Blood immune cell biomarkers in lung cancer

D Riemann; M Cwikowski; S Turzer; T Giese; M Grallert; W Schütte; B Seliger

doi:10.1111/cei.13219

. 2018 Nov 9;195(2):179–189. doi: 10.1111/cei.13219

Blood immune cell biomarkers in lung cancer

D Riemann ^1,^✉, M Cwikowski ¹, S Turzer ¹, T Giese ², M Grallert ³, W Schütte ⁴, B Seliger ¹

PMCID: PMC6330648 PMID: 30246868

Summary

Characterization of host immune cell parameters prior to treatment is expected to identify biomarkers predictive of clinical outcome as well as to elucidate why some patients fail to respond to immunotherapy. We monitored blood immune cells from 58 patients with non‐small‐ cell lung cancer (NSCLC) undergoing surgery of the primary tumor and from 50 age‐matched healthy volunteers. Complete leukocyte blood count, the number of circulating dendritic cells (DC), HLA‐DR^low monocytes and several lymphocytic subpopulations were determined by eight‐color flow cytometry. Furthermore, the prognostic value of the immune cell parameters investigated was evaluated by patients’ survival analysis. Compared to the control group, blood of NSCLC patients contained more neutrophils resulting in a higher neutrophil‐to‐lymphocyte ratio (NLR), but a lower number of blood DC, in particular of plasmacytoid DC (pDC), natural killer (NK) cells and naive CD4⁺ and CD8⁺ T cells. Furthermore, a higher frequency of CD4⁺ regulatory T cells (Treg) and HLA‐DR^low monocytes was detected, and smoking had a significant impact on these values. HLA‐DR^low monocytes were positively correlated to the number of neutrophils, monocytes and NLR, but negatively associated with the number of pDC and naive CD4⁺ T cells. The frequency of Treg, HLA‐DR^low monocytes and naive CD4⁺ and CD8⁺ T cells as well as the ratios of CD4/HLA‐DR^low monocytes and HLA‐DR^low monocytes/pDC correlated with patient’s overall survival. Next to Treg, HLA‐DR^low monocytes and naive T cells represent prognostic markers for NSCLC patients and might be useful for monitoring of patients’ responses to immunotherapies in future studies.

Keywords: biomarker, dendritic cells, flow cytometry, immune monitoring, lung cancer, MDSC, naive T cells, NLR, overall survival

Introduction

Lung cancer is the most common cause of cancer death worldwide 1, with cigarette smoking as the main risk factor 2. Non‐small‐cell lung cancer (NSCLC), accounting for approximately 83% of all patients with lung cancer, is typically subdivided into adenocarcinoma (50–70%), squamous cell carcinoma (20–30%) and other subtypes (< 10%) 3. At diagnosis approximately 75% of NSCLC patients have an advanced disease associated with a bad prognosis and low survival rate 4. For stages I and II NSCLC the majority of patients (approximately 70%) undergo surgery, while most patients with stages III and IV NSCLC receive chemotherapy with or without radiation 5. Despite the tremendous developments in early detection and novel treatment modalities, the overall survival (OS) of lung cancer patients has not improved greatly during the past decades. Current studies have shown benefits of immunotherapy in lung tumors 6, in particular those targeting the immune checkpoint proteins programmed cell death 1/programmed cell death ligand 1 (PD‐1/PD‐L1). Many of these therapies are of greater benefit to immune‐competent patients 7, whereas systemic immune suppression is often observed in patients with advanced NSCLC 8.

Prediction of long‐term disease‐free survival of cancer patients immediately after surgical resection of clinically localized and advanced disease is of great value for patient counseling, scheduling follow‐up imaging and identifying poor risk‐group patients who might benefit from enrollment in adjuvant therapy protocols. Prognosis of cancer patients is based not only on tumor‐related factors, but also on host‐related factors, including systemic immune cell activation 9. It has been recently demonstrated that immune monitoring of peripheral blood cells might lead to the identification of biomarkers, which could serve to predict prognosis and/or therapy response/resistance 10.

The aim of this study was the identification of surrogate markers of host immunity in NSCLC patients. In total, blood samples collected from a cohort of 58 NSCLC patients undergoing surgery of the primary tumor and from 50 volunteers of an age‐matched control group were investigated. Multi‐color flow cytometry was used to determine the frequency, composition and activity of different immune cell subpopulations in peripheral blood. The impact of smoking, gender and age was analyzed and their prognostic value evaluated.

Patients and methods

Blood samples

This study was approved by the institutional review board of the Medical Faculty of the Martin Luther University and informed consent was obtained from all participants. Ethylenediamine tetraacetic acid (EDTA) peripheral blood samples were obtained from 58 patients with lung cancer undergoing primary surgery at the Department of Thorax Surgery of the Hospital Martha‐Maria Halle‐Dölau (Table 1). Patients included in the analysis met the following criteria: (1) curative resection with lymph node dissection; and (2) neither radiotherapy nor chemotherapy administered prior to the surgery. Nearly all patients had a Karnofsky performance status > 80%. All surgical tissues were verified by a pathologist. For the analysis of the OS, patients were censored at the time of their last clinical follow‐up appointment or their date of death. The median follow‐up duration for the detection of OS was 26·5 months (range 1–49 months).

Table 1.

Characteristics of patients and control group

	Patients n = 58	Control group n = 50
Age (median)	67 ± 9	60·9 ± 8·6
Sex Male Female	35 23	17 33
Smoking status Current smoker or former smoker < 6 months Former smoker > 6 months Never smoker	20 (34·5%) 26 (45%) 12 (21%)	5 (10%) 13 (26%) 32 (64%)
Histology Adenoca Squamous Other histological subtypes	34 (59%) 13 (22%) 11 (19%)
Stage pT1/2 pT3/4	42 (72%) 14 (24%)

Open in a new tab

Fifty age‐matched normal healthy donors without any known tumor disease and without any systemic immune‐suppressive therapy served as a control group (Table 1); 79% of the patients and 36% of the control group were smokers or had a smoking history (Table 1).

Flow cytometry and antibody staining

Immune cell profiling in peripheral blood is an important tool for biomarker identification. A lysed whole blood technique with eight‐color staining of blood cells was used. The monoclonal antibodies (mAb) purchased were CD56‐phycoerythrin (PE) from Beckmann Coulter (Hamburg, Germany), CD57‐allophycocyanin (APC) and CD27‐brilliant violet (BV) 421 from BioLegend (Fell, Germany); all other mAbs were from BD Biosciences (Heidelberg, Germany) (Multitest™ IMK kit, CD45‐fluorescein isothiocyanate (FITC), CCR7‐PE, CD25‐PE, CD4‐peridinin chlorophyll (PerCP), CD19‐PerCP, CD123‐PerCP, CD45RA‐PE‐cyanin 7 (Cy7), CCR4‐PE‐Cy7, CD11c‐PE‐Cy7, CD38‐APC, CD16‐APC, CD127‐Alexa Fluor 647, CD8‐APC‐H7, CD20‐APC‐H7, CD3‐APC‐H7, CD19‐APC‐H7, CD45RO‐APC‐H7, CD3‐V450, CD14‐V450, HLA‐DR‐V500; Supporting information, Table S1). Samples were measured with a fluorescence‐activated cell sorter (FACS) CANTO II flow cytometer (BD Biosciences). Data analyses were performed with BD FACS DIVA^TM software. Because standardized procedures are essential to allow for interindividual comparisons in the context of studies persisting for several months, Cytometer Setup and Tracking (CST) Beads (BD Biosciences) were used daily to set standardized geometric mean fluorescence intensity (MFI) ranges in the fluorescence channels used. Absolute values of CD4⁺ and CD8⁺ T cells, B cells and natural killer (NK) cells were determined using the BD Multitest™ IMK kit and BD Trucount^TM tubes with a no‐wash procedure, according to the manufacturer’s instructions (Supporting information, Table S1). Additionally, the relative percentages of CD45RA⁺CCR7⁺ naive, CCR7^–CD45RA^– effector memory (EM) and CD57⁺ immunosenescent T cells, as well as CD38^–HLA‐DR⁺ T cells, were evaluated. CD4⁺CD127^–/^dimCD25⁺⁺regulatory T cells (Treg) were identified as described by Liu et al. 11. Furthermore, the total number of dendritic cells (DC; HLA‐DR⁺ and lineage negative for CD3, CD19, CD20), with CD11c⁺ myeloid and CD123⁺ plasmacytoid DC (pDC) as well as the percentage of CD16⁺CD14^dim non‐classical monocytes, was measured. HLA‐DR expression on monocytes was quantified using mAbs labeled on a protein/fluorophore ratio of 1/1 (QuantiBRITE™ reagents; BD Biosciences). The anti‐HLA‐DR 1/1 PE (clone L243)/anti‐CD14 PerCP‐Cy5.5 mAb was used, according to the manufacturer’s instructions. A standard curve for antigen quantification was established using multi‐level calibrated QuantiBRITE beads. The measured geometric MFI of the gated population was converted into antibody molecules bound per cell (ABC) using a linear regression analysis. HLA‐DR MFI values of ≤ 5000 ABC for the whole monocyte population have been designated as ‘immunoparalysis’ in former studies, as the patients are at high risk of infectious diseases 12. Taking an MFI of 5000 ABC as borderline value for a low HLA‐DR intensity, the amount of HLA‐DR^low monocytes was estimated as a percentage of CD14⁺ cells and cell numbers/μl blood. The gating strategy for DC is illustrated in Supporting information, Fig. S1; the gating and calculation of HLA‐DR^low monocytes is shown in Supporting information, Fig. S2.

