Background
Granulocyte colony-stimulating factor (G-CSF) can be administered to healthy individuals donating hematopoietic stem cells (HSC) for transplantation and to cancer patients with the aim to prevent and/or treat chemotherapy-induced neutropenia. Currently, primary prophylaxis with G-CSF is recommended in patients at high risk for febrile neutropenia based on age, medical history, disease characteristics and myelotoxicity of the chemotherapy regimen.
Filgrastim is a recombinant human G-CSF derived from
Escherichia coli. Filgrastim has a short elimination half-life and requires daily subcutaneous injections for each chemotherapy cycle. The inconvenience associated with filgrastim administration has prompted the development of its covalent conjugation with monomethoxypolyethylene glycol (PEG) to obtain a longer-acting form (pegfilgrastim). The covalent attachment of PEG to the N-terminal amine group of the parent molecule increases its size, so that neutrophil-mediated clearance predominates over renal clearance in elimination of the drug, extending the median serum half-life of pegfilgrastim to 42 hours, compared with 3.5-3.8 hours for filgrastim [
1]. However, the half-life is variable, depending on the absolute neutrophil count (ANC), which in turn reflects the ability of pegfilgrastim to sustain neutrophil production. The PEG group in the pegfilgrastim molecule is a relatively inert adduct and is expected not to alter granulocyte function significantly compared with filgrastim. In line with this assumption, pegfilgrastim retains the same biological activity as filgrastim, and binds to the same G-CSF receptor, stimulating neutrophil proliferation, differentiation and activation.
The long-term effects of long-acting growth factors such as pegfilgrastim are unknown. Because an increasing number of healthy donors and cancer patients are exposed to pharmacologic doses of G-CSF, a thorough understanding of G-CSF effects is imperative to safeguard donor and patient safety. In this respect, there is accumulating evidence that the biological activities of G-CSF are not limited to the myeloid lineage but extend to cell types and cytokine networks implicated in inflammation, immunity and angiogenesis [
2]. Initial studies in mice supported a role for G-CSF in immune deviation towards T helper type 2 (Th2) cytokine production [
3]. In humans, G-CSF increases interleukin (IL)-4 release and decreases interferon (IFN)-γ production [
4], induces immune modulatory genes in T cells, including the Th2 master transcription factor GATA-3 [
5], and promotes the differentiation of type 1 regulatory T cells (Treg), endowed with the ability to release IL-10 and transforming growth factor (TGF)-β1, and to suppress T-cell proliferation in a cytokine-dependent manner [
6]. Furthermore, G-CSF induces the release of hepatocyte growth factor (HGF) [
7], a pleiotropic cytokine that inhibits dendritic cell (DC) maturation [
8] and down-regulates immune responses
in vivo[
9]. Finally, G-CSF mobilizes human type 2 DC (DC2) [
10] and promotes the
in vitro differentiation of regulatory DC through the stimulation of IL-10 and IFN-α production [
11]. On a molecular level, G-CSF may determine mitochondrial dysfunction and proliferation arrest in T cells [
12]. G-CSF-mobilized monocytes acquire the ability to release large quantities of immunosuppressive IL-10 and impair the induction of CD28-responsive complex in CD4
+ T cells [
13]. Similar to filgrastim, pegylated G-CSF enhances the lipopolysaccharide (LPS)-stimulated production of immune suppressive IL-10 and favorably affects the clinical course of graft-versus-host disease (GVHD) in mice [
14].
It is presently unknown whether pegylated G-CSF modulates human T-cell and DC function to a similar extent as unconjugated G-CSF. The hypothesis that the two formulations of G-CSF may target distinct cell populations
in vivo and that, in spite of structural similarities, the spectrum of their biological activities may diverge is supported by investigations with human pegfilgrastim-mobilized HSC, which display unique features compared with HSC mobilized by filgrastim [
15]. The present study provides evidence that pegylated G-CSF mobilizes both DC1 and DC2 precursors and, at variance with filgrastim, promotes monocytic IL-12 release. These findings portend favorable implications for pegfilgrastim administration to cancer patients.
