Nội dung

McDonnell et al. BMC Cancer (2017) 17:417 DOI 10.1186/s12885-017-3403-5 RESEARCH ARTICLE Open Access Serial immunomonitoring of cancer patients receiving combined antagonistic anti-CD40 and chemotherapy reveals consistent and cyclical modulation of T cell and dendritic cell parameters Alison M. McDonnell1,2†, Alistair Cook1,2†, Bruce W. S. Robinson1,2,4, Richard A. Lake1,2† and Anna K. Nowak1,2,3*† Abstract Background: CD40 signalling can synergise with chemotherapy in preclinical cancer models, and early clinical studies are promising. We set out to define the immunological changes associated with this therapeutic combination to identify biomarkers for a response to the therapy. Here, we present serial immunomonitoring examining dendritic cell and T cell subpopulations over sequential courses of chemoimmunotherapy. Methods: Fifteen patients with mesothelioma received up to six 21-day cycles of pemetrexed plus cisplatin chemotherapy and anti-CD40 (CP-870,893). Peripheral blood was collected weekly, and analysed by flow cytometry. Longitudinal immunophenotyping data was analysed by linear mixed modelling, allowing for variation between patients. Exploratory analyses testing for any correlation between overall survival and immunophenotyping data were undertaken up to the third cycle of treatment. Results: Large statistically significant cyclical variations in the proportions of BDCA-1+, BDCA-2+ and BDCA-3+ dendritic cells were observed, although all subsets returned to baseline levels after each cycle and no significant changes were observed between start and end of treatment. Expression levels of CD40 and HLA-DR on dendritic cells were also cyclically modulated, again without significant change between start and end of treatment. CD8 and CD4 T cell populations, along with regulatory T cells, effector T cells, and markers of proliferation and activation, showed similar patterns of statistically significant cyclical modulation in response to therapy without changes between start and end of treatment. Exploratory analysis of endpoints revealed that patients with a higher than average proportion of BDCA-2+ dendritic cells (p = 0.010) or a higher than average proportion of activated (ICOS+) CD8 T cells (0.022) in pretreatment blood samples had better overall survival. A higher than average proportion of BDCA-3+ dendritic cells was associated with poorer overall survival at both the second (p = 0.008) and third (p = 0.014) dose of anti-CD40. (Continued on next page) * Correspondence: anna.nowak@uwa.edu.au † Equal contributors 1 School of Medicine and Pharmacology, The University of Western Australia, Crawley, WA 6009, Australia 2 National Centre for Asbestos Related Diseases, The University of Western Australia, Crawley, WA 6009, Australia Full list of author information is available at the end of the article © The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. McDonnell et al. BMC Cancer (2017) 17:417 Page 2 of 15 (Continued from previous page) Conclusions: Substantial cyclical variations in DC and T cell populations during sequential cycles of chemoimmunotherapy highlight the critical importance of timing of immunological biomarker assessments in interpretation of results and the value of linear mixed modelling in interpretation of longitudinal change over a full treatment course. Trial registration: Australia New Zealand Clinical Trials Registry number ACTRN12609000294257 (18th May 2009). Keywords: Mesothelioma, Dendritic cells, Prognosis Background As a variety of immunotherapies progress toward clinical approval, it is becoming more important to identify biomarkers to assess the clinical activity of these drugs; both to begin to understand what immunobiological changes are induced, and to identify those patients who are likely to benefit from these potentially toxic and often costly treatments. The addition of chemotherapy to immunotherapy in combination treatments is under intense investigation, however there is limited understanding of how concurrent chemotherapy may affect putative biomarkers of immunotherapy, and how to analyse and interpret these in the context of the cyclical changes in immunological