OncoImmunology

ISSN: (Print) 2162-402X (Online) Journal homepage: https://www.tandfonline.com/loi/koni20

Predictive role of plasmatic biomarkers in

advanced non-small cell lung cancer treated by
nivolumab

Adrien Costantini, Catherine Julie, Coraline Dumenil, Zofia Hélias-Rodzewicz,

Julie Tisserand, Jennifer Dumoulin, Violaine Giraud, Sylvie Labrune, Thierry
Chinet, Jean-François Emile & Etienne Giroux Leprieur

To cite this article: Adrien Costantini, Catherine Julie, Coraline Dumenil, Zofia Hélias-Rodzewicz,
Julie Tisserand, Jennifer Dumoulin, Violaine Giraud, Sylvie Labrune, Thierry Chinet, Jean-
François Emile & Etienne Giroux Leprieur (2018) Predictive role of plasmatic biomarkers in
advanced non-small cell lung cancer treated by nivolumab, OncoImmunology, 7:8, e1452581, DOI:
10.1080/2162402X.2018.1452581

To link to this article: https://doi.org/10.1080/2162402X.2018.1452581

View supplementary material Published online: 20 Apr 2018.

Submit your article to this journal Article views: 1098

View related articles View Crossmark data

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=koni20
ONCOIMMUNOLOGY
2018, VOL. 7, NO. 8, e1452581 (11 pages)
https://doi.org/10.1080/2162402X.2018.1452581

ORIGINAL RESEARCH

Predictive role of plasmatic biomarkers in advanced non-small cell lung cancer

treated by nivolumab
Adrien Costantinia,b, Catherine Juliea,c, Coraline Dumenila,b, Zofia He
lias-Rodzewicza,c, Julie Tisseranda,c,
Jennifer Dumoulina,b, Violaine Girauda,b, Sylvie Labrunea,b, Thierry Chineta,b, Jean-François Emilea,c,
and Etienne Giroux Leprieura,b
a
EA4340, UVSQ, Paris-Saclay University, Boulogne-Billancourt, France; bDepartment of Respiratory Diseases and Thoracic Oncology, APHP – Ambroise
Pare Hospital, Boulogne-Billancourt, France; cDepartment of Pathology, APHP – Ambroise Pare Hospital, Boulogne-Billancourt, France

ABSTRACT ARTICLE HISTORY

Immune checkpoint inhibitors, as nivolumab, are used in advanced non-small cell lung cancer (NSCLC). Received 11 January 2018
However, no associated biomarker is validated in clinical practice with this drug. We investigated herein Revised 7 March 2018
immune-related blood markers in patients with advanced NSCLC treated with nivolumab. Plasma of 43 Accepted 9 March 2018
consecutive patients were prospectively collected at time of the diagnosis of cancer, at the initiation of KEYWORDS
nivolumab and at the first tumour evaluation (2 months). Concentrations of PD-L1 (sPD-L1), soluble PD-L2 nivolumab; non-small cell
(sPD-L2), Interleukine-2 (sIl-2), Interferon-gamma (sIFN-g), and Granzyme B (sGranB) were quantified by lung cancer; plasma; PD-L1;
ELISA. Cell free RNA was quantified by Reverse Transcriptase -PCR), and plasmatic microRNAs (miRNAs) PD-L2; Granzyme B;
were evaluated by targeted sequencing. Expression of PD-L1 on tumour biopsies was performed by Interleukine-2; Interferon-g;
immunohistochemistry using E13LN. High sPD-L1 at 2 months and increase of sPD-L1 concentrations were microRNA; plasma;
associated with poor response and absence of clinical benefit (nivolumab treatment less than 6 months). biomarker
The variation of sPD-L1 concentrations were confirmed by RNA quantification. sPD-L1 concentrations were
not correlated with PD-L1 expression on corresponding tumour samples. Low sGranB at nivolumab
initiation was also associated with poor response. High sPD-L1 and low sGranB were associated with poor
progression-free survival (PFS) and overall survival (OS). Low sPD-L2, low sIl-2 and high sIFN-g were
associated with grade 3–4 toxicities. Finally, miRNA screening showed that patients with clinical benefit
(n D 9) had down-expression of miRNA-320b and -375 compared to patients with early progression at
2 months (n D 9). In conclusion, our results highlight the interest of circulating biomarkers in patients
treated with nivolumab.

Introduction
currently used for second-line treatment in ALK- and EGFR-
Lung cancer is the leading cause of cancer related death world- wild type advanced NSCLC.4,5 There is however no biomarker
wide.1 Its prognosis, especially at the advanced stage, is poor predictive of nivolumab efficacy validated for clinical practice
with limited efficacy of cytotoxic chemotherapy (CT). Immune in this setting. In the two pivotal Checkmate studies,4,5 PD-L1
checkpoint inhibitors (ICIs), humanised monoclonal antibodies immunohistochemistry (IHC) was performed on the tumour
targeting programmed death 1 (PD-1) or programmed death- specimens obtained at the time of diagnosis. This analysis
ligand 1 (PD-L1), have recently been developed. PD-L1 and found that patients with non-squamous histology and high
programmed death-ligand (PD-L2) are membranous proteins PD-L1 expression determined by IHC had longer progression-
expressed by malignant cells that interact with PD-1 expressed free survival (PFS) and overall survival (OS) than those with
by T-cells. When PD-L1/PD-L2 and PD-1 bind, the T-cells’ low PD-L1 expression.5 These findings were not replicated in
cytotoxic anti-tumour activity is down-regulated. By blocking the squamous histology setting.4 Patients were not stratified at
the interaction between PD-L1 and PD-1, ICIs restore cytotoxic randomisation according to PD-L1 IHC, which was performed
immune response. The Type 1 T helper (Th1)-related cytotoxic retrospectively, rendering the interpretation of these results dif-
lymphocyte activation is mainly mediated by two effectors, ficult. In addition, IHC was performed on the diagnostic
interleukine-2 (IL-2) and interferon-gamma (IFN-g).2,3 After tumour specimens whereas PD-L1 expression varies with time
recognition of tumour antigens, activated CD8C lymphocytes and especially after CT.6,7 Moreover, PD-L1 IHC interpretation
secrete perforins and granzymes (mainly Granzyme B) that can be difficult, as there is a spatial heterogeneity of PD-L1
induce tumour cell death.2 expression, within one same tumour region or between two dif-
ICIs have shown their efficacy in advanced non-small cell ferent tumour regions (primary and metastatic).8-14 There is
lung cancer (NSCLC). Nivolumab, an anti-PD-1 antibody, is therefore an unmet need for developing new biomarkers to

CONTACT Dr. Etienne Giroux Leprieur Etienne.giroux-leprieur@aphp.fr Department of Respiratory Diseases and Thoracic Oncology, APHP – Ambroise Pare
Hospital, 9 avenue Charles de Gaulle, 92100 Boulogne-Billancourt, France.
Supplemental data for this article can be accessed at: https://doi.org/10.1080/2162402X.2018.1452581.
© 2018 Taylor & Francis Group, LLC
e1452581-2 A. COSTANTINI ET AL.