Statistical analysis

The statistical analysis was performed using the commercial software spss 22.0 (SPSS Inc., Munich, Germany). Differences in the absolute number of immune cells between patients and healthy volunteers or between patients with different tumor stages were analyzed using Student’s t‐test or Wilcoxon–Mann–Whitney test, as appropriate. Analysis of variance (anova) was used to consider the effect of smoking, age or gender. Bonferroni–Holm correction was used for multiple comparisons. All P‐values are exploratory. To evaluate correlations between Treg and HLA‐DR^low monocytes with other immune cell parameters, Spearman’s correlation coefficients (CC) were calculated. OS was defined as the length of time from surgery to death or last follow‐up. Survival analysis first comprised a descriptive presentation of the cumulative survival functions according to Kaplan–Meier, and differences among the curves were evaluated using the log‐rank test. Univariate and multivariate analyses were performed using Cox proportional hazards model, adjusted to tumor stage. Two‐sided P‐values of less than 0·05 were considered statistically significant.

Results

Comparison of immune cells in NSCLC patients compared to an age‐matched healthy control group

Granulocyte counts and neutrophil‐to‐lymphocyte ratio (NLR)

NSCLC patients when compared to healthy donors had a significantly higher number of leukocytes due to higher neutrophil counts, as illustrated in Table 2. In most cases, cancer patients had a mild granulocytosis; only one patient had a value > 10 000 neutrophils/μl blood. Correlation of neutrophil counts with smoking as a risk factor for lung cancer showed that smoking (former and current smokers as one group compared to never smokers) significantly influences the neutrophil counts (P = 0·032), although the distinction between control and tumor group had a higher impact on neutrophil counts (P = 0·003). Patients with squamous cell carcinoma had higher neutrophil counts than adenocarcinoma patients (Fig. 1a). No differences in neutrophil numbers could be detected in patients with early (T1/2) versus late (T3/4) tumor stages and with respect to gender and age. The number of blood eosinophils showed a high standard deviation (s.d.), and there was no significant difference between the control and tumor group (P = 0·650). In patients with squamous cell carcinoma, the number of blood eosinophils was elevated in late tumor stages (117 ± 150 cells/μl in T1/2; 291 ± 201 cells/μl blood in T3/4). For several malignancies, a higher NLR has been discussed as a predictor of patients’ outcome 13. As shown in Fig. 1b, NSCLC patients had higher NLR values than the control group, although with a high s.d. in T1/2 stages. Although smokers had higher NLR values, smoking had no impact on NLR in our analyses (P = 0·212). Furthermore, there was no difference in NLR values between adenocarcinoma and squamous cell carcinoma (3·08 ± 1·43 versus 3·38 ± 1·49).

Table 2.

Comparison of blood immune cells in lung cancer patients [mean ± standard deviation (s.d.); n = 57] and a control group (n = 50) as result of flow cytometric analyses (t‐test with Bonferroni–Holm correction).

		Control	Tumor	p
Leukocyte counts*	Cells/μl	6715 ± 1642	8128 ± 2303	< 0·001
Neutrophils*	Cells/μl	3854 ± 1335	5425 ± 2018	< 0·001
Monocytes*	Cells/μl	498 ± 192	559 ± 220
Lymphocytes*	Cells/μl	2115 ± 614	1914 ± 683
NLR*	Ratio (cells/μl)	1·95 ± 0·9	3·1 ± 1·36	< 0·001
HLA‐DR^low MDSC	% of monocytes	2·61 ± 2·54	5·31 ± 4·1	0·007
CD4⁺T cells/MDSC	Ratio (cells/μl)	211 ± 260	62·9 ± 73·5	< 0·001
Naive CD4⁺T/MDSC	Ratio (cells/μl)	94·4 ± 138	19·4 ± 31·1	0·001
CD16⁺CD14^dim monocytes	% monocytes	4·9 ± 2·3	5·65 ± 3·1
Total DC count	Cells/μl	80·6 ± 41·8	64·7 ± 32·5	0·031
pDC	Cells/μl	7·5 ± 4·2	4·2 ± 2·35	< 0·001
Ratio MDSC/pDC	Ratio (Cells/μl)	2·5 ± 3·9	9·7 ± 15·8	0·002
T cells	Cells/μl	1396 ± 70	1320 ± 69
B cells	Cells/μl	217 ± 138	183 ± 103
NK cells	Cells/μl	352 ± 178	234 ± 120	< 0·001
CD4⁺ T cells	Cells/μl	953 ± 374	849 ± 358
CD8⁺ T cells	Cells/μl	378 ± 24	408 ± 32
Naive CD4⁺ T cells	Cells/μl	415 ± 254	251 ± 209	< 0·001
Naive CD8⁺ T cells	Cells/μl	92 ± 62	58 ± 39	0·001
CD4⁺ Treg	% CD4⁺ T cells	7·5 ± 1·6	8·6 ± 2·5	0·006
CD38^neg DR⁺CD4⁺ T cells	% CD4⁺ T cells	3·9 ± 1·9	6·6 ± 3·0	< 0·001
CD57⁺CD4⁺ T cells	% CD4⁺ T cells	2·4 ± 3·3	5·5 ± 8·1	0·012
CD4⁺ EM T cells	% CD4⁺ T cells	19·4 ± 8·8	26 ± 11·3	0·001
CD57⁺CD8⁺ T cells	% CD8⁺ T cells	17·7 ± 14·9	22·7 ± 18·7

Open in a new tab

NLR = neutrophil‐to‐lymphocyte‐ratio; HLA‐DR = human leucocyte antigen D‐related; MDSC = myeloid‐derived suppressor cells; pDC = plasmacytoid DC; NK = natural killer; EM = effector memory.

Comparison of blood immune cells in non‐small‐cell lung cancer (NSCLC) patients and a control group. Box‐plots shows a significant difference in the neutrophil counts between control and tumor groups with adenocarcinoma (AC) and squamous cell carcinoma (SCC) (a), an increase in the neutrophil‐to‐lymphocyte‐ratio (NLR) (b) and in monocyte counts, especially in late tumor stages (c), and a decrease of the plasmacytoid DC (pDC) numbers with age (d; 1 < 60; 2 60 < × < 72; 3 ≥ 72 years old). Results are given as median ± standard deviation (s.d.). Asterisks mark significant differences [analysis of variance (anova), Bonferroni].