parameters induced by cytotoxic treatments. Here, we present further immune biomarker data from a recent chemoimmunotherapy clinical trial conducted in patients with mesothelioma, and discuss the complexity of interpreting this information in the context of prediction and prognosis. CD40 is a member of the TNF receptor superfamily primarily expressed on antigen-presenting cells (APC), e.g. dendritic cells (DC), B cells and monocytes, but also found on some non-lymphoid cells such as epithelial and endothelial cells, fibroblasts, and some tumours [1]. During the T cell-mediated immune response, CD40 ligand (CD40L; CD154) -expressing CD4 helper T cells can activate APC through CD40 signalling. These APC can in turn provide a ‘licence-to-kill’ signal to CD8 cytotoxic T cells - the main effectors in immune-mediated tumour regression [2]. Extensive preclinical studies of anti-CD40 therapy have shown efficacy in various tumour model systems, and several clinical agents targeting the CD40 signalling axis have been or are currently under investigation [3]. CP-870,893 is a fully human IgG CD40 agonist antibody that has shown promise as a single agent in patients with solid tumours, although overall response rates are still low [4]. Although CP-870,893 infusion was predicted from preclinical studies to induce tumour-specific cell-mediated immune responses, this remains to be fully confirmed in the clinical setting. Agonistic anti-CD40 can synergise with chemotherapy and cure advanced tumours in mice, especially when administered after chemotherapy [5, 6]. This post- chemotherapy activity of agonistic anti-CD40 is hypothesised to occur by activating DC that have become ‘loaded’ with antigen from chemotherapy-induced tumour cell death, inducing expression of costimulatory molecules CD80 and CD86 and increased production of IL12 amongst other cytokines [7]. CP-870,893 has been investigated in conjunction with chemotherapy in early phase clinical trials, mostly in patients with advanced, treatment-resistant disease [7, 8]. In studies in pretreated patients, around 20% of participants achieved objective tumour regression. In our recent first-line mesothelioma trial (a phase Ib dose escalation study in combination with chemotherapy), 40% of patients achieved a partial response [9]. Currently identified pharmacodynamic effects of CP870,893 as a monotherapy are most obvious in the B cell compartment, with depletion and activation of peripheral B-cells occurring within 72 h of infusion [4]. Previous studies have also reported detectable modulation of DC by CP-870,893; in particular, depletion of CD11c + CD123dimCD14− DC from peripheral blood, and in vitro increases in HLA-DR expression by monocytederived DC [10, 11]. Weekly CP-870,893 monotherapy halved circulating lymphocyte concentrations after 48 h before returning to pre-treatment levels; this depletion was not observed when dosing occurred every three weeks [4, 10]. Thus, the pharmacodynamic effects of CP-870,893 are still somewhat undefined – particularly when combined with chemotherapy. Specifically, prolonged immune modulation over the longer term of a full course of treatment has not been characterised. We recently showed that B-cell depletion and activation also occurs over several cycles of combined CP870,893 plus pemetrexed and cisplatin chemotherapy in patients with mesothelioma [9]. Here, we present further in-depth flow cytometric analysis of patient peripheral blood mononuclear cells (PBMC) collected longitudinally throughout this study, in order to enhance our understanding of the immunobiology of combination chemoimmunotherapy and the unique challenges and considerations of analysis in this setting. We identify cyclical variations in PBMC subpopulations repeated with each cycle of chemo-immunotherapy, and identify potential relevant biomarkers of clinical activity in McDonnell et al. BMC Cancer (2017) 17:417 Page 3 of 15 dendritic cells, CD8+ effector cells and regulatory T cells in response to anti-CD40 agonist treatment in the context of chemotherapy. We present statistical analysis techniques which may inform other investigators in serial immunomonitoring of