predict nivolumab efficacy. Tumour mutation load or IFN-g Nivolumab was given as second-line treatment in 67% of cases
signature have recently been evaluated,15,16 but their use in clin- (further line in 33%), and patients had good PS (0 – 1) at nivo-
ical practice is still challenging. lumab initiation in 58% of cases.
Compared with tumour specimens, plasma has the advan- Objective Response Rate (ORR) with nivolumab in the
tage of being easily accessible, allowing sequential analysis dur- global population was 40% (n D 17), and 35% (n D 15) had
ing follow-up. Plasmatic biomarkers also have the advantage of clinical benefit under nivolumab (as defined as still receiving
reflecting different tumour clones present throughout the body. nivolumab at 6 months). With a median follow-up of
Circulating tumour DNA (ctDNA) has shown to be a good pre- 16.3 months (IQR 11.7 – 21.1), 24 patients (56%) were deceased
dictive marker of ICIs efficacy.17-19 However, no study has been at the time of cut-off due to tumour progression, 11 patients
dedicated so far to other circulating biomarkers such as ICI- (26%) had controlled disease, 5 (12%) were still receiving fur-
related proteins or circulating microRNA (miRNA). The pres- ther treatment after nivolumab progression and 3 (6%) were
ence of soluble PD-L1 (sPD-L1) has already been established in lost during follow-up. The median nivolumab progression-free
patients with NSCLC with a prognostic impact of sPD-L1 con- survival (PFS) was 3.0 months (IQR 1.6 – 10.1) and median
centrations.20-22 Its prognostic and predictive impact with ICIs overall survival (OS) was 6.2 months (IQR 2.2 – NR).
is however still unknown. At initial diagnosis, nivolumab initiation and first tumour
In this study, we propose to evaluate new plasmatic bio- evaluation (2 months of treatment), median sPD-L1 concentra-
markers as putative predictive biomarkers associated with nivo- tions were 39.81 pg/ml (IQR 29.75 – 59.21), 49.86 pg/ml (IQR
lumab efficacy and toxicity in advanced NSCLC: sPD-L1, sPD- 36.11 – 65.91) and 51.57 pg/ml (IQR 31.91 – 72.06), respectively;
L2, sGranzyme B (sGran B), sIl, sIFN-g , and circulating median sPD-L2 concentrations were 16390.00 pg/ml (IQR
miRNA. 11185.00 – 22335.50), 18250.00 pg/ml (13963.00 – 21750.00) and
17567.00 pg/ml (14384.50 – 22166.50), respectively; median
sGranB concentrations were 14.06 pg/ml (IQR 9.84 – 21.13),
Results 13.24 pg/ml (IQR 8.53 – 18.95) and 17.24 pg/ml (IQR 8.63 –
25.79), respectively; median sIL-2 concentrations were 188.00 pg/
Characteristics of the 43 patients are presented in Table 1.
ml (IQR 113.00 – 227.50), 162.00 pg/ml (IQR 96.75 – 242.00)
Patients were mostly male (67%), current or former smokers
and 152.00 pg/ml (IQR 91.50 – 176.50), respectively ; median
(88%) and with adenocarcinoma histology (65%). They had
sIFN-g concentrations were 0.13 pg/ml (IQR -0.03 – 0.29),
advanced stage disease at diagnosis (93%) and mainly without
0.05 pg/ml (IQR -0.03 – 0.20) and 0.11 pg/ml (IQR -0.06 – 0.27),
EGFR-, KRAS- mutations or ALK -rearrangement (63%).
respectively.

Table 1. Patients’ characteristics.

Patient charcateristics Patients (n D 43) ORR, clinical benefit and survival
Sex (%)
Male 29 (67%) sPD-L1
Female 14 (33%) At initial diagnosis and at nivolumab initiation, there was no
Smoking history statistical difference in sPD-L1 concentrations in patients who
Current 14 (33%)
Former 24 (55%) were responders compared to non-responders to nivolumab
Non-smoker 5 (12%) (Supplementary Fig. 1A and 1B), and in patients presenting
Histology with clinical benefit compared to patients who did not present
Adenocarcinoma 28 (65%)
Squamous-cell carcinoma 9 (21%) with clinical benefit (Supplementary Fig. 1C and 1D). However,
Sarcomatoid carcinoma 1 (2%) at first tumour evaluation under nivolumab, sPD-L1 concentra-
Large cell carcinoma 4 (9%) tions were significantly higher in non-responders with a
Clear cell carcinoma 1 (2%)
Stage at diagnosis median value of 67.64 pg/ml (IQR 46.36 – 75.14) compared to
I-II 3 (7%) 32.94 pg/ml (IQR 24.89 – 58.91) in responders (p D 0.031)
III-IV 40 (93%) (Fig. 1A). In the same way, median sPD-L1 concentrations
EGFR mutation 1 (2%)
KRAS mutation 15 (35%) were significantly higher in patients without clinical benefit
ALK rearrangement 0 (0%) with a median value of 67.64 pg/ml (IQR 42.74 – 75.45) com-
Other* 4 (10%) pared to 34.14 pg/ml (IQR 24.67 – 56.48) in patients with clini-
Wild type 23 (53%)
Number of lines before nivolumab initiation cal benefit (p D 0.024) (Fig. 1B). Moreover, in case of increase
1 29 (67%) of sPD-L1 concentrations between the initiation of nivolumab
2 10 (23%) and first tumour evaluation (n D 12), ORR was 17% (n D 2)
3 2 (5%)
4 2 (5%) versus 68% (n D 13) in case of decrease or stability of sPD-L1
Median age at nivolumab initiation 68 (IQR 62 – 71.5) concentrations (n D 19) (p D 0.005). The clinical benefit rate
PS at nivolumab initiation was 10% (n D 1) in case of increase of sPD-L1 concentrations,
0–1 25 (58%)
2 17 (40%) versus 47% (n D 8) in case of decrease or stability of sPD-L1
3 1 (2%) concentrations (n D 17) (p D 0.049).
Using ROC curves, we found a sPD-L1 cut-off concentration
EGFR: Epidermal Growth Factor Receptor, ALK : Anaplastic Lymphoma Kinase, PS :
Performance Status. of 33.97 pg/ml associated with a sensitivity of 94%, a specificity

Other mutations: HER 2 (n D 1; 2%), BRAF (n D 3; 7%). of 56%, a positive predictive value (PPV) of 70% and a negative
 ONCOIMMUNOLOGY e1452581-3

Figure 1. sPD-L1 and tumour response, clinical benefit, progression-free survival (PFS) and overall survival (OS). A: sPD-L1 concentrations at first tumour evaluation in
patients with tumour response and patients without tumour response. B: sPD-L1 concentrations at first tumour evaluation in patients with clinical benefit and patients
without clinical benefit. C and D: PFS (C) and OS (D) according to sPD-L1 concentration at first tumour evaluation. E and F: PFS (E) and OS (F) according to sPD-L1 variation
between nivolumab initiation and first tumour evaluation. P-values were calculated by Mann-Whitney test (A and B) or log-rank test (C-F).

predictive value (NPV) of 90% to predict response at first increase of PD-L1 expression (n D 3; 2 patients without evalu-
tumor evaluation. Using this cut-off, we determined that able RT-PCR results). In the same way, patients with sPD-L1
patients with low sPD-L1 (n D 10) had an ORR of 90% whilst decrease in ELISA had a significant decrease of PD-L1 expres-
patients who had high sPD-L1 concentrations (n D 23) had an sion (n D 2; 3 patients without evaluable RT-PCR results)
ORR of 30% (p D 0.002). In the same way, we found a cut-off (Table 2).
concentration of 36.36 pg/ml associated with a sensitivity of High sPDL-L1 concentrations at the first tumour evaluation
84%, a specificity of 57%, a PPV of 73% and an NPV of 73% to and increase of sPD-L1 concentrations were associated with
predict clinical benefit. Using this cut-off, we determined that worse PFS and OS (Fig. 1 C–F). Patients with low sPD-L1 con-
patients with low sPD-L1 concentrations (n D 11) had clinical centration at first tumour evaluation had a median PFS of
benefit in 73% of cases, whereas patients who had high sPD-L1 11.8 months (IQR 6.5 – NR) and median OS NR (IQR 13.6 –
concentrations (n D 22) had clinical benefit in 38% of cases NR), versus median PFS of 2.2 months (IQR 1.6 – 9.1) and
(p D 0.013). median OS of 6.2 months (IQR 2.4 -NR) for patients with high
We confirmed the variation of sPD-L1 by performing RT- sPD-L1 concentrations at first tumour evaluation (p D 0.041
PCR analysis on circulating RNA. Selecting the patients with for PFS comparison and p D 0.087 for OS comparison)
the most important increase (n D 5; ranging from C78% to (Fig. 1C and 1D). Patients presenting with an increase of sPD-
C101%) and decrease (n D 5; ranging from -39% to -69%) of L1 concentrations had a median PFS of 1.8 months (IQR 0.9 –
sPD-L1 concentrations evaluated by ELISA, we showed that 3.0), versus 6.5 months (IQR 3.0 – NR) in patients presenting
patients with sPDL-L1 increase in ELISA had a significant with a decrease or a stability of sPD-L1 concentrations (p D
e1452581-4 A. COSTANTINI ET AL.