Methods
Patient eligibility and treatment plan
The study population was comprised of 12 patients with gynecological malignancies (7 ovarian, 4 endometrial, 1 cervical cancer) ranging in age from 38 to 78 years (median age = 68 years). All patients received a conventional chemotherapeutic regimen, consisting of carboplatin (AUC5) and paclitaxel (175 mg/square meter). The patients' clinical characteristics are summarized in Table
1. After the completion of chemotherapy, patients were given a single dose (6 mg) of subcutaneous pegfilgrastim (Neulasta
®; Amgen Dompè, Milan, Italy), as prophylaxis of febrile neutropenia. The investigations were approved by the Institutional Review Board. A retrospective analysis of 7 registrational clinical trials that examined the safety and efficacy of pegfilgrastim indicated that serum pegfilgrastim concentrations are consistently sub-therapeutic (< 2 ng/ml) by day +12 from the commencement of treatment [
16]. Taking advantage of this knowledge, we collected blood samples from each consented patient on day 0 (the day before chemotherapy), and on days +7, +11 and +21.
Table 1
Patients' characteristics
UPN #1 | Endometrial carcinoma (endometrioid) | Ic | G3 | 4 |
UPN #2 | Endometrial carcinoma (serous) | IV | G3 | 5 |
UPN #3 | Ovarian carcinoma (serous) | IIIb | G3 | 4 |
UPN #4 | Cervical carcinoma (squamous) | Ib2 | G2 | 2 |
UPN #5 | Ovarian carcinoma (serous) | IIIc | G3 | 3 |
UPN #6 | Endometrial carcinoma (mixed) | Ic | G2 | 1 |
UPN #7 | Ovarian carcinoma (serous) | Ic | G3 | 4 |
UPN #8 | Ovarian carcinoma | IIIc | G3 | 4 |
UPN #9 | Ovarian carcinoma (serous) | IIIc | G3 | 4 |
UPN #10 | Endometrial carcinoma (endometrioid) | Ic | G3 | 4 |
UPN #11 | Ovarian carcinoma (endometrioid) | IIIc | G3 | 3 |
UPN #12 | Ovarian carcinoma (endometrioid) | IIIb | G2 | 4 |
A control group of 7 patients with gynecological malignancies received the same carboplatin/paclitaxel chemotherapy regimen, followed by daily filgrastim (5 μg/kg of body weight) from day +2 to day +10. Blood samples for
ex vivo studies were drawn on day 0 (the day before chemotherapy) and on days +7, +11 (24 hours after the last filgrastim administration) and +21. For both groups of patients, serum was obtained by centrifugation at 4,000 rpm for 15 minutes shortly after blood collection, was divided into aliquots and stored at -80°C until used. Peripheral blood mononuclear cells (PBMC) were separated by Ficoll-Hypaque density gradient centrifugation, as previously reported [
11], and were used as detailed below.
Generation of monocyte-derived DC (Mo-DC) and evaluation of DC endocytic activity
CD14+ monocytes were purified by negative selection (Monocyte Isolation Kit II, Miltenyi Biotec, Bergisch Gladbach, Germany) and were cultured in RPMI-1640 medium for 6 days at 37°C under serum-free conditions (10% BIT-9500; StemCell Technologies, Vancouver, BC) but in the presence of 500 IU/ml recombinant human GM-CSF and 25 ng/ml IL-4 (both cytokines were from R&D Systems, Oxon, Cambridge, UK). When indicated, the DC preparations were matured with 500 IU/ml tumour necrosis factor-α (TNF-α; R&D Systems) for 48 hours. Patient serum obtained before (pre-G) or after G-CSF administration (post-G) was supplemented to freshly isolated monocytes at 20% (v/v). In selected experiments, monocytes were stimulated in vitro with LPS (1 μg/ml) for 24 hours, prior to the measurement of secreted IL-12p40/IL-12p70 and IL-10 by ELISA.
To evaluate DC endocytic activity [
17], monocyte-derived DC populations were suspended in culture medium supplemented with 10% fetal calf serum (FCS) in the presence of 100 μg/ml FITC-dextran (Sigma Chemical Co., St. Louis, MO) for 1 hour at 37°C. Control DC cultures were pulsed with FITC-dextran at 4°C, as previously detailed [
8]. The extent of FITC-dextran incorporation was expressed as the ratio between the mean fluorescence intensity (MFI) of samples kept at 37°C and the MFI of samples cultured at 4°C, as detailed in the Figure legends.
