
378
:
1277
90
.