Abstract |
Introduction
Lung cancer (LC) is among the most frequent type of malignancies,
accounting for 13% of the newly diagnosed cases worldwide in 2012. Nonsmall-
cell lung cancer (NSCLC) represents almost 85% of the total number of
LC diagnoses and can be further classified into different subtypes based on the
expression of certain molecular and histological features.
Cancer-related Inflammation has been identified as a hallmark of cancer. Host
immune system interacts with tumor cells in a complex and paradoxical
process that has been shown to consist of three phases: elimination,
equilibrium and escape. During the escape phase tumor cells have acquired the
ability to evade immune surveillance by establishing an immunosuppressive
tumor microenvironment and begin to grow uncontrollably. A wide range of
cytokines released by either tumor cells or other cells in the tumor stroma
appear to disrupt the balance between the antitumor immune response and
immune tolerance and play a crucial role in the development of potent
immunosuppressive mechanisms by inducing immunosuppressive cells,
including myeloid-derived suppressive cells (MDSCs) and regulatory T cells
(Treg), in both the tumor microenvironment and the peripheral blood of cancer
patients.
Vascular endothelial growth factor (VEGF) is the principal mediator and key
promoter of tumor angiogenesis. Recent data suggest a multifaceted inhibitory
effect of VEGF on the development of a potent antitumor immune response.
Mechanisms underlying the immunosuppressive capacity of VEGF have been
found to include inhibition of maturation of dendritic cells, promotion of
MDSCs migration and selective accumulation in tumor sites and induction of
Treg proliferation and enhancement of their suppressive functionality. Taking
into consideration the proposed link between VEGF and immunosuppression,
it can be hypothesized that, the monoclonal antibody against VEGF,
bevacizumab may be useful in reversing tumor-induced immune tolerance.
118
There is a growing body of evidence that links MDSC-, Treg- and VEGFinduced
immunosuppression to tumor development and progression.
Accumulating data suggest that the presence of these cells in the peripheral
blood and in the tumor microenvironment is significantly associated with the
patients’ clinical outcome and survival. However, there is no consensus on the
phenotypic and functional characterization of these cells in NSCLC patients.
Furthermore, there are limited data regarding the effect of first-line
chemotherapy, with or without anti-VEGF therapy, on the frequency and
functionality of these cells in patients with NSCLC. The aim of this study was
to identify, phenotypically and functionally characterize subpopulations of
MDSCs and Treg in the peripheral blood of NSCLC patients, determine their
frequency at diagnosis and evaluate their prognostic significance.
Additionally, the methodology developed allowed the study of the effect of
chemotherapy, with or without the addition of bevacizumab, on the
functionality and the levels of circulating immunosuppressive cells
(Treg/MDSCs) and the correlation of their changes with the clinical outcome.
Patients and methods
Peripheral blood was collected from newly diagnosed, treatment-naïve patients
with stage IIIB/IV NSCLC before initiation of treatment, after the 3rd cycle of
chemotherapy, as well as at the time of disease progression. Blood samples
from aged- and sex-matched healthy volunteers were used as normal controls.
Various subpopulations of MDSCs and Treg were identified and analyzed
using flow cytometry. Their immunosuppressive activity was assessed by
demonstrating their inhibitory effect on IFN-γ production by activated T-cells.