T-cell isolation and primary MLR
CD4+ T cells were isolated from the peripheral blood with an indirect magnetic labeling system (CD4+ T Cell Isolation Kit II; Miltenyi Biotec). Briefly, PBMC were labeled with a cocktail of biotin-conjugated antibodies against CD8, CD14, CD16, CD19, CD36, CD56, CD123, TCR γ/δ and CD235a. Anti-biotin microbeads were used for depletion, yielding a population of highly pure, untouched CD4+ T cells. CD25 microbeads II (Miltenyi Biotec) were subsequently used for positive selection or depletion of CD25+ cells, following the manufacturer's instructions.
CD4
+CD25
- T cells were re-suspended in RPMI-1640 containing carboxyfluorescein-diacetate succinimidyl-ester (CFDA-SE, 2.5 μM; Molecular Probes, Eugene, OR) for 10 minutes at 37°C. To quench the labeling process, an equal volume of FCS was added. After washings in RPMI-1640 medium supplemented with 10% FCS, CD4
+CD25
- T cells were activated with the mixed leukocyte reaction (MLR), as reported elsewhere [
6]. Briefly, 5 × 10
4 allogeneic CD4
+CD25
- T cells were cultured with fixed numbers of irradiated (25 Gy) DC or monocytes for 7 days, in RPMI-1640 medium supplemented with 20% BIT serum substitute. In selected experiments, serum from patients given either pegfilgrastim or filgrastim was supplemented at 20% (v/v) to the allogeneic MLR containing T cells and monocytes/DC from third-party healthy donors, as previously detailed [
18].
Immunological markers, four-color flow cytometry and data analysis
Mo-DC and monocytes were incubated for 20 minutes at 4°C with the following FITC-, PE-, PerCP- or PE-Cy7-conjugated monoclonal antibodies (mAb): CD1a, CD11c, CD14, CD80, CD86, CD83 (Caltag Laboratories, Burlingame, CA), HLA-DR, CD11c and IL-3 receptor α-chain or CD123 (BD Biosciences, Mountain View, CA), immunoglobulin-like transcript 3 (ILT3), DC-SIGN (DC-specific ICAM-3 grabbing non-integrin; CD209; Immunotech, Marseille, France), or with the appropriate fluorochrome-conjugated, isotype-matched irrelevant mAb to establish background fluorescence.
To monitor DC mobilization, peripheral blood samples were stained with a cocktail of FITC-conjugated mAb directed against lineage-specific antigens (CD4, CD14, CD16, CD19, CD20, CD56; Lineage Cocktail 1, BD Biosciences), and with anti-CD123, anti-HLA-DR and anti-CD11c mAb (BD), in order to discriminate type 1 DC (DC1) from DC2. Cells were then incubated with ammonium chloride lysis buffer for 5 minutes to remove residual red blood cells. Unfractionated whole blood samples were gated on the basis of forward and side scatter characteristics. After gating on lineage
-HLA-DR
+ events, two populations of DC were identified, corresponding to HLA-DR
+CD11c
+ DC (DC1) and HLA-DR
+CD123
+ DC (DC2), as previously published [
10]. The proportion of DC1 and DC2 within lineage
-/dim cells was enumerated and expressed as a percentage of total leukocytes.
The analysis of CFDA-SE fluorescence in cell proliferation tracking assays was performed with the proliferation wizard of the ModFit™ LT 2.0 software (Verity Software House Inc., Topsham, ME). Replication data were expressed in terms of proliferation index (PI), which was calculated as previously detailed [
12].
The frequency of CD4+FoxP3+ Treg cells in the peripheral blood of G-CSF-treated patients and in MLR cultures was estimated with an anti-FoxP3 mAb (PCH101 clone; eBioscience, San Diego, CA). Cells were initially stained with fluorochrome-conjugated anti-CD4 and anti-CD25 mAb (BD Biosciences), followed by sequential cell fixation and permeabilization and by labeling with the Alexa-Fluor® 488-conjugated anti-human FoxP3 mAb.
All samples were run through a FACS Canto® flow cytometer (BD Biosciences) with standard equipment.
Analysis of cytokine production
IL-12p40, IL-12p70, IL-10, TGF-β1 and HGF levels in patient serum and in culture supernatants were quantified by ELISA, using commercially available reagents (R&D Systems). The limits of detection were < 15 pg/ml IL-12p40, 0.625 pg/ml IL-12p70, 7.8 pg/ml IL-10, 7 pg/ml TGF-β1 and <40 pg/ml HGF.
