Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A multicenter study of body mass index in cancer patients treated with anti-PD-1/PD-L1 immune checkpoint inhibitors: when overweight becomes favorable

A multicenter study of body mass index in cancer patients treated with anti-PD-1/PD-L1 immune... Background: Recent evidence suggested a potential correlation between overweight and the efficacy of immune checkpoint inhibitors (ICIs) in cancer patients. Patients and methods: We conducted a retrospective study of advanced cancer patients consecutively treated with anti-PD-1/PD-L1 inhibitors, in order to compare clinical outcomes according to baseline BMI levels as primary analysis. Based on their BMI, patients were categorized into overweight/obese (≥ 25) and non-overweight (< 25). A gender analysis was also performed, using the same binomial cut-off. Further subgroup analyses were performed categorizing patients into underweight, normal weight, overweight and obese. Results: Between September 2013 and May 2018, 976 patients were evaluated. The median age was 68 years, male/female ratio was 663/313. Primary tumors were: NSCLC (65.1%), melanoma (18.7%), renal cell carcinoma (13.8%) and others (2.4%). ECOG-PS was ≥2 in 145 patients (14.9%). PD-1/PD-L1 inhibitors were administered as first-line treatment in 26.6% of cases. Median BMI was 24.9: 492 patients (50.6%) were non-overweight, 480 patients (50.4%) were overweight/obese. 25.2% of non-overweight patients experienced irAEs of any grade, while 55.6% of overweight/obese patients (p < 0.0001). ORR was significantly higher in overweight/obese patients compared to non-overweight (p < 0.0001). Median follow-up was 17.2 months. Median TTF, PFS and OS were significantly longer for overweight/obese patients in univariate (p < 0.0001, for all the survival intervals) and multivariate models (p = 0.0009, p < 0.0001 and p < 0.0001 respectively). The significance was confirmed in both sex, except for PFS in male patients (p = 0.0668). (Continued on next page) * Correspondence: alessiocortellini@gmail.com Medical Oncology, St. Salvatore Hospital, L’Aquila, Italy Department of Biotechnological and Applied Clinical Sciences, University of L’Aquila, L’Aquila, Italy Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 2 of 11 (Continued from previous page) Conclusions: Overweight could be considered a tumorigenic immune-dysfunction that could be effectively reversed by ICIs. BMI could be a useful predictive tool in clinical practice and a stratification factor in prospective clinical trials with ICIs. Keywords: BMI, Anti-PD-1/PD-L1, Overweight, Obesity, Cancer, Immunotherapy st Key message treatment with single agent anti-PD-1/PD-L1 as 1 or Recent evidence revealed that adipose tissue might affect subsequent line, at the medical oncology departments of the response to immune checkpoint inhibitors (ICIs) in 17 Italian centers (Additional file 1), between September cancer patients. In this retrospective transverse study, 2013 and May 2018. enrolling 976 advanced cancer patients treated with anti-PD-1/PD-L1 immunotherapy, we found a significant Anthropometric measurements association between overweight (BMI ≥ 25) and improved Weight and height were obtained from the patient’s clinical outcomes to ICIs. medical records at the time of immunotherapy initiation. BMI was calculated using the formula of weight/height Introduction (kilograms per square meter) and classified according to Although the interaction between malnutrition and chronic the World Health Organization (WHO) categories: inflammation has been widely investigated, whether this underweight, BMI < 18.5; normal, 18.5 ≤ BMI ≤ 24.9; association is causative or correlative is still debated [1]. overweight, 25 ≤ BMI ≤ 29.9; obesity, BMI ≥ 30. For the Historically, body mass index (BMI) has been considered study purpose, the binomial cut-off for BMI </≥ 25 was the major surrogate of nutritional status and its correlation used, and patients were categorized into non-overweight with clinical outcomes in advanced cancer patients has (< 25) and overweight/obese (≥ 25) for the final analysis. already been investigated without conclusive results [2–5]. Underweight patients were included in the non-over- It is now becoming clear that the nutritional assess- weight group. ment, which should include BMI, could be seen in a "new light" in the era of immune checkpoint inhibitors Study design (ICIs). A large retrospective study has recently found an We conducted a “real-life”, multicenter, retrospective association between BMI and improved progression free observational study aimed at comparing the clinical out- survival (PFS) and overall survival (OS) in melanoma comes of cancer patients treated with ICIs according to patients treated with either targeted therapy or immuno- baseline BMI levels. therapy [6]. Another study has reported that overweight Primary outcomes measures were: objective response sarcopenic melanoma patients treated with anti-PD1 rate (ORR), time to treatment failure (TTF), PFS and (Programmed cell death protein 1) inhibitors expe- OS. ORR was defined as the proportion of patients rienced early acute limiting toxicity [7]. experiencing an objective response (either complete re- Additionally, another retrospective analysis by Richtig et sponse or partial response) as best response to immuno- al. revealed that overweight (BMI ≥ 25) melanoma patients therapy. TTF was defined as the time from treatment’s (76 total) treated with ipilimumab had significantly higher start to discontinuation for any reason. Progression-free response rate (p=0.024) andatrendforlongerOS (p= survival (PFS) was defined as the time from the start of 0.056), when compared to non-overweight patients [8]. immunotherapy to the date of disease progression or Lastly, Wang and colleagues have recently reported death, whichever occurred first. Patients who were alive an improvement in terms of PFS (p = 0.003) and OS without disease progression were censored on the date (p = 0.049) in a cohort of obese advanced cancer of their last disease assessment. Overall survival (OS) patients (BMI ≥ 30) treated with ICIs [9]. was defined as the time from the start of immunothe- To further dissect this question, we conducted a large, rapy to death. Patients who were still alive were cen- multicentre, retrospective transverse study to evaluate sored at the date of last contact. Patients were treated clinical outcomes of patients with advanced solid tumors according to the tumor type indication with pembrolizu- treated with ICIs according to baseline BMI. mab, nivolumab or atezolizumab with standard doses and schedules. Materials and methods In order to weighing the possible prognostic influence Patient eligibility of obesity (30 BMI) and malnutrition (or cachexia), two This study enrolled patients with confirmed diagnosis of subgroup analysis (according to each BMI categories) measurable stage IV cancer, who consecutively underwent were performed. In the first one, overweight (25-30 BMI) Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 3 of 11 and obese (≥ 30 BMI) patients were respectively com- Table 1 Patients’ characteristics pared to non-overweight (< 25) patients, in the second N° (%) one overweight (25-30 BMI) and obese (≥ 30 BMI) patients were respectively compared to normal weight AGE, (years) patients (18.5-25 BMI). Median 68 A subgroup analysis comparing clinical outcomes in Range 24–92 males and females patients, using the binomial cut-off Elderly (≥ 70) 445 (45.6) (BMI </≥ 25) was also conducted as secondary analysis. The following covariates were considered for the SEX multivariate analyses: primary tumor (NSCLC, mela- Male 663 (67.9) noma, kidney and others), sex (male vs female), Eastern Female 313 (32.1) Cooperative Oncology Group Performance Status ECOG PS (ECOG-PS) (0-1 vs. ≥ 2), age (< 70 vs. ≥ 70 years old) 0–1 831 (85.1) [10–13], number of metastatic sites (≤ 2 vs. > 2) and ≥ 2 145 (14.9) treatment line (first vs non-first). As in some indications the anti-PD-1/PD-L1 agents dosages had been weight- Primary Tumor based, weight was used as a continuous covariate in all NSCLC 635 (65.1) the analyses, considering the possible dose-depending Melanoma 183 (18.7) confounding effect on the clinical outcomes. Renal cell carcinoma 135 (13.8) Immune-related AEs (irAEs) were graduated according to Others 23 (2.4) theCommonToxicity Criteriafor AdverseEvents (CTCAE; No. of metastatic sites version 4.0) and cumulatively reported. Immune-related AEs were categorized on the basis of the organ/system ≤ 2 467 (47.9) involved as follows: endocrine irAEs (including thyroid > 2 509 (52.1) disorders), gastro-intestinal (GI) irAEs (excluding pan- Type of anti-PD-1/PD-L1 agent creatitis), skin irAEs, pneumological irAEs, hepatic Pembrolizumab 235 (24.1) irAEs, rheumatologic irAEs and others irAEs (including Nivolumab 706 (72.3) neuro-muscolar, pancreatitis, fever, asthenia and an- Atezolizumab 35 (3.6) orexia). The safety analysis was performed for irAEs of any grade and for G3/G4 irAEs. Treatment line of Immunotherapy To determine ORR and PFS, scans were reviewed by a First 260 (26.6) dedicated thoracic oncologist at each Institution using Non-First 716 (73.4) Response Evaluation Criteria In Solid Tumors (RECIST) Weight (Kg) version 1.1. [14]. χ2 was used to compare ORR and inci- Median 71 dence of irAEs among subgroups [15]. In the multivari- Range 35–139 ate analysis, logistic regression was used to evaluate the role of parameters proven to be significant at the univa- BMI (kg/m ) riate analysis of ORR [16]. Median TTF, median PFS, Median (range) 24.9 (13.5–46.6) and median OS were evaluated using the Kaplan-Meier Underweight (BMI ≤ 18.5), n°(%) 40 (4.1) method [17]. Median follow-up was calculated according Normal weight (BMI 18.5 < BMI ≤ 24.9), n°(%) 452 (46.3) to the reverse Kaplan-Meier method [18]. Cox propor- Overweight (25 < BMI ≤ 29.9), n°(%) 377 (38.6) tional hazards model [19] was used to evaluate predictor Obese (BMI ≥ 30), n° (%) 107 (11) variables in univariate and multivariate analysis for TTF, th PFS and OS. The data cut-off was October 29 , 2018. All statistical analyses were performed using MedCalc tumors were: NSCLC (635 patients), melanoma (183 Statistical Software version 18.6 (MedCalc Software patients), renal cell carcinoma (135 patients) and others bvba, Ostend, Belgium; http://www.medcalc.org; 2018). (23 patients). ECOG-PS was 0/1 in 831 patients (85.1%), and ≥ 2 in 145 patients (14.9%); 467 patients (47.9%) had Results ≤ 2 metastatic sites while 509 (52.1%) had more than 2 Patient characteristics metastatic sites. PD-1/PD-L1 inhibitors were adminis- Nine hundred and seventy-six, consecutive advanced tered as first-line treatment in 260 patients (26.6%). cancer patients were evaluated. Patient characteristics Median weight was 71 Kg, median BMI was 24.9; are summarized in Table 1. The median age was 68 years according to WHO classification 40 patients (4.1%) were (range: 24 – 92), male/female ratio was 663/313. Primary defined as underweight, 452 patients (46.3%) as having a Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 4 of 11 normal weight, 377 patients (38.6%) as overweight and Efficacy analysis 107 patients (11%) as obese. For the study purpose, 492 At median follow-up of 17.2 months, median TTF was patients were considered as non-overweight (50.4%) and 5.9 months (95% CI: 5.3 – 6.7; 681 events), median PFS 484 patients were categorized as overweight/obese was 6.5 months (95% CI: 6.1 – 