Cancer Immunology, Immunotherapy (2022) 71:45–55

https://doi.org/10.1007/s00262-021-02960-1

ORIGINAL ARTICLE

Predictive value of NLR, TILs (CD4+/CD8+) and PD‑L1 expression

for prognosis and response to preoperative chemotherapy in gastric
cancer
Ina Valeria Zurlo1 · Mattia Schino2 · Antonia Strippoli1 · Maria Alessandra Calegari1 · Alessandra Cocomazzi2 ·
Alessandra Cassano1,3 · Carmelo Pozzo1,3 · Mariantonietta Di Salvatore1 · Riccardo Ricci1,2 · Carlo Barone1,3 ·
Emilio Bria1,3 · Giampaolo Tortora1,3 · Luigi Maria Larocca · Michele Basso1 · Maurizio Martini1,2

Received: 4 January 2021 / Accepted: 6 May 2021 / Published online: 19 May 2021
© The Author(s) 2021

Abstract
The combination of perioperative chemotherapy plus complete surgical resection is currently accounted as the first-choice
strategy in patients with locally advanced Gastric Cancer (LAGC). Nevertheless, the partial response rate makes it necessary
to search biological parameters useful to select patients who would benefit most from neoadjuvant chemotherapy (NAD-CT).
We performed a retrospective analysis on a cohort of 65 LAGC cases, EBV negative and without MMR defect, submitted
to perioperative chemotherapy plus surgical resection. We evaluated the neutrophil-lymphocytes ratio (NLR) in peripheral
blood, the TILs density (reported as CD4/CD8 tissue ratio) and PD-L1 expression by immunohistochemistry on bioptic
tissues before the treatment. Results were correlated with the biological features, histological response (TRG) and clinical
outcome (PFS and OS). We found that NLR, TILs and PD-L1 expression showed a significant correlation with TNM stage,
lymphovascular invasion and response to NAD-CT (TRG). Correlating the NLR, TILs and PD-L1 expression with PFS and
OS, we found that patients with lower NLR levels (< 2.5 ratio), lower TILs (< 0.2 ratio) and higher PD-L1 level (CPS ≥ 1)
had a significantly better PFS and OS than those with higher NLR, higher TILs and lower PD-L1 expression (p < 0.0001).
Multivariate and multiple regression analyses confirmed the predictive and prognostic role of all three parameters, especially
when all three parameters are combined. Our study demonstrated that pre-treatment NLR, TILs and PD-L1 expression are
predictive and prognostic parameters in NAD-CT-treated LAGC suggesting a pivotal role of the systemic and tumor micro-
environment immunological profile in the response to chemotherapy.

Keywords Gastric cancer · Neoadjuvant chemotherapy · Predictive factors · Precision medicine · Immunological status

Abbreviations LACG​ Locally advanced gastric cancer

ALC Absolute lymphocytes count MSI Microsatellite instability
ANC Absolute neutrophil count MMR Mismatch repair
CPS Combined positive score NAD-CT Neoadjuvant chemotherapy
EBV Epstein–Barr virus NLR Neutrophil-lymphocytes ratio
GC Gastric cancer OS Overall survival
IHC Immunohistochemistry PFS Progression free survival
TIL Tumor-infiltrating lymphocytes
TME Tumor microenvironment
Ina Valeria Zurlo and Mattia Schino have contributed equally to TRG​ Tumor regression grade
this work.

Michele Basso and Maurizio Martini have shared senior

authorship.
Background

* Maurizio Martini Gastric cancer (GC) stands as the fifth most frequently diag-
maurizio.martini@unicatt.it nosed malignancy and the third leading cause of cancer death
Extended author information available on the last page of the article [1]; in Western countries, about two-thirds of GC patients are

13
Vol.:(0123456789)
46 Cancer Immunology, Immunotherapy (2022) 71:45–55

diagnosed with locally advanced cancer (LAGC) or metastatic Pathology of Fondazione Policlinico Universitario “Ago-
disease [2]. stino Gemelli”—IRCCS, Università Cattolica del Sacro
The combination of perioperative chemotherapy plus com- Cuore, Rome. Clinical and pathological records from 112
plete surgical resection (R0) is currently considered as the first- CT-naïve patients with LAGC treated with NAD-CT from
choice strategy to improve progression-free survival (PFS) and January 2012 to January 2017 were reviewed. Inclusion and
overall survival (OS) in LAGC patients [3–5]. exclusion criteria, chemotherapy regimen, post-treatment
Nevertheless, while it is true that latest years brought to follow-up examinations and other clinical parameters were
significant advances both in surgery and combined drug regi- reported in supplementary material. Finally, 65 patients who
mens, the overall response rate to chemotherapy is still less met the inclusion criteria were enrolled in this study, while
than 50%, keeping the prognosis rather dismal [3–5]. There- 47 patients were excluded. The main clinical and biological
fore, it would be advantageous to select patients who would characteristics of the cohort are shown in Table 1.
benefit most from neoadjuvant chemotherapy (NAD-CT) and
who might gain the best survival advantage [6, 7]. Blood parameters
In this way, several studies have highlighted that tumor
immune infiltrations [7–9], by defining neoplasms either as Venous blood samples were taken at diagnosis, before
immunologically “cold,” when they exhibit a low level of NAD-CT, and four weeks or more after the last dose
tumor infiltrating lymphocytes (TILs), or as immunologically of chemotherapy and within 1 week before the surgical
“hot,” when TILs’ level is high, could not only be a prognos- treatment. Hence, baseline or pre-treatment NLR (pre-
tic marker in GC in general terms—thus supporting the role NRL) and post-treatment NLR (post-NRL) were calculated
of immunotherapy (IT) in GC itself [10, 11]—but could also dividing the absolute neutrophil count (ANC) by the
play a role in predicting GC response to NAD-CT [7, 12, 13]. absolute lymphocyte count (ALC; NLR = ANC/ ALC).
However, this topic is still debated, since current literature data The cut-off values for white blood cells ANC (> 4000/mm3
are conflicting and ultimately leading to disagreement on both and ≤ 4000/mm3), ALC (> 1500/mm3 and ≤ 1500/mm3) and
type and association of biomarkers to be used to investigate NLR (> 2.5 and ≤ 2.5) were defined considering the median
GC-related “immunological status” [6, 7, 12]. values and data from previous studies [6, 7].
Recently, systemic immune-inflammation indexes, based
on routinely measurable peripheral blood parameters—such as HER2 gene amplification
neutrophil-to-lymphocyte ratio (NLR)—have been proposed
as manageable prognosticators and surrogates of cancer- HER2 amplification was performed using the INFORM
related inflammation in a variety of neoplasms [14], including HER2/neu Dual ISH DNA Probe Cocktail assay (Ventana
GC, also after NAD-CT treatment [6, 7, 15]. Medical Systems, Inc, Tucson, Arizona) as previously
Furthermore, lymphocyte sub-sets in GC have also been described (supplementary material) [20].
considered, hinting a significant correlation between TILs,
expressed as CD4+/CD8+ T cells tissue ratio, and survival TILs (CD4+/CD8+ T‑cells ratio) evaluation and PD‑L1
[12, 16, 17], while other studies focused on high microsatel- expression
lite instability, MSI-H and PD-L1 expression, as a potential
indicator of resistance to NAD-CT [12, 18, 19]. TILs were evaluated as the CD4+/CD8+ T-cells ratio. The
Based on the latest evidence, we performed a retrospective expression of CD4 + and CD8 + was assessed by immuno-
monocentric analysis, in a cohort of NAD-CT treated locally histochemistry (IHC; supplementary material) as previously
advanced GCs, EBV negative and with normal expression of described with few modifications [8, 9, 21].
mismatch repair (MMR) proteins, to evaluate the baseline PD-L1 expression was evaluated using immunohisto-
NRL, TILs density (reported as CD4+/CD8+ T cells tissue chemistry (IHC) and anti-PD-L1 rabbit monoclonal anti-
ratio) and PDL-1 expression, as indicators of the GC immuno- body (PD-L1 IHC 22C3 pharmDX; Agilent Technolologies,
logical status, and their correlation with the main clinical and Carpinteria, CA, USA). Detailed methods are reported in
biological features, including response to therapy. supplementary material. Samples were considered PD-L1
negative if CPS was less than 1 and PD-L1 positive if CPS
was 1 or more.
Methods
Statistical analysis
Patients’ features
The objective of this analysis was to explore the correla-
This exploratory monocentric retrospective analysis was tion between clinical and biological parameter, TILs den-
performed at the Unit of Medical Oncology and Unit of sity (reported as CD4/CD8 tissue ratio), PDL-1 expression