HPLC measurement of tryptophan (Trp) and kynurenine (Kyn)
Quantification of serum Trp and Kyn was obtained using reverse-phase (RP)-HPLC. The chromatographic procedure was similar to a method previously described, with minor modifications [
19]. In brief, sample aliquots (100 μL) were deproteinized with HClO
4 (0.3 M final concentration). After centrifugation (14,000 rpm for 15 minutes), the supernatants were spiked with 50 μM 3-L-nitrotyrosine and analyzed using a ReproSil-Pur C18-AQ (4 × 250 mm, 5 μM granulometry) RP-HPLC column (Dr. Maisch GmbH, Ammerbuch-Entringen, Germany), using a double-pump HPLC apparatus from Jasco (Tokyo, Japan) equipped with a mod. 2070 UV spectrophotometric detector and a FP-2020 fluorescence detector. Both detectors were connected in series to allow simultaneous measurements. The chromatographic peaks were detected by recording UV absorbance at 360 nm and emission fluorescence at 366 nm, after excitation at 286 nm. The elution solvent was: 2.7% CH
3CN in 15 mM acetate buffer, pH 4.00 (both HPLC-grade from Fluka, Milan, Italy). To control the set-up and for peak quantification, Borwin 1.5 and MS Excel software were used. The concentrations of components were calculated according to peak heights and were compared both with 3-nitro-L-tyrosine as the internal standard and with the reference curves constructed with Kyn and L-Trp, both purchased from Sigma-Aldrich.
Statistical analysis
The approximation of data distribution to normality was tested preliminarily using statistics for kurtosis and symmetry. Data were presented as median and interquartile range, and comparisons were performed with the Mann-Whitney test for paired or unpaired data, or with the Kruskal-Wallis test with Dunn's correction for multiple comparisons, as appropriate. The criterion for statistical significance was defined as p ≤ 0.05.
Discussion
It is conceivable that the G-CSF formulations currently available for clinical use differentially affect WBC number and function. For instance, a direct comparison of lenograstim (nonglycosylated G-CSF) and filgrastim or pegfilgrastim with regard to neutrophil phenotype and function indicated that neutrophils primed with lenograstim are less functional and structurally more immature compared with those primed with filgrastim and, to a lesser extent, pegfilgrastim [
26]. Importantly, randomized clinical trials evaluating single administration of pegfilgrastim vs. daily filgrastim as an adjunct to chemotherapy in patients with hematological and solid malignancies reported similar efficacy profiles [
27] or even a lower overall rate of febrile neutropenia in patients treated with pegfilgrastim compared with those given daily filgrastim [
28].
The present study aimed to address whether pegfilgrastim given as prophylaxis for chemotherapy-induced neutropenia affects the number and function of immune cells, a finding with potential implications for the treatment of cancer patients. The immune modulating actions of unconjugated G-CSF have been previously described both
in vitro and
ex vivo[
29]. This basic knowledge has been translated into animal models of autoimmune disorders to skew the immune response and to promote tolerance. For instance, G-CSF ameliorated experimental autoimmune encephalomyelitis [
30], type 1 diabetes [
31], experimental colitis [
32] and lupus nephritis [
33] through effects on adaptive and innate immune responses. A pilot clinical trial in Crohn's disease provided
proof of principle in favor of immune regulatory effects by filgrastim in the human setting [
34]. In this study, daily treatment with G-CSF for 4 weeks was correlated with an increase of IL-10-secreting type 1 Treg cells in the peripheral blood and with the accumulation of plasmacytoid DC in the gut
lamina propria[
34].
In the present report, WBC and ANC recovery in patients treated with pegfilgrastim occurred without the fluctuations associated with daily filgrastim injections. The administration of pegfilgrastim translated into a transient but significant elevation of CD34-expressing HSC, lymphocytes and monocytes. Lymphocyte recirculation is expected to favorably impact on the immune control of the underlying malignancy, and the observation that prompt lymphocyte recovery predicts a higher relapse-free survival in cancer patients [
35] underpins the potential clinical significance of the pegfilgrastim-induced changes in WBC subsets. Pegfilgrastim did not elicit any appreciable mobilization of Treg cells, as documented by serial measurements of the frequency of circulating CD4
+FoxP3
+ Treg cells. We cannot rule out the possibility that any G-CSF-induced recirculation of Treg cells was obscured by the high frequency of Treg cells already measured at baseline. Of interest, filgrastim has been reported to increase the frequency of CD4
+CD25
high Treg cells only when given to cancer patients in combination with cyclophosphamide as HSC mobilization regimen [
36]. In healthy donors, the phenotype and frequency of CD4
+CD25
highFoxP3
+ Treg cells may be unaffected by G-CSF [
37]. At variance with human data, filgrastim recruited functional TGF-β-expressing Treg cells to the pancreatic lymph nodes of NOD mice, with the likely aim to restrain the proliferation and function of diabetogenic T cells [
31]. It remains to be determined whether Treg recirculation and/or recruitment to sites of inflammation and tissue injury may also occur in humans as a result of pegfilgrastim administration.