Blood monocytes

NSCLC patients showed only marginally higher monocyte counts than the control group (Table 2), with a slight increase of monocytes in late tumor stages (Fig. 1c). Smoking had a significant influence on monocyte counts (P = 0·021). For the smoker subgroup, we observed significant higher monocyte counts in late compared to early tumor stages (717 ± 313 cells/μl in T3/4 versus 534 ± 184 cells/μl blood in T1/2; P = 0·020). Men had higher monocyte counts than women (608 ± 252 versus 457 ± 147). Male smokers had the highest monocyte counts in our analyses (642 ± 272 cells/μl) and female never‐smokers the lowest values (417 ± 119 cells/μl). CD14^lowCD16⁺ non‐classical monocytes represented approximately 5–6% of monocytes and their frequency was not altered between the control group and patients (Table 2). As HLA‐DR^low monocytes representing a subtype of myeloid‐derived suppressor cells (MDSC) are known to be elevated in cancer patients 14, the monocytic HLA‐DR intensity was quantified, resulting in an only marginally lower MFI in cancer patients (ABC = 30 134 ± 12 128) compared to healthy volunteers (ABC = 35 147 ± 9993). Smoking had a significant impact on monocytic HLA‐DR intensity (P < 0·001). Using a threshold of 5000 ABC to define monocytes with a strongly diminished HLA‐DR intensity, the lung cancer group had a higher percentage of those HLA‐DR^low monocytes (Table 2), with 4·6 ± 3·1% of MDSC in T1/2 and 7·5 ± 6·0 in T3/4. Smoking affected the value. Within the control group, never‐smokers had 1·6 ± 1·3%, but smokers (former and current) had 4·7 ± 3·5% HLA‐DR^low monocytes. Smokers in the tumor group had the highest percentages of HLA‐DR^low MDSC (5·9 ± 4·3%). Figure 2 illustrates the impact of smoking behavior and tumor diagnosis on the percentages of HLA‐DR^low monocytes. We found a positive correlation between HLA‐DR^low MDSC and neutrophil counts with a Spearman’s rank correlation of 0·434 (Table 3). A ratio of CD4⁺ T cells/HLA‐DR^low MDSC was – despite a high s.d. – significantly lower in the tumor compared to the control group (Table 2), and smokers had a lower ratio than never‐smokers. Never‐smoking women in the control group had the highest ratio: CD4⁺ T cells/HLA‐DR^low MDSC (305 ± 294), although the gender difference was not significant (P = 0·090). A ratio of naive CD4⁺ T cells/HLA‐DR^low MDSC also differed between control group and NSCLC patients (Table 2), again with a high s.d., and with a strong impact of smoking (97·2 ± 142·6 in never‐smokers, 24·8 ± 46·6 in smokers).

Impact of different parameters on immune cells, given as estimated marginal means from analysis of variance (anova) analysis. The percentages of HLA‐DR^low myeloid‐derived suppressor cells (MDSC) are more affected by the smoking behavior (non‐smoker means never‐smoker) than by tumor disease (a). Natural killer (NK) cell counts are lower in non‐small‐cell lung cancer (NSCLC) patients, with lower values in women in tumor and control group (b). Both naive CD4⁺ T cells (c) and naive CD8⁺ T cells (d) decrease with tumor stages, with lower values in people >70 years age.

Table 3.

Association of regulatory T cells (Treg) and HLA‐DR^low monocytes with several immune cell parameters analysed by Spearman’s rank correlation. Correlation coefficient (CC) and P‐value are given; data are significant with P < 0·05 after Bonferroni–Holm correction.

	CC	P
HLA‐DR^low MDSC (% of monocytes) Neutrophil counts (cells/μl blood)	0·434	< 0·001
HLA‐DR^low MDSC (cells/μl blood) Naive CD4⁺ T cells (cells/μl blood)	–0·248	0·010
HLA‐DR^low MDSC (% of monocytes) pDC (cells/μl blood)	–0·266	0·006
Treg (% of CD4⁺ T cells) NLR	0·258	0·007
Treg (% of CD4⁺ T cells) CD4⁺ T cells	–0·294	0·002
Treg (% of CD4⁺ T cells) Naive CD4⁺ T cells	–0·230	0·018

Open in a new tab

MDSC = myeloid‐derived suppressor cells; pDC = plasmacytoid DC; NLR = neutrophil‐to‐lymphocyte‐ratio.

Dendritic cells

Cancer patients had a lower amount of DC in blood compared to the control group (Table 2). The majority of DC was CD11c⁺ myeloid DC, with pDC accounting for only 9·5 ± 5·5% of total DC. The number of DC decreased with age, as illustrated for pDC in Fig. 1d. pDC counts correlated negatively with the percentage of HLA‐DR^low MDSC (Table 3). The ratio of HLA‐DR^low MDSC/pDC had significantly higher values in the tumor group (Table 2), especially in late tumor stages (T1/2 6·9 ± 8·3 versus T3/4 19·4 ± 28·1).

Blood lymphocytes and lymphocytic subpopulations

The lymphocyte count was not different between the control and tumor groups (Table 2) and was not altered in smokers. However, the lymphocyte counts declined with age, especially above age > 70 years (2150 ± 670 lymphocytes/μl blood in people < 70 years versus 1700 ± 510 lymphocytes ≥ 70 years, P = 0·001). By trend, women had higher lymphocyte numbers than men (2110 ± 700 versus 1890 ± 600 cells/μl blood, P = 0·077). Concerning lymphocytic subpopulations, there was no difference between control and tumor groups for B and T cell counts, as well as for the number of CD4⁺ and CD8⁺ T cells. In contrast, a lower amount of NK cells was found in the blood of cancer patients, with higher NK cell numbers in men compared to women (Fig. 2b). Smoking and age had no impact on NK cell counts in our analyses. A more detailed characterization of NK cells revealed no difference in the percentage of CD56^brightCD16^dim NK cells in the control and tumor groups (3·8 ± 0·3% of NK cells in control group versus 3·6 ± 0·3 in patients).

The counts of CD4⁺ and CD8⁺ T cells were highly variable, ranging for CD4⁺ T cells from 240 to 2320 cells/μl and for CD8⁺ T cells from 50 to 1230 cells/μl blood. Whereas no difference was found between tumor and control groups, CD4⁺ T cell numbers were significantly higher in women versus men (965 ± 390 versus 820 ± 310 cells/μl blood; P = 0·032), and declined with age with a significant drop in individuals aged > 70 years (980 ± 380 ≤ 70 years versus 720 ± 270 cells/μl blood > 70 years). In comparison to CD4⁺ T cells, the number of CD8⁺ T cells showed neither gender‐specific nor age‐dependent differences (data not shown).

Higher percentages of CD127^–CD25⁺⁺CD4⁺ Treg (% of CD4⁺ T cells) were found in cancer patients (Table 2) with a significant impact of smoking behavior. Smokers when compared with never‐smokers had higher percentages of Treg (8·7 ± 2·4 versus 7·3 ± 1·5% of CD4⁺ T cells, P = 0·017), while the subgroup ‘never‐smokers’ in control versus tumor group did not show any difference (7·3 ± 1·2 versus 7·2 ± 2·2% of CD4⁺ T cells). The absolute count of Treg was 71 ± 32 cells/μl blood, without any differences between patients and control group; 42·4 ± 19·9% of Treg were memory Treg (CD45RO⁺CCR4⁺). The percentage of Treg correlated negatively with the numbers of CD4⁺ T cells as well as with naive CD4⁺ T cells (Table 3). Based on their effect on lymphocytes, Treg correlated positively with the NLR (Table 3).

The counts of both naive CD4⁺ and CD8⁺ T cells were significantly lower in cancer patients compared to the control group (Table 2). In particular, the percentage of naive CD4⁺ T cells dropped in cancer patients (42 ± 13% of CD4⁺ T cells in the control group versus 28 ± 15% in the tumor group; P < 0·001). Naive CD4⁺ T cells correlated negatively both with the percentages of HLA‐DR^low MDSC as well as with Treg (Table 3). Furthermore, for NSCLC a decreased number of naive CD4⁺ T cells and naive CD8⁺ T cells was found in patients aged > 70 years (Fig. 2c,d). In contrast to naive CD4⁺T cells, the percentage of EM CD4⁺ T cells was significantly higher in cancer patients (Table 2). Furthermore, cancer patients had a higher number of HLA‐DR⁺ T cells (161 ± 108 cells/μl blood compared to 124 ± 73 in the control group, P = 0·04), as well as a higher percentage of CD38^–HLA‐DR⁺CD4⁺ T cells (Table 2).