13
Cancer Immunology, Immunotherapy (2022) 71:45–55 47

Table 1  Patient characteristics and MedCalc version 10.2.0.0 (MedCalc Software, Mari-
n = 65 akerke, Belgium; supplementary material) [20, 22].
The evaluation of the tumor response to neoadjuvant
Age, mean (± SD) 63 (9.3) treatment was performed using Mandard’s classification sys-
Gender [n (%)] tem (Tumor Regression Grade, TRG) [23]. Responders were
Male 41 (63.1) defined as TRG 1–2, and non-responders were defined as
Female 24 (36.9) TRG 3–5 [24]. Progression Free Survival (PFS) and Overall
TNM stage* [n (%)] Survival (OS) were the survival endpoints. PFS was calcu-
IIIB 35 (53.85) lated as the time from NAD-CT beginning to any evidence
IIIC 30 (46.15) of disease progression (either local/regional relapses or
Tumor site [n (%)] distant metastases) or death, whichever occurred first. OS
Upper 25 (38.5) was calculated as the time from NAD-CT beginning to the
Middle 18 (27.7) patient’s death, due to any cause.
Lower 22 (33.8)
Histological subtype [n (%)]
Intestinal 31 (47.7) Results
Diffuse 34 (52.3)
HER2 status [n (%)] Patient characteristics and treatment response
HER2+ 15 (23.1)
HER2− 50 (76.9) Main clinicopathologic characteristics of our cohort (65
Response to NAD-CT [n (%)] consecutive patients with LAGC treated with NAD-CT) are
TRG 1–2 34 (52.3) reported in Table 1.
TRG 3–5 31 (47.7) Mean age at the time of diagnosis was 63 years, and 63%
PFS, mean (months, SD) 26.3 (18.1) of patients were male. Clinical TNM stage was IIIB in 35
OS, mean (months, SD) 34.6 (18.7) cases (53.85%) and stage IIIC in 30 cases (46.15%), respec-
LVI [n (%)] tively. Twenty-five patients out of 65 (38.5%) had an upper
Yes 31 (47.7) site located GC (Siewert type 2 or 3), 18 out of 65 (27.7%)
No 34 (52.3) had a middle site located GC, and 22 out of 65 (33.8%) a
Perineural infiltration (%) lower site GC. Histologically, 31 out of 65 (47.7%) had an
Yes 20 (30.8) intestinal subtype GC, while 34 out of 65 (52.3%) a diffuse
No 45 (69.2) subtype GC. HER2 amplification was found in 15 out of 65
NLR pre-chemotherapy (%) cases (23.1%), while 50 patients (76.9%) were negative. The
≥ 2.5 35 (53.8) post- NAD-CT response was evaluated according to Mand-
< 2.5 30 (46.2) ard-TRG pathological response system: 34 out of 65 patients
NLR post-chemotherapy (%) (55.4%) had a good tumor regression (TRG 1–2), while 29
≥ 2.5 29 (44.6) out of 65 (44.6%) had a poor tumor regression (TRG 3–5).
< 2.5 36 (55.4)
CD4+/CD8 + ratio (%) Blood neutrophil–lymphocyte ratio
≥ 0.2 32 (49.2)
< 0.2 33 (50.8) Mean pre-chemotherapy neutrophil and lymphocyte
PD-L1 expression, CPS (%) counts were 4940 and 1801 per m ­ m3, respectively. Mean
<1 34 (52.3) pre-chemotherapy NLR was 3.5 (range 0.57–3.8); by
≥1 31 (47.7) considering 2.5 as the cut-off value for NLR, as reported in
* According to the 8th edition of the American Joint Committee on the literature data [6], we found that 30 out of 65 patients
Cancer (AJCC) guidelines. SD, standard deviation. PFS, progression- (46.2%) had a low NLR ratio, while 35 patients (53.8%)
free survival. OS, overall survival. LVI, lympho-vascular invasion. had a high NLR ratio (Table 1). On the contrary, mean post-
chemotherapy neutrophil and lymphocyte counts were 4010
and 1460 per m ­ m3, respectively. Mean post-chemotherapy
NLR was 2.74 (range 0.72–12.6); by considering 2.5 as the
and NLR, in a cohort of patients with LAGC treated with cut-off value for NLR, we found that 36 out of 65 patients
NAD-CT. (55.4%) had a low NLR ratio, while 29 patients (44.6%)
Statistical analysis was performed using GraphPad- had a high NLR ratio (Table 1). Correlating the ANC and
Prism 5 software (Graph Pad Software, San Diego, CA) ALC, in the pre-treatment blood samples, with the clinical