chemoimmunotherapy. hepatic and renal function tests, and toxicity assessment were performed weekly on combination treatment and three-weekly on CP-870,893 alone. Clinical, imaging, and time to event outcome assessments have been described previously [9]. Methods Cell preparation Peripheral blood volumetric cell counts Patients Clinical trial designs The clinical trial was a prospective, single-centre, phase Ib trial of cisplatin and pemetrexed with CP-870,893 [9]. A 3 + 3 phase I design was used with a 6-patient expansion cohort at the maximum tolerated dose (MTD) of CP-870,893. The primary endpoint was the MTD of CP870,893. Secondary endpoints included toxicity (NCI CTC Version 3.0), objective tumour response as measured by the modified RECIST criteria [12] and by fluorodeoxyglucose positron emission tomography (FDG-PET) [13], time to progression (TTP), and overall survival (OS). Eligibility Eligibility criteria have been previously described in detail [9]. In brief, all patients had confirmed malignant pleural mesothelioma, Eastern Co-operative Oncology Group (ECOG) performance status (PS) 0–1, and were planned for first-line cisplatin/pemetrexed. Exclusions specific to study drug were: history of venous thromboembolism or severe autoimmunity. The protocol was approved by the Institutional Human Research Ethics Committee and participants provided written informed consent. Australia New Zealand Clinical Trials Registry number ACTRN12609000294257. Treatment and outcome assessment Patients received cisplatin 75 mg/m2 and pemetrexed 500 mg/m2 on day 1 of a 21 day cycle to maximum 6 cycles with vitamin B12 and folate supplementation. CP-870,893 was given on day 8 each cycle, at three dose levels in consecutive patient cohorts: 0.1 mg/kg; 0.2 mg/kg; with 0.15 mg/kg as a deescalation level. Patients received premedication for cytokine release reaction before CP-870,893 administration as previously described [9]. Prophylactic medications for chemotherapy included corticosteroids (days −1 to 2) and antiemetics; other anti-cancer treatments were not allowed. Chemotherapy was stopped before 6 cycles in the event of progression, unacceptable toxicity, or patient withdrawal; in this event, CP-870,893 was also stopped. Patients with stable or responding tumour at 6 cycles could continue single agent CP-870,893 every 21 days for a further 6 cycles at the same dose level, ceasing on progression or toxicity. Complete blood count, Whole blood was analysed by flow cytometry on the day of collection to obtain absolute volumetric cell counts (cells per mL) of CD3+CD8+ and CD3+CD4+ T cells. Blood samples were stained using CD4-AlexaFluor488, CD3-PE and CD8-PECy7 antibodies as detailed in Table 1. Fixation and red blood cell lysis was performed using BD FACS lysing buffer, and data collected by three-color analysis using a Millipore Guava and Guava ExpressPro Software. PBMC Isolation Whole blood for PBMC isolation was collected into BD K2EDTA Vacutainers (BD Diagnostics, Australia) weekly during combined treatment (days 1, 8, 15), always prior to treatment administration (Fig. 1a). PBMC were isolated by Ficoll-Paque™ density gradient centrifugation, and cryopreserved in liquid nitrogen until analysis. Dual baseline samples were collected within 14 days of day 1, and pre-treatment on day 1 cycle 1. Serial analyses were performed on cryopreserved PBMCs, with all samples analysed concurrently for individual patients ensuring comparable experimental conditions across time points. PBMC flow cytometry PBMCs were thawed for 1 min at 37 °C and washed once in RPMI (Invitrogen), followed by two washes in PBS after putting cells into 96-well U-bottom plates (1 × 106 cells / well). PBMC were stained for expression of surface markers using specific anti-human monoclonal antibodies (mAb) comprising three 8-colour panels as detailed in Table 1. Dead cells were identified using LIVE/DEAD Fixable Dead Cell Stain Kit (Thermo Fisher Scientific, Waltham, MA, USA). DC were identified as staining with MHC Class II (HLA-DR-V500), and negative for a lineage cocktail of CD3-PerCP-Cy5.5, CD14-PerCP-Cy5.5, CD16-PerCPCy5.5, CD19-PerCP-Cy5.5, CD56-PerCP-Cy5.5 and LIVE/DEAD