While the percentage of CD57⁺CD4⁺ T cells was significantly higher in cancer patients, the percentage of CD57⁺CD8⁺ T cells did not differ significantly between control and tumor groups (Table 2).

Survival analyses

Survival analysis first comprised a descriptive presentation of the cumulative survival functions according to Kaplan–Meier. As expected, patients in late tumor stages exhibited shorter OS. Otherwise, never‐smoking was not correlated with a favorable prognostic risk (Supporting information, Table S2). Figure 3 illustrates that a higher percentage of HLA‐DR^low MDSC and Treg as well as a lower number of naive CD4⁺ T cells and naive CD8⁺ T cells correlated with shorter OS. Furthermore, a lower ratio CD4⁺ T cells/HLA‐DR^low MDSC as well as a higher ratio MDSC/pDC was associated with poor survival. Although Kaplan–Meier curves illustrated an adverse prognostic risk for patients with a higher NLR (≥ 4·5: survival time 26·9 ± 3·9 versus <4·5: 41·6 ± 2·1 months), NLR was not associated with a significant prognostic risk in log‐rank test in our analysis (Supporting information, Table S2). Furthermore, Kaplan–Meier curves showed no difference in the survival time for T cell numbers (data not shown), as well as for CD4⁺ T cells and CD8⁺ T cells (Supporting information, Table S2). Similarly, the percentages of pDC did not correlate with survival time (Supporting information, Table S2).

Relationship between immune cell parameters and overall survival (OS). A high percentage of HLA‐DR^low myeloid‐derived suppressor cells (MDSC) (a) and regulatory T cells (Treg) (b) and a low number of naive CD4⁺ and CD8⁺ T cells (e,f) were associated with poorer survival. Also the ratio of CD4⁺ T cells/MDSC (c) and the ratio of MDSC/plasmacytoid DC (pDC) (d) correlated with OS. Kaplan–Meier analysis is shown with median survival time in both groups; patients with censored survival times are denoted by tick‐marks. The P‐values are 0·019 (a), 0·037 (b), 0·027 (c), 0·019 (d), 0·024 (e) and 0·030 (f).

Univariate Cox proportional hazards regression on OS confirmed that a higher percentage of HLA‐DR^low MDSC (> 3% of monocytes) was associated with an increased OS risk; the hazard ratio (HR) for death was 8·301 (P = 0·046). Univariate Cox regression also confirmed a poor OS for patients with a lower number of naive CD4⁺ T cells (P = 0·037) and naive CD8⁺ T cells (P = 0·041). Adjusted to tumor stages (T1/2 against T3/4, multivariate Cox regression, Supporting information, Table S2), patients with > 3% HLA‐DR^low monocytes had a HR of 6·01 (P = 0·09), but due to the low case number (n = 51) the 95% confidence interval was imprecise (0·75–48·34). A higher Treg percentage showed an increased survival risk in Cox regression adjusted to tumor stages; patients with ≥ 10% Treg had a HR of 3·604 (1·012–12·832) (P = 0·048, Supporting information, Table S2).

Our blood‐based biomarker panel for the immune monitoring of cancer patients was affected by various factors. In summary, smoking increased the counts of leukocytes, neutrophils and monocytes as well as the percentages of HLA‐DR^low MDSC and Treg. Supporting information, Table S3 illustrates that current smokers have higher leukocyte counts and neutrophil numbers than former smokers (significant difference only between never smokers and current smokers). In contrast, the percentage of HLA‐DR^low monocytes and Treg remain increased in former smokers, indicating a long‐lasting impact of smoking on immunosuppressive cells. Treg data were only significant for the cancer patients group. The number of monocytes and of NK cells was higher in men, whereas the CD4⁺ T cell counts showed higher values in women. The numbers of DC, pDC, lymphocytes, CD4⁺ T cells, naive CD4⁺ T cells as well as naive CD8⁺ T cells dropped with age. Factors correlating with patients’ OS were the percentages of HLA‐DR^low MDSC and of Treg, the number of naive CD4⁺ and naive CD8⁺ T cells as well as the ratio MDSC/pDC.

Discussion

Flow cytometry serves as a powerful analytical platform for the rapid characterization of individual cells within heterogenic cell populations. In this study, multi‐color flow cytometry was used to investigate the clinical relevance of immune cell subpopulations and to identify predictive surrogate markers in the peripheral blood of NSCLC patients undergoing surgery of the primary tumor. With this approach, several differences were found compared to an age‐matched control group without any known tumor disease. Our data are mainly expressed as cells/μl blood; this allows a better comparison of values with known reference ranges. The blood of lung cancer patients contained a higher number of neutrophils resulting in higher leukocyte counts, and patients with squamous NSCLC had higher neutrophil counts than adenocarcinoma patients. NLR was higher in cancer patients, although the lymphocyte counts in NSCLC patients compared to the control group were only marginally lower. A mild blood granulocytosis is common in cancer patients, as tumors are known to drive myelopoiesis 15. Cancer‐associated myeloproliferation is not merely a paraneoplastic phenomenon of questionable importance, but leads to the suppression of host immunity and promotion of tumor angiogenesis. A phenotypical diversity and plasticity of circulating neutrophils has to be taken into account. Some granulocytes might exhibit an immunosuppressive phenotype; for example, CD14^–/CD15⁺ arginase‐producing MDSC have been reported as a granulocytic subpopulation in peripheral blood of NSCLC 16. Neutrophilia is regarded as a poor independent prognostic factor in tumor disease; e.g. Kasuga described leukocytosis to be linked to poor prognosis in NSCLC 17. Even within the NSCLC tumor tissue, where neutrophils account for approximately 20% of CD45⁺ cells 18, granulocytes have an immunosuppressive role, and a negative correlation between neutrophils and tumor‐infiltrating T lymphocytes has been described 18. Gentles and co‐authors identified neutrophil transcripts in the tumor as the strongest predictor of mortality in NSCLC patients 19. Despite higher neutrophil counts in our tumor group, a significant association of blood leukocytosis or NLR with overall survival could not be confirmed. However, it is noteworthy that blood neutrophil counts may be affected by various factors, such as smoking behavior and systemic infection, resulting in considerable variability, especially in a small group of patients. In contrast to our data, Shimizu et al. found that NSCLC patients with an NLR ≥ 2.5 had a significantly poorer survival outcome 20, which was confirmed in a meta‐analysis demonstrating elevated pretreatment NLR values to predict poorer OS in patients with NSCLC 13.

Smoking is the leading cause of lung cancer 2, and 79% of the NSCLC patients investigated in this study had a smoking history or were still active smokers. The impact of tobacco smoke on lung homeostasis is complex, with one of the predominant features being suppression of immune cells 21. In our study, when compared to never‐smokers, smokers not only had higher blood neutrophil counts, but also higher monocyte values and a higher percentage of Treg and HLA‐DR^low MDSC. Interestingly, even years after having quit smoking, Treg and HLA‐DR^low MDSC remained increased. However, smoking had no significant impact on the NLR, which is in contrast to data of Tulgar and co‐authors showing an increased NLR in smokers in correlation with pack/year 22. This discrepancy might be explained by the patient cohort analyzed, as we focused only on tumor diagnosis resulting in a more heterogeneous patient group, whereas Tulgar et al. excluded people with several chronic diseases known to affect the NLR 22.

An increase of HLA‐DR^low monocytes has been described in several tumor types (for review see 23). It is suggested that these cells suppress T cell function in cancer patients, as already described for HLA‐DR^low monocytes in sepsis 24 and major trauma 25. HLA‐DR^low monocytes suppress NK cell functions in patients with hepatocellular carcinoma inhibiting autologous NK cell cytotoxicity and cytokine secretion in co‐culture 26.