13
48 Cancer Immunology, Immunotherapy (2022) 71:45–55

and biological parameters of our cohort, we found that pre-treatment NLR value (Spearman r = − 0.781; p < 0.0001;
male gender was significantly correlated with the ANC Fig. 2S, panel B) and TILs (Spearman r = − 0.567;
(p < 0.001; Table 1S), while all other parameters did not p < 0.0001; Fig. 2S, panel C).
have any significant association, either with ANC or with
ALC (Table 1S). Moreover, NLR in pre-treatment blood Prognostic variables for PFS and OS
samples showed a significant correlation with TNM stage
(p = 0.013), lymphovascular invasion (LVI, p < 0.001) and Mean PFS and mean OS were 26.3 months and 34.6 months,
TRG (p = 0.001; Table 2). Twenty-four patients with a low respectively. When we correlate the pre-treatment NLR with
baseline NLR level remained in this group after first-line PFS and OS, we found that patients with lower NLR levels
chemotherapy (Table 3). By contrast, 6 patients from this had a significantly better PFS and OS than those with higher
group were transferred to the high NLR level group after NLR levels (Fig. 2 panel A for PFS: median PFS for lower
NAD-CT. Twenty-three patients with a high baseline NLR pre-treatment NLR level 66 months versus median PFS for
level retained this level after first-line chemotherapy. By higher pretreatment NLR level 29 months, p < 0.0001, HR
contrast, 12 patients with a high baseline NLR level were 7.21, 95% CI from 3.03 to 17.12; Fig. 3 panel A for OS:
transferred to the low NLR level group after NAD-CT. median OS for lower pre-treatment NLR level 73 months
When we correlated the changing in the NLR level with versus median OS for higher pre-treatment NLR level 37
the tumor response to neoadjuvant treatment (TRG), we months, p < 0.0001, HR 6.60, 95% CI from 2.86 to 15.23). A
found that patients who remained in or were transferred to similar but lower significant result was observed correlating
the low NLR level subgroup following NAD-CT exhibited the post-treatment NLR with PFS and OS (Fig. 1S panel B
improved response, compared to patients who remained in for PFS: median PFS for lower post-treatment NLR level 61
or were transferred to the high NLR level group (Table 3). months versus median PFS for higher post-treatment NLR
Interestingly, we noted a significant decrement in the NRL level 27 months, p = 0.001, HR 4.20, 95% CI from 1.78 to
values between pre- and post-chemotherapy (p = 0.0033, 9.91; Fig. 1S panel C for OS: median OS for lower post-
paired t-test; Fig. 1S, panel A). treatment NLR level 64 months versus median OS for higher
post-treatment NLR level 37 months, p = 0.015, HR 2.79,
Immunohistochemistry for TILs (CD4+/CD8+ T cells 95% CI from 1.22 to 6.42).
ratio) in GC tissue Patients with lower TILs had a significant association
with better PFS and OS than those with higher CD4+/
Mean pre-chemotherapy CD4+/CD8+ T cells tissue ratio CD8+ T-cell tissue ratio (Fig. 2 panel B for PFS: median
(TILs) was 0.2 (range from 0.03 to 5.53, Fig. 1). Using the PFS for patients with lower CD4+/CD8+ T cells ratio 65
TILs’ mean value as the cut-off value, we found that 33 out months versus median PFS for patients with higher CD4+/
of 65 patients (50.9%) had a low TILs, while 32 patients CD8+ T cell ratio patients 18 months, p < 0.0001, HR 11.88,
(49.2%) had a high TILs (Table 1). Correlating TILs with 95% CI from 4.65 to 30.33; Fig. 3 panel B for OS: median
the clinical and biological parameters of our cohort, we OS for patients with lower CD4+/CD8+ T cell ratio 75
found that TNM stage (p = 0.025), LVI (p < 0.001) and TRG months versus median OS for patients with higher CD4+/
(p = 0.003) were significantly correlated with TILs (Table 2), CD8 + T cell ratio 27 months, p < 0.0001, HR 11.57, 95%CI
while none of the other parameters showed any significant from 4.76 to 28.12).
association (Table 2). Interestingly, we found that TILs had Moreover, higher PD-L1 level (CPS ≥ 1) in pretreated GC
a direct and significant correlation with pre-treatment NLR tissue was significantly associated with a better PFS and OS
value (Spearman r = 0.6338; p < 0.0001; Fig. 2S, panel A). in comparison with those with lower PD-L1 score (CPS < 1;
Fig. 2 panel C for PFS: median PFS for PD-L1 CPS ≥ 1
Evaluation of PD‑L1 expression (CPS) patients 20 months versus median PFS for PD-L1 CPS < 1
patients 67 months, p < 0.0001, HR 0.09924, 95%CI from
We found that 34 out of 65 patients (52.3%) had a low PD-L1 0.04029 to 0.2445; Fig. 3 panel C for OS: median OS for
expression (CPS < 1), while 31 patients (47.7%) had a high PD-L1 CPS ≥ 1 patients 28 months versus median OS for
PD-L1 level (CPS ≥ 1; Table 1 and Fig. 1). Correlating the PD-L1 CPS < 1 patients 74 months, p < 0.0001, HR 0.1154,
PD-L1 expression with the clinical and biological param- 95%CI from 0.04865 to 0.2737).
eters of our cohort, we found that LVI (p < 0.001; Table 2), In addition, Kaplan–Meir analysis showed that variables
stage (IIIC vs IIIB; p = 0.006; Table 2) and TRG (p < 0.001; predicting improved PFS and OS were TNM-IIIB (Table 2S,
Table 2) were significantly correlated with PD-L1 level, PFS, p = 0.0379; OS, p = 0.0174), absence of LVI (Table 2S,
while none of other parameters showed any significant PFS, p = 0.0515; OS, p = 0.0196) and TRG 1–2 (Table 2S,
association (Table 2). We also found that PD-L1 expres- PFS, p = 0.0244; OS, p = 0.0112). No correlation was found
sion had an indirect and significant correlation with both

13
 Table 2  Correlation between clinical and biological parameters with pretreatment NLR, CD4+/CD8 + T-cell tissue ratio and PD-L1 expression
Pre-treatment NLR CD4+/CD8+ T-cell tissue ratio PD-L1, CPS