We were also interested in evaluating whether pegfilgrastim induced the release of immune suppressive HGF and TGF-β1. HGF is a pro-angiogenic and tumor-promoting cytokine. HGF reportedly skews DC function, driving an IL-10-secreting tolerogenic profile both in mice [
38] and in humans [
8]. We measured significantly elevated levels of HGF in patients treated with either pegfilgrastim or filgrastim. Furthermore, HGF secretion was significantly lower after pegfilgrastim compared with daily filgrastim administration. In contrast, serum TGF-β1 levels were not modified by either G-CSF formulation. The observation that HGF induces functional IDO1 in human monocyte-derived DC [
8] raised the previously unexplored possibility that pegfilgrastim may indirectly activate IDO1-mediated Trp breakdown into immune suppressive derivatives, collectively referred to as Kyn. Interestingly, serum Kyn were not significantly different when comparing samples at baseline with those obtained from patients receiving pegfilgrastim. It should be noted that baseline Kyn levels in our patient cohort were higher than those measured in healthy controls (median Kyn concentration = 1.86 μM; number of samples = 20), probably reflecting the expression of functional IDO1 by the ovarian and endometrial cancer cells [
39]. Also, mRNA signals for IDO1 in monocytes and granulocytes, a potential source of IDO1 activity [
40], were unchanged when comparing pre-G and post-G samples. These observations are backed by a recent study indicating that G-CSF-mobilized immature myeloid cells inhibit alloreactive responses in mice through an IDO-independent mechanism, and that G-CSF signaling is incapable of directly inducing IDO [
41].
The studies published so far suggest that the extent of DC1/DC2 mobilization by filgrastim crucially depends on the intensity of the mobilization regimen and on the underlying neoplastic disorder. In this respect, filgrastim preferentially mobilized DC2 in healthy donors [
10] but failed to impact on the DC1/DC2 ratio in patients with hematological and solid malignancies [
42]. In another study with healthy donors, low-dose G-CSF (8-10 μg/kg/day) increased the frequency of CD123
+ blood DC precursors but mobilized CD11c
+ DC only occasionally [
43]. Furthermore, high-dose G-CSF (30 μg/kg/day) mobilized CD123
+ DC in patients with multiple myeloma but only occasionally in those affected by non-Hodgkin's lymphoma, and exerted varying effects on CD11c
+ DC [
43]. We have shown herein that pegfilgrastim mobilized both DC1 and DC2 precursors into the peripheral blood of patients with gynecological malignancies treated with carboplatin and paclitaxel, suggesting lack of DC skewing
in vivo. The highest frequencies of DC1 precursors were measured on day +7 from pegfilgrastim administration, whereas DC2 precursors were higher in day +11 samples and declined thereafter. It is conceivable that different chemotherapy/growth factor combinations and doses and/or intrinsic characteristics of the underlying neoplastic disorder account for differences in the relative proportion of DC1/DC2 precursors and in their mobilization kinetics. It is presently unknown whether the transient DC1 mobilization induced by pegfilgrastim will impact on the host immune system's ability to control disease progression.