Results
Phenotypic characterization of MDSCs subpopulations with clinical
significance and their prognostic role in treatment-naïve patients with NSCLC
A total of 110 patients were enrolled in this study. The patients’ median age
was 68 years. Approximately, 84.5% were men, 51.8% had adenocarcinoma
and 74.5% had stage IV disease. Briefly, three new subpopulations of MDSCs
were identified in patients with NSCLC: two monocytic
[CD14+CD15−CD11b+CD33+HLA-DR−Lin− (CD15−M-MDSCs) and
119
CD14+CD15+CD11b+CD33+HLA-DR−Lin− (CD15+ M-MDSCs)] and one
granulocytic [CD14−CD15+CD11b+CD33+HLA-DR−Lin− (G-MDSCs)]. The
levels of these MDSCs subpopulations were significantly increased in the
peripheral blood of the patients compared to healthy donors [(CD15+ MMDSCs,
3.5 ± 0.5% versus 0.5 ± 0.2%, p≤ 0.0001), (CD15− M-MDSCs, 5.2 ±
0.5% versus 3 ± 0.8%, p= 0.04) and (G-MDSCs, 2 ± 0.5% versus 0.1 ± 0.02%,
p≤ 0.0001)]. However, the difference in the frequency of these subpopulations
among patients with stage III disease and those with stage IV disease was not
statistically significant. Patients who progressed after 3 cycles of treatment had
higher M-MDSCs levels prior to initiation of treatment (CD14+CD15+HLADR
−Lin−: 1.1 ± 0.3% and CD14+CD15−HLA-DR−Lin−: 5.5 ± 1.1%) compared
to those with disease control (CR, PR and SD) (0.6 ± 0.07%, p=0.02 and 2.9 ±
0.3%, p=0.02, respectively). Moreover, MDSCs’ percentage within the
“normal range” (levels within the 90% of the healthy controls) prior to
initiation of therapy was associated with a longer progression-free survival
(PFS) and overall survival (OS) compared to increased (>90% of the healthy
controls) MDSCs’ levels (10.87 versus 5.3 months, p=0.005 and 12.9 versus
7.1 months, p=0.008, respectively). In the multivariate analysis, increased
frequency of circulating CD15+ Μ-MDSCs before treatment initiation was
identified as an independent prognostic factor for poor survival (PFS: HR =
2.41, 95% CI: 1.37–4.24, p=0.002 and OS: HR = 2.35, 95% CI: 1.25–4.41,
p=0.008). Finally, it was demonstrated that these MDSCs subpopulations are
functional and suppress the production of IFN-γ by activated CD3+ T cells.
Effect of treatment on the MDSCs’ subpopulations in the peripheral blood of
patients with NSCLC
A total of 46 patients were included, 33 on the chemotherapy arm and 13 on
the chemotherapy plus bevacizumab arm. The patients’ median age was 68
and 85% were male. After 3 cycles of treatment, 24% of the patients
experienced PR, 67% SD and 9% PD.
Overall, chemotherapy did not appear to have a uniform or significant impact
on the levels or the functionality of the various subpopulations of circulating
MDSCs. However, three cycles of bevacizumab-based chemotherapy resulted
120
in a statistically significant decrease in the levels of the G-MDSCs
subpopulation compared to regimens that did not include bevacizumab (mean
change in percentages of G-MDSCs with the bevacizumab-based versus non–
bevacizumab-based regimens: –0.68% ± 0.34% versus 0.69% ± 0.38%,
p=0.0086). Furthermore, after 3 cycles of treatment, patients experiencing PD
had significantly increased levels of all three MDSCs subpopulations
compared to the levels in patients that responded (CR+PR+SD) to treatment.
Interestingly, patients with PD had a statistically significant increase in the
percentage of CD15+ M-MDSCs compared to the pre-treatment levels
(p=0.03).
Phenotypic characterization of Treg subpopulations with clinical significance
and their prognostic role in treatment-naïve patients with NSCLC
A total of 156 patients with NSCLC were enrolled in this study. The patients’
median age was 62 years, 82.1% were male, 57.6% had adenocarcinoma and
82.1% had stage IV disease.
Briefly, the levels of CD4+CD25+ Treg were significantly increased in the
peripheral blood of the patients compared to healthy donors (24.81 ± 1%
versus 14.67 ± 1.5%, respectively, p= 0.0002). Depending on the stage of
activation and differentiation, three CD4+ Treg subtypes were further
analyzed: naive, effector and terminal effector Treg. Patients with PD as best
response to first line treatment had increased percentage of naive Treg
(CD25highCD127-/lowCD152-FoxP3lowCD45RO–) prior to treatment initiation
compared to those with disease control (CR, PR and SD) (3.17 ± 0.58% versus
0.81 ± 0.33%, respectively, p=0.003). Moreover, increased (>90% of the
healthy controls) levels of naïve Treg at baseline were associated with
significantly shorter OS compared to that of patients with Treg percentage
within the normal range (levels within the 90% of the healthy controls) (OS:
18.37 versus 40.47 months, respectively, p=0.039). Similarly, patients with
increased levels of effector Treg(CD25highCD127lowCD152+FoxP3+CD45RO+)
before treatment initiation had significantly shorter PFS (6.8 versus 8.53
months, p=0.046) and OS (5 versus 15.37 months, p=0.037) relative to that of
patients with Treg levels within the normal range. In contrast, patients
121
experiencing PD had significantly lower levels of terminal effector Treg
(CD25highCD127-CD152+FoxP3+CD45RO+) at baseline compared to those
with disease control (PD versus CR+PR+SD: 7.48 ± 1.27% versus
14.08 ± 2.74%, respectively, p=0.04). Moreover, increased levels of terminal
effector Treg were associated with longer PFS (16.2 versus 7.5 months
respectively, p=0.03) and OS (undefined versus 12.63 months, respectively,
p=0.049). Multivariate analysis revealed that increased frequency of
circulating naive Treg at baseline is an independent prognostic factor for poor
OS (HR = 8.632, 95% CI: 2.226-33.468, p=0.002). Finally, it was found that,
all the above described, Treg subpopulations, regardless of their activation and
differentiation stage, are functional and suppress the production of IFN-γ by
activated CD4+ T cells.