≥ 2.5 (n; %) < 2.5 (n; %) p OR (95% CI) ≥ 0.2 (n; %) < 0.2 (n; %) p OR (95% CI) CPS ≥ 1 (n; %) CPS < 1 (n; %) p OR (95% CI)

Age
≥ 63 21; 53.85 18; 46.15 1.00 1.00 (0.37; 2.71) 19; 48.72 20; 51.28 1.00 0.97 (0.35; 2.56) 13; 50.00 13; 50.00 0.80 1.17 (0.43; 3.15)
< 63 14; 53.85 12; 46.15 13; 50.00 13; 50.00 18; 46.15 21; 53.85
Gender
Female 11; 45.83 13; 54.17 0.44 0.60 (0.22; 1.65) 10; 41.67 14; 58.33 0.44 0.78 (0.22; 1.71) 14; 58.33 10; 41.67 0.21 1.98 (0.71; 5.49)
Male 24; 58.54 17; 41.46 22; 53.66 19; 46.34 17; 41.46 24; 58,54
TNM stage
IIIB 24; 68.57 11; 31.43 0.013 3.77 (1.34; 10.56) 22; 62.86 13; 37.14 0.025 3.38 (1.22; 9.42) 20; 66.66 10; 33.34 0.006 4.36 (1.54; 12.37)
IIIC 11; 36.67 19; 63.33 10; 33.33 20; 66.67 11; 31.42 24; 68.58
Tumor site
Cancer Immunology, Immunotherapy (2022) 71:45–55

Upper 12; 48.00 13; 52.00 1.00 0.83 (0.35; 1.97) 11; 44.00 14; 56.00 0.60 0.84 (0.35; 2.03) 13; 52.00 12; 48.00 0.49 1.15 (0.48; 2.76)
Middle 9; 50.00 9; 50.00 7; 38.89 11; 61.11 10; 55.55 8; 44.45
Lower 14; 63.64 8; 36.36 14; 63.64 8; 36.36 8; 36.36 14; 63.64
Histotype
Intestinal 17; 54.84 14; 45.16 1.00 1.08 (0.41; 2.87) 15; 48.39 16; 51.61 1.00 0.94 (0.35; 2.48) 14; 45.16 17, 54.84 0.80 0.82 (0.31; 2.19)
Diffuse 18; 52.94 16; 47.06 17; 50.00 17; 50.00 17; 50.00 17; 50.00
HER2 status
HER2 + 11; 73.33 4; 26.67 0.14 2.98 (0.83; 10.63) 9; 60.00 6; 40.00 0.39 1.76 (0.54; 5.69) 4; 26.67 11; 73.33 0.08 0.31 (0.09; 1.11)
HER2 − 24; 48.00 26; 52 23; 46.00 27; 54.00 27; 54.00 23; 46.00
LVI
Yes 25; 80.65 6; 19.35 < 0.001 10 (3.14; 31.81) 23; 74.19 8; 25.81 < 0.001 7.98 (2.64; 24.19) 7; 22.58 24; 77.42 < 0.001 0.12 (0.04; 0.37)
No 10; 29.41 24; 70.59 9; 26.47 25; 73.53 24; 77.42 10; 22.58
PNI
Yes 14; 70.00 6; 30.00 0.11 2.67 (0.87; 8.18) 19; 42.22 26; 57.78 0.11 0.39 (0.13; 1.74) 7; 35.00 13; 65.00 0.19 0.47 (0.16; 1.40)
No 21; 46.67 24; 53.33 13; 65 7; 35 24; 5.33 21; 94.67
TRG​
TRG 3–5 21; 67.74 10; 32.26 0.001 5.83 (1.99; 17.03) 22; 70.97 9; 29.03 0.003 5.11 (1.77; 14.72) 9; 29.03 22; 70.97 < 0.001 0.15 (0.05; 0.44)
TRG 1–2 9; 26.47 25; 73.53 11; 32.35 23; 67.65 25; 73.53 9; 26.47
PFS, months
≥ 26.3 4; 16.67 20; 83.33 < 0.001 0.06 (0.18; 0.23) 9; 35 27; 75 < 0.001 0.09 (0.03; 0.28) 17; 85.00 3; 15.00 < 0.001 12.55 (3.15; 49.9)
< 26.3 31; 75.61 10; 24.39 23; 79.31 6; 20.69 14; 31.11 31; 68.89
OS, months
≥ 34.6 4; 17.39 19; 82.61 < 0.001 0.07 (0.21; 0.27) 1; 4.35 22; 95.65 < 0.001 0.02 (0.001; 0.13) 20; 80.00 5; 20.00 < 0.001 10.55 (3.17; 35.0)
< 34.6 31; 73.81 11; 26.19 31; 73.81 11; 26.19 11; 27.50 29; 72.50

LVI lymphovascular invasion, PNI perineural infiltration

Bold indicates the only significant parameters

13
49
50 Cancer Immunology, Immunotherapy (2022) 71:45–55

Table 3  Relationship between Pre-chemotherapy Post-chemotherapy TRG 1–2 TRG 3–5 OR p value
changes in the NLR level and (n = 34) (n = 31)
tumor response to neoadjuvant
treatment (TRG) NLR < 2.5 (n = 30) NLR < 2.5 (n = 24) 15 9 21.21 0.017
NLR ≥ 2.5 (n = 6) 0 6
NLR ≥ 2.5 (n = 35) NLR < 2.5 (n = 12) 11 1 20.63 0.002
NLR ≥ 2.5 (n = 23) 8 15

Fig. 1  The figure shows two LAGC samples of intestinal and diffuse fication, bare scale 150 mm; the box in the panel C shows a detail
histotype cancer (panel A and E, respectively, E&E, ×200 magnifica- of the image at ×400 magnification, where the arrows indicate the
tion, bare scale 150 mm); the CD4 + and CD8 + cell counts (figure B CD4 + cells); the PD-L1 expression, evaluated as CPS score (figure
and F for CD8 + cells; figure C and G for CD4 + cells; ×200 magni- D, CPS ≥ 1 and H, CPS < 1; × 200 magnification, bare scale 150 mm)