For the quantification of HLA‐DR^low monocytes, the QuantiBRITE^TM system (BD Biosciences) with multi‐level calibration beads and an HLA‐DR‐specific antibody with a 1/1 fluorochrome‐to‐protein ratio was used, an approach to reduce variability leading to highly reproducible results among cytometers and institutions 12. Using the geometric mean representing 5000 ABC as borderline value for a low monocytes HLA‐DR intensity, the age‐matched control group had 2·3% HLA‐DR^low monocytes and never‐smokers in the control group only 1.6% of total monocytes. In contrast, NSCLC patients had 5·3 ± 4·1% of these monocytes, which is comparable with earlier results in renal cell cancer patients showing 7·5 ± 7·4% HLA‐DR^low monocytes in blood 27. It is noteworthy that in the age‐matched control group individual people showed a somewhat high amount of HLA‐DR^low monocytes. Although the existence of an unnoticed tumor cannot be excluded, circulating MDSC are not tumor‐specific and have been described in several other pathological conditions, such as viral infections 28, various chronic infections 29 or dilated cardiomyopathy 30. These reports suggest that HLA‐DR^low monocytes might be part of a conserved response to different endogenous and exogenous stress signals, including inflammation 31.

Although several authors described HLA‐DR^low monocytes in lung cancer patients, HLA‐DR expression was not quantified, which hampers the comparability of data. Furthermore, absolute numbers of monocytes are rarely shown in the literature because blood cells are used after serum withdrawal or after density gradient centrifugation. In addition, the impact of smoking was seldom considered. A high variability in the reported frequency of HLA‐DR^low MDSC in blood is striking: Vetsika and co‐authors reported 25·2 ± 2% of HLA‐DR^low monocytes without any difference between NSCLC patients and controls 32, Huang and co‐authors found 9·4% HLA‐DR^low monocytes in the blood of 89 NSCLC patients 33 and Chen and co‐authors reported 7·7% HLA‐DR^low monocytes in patients with squamous cell carcinoma 34. The latter value matches well with the 8·1% HLA‐DR^low MDSC of our small patient subgroup with squamous cell carcinoma. MDSC percentages correlated positively with the number of neutrophils and negatively with pDC counts in our analysis. The percentage of HLA‐DR^low monocytes has clinical relevance, as a higher frequency of this immune subpopulation was associated with shorter OS of NSCLC patients. Corresponding to our data, Huang et al. reported a shorter progression‐free survival in patients with a higher percentage of HLA‐DR^low MDSC 33. The authors demonstrated that HLA‐DR^low monocytes inhibit autologous T cell proliferation and interferon (IFN)‐γ production in a cell–cell contact‐dependent manner 33. Increased percentages of monocytes MDSC have been associated with a worse response to treatment in NSCLC, patients confirming their value as a biomarker 32.

Most tumors emerge in aging populations at a time when T cell function is declining. In our investigations, patients aged > 70 years had a lower number of lymphocytes, especially of CD4⁺ T cells. Cancer patients had lower counts of naive T cells compared to the control group, which dropped with age. Whereas CD8⁺ T cell numbers were highly variable, without any difference for the tumor group and with no age‐dependency, naive CD8⁺ T cells, similar to naive CD4⁺ T cells, were significantly lower in tumor disease and dropped with age. In addition to the decline in numbers, functional defects have been described in NSCLC patients, e.g. naive T cells show reduced IFN‐γ and tumour necrosis factor (TNF)‐α production after mitogenic stimulation 35 or a shorter telomere length 36. Nevertheless, higher numbers of both naive CD4⁺ and CD8⁺ T cells correlated with a better survival in our analyses, suggesting that the number of naive T cells is an interesting biomarker. Our results confirm the data of Yang and co‐authors, demonstrating the naive CD4⁺ and CD8⁺ T cells as predictors for survival 37. However, these authors defined ‘naive’ simply as CD45RA⁺, whereas we described CD45RA⁺CCR7⁺ cells as naive T cells, as proposed by Sallusto 38. In our investigations, neither numbers of NK cells nor CD8⁺ T cells correlated with patients’ OS; however, NSCLC patients had lower NK cell numbers in blood compared to the controls, and interestingly, female patients had lower numbers compared to males. NSCLC patients had a higher percentage of CD38^–HLA‐DR⁺CD4⁺ T cells, which have a clear bias towards secretion of the T helper type 1 (Th1)‐associated cytokines IFN‐γ and TNF 39, but data on their function in cancer patients are still lacking. The percentage of immunosenescence‐associated CD57⁺CD8⁺ T cells did not differ between the control and tumor group, but cancer patients had a higher proportion of CD57⁺CD4⁺T cells. The most differentiated CD4⁺ T cell subset, the CD27^–CD28^– terminally differentiated effector memory (TEMRA) cells, express CD57 40. As tumor‐infiltrating CD8⁺CD57⁺ T cells with functional properties of both early effector memory cells and terminally differentiated effector cells have been described 41, the role of CD4⁺CD57⁺ T cells in NSCLC has still to be elucidated.

Several studies have demonstrated that NSCLC patients show increased proportions of Treg in the peripheral blood 42, 43, although multiple markers are used for their identification in different studies, and absolute Treg numbers are rarely published. Elevated Treg numbers with their highly suppressive function help tumors to avoid a productive immune surveillance 44, and represent major hurdles towards successful immunotherapy. In contrast to data from literature, we could not find significant differences in the absolute values of CD25⁺⁺CD127^–/dimCD4⁺ Treg between the control and tumor groups. The higher percentages of Treg in cancer patients were caused by smoking behavior. However, lower Treg percentages predicted better survival in our patient cohort, which is in line with other reports 45. This Treg‐linked adverse prognosis could be a result of lymphocytic suppression. Indeed, a negative correlation of Treg percentages with the number of CD4⁺ T cells as well as of naive CD4⁺ T cells could be found. Additionally, we observed a negative correlation of naive CD4⁺ T cells with the percentage of HLA‐DR^low MDSC. MDSC suppress T cells via sequestration of cysteine and via down‐regulation of L‐selectin, which impairs homing of naive T cells 46. Thus, decreased numbers of naive T cells in cancer patients could be the result of an impact of both suppressive Treg and HLA‐DR^low MDSC.

In line with earlier studies with renal cancer patients 27, NSCLC patients had lower numbers of blood DC, especially of pDC. We confirmed data of others 47, that age had a significant impact on circulating DC numbers. Whereas in melanoma patients low levels of circulating pDC indicate a worse prognosis 48, the numbers of pDC did not correlate with OS in our analysis. However, patients with a lower ratio HLA‐DR^low MDSC/pDC had a longer median survival in Kaplan–Meier analysis.

In summary, peculiarities in peripheral blood immune cell populations in NSCLC patients were analyzed by flow cytometry. Among the parameters useful for an immune monitoring are HLA‐DR^low MDSC, which correlate positively with the neutrophil count and negatively with the number of pDC and naïve CD4⁺ T cells. Our data provide the first clues as to why comorbidity in lung cancer has been found associated, for example, with older age, smoking behavior and female gender 49. Thus, in older age, for example, the number of naive T cells and DC is decreased. Otherwise, smokers have higher percentages of Treg and HLA‐DR^low MDSC. Lower NK cell numbers were observed in women. Of prognostic value are naive T cell counts, the percentages of Treg and HLA‐DR^low monocytes, as well as the ratio HLA‐DR^low MDSC/pDC. These markers should be investigated with higher patient numbers in future studies.

Grant support

This work was supported by the GRK PromoAGE (B. S.).

Disclosures

All authors have no potential conflicts of interest.