Table 2. Relative PD-L1 gene expression in plasma (evaluated by RT-PCR) in Table 4. Multivariate analysis (Cox model) on overall survival.
patients with the largest increase or decrease of sPD-L1 concentrations (evaluated
by ELISA) between nivolumab initiation and first tumour evaluation. Variable Hazard ratio (HR) IC 95% p-value

patients relative change of sPD- relative PD-L1 gene PS > 1 36.49 4.56–291.76 0.001
L1 concentration (ELISA) expression in plasma sPD-L1 at first tumor evaluation 0.98 0.95–1.01 0.129
(RT-PCR) (continuous variable)
sGranzyme B at nivolumab 1.00 0.94–1.07 0.966
patients with the patient 143% NA initiation (continuous variable)
largest sPD-L1 #1 Increase of sPD-L1 4.15 0.64–27.03 0.136
increase concentrations
patient 101% 5.60 (§ 2.78) Increase sGranzyme B 1.50 0.36–6.37 0.579
#2 concentrations
patient 93% 3.14 (§ 0.12) IHC PD-L1 positive 0.43 0.10–1.90 0.268
#3
patient 88% NA
#4
patient 78% 2.05 (§ 0.65) was observed between IHC positivity and sPD-L1 expression in
#5 IHC at the time of diagnosis: median sPD-L1 was 30.86 pg/ml
patients with the patient -59% NA
largest sPD-L1 #1
(IQR 22.13 – 48.23) in patients with positive IHC and 36.68 pg/
decrease ml (IQR 26.10 – 55.34) in patients with negative IHC (p D 0.604).
patient -51% 2.19E-17 (§ 6.95E-19) There was no correlation between sPD-L1 concentrations at
#2
patient -45% 1.90E-19 (§ 6.58E-21)
diagnosis and level of expression of PD-L1 in IHC according to
#3 different cut-offs (Supplementary Fig. 3).
patient -41% NA
#4
patient -11% NA sGranzyme B
#5
There was no statistical difference in sGran B concentrations
Relative change of sPD-L1 concentration (ELISA) is between nivolumab initiation measured at initial diagnosis and at first tumor evaluation in
and first tumour evaluation. Relative PD-L1 gene expression in plasma (RT-PCR) patients who were responders compared to non-responders
at first tumour evaluation is evaluated according to gene expression at nivolu-
mab initiation. NA: non amplifiable RNA. (Supplementary Fig. 4A and 4B) and in patients presenting with
clinical benefit compared to patients who did not present with
clinical benefit (Supplementary Fig. 4C and 4D). However, at
0.008). (Fig. 1E). Median OS was 5.4 months (IQR 1.1 – NR) nivolumab initiation, responders had significantly higher median
in patients with increasing sPD-L1 concentrations, versus NR sGranzyme B concentrations than non-responders: 18.44 pg/ml
(6.0 – NR) in patients with stable or decreasing sPD-L1 concen- (IQR 11.71 – 33.92) versus 11.88 pg/ml (IQR 7.94 – 15.41) (p D
trations (p D 0.028). (Fig. 1F). Multivariate analysis on PFS 0.039) (Fig. 2A). In the same way, median sGran B concentra-
(Table 3) confirmed the independent role of the increase of tions tended to be higher in patients with clinical benefit with a
sPD-L1, with a hazard ration (HR) at 4.85 (IC95% 1.02-NR; p median value of 67.64 pg/ml (IQR 42.74 – 75.45) compared to
D 0.048). The only significant factor for OS in multivariate 34.14 pg/ml (IQR 24.67 – 56.48) in patients without clinical ben-
analysis was PS (PS more than 1 associated with a HR D 36.49; efit (p D 0.116) (Fig. 2B). There was also a trend towards a differ-
IC95% 4.56-291.76; p D 0.001) (Table 4). ential evolution of sGran B concentrations according to tumour
Finally, we analysed the outcome in patients with stable dis- response and clinical benefit: ORR was 29% (n D 4) in case of
ease as best tumour response according to iRECIST (n D 5). increase of sGran B concentrations (n D 14), versus 63% (n D
Interestingly, the two patients who had decreasing sPD-L1 levels 10) in case of decrease or stability of sGran B concentrations (n
had longer PFS (5.51 and 21.08 months) than patients who had D 16) (p D 0.063). In case of increase of sGran B concentrations
increasing sPD-L1 levels (3.74, 2.33 and 2.07 months). Two (n D 14), the clinical benefit rate was 21% (n D 3), versus 56%
examples of stable patients with differential outcome according (n D 9) in case of decrease or a stability of sGran B concentra-
to sPD-L1 measurements are shown in Supplementary Fig. 2. tions (n D 16) (p D 0.052).
Of the 43 patients included in the study, PD-L1 IHC was per- Using ROC curves, we found a cut-off concentration of
formed in 34 cases (79%) on diagnostic samples. Twenty-four 18.24 pg/ml associated with a sensitivity of 53%, a specificity
patients (71%) had a positive PD-L1 expression. No association of 87%, a PPV of 73% and an NPV of 74% to predict tumour
response at nivolumab initiation. Using this cut-off, we
Table 3. Multivariate analysis (Cox Model) on progression-free survival. determined that patients with high sGran B concentrations
Variable Hazard ratio (HR) IC 95% p-value (n D 11) had an ORR of 73%, whereas patients who had low
sGran B concentrations (n D 27) had an ORR of 26% (p D
PS > 1 4.85 1.28–18.35 0.020 0.007).
sPD-L1 at first tumor evaluation 0.99 0.97–1.01 0.467
(continuous variable) Low sGran B concentrations at nivolumab initiation and
sGranzyme B at nivolumab 0.99 0.94–1.04 0.628 increase of sGran B concentrations were also associated with
initiation (continuous variable) worse PFS and OS (Fig. 2 C-F). Patients with high sGran B con-
Increase of sPD-L1 4.85 1.02-NR 0.048
concentrations centrations at nivolumab initiation had a median PFS of
Increase sGranzyme B 1.82 0.61–5.43 0.284 8.8 months (IQR 3.7 -NR) and median OS NR (IQR 9.0 – NR),
concentrations versus median PFS of 1.8 months (IQR 1.1 -3.7) and median OS
IHC PD-L1 positive 0.79 0.26–2.42 0.681
of 4.5 months (IQR 1.8 – NR) for patients with low sGran B
 ONCOIMMUNOLOGY e1452581-5

Figure 2. sGranzyme B (sGran B) and tumour response, clinical benefit, progression-free survival (PFS) and overall survival (OS). A: sGran B concentrations at nivolumab
initiation in patients with tumour response and patients without tumour response. B: sGran B concentrations at nivolumab initiation in patients with clinical benefit and
patients without clinical benefit. C and D: PFS (C) and OS (D) according to sGran B concentrations at nivolumab initiation. E and F: PFS (E) and OS (F) according to sGran B
variation between nivolumab initiation and first tumour evaluation. P-values were calculated by Mann-Whitney test (A and B) or log-rank test (C-F).

concentrations at nivolumab initiation (p D 0.018 for PFS com- Immune related adverse events (irAEs)
parison, and p D 0.096 for OS comparison) (Fig. 3C and 3D).
Eight patients (19%) experienced grade 3 – 4 irAEs whilst receiv-
Patients presenting with an increase in sGran B concentrations
ing nivolumab: hypophysitis (n D 1), arthro-myalgia (n D 1),
had a median PFS of 2.0 months (IQR 1.6 – 4.1), versus
auto-immune cholangitis (n D 1), auto-immune hepatitis (n D
8.8 months (IQR 1.8 -NR) in patients presenting with a decrease
1), auto-immune kidney failure (n D 1), interstitial pneumonia
or a stability of sGran B concentrations (p D 0.019) (Fig. 2E).
(n D 1), skin toxicity (n D 1), auto-immune colitis (n D 1).
Median OS was 4.5 months (IQR 2.2 – NR) in patients with
Of all the tested biomarkers, sPD-L2 at initial diagnosis and at
increasing sGran B concentrations, versus NR (IQR 6.2 – NR) in
nivolumab initiation, sIL-2 at nivolumab initiation and sIFN-g at
patients with stable or decreasing sGranzyme B concentrations
first tumor evaluation were significantly associated with grade 3–
(p D 0.043) (Fig. 2F).
4 irAEs with nivolumab (Fig. 3 and Supplementary Fig. 7).

sPD-L2, sIL-2 and sIFN-g

miRNA
There was no impact of sPD-L2, sIL-2 and sIFN-g concentra-
tions and variations on ORR, clinical benefit, PFS or OS, at any We performed miRNA screening on plasma at nivolumab initi-
time of analysis (Supplementary Figs. 5 and 6). ation, and compared the miRNA profile between patients with
e1452581-6 A. COSTANTINI ET AL.