Fig. 2  Kaplan–Meier curves for PFS of NAD-CT-treated LAGC with PD-L1 CPS ≥ 1 (blue-line) were significantly associated with
patients stratified by pre-treatment NLR (panel A), CD4+/CD8+ T a better PFS (p < 0.0001) respect to those patients with NLR ≥ 2.5,
cells tissue ratio (TILs; panel B) and PD-L1 expression (evaluated as TILs ≥ 0.2 and with PD-L1 CPS < 1 (red-line)
CPS score; panel C). Patients with lower NLR < 2.5, TILs < 0.2 and

with gender, age, tumor site, histological subtypes, HER2 p = 0.0013 for PD-L1 expression; p = 0.0036 for pretreat-
expression nor perineural infiltration (Table 2S). ment NLR). Similarly, multivariate analysis of OS, includ-
Multivariate analysis of PFS including pre-treatment ing pre-treatment NLR, TILs and PD-L1 expression, stage,
NLR, TILs and PD-L1 expression, stage, histological histological subtypes, TRG and lymphovascular invasion,
subtypes, TRG and lymphovascular invasion showed showed that the independent prognostic variables were
that NLR, TILs, PD-L1 expression before NAD-CT pretreatment NLR, TILs and PD-L1 expression (Table 4;
were significant predictors (Table 4; p = 0.0001 for TILs;

13
Cancer Immunology, Immunotherapy (2022) 71:45–55 51

Fig. 3  Kaplan–Meier curves for OS of NAD-CT-treated LAGC with PD-L1 CPS ≥ 1 (blue-line) were significantly associated with
patients stratified by pre-treatment NLR (panel A), CD4+/CD8+ T a better OS (p < 0.0001) respect to those patients with NLR ≥ 2.5,
cells tissue ratio (TILs; panel B) and PD-L1 expression (evaluated as TILs ≥ 0.2 and with PD-L1 CPS < 1 (red-line)
CPS score; panel C). Patients with lower NLR < 2.5, TILs < 0.2 and

Table 4  Multivariate analysis b SE Wald p Exp(b) 95% CI of Exp(b)

for PFS and OS
Covariate for PFS
CD4+/CD8+ ratio 2.7528 0.7025 15.3540 0.0001 15.6868 3.9584–62.1645
PD-L1 (CPS) 2.1720 0.6765 10.3092 0.0013 0.1139 0.0303–0.4291
Pre-treatment NLR 1.8501 0.6357 8.4692 0.0036 6.3605 1.8295–22.1123
Covariate for OS
CD4+/CD8+ ratio 2.8712 0.8072 12.6524 0.0004 17.6577 3.6295–85.9058
PD-L1 (CPS) 2.5776 0.7249 12.6431 0.0004 0.0760 0.0183–0.3145
Pre-treatment NLR 2.0307 0.6266 10.5042 0.0012 7.6195 2.2314–26.0178

b = coefficient estimates; SE = standard error for coefficient estimates b; Exp(b) = Hazard Ratio value;
95%CI of Exp(b) = 95% confidence interval of Hazard Ratio

p = 0.0004 for both TILs and PD-L1 expression; p = 0.0012 and OS (both p < 0.0001), also finding a significant associa-
for pretreatment NLR). tion between low NLR and TRG (p = 0.001) and TNM stage
The predictive value for PFS and OS of each parame- (p = 0.013) and LVI (p ≤ 0.001). At the same time, we evalu-
ter (pre-treatment NLR, PD-L1 expression and TILs) was ated the peritumoral microenvironment, where the immune
evaluated, first individually, then combined (two or three host cells (mainly lymphocytes) strictly interacted with neo-
parameters), performing a multiple regression analysis. The plasm, demonstrating, in agreement with other authors [12,
predictive value was significantly higher only when the three 16, 17], the significant association with low CD4+/CD8+ T
parameters were considered jointly combined (p < 0.0001 cell ratio with PFS, OS (both p < 0.001), and the pathologi-
for both PFS and OS; r partial 0.7329 for PFS and r partial cal response to NAD-CT (as TRG) in our cohort (p = 0.003).
0. 6157 for OS). Similarly, investigating the expression of PD-L1, we also
found, for the first time, that higher level of PD-L1 is associ-
ated with better PFS and OS (both p < 0.001) and response
Discussion to neoadjuvant therapy in LAGC (p < 0.001). In addition,
higher level of CPS (CPS ≥ 1) also had significant associa-
This monocentric study retrospectively investigated the tion with TNM stage (p = 0.006), TRG (p < 0.001) and LVI
relationship between systemic and tumor microenvironment (p ≤ 0.001). Lastly, we found that post-treatment NLR levels
(TME) immunological profile in patients with LAGC before were consistent with chemotherapeutic efficacy and clinical
receiving NAD-CT and their clinicopathologic outcome to outcome, suggesting NLR levels following treatment, though
identify some parameters which could help in selecting less significantly than pre-treatment NRL, may also provide
patients who might respond to CT. valuable prognostic and predictive information.
According to recent evidence [6, 7, 15], we demonstrated Several studies highlighted that the tumor immune status
low NLR in the peripheral blood of pre-treated NAD-CT plays a role also in response to neoadjuvant radio-chemo-
LAGC was significantly associated with a favorable PFS therapy treatments in several human cancers [6–8, 12, 17,