Supporting information

Fig. S1. Gating strategy for DC and monocytes. Ca. 1x10E6 mononuclear cells (MNC) were measured (storage of all cells). After doublet discrimination (R1) a gate on antibody‐negative MNC (R2) was used to exclude B and T cells (R3) and CD56+ NK cells (R4). Monocytes were defined as CD14+ cells and subdivided into CD16neg classical and CD16+ non‐classical cells. DC were defined as CD14neg cells (R5) with HLA‐DR expression (R6) and subdivided into CD11cneg CD123+ pDC and CD11c+ mDC.

Click here for additional data file.^{(7.6MB, tif)}

Fig. S2. Gating strategy and calculation for the estimation of HLA‐DRlow monocytes using BD QuantiBRITE™ reagents. Upper area: 10 000 events of the BD QuantiBRITE™ beads were collected. A gate was set around the bead population (R1). In a PE fluorescence histogram (geometric mean) markers were adapted to the four bead peaks. After log10 calculation, the lot‐specific PE values of the beads were plotted against the geometric mean values to perform a linear regression analysis using the equation y = mx + c. Lower area: Using the same instrument settings of fluorescences, approximately 50 000 leukocytes of the stained blood samples were measured to analyse >3000 monocytes. After doublet discrimination (R1) a gate on monocytes (R2) was defined in a plot CD14 against side scatter SSC. The geometric mean value of HLA‐DR‐PE fluorescence of the whole monocytic population (P1) was estimated in a PE histogram. After log10 calculation the HLA‐DR‐PE mean value was converted to the term PE molecules/cell using the linear regression curve (A). Additionally, the regression curve was used to determine the PE fluorescence channel that corresponds to 5000 PE molecules/cell. In the PE histogram, P2 illustrates the percentage of monocytes ≤5000 PE molecules/cell corresponding to HLA‐DRlow monocytes (B).

Click here for additional data file.^{(6.1MB, tif)}

Table S1. Antibodies and reagents used for flow cytometry.

Click here for additional data file.^{(15.3KB, docx)}

Table S2. Relationship between blood immune cell parameters with patient's survival. Data of univariate (Kaplan‐Meier) and multivariate (Cox regression) prognostic factor analysis are shown.

Click here for additional data file.^{(21.1KB, docx)}

Table S3. The effect of smoking status (with never smoker, former smoker >6months, and current smoker together with former smoker <6months) on blood immune cells. Data were given as mean±SD. Significant differences between the 3 groups of smoking status are indicated.

Click here for additional data file.^{(15.6KB, docx)}

References

1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 2015;65:5–29. [DOI] [PubMed] [Google Scholar]
2. Gibson GJ, Loddenkemper R, Lundback B, Sibille Y. Respiratory health and disease in Europe: the new European Lung White Book. Eur Respir J 2013;42:559–563. [DOI] [PubMed] [Google Scholar]
3. Travis WD, Brambilla E, Noguchi M et al International association for the study of lung cancer/american thoracic society/european respiratory society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol 2011;6:244–285. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Abbasi S, Badheeb A. Prognostic factors in advanced non‐small‐cell lung cancer patients: patient characteristics and type of chemotherapy. Lung Cancer Int 2011;2011:152125. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Miller KD, Siegel RL, Lin CC et al Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin 2016;66:271–289. [DOI] [PubMed] [Google Scholar]
6. Scagliotti GV, Bironzo P, Vansteenkiste JF. Addressing the unmet need in lung cancer: the potential of immuno‐oncology. Cancer Treat Rev 2015;41:465–475. [DOI] [PubMed] [Google Scholar]
7. Srivastava MK, Bosch JJ, Wilson AL, Edelman MJ, Ostrand‐Rosenberg S. MHC II lung cancer vaccines prime and boost tumor‐specific CD4+ T cells that cross‐react with multiple histologic subtypes of nonsmall cell lung cancer cells. Int J Cancer 2010;127:2612–2621. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Aerts JG, Lievense LA, Hoogsteden HC, Hegmans JP. Immunotherapy prospects in the treatment of lung cancer and mesothelioma. Transl Lung Cancer Res 2014;3:34–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Showe MK, Kossenkov AV, Showe LC. The peripheral immune response and lung cancer prognosis. Oncoimmunology 2012;1:1414–1416. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Gustafson MP, Lin Y, LaPlant B et al Immune monitoring using the predictive power of immune profiles. J Immunother Cancer 2013;1:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Liu W, Putnam AL, Xu‐Yu Z et al CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ Treg cells. J Exp Med 2006;203:1701–1711. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Docke WD, Hoflich C, Davis KA et al Monitoring temporary immunodepression by flow cytometric measurement of monocytic HLA‐DR expression: a multicenter standardized study. Clin Chem 2005;51:2341–2347. [DOI] [PubMed] [Google Scholar]
13. Gu XB, Tian T, Tian XJ, Zhang XJ. Prognostic significance of neutrophil‐to‐lymphocyte ratio in non‐small cell lung cancer: a meta‐analysis. Sci Rep 2015;5:12493. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Najjar YG, Finke JH. Clinical perspectives on targeting of myeloid derived suppressor cells in the treatment of cancer. Front Oncol 2013;3:49. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Wilcox RA. Cancer‐associated myeloproliferation: old association, new therapeutic target. Mayo Clin Proc 2010;85:656–663. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Liu CY, Wang YM, Wang CL et al Population alterations of L‐arginase‐ and inducible nitric oxide synthase‐expressed CD11b+/CD14(–)/CD15+/CD33+ myeloid‐derived suppressor cells and CD8+ T lymphocytes in patients with advanced‐stage non‐small cell lung cancer. J Cancer Res Clin Oncol 2010;136:35–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Kasuga I, Makino S, Kiyokawa H, Katoh H, Ebihara Y, Ohyashiki K. Tumor‐related leukocytosis is linked with poor prognosis in patients with lung carcinoma. Cancer 2001;92:2399–2405. [DOI] [PubMed] [Google Scholar]
18. Kargl J, Busch SE, Yang GH et al Neutrophils dominate the immune cell composition in non‐small cell lung cancer. Nat Commun 2017;8:14381. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Gentles AJ, Newman AM, Liu CL et al The prognostic landscape of genes and infiltrating immune cells across human cancers. Nat Med 2015;21:938–945. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Shimizu K, Okita R, Saisho S, Maeda A, Nojima Y, Nakata M. Preoperative neutrophil/lymphocyte ratio and prognostic nutritional index predict survival in patients with non‐small cell lung cancer. World J Surg Oncol 2015;13:291. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Goncalves RB, Coletta RD, Silverio KG et al Impact of smoking on inflammation: overview of molecular mechanisms. Inflamm Res 2011;60:409–424. [DOI] [PubMed] [Google Scholar]
22. Tulgar YK, Cakar S, Tulgar S, Dalkilic O, Cakiroglu B, Uyanik BS. The effect of smoking on neutrophil/lymphocyte and platelet/lymphocyte ratio and platelet indices: a retrospective study. Eur Rev Med Pharmacol Sci 2016;20:3112–3118. [PubMed] [Google Scholar]
23. Greten TF, Manns MP, Korangy F. Myeloid derived suppressor cells in human diseases. Int Immunopharmacol 2011;11:802–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Docke WD, Randow F, Syrbe U et al Monocyte deactivation in septic patients: restoration by IFN‐gamma treatment. Nat Med 1997;3:678–681. [DOI] [PubMed] [Google Scholar]
25. Hershman MJ, Cheadle WG, Wellhausen SR, Davidson PF, Polk HC Jr. Monocyte HLA‐DR antigen expression characterizes clinical outcome in the trauma patient. Br J Surg 1990;77:204–207. [DOI] [PubMed] [Google Scholar]
26. Hoechst B, Voigtlaender T, Ormandy L et al Myeloid derived suppressor cells inhibit natural killer cells in patients with hepatocellular carcinoma via the NKp30 receptor. Hepatology 2009;50:799–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Hase S, Weinitschke K, Fischer K et al Monitoring peri‐operative immune suppression in renal cancer patients. Oncol Rep 2011;25:1455–1464. [DOI] [PubMed] [Google Scholar]
28. Pang X, Song H, Zhang Q, Tu Z, Niu J. Hepatitis C virus regulates the production of monocytic myeloid‐derived suppressor cells from peripheral blood mononuclear cells through PI3K pathway and autocrine signaling. Clin Immunol 2016;164:57–64. [DOI] [PubMed] [Google Scholar]
29. Nagaraj S, Youn JI, Gabrilovich DI. Reciprocal relationship between myeloid‐derived suppressor cells and T cells. J Immunol 2013;191:17–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Zhang WC, Wang YG, Wei WH et al Activated circulating myeloid‐derived suppressor cells in patients with dilated cardiomyopathy. Cell Physiol Biochem 2016;38:2438–2451. [DOI] [PubMed] [Google Scholar]
31. Bronte V. Myeloid‐derived suppressor cells in inflammation: uncovering cell subsets with enhanced immunosuppressive functions. Eur J Immunol 2009;39:2670–2672. [DOI] [PubMed] [Google Scholar]
32. Vetsika EK, Koinis F, Gioulbasani M et al A circulating subpopulation of monocytic myeloid‐derived suppressor cells as an independent prognostic/predictive factor in untreated non‐small lung cancer patients. J Immunol Res 2014;2014:659294. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Huang A, Zhang B, Wang B, Zhang F, Fan KX, Guo YJ. Increased CD14(+)HLA‐DR (–/low) myeloid‐derived suppressor cells correlate with extrathoracic metastasis and poor response to chemotherapy in non‐small cell lung cancer patients. Cancer Immunol Immunother 2013;62:1439–1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Chen Y, Pan G, Tian D, Zhang Y, Li T. Functional analysis of CD14+HLA‐DR‐/low myeloid‐derived suppressor cells in patients with lung squamous cell carcinoma. Oncol Lett 2017;14:349–354. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Sheng SY, Gu Y, Lu CG, Zou JY, Hong H, Wang R. The distribution and function of human memory T cell subsets in lung cancer. Immunol Res 2017;65:639–650. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Qian Y, Ding T, Wei L, Cao S, Yang L. Shorter telomere length of T‐cells in peripheral blood of patients with lung cancer. Onco Targets Ther 2016;9:2675–2682. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Yang P, Ma J, Yang X, Li W. Peripheral CD4+ naive/memory ratio is an independent predictor of survival in non‐small cell lung cancer. Oncotarget 2017;8:83650–83659. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 1999;401:708–712. [DOI] [PubMed] [Google Scholar]
39. Scalzo‐Inguanti K, Plebanski M. CD38 identifies a hypo‐proliferative IL‐13‐secreting CD4+ T‐cell subset that does not fit into existing naive and memory phenotype paradigms. Eur J Immunol 2011;41:1298–1308. [DOI] [PubMed] [Google Scholar]
40. Koch S, Larbi A, Derhovanessian E, Ozcelik D, Naumova E, Pawelec G. Multiparameter flow cytometric analysis of CD4 and CD8 T cell subsets in young and old people. Immun Ageing 2008;5:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Wu RC, Hwu P, Radvanyi LG. New insights on the role of CD8(+)CD57(+) T‐cells in cancer. Oncoimmunology 2012;1:954–956. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Li L, Chao QG, Ping LZ et al The prevalence of FOXP3+ regulatory T‐cells in peripheral blood of patients with NSCLC. Cancer Biother Radiopharm 2009;24:357–367. [DOI] [PubMed] [Google Scholar]
43. Erfani N, Mehrabadi SM, Ghayumi MA et al Increase of regulatory T cells in metastatic stage and CTLA‐4 over expression in lymphocytes of patients with non‐small cell lung cancer (NSCLC). Lung Cancer 2012;77:306–311. [DOI] [PubMed] [Google Scholar]
44. Facciabene A, Santoro S, Coukos G. Know thy enemy: Why are tumor‐infiltrating regulatory T cells so deleterious? Oncoimmunology 2012;1:575–577. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Kotsakis A, Koinis F, Katsarou A et al Prognostic value of circulating regulatory T cell subsets in untreated non‐small cell lung cancer patients. Sci Rep 2016;6:39247. [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Ostrand‐Rosenberg S. Myeloid‐derived suppressor cells: more mechanisms for inhibiting antitumor immunity. Cancer Immunol Immunother 2010;59:1593–1600. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Agrawal A, Agrawal S, Tay J, Gupta S. Biology of dendritic cells in aging. J Clin Immunol 2008;28:14–20. [DOI] [PubMed] [Google Scholar]
48. Chevolet I, Speeckaert R, Schreuer M et al Clinical significance of plasmacytoid dendritic cells and myeloid‐derived suppressor cells in melanoma. J Transl Med 2015;13:9. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Tammemagi CM, Neslund‐Dudas C, Simoff M, Kvale P. In lung cancer patients, age, race–ethnicity, gender and smoking predict adverse comorbidity, which in turn predicts treatment and survival. J Clin Epidemiol 2004;57:597–609. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Click here for additional data file.^{(7.6MB, tif)}