Our results suggest that sPD-L1 may play an important part

in evaluating response, survival and clinical benefit under nivo-
lumab. sPD-L1 concentrations at first tumour evaluation and
the variation of sPD-L1 concentrations at the beginning of
treatment were associated with ORR and clinical benefit at six
months. sPD-L1 also strongly impacted PFS and OS, with lon-
ger median PFS and OS in case of low sPD-L1 concentrations
at first tumour evaluation or in case of decrease or stability of
sPD-L1 concentrations between initiation of nivolumab and
first tumour evaluation. sPD-L1 has already been studied in
NSCLC, with results showing a higher sPD-L1 expression in
patients with NSCLC than in healthy controls, and that patients
had a better prognosis in case of higher sPD-L1 concentra-
tions.20,21 However, patients did not receive ICIs treatment in
these studies. Furthermore, the patients did not have multiple
sPD-L1 measurements and no dynamic analysis was per-
formed. Finally, the moment of sampling was not specified
beforehand with samples taken at different times (diagnosis,
during treatment) rendering comparisons difficult. Another
report studied the variation of sPD-L1 in patients with locally
advanced or inoperable NSCLC undergoing radiotherapy.22
This study showed that sPD-L1 levels measured after 2 and
4 weeks of radiotherapy significantly decreased compared to
pre-radiotherapy levels and that patients with lower baseline
sPD-L1 levels had longer OS than those with higher sPD-L1
Figure 3. sPD-L2, sIL-2, sIFN-g and grade 3–4 toxicity with nivolumab. A and B:
levels.
sPD-L2 concentrations at initial diagnosis (A) and at nivolumab initiation (B) in We did not show a correlation between the sPD-L1 concen-
patients with grade 3–4 toxicity and patients without grade 3–4 toxicity. C: sIL-2 trations and the expression of PD-L1 in IHC performed on the
concentrations at nivolumab initiation in patients with grade 3–4 toxicity and
patients without grade 3–4 toxicity. D: sIFN-g concentrations at first tumour evalu-
initial biopsy. This can in part be explained by the use of two
ation in patients with grade 3–4 toxicity and patients without grade 3–4 toxicity. different antibodies for ELISA (28-8 clone) and for IHC
P-values were calculated by Mann-Whitney test. (E13LN clone). However, recent publications have shown a
good concordance between these two antibodies for PD-L1
expression in tumour cells.23 The difference between ELISA
clinical benefit (n D 9) and patients with early progression (at and IHC could also be explained by the fact that plasma reflects
first tumour evaluation) under nivolumab (n D 9). We showed the entire tumour heterogeneity, whereas PD-L1 expression
a differential expression of 2 miRNA between the 2 groups of can be heterogeneous within one tumour site, but also between
patients (Supplementary Fig. 8), corresponding to miRNA- different tumour sites.8-14 Finally, sPD-L1 could also be
320b and miRNA-375. Patients with clinical benefit had a secreted by cells other than tumour cells, such as immune cells.
down-expression of both miRNA-320b (fold change -3.1) and Interestingly, we confirmed the variation of sPD-L1 concentra-
-375 (fold change -3.2), compared to patients with early pro- tions by RT-PCR from circulating RNA in patients with the
gression (p<0.05 for both miRNA). When analyzing target largest variations of sPD-L1 concentrations. Circulating RNA
genes using miRecords, miRTarBase and TarBase databases, we is a mix of RNA from tumour cells and from other cells
found that miRNA-320b was notably associated with prolifera- (immune cells, stroma cells…), so our results cannot precisely
tion (MYC, TUBB1) genes, while miRNA-375 was notably determine the source of sPD-L1. Little data is currently avail-
associated with immune-related genes (JAK2, TGF-b2), Wnt able on the origin and function of sPD-L1. The first hypothesis
pathway (FZD4, FZD8), proliferation (MYC), Hippo pathway is that sPD-L1 concentrations reflect the tumour mass and that
(YAP1) and tumor migration (NCAM1, CDH2) genes. its presence in peripheral blood is due to the lysis of tumour
cells. A high level of sPD-L1 would be the reflection of high
tumour mass and the increase of sPD-L1 concentrations under
Discussion
nivolumab associated with poorer response could simply reflect
This study reports for the first time the predictive role of plas- the increase of the tumour volume. The second hypothesis is
matic immune-related biomarkers and circulating miRNA in that sPD-L1 has its own biological effect. It could interfere with
NSCLC patients treated with nivolumab. We showed that sPD- nivolumab by binding with PD-1 expressed by T lymphocytes
L1 and sGran B were associated with outcome, whereas sPD- and induce a competing effect with nivolumab. The low sPD-
L2, sIL-2 and sIFN-g were associated with nivolumab-related L1 concentrations at the beginning of treatment associated
grade 3 – 4 toxicity. Finally, we observed that the plasmatic with better efficacy of nivolumab is compatible with such a
miRNA profile at the beginning of treatment was different hypothesis. A study is currently underway to try to determine
between patients with objective response and patients with which cells produce sPD-L1 in various malignancies
early progression under nivolumab. (NCT01660776).
 ONCOIMMUNOLOGY e1452581-7

Granzyme B is a serine protease that is a mediator of target- tumour immune response with ICIs, and the potential utility of
cell apoptosis by cells such as NK cells and cytotoxic CD8C T screening for these miRNA in plasma before beginning nivolu-
cells. Granzymes are delivered to the target cells via cytotoxic mab to select patients who will have response and clinical bene-
granules and are responsible for caspase-dependant fit with this treatment.
apoptosis.24,25 Soluble Granzyme B has already been explored This study has several limitations. It is a monocentric
in the context of auto-immune diseases, showing that high lev- study of small size, with a possibility of lack of power for
els of sGran B were associated with rheumatoid arthritis, myo- some statistical analyses, especially for multivariate analyses.
cardial infarction and lipid-rich carotid plaques.26-28 To our Nevertheless, despite this small population, we observed sig-
knowledge, this is the first time that sGran B concentrations nificant results on ORR, clinical benefit, survival and toxic-
have been evaluated in the plasma of NSCLC patients treated ity. Finally, to confirm the predictive value of sPD-L1 and
with nivolumab. We found that patients who presented with an sGran B it would be necessary to use a validation cohort
objective response to nivolumab had significantly higher sGran with a control group. This study also has several strengths.
B concentrations at nivolumab initiation than patients who We used a prospective cohort of patients with plasma sam-
were non-responders. This could reflect the activation of the ples. These samples were well characterised beforehand and
CD8C cytotoxic immune response, known to be associated underwent rigorous pre-analytical conditioning. Further-
with better response with ICIs. When analysing the variation of more, our study presents multiple samples for each patient
sGran B concentrations between nivolumab initiation and first allowing dynamic analyses.
tumour evaluation we found a tendency favouring patients In conclusion, sPD-L1 and sGran B seem to be promis-
with stable or decreasing concentrations for ORR, clinical bene- ing biomarkers associated with tumour response and clini-
fit, OS and PFS. As for sPD-L1, an increase of sGran B at the cal benefit with nivolumab, while sPD-L2, sIL-2 and sIFN-
beginning of the treatment could be the reflection of the persis- g were effective in predicting immune-related toxicities.
tence of a high tumour volume, with persistent CD8 lympho- Furthermore, the screening of circulating miRNA also
cytes activation and Granzyme B secretion. However, these seems to be an exciting research possibility. If confirmed,
hypotheses need to be validated in further studies. our results suggest that an approach integrating plasmatic
The prediction of severe adverse events with ICIs remains a immune-related biomarkers such as sPD-L1, sGran B or cir-
daily challenge. Even if grade 3 – 4 adverse events are rare with culating miRNA could help us to rapidly detect patients
nivolumab,4,5 they can be severe, impacting quality of life, who truly benefit from immunotherapy and, on the con-
sometimes life-threatening, and can lead to treatment interrup- trary, consider a rapid change of strategy in case of no
tion. We found that several immune-related plasmatic bio- anticipated clinical benefit. Further studies are needed, for
markers were associated with grade 3 – 4 toxicity. Patients validation of these biomarkers in a different cohort of
presenting grade 3 – 4 toxicity under nivolumab had lower patients and evaluation in other ICIs treatment settings
sPD-L2 concentrations at initial diagnosis and at nivolumab (first-line treatment, combination of ICIs, ICIs and chemo-
initiation, lower sIL-2 concentrations at nivolumab initiation, therapy combined treatment).
and higher sIFN-g concentrations at first tumour evaluation. If
confirmed, these results could help predict which patients are
at risk for high-grade toxicity as early as the beginning of treat- Materials and methods
ment, leading to close follow-up of such patients. Experimental design
miRNA are non-coding RNA fragments with biological
activity. Their role in solid cancers has been widely studied, This study was an exploratory study, based on the analysis
and they have been shown to be involved in chemo-resis- of consecutive patients prospectively included in the
tance.29 Some publications have suggested an implication of Department of Respiratory Medicine and Thoracic Oncol-
miRNA in immune response regulation as well as PD-L1 and ogy (APHP – Ambroise Pare Hospital) between July 2015
PD-1 expression regulation.30-41 We report for the first time and September 2017. Exploratory endpoints were ORR,
the association of down-regulation of circulating miRNA-320b clinical benefit (i.e. complete response, partial response or
and -375 expression and response to ICIs. miRNA-320b is stability, according to iRECIST, lasting 6 months or more
down-regulated in various cancers, and it is associated with after initiation of nivolumab treatment), PFS, OS, grade 3 –
tumorigenesis and poor prognosis in glioma.42 miRNA-375 4 toxicity (according to CTCAE v4.0), according to plas-
was also shown to be often down-regulated in NSCLC,43,44 matic concentrations of various circulating biomarkers. Dif-
breast cancer,45 colorectal cancer,46-49 oesophageal cancer,50,51 ferential analysis of plasmatic miRNA profiles between
gastric cancer,52,53 and pancreatic cancer.54,55 It is associated responders and patients with early progression with nivolu-
with tumour proliferation and metastases.44,48,51,54,56-59 In mab was also planned.
NSCLC, miRNA-375 is associated with poor prognosis.43,44,60
Interestingly, miRNA-375 is strongly associated with the Wnt/
Patients and plasma
b-catenine pathway49 and the Hippo pathway,59,61-63 known to
be involved in ICIs resistance.64-70 Moreover, miRNA-375 tar- Tumour response was evaluated every two months using iRE-
gets the JAK2 gene,62,71 involved in the IFN-g and TNF-a path- CIST criteria. Medical records were reviewed, and data retro-
ways, key-regulators of cytotoxic CD8C immune response.72 spectively extracted on clinical and pathological features as well
Taken together, these data, highlighted by our results, suggest as treatment history. Plasma samples were taken at diagnosis,
an implication of both miRNA-320b and miRNA-375 in anti- just before the first injection of nivolumab (C1), and at the first
e1452581-8 A. COSTANTINI ET AL.