Fixable Red viability stain. Within this population, DC subsets were identified by antibodies against BDCA-1-PE, BDCA-2-FITC and BDCA-3-APC. Expression of CD40 was assessed within these subpopulations by CD40-APC-H7 staining. CD4+ T cells were identified by positive staining for CD3-V450 and CD4-APCH7, and negative for a lineage cocktail of CD19-V500, CD14-V500 and LIVE/DEAD McDonnell et al. BMC Cancer (2017) 17:417 Page 4 of 15 Table 1 List of antibodies Antigen Fluor Clone Isotype Supplier Catalogue # Panel BDCA-1 (CD1c) PE AD5-8E7 ms IgG1 Miltenyi 130–090-508 1 Dilution 1/10 BDCA-2 (CD303) FITC AC144 ms IgG1 Miltenyi 130–090-510 1 1/10 BDCA-3 (CD141) APC AD5-14H12 ms IgG1 Miltenyi 130–090-907 1 1/10 HLA-DR V500 G46.6 ms IgG2a BD Biosciences 561,224 1 1/100 CD3 PerCP-Cy5.5 SK7 ms IgG1 Biolegend 344,808 1 1/100 CD14 PerCP-Cy5.5 HCD14 ms IgG1 Biolegend 325,622 1 1/100 CD16 PerCP-Cy5.5 3G8 ms IgG1 Biolegend 302,028 1 1/100 CD19 PerCP-Cy5.5 HIB19 ms IgG1 Biolegend 2,072,925 1 1/100 CD56 PerCP-Cy5.5 HCD56 ms IgG1 Biolegend 318,322 1 1/100 CD40 APC-H7 5C3 ms IgG1 BD Biosciences 561,211 1 1/10 CD4 AF488 RPA-T4 ms IgG1 BD Biosciences 557,695 2 1/20 CD3 PE SK7 ms IgG1 BD Biosciences 347,347 2 1/50 CD8 PECy7 RPA-T8 ms IgG1 BD Biosciences 555,368 2 1/50 1/40 CD4 APC-H7 RPA-T4 ms IgG1 BD Biosciences 560,158 3 Foxp3 PE PCH101 rt IgG2a eBioscience 12–4776-42 3 1/20 CD25 APC M-A251 ms IgG1 BD Biosciences 555,434 3 1/5 CD127 PECy7 eBioRDR5 ms IgG1 eBioscience 25–1278-42 3 1/100 Ki67 FITC B56 ms IgG1 BD Biosciences 556,026 3,4 1/10 ICOS PerCP-Cy5.5 C398.4A ha IgG Biolegend 313,518 3,4 1/80 CD14 V500 M5E2 ms IgG2a BD Biosciences 561,391 3,4 1/80 CD19 V500 HIB19 ms IgG1 BD Biosciences 561,121 3,4 1/80 CD3 V450 UCHT1 ms IgG1 BD Biosciences 560,365 3,4 1/40 CD8 APC-H7 SK1 ms IgG1 BD Biosciences 560,179 4 1/40 CD38 AF647 HIT2 ms IgG1 Biolegend 303,514 4 1/40 HLA-DR PECy7 L243 ms IgG2a BD Biosciences 335,795 4 1/80 Bcl-2 PE Bcl-2/100 ms IgG1 BD Biosciences 556,535 4 1/10 List of monoclonal antibodies used for flow cytometric staining. Panels were used for dendritic cell staining of PBMC (panel 1), absolute cell counts of whole blood (panel 2), Treg staining of PBMC (panel 3), or CD8 T cell staining of PBMC (panel 4). Abbreviations: AF = AlexaFluor, ms = mouse, rt. = rat, ha = hamster Fixable Aqua viability stain. Tregs were identified within the CD3+ CD4+ T cell population by staining with CD25-APC, Foxp3-PE, and low CD127-PECy7. Within the Treg subset, cells were further assessed for staining by Ki67-FITC and ICOS-PerCP-Cy5.5. CD8+ T cells were identified by positive staining for CD3-V450 and CD8-APC-H7, and negative for a lineage cocktail of CD19-V500, CD14-V500 and LIVE/DEAD Fixable Aqua viability stain. Within the CD8+ population, activated effector cells were identified by HLA-DRPECy7 and CD38-AlexaFluor647. These cells were also assessed for staining by Ki67-FITC, ICOS-PerCP-Cy5.5 and Bcl2-PE. Samples were run on a FACSCanto II flow cytometer using FACSDiva software (both BD Biosciences). At least 50,000 lymphocyte events were collected per sample. Data were analysed using FlowJo software (Tree Star Inc., Ashland, OR, USA). Statistical analysis Linear mixed models were used to analyse the relationship between time and lymphocyte subsets, in addition to testing for interaction. Analysis used the R environment for statistical computing and IBM SPSS for Windows statistical package version 23. Bar graphs showing treatment means were analysed using ANOVA, P values are multiplicity adjusted using Dunn multiple comparisons tests. Exploratory analysis for correlation of DC or T cell subtypes with overall survival (OS) was performed separately for each sample collection time point between baseline and Cycle 3 Day 8. Patients were grouped above or below the median according to proportional presence of the DC or T cell subset under investigation. OS was analysed using the Kaplan Meier method, differences in OS between groups were calculated using the Mantel-Cox Log Rank test. No corrections for multiple comparisons were performed here as analyses were exploratory and hypothesis-generating. McDonnell et al. BMC Cancer (2017) 17:417 Fig. 1 (See