The influence of anti-cancer therapies on lymphocyte subpopulations of lung cancer patients

Introduction There are limited data on the influence of different anti-cancer therapies on lymphocyte subpopulations and their relationships to survival of non-small cell lung cancer (NSCLC) patients. This study aimed to assess the effect of immunotherapy, chemotherapy, immunochemotherapy, adjuvant chemotherapy after surgery, and antibodies against Vascular Endothelial Growth Factors (VEGF) on B cell, T cell, and NK cell subpopulations, and the survival time of NSCLC patients. Methods A total of 32 consecutive NSCLC patients were recruited at Pulmonology Clinic, Leipzig from January 2018 to March 2020 and enrolled in this study. Immunophenotyping was done using a FACS Canto II flow cytometer (BD Biosciences) before the administration of the planned therapy and during therapy with up to 7 observational windows for each patient targeting 130 immunologic parameters. Results Absolute transitional B cells was significantly increased after immunotherapy (p = 0.032), immunochemotherapy (p = 0.030), and antibodies against VEGF (p = 0.024). Similarly, absolute counts and percentage of B cells were significantly increased after adjuvant chemotherapy (p = 0.023). However, absolute counts and percentage of transitional B cells are significantly decreased after chemotherapy (p = 0.001). Activated cytotoxic T cells were significantly increased after immunotherapy (p = 0.031) and immunochemotherapy (p = 0.030). The overall survival rate of NSCLC patients was 31%. Conclusions In conclusion, this study suggests that different types of anti-cancer therapies affect lymphocyte subpopulations of NSCLC patients. Further large-scale and multicentre studies are required to confirm our results and to evaluate the prognostic value of lymphocyte subpopulations.


Introduction
Lung cancer is one of the most frequent malignancies with 2.2 million new cases a year and the most common cause of cancer death in the world (1).Non-small cell lung cancer (NSCLC) is about 80-85% of all lung cancer cases (2).Majority (> 55%) of NSCLC patients are diagnosed late at the advanced stages of the disease (3).Currently, the treatment options for NSCLC are systemic therapies such as chemotherapies, drugs targeting commonly mutated pathways in lung cancer, and immune checkpoint inhibitors, surgical resection of the primary tumor or metastatic lesion and radiation therapy (4).The management and therapy of patients with advanced NSCLC have changed markedly in the last few years.Early detection methods and therapeutic options have been improved tremendously.However, the overall survival rate of patients with advanced NSCLC did not improve much and is still dismal (5).The reported 5-year survival rate of NSCLC patients was 17.8%, which is one of the highest fatality rates in non-communicable diseases (6).Reliable markers of treatment response and survival outcomes are urgently needed to enable early adaptation of treatment strategies in advanced NSCLC patients.Tumorigenesis and programmed cell death ligand 1 (PD-L1) status became standard markers influencing therapy regimes (7).Expression of PD-L1 has shown a significant predictive role (8).Other biomarkers such as lung immune prognostic index (LIPI) and tumor mutation burden (TMB) have revealed inconsistent results (9).Several studies have also demonstrated that lung cancer patients have lower levels of CD4+/CD8+ ratio, CD4+T cells, NK cell levels, and higher regulatory T cells (Tregs) than healthy subjects (10,11).Patient-related factors such as age and comorbidities also affect peripheral blood lymphocyte subpopulations and treatment outcomes of NSCLC patients (12).Anticancer therapies, tumor surgery and wound healing may also have an impact on the immunophenotyping of lymphocyte subpopulations.So far, there are limited data concerning changes or developmental shifts in the most relevant B cell, T cell, and NK cell subpopulations caused by different anti-cancer therapies.Understanding the potential changes in lymphocyte subpopulations during different therapies is crucial to understand the effect of different therapies on the immune system and to adapt effective therapy regimens for better management of NSCLC patients.Therefore, in this study, we investigated the influence of immunotherapy, chemotherapy, immunochemotherapy, adjuvant chemotherapy after surgery, and antibodies against Vascular Endothelial Growth Factors (VEGF) on B cell, T cell, and NK cell subpopulations.Additionally, the relationship between therapy regimens and survival was assessed.