tumour evaluation (at 2 months, M2) (Supplementary Fig. 9). miRNA screening
Two 10ml-EDTA tubes of peripheral blood were taken, and
Plasmatic miRNA were extracted using miRNeasy Serum/
plasma was isolated within one hour after and immediately
Plasma kit (Qiagen), according to manufacturer’s instructions.
conserved at ¡80 C.
miRNA concentrations were evaluated by BioAnalyzer. Screen-
ing of plasmatic miRNA was performed by targeted sequencing
using TruSeq Small RNA kit (Illumina). Briefly, after a ligation
Ethical considerations step of miRNA with specific Illumina adapters, a RT-PCR was
All patients signed an informed consent allowing blood to be run. Banks of sequences were then analysed on HiSeq2500 (sin-
drawn and stored within the Centre de Ressources Biologiques gle read mode), with reading of 50 nucleotides (enough to cover
(CRB) of the Ambroise Par
e University Hospital during their the 19 to 22 bases of miRNA). After normalization and a
follow-up and treatment. The protocol was approved by the trimmed mean calculation step,73 a differential analysis of
Institutional Review Board CPP IDF n 8 (ID CRB 2014- expressed miRNA between patients with clinical benefit and
A00187-40). patients with early progression with nivolumab was per-
formed.74,75 After identification of miRNA differentially
expressed, corresponding target genes were identified using
miRecords, miRTarBase and TarBase databases.76
ELISA technique
sPD-L1, sPD-L2, sGran B, sIL-2, sIFN-g concentrations were
calculated by ELISA. ELISA tests were performed using com- Statistical analysis
mercial kits (ab214565 Human PD-L1 [28-8] ELISA Kit, Median soluble concentrations of all the tested biomarkers were
Abcam; BMS 2215 Human PD-L2 Platinum ELISA, Thermo analysed according to ORR, clinical benefit, grade 3–4 toxic-
Fisher Scientific; BMS 2027 Human Granzyme b Coated ELISA ities. The comparison of median biomarker levels between
Kit, Thermo Fisher Scientific; ab174443 Human IFN gamma groups was performed using Mann-Whitney test and inter-
ELISA Kit, Abcam; ab174444 Human Il-2 ELISA Kit, Abcam) quartile range (IQR) is given for each value. Receiving Operat-
according to manufacturer’s instructions. Corresponding ing Curve (ROC) method was used to determine a cut-off level
recombinant proteins were used for each test at pre-specified for each biomarker with a significant difference for endpoints
concentrations to build standard curves. The results were with the Mann-Whitney test. Kaplan-Meier method was used
obtained using a spectrophotometer (reading at 450nm), and to determine OS and PFS. Comparison between survival curves
concentrations were calculated according to the standard was performed using log-rank method. Multivariate analysis
curves. All samples, standards and negative controls were tested was performed using Cox proportional hazards model (PFS,
in duplicate. OS), integrating experimental variables with significant result
in univariate abalyses, and known prognostic and predictive
associated with ICIs (PS, PD-L1 IHC). Data analysis was com-
IHC technique puted using XLStat v 19.4 (Addinsoft). P-values were consid-
ered as significant if <0.05.
IHC was performed using an automated method (Leica) and
the E13LN anti-PD-L1 antibody (Cell signalling Technology)
diluted to the 1/80th on 4mm-slides from the treatment-na€ıve Disclosure statement
diagnostic samples. The assay was performed using human
EGL received research funding from Bristol-Myers Squibb. Other authors
amygdala as positive control, and IgG as isotype negative con- did not have conflict of interest related to this project.
trol. The IHC was considered as being positive if at least one
tumour cell out of 100 analysed tumour cells was positively
stained. Funding
This work was supported by Bristol-Myers-Squibb.

Quantitative reverse-transcriptase polymerase chain

reaction (RT-PCR) References
Plasmatic RNAs were extracted using miRNeasy Serum/Plasma 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J
kit (Qiagen), according to manufacturer’s instructions. RNA Clin. 2016;66:7–30. doi:10.3322/caac.21332.
concentrations were evaluated by Nanodrop. cDNA was syn- 2. Martinez-Lostao L, Anel A, Pardo J. How Do Cytotoxic Lymphocytes
Kill Cancer Cells? Clin Cancer Res. 2015;21:5047–56. doi:10.1158/
thesized using iScript cDNA Synthesis Kit (Bio-Rad) according 1078-0432.CCR-15-0685. PMID:26567364.
to the manufacturer’s instructions. RT-PCR for PD-L1 gene 3. Balkhi MY, Ma Q, Ahmad S, Junghans RP. T cell exhaustion and
expression was performed using specific Taqman primers and Interleukin 2 downregulation. Cytokine. 2015;71:339–47. doi:10.1016/
probes (Hs00204257_m1, ThermoFisher) on 7900HT Fast j.cyto.2014.11.024. PMID:25516298.
Real-Time PCR System (Applied Bisosystems). Gene expres- 4. Brahmer J, Reckamp KL, Baas P, Crin o L, Eberhardt WEE, Poddub-
skaya E, Antonia S, Pluzanski A, Vokes EE, Holgado E, et al. Nivolu-
sion analysis was calculated with the delta-delta CT method mab versus Docetaxel in Advanced Squamous-Cell Non-Small-Cell
normalized to an endogenous control (RPLP0). All samples Lung Cancer. N Engl J Med. 2015;373:123–35. doi:10.1056/
were tested in triplicate. NEJMoa1504627. PMID:26028407.
 ONCOIMMUNOLOGY e1452581-9