IL-12, a prototype member of a family of IL-12-related cytokines that includes IL-23 and IL-27, is an instigator of Th1 immune responses and possesses
in vivo anti-tumor activities [
44]. IL-12 is a heterodimer formed by a 35-kDa light chain (known as p35 or IL-12α) and a 40-kDa heavy chain (known as p40 or IL-12β). Messenger RNA encoding IL-12p35 is present in many cell types, whereas mRNA encoding IL-12p40 is restricted to cells that produce the biologically active heterodimer [
45]. DC and monocytes have been reported to secrete a 10-1,000 fold excess of IL-12p40 compared with IL-12p75 [
44]. A report on post-transplantation immune functions in 43 patients receiving filgrastim has shown that cytokine administration delays the reconstitution of CD4
+ T cells and blunts anti-fungal T-cell responses [
25]. These abnormalities were correlated with the inability of DC and monocytes from G-CSF-treated patients to release IL-12p40 [
25]. Interestingly, the
in vivo immune modulating effects of G-CSF were replicated
in vitro when monocytes from normal volunteers were differentiated along the DC lineage after their 24-hour pre-treatment with exogenous G-CSF. Under these conditions, IL-12p40 production was inhibited both at the mRNA and protein level [
25]. In our study, pegfilgrastim administration was associated with a significant increase of the inducible IL-12p40 subunit in patient serum. In patients given filgrastim, IL-12p40 slightly declined and returned to baseline values by day +11 from the commencement of cytokine treatment. Interestingly, neutrophil-derived serine proteases have been reported to inactivate human growth factors such as TNF-α at sites of inflammation and to promote the formation of cytokine split products [
46]. It is tempting to speculate that immunoreactive IL-12 in patients given filgrastim may have been degraded as a result of sharp increases in circulating PMN capable of releasing proteolytic enzymes. Intriguingly, monocytes from patients treated with pegfilgrastim released higher amounts of both IL-12p40 and IL-12p70
in vitro compared with monocytes from filgrastim-treated patients. In contrast, the LPS-induced release of IL-10 increased to a similar extent in cultures established with monocytes from patients given pegfilgrastim and filgrastim. IL-12p40 homodimers may behave as IL-12 receptor antagonists both in mice and in humans, inhibiting IL-12-induced T-cell proliferation [
47,
48]. Our observation that post-pegfilgrastim monocytes release higher amounts of bioactive IL-12p70 compared with post-filgrastim monocytes supports the conclusion that pegfilgrastim may not dampen
in vivo anti-tumor immunity and/or host defense against infectious agents, a response that crucially depends on the balance between IL-12 and IL-10 production. It has been reported that 6-sulfo LacNAc
+ DC, a major source of IL-12 and potent inducers of T-cell responses
in vitro, are efficiently mobilized in healthy donors given G-CSF at 7.5 μg/kg of body weight [
49]. Conceivably, pegfilgrastim might also favor the mobilization of 6-sulfo LacNAc
+ DC or other as yet unrecognized monocyte/DC populations with a unique ability to produce bioactive IL-12.
It is known that unconjugated G-CSF promotes the development of tolerogenic DC
in vitro[
11] and
in vivo[
31]. We showed herein that pegfilgrastim-induced soluble factors promoted the emergence of mature DC-like populations with high expression of costimulatory molecules (CD80, CD86), CD83 and CD209, and with low endocytic capacity. Post-pegfilgrastim DC-like cells also up-regulated ILT3, an inhibitory receptor detected on anergizing DC preparations [
50,
51], and yet activated the proliferation of allogeneic naïve T cells to a similar extent as immunogenic DC. It should be noted that ILT3 expression may be dispensable for the induction of CD4
+CD25
+ Treg cells by 1,25-dihydroxyvitamin D3 [
52], indicating that molecular determinants of T-cell suppression other than ILT3 may be operational depending upon the experimental system. Of potential interest, we measured high levels of IL-10 in post-pegfilgrastim DC cultures (317 ± 140 pg/ml compared with 27.1 ± 2.3 pg/ml in control cultures of immunogenic
GM4DC). IL-10 secretion may have been responsible for ILT3 up-regulation on post-pegfilgrastim monocytes, in line with the effect of exogenous IL-10 on ILT3 expression by human vascular endothelial cells [
53]. We also evaluated the ability of post-pegfilgrastim DC to activate allogeneic T-cell responses
in vitro. Interestingly, monocytes from patients given pegfilgrastim induced T-cell proliferation to a similar extent as immunogenic DC. In line with this, T-cell proliferation in response to allogeneic monocytes was not inhibited by the provision of post-pegfilgrastim serum to the MLR culture. Our observations on
in vitro DC phenotype and function reinforce the view that pegfilgrastim and filgrastim differ in their ability to skew monocyte function, with the former supporting the
in vitro development of activating DC and the latter favoring the emergence of tolerogenic DC preparations [
18].
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
GB carried out the experiments and participated in the design of the study. AM, AP and MC carried out the experiments. AP and GS participated in the design of the study and were responsible for patient care and sample procurement. RDC carried out the experiments and contributed to manuscript drafting. SD gave intellectual input and advice. SR participated in the design of the study, carried out the experiments, performed the statistical analysis and drafted the manuscript. All authors read and approved the final manuscript.