Study design, population, and period
A longitudinal cohort study was performed on 32 consecutive NSCLC patients who visited the Pulmonology Clinic, Leipzig from January 10, 2018 to March 3, 2020.Socio-demographic information such as age, sex, smoking status, and clinical information such as type of cancer and the stage of cancer were collected from patients using a standardized questionnaire.Active smokers were defined as patients actively smoking at the time of data collection or up to 6 months before the data collection.To be qualified as former smoker, it was required that the patient quit smoking more than 6 months prior to data collection.Immunophenotyping was done before the administration of the planned therapy and during therapy with up to 7 observational windows (3 to 4 weeks each window) in the laboratory of the Institute of Clinical Immunology, University Hospital Leipzig resulting in a total of 135 observations with 130 immunologic parameters assessed.The median follow-up period for the study participants included in this study was 1,220 days.

Ethical considerations
Ethical approval was made by the Medical Faculty of the University of Leipzig ethics committee (Ref.No: 201/18-ek).The sociodemographic information, clinical data, and venous blood were collected from NSCLC patients following the ethical considerations recommended by the Declaration of Helsinki of 1975, revised in 2013.Data identifying study participants were removed prior to analyses.Safeguards for appropriate and ethical use of data were in place.Informed consent was acquired from all NSCLC patients included in the study.

Blood collection
Ethylenediaminetetraacetic acid (EDTA) venous blood was collected from each patient in the pulmonology clinic before the administration of the planned therapy at the time of treatment start and during observational windows.EDTA blood specimen was immediately brought to the laboratory of the Institute of Clinical Immunology for analysis of the cellular immune status.Blood samples were processed and analyzed within 3 hours of collection.

Cell surface markers staining
The staining and flow cytometry analysis of B cell, T cell, and NK cell subpopulations were performed according to previously published protocols (13,14).In brief, whole blood sample was divided into 8 different panel tests to analyze the following specific cell populations: (i) general lymphocytes, (ii) B cell subsets, (iii) CD4 T cell subsets, (iv) CD8 T cell subsets, (v) regulatory T cells, (vi) recent thymic emigrants (RTEs), (vii) NK cell subsets, and (viii) NK cell activation markers.To each panel test, a 100 µl whole blood specimen was added and incubated with specific antibodies to the chosen cell subsets.The antibody mixes, cell markers, fluorochromes, IgG subtypes, antibody concentration, and clones used for every panel test are explained in Supplementary Table S1.
To decide the optimum dilution for each staining, each antibody for eight-color analysis panels was titrated separately and compared with isotype controls.Cell surface markers were stained for 15 minutes in the dark at room temperature and then erythrocytes were lysed by lysis buffer (BD Biosciences, Heidelberg, Germany) after incubation or 10 minutes.After centrifugation and washing with phosphate-buffered saline (PBS) (Biochrom, Berlin, Germany), cells were fixed with 200 µl PBS that contains 1% formaldehyde and ready for flow cytometry analysis.

Flow cytometry analysis
The flowcytometry data acquisition was made using FACS Canto II (BD Biosciences) machine.The FACS Canto II is has three lasers: a 647 nm red laser, a 488 nm blue laser, and a 405 nm violet laser.During the analysis of all panel tests, automatic standard compensation was done, except for the B cell subsets panel, where the spectral overlap of the allophycocyanin (APC) toward peridinin-chlorophyll-protein (PerCP) was set to 12.66, to achieve optimal separation of memory and naive B cells.A total of 50,000 cell events were collected in the lymphocyte collection gate (CD45+ versus SSClow) for eight-color analysis, and the FACS DIVA (BD Biosciences) software version was used for data analysis 8. The absolute cell counts were determined using the relative flow cytometric cell counts measured by BD FACS Canto II Flow Cytometer set in relation to absolute cell counts analyzed by Sysmex XP-300 automated hematology analyzer (Sysmex Germany GmbH, Bornbarch, Germany) based on the standard protocol.

General lymphocytes gating strategy
The general lymphocyte gating strategy gives an overview of changes in the lymphocyte subpopulations.B cells (CD19+), T cells (CD3+), and NK cells (CD16+/56+) as well as HLA-DR+CD38+ activated T cells were gated from the lymphocyte gate using specific markers.Moreover, NK-T cells (T cells co-expressing CD16 molecule) were separated by gating CD3 versus CD16/56.The CD8 versus CD4 plot was used to separate 4 different T cell subsets: CD4+ and CD8+ T cells, CD4-/CD8-double-negative and CD4+/CD8+ double-positive T cells (Figure 1).from the CD27 gate by IgD versus IgM demarcation.The CD38low versus CD21low plot creates the gates for the activated CD21lowCD38low B cells and immature transitional B cells, which are specified to be CD21lowCD38+ B cells.IgM+ and CD38+ co-expressions were used for gating Transitional B cells.From this plot, plasma-blasts were separated using CD38+ expression and the absence of surface IgM.Additionally, CD27 +CD138+ peripheral plasma cells were identified (Supplementary Figure S1(I)).