13
52 Cancer Immunology, Immunotherapy (2022) 71:45–55

18, 25, 26]. Accordingly, considering the need to identify proteins in gastric tumors is due to the tumor infiltration
and select those patients who would benefit most from a of effector T cells (especially the CD8 + cells), defined as
neoadjuvant treatment, recent studies have tried to use dif- “T cell inflamed phenotype,” that in turn determine the
ferent immune parameters as indicators of response to neo- upregulation of immune checkpoints, and not to a genomic
adjuvant treatments, sometimes finding inconsistent results, instability [29]. This way, the PD-L1 target in gastric can-
especially regarding the role of T cells subtypes [8–12, 17, cer could only be clinically effective (also in NAD-CT)
18, 27]. This probably depends on different factors such as for the subgroup of tumors that contain tumor-infiltrating
the inhomogeneity of the analyzed cohorts, different neoad- immune cells and could explain the controversial results in
juvant scheme treatments, as well as the use of different and the predictive effects of PD-L1 in response to PD-1/PD-L1
single parameter to investigate the immune status which is antibodies in GC [32].
inherently complex and variegated, and whose function is Although over the last year, the standard NAD-CT in
scarcely reducible to only one specific parameter. LAGC changed, after the publication of the FLOT4-trial
Investigating three immunological prechemotherapy data [33], we hypothesized that pre-treatment NLR, TILs
parameters, we found that the latter, besides being signifi- and PD-L1 expression, even if requires demonstrating, could
cantly correlated with better PFS and OS (both p < 0.005) be a predictive and prognostic parameter also in this new
in univariate and multivariate analysis, had likewise a direct fluorouracil-based regimen.
and significant correlation with each other. In fact, NLR, Unlike Wang et al. [34] who described a significant asso-
an expression of the immunosurveillance capacity of the ciation between lower PD-L1 expression and HER2 gene
host, the presence of higher levels of CD8 + in the tumor amplification in GC, we did not find any significant associa-
microenvironment, as an index of a patient’s better immune- tion between this molecular feature and pre-treatment NLR,
response, and PD-L1 expression, indicating the tumor’s TILs and PD-L1 expression, suggesting that this alteration
intrinsic immune-escape capability and consequently also did not have a predictive role in NAD-CT-treated LAGC.
the TME CD8 + T cells level, source of cytokines such as In addition, our data have shown that both the NRL levels
interferon gamma that induced the expression of PD-L1 [28, before and after NAD-CT have a predictive and prognostic
29], mirrored some crucial aspects of the tumor/immunity value in patients with LAGC. However, the post-treatment
interaction which, if considered together, better select LAGC NRL shows less significance than the pre-treatment NRL.
patients who will benefit most from NAD-CT treatment. This could be partly explained by the bone morrow sup-
In fact, we demonstrated that joining the aforementioned pression of the NAD-CT and other biological parameters,
parameters with one another makes stronger correlation as already reported by several authors [9, 35, 36].
between pre-chemotherapy immune status and clinical out- Heterogeneity in the immunohistochemical assessment
comes (p < 0.0001). for PD-L1 and TILs (CD4 and CD8) expression both within
Although the chemotherapy tends to deeply modify the and between tumor sites is a well-documented phenomenon
host immunity, often with detrimental and myelosuppressive that could have important implications, especially for PD-L1
effects, accumulating evidence indicates that the efficacy of accuracy as a predictive biomarker [37–39]. In addition,
conventional anticancer agents does not only involve direct other factors, such as the use of different antibodies, of dif-
cytostatic/cytotoxic effects, but also relies on the (re)activa- ferent cut-off, of different immunohistochemical platforms
tion of tumor-targeting immune responses, similarly to the and the inter-observer variability could play a role in the
abscopal effect due to the radiotherapy [30]. This chemo- evaluation of these two parameters [37, 38]. Notwithstand-
therapeutic effect might act in synergy with a most reactive, ing, the immunohistochemical assay for PD-L1 and TILs
less depressive immune system and higher TME CD8 + [29], remains today one of the most used markers, especially for
whose pre-chemotherapy status could be assessed with NLR, immunotherapy, and several studies have shown that it is
TILs and PD-L1 expression. possible to reduce the impact of the heterogeneity of expres-
By ruling out both altered-MMRP carriers and EBV sion in the evaluation of these two parameters. In this way,
positive tumors, we managed to minimize confounding we perform the PD-L1 and TILs expression on baseline
effects of other tumor variables on the outcomes and fluo- multisite sampling tumor tissue biopsies (median 4.02, with
rouracil-based NAD-CT [18, 31]. Interestingly, PD-L1 at least 50% tumor tissue), using the same antibodies and
expression, which was typically associated with MSI-H immunohistochemical platform, following standardized and
status and maybe indirectly to the fluorouracil-based well-defined conditions, and calculating the agreement indi-
NAD-CT outcome, as demonstrated in several tumors, ces (Cohen’s K index) between the two pathologists (M.M.
showed here a significant correlation with PFS, OS and and R.R., who also went through a formal training program
response to therapy only in LAGC patients displayed to evaluate CPS by the 22C3 pharmDx assay) in the evalu-
higher levels of this immune-suppressing protein. Prob- ation of TILs and PD-L1 expression (that were very good:
ably, the upregulation of negative immune checkpoint k = 0.82 and k = 0.87, respectively).

13
Cancer Immunology, Immunotherapy (2022) 71:45–55 53

Finally, although the evaluation of NRL, TIL and PD-L1 Declarations

parameters can be an easily executable analysis and several
studies have shown that these parameters, even if individu- Conflict of interest The authors declare that they have no competing
ally, could be useful in the selection of LAGC patients for interests.
NAD-CT; however, it appears necessary to better under- Ethics approval and consent to participate Written informed consent
stand the molecular mechanisms underlying this association, was provided by the patient prior to the first study specific interven-
before these being adopting in clinical practice. To this end, tion. Ethical approval for study was provided by the ethics commit-
in-vitro models such as co-cultures of patient’ immune cells tee of Università Cattolica del Sacro Cuore, Roma (Roma PROT. N.
0033623/18 and 29042/18). The report does not present identifying
and GC organoid could represent a valid model, in which images or other personal or clinical details of participants that com-
different types of neoadjuvant treatment could be tested, also promise anonymity.
in combination with immunotherapy [40].
Consent for publication All authors reached an agreement to publish
The main limitations of our study are the retrospective
the study in this journal.
monocentric design and the relatively small though homoge-
neous-treated cohort of patients, so that our data need to be
Open Access This article is licensed under a Creative Commons Attri-
confirmed in other independent studies including NAD-CT-
bution 4.0 International License, which permits use, sharing, adapta-
treated LAGC patients, and our immune parameters should tion, distribution and reproduction in any medium or format, as long
be validated as a prognostic tool. as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated
Conclusions otherwise in a credit line to the material. If material is not included in
the article’s Creative Commons licence and your intended use is not
In summary, we propose that pre-treatment NLR, TILs and permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
PD-L1 expression may be predictive and prognostic param-
copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.
eters in NAD-CT-treated LAGC suggesting a pivotal role of
the tumor inflammatory microenvironment in the response to
chemotherapy. These three parameters, as indicators of the
References
patient tumor/immune system status, significantly correlated
each other and could help the clinicians to recognize patients 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A
who might benefit from a NAD-CT with LAGC. (2018) Global cancer statistics 2018: GLOBOCAN estimates of
incidence and mortality worldwide for 36 cancers in 185 coun-
tries. CA A Cancer J Clin 68:394–424
2. Orditura M, Galizia G, Sforza V, Gambardella V, Fabozzi A,
Supplementary Information The online version contains supplemen-
Laterza MM et al (2014) Treatment of gastric cancer. World J
tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 00262-0​ 21-0​ 2960-1.
Gastroenterol 20:1635–1649
3. Bang YJ, Kim YW, Yang HK, Chung HC, Park YK, Lee KH et al
Acknowledgements We thank Dr. Sara Capodimonti and Dr. Roberta (2012) Adjuvant capecitabine and oxaliplatin for gastric cancer
Benvenuto for their technical support. after D2 gastrectomy (classic): a phase 3 open-label, randomised
controlled trial. Lancet 379:315–321
Authors’ contribution IVZ, MS, MB and MM were the principal 4. Fuse N, Bando H, Chin K, Ito S, Yoshikawa T, Tsuburaya A et al
authors and the main contributors in writing the manuscript. AS, (2017) Adjuvant capecitabine plus oxaliplatin after D2 gastrec-
MAC, AC, AC and MDS analyzed and interpreted the patient data and tomy in Japanese patients with gastric cancer: a phase II study.
reviewed the literature. MM, AC, RR and IVZ performed the immuno- Gastric Cancer 20:332–340
histochemistry analysis. EB, CB and GT read and corrected the manu- 5. Fuentes E, Ahmad R, Hong TS, Clark JW, Kwak EL, Rattner DW
script. RR and LML corrected the English language errors. All authors et al (2016) Adjuvant therapy completion rates in patients with
read and approved the final manuscript. gastric cancer undergoing perioperative chemotherapy versus a
surgery-first approach. J Gastrointest Surg 20:172–179
Funding Open access funding provided by Università Cattolica del 6. Jin H, Zhang G, Liu X, Liu X, Chen C, Yu H et al (2013) Blood
Sacro Cuore within the CRUI-CARE Agreement. This study was sup- neutrophil-lymphocyte ratio predicts survival for stages III–IV
ported by Università Cattolica del Sacro Cuore, Fondi d’Ateneo, Linea gastric cancer treated with neoadjuvant chemotherapy. World J
D1 (2017 and 2018; MM). Surg Oncol 11:112
7. He Q, Li G, Ji X, Ma L, Wang X, Li Y et al (2017) Impact of
Availability of data and materials The datasets used and/or analyzed the immune cell population in peripheral blood on response and
during the current study are available from the corresponding author survival in patients receiving neoadjuvant chemotherapy for
on reasonable request. advanced gastric cancer. Tumor Biol 5:1–8
8. Fridman WH, Galon J, Dieu-Nosjean M, Cremer I, Fisson S,
Damotte D et al (2010) Immune infiltration in human cancer:
prognostic significance and disease control. In: Dranoff G (ed)
Cancer immunology and immunotherapy, current topics in micro-
biology and immunology, vol 344. Springer, Berlin, Heidelberg