5. Borghaei H, Paz-Ares L, Horn L, Spigel DR, Steins M, Ready NE, efficacy of nivolumab treatment in patients with non-small cell lung
Chow LQ, Vokes EE, Felip E, Holgado E, et al. Nivolumab versus cancer. Eur J Cancer Oxf Engl 1990. 2017;86:349–57.
Docetaxel in Advanced Nonsquamous Non-Small-Cell Lung Cancer. 19. Giroux Leprieur E, Herbretau G, Dumenil C, Julie C, Giraud V, Lab-
N Engl J Med. 2015;373:1627–39. doi:10.1056/NEJMoa1507643. rune S, Dumoulin J, Tisserand J, Emile J-F, Blons H, et al. Circulating
PMID:26412456. tumor DNA evaluated by Next-Generation Sequencing is predictive
6. Sheng J, Fang W, Yu J, Chen N, Zhan J, Ma Y, Yang Y, Huang Y, Yan- of tumor response and prolonged clinical benefit with nivolumab in
huang null, Zhao H, et al. Expression of programmed death ligand-1 advanced non-small cell lung cancer. OncoImmunology. 2018.
on tumor cells varies pre and post chemotherapy in non-small cell doi:10.1080/2162402X.2018.1424675.
lung cancer. Sci Rep. 2016;6:20090. doi:10.1038/srep20090. 20. Zhang J, Gao J, Li Y, Nie J, Dai L, Hu W, Chen X, Han J, Ma X, Tian
PMID:26822379. G, et al. Circulating PD-L1 in NSCLC patients and the correlation
7. Lim SH, Hong M, Ahn S, Choi Y-L, Kim K-M, Oh D, Ahn YC, Jung S- between the level of PD-L1 expression and the clinical characteristics.
H, Ahn M-J, Park K, et al. Changes in tumour expression of pro- Thorac Cancer. 2015;6:534–8. doi:10.1111/1759-7714.12247. PMID:
grammed death-ligand 1 after neoadjuvant concurrent chemoradio- 26273411.
therapy in patients with squamous oesophageal cancer. Eur J Cancer 21. Okuma Y, Hosomi Y, Nakahara Y, Watanabe K, Sagawa Y, Homma S.
Oxf Engl 1990. 2016;52:1–9. High plasma levels of soluble programmed cell death ligand 1 are
8. Kim S, Kim M-Y, Koh J, Go H, Lee DS, Jeon YK, Chung DH. Pro- prognostic for reduced survival in advanced lung cancer. Lung Cancer
grammed death-1 ligand 1 and 2 are highly expressed in pleomorphic Amst Neth. 2017;104:1–6. doi:10.1016/j.lungcan.2016.11.023.
carcinomas of the lung: Comparison of sarcomatous and carcinoma- 22. Zhao J, Zhang P, Wang J, Xi Q, Zhao X, Ji M, Hu G. Plasma levels of
tous areas. Eur J Cancer Oxf Engl 1990. 2015;51:2698–707.. soluble programmed death ligand-1 may be associated with overall
9. Ilie M, Long-Mira E, Bence C, Butori C, Lassalle S, Bouhlel L, Fazzalari survival in nonsmall cell lung cancer patients receiving thoracic radio-
L, Zahaf K, Lalv
ee S, Washetine K, et al. Comparative study of the PD- therapy. Medicine (Baltimore). 2017;96:e6102. doi:10.1097/
L1 status between surgically resected specimens and matched biopsies MD.0000000000006102. PMID:28207525.
of NSCLC patients reveal major discordances: a potential issue for 23. Adam J, Rouquette I, Damotte D, Badoual C, Danel C, Damiola F,
anti-PD-L1 therapeutic strategies. Ann Oncol Off J Eur Soc Med Penault-Llorca F, Lantuejoul S. Multicentric French Harmonization
Oncol. 2016;27:147–53. doi:10.1093/annonc/mdv489. Study for PD-L1 IHC Testing in NSCLC. J Thorac Oncol Off Publ Int
10. Li C, Huang C, Mok TS, Zhuang W, Xu H, Miao Q, Fan X, Zhu W, Assoc Study Lung Cancer. 2017;12 (1s):PL04a.04.
Huang Y, Lin X, et al. Comparison of 22C3 PD-L1 Expression 24. Trapani JA, Sutton VR. Granzyme B: pro-apoptotic, antiviral and
between Surgically Resected Specimens and Paired Tissue Microarrays antitumor functions. Curr Opin Immunol. 2003;15:533–43.
in Non-Small Cell Lung Cancer. J Thorac Oncol Off Publ Int Assoc doi:10.1016/S0952-7915(03)00107-9. PMID:14499262.
Study Lung Cancer. 2017;12:1536–43. 25. Waterhouse NJ, Sutton VR, Sedelies KA, Ciccone A, Jenkins M,
11. Casadevall D, Clav
e S, Taus A,
Hardy-Werbin M, Rocha P, Lorenzo Turner SJ, Bird PI, Trapani JA. Cytotoxic T lymphocyte-induced kill-
M, Men
endez S, Salido M, Albanell J, Pijuan L, et al. Heterogeneity of ing in the absence of granzymes A and B is unique and distinct from
Tumor and Immune Cell PD-L1 Expression and Lymphocyte Counts both apoptosis and perforin-dependent lysis. J Cell Biol. 2006;
in Surgical NSCLC Samples. Clin Lung Cancer. 2017;18:682–91.e5. 173:133–44. doi:10.1083/jcb.200510072. PMID:16606695.
doi:10.1016/j.cllc.2017.04.014. PMID:28549836. 26. Tak PP, Spaeny-Dekking L, Kraan MC, Breedveld FC, Froelich CJ,
12. Uruga H, Bozkurtlar E, Huynh TG, Muzikansky A, Goto Y, Gomez- Hack CE. The levels of soluble granzyme A and B are elevated in
Caraballo M, Hata AN, Gainor JF, Mark EJ, Engelman JA, et al. Pro- plasma and synovial fluid of patients with rheumatoid arthritis (RA).
grammed Cell Death Ligand (PD-L1) Expression in Stage II and III Clin Exp Immunol 1999;116:366–70. doi:10.1046/j.1365-2249.1999.
Lung Adenocarcinomas and Nodal Metastases. J Thorac Oncol Off 00881.x. PMID:10337032.
Publ Int Assoc Study Lung Cancer. 2017;12:458–66. 27. Kondo H, Hojo Y, Tsuru R, Nishimura Y, Shimizu H, Takahashi N,
13. Pinato DJ, Shiner RJ, White SDT, Black JRM, Trivedi P, Stebbing J, Hirose M, Ikemoto T, Ohya K-I, Katsuki T, et al. Elevation of plasma
Sharma R, Mauri FA. Intra-tumoral heterogeneity in the expression of granzyme B levels after acute myocardial infarction. Circ J Off J Jpn
programmed-death (PD) ligands in isogeneic primary and metastatic Circ Soc. 2009;73:503–7.
lung cancer: Implications for immunotherapy. Oncoimmunology. 28. Skjelland M, Michelsen AE, Krohg-Sørensen K, Tennøe B, Dahl A,