NK cells subsets gating strategy
For NK cells subsets analysis, T cells were excluded by gating the CD3 negative cells.From the CD3 negative cells, CD16 and CD56 markers were used to separate mature NK cells (CD16+CD56dim) and CD56 bright NK cells.In the next step, the expression of NK cells activation markers such as Nkp30, Nkp44, Nkp46, and CD57 was analyzed on mature NK cells (Supplementary Table S1(III)).

Statistical analysis
Descriptive statistics were used to analyze demographic and clinical characteristics of patients before and after anti-cancer treatment.Treatment effects on the immunologic parameters compared to their baseline values were determined by applying linear mixed models (random intercept and random slope) with the statistical software suite R (version 4.2.0,R packages "lme4" and "lmerTest") to consider multiple measurements per subject.Time points were assumed to be equidistant for this purpose.We further adjusted for age.Parameters were initially log-or logit-transformed if they were on an absolute or percentage scale, respectively.We corrected for multiple significance testing of changes in lymphocyte subpopulations after anti-cancer treatment by controlling the False Discovery Rate (FDR, Benjamini-Hochberg) for the 130 tests performed, so that FDR is less than or equal to 5%.In the following we denote the corrected p-values (according to FDR) as "q-values".For survival analysis, the ratios of survivors to all included patients within each treatment option were calculated.

Effects of anti-cancer therapies on B cell subpopulations
Absolute number of transitional B cells was significantly increased after therapy with immunotherapy (q=0.032),immunochemotherapy (q=0.030), and antibodies against VEGF (q=0.024).The transitional B cells percentage was significantly increased after immunotherapy (q=0.033).The percentage of B cells and absolute number of B cells were significantly increased after adjuvant chemotherapy (q=0.023).However, the percentage and absolute number of transition cells were significantly decreased after chemotherapy (q=0.001).Likewise, the absolute number of memory B cells and non-switched B cells were decreased significantly after adjuvant chemotherapy (q=0.023, and q=0.018, respectively) (Table 2, Figure 2).

Effects of anti-cancer therapies on NK cells subsets
In our study, none of the treatment options used had a statistically significant effect on NK cell subsets.However, an increased trend in the percentage of immature NK cells II (CD56 +CD16-) after immunotherapy (q=0.057) and a decreased trend in the percentage of CD16+CD56 NK cells (cytolytic activity) after immunochemotherapy (q=0.059) was observed (Table 4; Figure 2).

Effects of anti-cancer therapies on monocytes and granulocytes
In addition to lymphocytes, we assessed possible changes in monocytes and granulocytes after our therapy options.Percentages of monocytes significantly increased from 3.7% to 11.7% (q=0.033) after immunotherapy.Moreover, in the chemotaxis of granulocytes, the number of migrated granulocytes was significantly increased (q=0.023), while chemotaxis of granulocytes and percentages of active granulocytes (L-selectin) significantly decreased (q=0.018) after adjuvant chemotherapy.Other therapy options did not show significant changes in monocyte and granulocyte parameters (Supplementary Table S2).

Effect of age on T cell, B cell, and NK cell subpopulations
Age showed significant association with CD4+ T-lymphocytes, naive (CD45RA+CCR7+) CD8+T cells, and CD4-recent thymus migrations cells (q<0.05).However, age was not significantly associated with lymphocyte subpopulations showing significant changes after anti-cancer therapies.These findings suggest that the observed changes in lymphocyte subpopulations after anticancer therapies are not confounded by age (Figure 2, Supplementary Figure S2).