13
54 Cancer Immunology, Immunotherapy (2022) 71:45–55

9. Kim JW, Nam KH, Ahn S, Park DJ, Kim H, Kim SH et al (2016) lymphocytes in patients with pancreatic ductal adenocarcinoma
Prognostic implications of immunosuppressive protein expression treated with neoadjuvant chemotherapy. Pancreas 46:1180–1187
in tumors as well as immune cell infiltration within the tumor 27. Li Y, Wei Y, He Q, Wang X, Fan C, Li G (2018) Clinicopathologi-
microenvironment in gastric cancer. Gastric Cancer 19:42–52 cal and prognostic significance of high circulating lymphocyte
10. Zhou Y, Zhu G, Lu X, Zheng KI, Wang Q, Chen J et al (2020) ratio in patients receiving neoadjuvant chemotherapy for advanced
Identification and validation of tumour microenvironment-based gastric cancer. Sci Rep 8:6223
immune molecular subgroups for gastric cancer: immunothera- 28. Taube JM, Anders RA, Young GD, Xu H, Sharma R, McMiller
peutic implications. Cancer Immunol Immunother 69:1057–1069 TL et al (2012) Colocalization of inflammatory response with
11. Kelly RJ (2017) Immunotherapy for esophageal and gastric can- B7–h1 expression in human melanocytic lesions supports an
cer. In: Dizon DS (ed) American society of clinical oncology edu- adaptive resistance mechanism of immune escape. Sci Trans Med
cational book, vol 37. pp 292–300 4:127ra37
12. Yu Y, Ma X, Zhang Y, Zhang Y, Ying J, Zhang W et al (2019) 29. Thompson ED, Zahurak M, Murphy A, Cornish T, Cuka N,
Changes in expression of multiple checkpoint molecules and infil- Abdelfatah E et al (2017) Patterns of PD-L1 expression and CD8
tration of tumor immune cells after neoadjuvant chemotherapy in T cell infiltration in gastric adenocarcinomas and associated
gastric cancer. J Cancer 10:2754–2763 immune stroma. Gut 66:794–801
13. Li Z, Gao X, Peng X, Chen MM, Zhe Li, Wei B et al (2020) 30. Galluzzi L, Buqué A, Kepp O, Zitvogel L, Kroemer G (2015)
Multi-omics characterization of molecular features of gastric can- Immunological effects of conventional chemotherapy and targeted
cer correlated with response to neoadjuvant chemotherapy. Sci anticancer agents. Cancer Cell 28:690–714
Adv 6:e4211 31. Saito R, Abe H, Kunita A, Yamashita H, Seto Y, Fukayama M
14. Ocana A, Nieto-Jiménez C, Pandiella A, Templeton AJ (2017) (2017) Overexpression and gene amplification of PD-L1 in can-
Neutrophils in cancer: prognostic role and therapeutic strategies. cer cells and PD-L1+ immune cells in Epstein–Barr virus-asso-
Mol Cancer 16:137 ciated gastric cancer: the prognostic implications. Mod Pathol
15. Mellor KL, Powell AGMT, Lewis WG (2018) Systematic review 30:427–439
and meta-analysis of the prognostic significance of neutrophil- 32. Gu L, Chen M, Guo D, Zhu H, Zhang W, Pan J et al (2017) PD-L1
lymphocyte ratio (NLR) after r0 gastrectomy for cancer. J Gas- and gastric cancer prognosis: a systematic review and meta-anal-
trointest Cancer 49:237–244 ysis. PLoS ONE 12:e0182692
16. Fukuda K, Tsujitani S, Maeta Y, Yamaguchi K, Ikeguchi M, Kai- 33. Al-Batran SE, Homann N, Pauligk C, Goetze TO, Meiler J, Kasper
bara N (2002) The expression of RCAS1 and tumor infiltrating S et al (2019) Perioperative chemotherapy with fluorouracil plus
lymphocytes in patients with T3 gastric carcinoma. Gastric Cancer leucovorin, oxaliplatin, and docetaxel versus fluorouracil or
5:220–227 capecitabine plus cisplatin and epirubicin for locally advanced,
17. Li F, Sun Y, Huang J, Xu W, Liu J, Yuan Z (2019) CD4/CD8 + resectable gastric or gastro-oesophageal junction adenocarcinoma
T cells, DC subsets, Foxp3, and IDO expression are predictive (FLOT4): a randomised, phase 2/3 trial. Lancet 393:1948–1957
indictors of gastric cancer prognosis. Cancer Med 8:7330–7344 34. Wang L, Zhang Q, Ni S, Tan C, Cai X, Huang D et al (2018) Pro-
18. Hashimoto T, Kurokawa Y, Takahashi T, Miyazaki Y, Tanaka grammed death-ligand 1 expression in gastric cancer: correlation
K, Makino T et al (2019) Predictive value of MLH1 and PD-L1 with mismatch repair deficiency and Her2-negative status. Cancer
expression for prognosis and response to preoperative chemo- Med 7:2612–2620
therapy in gastric cancer. Gastric Cancer 22:785–792 35. Wang F, Liu Z, Xia Y, Zhou C, Shen X, Li X et al (2015) Changes
19. Wang Y, Zhu C, Song W, Li J, Zhao G, Cao H (2018) PD-L1 in neutrophil/lymphocyte and platelet/lymphocyte ratios after
expression and CD8+ T Cell infiltration predict a favorable prog- chemotherapy correlate with chemotherapy response and pre-