2016;5:e1213934. doi:10.1080/2162402X.2016.1213934. PMID:2775 Bakke S, Brosstad F, Damas JK, Russell D, Halvorsen B, et al. Plasma
7309. levels of granzyme B are increased in patients with lipid-rich carotid
14. Mansfield AS, Aubry MC, Moser JC, Harrington SM, Dronca RS, Park plaques as determined by echogenicity. Atherosclerosis. 2007;195:
SS, Dong H. Temporal and spatial discordance of programmed cell e142–146. doi:10.1016/j.atherosclerosis.2007.05.001. PMID:17568588.
death-ligand 1 expression and lymphocyte tumor infiltration between 29. Naidu S, Garofalo M. microRNAs: An Emerging Paradigm in Lung
paired primary lesions and brain metastases in lung cancer. Ann Cancer Chemoresistance. Front Med. 2015;2:77. doi:10.3389/fmed.
Oncol Off J Eur Soc Med Oncol. 2016;27:1953–8. doi:10.1093/ 2015.00077..
annonc/mdw289. 30. Zhou S, Dong X, Zhang C, Chen X, Zhu J, Li W, Song X, Xu Z, Zhang
15. Van Allen EM, Miao D, Schilling B, Shukla SA, Blank C, Zimmer W, Yang X, et al. MicroRNAs are implicated in the suppression of
L, Sucker A, Hillen U, Foppen MHG, Goldinger SM, et al. Geno- CD4CCD25¡ conventional T cell proliferation by CD4CCD25C reg-
mic correlates of response to CTLA-4 blockade in metastatic mel- ulatory T cells. Mol Immunol. 2015;63:464–72. doi:10.1016/j.
anoma. Science. 2015;350:207–11. doi:10.1126/science.aad0095. molimm.2014.10.001. PMID:25457879.
PMID:26359337. 31. Wei J, Nduom EK, Kong L-Y, Hashimoto Y, Xu S, Gabrusiewicz K,
16. Garcia-Diaz A, Shin DS, Moreno BH, Saco J, Escuin-Ordinas H, Ling X, Huang N, Qiao W, Zhou S, et al. MiR-138 exerts anti-glioma
Rodriguez GA, Zaretsky JM, Sun L, Hugo W, Wang X, et al. Interferon efficacy by targeting immune checkpoints. Neuro-Oncol.
Receptor Signaling Pathways Regulating PD-L1 and PD-L2 Expres- 2016;18:639–48. doi:10.1093/neuonc/nov292. PMID:26658052.
sion. Cell Rep. 2017;19:1189–201. doi:10.1016/j.celrep.2017.04.031. 32. Ali MA, Matboli M, Tarek M, Reda M, Kamal KM, Nouh M,
PMID:28494868. Ashry AM, El-Bab AF, Mesalam HA, Shafei AE-S, et al. Epige-
17. Cabel L, Riva F, Servois V, Livartowski A, Daniel C, Rampanou A, netic regulation of immune checkpoints: another target for cancer
Lantz O, Romano E, Milder M, Buecher B, et al. Circulating tumor immunotherapy? Immunotherapy. 2017;9:99–108. doi:10.2217/imt-
DNA changes for early monitoring of anti-PD1 immunotherapy: a 2016-0111. PMID:28000527.
proof-of-concept study. Ann Oncol. 2017;28:1996–2001. doi:10.1093/ 33. Cioffi M, Trabulo SM, Vallespinos M, Raj D, Kheir TB, Lin M-L,
annonc/mdx212. PMID:28459943. Begum J, Baker A-M, Amgheib A, Saif J, et al. The miR-25-93-106b
18. Iijima Y, Hirotsu Y, Amemiya K, Ooka Y, Mochizuki H, Oyama T, cluster regulates tumor metastasis and immune evasion via modula-
Nakagomi T, Uchida Y, Kobayashi Y, Tsutsui T, et al. Very early tion of CXCL12 and PD-L1. Oncotarget. 2017;8:21609–25. doi:10.
response of circulating tumour-derived DNA in plasma predicts 18632/oncotarget.15450. PMID:28423491.
e1452581-10 A. COSTANTINI ET AL.