7.1; 644 events) and median (49.6%) according to a BMI cut-off of 25 (<25 vs. ≥25). OS was 13.4 months (95% CI: 11.0 – 16.5; 488 censored Among male patients median age was 69 years, median patients) in the entire cohort. weight was 72 Kg (range: 35 – 139) and median BMI was When these outcomes where analyzed according to 24.8 (range: 14 – 46.6). Among female patients median BMI, we found that median TTF was significantly longer age was 67, median weight was 70 Kg (range: 40 – 130) in overweight/obese patients compared to non-overweight and median BMI was 25.4 (range: 13.6 – 46.1). patients (9.3 [95% CI: 8.1 – 11.6; 318 events] vs. 3.6 months [95% CI: 3.2 – 4.1; 363 events]; HR= 0.51 [95% CI: 0.44 – 0.60], p < 0.0001) (Fig. 1a). Similarly, Safety analysis median PFS was significantly improved in the over- In the entire cohort, 393 patients (40.3%) experienced weight/obese group compared to the non-overweight irAEs of any grade. Sixty-three patients (6.5%) expe- group (11.7 months [95% CI: 9.4 – 15; 286 events] vs. rienced G3/G4 irAEs. Overweight/obese patients were 3.7 months [95% CI: 3.2 – 4.1; 358 events]; HR= 0.46 significantly more likely to experience any grade irAEs [95%CI: 0.39 – 0.54], p < 0.0001) (Fig. 1b). Consistently compared to non-overweight patients (55.6% vs. 25.2%, we also found a significantly prolonged median OS among p < 0.0001). However, no difference in the rate of G3/G4 overweight/obese patients compared to non-over irAEs was observed between Overweight/obese patients weight patients (26.6 months [95% CI: 21.4 – 36.8; 286 and non-overweight patients (7.6 vs. 5.3%, p = 0.1338). censored patients] vs. 6.6 months [95% CI: 5.8 – 8.5; The safety profile of ICIs according to BMI is summa- 182 censored patients]; HR= 0.33 [95%CI: 0.28 – 0.41], rized in Additional file 2. p < 0.0001) (Fig. 1c). After adjusting for PS, treatment line, n° of metastatic sites, gender, primary tumor subtype and development Activity analysis of irAEs, a BMI of ≥25 retained a significant association Univariate and multivariate analyses for ORR are de- with a longer TTF (p = 0.0009), PFS (p < 0.0001) and tailed in Additional file 3. Among 910 patients evaluable OS (p < 0.0001) in multivariate models (Table 2, Table 3, for activity, 283 patients had a response to ICIs (ORR: Table 4) 31.1%). Overweight/obese patients had a significantly higher ORR compared non-overweight patients (41.3% vs. 20.9%, p < 0.0001). Similarly, we found a significantly Subgroup analyses higher ORR among patients who experienced at least Table 5 reports the univariate and multivariate gender ana- 1 irAE compared to those without irAEs (45.1% vs. lyses for TTF, PFS and OS of male patients (Table 5A) and 21.1%, p < 0.0001). Both BMI (overweight/obese vs. female patients (Table 5B). As shown overweight/obese non-overweight) and the development of irAEs of any male patients had significantly longer TTF (p = 0.0330) and grade, were independently associate with higher ORR OS (p = 0.0013), but not PFS (p = 0.0668), when compared in the multivariate analysis (p = 0.0239 and p < 0.0001, with non-overweight patients, while overweight/obese fe- respectively). male patients had significantly longer TTF (p = 0.0037), AB C Fig. 1 Kaplan-Meier survival curves according to binomial BMI levels (cut-off 25). (a) Time to Treatment Failure. BMI < 25: 3.6 months (95% CI: 3.2–4.1); BMI ≥ 25: 9.3 months (95%CI: 8.1–11.6). (b) Progression Free Survival. BMI < 25: 3.7 months (95% CI: 3.2–4.1); BMI ≥ 25: 11.7 months (95% CI: 9.4–15). (C) Overall Survival. BMI < 25: 6.6 months (95% CI: 5.8–8.5); BMI ≥ 25: 26.6 months (95% CI: 21.4–36.8) Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 5 of 11 Table 2 Cox proportional-hazards regression: univariate and multivariate analyses of Time to Treatment Failure Time to Treatment Failure Univariate Analysis Multivariate Analysis VARIABLE (Comparator) HR (95% CI); p - value HR (95% CI); p - value BMI 0.51 (0.44–0.60); p < 0.0001 0.67 (0.53–0.85); p = 0.0009 ≥ 25 vs < 25 Weight 0.98 (0.97–0.99); p < 0.0001 0.99 (0.98–1.01); p = 0.8422 irAEs of any grade 0.57 (0.48–0.66); p < 0.0001 0.79 (0.65–0.97); p = 0.0295 Yes vs No Primary Tumor (NSCLC) Melanoma 0.62 (0.50–0.76); p < 0.0001 0.79 (0.64–1.01); p = 0.0517 Kidney 0.73 (0.59–0.92); p = 0.0077 0.71 (0.56–0.88); p = 0.0025 Others 1.15 (0.71–1.87); p = 0.5560 0.78 (0.48–1.28); p = 0.3389 Sex 1.22 (1.04–1.43); p = 0.0147 1.10 (0.93–1.30); p = 0.2607 Male vs Female Age 1.04 (0.90–1.21); p = 0.5366 – Elderly vs Non-elderly Treatment line 1.36 (1.13–1.64); p = 0.0008 1.51 (1.25–1.81); p < 0.0001 Non-first vs First N° of metastatic sites 1.54 (1.34–1.77); p < 0.0001 1.52 (1.30–1.77); p < 0.0001 >2 vs ≤ 2 ECOG PS 2.86 (2.36–3.48); p < 0.0001 2.35 (1.92–2.88); p < 0.0001 ≥2vs0–1 Weight was used as a continuous variable Table 3 Cox proportional-hazards regression: univariate and multivariate analyses of Progression Free Survival Progression Free Survival Univariate Analysis Multivariate Analysis VARIABLE (Comparator) HR (95% CI); p - value HR (95% CI); p - value BMI 0.46 (0.39–0.54); p < 0.0001 0.71 (0.56–0.90); p < 0.0001 ≥ 25 vs < 25 Weight 0.97 (0.96–0.98); p < 0.0001 0.99 (0.98–1.01); p = 0.1580 irAEs of any grade 0.48 (0.41–0.57); p < 0.0001 0.67 (0.54–0.83); p = 0.0002 Yes vs No Primary Tumor (NSCLC) Melanoma 0.52 (0.42–0.66); p < 0.0001 0.67 (0.53–0.85); p = 0.0008 Kidney 0.72 (0.58–0.91); p = 0.0062 0.67 (0.53–0.84); p = 0.0008 Others 1.08 (0.65–1.78); p = 0.7556 0.69 (0.41–1.15); p = 0.1533 Sex 1.20 (1.01–1.42); p = 0.0314 1.03 (0.86–1.22); p = 0.7252 Male vs Female Age 0.96 (0.82–1.12); p = 0.6394 – Elderly vs Non-elderly Treatment line 1.62 (1.33–1.96); p < 0.0001 1.61 (1.32–1.93); p < 0.0001 Non-first vs First N° of metastatic sites 1.46 (1.27–1.68); p < 0.0001 1.42 (1.21–1.67); p < 0.0001 >2 vs ≤ 2 ECOG PS 2.60 (2.13–3.17); p < 0.0001 2.06 (1.67–2.52); p < 0.0001 ≥2vs 0–1 Weight was used as a continuous variable Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 6 of 11 Table 4 Cox proportional-hazards regression: univariate and multivariate analyses of Overall Survival Overall Survival Univariate Analysis Multivariate Analysis VARIABLE HR (95% CI); p – value HR (95% CI); p - value (Comparator) BMI 0.33 (0.28–0.41); p < 0.0001 0.49 (0.38–0.64); p < 0.0001 ≥ 25 vs < 25 Weight 0.97 (0.96–0.97); p < 0.0001 0.99 (0.99–1.01); p = 0.1884 irAEs of any grade 0.45 (0.37–0.54); p < 0.0001 0.82 (0.65–1.04); p = 0.1085 Yes vs No Primary Tumor (NSCLC) Melanoma 0.49 (0.38–0.64); p < 0.0001 0.67 (0.51–0.87); p = 0.0036 Kidney 0.56 (0.42–0.74); p = 0.0001 0.61 (0.45–0.80); p = 0.0005 Others 1.11 (0.62–1.96); p = 0.7337 0.71 (0.40–1.28); p = 0.2632 Sex 1.50 (1.23–1.83); p < 0.0001 1.33 (1.09–1.63); p = 0.0044 Male vs Female Age 1.11 (0.93–1.32); p = 0.2401 – Elderly vs Non-elderly Treatment line 1.58 (1.26–1.97); p = 0.0001 1.42 (1.15–1.77); p = 0.0012 Non-first vs First N° of metastatic sites 1.52 (1.29–1.78); p < 0.0001 1.41 (1.17–1.69); p = 0.0002 >2 vs ≤ 2 ECOG PS 2.07 (1.87–2.29); p < 0.0001 2.59 (2.09–3.21); p < 0.0001 ≥2vs 0–1 Weight was used as a continuous variable PFS (p = 0.0132) and OS (p < 0.0001), when compared to also found a significant prolonged median OS among nor- non-overweight patients. mal weight compared to underweight patients (7.9 months Median TTF was not significantly different between [95%CI: 6.4 – 9.8; 178 censored patients] vs. 2.8 months overweight and obese patients (10.3 months [95%CI: [95%CI: 1.8 – 3.6; 4 censored patients], HR= 0.33 [95%CI: 8.2 – 4.1; 238 events] vs. 7.3 [95%CI: 5.5 – 11.7; 80 0.23 – 0.48], p < 0.0001). Table 7 reports the univariate and events], HR=1.23 [95%CI: 0.95 – 1.58], p = 0.1087). multivariate analyses of TTF, PFS and OS, comparing Similarly, we found no significant differences in median overweight (non-obese) patients and obese patients PFS (11.2 months [95%CI: 9.1 – 15.6; 223 events pa- with normal weight patients. Figure 3 reports the tients] vs. 12.9 months [95%CI: 7.1 – 18; 63 events], Kaplan-Meier survival curves of obese, overweight and HR=0.99 [95%CI: 0.75 – 1.31], p = 0.9798) and median normal weight patients. OS (26.6 months [95%CI: 21.4 – 36.8; 223 censored patients] vs. not reached [63 censored patients], Discussion HR=1.04 [95%CI: 0.75 – 1.46], p = 0.7767) between over- In this study we demonstrated that patients with a weight and obese patients. Table 6 reports the univariate BMI ≥ 25 experienced a better clinical outcome com- and multivariate analyses of TTF, PFS and OS, comparing pared to those with a BMI < 25. Recently, the association overweight (non-obese) patients and obese patient with between BMI and OS of metastatic renal cell carcinoma non-overweight patients. Figure 2 reports the Kaplan-Meier patients, has been reported regardless of the use of survival curves of obese, overweight and non-over- anti-PD-1/PD-L1 therapy [4, 20]. However, in our study weight patients. we found a strong correlation between overweight and When we analyzed the clinical outcomes of normal improved clinical outcomes with anti-PD-1/PD-L1. weight vs. underweight patients, we found a significantly Some authors have already speculated about the nega- longer median TTF (3.9 months [95%CI: 3.4 – 5.0; 327 tive impact of body composition alteration on immune events] vs. 1.8 [95%CI: 1.7 – 2.9; 36 events], HR= 0.51 cells activity [21]. Interestingly, it has been increasingly [95%CI: 0.35 – 0.71], p = 0.0001 and median PFS recognized that white adipose tissue, which is the most (4.4 months [95%CI: 3.6 – 5.3; 322 events] vs. 1.9 related to the fattening process [22], is also involved in months [95%CI: 1.7 – 2.9; 36 events] HR= 0.45; the induction and/or coordination of host defenses, 95%CI: 0.32 – 0.64], p < 0.0001) in normal weight being a source of cytokines and chemokines [23]. In fact, patients compared with underweight patients. We adipose tissue modulates the Th1/Th2 balance, decreases Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 7 of 11 Table 5 Cox proportional-hazards regression: univariate and Table 7 Cox proportional-hazards regression: univariate and multivariate analyses multivariate analyses according to normal weight (18.5-25), overweight (25-30) and obese (≥ 30) BMI levels A Univariate Analysis Multivariate Analysis Univariate Analysis Multivariate Analysis VARIABLE HR (95% CI); p - value HR (95% CI); p - value VARIABLE HR (95% CI); p - value HR (95% CI); p - value Time to Treatment Failure (Comparator) BMI 0.54 (0.45–0.66); p < 0.0001 0.74 (0.56–0.97); p = 0.0330 BMI Time to Treatment Failure ≥ 25 vs < 25 (18.5–25) Progression Free Survival 25–30 0.51 (0.43–0.61); p < 0.0001 0.65 (0.51–0.82); p = 0.0004 BMI 0.49 (0.40–0.59); p < 0.0001 0.77 (0.58–1.01); p = 0.0668 ≥ 30 0.63 (0.49–0.81); p = 0.0003 0.79 (0.55–1.15); p = 0.2300 ≥ 25 vs < 25 BMI Progression Free Survival Overall Survival (18.5–25) BMI 0.38 (0.31–0.48); p < 0.0001 0.59 (0.43–0.81); p = 0.0013 25–30 0.49 (0.41–0.58); p < 0.0001 0.68 (0.53–0.87); p = 0.0016 ≥ 25 vs < 25 ≥ 30 0.48 (0.37–0.64); p < 0.0001 0.72 (0.49–1.06); p = 0.0991 BMI Overall Survival B Univariate Analysis Multiavariate Analysis (18.5–25) VARIABLE HR (95% CI); p - value HR (95% CI); p - value 25–30 0.35 (0.29–0.43); p < 0.0001 0.46 (0.35–0.61); p < 0.0001 ≥ 30 0.37 (0.27–0.51); p < 0.0001 0.50 (0.32–0.79); p = 0.0029 Time to Treatment Failure The used covariates (not shown) were: weight (continuous), irAEs of any grade, BMI 0.45 (0.35–0.61); p < 0.0001 0.51 (0.32–0.80); p = 0.0037 primary tumors, sex, line of treatment, ECOG-PS, number of metastatic