Click here for additional data file.^{(6.1MB, tif)}

Table S1. Antibodies and reagents used for flow cytometry.

Click here for additional data file.^{(15.3KB, docx)}

Table S2. Relationship between blood immune cell parameters with patient's survival. Data of univariate (Kaplan‐Meier) and multivariate (Cox regression) prognostic factor analysis are shown.

Click here for additional data file.^{(21.1KB, docx)}

Click here for additional data file.^{(15.6KB, docx)}

[cei13219-bib-0001] 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 2015;65:5–29. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0002] 2. Gibson GJ, Loddenkemper R, Lundback B, Sibille Y. Respiratory health and disease in Europe: the new European Lung White Book. Eur Respir J 2013;42:559–563. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0003] 3. Travis WD, Brambilla E, Noguchi M et al International association for the study of lung cancer/american thoracic society/european respiratory society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol 2011;6:244–285. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0004] 4. Abbasi S, Badheeb A. Prognostic factors in advanced non‐small‐cell lung cancer patients: patient characteristics and type of chemotherapy. Lung Cancer Int 2011;2011:152125. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0005] 5. Miller KD, Siegel RL, Lin CC et al Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin 2016;66:271–289. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0006] 6. Scagliotti GV, Bironzo P, Vansteenkiste JF. Addressing the unmet need in lung cancer: the potential of immuno‐oncology. Cancer Treat Rev 2015;41:465–475. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0007] 7. Srivastava MK, Bosch JJ, Wilson AL, Edelman MJ, Ostrand‐Rosenberg S. MHC II lung cancer vaccines prime and boost tumor‐specific CD4+ T cells that cross‐react with multiple histologic subtypes of nonsmall cell lung cancer cells. Int J Cancer 2010;127:2612–2621. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0008] 8. Aerts JG, Lievense LA, Hoogsteden HC, Hegmans JP. Immunotherapy prospects in the treatment of lung cancer and mesothelioma. Transl Lung Cancer Res 2014;3:34–45. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0009] 9. Showe MK, Kossenkov AV, Showe LC. The peripheral immune response and lung cancer prognosis. Oncoimmunology 2012;1:1414–1416. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0010] 10. Gustafson MP, Lin Y, LaPlant B et al Immune monitoring using the predictive power of immune profiles. J Immunother Cancer 2013;1:7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0011] 11. Liu W, Putnam AL, Xu‐Yu Z et al CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ Treg cells. J Exp Med 2006;203:1701–1711. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0012] 12. Docke WD, Hoflich C, Davis KA et al Monitoring temporary immunodepression by flow cytometric measurement of monocytic HLA‐DR expression: a multicenter standardized study. Clin Chem 2005;51:2341–2347. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0013] 13. Gu XB, Tian T, Tian XJ, Zhang XJ. Prognostic significance of neutrophil‐to‐lymphocyte ratio in non‐small cell lung cancer: a meta‐analysis. Sci Rep 2015;5:12493. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0014] 14. Najjar YG, Finke JH. Clinical perspectives on targeting of myeloid derived suppressor cells in the treatment of cancer. Front Oncol 2013;3:49. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0015] 15. Wilcox RA. Cancer‐associated myeloproliferation: old association, new therapeutic target. Mayo Clin Proc 2010;85:656–663. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0016] 16. Liu CY, Wang YM, Wang CL et al Population alterations of L‐arginase‐ and inducible nitric oxide synthase‐expressed CD11b+/CD14(–)/CD15+/CD33+ myeloid‐derived suppressor cells and CD8+ T lymphocytes in patients with advanced‐stage non‐small cell lung cancer. J Cancer Res Clin Oncol 2010;136:35–45. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0017] 17. Kasuga I, Makino S, Kiyokawa H, Katoh H, Ebihara Y, Ohyashiki K. Tumor‐related leukocytosis is linked with poor prognosis in patients with lung carcinoma. Cancer 2001;92:2399–2405. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0018] 18. Kargl J, Busch SE, Yang GH et al Neutrophils dominate the immune cell composition in non‐small cell lung cancer. Nat Commun 2017;8:14381. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0019] 19. Gentles AJ, Newman AM, Liu CL et al The prognostic landscape of genes and infiltrating immune cells across human cancers. Nat Med 2015;21:938–945. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0020] 20. Shimizu K, Okita R, Saisho S, Maeda A, Nojima Y, Nakata M. Preoperative neutrophil/lymphocyte ratio and prognostic nutritional index predict survival in patients with non‐small cell lung cancer. World J Surg Oncol 2015;13:291. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0021] 21. Goncalves RB, Coletta RD, Silverio KG et al Impact of smoking on inflammation: overview of molecular mechanisms. Inflamm Res 2011;60:409–424. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0022] 22. Tulgar YK, Cakar S, Tulgar S, Dalkilic O, Cakiroglu B, Uyanik BS. The effect of smoking on neutrophil/lymphocyte and platelet/lymphocyte ratio and platelet indices: a retrospective study. Eur Rev Med Pharmacol Sci 2016;20:3112–3118. [PubMed] [Google Scholar]