legend on next page.) Page 5 of 15 McDonnell et al. BMC Cancer (2017) 17:417 Page 6 of 15 (See figure on previous page.) Fig. 1 a Patient treatment schedule showing timings of study drug administration and PBMC collection; Chemo = pemetrexed/cisplatin chemotherapy, CP = CP-870,893. Blood was collected at baseline, then days 1, 8 and 15 of each treatment cycle for a maximum of 6 cycles combined therapy. b Representative flow cytometry data demonstrating gating strategy for DC. Forward scatter (FSC) area vs. FSC-height was used for doublet discrimination. A ‘dump’ channel was used to gate out dead cells (LIVE/DEAD viability stain) plus those staining positively with a CD3/CD14/CD16/CD19/CD56 lineage cocktail (lin+). DCs were identified as lin−HLA-DR+ cells, and respective DC subpopulations identified by BDCA-1, BDCA-2 or BDCA-3. c Longitudinal flow cytometry data on DC across six cycles of chemoimmunotherapy, for BDCA-1, BDCA-2 and BDCA-3 subpopulations as a proportion of total PBMC. Left-hand panels show observed values from individual patients, together with their empirical means (solid line), mean and SD at baseline are quoted. Centre panels show results of fitting a linear mixed model; a linear trend over time and additive treatment effects of the day of the treatment yield the corresponding population average curves. Black and white numbered bars on X-axes represent the number of treatment cycles undertaken, time point ‘B’ represent pre-study baseline samples. Y axis scales have been modified from left-hand panels for clarity to show cyclical changes highlighted by modelling. Average change over the duration of the study is described. Right-hand panels show estimated treatment means, showing differences between day 1, day 8, and day 15 of the chemoimmunotherapy treatment over 6 cycles (P-values: * <0.05, ** <0.01, *** <0.001, **** < 0.0001). d Longitudinal data on the ratio of BDCA-1 to BDCA-2 dendritic cells Results Sixteen patients with radiologically assessable malignant mesothelioma were enrolled as described previously [9]. Patients were treated and PBMC samples collected as described in the materials and methods section and Fig. 1a. We analysed all patient PBMC samples by flow cytometry and here report on DC, and CD4 or CD8 T cell subsets. Linear mixed modelling of the data allowed us to not only assess immunological changes from week to week in response to different components of the treatment regimen, but also to examine longitudinal changes across 6 cycles of treatment. Dendritic cell subpopulations Blood DC subpopulations were identified by flow cytometry on the basis of BDCA marker expression using the gating strategy as shown in Fig. 1b (see discussion for DC subpopulation characteristics and roles). Whilst our data on the proportion of these DC subpopulations as a fraction of total PBMC showed a wide variability, both within and between individual patients over 6 cycles of chemoimmunotherapy treatment, the pattern of change was cyclical and consistent for BDCA-1+ and BDCA-2+ (CD303+ plasmacytoid) DC (Fig. 1c). In both these DC subsets, a marked proportional decrease of around 50% was observed prior to Day 1 of each cycle, coinciding with pre-chemotherapy medication with the glucocorticoid steroid dexamethasone [14]. The proportion of BDCA-2+ DC remained low until Day 15 of each cycle before returning to baseline levels, whereas a recovery in BDCA-1+ DCs was seen a week earlier at Day 8 (Fig. 1c). The less numerous, BDCA-3+ (CD141+ myeloid), DC showed a more complex profile, with linear mixed modelling indicating a rebound significantly above, baseline levels by Day 1 of each treatment cycle. The ratio of BDCA-1+ to BDCA-2+ DC was variable, reflecting the differential recovery of BDCA-1+ and BDCA-2+ subsets on Day 8 and Day 15 respectively, with these differences becoming less pronounced as the number of treatment cycles progressed (Fig. 1d). Dendritic cell functional markers Relative expression levels of both CD40 and HLA-DR were assessed by mean fluorescence intensity (MFI) of staining (Fig. 2). CD40 was expressed at highest levels on BDCA-1+ DC, and lowest levels on BDCA-3+ DC. Linear mixed modelling