sites ≥ 25 vs < 25 Progression Free Survival the activation of Treg through adiponectin, increases BMI 0.41 (0.31–0.56); p < 0.0001 0.56 (0.35–0.88); p = 0.0132 pro-inflammatory macrophages, activates T-cells with ≥ 25 vs < 25 the binding between LIGHT-HVEM (herpesvirus entry Overall Survival mediator) and increases the inflammatory status through BMI 0.25 (0.17–0.36); p < 0.0001 0.27 (0.15–0.48); p < 0.0001 CD40 pathway [24–26]. ≥ 25 vs < 25 Moreover, a recent preclinical study revealed that (A) male patients (B) female patients. The used covariates (not shown) were: white adipose tissue might also play a role in immune weight (continuous), irAEs of any grade, primary tumors, line of treatment, homeostasis [27]. In this study, white adipose tissue of ECOG-PS, number of metastatic sites mice was reported to accumulate pathogen-specific memory T-cells after a microbial infection, including tissue-resident cells expressing a distinct metabolic pro- file. Intriguingly, these data support the hypothesis that adipose tissue can act as a reservoir of tissue-specific memory T-cells, which can undergo a rapid response to reactivation against exogenous stimuli. This evidence Table 6 Cox proportional-hazards regression: univariate and raises an interesting question, can these adipose tissue- multivariate analyses according to non-overweight (< 25), specific T-cells be promptly reactivated against cancer- overweight (25-30) and obese (≥ 30) BMI levels specific antigens as they do against microbial antigens? Univariate Analysis Multivariate Analysis In a recent meta-analysis of patients with immune-me- VARIABLE HR (95% CI); p - value HR (95% CI); p - value diated inflammatory diseases treated with anti-TNF (Comparator) (tumor necrosis factor), the authors reported a trend BMI Time to Treatment Failure towards a lower response rate to treatment among over- (< 25) weight patients [28]. This is likely to reflect the reduced 25–30 0.49 (0.41–0.58); p < 0.0001 0.67 (0.53–0.84); p = 0.0008 responsivity of T-cells of obese individuals, which has ≥ 30 0.60 (0.47–0.77); p = 0.0001 0.86 (0.61–1.24); p = 0.4414 also been confirmed in preclinical models showing a BMI Progression Free Survival significant increase in dysfunctional exhausted T-cells (< 25) in obese mice [9]. Nevertheless, such inflamed and 25–30 0.46 (0.39–0.55); p < 0.0001 0.71 (0.56–0.89); p = 0.0044 immune-exhausted status may be more likely suscep- ≥ 30 0.46 (0.35–0.61); p < 0.0001 0.80 (0.55–1.17); p = 0.2669 tible to the immune checkpoint blockade. In support of BMI Overall Survival this, in preclinical models, T-cell dysfunction in obese (< 25) mice was proven to be partly mediated by the PD-1 axis 25–30 0.33 (0.27–0.41); p < 0.0001 0.49 (0.37–0.64); p < 0.0001 and driven by leptin, strengthening the already known ≥ 30 0.34 (0.25–0.48); p < 0.0001 0.61 (0.39–0.94); p = 0.0258 correlation between JAK/STAT pathway and immune The used covariates (not shown) were: weight (continuous), irAEs of any grade, primary tumors, sex, line of treatment, ECOG-PS, number of metastatic sites checkpoint inhibition [9, 29]. Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 8 of 11 ABC Fig. 2 Kaplan-Meier survival curves according to BMI levels (non-overweight BMI < 25, overweight BMI 25–30, obese BMI ≥ 30). (a) Time to Treatment Failure. BMI < 25: 3.6 months (95% CI: 3.2–4.1); BMI 25–30: 10.3 months (95%CI: 8.2–4.1); BMI ≥ 30: 7.3 months (95%CI: 5.5–11.7). (b) Progression Free Survival. BMI < 25: 3.7 months (95% CI: 3.2–4.1); BMI 25–30: 11.2 months (95%CI: 9.1–15.6); BMI ≥ 30: 12.9 months (95%CI: 7.1–18). (c) Overall Survival. BMI < 25: 6.6 months (95% CI: 5.8–8.5); BMI 25–30: 26.6 months (95%CI: 21.4–36.8); BMI ≥ 30: not reached Importantly, in our study we also found a significantly analysis, compared to obese patients, thus supporting the higher incidence of irAEs of any grade among overweight/ hypothesis that the prognostic weight of obesity, could obese patients. In light of the emerging association be- have partially influenced the final results. tween the development of irAEs and improved clinical out- On the other hand, despite the small sample size (4.1% comes with ICIs across different tumor types, our findings of the entire population), underweight patients had are not unexpected [30–35]. In our cohort, the develop- significantly shorter TTF, PFS and OS, when compared ment of irAEs of any grade was independently associated to normal weight patients, confirming that malnutrition with improved clinical outcomes along with a BMI ≥25 in (and cachexia) is an independent negative prognostic multivariate analyses. factor. Nonetheless, when we compared obese and over- The analysis performed by separating overweight and weight patients (Table 7)with normal weightpatientswe obese patients, demonstrated that a linear relationship be- observed significantly improved clinical outcomes favou- tween BMI and positive outcomes cannot be assumed. ring the overweight group, which suggests that over- Even though we found no statistically significant diffe- weightness has a direct impact on the efficacy of ICIs. rences in TTF, PFS and OS between overweight and obese In our study we also carried a gender-based analysis. Pre- patients, when separately comparing obese patients to viously, it has been reported that female patients tend to non-overweight patients (Table 6), we observed the loss of have lower benefit from ICI compared to males [36, 37]. significance regarding TTF and PFS, while not regarding However, whether the gender plays a key role in determi- OS. This result is of particular interest, considering the ning the clinical outcome to immunotherapy is still in need possibility of a negative impact on survival of obese pa- of further investigation. In our study, we found that over- tients due to cardiovascular and metabolic complications weight female patients derived a greater clinical benefit of obesity itself. Noteworthy, the HRs are concordantly form immunotherapy as compared to the male counterpart lower for overweight (non-obese) patients in each survival (Table 5). However, it should be highlighted that ABC Fig. 3 Kaplan-Meier survival curves according to BMI levels (normal weight BMI 18.5–25, overweight BMI 25–30, obese BMI ≥ 30). (a) Time to Treatment Failure. BMI 18.5–25: 3.9 months (95% CI: 3.4–5.0); BMI 25–30: 10.3 months (95%CI: 8.2–4.1); BMI ≥ 30: 7.3 months (95%CI: 5.5–11.7). (b) Progression Free Survival. BMI 18.5–25: 4.4 months (95% CI: 3.6–5.3); BMI 25–30: 11.2 months (95%CI: 9.1–15.6); BMI ≥ 30: 12.9 months (95%CI: 7.1–18). (c) Overall Survival. BMI 18.5–25: 7.9 months (95% CI: 6.4–9.8); BMI 25–30: 26.6 months (95%CI: 21.4–36.8); BMI ≥ 30: not reached Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 9 of 11 overweightness was associated with improved outcomes in Acknowledgements This work was supported by the Consorzio Interuniversitario Nazionale per la both males and females in the multivariate analysis. This Bio-Oncologia (CINBO). led us to speculate that in our population the predictive role of BMI was to be stronger than the predictive role of Funding No funding was received. gender. Certainly, the relationship between sex, adipose tissue Availability of data and materials and immunity is complex and ambiguous. Sex-hormones, The datasets used during the present study are available from the in particular estrogens, could affect adipose tissue func- corresponding author upon reasonable request. tions [38], but in some respects their influence on the im- Authors’ contributions mune systems does not seem unidirectional [39]. All authors contributed to the publication according to the ICMJE guidelines Furthermore, the median age of female patients in our for the authorship. All authors read and approved the manuscript and agree to be accountable for all aspects of the research in ensuring that the study was 67, indicating a prevalence of postmenopausal accuracy or integrity of any part of the work are appropriately investigated patients. In this specific population the adipose tissue and resolved. becomes a major source of circulating estrogens [40]. Ethics approval and consent to participate Another way to explain how BMI might affect sex All patients provided written, informed consent to treatment with hormones levels and the immune response, is through immunotherapy. All patients alive at the time of data collection provided an diet regimens which underpin the weight gain. Indeed, informed consent for the present retrospective analysis. The procedures followed were in accordance with the precepts of Good Clinical Practice and gut microbiota may be influenced by the different the declaration of Helsinki. The study was approved by the respective local "modifying pressures" of various diet types. Interestingly, ethical committees on human experimentation of each institution, after males and females have recently been reported to have previous approval by the coordinating center (University of L’Aquila, Internal th Review Board protocol number 32865, approved on July 24 , 2018). gender-specific differences in their immune system and gut microbiota composition. Whether these differences in Consent for publication gut microbiota composition might impact the efficacy or Not applicable. the safety profile of immunotherapy is subject of intense Competing interests research and is expected to provide us further insight in the Dr Alessio Cortellini received grants as speaker by MSD, Astra-Zeneca and optimal management of our patients [41–43]. Boehringer Ingelheim, gran consultancies by BMS and Ipsen; dr Marcello Our study is certainly flawed by several caveats, Tiseo received grant as speaker and advisory role by Astra-Zeneca, Pfizer, Eli- Lilly, BMS, Novartis, Roche, MSD, Boehringer Ingelheim, Otsuka, Pierre Fabre; including the retrospective design with the risk of selec- dr Maria Giuseppa Vitale received travel grants and speaker fees by BMS, tion and data collection biases, the heterogeneity of the Ipsen Astellas, Jansen, Novartis and Pfizer; dr Sebatiano Buti received grants analyzed population, the lack of a centralized imaging as speaker and advisory role by BMS, Pfizer, MSD, Ipsen, Novartis, Astra- Zeneca; dr. Melissa Bersanelli received honoraria as speaker at scientific review for response assessment and the lack of data events and as consultant for advisory role by BMS and Pfizer. about patient comorbidities. In addition, the lack of con- trol group of patients who did not received ICIs further Publisher’sNote limit the power of our analysis. On the other hand, a Springer Nature remains neutral with regard to jurisdictional claims in unique strength of our study is that we evaluated the published maps and institutional affiliations. predictive role of baseline assessment of BMI in a “real Author details life” population of individuals candidate to receive ICIs. 1 2 Medical Oncology, St. Salvatore Hospital, L’Aquila, Italy. Department of Biotechnological and Applied Clinical Sciences, University of L’Aquila, Conclusion L’Aquila, Italy. Medical Oncology, University Hospital of Parma, Parma, Italy. Department of Medicine and Surgery, University of Parma, Parma, Italy. In this study we demonstrated that patients with a 5 6 Medical Oncology, Campus Bio-Medico University, Rome, Italy. Department BMI ≥ 25 experienced better clinical outcomes with of Clinical and Molecular Medicine, Sant’Andrea Hospital, Sapienza University anti-PD-1/PD-L1 agents, compared to those with a of Rome, Rome, Italy. Clinical Oncology Unit, S.S. Annunziata Hospital, Chieti, Italy. Department of Medical, Oral & Biotechnological Sciences University G. BMI < 25. Our results suggest that BMI could be a useful D’Annunzio, Chieti-Pescara, Italy. Medical Oncology, Santa Maria Goretti predictive tool in clinical practice as well as a reliable strati- 10 Hospital, Latina, Italy. Department of Oncology, University Hospital of Pisa, fication variable for prospective clinical trials with ICIs. Istituto Toscano Tumori, Pisa, Italy. Medical Oncology, “Santo Spirito” Hospital, Pescara, Italy. Istituto Dermopatico dell’Immacolata, IDI-IRCCS, Rome, Italy. Medical Oncology, University Hospital of Modena, Modena, 14 15 Italy. Medical Oncology, Fermo Area Vasta 4, Fermo, Italy. Oncology Clinic, Additional files Università Politecnica delle Marche, Ospedali Riuniti di Ancona, Ancona, Italy. 