Discussion
This study was aimed to analyse the impact of different therapeutic regimens on the immune system of NSCLC patients investigating changes in B-, T-, and NK cell subsets during therapy.
The absolute numbers of transitional B cells were significantly increased after therapy with immunotherapy, immunochemotherapy, and antibodies against VEGF.Percentages of transitional B cells also significantly increased after immunotherapy.Similarly, absolute numbers and percentages of B cells were significantly increased after adjuvant chemotherapy.The observed upregulation of B cells and transitional B cells during these therapies suggest activation of immune defense.These effects of therapies on B cells also imply the potential use of these cells as a target for vaccination against lung cancer.In line with this, targeted strategies to enhance B cells by using stimulated B cell ligands have been applied to inhibit lung metastasis and tumor growth (15,16).
B cells are important immune cells that support an effective antitumor immune response by their capability to present tumor antigens, activate cytotoxic T cell and T-lymphocytes, response, as well as their ability to generate cytokines and anti-tumor antibodies (17)(18)(19).In line with the findings of this study, previous studies suggest that B cell counts increase in tumors of patients responding to immunotherapy, and tumor-infiltrating B cells were associated with, tumor stage and longer survival in NSCLC (20,21).Even though the importance of B cells in making anti-cancer immune responses was progressively unraveled in the last years, a better understanding about the role of different B cell subsets, particularly transitional B cells is still required in cancer patients (17).
Interestingly, absolute numbers and percentages of transitional B cells significantly decreased after chemotherapy.Similarly, the absolute number of memory B cells and non-class switched B cells significantly decreased after adjuvant chemotherapy.Downregulation of B cells during chemotherapy suggests a potential suppression of immune defense due to cytotoxic side-effects on the bone marrow.A previous study also reported that chemotherapy reduced the absolute counts of B cells (22).Another study showed that chemotherapy reduced absolute counts of B cells followed by a slow increase but not reaching baseline levels.This suggests that the chemotherapy has a long-term effect on lymphocyte subsets (23).
In this study, we noticed a significant increase in activated cytotoxic T cells in patients treated with immunotherapy and immunochemotherapy.The immune system activation by immunotherapy and immunochemotherapy is a desired therapeutic effect.T lymphocytes are the most important inflammatory cells that infiltrate the tumor, exert a direct cytotoxic effect, or cause lysis of tumor by cytokine release (24-26).CD8+ T cells can also increase and differentiate into CD8+ T Heatmap showing changes in lymphocyte subpopulations after treatment with different anti-cancer therapies and aging.Statistically significant changes in lymphocyte subpopulations are indicated by dark red and red color boxes (FDR = 0 -1% and 1 -5%, respectively).

Immunotherapy
Chemotherapy Immuno-chemotherapy Adjuvant chemotherapy Antibodies against VEGF before after qvalue before after qvalue before after qvalue before after qvalue before after qvalue T cells (cells/µL) 575