34. Khorrami S, Zavaran Hosseini A, Mowla SJ, Soleimani M, Rakhshani 51. Hu C, Lv L, Peng J, Liu D, Wang X, Zhou Y, Huo J. MicroRNA-375
N, Malekzadeh R. MicroRNA-146a induces immune suppression and suppresses esophageal cancer cell growth and invasion by repressing
drug-resistant colorectal cancer cells. Tumour Biol J Int Soc Oncode- metadherin expression. Oncol Lett. 2017;13:4769–75. doi:10.3892/
velopmental Biol Med. 2017;39:1010428317698365. doi:10.1177/ ol.2017.6098. PMID:28599478.
1010428317698365. 52. Lee SW, Park KC, Kim JG, Moon SJ, Kang SB, Lee DS, Sul HJ, Ji JS,
35. Xu C, Zhang Y, Wang Q, Xu Z, Jiang J, Gao Y, Gao M, Kang J, Wu M, Jeong HY. Dysregulation of MicroRNA-196b-5p and MicroRNA-375
Xiong J, et al. Long non-coding RNA GAS5 controls human embry- in Gastric Cancer. J Gastric Cancer. 2016;16:221–9. doi:10.5230/
onic stem cell self-renewal by maintaining NODAL signalling. Nat jgc.2016.16.4.221. PMID:28053808.
Commun. 2016;7:13287. doi:10.1038/ncomms13287. PMID:27811843. 53. Lian S, Park JS, Xia Y, Nguyen TT, Joo YE, Kim KK, Kim HK, Jung
36. Yee D, Shah KM, Coles MC, Sharp TV, Lagos D. MicroRNA-155 YD. MicroRNA-375 Functions as a Tumor-Suppressor Gene in Gas-
induction via TNF-a and IFN-g suppresses expression of pro- tric Cancer by Targeting Recepteur d’Origine Nantais. Int J Mol Sci.
grammed death ligand-1 (PD-L1) in human primary cells. J Biol 2016;17:p. 1633. doi:10.3390/ijms17101633.
Chem. 2017;292:20683–93. doi:10.1074/jbc.M117.809053. PMCID: 54. Zhou J, Song S, Cen J, Zhu D, Li D, Zhang Z. MicroRNA-375 is
PMC5733604. downregulated in pancreatic cancer and inhibits cell proliferation
37. Cortez MA, Ivan C, Valdecanas D, Wang X, Peltier HJ, Ye Y, Araujo in vitro. Oncol Res. 2012;20:197–203. doi:10.3727/096504013X13
L, Carbone DP, Shilo K, Giri DK, et al. PDL1 Regulation by p53 via 589503482734. PMID:23581226.
miR-34. J Natl Cancer Inst. 2016;108:djv303. doi:10.1093/jnci/djv303. 55. Song S, Zhou J, He S, Zhu D, Zhang Z, Zhao H, Wang Y, Li D. Expres-
PMID:26577528. sion levels of microRNA-375 in pancreatic cancer. Biomed Rep.
38. Zhao L, Yu H, Yi S, Peng X, Su P, Xiao Z, Liu R, Tang A, Li X, Liu F, 2013;1:393–8. doi:10.3892/br.2013.88. PMID:24648956.
et al. The tumor suppressor miR-138-5p targets PD-L1 in colorectal 56. Cui F, Wang S, Lao I, Zhou C, Kong H, Bayaxi N, Li J, Chen Q, Zhu T,
cancer. Oncotarget. 2016;7:45370–84. PMID:27248318. Zhu H. miR-375 inhibits the invasion and metastasis of colorectal can-
39. Zhu J, Chen L, Zou L, Yang P, Wu R, Mao Y, Zhou H, Li R, Wang K, cer via targeting SP1 and regulating EMT-associated genes. Oncol
Wang W, et al. MiR-20b, -21, and -130b inhibit PTEN expression Rep. 2016;36:487–93. doi:10.3892/or.2016.4834. PMID:27222350.
resulting in B7-H1 over-expression in advanced colorectal cancer. 57. Yang D, Yan R, Zhang X, Zhu Z, Wang C, Liang C, Zhang X. Deregu-
Hum Immunol. 2014;75:348–53. doi:10.1016/j.humimm.2014.01.006. lation of MicroRNA-375 inhibits cancer proliferation migration and
PMID:24468585. chemosensitivity in pancreatic cancer through the association of
40. Li Q, Johnston N, Zheng X, Wang H, Zhang X, Gao D, Min W. miR- HOXB3. Am J Transl Res. 2016;8:1551–9. PMID:27186281.
28 modulates exhaustive differentiation of T cells through silencing 58. Kong KL, Kwong DLW, Chan TH-M, Law SY-K, Chen L, Li Y, Qin Y-
programmed cell death-1 and regulating cytokine secretion. Oncotar- R, Guan X-Y. MicroRNA-375 inhibits tumour growth and metastasis
get. 2016;7:53735–50. PMID:27447564. in oesophageal squamous cell carcinoma through repressing insulin-
41. Wang X, Li J, Dong K, Lin F, Long M, Ouyang Y, Wei J, Chen X, like growth factor 1 receptor. Gut. 2012;61:33–42. doi:10.1136/gutjnl-
Weng Y, He T, et al. Tumor suppressor miR-34a targets PD-L1 and 2011-300178. PMID:21813472.
functions as a potential immunotherapeutic target in acute myeloid 59. Liu AM, Poon RTP, Luk JM. MicroRNA-375 targets Hippo-sig-
leukemia. Cell Signal. 2015;27:443–52. doi:10.1016/j.cellsig.2014. naling effector YAP in liver cancer and inhibits tumor properties.
12.003. PMID:25499621. Biochem Biophys Res Commun. 2010;394:623–7. doi:10.1016/j.
42. Lv Q-L, Du H, Liu Y-L, Huang Y-T, Wang G-H, Zhang X, Chen S-H, bbrc.2010.03.036. PMID:20226166.
Zhou H-H. Low expression of microRNA-320b correlates with tumor- 60. Shao Y, Geng Y, Gu W, Huang J, Ning Z, Pei H. Prognostic sig-
igenesis and unfavorable prognosis in glioma. Oncol Rep. 2017; nificance of microRNA-375 downregulation in solid tumors: a
38:959–66. doi:10.3892/or.2017.5762. PMID:28656255. meta-analysis. Dis Markers. 2014;2014:626185. doi:10.1155/2014/
43. Li Y, Jiang Q, Xia N, Yang H, Hu C. Decreased expression of micro- 626185. PMID:25404787.
RNA-375 in nonsmall cell lung cancer and its clinical significance. J 61. Hu Y, Wang L, Gu J, Qu K, Wang Y. Identification of microRNA dif-
Int Med Res. 2012;40:1662–9. doi:10.1177/030006051204000505. ferentially expressed in three subtypes of non-small cell lung cancer
PMID:23206448. and in silico functional analysis. Oncotarget. 2017;8:74554–66.
44. Chen L-J, Li X-Y, Zhao Y-Q, Liu W-J, Wu H-J, Liu J, Mu X-Q, Wu H- PMID:29088807.
B. Down-regulated microRNA-375 expression as a predictive bio- 62. Chen X, Li B, Luo R, Cai S, Zhang C, Cao X. Analysis of the function
marker in non-small cell lung cancer brain metastasis and its prognos- of microRNA-375 in humans using bioinformatics. Biomed Rep.
tic significance. Pathol Res Pract. 2017;213:882–8. doi:10.1016/j. 2017;6:561–6. doi:10.3892/br.2017.889. PMID:28515914.
prp.2017.06.012. PMID:28688608. 63. Selth LA, Das R, Townley SL, Coutinho I, Hanson AR, Centenera
45. Zou Q, Yi W, Huang J, Fu F, Chen G, Zhong D. MicroRNA-375 tar- MM, Stylianou N, Sweeney K, Soekmadji C, Jovanovic L, et al. A
gets PAX6 and inhibits the viability, migration and invasion of human ZEB1-miR-375-YAP1 pathway regulates epithelial plasticity in pros-
breast cancer MCF-7 cells. Exp Ther Med. 2017;14:1198–204. tate cancer. Oncogene. 2017;36:24–34. doi:10.1038/onc.2016.185.
doi:10.3892/etm.2017.4593. PMID:28810579. PMID:27270433.
46. Dai X, Chiang Y, Wang Z, Song Y, Lu C, Gao P, Xu H. Expression lev- 64. Spranger S, Gajewski TF. A new paradigm for tumor immune escape:
els of microRNA-375 in colorectal carcinoma. Mol Med Rep. b-catenin-driven immune exclusion. J Immunother Cancer.
2012;5:1299–304. PMID:22377847. 2015;3:43. doi:10.1186/s40425-015-0089-6. PMID:26380088.
47. Xu L, Li M, Wang M, Yan D, Feng G, An G. The expression of micro- 65. Ramos RN, Piaggio E, Romano E. Mechanisms of Resistance to
RNA-375 in plasma and tissue is matched in human colorectal cancer. Immune Checkpoint Antibodies. Handb Exp Pharmacol. 2017;
BMC Cancer. 2014;14:714. doi:10.1186/1471-2407-14-714. PMID: doi:10.1007/164_2017_11. PMID:28315073. [Epub ahead of print]
25255814. 66. Moroishi T, Hayashi T, Pan W-W, Fujita Y, Holt MV, Qin J, Carson
48. Alam KJ, Mo J-S, Han S-H, Park W-C, Kim H-S, Yun K-J, Chae S-C. DA, Guan K-L. The Hippo Pathway Kinases LATS1/2 Suppress Can-
MicroRNA 375 regulates proliferation and migration of colon cancer cer Immunity. Cell. 2016;167:1525–39.e17.
cells by suppressing the CTGF-EGFR signaling pathway. Int J Cancer. 67. Siemers NO, Holloway JL, Chang H, Chasalow SD, Ross-MacDon-
2017;141:1614–29. doi:10.1002/ijc.30861. PMID:28670764. ald PB, Voliva CF, Szustakowski JD. Genome-wide association
49. Xu L, Wen T, Liu Z, Xu F, Yang L, Liu J, Feng G, An G. MicroRNA- analysis identifies genetic correlates of immune infiltrates in solid
375 suppresses human colorectal cancer metastasis by targeting Friz- tumors. PloS One. 2017;12:e0179726. doi:10.1371/journal.pone.
zled 8. Oncotarget. 2016;7:40644–56. PMID:27276676. 0179726. PMID:28749946.
50. Lv H, He Z, Wang H, Du T, Pang Z. Differential expression of miR-21 68. Kakavand H, Rawson RV, Pupo GM, Yang JYH, Menzies AM, Carlino
and miR-75 in esophageal carcinoma patients and its clinical implica- MS, Kefford RF, Howle JR, Saw RPM, Thompson JF, et al. PD-L1
tion. Am J Transl Res. 2016;8:3288–98. PMID:27508050. Expression and Immune Escape in Melanoma Resistance to MAPK
 ONCOIMMUNOLOGY e1452581-11

Inhibitors. Clin Cancer Res Off J Am Assoc Cancer Res. Maxillofac Surg. 2013;42:949–55. doi:10.1016/j.ijom.2013.04.
2017;23:6054–61. doi:10.1158/1078-0432.CCR-16-1688.. 016. PMID:23726271.
69. Pai SG, Carneiro BA, Mota JM, Costa R, Leite CA, Barroso-Sousa R, 73. Robinson MD, Oshlack A. A scaling normalization method for differ-
Kaplan JB, Chae YK, Giles FJ. Wnt/beta-catenin pathway: modulating ential expression analysis of RNA-seq data. Genome Biol. 2010;11:
anticancer immune response. J Hematol OncolJ Hematol Oncol. R25. doi:10.1186/gb-2010-11-3-r25. PMID:20196867.
2017;10:101. doi:10.1186/s13045-017-0471-6.. 74. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor pack-
70. Massi D, Romano E, Rulli E, Merelli B, Nassini R, De Logu F, Bieche I, age for differential expression analysis of digital gene expression data.
Baroni G, Cattaneo L, Xue G, et al. Baseline b-catenin, programmed Bioinformatics. 2010;26:139–40. doi:10.1093/bioinformatics/btp616.
death-ligand 1 expression and tumour-infiltrating lymphocytes pre- PMID:19910308.
dict response and poor prognosis in BRAF inhibitor-treated mela- 75. McCarthy DJ, Chen Y, Smyth GK. Differential expression analysis of
noma patients. Eur J Cancer Oxf Engl 1990. 2017;78:70–81. multifactor RNA-Seq experiments with respect to biological variation.
71. Wei R, Yang Q, Han B, Li Y, Yao K, Yang X, Chen Z, Yang S, Zhou J, Nucleic Acids Res. 2012;40:4288–97. doi:10.1093/nar/gks042. PMID:
Li M, et al. microRNA-375 inhibits colorectal cancer cells proliferation 22287627.
by downregulating JAK2/STAT3 and MAP3K8/ERK signaling path- 76. Ru Y, Kechris KJ, Tabakoff B, Hoffman P, Radcliffe RA, Bowler R,
ways. Oncotarget. 2017;8:16633–41. PMID:28186962. Mahaffey S, Rossi S, Calin GA, Bemis L, et al. The multiMiR R pack-
72. Wang J, Huang H, Wang C, Liu X, Hu F, Liu M. MicroRNA-375 age and database: integration of microRNA–target interactions along
sensitizes tumour necrosis factor-alpha (TNF-a)-induced apoptosis with their disease and drug associations. Nucleic Acids Res. 2014;42:
in head and neck squamous cell carcinoma in vitro. Int J Oral e133¡e133. doi:10.1093/nar/gku631. PMID:25063298.