16 17 Medical Oncology, AV2 Fabriano ASUR Marche, Pescara, Italy. Medical Additional file 1: List of oncological institutions of the study. 18 Oncology Unit, University Hospital of Cagliari, Cagliari, Italy. Medical (DOCX 15 kb) 19 Oncology, Santa Maria della Misericordia Hospital, Perugia, Italy. Pulmonary Additional file 2: Immune-related adverse events of any grade and Oncology Unit, St. Camillo Forlanini Hospital, Rome, Italy. Medical Oncology G3/G4 immune-related adverse events. (DOCX 15 kb) (B), Policlinico Umberto I, “Sapienza” University of Rome, Rome, Italy. 21 22 Dermatology, San Salvatore Hospital, L’Aquila, Italy. Melanoma, Cancer Additional file 3: Univariate and multivariate analyses with logistic Immunotherapy and Development Therapeutics Unit, Istituto Nazionale regression of Objective Response Rate. (DOC 49 kb) Tumori-IRCCS Fondazione “G. Pascale”, Naples, Italy. Medical Oncology Unit, Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 10 of 11 St. Salvatore Hospital, Department of Biotechnological and Applied Clinical Physiol Endocrinol Metab. 2008;295(5):E1076–83. https://doi.org/10.1152/ Sciences, University of L’Aquila, Via Vetoio, 67100 L’Aquila, Italy. ajpendo.90408.2008 Epub 2008 Sep 9. 23. Ouchi N, Parker JL, Lugus JJ, et al. Adipokines in inflammation and Received: 4 December 2018 Accepted: 5 February 2019 metabolic disease. Nat Rev Immunol. 2011;11(2):85–97. https://doi.org/10. 1038/nri2921. Epub 2011 Jan. 24. Jordan BF, Gourgue F, Cani PD. Adipose Tissue metabolism and Cancer progression: novel insights from gut microbiota? Curr References Pathobiol Rep. 2017;5(4):315–22. https://doi.org/10.1007/s40139-017- 1. Laviano A, Koverech A, Mari A. Cachexia: clinical features when 0154-6 Epub 2017 Oct 24. inflammation drives malnutrition. Proc Nutr Soc. 2015 Nov;74(4):348–54. 25. Magrone T, Jirillo E. Childhood obesity: immune response and nutritional 2. Taghizadeh N, Boezen HM, Schouten, et al. BMI and lifetime changes in BMI approaches. Front Immunol. 2015;6:76. https://doi.org/10.3389/fimmu.2015. and cancer mortality risk. PLoS One. 2015;10(4):e0125261. https://doi.org/10. 00076 eCollection 2015. 1371/journal.pone.0125261 eCollection 2015. 26. Seijkens T, Kusters P, Chatzigeorgiou A, Chavakis T, Lutgens E. Immune cell 3. Lennon H, Sperrin M, Badrick E, et al. The obesity paradox in Cancer: crosstalk in obesity: a key role for costimulation? Diabetes. 2014;63(12): a review. Curr Oncol Rep. 2016;18(9):56. https://doi.org/10.1007/ 3982–91. https://doi.org/10.2337/db14-0272. s11912-016-0539-4. 27. Han SJ, Glatman Zaretsky A, Andrade-Oliveira V, et al. White Adipose Tissue 4. Albiges L, Hakimi AA, Xie W, et al. Body mass index and metastatic renal cell Is a Reservoir for Memory T Cells and Promotes Protective Memory carcinoma: clinical and biological correlations. J Clin Oncol. 2016;34(30):3655–63. Responses to Infection. Immunity. 2017;47(6):1154–1168.e6. https://doi.org/ https://doi.org/10.1200/JCO.2016.66.7311. 10.1016/j.immuni.2017.11.009 Epub 2017 Dec 5. 5. Antoun S, Bayar A, Ileana E, et al. High subcutaneous adipose tissue predicts 28. Singh S, Facciorusso A, Singh AG, et al. Obesity and response to anti- the prognosis in metastatic castration-resistant prostate cancer patients in tumor necrosis factor-α agents in patients with select immune- post chemotherapy setting. Eur J Cancer. 2015;51(17):2570–7. https://doi. mediated inflammatory diseases: a systematic review and meta- org/10.1016/j.ejca.2015.07.042 Epub 2015 Aug 13. analysis. PLoS One. 2018;13(5):e0195123. https://doi.org/10.1371/ 6. McQuade JL, Daniel CR, Hess KR, et al. Association of body-mass index and journal.pone.0195123. outcomes in patients with metastatic melanoma treated with targeted 29. Mullen M, Gonzalez-Perez RR. Leptin-Induced JAK/STAT Signaling and therapy, immunotherapy, or chemotherapy: a retrospective, multicohort Cancer Growth. Vaccines (Basel). 2016;4(3). pii: E26. doi: https://doi.org/10. analysis. Lancet Oncol. 2018;19(3):310–22 Epub 2018 Feb 12. 3390/vaccines4030026. 7. Heidelberger V, Goldwasser F, Kramkimel N, et al. Sarcopenic overweight is 30. Sanlorenzo M, Vujic I, Daud A, et al. Pembrolizumab Cutaneous Adverse associated with early acute limiting toxicity of anti-PD1 checkpoint Events and Their Association With Disease Progression JAMA Dermatol. inhibitors in melanoma patients. Investig New Drugs. 2017;35(4):436–41. 2015;151(11):1206–12. https://doi.org/10.1001/jamadermatol.2015.1916. https://doi.org/10.1007/s10637-017-0464-x Epub 2017 Apr 10. 31. Freeman-Keller M, Kim Y, Cronin H, et al. Nivolumab in resected and 8. Richtig G, Hoeller C, Wolf M, et al. Body mass index may predict the Unresectable metastatic melanoma: characteristics of immune-related response to ipilimumab in metastatic melanoma: An observational multi- adverse events and association with outcomes. Clin Cancer Res. 2016;22(4): centre study. PLoS One. 2018 ;13(10):e0204729. https://doi.org/10.1371/ 886–94. https://doi.org/10.1158/1078-0432.CCR-15-1136. journal.pone.0204729. eCollection 2018.13. 32. Min Lee CK, Li S, Tran DC, et al. Characterization of dermatitis after PD-1/PD- 9. Wang Z, Aguilar EG, Luna JI, et al. Paradoxical effects of obesity on T cell L1 inhibitor therapy and association with multiple oncologic outcomes: A function during tumor progression and PD-1 checkpoint blockade. Nat retrospective case-control study. J Am Acad Dermatol. 2018. pii: S0190– Med. 2018. https://doi.org/10.1038/s41591-018-0221-5 [Epub ahead of print]. 9622(18)30831–4. doi: https://doi.org/10.1016/j.jaad.2018.05.035. [Epub 10. Minana B, Cozar JM, Palou J, et al. Bladder cancer in Spain 2011: population- ahead of print]. based study. J Urol. 2014;191(2):323–8. 33. Osorio JC, Ni A, Chaft JE, et al. Antibody-mediated thyroid dysfunction during 11. Ciocan D, Barbe C, Aubin F, et al. Distinctive features of melanoma and its T-cell checkpoint blockade in patients with non-small-cell lung cancer. Ann management in elderly patients: a population-based study in France. JAJA Oncol. 2017;28(3):583–9. https://doi.org/10.1093/annonc/mdw640. Dermatol. 2013;149(10):1150–7. 34. Haratani K, Hayashi H, Chiba Y, et al. Association of Immune-Related 12. Gridelli C, Balducci L, Ciardiello F, et al. Treatment of elderly patients with non- Adverse Events with Nivolumab Efficacy in non-small-cell lung Cancer. small-cell lung Cancer: results of an international expert panel meeting of the JAMA Oncol. 2018;4(3):374–8. https://doi.org/10.1001/jamaoncol.2017.2925. Italian Association of Thoracic Oncology. Clin Lung Cancer. 2015;16(5):325–33. 35. Teraoka S, Fujimoto D, Morimoto T, et al. Early immune-related adverse 13. Azawi NH, Joergensen SM, Jensen NV, et al. Trends in kidney cancer among events and association with outcome in advanced non-small cell lung the elderly in Denmark, 1980-2012. Acta Oncol. 2016;55(Suppl 1):79–84. Cancer patients treated with Nivolumab: a prospective cohort study. J 14. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation Thorac Oncol. 2017;12(12):1798–805. https://doi.org/10.1016/j.jtho.2017.08. criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J 022 Epub 2017 Sep 20. Cancer. 2009;45:228–47. 36. Conforti F, Pala L, Bagnardi V, et al. Cancer immunotherapy efficacy 15. Mantel N. Chi-square tests with one degree of freedom: extensions of the and patients' sex: a systematic review and meta-analysis. Lancet Mendel-Haenszel procedure. J Am Stat Assoc. 1963;58:690–700. Oncol. 2018;19(6):737–46. https://doi.org/10.1016/S1470-2045(18)30261- 16. Hosmer DW Jr, Lemeshow S, Sturdivant RX. Applied logistic regression. 4 Epub 2018 May 16. Third Edition. New Jersey: John Wiley & Sons (2013). 37. Botticelli A, Onesti CE, Zizzari I, et al. The sexist behaviour of immune 17. Kaplan EL, Meier P. Nonparametric estimation of incomplete observations. J checkpoint inhibitors in cancer therapy? Oncotarget. 2017;8(59):99336–46. Am Stat Assoc. 1958;53:457–81. https://doi.org/10.18632/oncotarget.22242. 18. Schemper M, Smith TL. A note on quantifying follow-up in studies of failure 38. Pallottini V, Bulzomi P, Galluzzo P, et al. Estrogen regulation of adipose time. Control Clin Trials. 1997;17:343–6. tissue functions: involvement of estrogen receptor isoforms. Infect Disord 19. Cox DR. Regression models and life tables (with discussion). J R Stat Soc Ser Drug Targets. 2008 Mar;8(1):52–60. B. 1972;74(187–200). 39. Khan D, Ahmed SA. The Immune System Is a Natural Target for 20. Goebell PJ, Müller L, Hübner A, et al. Body mass index as independent Estrogen Action: Opposing Effects of Estrogen in Two Prototypical predictor of overall survival in patients with advanced renal cell carcinoma Autoimmune Diseases. Front Immunol. 2015;6:635. https://doi.org/10. at start of systemic treatment-Analyses from the German clinical RCC- 3389/fimmu.2015.00635. Registry. Urol Oncol. 2018;(10):470.e1–9. https://doi.org/10.1016/j.urolonc. 40. Stocco C. Tissue Physiology and Pathology of Aromatase Steroids. 2012 Jan; 2018.07.007 Epub 2018 Aug 18. 77(1-2): 27–35. Published online 2011 Nov 13. doi: https://doi.org/10.1016/j. 21. Lutz CT, Quinn LS. Sarcopenia, obesity, and natural killer cell immune steroids.2011.10.013. senescence in aging: altered cytokine levels as a common mechanism. Aging (Albany NY). 2012;4(8):535–46. 41. Markle JG, Frank DN, Mortin-Toth S, Robertson CE, Feazel LM, Rolle- 22. Sutherland LN, Capozzi LC, Turchinsky NJ, et al. Time course of high-fat Kampczyk U, et al. Sex differences in the gut microbiome drive hormone- diet-induced reductions in adipose tissue mitochondrial proteins: dependent regulation of autoimmunity. Science (2013) 339(6123):1084–8. potential mechanisms and the relationship to glucose intolerance. Am J https://doi.org/10.1126/science.1233521. Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 11 of 11 42. Yurkovetskiy L, Burrows M, Khan AA, Graham L, Volchkov P, Becker L, et al. Gender bias in autoimmunity is influenced by microbiota. Immunity. 2013; 39(2):400–12. https://doi.org/10.1016/j.immuni.2013.08.013. 43. Routy B, Le Chatelier E, Derosa L, et al. Gut microbiome influences efficacy of PD-1-based immunotherapy against epithelial tumors. Science. 2018; 359(6371):91–7. https://doi.org/10.1126/science.aan3706. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal for ImmunoTherapy of Cancer Springer Journals

A multicenter study of body mass index in cancer patients treated with anti-PD-1/PD-L1 immune checkpoint inhibitors: when overweight becomes favorable

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/a-multicenter-study-of-body-mass-index-in-cancer-patients-treated-with-odYvqxRD5p

References (54)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s).