NK cell subpopulations
Immunotherapy Chemotherapy Immunochemotherapy Adjuvant chemotherapy Antibodies against VEGF before after q-value before after q-value before after q-value before after q-value before after q-value NK cells (CD16/56) ( cells that can infiltrate tumors by peripheral blood migration and play an essential role in the direct killing of tumor cells (27).Our finding is in line with the reported increase of CD8+ T cells after immunotherapy in lung cancer patients (28).Another study also suggested that as part of a successful immunotherapy response, effector memory tumor antigen-specific cytotoxic T cells are increased (29).However, in contrast to our findings, other studies reported decreased proportion of CD8+ T cells after immunotherapy as compared to baseline values (30,31).The observed differences in cytotoxic T cell response after immunotherapy could be explained as follows.First, lung cancer patients treated with PD-1 inhibitors are characterized by expressing effector-like phenotypes (HLA-DR+, CD38+, Bcl-2lo), co-stimulatory molecules (CD28, CD27, ICOS), a highly specific subset of proliferating CD8+ T cells in the peripheral blood, and high levels of PD-1 (32).Second, different timings of blood sampling could contribute to the observed contradictory findings across studies.CD8+ T cell response induction by blocking the PD-1 pathway in the peripheral blood is transient and detected during the first four weeks after treatment start.Then these specific CD8+ T cells migrate to the tumor sites (32).
Although none of the therapies investigated in this study showed a significant effect on NK cells, a trend increase in the percentage of immature NK cells was observed after immunotherapy.A decrease in the percentage of CD16+CD56 NK cells was also observed after immunochemotherapy.A previous study demonstrated that the PD-L1 leads blockage of causes activation of NK cells and increases the direct antitumor effect of NK cells (33).Another study showed that in NSCLC patients, NK cell activity is a marker to predict the immunotherapy response (34).Drugs modulating the function and proliferative activity of NK cells have been developed to produce synergistic effect in combination with immunotherapy (35, 36).
In addition to lymphocyte subpopulations, we assessed the possible effects of anti-cancer therapies on monocytes and granulocytes.Percentage of monocytes showed a significant increase after immunotherapy.This could be interpreted as the effect of a T cell interaction and activation by immunotherapy as described in a previous study (37).In several tumor types, an increase in HLA-DR low monocytes has been reported (38).As already described for HLA-DR low monocytes in sepsis, these monocytic cells might decrease T cell function in patients with cancer (39).Moreover, the number of migrated granulocytes in this chemotaxis study was significantly increased, while the percentage of active granulocytes (L-selectin) decreased after adjuvant chemotherapy.This is possibly due to the consequence of high baseline activity, postoperative inflammation, and wound healing processes.Previous studies also reported that the number of neutrophils in tumor tissues and blood is correlated with poor patient outcomes and disease progression (40).Neutrophil count was associated with the percentage of HLA-DR low monocytes in patients with lung cancer managed by surgery of the primary tumor as an important subpopulation of the myeloid-derived suppressor cells (MDSC).Nevertheless, in the late tumor stages, this observation was not confirmed (41).
In the present study, NSCLC patients treated with adjuvant chemotherapy showed the highest survival of 80%.In contrast, patients treated with chemotherapy alone showed the lowest survival with 0%.This could be explained by the fact that adjuvant chemotherapy is usually applied for early-stage NSCLC; stage IIA, IIB, and IIIA NSCLC to kill remaining cancer cells after surgery (42).Adding immunotherapy to chemotherapy in a palliative setting increased observed survival to 22% compared to chemotherapy alone.This is consistent with previous reports showing a significant increase in overall response rates, overall survival and progressionfree survival in NSCLC patients treated with immunotherapy compared to chemotherapy (43).Five-year follow-up study showed a significantly increased median time of overall survival, which doubled from 13.4 months in chemotherapy to 26.3 months in immunotherapy; and a higher overall survival rate, 31.9% in the immunotherapy group compared to 16.3% in the chemotherapy group (44).Immunotherapy enhances the immune system that targets cancer cells and slows or stops the cancer cells metastasis and growth (45).Previous studies also showed that improved survival is correlated with a strong immune response to tumers.Moreover, higher numbers of CD8+ T cells and natural killer cells are linked with better patient survival (46,47).
We repeated all the analyses twice, with further adjustment for tumor stage, and with further adjustment for tumor stage and exclusion of patients with adjuvant therapy to check that stage of tumor did not bias our initial analysis with adjustment for age only.Indeed, the results were comparable yielding no significant additional information.
This study has some limitations to be mentioned.Firstly, despite multiple measurements, due to the limited sample size our study might be underpowered to identify smaller changes in the lymphocyte subpopulations before and after treatment with different anti-cancer therapies.Likewise, survival analysis is limited by the small sample size.Secondly, this study was conducted at a single health care facility, therefore, selection bias cannot be fully excluded.
Since this study was the first one investigating such a large spectrum of parameters in cytomics, further research is needed to determine prognostic significance, further potential confounding factors and long-term survival.
-small cell lung cancer; VEGF, Vascular endothelial growth factors; N, number.Baseline = Measurement before therapy initiation.

TABLE 1
Socio-demographic and clinical characteristics of study participants by treatment group (N = 32).

TABLE 2
Changes in B cell subpopulations in NSCLC patients during different anti-cancer therapies.
Values "before" and "after" correspond to the measured values before the initial treatment and the measured value after the last treatment, respectively, after an average observation time of 718/459/473/1,060/1,462 days of immunotherapy/chemotherapy/ immunochemotherapy/adjuvant chemotherapy/antibodies against VEGF.Q-values correspond to the significance of the estimated change over time from linear mixed modelling.VEGF, Vascular endothelial growth factors; NA, Not available.

TABLE 4
"before" and "after" correspond to the measured values before the initial treatment and the measured value after the last treatment, respectively, after an average observation time of 718/459/473/1,060/1,462 days of immunotherapy/chemotherapy/immunochemotherapy/ adjuvant chemotherapy/antibodies against VEGF.Q-values correspond to the significance of the estimated change over time from linear mixed modelling.VEGF, Vascular endothelial growth factors; HLA-DR, Human Leukocyte Antigen-DR isotype.Changes in the NK cell subpopulations in NSCLC patients after different anti-cancer therapies. Values