Effect of treatment on the Treg subtypes in the peripheral blood of patients
with NSCLC
A total of 46 patients were included in this study, the same patients analyzed
for the effect of treatment on MDSCs. Blood samples were obtained before
initiation of first-line therapy and after 3 cycles of treatment.
Overall, chemotherapy was not associated with a significant change in the
levels of the CD4+CD25highCD127−CD152+FoxP3+ Treg, in the peripheral
blood of the patients. However, after three cycles of bevacizumab-based
chemotherapy, patients receiving bevacizumab had statistically significant
decreased levels of CD4+CD25highCD127−CD152+FoxP3+ Treg compared to
patients who did not receive bevacizumab (percentages after 3 cycles of
treatment with bevacizumab-based versus non-bevacizumab-based regimens:
14.81 ± 4.615% versus 36.44 ± 4.506%, p=0.0058). In particular, regarding
naive Treg, patients who received 3 cycles of bevacizumab-based treatment
had statistically significantly lower cell counts compared to patients who did
not receive bevacizumab (naive Treg percetnage after 3 cycles of treatment,
bevacizumab versus non-bevacizumab based treatment: 0.6273 ± 0.345%
versus 3.44 ± 1.076%, p=0.019). In contrast, non-bevacizumab based
chemotherapy was associated with a statistically significant increase in the
naive Treg levels in the peripheral blood of the patients (naive Treg pre- and
122
after 3 cycles of treatment without bevacizumab: 1.456 ± 0.353% versus 3.44
± 1.076% respectively, p=0.045). There was no statistically significant
difference in the levels of effector Treg between patients treated with or
without the addition of bevacizumab (mean change in effector Treg
percentage, bevacizumab based versus non-bevacizumab based regimens:
0.3818 ± 1.54% versus -1.59 ± 3.2%, p=0.49). Finally, after 3 cycles of
treatment, patients treated with bevacizumab had decreased levels of terminal
effector Treg compared to patients who did not receive bevacizumab (terminal
effector Treg percentage after 3 cycles of treatment, bevacizumab based versus
non-bevacizumab regimens: 7.3 ± 2.958% versus 15.72 ± 3.128%, p=0.059).
Conclusions
In conclusion, this study precisely defines, phenotypically and functionally,
distinct MDSCs and Treg subpopulations with prognostic significance in the
peripheral blood of patients with NSCLC. Quantification of their frequency in
the peripheral blood has emerged as an independent prognostic factor for
survival in patients with NSCLC, highlighting the potential clinical utility of
the particular biomarker. The selective targeting of specific MDSCs and Treg
subpopulations through bevacizumab-based treatment provides evidence for
the immunomodulating effect of anti-angiogenic therapy and bevacizumab
emerges as a promising agent in “taking the foot off the brake of
immunosuppression”. Nevertheless, these experimental and clinical findings
remain to be confirmed prospectively in future randomized clinical trials
enrolling more patients with more homogeneity in the histological type, the
disease burden and the treatment that they receive. At the same time, efforts
should be mounted toward linking the levels of the circulating MDSCs and
Treg with their frequency in the tumor microenvironment, where they exert
their immunosuppressive action. Improving the understanding of the biology
and the functionality of MDSCs and Treg is considered to be of paramount
importance for the development of novel strategies to efficiently detect,
quantify and, finally, target tumor-induced immunosuppression.
|