Subject
Medicine & Public Health; Oncology; Immunology
eISSN
2051-1426
DOI
10.1186/s40425-019-0527-y
Publisher site
See Article on Publisher Site

Abstract

Background: Recent evidence suggested a potential correlation between overweight and the efficacy of immune checkpoint inhibitors (ICIs) in cancer patients. Patients and methods: We conducted a retrospective study of advanced cancer patients consecutively treated with anti-PD-1/PD-L1 inhibitors, in order to compare clinical outcomes according to baseline BMI levels as primary analysis. Based on their BMI, patients were categorized into overweight/obese (≥ 25) and non-overweight (< 25). A gender analysis was also performed, using the same binomial cut-off. Further subgroup analyses were performed categorizing patients into underweight, normal weight, overweight and obese. Results: Between September 2013 and May 2018, 976 patients were evaluated. The median age was 68 years, male/female ratio was 663/313. Primary tumors were: NSCLC (65.1%), melanoma (18.7%), renal cell carcinoma (13.8%) and others (2.4%). ECOG-PS was ≥2 in 145 patients (14.9%). PD-1/PD-L1 inhibitors were administered as first-line treatment in 26.6% of cases. Median BMI was 24.9: 492 patients (50.6%) were non-overweight, 480 patients (50.4%) were overweight/obese. 25.2% of non-overweight patients experienced irAEs of any grade, while 55.6% of overweight/obese patients (p < 0.0001). ORR was significantly higher in overweight/obese patients compared to non-overweight (p < 0.0001). Median follow-up was 17.2 months. Median TTF, PFS and OS were significantly longer for overweight/obese patients in univariate (p < 0.0001, for all the survival intervals) and multivariate models (p = 0.0009, p < 0.0001 and p < 0.0001 respectively). The significance was confirmed in both sex, except for PFS in male patients (p = 0.0668). (Continued on next page) * Correspondence: alessiocortellini@gmail.com Medical Oncology, St. Salvatore Hospital, L’Aquila, Italy Department of Biotechnological and Applied Clinical Sciences, University of L’Aquila, L’Aquila, Italy Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 2 of 11 (Continued from previous page) Conclusions: Overweight could be considered a tumorigenic immune-dysfunction that could be effectively reversed by ICIs. BMI could be a useful predictive tool in clinical practice and a stratification factor in prospective clinical trials with ICIs. Keywords: BMI, Anti-PD-1/PD-L1, Overweight, Obesity, Cancer, Immunotherapy st Key message treatment with single agent anti-PD-1/PD-L1 as 1 or Recent evidence revealed that adipose tissue might affect subsequent line, at the medical oncology departments of the response to immune checkpoint inhibitors (ICIs) in 17 Italian centers (Additional file 1), between September cancer patients. In this retrospective transverse study, 2013 and May 2018. enrolling 976 advanced cancer patients treated with anti-PD-1/PD-L1 immunotherapy, we found a significant Anthropometric measurements association between overweight (BMI ≥ 25) and improved Weight and height were obtained from the patient’s clinical outcomes to ICIs. medical records at the time of immunotherapy initiation. BMI was calculated using the formula of weight/height Introduction (kilograms per square meter) and classified according to Although the interaction between malnutrition and chronic the World Health Organization (WHO) categories: inflammation has been widely investigated, whether this underweight, BMI < 18.5; normal, 18.5 ≤ BMI ≤ 24.9; association is causative or correlative is still debated [1]. overweight, 25 ≤ BMI ≤ 29.9; obesity, BMI ≥ 30. For the Historically, body mass index (BMI) has been considered study purpose, the binomial cut-off for BMI </≥ 25 was the major surrogate of nutritional status and its correlation used, and patients were categorized into non-overweight with clinical outcomes in advanced cancer patients has (< 25) and overweight/obese (≥ 25) for the final analysis. already been investigated without conclusive results [2–5]. Underweight patients were included in the non-over- It is now becoming clear that the nutritional assess- weight group. ment, which should include BMI, could be seen in a "new light" in the era of immune checkpoint inhibitors Study design (ICIs). A large retrospective study has recently found an We conducted a “real-life”, multicenter, retrospective association between BMI and improved progression free observational study aimed at comparing the clinical out- survival (PFS) and overall survival (OS) in melanoma comes of cancer patients treated with ICIs according to patients treated with either targeted therapy or immuno- baseline BMI levels. therapy [6]. Another study has reported that overweight Primary outcomes measures were: objective response sarcopenic melanoma patients treated with anti-PD1 rate (ORR), time to treatment failure (TTF), PFS and (Programmed cell death protein 1) inhibitors expe- OS. ORR was defined as the proportion of patients rienced early acute limiting toxicity [7]. experiencing an objective response (either complete re- Additionally, another retrospective analysis by Richtig et sponse or partial response) as best response to immuno- al. revealed that overweight (BMI ≥ 25) melanoma patients therapy. TTF was defined as the time from treatment’s (76 total) treated with ipilimumab had significantly higher start to discontinuation for any reason. Progression-free response rate (p=0.024) andatrendforlongerOS (p= survival (PFS) was defined as the time from the start of 0.056), when compared to non-overweight patients [8]. immunotherapy to the date of disease progression or Lastly, Wang and colleagues have recently reported death, whichever occurred first. Patients who were alive an improvement in terms of PFS (p = 0.003) and OS without disease progression were censored on the date (p = 0.049) in a cohort of obese advanced cancer of their last disease assessment. Overall survival (OS) patients (BMI ≥ 30) treated with ICIs [9]. was defined as the time from the start of immunothe- To further dissect this question, we conducted a large, rapy to death. Patients who were still alive were cen- multicentre, retrospective transverse study to evaluate sored at the date of last contact. Patients were treated clinical outcomes of patients with advanced solid tumors according to the tumor type indication with pembrolizu- treated with ICIs according to baseline BMI. mab, nivolumab or atezolizumab with standard doses and schedules. Materials and methods In order to weighing the possible prognostic influence Patient eligibility of obesity (30 BMI) and malnutrition (or cachexia), two This study enrolled patients with confirmed diagnosis of subgroup analysis (according to each BMI categories) measurable stage IV cancer, who consecutively underwent were performed. In the first one, overweight (25-30 BMI) Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 3 of 11 and obese (≥ 30 BMI) patients were respectively com- Table 1 Patients’ characteristics pared to non-overweight (< 25) patients, in the second N° (%) one overweight (25-30 BMI) and obese (≥ 30 BMI) patients were respectively compared to normal weight AGE, (years) patients (18.5-25 BMI). Median 68 A subgroup analysis comparing clinical outcomes in Range 24–92 males and females patients, using the binomial cut-off Elderly (≥ 70) 445 (45.6) (BMI </≥ 25) was also conducted as secondary analysis. The following covariates were considered for the SEX multivariate analyses: primary tumor (NSCLC, mela- Male 663 (67.9) noma, kidney and others), sex (male vs female), Eastern Female 313 (32.1) Cooperative Oncology Group Performance Status ECOG PS (ECOG-PS) (0-1 vs. ≥ 2), age (< 70 vs. ≥ 70 years old) 0–1 831 (85.1) [10–13], number of metastatic sites (≤ 2 vs. > 2) and ≥ 2 145 (14.9) treatment line (first vs non-first). As in some indications the anti-PD-1/PD-L1 agents dosages had been weight- Primary Tumor based, weight was used as a continuous covariate in all NSCLC 635 (65.1) the analyses, considering the possible dose-depending Melanoma 183 (18.7) confounding effect on the clinical outcomes. Renal cell carcinoma 135 (13.8) Immune-related AEs (irAEs) were graduated according to Others 23 (2.4) theCommonToxicity Criteriafor AdverseEvents (CTCAE; No. of metastatic sites version 4.0) and cumulatively reported. Immune-related AEs were categorized on the basis of the organ/system ≤ 2 467 (47.9) involved as follows: endocrine irAEs (including thyroid > 2 509 (52.1) disorders), gastro-intestinal (GI) irAEs (excluding pan- Type of anti-PD-1/PD-L1 agent creatitis), skin irAEs, pneumological irAEs, hepatic Pembrolizumab 235 (24.1) irAEs, rheumatologic irAEs and others irAEs (including Nivolumab 706 (72.3) neuro-muscolar, pancreatitis, fever, asthenia and an- Atezolizumab 35 (3.6) orexia). The safety analysis was performed for irAEs of any grade and for G3/G4 irAEs. Treatment line of Immunotherapy To determine ORR and PFS, scans were reviewed by a First 260 (26.6) dedicated thoracic oncologist at each Institution using Non-First 716 (73.4) Response Evaluation Criteria In Solid Tumors (RECIST) Weight (Kg) version 1.1. [14]. χ2 was used to compare ORR and inci- Median 71 dence of irAEs among subgroups [15]. In the multivari- Range 35–139 ate analysis, logistic regression was used to evaluate the role of parameters proven to be significant at the univa- BMI (kg/m ) riate analysis of ORR [16]. Median TTF, median PFS, Median (range) 24.9 (13.5–46.6) and median OS were evaluated using the Kaplan-Meier Underweight (BMI ≤ 18.5), n°(%) 40 (4.1) method [17]. Median follow-up was calculated according Normal weight (BMI 18.5 < BMI ≤ 24.9), n°(%) 452 (46.3) to the reverse Kaplan-Meier method [18]. Cox propor- Overweight (25 < BMI ≤ 29.9), n°(%) 377 (38.6) tional hazards model [19] was used to evaluate predictor Obese (BMI ≥ 30), n° (%) 107 (11) variables in univariate and multivariate analysis for TTF, th PFS and OS. The data cut-off was October 29 , 2018. All statistical analyses were performed using MedCalc tumors were: NSCLC (635 patients), melanoma (183 Statistical Software version 18.6 (MedCalc Software patients), renal cell carcinoma (135 patients) and others bvba, Ostend, Belgium; http://www.medcalc.org; 2018). (23 patients). ECOG-PS was 0/1 in 831 patients (85.1%), and ≥ 2 in 145 patients (14.9%); 467 patients (47.9%) had Results ≤ 2 metastatic sites while 509 (52.1%) had more than 2 Patient characteristics metastatic sites. PD-1/PD-L1 inhibitors were adminis- Nine hundred and seventy-six, consecutive advanced tered as first-line treatment in 260 patients (26.6%). cancer patients were evaluated. Patient characteristics Median weight was 71 Kg, median BMI was 24.9; are summarized in Table 1. The median age was 68 years according to WHO classification 40 patients (4.1%) were (range: 24 – 92), male/female ratio was 663/313. Primary defined as underweight, 452 patients (46.3%) as having a Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 4 of 11 normal weight, 377 patients (38.6%) as overweight and Efficacy analysis 107 patients (11%) as obese. For the study purpose, 492 At median follow-up of 17.2 months, median TTF was patients were considered as non-overweight (50.4%) and 5.9 months (95% CI: 5.3 – 6.7; 681 events), median PFS 484 patients were categorized as overweight/obese was 6.5 months (95% CI: 6.1 – 7.1; 644 events) and median (49.6%) according to a BMI cut-off of 25 (<25 vs. ≥25). OS was 13.4 months (95% CI: 11.0 – 16.5; 488 censored Among male patients median age was 69 years, median patients) in the entire cohort. weight was 72 Kg (range: 35 – 139) and median BMI was When these outcomes where analyzed according to 24.8 (range: 14 – 46.6). Among female patients median BMI, we found that median TTF was significantly longer age was 67, median weight was 70 Kg (range: 40 – 130) in overweight/obese patients compared to non-overweight and median BMI was 25.4 (range: 13.6 – 46.1). patients (9.3 [95% CI: 8.1 – 11.6; 318 events] vs. 3.6 months [95% CI: 3.2 – 4.1; 363 events]; HR= 0.51 [95% CI: 0.44 – 0.60], p < 0.0001) (Fig. 1a). Similarly, Safety analysis median PFS was significantly improved in the over- In the entire cohort, 393 patients (40.3%) experienced weight/obese group compared to the non-overweight irAEs of any grade. Sixty-three patients (6.5%) expe- group (11.7 months [95% CI: 9.4 – 15; 286 events] vs. rienced G3/G4 irAEs. Overweight/obese patients were 3.7 months [95% CI: 3.2 – 4.1; 358 events]; HR= 0.46 significantly more likely to experience any grade irAEs [95%CI: 0.39 – 0.54], p < 0.0001) (Fig. 1b). Consistently compared to non-overweight patients (55.6% vs. 25.2%, we also found a significantly prolonged median OS among p < 0.0001). However, no difference in the rate of G3/G4 overweight/obese patients compared to non-over irAEs was observed between Overweight/obese patients weight patients (26.6 months [95% CI: 21.4 – 36.8; 286 and non-overweight patients (7.6 vs. 5.3%, p = 0.1338). censored patients] vs. 6.6 months [95% CI: 5.8 – 8.5; The safety profile of ICIs according to BMI is summa- 182 censored patients]; HR= 0.33 [95%CI: 0.28 – 0.41], rized in Additional file 2. p < 0.0001) (Fig. 1c). After adjusting for PS, treatment line, n° of metastatic sites, gender, primary tumor subtype and development Activity analysis of irAEs, a BMI of ≥25 retained a significant association Univariate and multivariate analyses for ORR are de- with a longer TTF (p = 0.0009), PFS (p < 0.0001) and tailed in Additional file 3. Among 910 patients evaluable OS (p < 0.0001) in multivariate models (Table 2, Table 3, for activity, 283 patients had a response to ICIs (ORR: Table 4) 31.1%). Overweight/obese patients had a significantly higher ORR compared non-overweight patients (41.3% vs. 20.9%, p < 0.0001). Similarly, we found a significantly Subgroup analyses higher ORR among patients who experienced at least Table 5 reports the univariate and multivariate gender ana- 1 irAE compared to those without irAEs (45.1% vs. lyses for TTF, PFS and OS of male patients (Table 5A) and 21.1%, p < 0.0001). Both BMI (overweight/obese vs. female patients (Table 5B). As shown overweight/obese non-overweight) and the development of irAEs of any male patients had significantly longer TTF (p = 0.0330) and grade, were independently associate with higher ORR OS (p = 0.0013), but not PFS (p = 0.0668), when compared in the multivariate analysis (p = 0.0239 and p < 0.0001, with non-overweight patients, while overweight/obese fe- respectively). male patients had significantly longer TTF (p = 0.0037), AB C Fig. 1 Kaplan-Meier survival curves according to binomial BMI levels (cut-off 25). (a) Time to Treatment Failure. BMI < 25: 3.6 months (95% CI: 3.2–4.1); BMI ≥ 25: 9.3 months (95%CI: 8.1–11.6). (b) Progression Free Survival. BMI < 25: 3.7 months (95% CI: 3.2–4.1); BMI ≥ 25: 11.7 months (95% CI: 9.4–15). (C) Overall Survival. BMI < 25: 6.6 months (95% CI: 5.8–8.5); BMI ≥ 25: 26.6 months (95% CI: 21.4–36.8) Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 5 of 11 Table 2 Cox proportional-hazards regression: univariate and multivariate analyses of Time to Treatment Failure Time to Treatment Failure Univariate Analysis Multivariate Analysis VARIABLE (Comparator) HR (95% CI); p - value HR (95% CI); p - value BMI 0.51 (0.44–0.60); p < 0.0001 0.67 (0.53–0.85); p = 0.0009 ≥ 25 vs < 25 Weight 0.98 (0.97–0.99); p < 0.0001 0.99 (0.98–1.01); p = 0.8422 irAEs of any grade 0.57 (0.48–0.66); p < 0.0001 0.79 (0.65–0.97); p = 0.0295 Yes vs No Primary Tumor (NSCLC) Melanoma 0.62 (0.50–0.76); p < 0.0001 0.79 (0.64–1.01); p = 0.0517 Kidney 0.73 (0.59–0.92); p = 0.0077 0.71 (0.56–0.88); p = 0.0025 Others 1.15 (0.71–1.87); p = 0.5560 0.78 (0.48–1.28); p = 0.3389 Sex 1.22 (1.04–1.43); p = 0.0147 1.10 (0.93–1.30); p = 0.2607 Male vs Female Age 1.04 (0.90–1.21); p = 0.5366 – Elderly vs Non-elderly Treatment line 1.36 (1.13–1.64); p = 0.0008 1.51 (1.25–1.81); p < 0.0001 Non-first vs First N° of metastatic sites 1.54 (1.34–1.77); p < 0.0001 1.52 (1.30–1.77); p < 0.0001 >2 vs ≤ 2 ECOG PS 2.86 (2.36–3.48); p < 0.0001 2.35 (1.92–2.88); p < 0.0001 ≥2vs0–1 Weight was used as a continuous variable Table 3 Cox proportional-hazards regression: univariate and multivariate analyses of Progression Free Survival Progression Free Survival Univariate Analysis Multivariate Analysis VARIABLE (Comparator) HR (95% CI); p - value HR (95% CI); p - value BMI 0.46 (0.39–0.54); p < 0.0001 0.71 (0.56–0.90); p < 0.0001 ≥ 25 vs < 25 Weight 0.97 (0.96–0.98); p < 0.0001 0.99 (0.98–1.01); p = 0.1580 irAEs of any grade 0.48 (0.41–0.57); p < 0.0001 0.67 (0.54–0.83); p = 0.0002 Yes vs No Primary Tumor (NSCLC) Melanoma 0.52 (0.42–0.66); p < 0.0001 0.67 (0.53–0.85); p = 0.0008 Kidney 0.72 (0.58–0.91); p = 0.0062 0.67 (0.53–0.84); p = 0.0008 Others 1.08 (0.65–1.78); p = 0.7556 0.69 (0.41–1.15); p = 0.1533 Sex 1.20 (1.01–1.42); p = 0.0314 1.03 (0.86–1.22); p = 0.7252 Male vs Female Age 0.96 (0.82–1.12); p = 0.6394 – Elderly vs Non-elderly Treatment line 1.62 (1.33–1.96); p < 0.0001 1.61 (1.32–1.93); p < 0.0001 Non-first vs First N° of metastatic sites 1.46 (1.27–1.68); p < 0.0001 1.42 (1.21–1.67); p < 0.0001 >2 vs ≤ 2 ECOG PS 2.60 (2.13–3.17); p < 0.0001 2.06 (1.67–2.52); p < 0.0001 ≥2vs 0–1 Weight was used as a continuous variable Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 6 of 11 Table 4 Cox proportional-hazards regression: univariate and multivariate analyses of Overall Survival Overall Survival Univariate Analysis Multivariate Analysis VARIABLE HR (95% CI); p – value HR (95% CI); p - value (Comparator) BMI 0.33 (0.28–0.41); p < 0.0001 0.49 (0.38–0.64); p < 0.0001 ≥ 25 vs < 25 Weight 0.97 (0.96–0.97); p < 0.0001 0.99 (0.99–1.01); p = 0.1884 irAEs of any grade 0.45 (0.37–0.54); p < 0.0001 0.82 (0.65–1.04); p = 0.1085 Yes vs No Primary Tumor (NSCLC) Melanoma 0.49 (0.38–0.64); p < 0.0001 0.67 (0.51–0.87); p = 0.0036 Kidney 0.56 (0.42–0.74); p = 0.0001 0.61 (0.45–0.80); p = 0.0005 Others 1.11 (0.62–1.96); p = 0.7337 0.71 (0.40–1.28); p = 0.2632 Sex 1.50 (1.23–1.83); p < 0.0001 1.33 (1.09–1.63); p = 0.0044 Male vs Female Age 1.11 (0.93–1.32); p = 0.2401 – Elderly vs Non-elderly Treatment line 1.58 (1.26–1.97); p = 0.0001 1.42 (1.15–1.77); p = 0.0012 Non-first vs First N° of metastatic sites 1.52 (1.29–1.78); p < 0.0001 1.41 (1.17–1.69); p = 0.0002 >2 vs ≤ 2 ECOG PS 2.07 (1.87–2.29); p < 0.0001 2.59 (2.09–3.21); p < 0.0001 ≥2vs 0–1 Weight was used as a continuous variable PFS (p = 0.0132) and OS (p < 0.0001), when compared to also found a significant prolonged median OS among nor- non-overweight patients. mal weight compared to underweight patients (7.9 months Median TTF was not significantly different between [95%CI: 6.4 – 9.8; 178 censored patients] vs. 2.8 months overweight and obese patients (10.3 months [95%CI: [95%CI: 1.8 – 3.6; 4 censored patients], HR= 0.33 [95%CI: 8.2 – 4.1; 238 events] vs. 7.3 [95%CI: 5.5 – 11.7; 80 0.23 – 0.48], p < 0.0001). Table 7 reports the univariate and events], HR=1.23 [95%CI: 0.95 – 1.58], p = 0.1087). multivariate analyses of TTF, PFS and OS, comparing Similarly, we found no significant differences in median overweight (non-obese) patients and obese patients PFS (11.2 months [95%CI: 9.1 – 15.6; 223 events pa- with normal weight patients. Figure 3 reports the tients] vs. 12.9 months [95%CI: 7.1 – 18; 63 events], Kaplan-Meier survival curves of obese, overweight and HR=0.99 [95%CI: 0.75 – 1.31], p = 0.9798) and median normal weight patients. OS (26.6 months [95%CI: 21.4 – 36.8; 223 censored patients] vs. not reached [63 censored patients], Discussion HR=1.04 [95%CI: 0.75 – 1.46], p = 0.7767) between over- In this study we demonstrated that patients with a weight and obese patients. Table 6 reports the univariate BMI ≥ 25 experienced a better clinical outcome com- and multivariate analyses of TTF, PFS and OS, comparing pared to those with a BMI < 25. Recently, the association overweight (non-obese) patients and obese patient with between BMI and OS of metastatic renal cell carcinoma non-overweight patients. Figure 2 reports the Kaplan-Meier patients, has been reported regardless of the use of survival curves of obese, overweight and non-over- anti-PD-1/PD-L1 therapy [4, 20]. However, in our study weight patients. we found a strong correlation between overweight and When we analyzed the clinical outcomes of normal improved clinical outcomes with anti-PD-1/PD-L1. weight vs. underweight patients, we found a significantly Some authors have already speculated about the nega- longer median TTF (3.9 months [95%CI: 3.4 – 5.0; 327 tive impact of body composition alteration on immune events] vs. 1.8 [95%CI: 1.7 – 2.9; 36 events], HR= 0.51 cells activity [21]. Interestingly, it has been increasingly [95%CI: 0.35 – 0.71], p = 0.0001 and median PFS recognized that white adipose tissue, which is the most (4.4 months [95%CI: 3.6 – 5.3; 322 events] vs. 1.9 related to the fattening process [22], is also involved in months [95%CI: 1.7 – 2.9; 36 events] HR= 0.45; the induction and/or coordination of host defenses, 95%CI: 0.32 – 0.64], p < 0.0001) in normal weight being a source of cytokines and chemokines [23]. In fact, patients compared with underweight patients. We adipose tissue modulates the Th1/Th2 balance, decreases Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 7 of 11 Table 5 Cox proportional-hazards regression: univariate and Table 7 Cox proportional-hazards regression: univariate and multivariate analyses multivariate analyses according to normal weight (18.5-25), overweight (25-30) and obese (≥ 30) BMI levels A Univariate Analysis Multivariate Analysis Univariate Analysis Multivariate Analysis VARIABLE HR (95% CI); p - value HR (95% CI); p - value VARIABLE HR (95% CI); p - value HR (95% CI); p - value Time to Treatment Failure (Comparator) BMI 0.54 (0.45–0.66); p < 0.0001 0.74 (0.56–0.97); p = 0.0330 BMI Time to Treatment Failure ≥ 25 vs < 25 (18.5–25) Progression Free Survival 25–30 0.51 (0.43–0.61); p < 0.0001 0.65 (0.51–0.82); p = 0.0004 BMI 0.49 (0.40–0.59); p < 0.0001 0.77 (0.58–1.01); p = 0.0668 ≥ 30 0.63 (0.49–0.81); p = 0.0003 0.79 (0.55–1.15); p = 0.2300 ≥ 25 vs < 25 BMI Progression Free Survival Overall Survival (18.5–25) BMI 0.38 (0.31–0.48); p < 0.0001 0.59 (0.43–0.81); p = 0.0013 25–30 0.49 (0.41–0.58); p < 0.0001 0.68 (0.53–0.87); p = 0.0016 ≥ 25 vs < 25 ≥ 30 0.48 (0.37–0.64); p < 0.0001 0.72 (0.49–1.06); p = 0.0991 BMI Overall Survival B Univariate Analysis Multiavariate Analysis (18.5–25) VARIABLE HR (95% CI); p - value HR (95% CI); p - value 25–30 0.35 (0.29–0.43); p < 0.0001 0.46 (0.35–0.61); p < 0.0001 ≥ 30 0.37 (0.27–0.51); p < 0.0001 0.50 (0.32–0.79); p = 0.0029 Time to Treatment Failure The used covariates (not shown) were: weight (continuous), irAEs of any grade, BMI 0.45 (0.35–0.61); p < 0.0001 0.51 (0.32–0.80); p = 0.0037 primary tumors, sex, line of treatment, ECOG-PS, number of metastatic sites ≥ 25 vs < 25 Progression Free Survival the activation of Treg through adiponectin, increases BMI 0.41 (0.31–0.56); p < 0.0001 0.56 (0.35–0.88); p = 0.0132 pro-inflammatory macrophages, activates T-cells with ≥ 25 vs < 25 the binding between LIGHT-HVEM (herpesvirus entry Overall Survival mediator) and increases the inflammatory status through BMI 0.25 (0.17–0.36); p < 0.0001 0.27 (0.15–0.48); p < 0.0001 CD40 pathway [24–26]. ≥ 25 vs < 25 Moreover, a recent preclinical study revealed that (A) male patients (B) female patients. The used covariates (not shown) were: white adipose tissue might also play a role in immune weight (continuous), irAEs of any grade, primary tumors, line of treatment, homeostasis [27]. In this study, white adipose tissue of ECOG-PS, number of metastatic sites mice was reported to accumulate pathogen-specific memory T-cells after a microbial infection, including tissue-resident cells expressing a distinct metabolic pro- file. Intriguingly, these data support the hypothesis that adipose tissue can act as a reservoir of tissue-specific memory T-cells, which can undergo a rapid response to reactivation against exogenous stimuli. This evidence Table 6 Cox proportional-hazards regression: univariate and raises an interesting question, can these adipose tissue- multivariate analyses according to non-overweight (< 25), specific T-cells be promptly reactivated against cancer- overweight (25-30) and obese (≥ 30) BMI levels specific antigens as they do against microbial antigens? Univariate Analysis Multivariate Analysis In a recent meta-analysis of patients with immune-me- VARIABLE HR (95% CI); p - value HR (95% CI); p - value diated inflammatory diseases treated with anti-TNF (Comparator) (tumor necrosis factor), the authors reported a trend BMI Time to Treatment Failure towards a lower response rate to treatment among over- (< 25) weight patients [28]. This is likely to reflect the reduced 25–30 0.49 (0.41–0.58); p < 0.0001 0.67 (0.53–0.84); p = 0.0008 responsivity of T-cells of obese individuals, which has ≥ 30 0.60 (0.47–0.77); p = 0.0001 0.86 (0.61–1.24); p = 0.4414 also been confirmed in preclinical models showing a BMI Progression Free Survival significant increase in dysfunctional exhausted T-cells (< 25) in obese mice [9]. Nevertheless, such inflamed and 25–30 0.46 (0.39–0.55); p < 0.0001 0.71 (0.56–0.89); p = 0.0044 immune-exhausted status may be more likely suscep- ≥ 30 0.46 (0.35–0.61); p < 0.0001 0.80 (0.55–1.17); p = 0.2669 tible to the immune checkpoint blockade. In support of BMI Overall Survival this, in preclinical models, T-cell dysfunction in obese (< 25) mice was proven to be partly mediated by the PD-1 axis 25–30 0.33 (0.27–0.41); p < 0.0001 0.49 (0.37–0.64); p < 0.0001 and driven by leptin, strengthening the already known ≥ 30 0.34 (0.25–0.48); p < 0.0001 0.61 (0.39–0.94); p = 0.0258 correlation between JAK/STAT pathway and immune The used covariates (not shown) were: weight (continuous), irAEs of any grade, primary tumors, sex, line of treatment, ECOG-PS, number of metastatic sites checkpoint inhibition [9, 29]. Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 8 of 11 ABC Fig. 2 Kaplan-Meier survival curves according to BMI levels (non-overweight BMI < 25, overweight BMI 25–30, obese BMI ≥ 30). (a) Time to Treatment Failure. BMI < 25: 3.6 months (95% CI: 3.2–4.1); BMI 25–30: 10.3 months (95%CI: 8.2–4.1); BMI ≥ 30: 7.3 months (95%CI: 5.5–11.7). (b) Progression Free Survival. BMI < 25: 3.7 months (95% CI: 3.2–4.1); BMI 25–30: 11.2 months (95%CI: 9.1–15.6); BMI ≥ 30: 12.9 months (95%CI: 7.1–18). (c) Overall Survival. BMI < 25: 6.6 months (95% CI: 5.8–8.5); BMI 25–30: 26.6 months (95%CI: 21.4–36.8); BMI ≥ 30: not reached Importantly, in our study we also found a significantly analysis, compared to obese patients, thus supporting the higher incidence of irAEs of any grade among overweight/ hypothesis that the prognostic weight of obesity, could obese patients. In light of the emerging association be- have partially influenced the final results. tween the development of irAEs and improved clinical out- On the other hand, despite the small sample size (4.1% comes with ICIs across different tumor types, our findings of the entire population), underweight patients had are not unexpected [30–35]. In our cohort, the develop- significantly shorter TTF, PFS and OS, when compared ment of irAEs of any grade was independently associated to normal weight patients, confirming that malnutrition with improved clinical outcomes along with a BMI ≥25 in (and cachexia) is an independent negative prognostic multivariate analyses. factor. Nonetheless, when we compared obese and over- The analysis performed by separating overweight and weight patients (Table 7)with normal weightpatientswe obese patients, demonstrated that a linear relationship be- observed significantly improved clinical outcomes favou- tween BMI and positive outcomes cannot be assumed. ring the overweight group, which suggests that over- Even though we found no statistically significant diffe- weightness has a direct impact on the efficacy of ICIs. rences in TTF, PFS and OS between overweight and obese In our study we also carried a gender-based analysis. Pre- patients, when separately comparing obese patients to viously, it has been reported that female patients tend to non-overweight patients (Table 6), we observed the loss of have lower benefit from ICI compared to males [36, 37]. significance regarding TTF and PFS, while not regarding However, whether the gender plays a key role in determi- OS. This result is of particular interest, considering the ning the clinical outcome to immunotherapy is still in need possibility of a negative impact on survival of obese pa- of further investigation. In our study, we found that over- tients due to cardiovascular and metabolic complications weight female patients derived a greater clinical benefit of obesity itself. Noteworthy, the HRs are concordantly form immunotherapy as compared to the male counterpart lower for overweight (non-obese) patients in each survival (Table 5). However, it should be highlighted that ABC Fig. 3 Kaplan-Meier survival curves according to BMI levels (normal weight BMI 18.5–25, overweight BMI 25–30, obese BMI ≥ 30). (a) Time to Treatment Failure. BMI 18.5–25: 3.9 months (95% CI: 3.4–5.0); BMI 25–30: 10.3 months (95%CI: 8.2–4.1); BMI ≥ 30: 7.3 months (95%CI: 5.5–11.7). (b) Progression Free Survival. BMI 18.5–25: 4.4 months (95% CI: 3.6–5.3); BMI 25–30: 11.2 months (95%CI: 9.1–15.6); BMI ≥ 30: 12.9 months (95%CI: 7.1–18). (c) Overall Survival. BMI 18.5–25: 7.9 months (95% CI: 6.4–9.8); BMI 25–30: 26.6 months (95%CI: 21.4–36.8); BMI ≥ 30: not reached Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 9 of 11 overweightness was associated with improved outcomes in Acknowledgements This work was supported by the Consorzio Interuniversitario Nazionale per la both males and females in the multivariate analysis. This Bio-Oncologia (CINBO). led us to speculate that in our population the predictive role of BMI was to be stronger than the predictive role of Funding No funding was received. gender. Certainly, the relationship between sex, adipose tissue Availability of data and materials and immunity is complex and ambiguous. Sex-hormones, The datasets used during the present study are available from the in particular estrogens, could affect adipose tissue func- corresponding author upon reasonable request. tions [38], but in some respects their influence on the im- Authors’ contributions mune systems does not seem unidirectional [39]. All authors contributed to the publication according to the ICMJE guidelines Furthermore, the median age of female patients in our for the authorship. All authors read and approved the manuscript and agree to be accountable for all aspects of the research in ensuring that the study was 67, indicating a prevalence of postmenopausal accuracy or integrity of any part of the work are appropriately investigated patients. In this specific population the adipose tissue and resolved. becomes a major source of circulating estrogens [40]. Ethics approval and consent to participate Another way to explain how BMI might affect sex All patients provided written, informed consent to treatment with hormones levels and the immune response, is through immunotherapy. All patients alive at the time of data collection provided an diet regimens which underpin the weight gain. Indeed, informed consent for the present retrospective analysis. The procedures followed were in accordance with the precepts of Good Clinical Practice and gut microbiota may be influenced by the different the declaration of Helsinki. The study was approved by the respective local "modifying pressures" of various diet types. Interestingly, ethical committees on human experimentation of each institution, after males and females have recently been reported to have previous approval by the coordinating center (University of L’Aquila, Internal th Review Board protocol number 32865, approved on July 24 , 2018). gender-specific differences in their immune system and gut microbiota composition. Whether these differences in Consent for publication gut microbiota composition might impact the efficacy or Not applicable. the safety profile of immunotherapy is subject of intense Competing interests research and is expected to provide us further insight in the Dr Alessio Cortellini received grants as speaker by MSD, Astra-Zeneca and optimal management of our patients [41–43]. Boehringer Ingelheim, gran consultancies by BMS and Ipsen; dr Marcello Our study is certainly flawed by several caveats, Tiseo received grant as speaker and advisory role by Astra-Zeneca, Pfizer, Eli- Lilly, BMS, Novartis, Roche, MSD, Boehringer Ingelheim, Otsuka, Pierre Fabre; including the retrospective design with the risk of selec- dr Maria Giuseppa Vitale received travel grants and speaker fees by BMS, tion and data collection biases, the heterogeneity of the Ipsen Astellas, Jansen, Novartis and Pfizer; dr Sebatiano Buti received grants analyzed population, the lack of a centralized imaging as speaker and advisory role by BMS, Pfizer, MSD, Ipsen, Novartis, Astra- Zeneca; dr. Melissa Bersanelli received honoraria as speaker at scientific review for response assessment and the lack of data events and as consultant for advisory role by BMS and Pfizer. about patient comorbidities. In addition, the lack of con- trol group of patients who did not received ICIs further Publisher’sNote limit the power of our analysis. On the other hand, a Springer Nature remains neutral with regard to jurisdictional claims in unique strength of our study is that we evaluated the published maps and institutional affiliations. predictive role of baseline assessment of BMI in a “real Author details life” population of individuals candidate to receive ICIs. 1 2 Medical Oncology, St. Salvatore Hospital, L’Aquila, Italy. Department of Biotechnological and Applied Clinical Sciences, University of L’Aquila, Conclusion L’Aquila, Italy. Medical Oncology, University Hospital of Parma, Parma, Italy. Department of Medicine and Surgery, University of Parma, Parma, Italy. In this study we demonstrated that patients with a 5 6 Medical Oncology, Campus Bio-Medico University, Rome, Italy. Department BMI ≥ 25 experienced better clinical outcomes with of Clinical and Molecular Medicine, Sant’Andrea Hospital, Sapienza University anti-PD-1/PD-L1 agents, compared to those with a of Rome, Rome, Italy. Clinical Oncology Unit, S.S. Annunziata Hospital, Chieti, Italy. Department of Medical, Oral & Biotechnological Sciences University G. BMI < 25. Our results suggest that BMI could be a useful D’Annunzio, Chieti-Pescara, Italy. Medical Oncology, Santa Maria Goretti predictive tool in clinical practice as well as a reliable strati- 10 Hospital, Latina, Italy. Department of Oncology, University Hospital of Pisa, fication variable for prospective clinical trials with ICIs. Istituto Toscano Tumori, Pisa, Italy. Medical Oncology, “Santo Spirito” Hospital, Pescara, Italy. Istituto Dermopatico dell’Immacolata, IDI-IRCCS, Rome, Italy. Medical Oncology, University Hospital of Modena, Modena, 14 15 Italy. Medical Oncology, Fermo Area Vasta 4, Fermo, Italy. Oncology Clinic, Additional files Università Politecnica delle Marche, Ospedali Riuniti di Ancona, Ancona, Italy. 