[cei13219-bib-0023] 23. Greten TF, Manns MP, Korangy F. Myeloid derived suppressor cells in human diseases. Int Immunopharmacol 2011;11:802–807. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0024] 24. Docke WD, Randow F, Syrbe U et al Monocyte deactivation in septic patients: restoration by IFN‐gamma treatment. Nat Med 1997;3:678–681. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0025] 25. Hershman MJ, Cheadle WG, Wellhausen SR, Davidson PF, Polk HC Jr. Monocyte HLA‐DR antigen expression characterizes clinical outcome in the trauma patient. Br J Surg 1990;77:204–207. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0026] 26. Hoechst B, Voigtlaender T, Ormandy L et al Myeloid derived suppressor cells inhibit natural killer cells in patients with hepatocellular carcinoma via the NKp30 receptor. Hepatology 2009;50:799–807. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0027] 27. Hase S, Weinitschke K, Fischer K et al Monitoring peri‐operative immune suppression in renal cancer patients. Oncol Rep 2011;25:1455–1464. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0028] 28. Pang X, Song H, Zhang Q, Tu Z, Niu J. Hepatitis C virus regulates the production of monocytic myeloid‐derived suppressor cells from peripheral blood mononuclear cells through PI3K pathway and autocrine signaling. Clin Immunol 2016;164:57–64. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0029] 29. Nagaraj S, Youn JI, Gabrilovich DI. Reciprocal relationship between myeloid‐derived suppressor cells and T cells. J Immunol 2013;191:17–23. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0030] 30. Zhang WC, Wang YG, Wei WH et al Activated circulating myeloid‐derived suppressor cells in patients with dilated cardiomyopathy. Cell Physiol Biochem 2016;38:2438–2451. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0031] 31. Bronte V. Myeloid‐derived suppressor cells in inflammation: uncovering cell subsets with enhanced immunosuppressive functions. Eur J Immunol 2009;39:2670–2672. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0032] 32. Vetsika EK, Koinis F, Gioulbasani M et al A circulating subpopulation of monocytic myeloid‐derived suppressor cells as an independent prognostic/predictive factor in untreated non‐small lung cancer patients. J Immunol Res 2014;2014:659294. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0033] 33. Huang A, Zhang B, Wang B, Zhang F, Fan KX, Guo YJ. Increased CD14(+)HLA‐DR (–/low) myeloid‐derived suppressor cells correlate with extrathoracic metastasis and poor response to chemotherapy in non‐small cell lung cancer patients. Cancer Immunol Immunother 2013;62:1439–1451. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0034] 34. Chen Y, Pan G, Tian D, Zhang Y, Li T. Functional analysis of CD14+HLA‐DR‐/low myeloid‐derived suppressor cells in patients with lung squamous cell carcinoma. Oncol Lett 2017;14:349–354. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0035] 35. Sheng SY, Gu Y, Lu CG, Zou JY, Hong H, Wang R. The distribution and function of human memory T cell subsets in lung cancer. Immunol Res 2017;65:639–650. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0036] 36. Qian Y, Ding T, Wei L, Cao S, Yang L. Shorter telomere length of T‐cells in peripheral blood of patients with lung cancer. Onco Targets Ther 2016;9:2675–2682. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0037] 37. Yang P, Ma J, Yang X, Li W. Peripheral CD4+ naive/memory ratio is an independent predictor of survival in non‐small cell lung cancer. Oncotarget 2017;8:83650–83659. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0038] 38. Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 1999;401:708–712. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0039] 39. Scalzo‐Inguanti K, Plebanski M. CD38 identifies a hypo‐proliferative IL‐13‐secreting CD4+ T‐cell subset that does not fit into existing naive and memory phenotype paradigms. Eur J Immunol 2011;41:1298–1308. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0040] 40. Koch S, Larbi A, Derhovanessian E, Ozcelik D, Naumova E, Pawelec G. Multiparameter flow cytometric analysis of CD4 and CD8 T cell subsets in young and old people. Immun Ageing 2008;5:6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0041] 41. Wu RC, Hwu P, Radvanyi LG. New insights on the role of CD8(+)CD57(+) T‐cells in cancer. Oncoimmunology 2012;1:954–956. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0042] 42. Li L, Chao QG, Ping LZ et al The prevalence of FOXP3+ regulatory T‐cells in peripheral blood of patients with NSCLC. Cancer Biother Radiopharm 2009;24:357–367. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0043] 43. Erfani N, Mehrabadi SM, Ghayumi MA et al Increase of regulatory T cells in metastatic stage and CTLA‐4 over expression in lymphocytes of patients with non‐small cell lung cancer (NSCLC). Lung Cancer 2012;77:306–311. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0044] 44. Facciabene A, Santoro S, Coukos G. Know thy enemy: Why are tumor‐infiltrating regulatory T cells so deleterious? Oncoimmunology 2012;1:575–577. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0045] 45. Kotsakis A, Koinis F, Katsarou A et al Prognostic value of circulating regulatory T cell subsets in untreated non‐small cell lung cancer patients. Sci Rep 2016;6:39247. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0046] 46. Ostrand‐Rosenberg S. Myeloid‐derived suppressor cells: more mechanisms for inhibiting antitumor immunity. Cancer Immunol Immunother 2010;59:1593–1600. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0047] 47. Agrawal A, Agrawal S, Tay J, Gupta S. Biology of dendritic cells in aging. J Clin Immunol 2008;28:14–20. [DOI] [PubMed] [Google Scholar]

[cei13219-bib-0048] 48. Chevolet I, Speeckaert R, Schreuer M et al Clinical significance of plasmacytoid dendritic cells and myeloid‐derived suppressor cells in melanoma. J Transl Med 2015;13:9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cei13219-bib-0049] 49. Tammemagi CM, Neslund‐Dudas C, Simoff M, Kvale P. In lung cancer patients, age, race–ethnicity, gender and smoking predict adverse comorbidity, which in turn predicts treatment and survival. J Clin Epidemiol 2004;57:597–609. [DOI] [PubMed] [Google Scholar]

PERMALINK

Blood immune cell biomarkers in lung cancer

D Riemann

M Cwikowski

S Turzer

T Giese

M Grallert

W Schütte

B Seliger

Summary

Introduction