nosis in advanced gastric cancer. J Immunol Res 2018:4180517 diction of prognosis in patients with unresectable gastric cancer.
20. Strippoli A, Cocomazzi A, Basso M, Cenci T, Ricci R, Pierconti Oncol Lett 10:3411–3418
F et al (2020) c-MYC expression is a possible keystone in the 36. Li Z, Li S, Ying S, Zhang L, Shan F, Jia Y et al (2020) The
colorectal cancer resistance to EGFR inhibitors. Cancers 12:E638 clinical value and usage of inflammatory and nutritional markers
21. Rossi ED, Martini M, Capodimonti S, Cenci T, Straccia P, in survival prediction for gastric cancer patients with neoadju-
Angrisani B et al (2014) Analysis of immunocytochemical and vant chemotherapy and D2 lymphadenectomy. Gastric Cancer
molecular BRAF expression in thyroid carcinomas: a cytohisto- 23:540–549
logic institutional experience. Cancer Cytopathol 122:527–535 37. Elfving H, Mattsson JSM, Lindskog C, Backman M, Menzel U,
22. Martini M, Cenci T, D’Alessandris GQ, Cesarini V, Cocomazzi A, Micke P (2019) Programmed cell death ligand 1 immunohis-
Ricci-Vitiani L et al (2013) Epigenetic silencing of Id4 identifies tochemistry: a concordance study between surgical specimen,
a glioblastoma subgroup with a better prognosis as a consequence biopsy, and tissue microarray. Clin Lung Cancer 20:258–262
of an inhibition of angiogenesis. Cancer 119:1004–1012 38. Ye M, Huang D, Zhang Q, Weng W, Tan T, Qin G et al (2020)
23. Zhu Y, Sun Y, Hu S, Jiang Y, Yue J, Xue X et al (2017) Compari- Heterogeneous programmed death-ligand 1 expression in gastric
son of five tumor regression grading systems for gastric adenocar- cancer: comparison of tissue microarrays and whole sections.
cinoma after neoadjuvant chemotherapy: a retrospective study of Cancer Cell Int 24(20):186
192 cases from National Cancer Center in China. BMC Gastro- 39. Lee JS, Won HS, Sun DS, Hong JH, Ko YH (2018) Prognostic role
enterol 17:41 of tumor-infiltrating lymphocytes in gastric cancer: a systematic
24. Noble F, Lloyd MA, Turkington R, Griffiths E, O’Donovan M, review and meta-analysis. Medicine (Baltimore) 97:e11769
O’Neill JR et al (2017) Multicentre cohort study to define and val- 40. Chakrabarti J, Holokai L, Syu L, Steele N, Chang J, Dlugosz
idate pathological assessment of response to neoadjuvant therapy A et al (2018) Mouse-derived gastric organoid and immune cell
in oesophagogastric adenocarcinoma. Br J Surg 104:1816–1828 co-culture for the study of the tumor microenvironment. Methods
25. Corbeau I, Jacot W, Guiu S (2020) Neutrophil to lymphocyte ratio Mol Biol 1817:157–168
as prognostic and predictive factor in breast cancer patients: a
systematic review. Cancers 12:958 Publisher’s Note Springer Nature remains neutral with regard to
26. Nejati R, Goldstein JB, Halperin DM, Wang H, Hejazi N, Rashid jurisdictional claims in published maps and institutional affiliations.
A et al (2017) Prognostic significance of tumor-infiltrating

13
Cancer Immunology, Immunotherapy (2022) 71:45–55 55

Authors and Affiliations

Ina Valeria Zurlo1 · Mattia Schino2 · Antonia Strippoli1 · Maria Alessandra Calegari1 · Alessandra Cocomazzi2 ·
Alessandra Cassano1,3 · Carmelo Pozzo1,3 · Mariantonietta Di Salvatore1 · Riccardo Ricci1,2 · Carlo Barone1,3 ·
Emilio Bria1,3 · Giampaolo Tortora1,3 · Luigi Maria Larocca · Michele Basso1 · Maurizio Martini1,2

Ina Valeria Zurlo Carlo Barone

valeriazurlo26@gmail.com carloantonio.barone@policlinicogemelli.it
Mattia Schino Emilio Bria
mattia.schino01@icatt.it emilio.bria@unicatt.it
Antonia Strippoli Giampaolo Tortora
strefoantonia@gmail.com giampaolo.tortora@unicatt.it
Maria Alessandra Calegari Luigi Maria Larocca
mariaalessandra.calegari@policlinicogemelli.it luigimaria.larocca@unicatt.it
Alessandra Cocomazzi Michele Basso
alessandra.cocomazzi@libero.it michele.basso@policlinicogemelli.it
Alessandra Cassano 1
Fondazione Policlinico Universitario A. Gemelli IRCCS,
alessandra.cassano@unicatt.it
Largo A. Gemelli 8, 00168 Rome, Italy
Carmelo Pozzo 2
Division of Pathology, Department of Health Science
carmelo.pozzo@unicatt.it
and Public Health, Università Cattolica del Sacro Cuore,
Mariantonietta Di Salvatore Largo Francesco Vito 1, 00168 Rome, Italy
disalvatore.mariantonietta@gmail.com 3
Division of Oncology, Department of Medicine
Riccardo Ricci and Translational Surgery, Università Cattolica del Sacro
riccardo.ricci@unicatt.it Cuore, Largo F. Vito 1, 00168 Rome, Italy

13