16 17 Medical Oncology, AV2 Fabriano ASUR Marche, Pescara, Italy. Medical Additional file 1: List of oncological institutions of the study. 18 Oncology Unit, University Hospital of Cagliari, Cagliari, Italy. Medical (DOCX 15 kb) 19 Oncology, Santa Maria della Misericordia Hospital, Perugia, Italy. Pulmonary Additional file 2: Immune-related adverse events of any grade and Oncology Unit, St. Camillo Forlanini Hospital, Rome, Italy. Medical Oncology G3/G4 immune-related adverse events. (DOCX 15 kb) (B), Policlinico Umberto I, “Sapienza” University of Rome, Rome, Italy. 21 22 Dermatology, San Salvatore Hospital, L’Aquila, Italy. Melanoma, Cancer Additional file 3: Univariate and multivariate analyses with logistic Immunotherapy and Development Therapeutics Unit, Istituto Nazionale regression of Objective Response Rate. (DOC 49 kb) Tumori-IRCCS Fondazione “G. Pascale”, Naples, Italy. Medical Oncology Unit, Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 10 of 11 St. Salvatore Hospital, Department of Biotechnological and Applied Clinical Physiol Endocrinol Metab. 2008;295(5):E1076–83. https://doi.org/10.1152/ Sciences, University of L’Aquila, Via Vetoio, 67100 L’Aquila, Italy. ajpendo.90408.2008 Epub 2008 Sep 9. 23. Ouchi N, Parker JL, Lugus JJ, et al. Adipokines in inflammation and Received: 4 December 2018 Accepted: 5 February 2019 metabolic disease. Nat Rev Immunol. 2011;11(2):85–97. https://doi.org/10. 1038/nri2921. Epub 2011 Jan. 24. Jordan BF, Gourgue F, Cani PD. Adipose Tissue metabolism and Cancer progression: novel insights from gut microbiota? Curr References Pathobiol Rep. 2017;5(4):315–22. https://doi.org/10.1007/s40139-017- 1. Laviano A, Koverech A, Mari A. Cachexia: clinical features when 0154-6 Epub 2017 Oct 24. inflammation drives malnutrition. Proc Nutr Soc. 2015 Nov;74(4):348–54. 25. Magrone T, Jirillo E. Childhood obesity: immune response and nutritional 2. Taghizadeh N, Boezen HM, Schouten, et al. BMI and lifetime changes in BMI approaches. Front Immunol. 2015;6:76. https://doi.org/10.3389/fimmu.2015. and cancer mortality risk. PLoS One. 2015;10(4):e0125261. https://doi.org/10. 00076 eCollection 2015. 1371/journal.pone.0125261 eCollection 2015. 26. Seijkens T, Kusters P, Chatzigeorgiou A, Chavakis T, Lutgens E. Immune cell 3. Lennon H, Sperrin M, Badrick E, et al. The obesity paradox in Cancer: crosstalk in obesity: a key role for costimulation? Diabetes. 2014;63(12): a review. Curr Oncol Rep. 2016;18(9):56. https://doi.org/10.1007/ 3982–91. https://doi.org/10.2337/db14-0272. s11912-016-0539-4. 27. Han SJ, Glatman Zaretsky A, Andrade-Oliveira V, et al. White Adipose Tissue 4. Albiges L, Hakimi AA, Xie W, et al. Body mass index and metastatic renal cell Is a Reservoir for Memory T Cells and Promotes Protective Memory carcinoma: clinical and biological correlations. J Clin Oncol. 2016;34(30):3655–63. Responses to Infection. Immunity. 2017;47(6):1154–1168.e6. https://doi.org/ https://doi.org/10.1200/JCO.2016.66.7311. 10.1016/j.immuni.2017.11.009 Epub 2017 Dec 5. 5. Antoun S, Bayar A, Ileana E, et al. High subcutaneous adipose tissue predicts 28. Singh S, Facciorusso A, Singh AG, et al. Obesity and response to anti- the prognosis in metastatic castration-resistant prostate cancer patients in tumor necrosis factor-α agents in patients with select immune- post chemotherapy setting. Eur J Cancer. 2015;51(17):2570–7. https://doi. mediated inflammatory diseases: a systematic review and meta- org/10.1016/j.ejca.2015.07.042 Epub 2015 Aug 13. analysis. PLoS One. 2018;13(5):e0195123. https://doi.org/10.1371/ 6. McQuade JL, Daniel CR, Hess KR, et al. Association of body-mass index and journal.pone.0195123. outcomes in patients with metastatic melanoma treated with targeted 29. Mullen M, Gonzalez-Perez RR. Leptin-Induced JAK/STAT Signaling and therapy, immunotherapy, or chemotherapy: a retrospective, multicohort Cancer Growth. Vaccines (Basel). 2016;4(3). pii: E26. doi: https://doi.org/10. analysis. Lancet Oncol. 2018;19(3):310–22 Epub 2018 Feb 12. 3390/vaccines4030026. 7. Heidelberger V, Goldwasser F, Kramkimel N, et al. Sarcopenic overweight is 30. Sanlorenzo M, Vujic I, Daud A, et al. Pembrolizumab Cutaneous Adverse associated with early acute limiting toxicity of anti-PD1 checkpoint Events and Their Association With Disease Progression JAMA Dermatol. inhibitors in melanoma patients. Investig New Drugs. 2017;35(4):436–41. 2015;151(11):1206–12. https://doi.org/10.1001/jamadermatol.2015.1916. https://doi.org/10.1007/s10637-017-0464-x Epub 2017 Apr 10. 31. Freeman-Keller M, Kim Y, Cronin H, et al. Nivolumab in resected and 8. Richtig G, Hoeller C, Wolf M, et al. Body mass index may predict the Unresectable metastatic melanoma: characteristics of immune-related response to ipilimumab in metastatic melanoma: An observational multi- adverse events and association with outcomes. Clin Cancer Res. 2016;22(4): centre study. PLoS One. 2018 ;13(10):e0204729. https://doi.org/10.1371/ 886–94. https://doi.org/10.1158/1078-0432.CCR-15-1136. journal.pone.0204729. eCollection 2018.13. 32. Min Lee CK, Li S, Tran DC, et al. Characterization of dermatitis after PD-1/PD- 9. Wang Z, Aguilar EG, Luna JI, et al. Paradoxical effects of obesity on T cell L1 inhibitor therapy and association with multiple oncologic outcomes: A function during tumor progression and PD-1 checkpoint blockade. Nat retrospective case-control study. J Am Acad Dermatol. 2018. pii: S0190– Med. 2018. https://doi.org/10.1038/s41591-018-0221-5 [Epub ahead of print]. 9622(18)30831–4. doi: https://doi.org/10.1016/j.jaad.2018.05.035. [Epub 10. Minana B, Cozar JM, Palou J, et al. Bladder cancer in Spain 2011: population- ahead of print]. based study. J Urol. 2014;191(2):323–8. 33. Osorio JC, Ni A, Chaft JE, et al. Antibody-mediated thyroid dysfunction during 11. Ciocan D, Barbe C, Aubin F, et al. Distinctive features of melanoma and its T-cell checkpoint blockade in patients with non-small-cell lung cancer. Ann management in elderly patients: a population-based study in France. JAJA Oncol. 2017;28(3):583–9. https://doi.org/10.1093/annonc/mdw640. Dermatol. 2013;149(10):1150–7. 34. Haratani K, Hayashi H, Chiba Y, et al. Association of Immune-Related 12. Gridelli C, Balducci L, Ciardiello F, et al. Treatment of elderly patients with non- Adverse Events with Nivolumab Efficacy in non-small-cell lung Cancer. small-cell lung Cancer: results of an international expert panel meeting of the JAMA Oncol. 2018;4(3):374–8. https://doi.org/10.1001/jamaoncol.2017.2925. Italian Association of Thoracic Oncology. Clin Lung Cancer. 2015;16(5):325–33. 35. Teraoka S, Fujimoto D, Morimoto T, et al. Early immune-related adverse 13. Azawi NH, Joergensen SM, Jensen NV, et al. Trends in kidney cancer among events and association with outcome in advanced non-small cell lung the elderly in Denmark, 1980-2012. Acta Oncol. 2016;55(Suppl 1):79–84. Cancer patients treated with Nivolumab: a prospective cohort study. J 14. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation Thorac Oncol. 2017;12(12):1798–805. https://doi.org/10.1016/j.jtho.2017.08. criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J 022 Epub 2017 Sep 20. Cancer. 2009;45:228–47. 36. Conforti F, Pala L, Bagnardi V, et al. Cancer immunotherapy efficacy 15. Mantel N. Chi-square tests with one degree of freedom: extensions of the and patients' sex: a systematic review and meta-analysis. Lancet Mendel-Haenszel procedure. J Am Stat Assoc. 1963;58:690–700. Oncol. 2018;19(6):737–46. https://doi.org/10.1016/S1470-2045(18)30261- 16. Hosmer DW Jr, Lemeshow S, Sturdivant RX. Applied logistic regression. 4 Epub 2018 May 16. Third Edition. New Jersey: John Wiley & Sons (2013). 37. Botticelli A, Onesti CE, Zizzari I, et al. The sexist behaviour of immune 17. Kaplan EL, Meier P. Nonparametric estimation of incomplete observations. J checkpoint inhibitors in cancer therapy? Oncotarget. 2017;8(59):99336–46. Am Stat Assoc. 1958;53:457–81. https://doi.org/10.18632/oncotarget.22242. 18. Schemper M, Smith TL. A note on quantifying follow-up in studies of failure 38. Pallottini V, Bulzomi P, Galluzzo P, et al. Estrogen regulation of adipose time. Control Clin Trials. 1997;17:343–6. tissue functions: involvement of estrogen receptor isoforms. Infect Disord 19. Cox DR. Regression models and life tables (with discussion). J R Stat Soc Ser Drug Targets. 2008 Mar;8(1):52–60. B. 1972;74(187–200). 39. Khan D, Ahmed SA. The Immune System Is a Natural Target for 20. Goebell PJ, Müller L, Hübner A, et al. Body mass index as independent Estrogen Action: Opposing Effects of Estrogen in Two Prototypical predictor of overall survival in patients with advanced renal cell carcinoma Autoimmune Diseases. Front Immunol. 2015;6:635. https://doi.org/10. at start of systemic treatment-Analyses from the German clinical RCC- 3389/fimmu.2015.00635. Registry. Urol Oncol. 2018;(10):470.e1–9. https://doi.org/10.1016/j.urolonc. 40. Stocco C. Tissue Physiology and Pathology of Aromatase Steroids. 2012 Jan; 2018.07.007 Epub 2018 Aug 18. 77(1-2): 27–35. Published online 2011 Nov 13. doi: https://doi.org/10.1016/j. 21. Lutz CT, Quinn LS. Sarcopenia, obesity, and natural killer cell immune steroids.2011.10.013. senescence in aging: altered cytokine levels as a common mechanism. Aging (Albany NY). 2012;4(8):535–46. 41. Markle JG, Frank DN, Mortin-Toth S, Robertson CE, Feazel LM, Rolle- 22. Sutherland LN, Capozzi LC, Turchinsky NJ, et al. Time course of high-fat Kampczyk U, et al. Sex differences in the gut microbiome drive hormone- diet-induced reductions in adipose tissue mitochondrial proteins: dependent regulation of autoimmunity. Science (2013) 339(6123):1084–8. potential mechanisms and the relationship to glucose intolerance. Am J https://doi.org/10.1126/science.1233521. Cortellini et al. Journal for ImmunoTherapy of Cancer (2019) 7:57 Page 11 of 11 42. Yurkovetskiy L, Burrows M, Khan AA, Graham L, Volchkov P, Becker L, et al. Gender bias in autoimmunity is influenced by microbiota. Immunity. 2013; 39(2):400–12. https://doi.org/10.1016/j.immuni.2013.08.013. 43. Routy B, Le Chatelier E, Derosa L, et al. Gut microbiome influences efficacy of PD-1-based immunotherapy against epithelial tumors. Science. 2018; 359(6371):91–7. https://doi.org/10.1126/science.aan3706.

Journal

Journal for ImmunoTherapy of Cancer – Springer Journals

Published: Feb 27, 2019

There are no references for this article.