revealed a cyclical pattern of CD40 expression on all three subsets. BDCA-1+ DC expressed most CD40 at Day 1 of each cycle (after dexamethasone premedication), returning to baseline (precorticosteroid) levels at Day 8 and Day 15. CD40 expression on BDCA-2+ DC was seen to decrease slightly from baseline levels at Day 1, with a more substantial decrease by Day 8 before returning to baseline levels by Day 15 of each cycle. BDCA-3+ DC displayed upregulation of CD40 at Day 1 and Day 8, decreasing back to baseline levels at Day 15. However, none of the DC subsets investigated here exhibited a significant overall change in the levels of CD40 expression across 6 cycles of combined chemoimmunotherapy. HLA-DR expression was highly variable. Linear mixed modelling indicated that HLA-DR expression levels were also modulated in a cyclical fashion to some degree, and on BDCA-1+ DC showed a statistically significant, but minor, decrease between the start and end of treatment. Cyclical changes in HLA-DR expression on BDCA-2+ DC were minimal, and general HLA-DR levels as measured by MFI were lower than the other DC subsets. HLA-DR expression on BDCA-1+ DC showed a marked decrease at Day 1 of each cycle before rebounding, whereas BDCA-3+ DC displayed an increase in HLADR levels at Day 8 followed by a sustained decrease by Day 15 of each cycle. Flow cytometry data on DC subset representation and relative expression of functional markers underwent exploratory analyses for correlation with overall survival (OS), separately for each time point between baseline and Cycle 3 Day 8. For a full list of parameters analysed, McDonnell et al. BMC Cancer (2017) 17:417 Page 7 of 15 CD40 expression BDCA-1+ HLA-DR expression Average change -1047.04 over 19 time points, p=0.022 BDCA-1+ CD40 MFI HLA-DR MFI Average change -6.37 over 19 time points, p=0.341 B BDCA-2+ 1 2 3 4 5 B 6 Average change +6.62 over 19 time points, p=0.087 2 3 4 5 6 Average change -435.67 over 19 time points, p=0.072 HLA-DR MFI CD40 MFI BDCA-2+ 1 B BDCA-3+ 1 2 3 4 5 6 B BDCA-3+ 2 3 4 5 6 Average change -643.32 over 19 time points, p=0.124 CD40 MFI HLA-DR MFI Average change +2.45 over 19 time points, p=0.774 1 B 1 2 3 4 5 6 B 1 2 3 4 5 6 Fig. 2 Longitudinal flow cytometry data on DC across six cycles of chemoimmunotherapy, detailing changes in mean fluorescence intensity (MFI) relating to expression levels of CD40 or HLA-DR in BDCA-1, BDCA-2 and BDCA-3 DC subsets. Left-hand panels show observed values from individual patients, together with their empirical means (solid line). Right hand panels show results of fitting a linear mixed model; a linear trend over time and additive treatment effects of the day of the treatment yield the corresponding population average curves. (P-values: * <0.05, ** <0.01, *** <0.001, **** < 0.0001) see Table 2. The majority of analyses showed no significant differences in OS above vs below the median, however those patients who had higher than the median proportion of BDCA-2+ DC at baseline have longer OS (Fig. 3a), and patients who have higher than the median proportion of BDCA-3+ DC when CP-870,893 was delivered (Day 8) during Cycle 2 and Cycle 3 had poorer OS (Fig. 3b). treatment dependent CD4 depletion. With each additional cycle of chemoimmunotherapy the scale of CD4+ T cell depletion became more pronounced, but remained transient such that the balance between CD4 and CD8 always returned to baseline by Day 15 of each cycle (Fig. 4c). Stability of T cell number and subset distribution We further analysed the CD8+ T cell compartment for markers of proliferation and activation (see Fig. 5a for flow cytometry gating strategy). Proliferation was assessed by intracellular staining for Ki67, expressed in dividing and recently-divided cells [15]. Proliferation was seen to be highly cyclical, with the lowest proportion of Ki67+ cells consistently observed one week after chemotherapy (at Day 8) and the highest degree of proliferation one week after immunotherapy (at Day 15) for each cycle (Fig. 5b). The inducible co-stimulator molecule (ICOS), a member of the CD28 co-stimulator family, is expressed on activated T cells and is associated with antigen recognition [16, 17]. ICOS expression exhibited a similar pattern to Ki67, with a sharp decrease T cell concentrations per microlitre of peripheral blood were recorded for each sample prior to PBMC isolation (Fig. 4a). Variation in overall number of CD3+ T cells, as well as the proportion of CD4+ and CD8+ cells fluctuated in a repetitive manner each treatment cycle, with the number of T cells halving at Day 1, and returning to baseline levels at Day 8 and Day 15 (Fig. 4b). Treatment did not affect CD4+ and CD8+ numbers in an identical manner, with the CD8+ to CD4+ ratio becoming skewed in favour of CD8+ T cells at Day 1 of each cycle after dexamethasone and immediately prior to chemotherapy, and returning to around baseline levels by Day 8 and Day 15 of each cycle, a pattern consistent with a CD8+ T cell proliferating, activated and effector populations McDonnell et al. BMC Cancer (2017) 17:417 Page 8 of 15 Table 2 Data analysed for correlation with overall survival DC parameters T cell parameters DC (HLA-DR+ lin-) % of PBMC CD8 + % of CD3+ BDCA-1+ DC % of total PBMC Ki67+% of CD8 T cells BDCA-2+ DC % of total PBMC ICOS+% of CD8 T cells BDCA-3+ DC % of total PBMC Teff (HLA-DR+CD38+) % of CD8 T cells BDCA-1+ % of DC CD3+% of lymphocytes BDCA-2+ % of DC CD4+% of CD3+ BDCA-3+ % of DC Ki67+% of CD4 T cells BDCA-1+:BDCA-2+ ratio ICOS+% of CD4 T cells BDCA-1+:BDCA-3+ ratio Treg+% of CD4 T cells BDCA-2+:BDCA-3+ ratio ki67 + % of Treg (CD25+CD127loFoxp3+) BDCA-1+ DC CD40 MFI ICOS + % of Treg (CD25+CD127loFoxp3+) BDCA-1+ DC HLA-DR MFI BDCA-2+ DC CD40 MFI BDCA-2+ DC HLA-DR MFI BDCA-3+ DC CD40 MFI BDCA-3+ DC HLA-DR MFI List of parameters undergoing exploratory analysis for correlation with OS. Flow cytometry data was used to group patients above or below the median for each parameter, and repeated using data from each sample collection time point from baseline through to Cycle 3 Day 8 coinciding with blood samples collected one week following chemotherapy (Day 8) before returning to baseline levels (Fig. 5b). We also examined “effector” CD38hiHLA-DRhi CD8+ T cells, which are activated in an antigen-specific manner as previously reported from studies of chronic viral infection, and are present at elevated levels in cancers including mesothelioma compared with healthy controls [18–22]. These cells are predominantly proliferating, and exhibit low expression of Bcl-2 (an anti-apoptotic protein downregulated A BDCA-2+ DC% of PBMC Baseline p=0.010 B following antigen-specific T cell activation) [19, 23]. Our raw data and subsequent linear mixed modelling data again show a marked cyclical pattern over each cycle of chemoimmunotherapy treatment, with the effector proportion of CD8+ T cells peaking at Day 1 of each cycle after dexamethasone prior to falling just below baseline levels at Day 8 following chemotherapy (Fig. 5b). None of the CD8+ T cell parameters described above showed a significant change over the 6 cycles of chemoimmunotherapy. CD4+ T cell proliferating, activated and regulatory populations CD4+ T cells were also assessed by flow cytometry for overall proliferation (Ki67), activation (ICOS), and for the proportion of CD25+CD127loFoxp3+ regulatory T cells (Tregs) present within the CD4+ T cell compartment, using the gating strategy described in Fig. 6a. Overall, CD4+ T cells exhibited a broadly similar profile for fluctuations in proliferation and activation as CD8+ T cells, with significant reductions in the proportion of both Ki67+ and ICOS+ cells at Day 8 of each cycle, one week following chemotherapy (Fig. 6b). Analysis of Tregs also revealed a cyclical pattern, albeit less pronounced, with Tregs decreasing by approximately 20% between Day 1 and Day 8 of each treatment cycle followed by a return to around baseline levels a week later (Fig. 6b). There was no change over the six treatment cycles in any CD4+ T cell populations. Flow cytometry data on T cell subset representation and expression of activational markers underwent exploratory analysis for correlation with OS, separately for each time point between baseline and Cycle 3 Day 8. For a full list of T cell parameters analysed, see BDCA-3+ DC% of PBMC C2D8 C3D8 p=0.008 p=0.014 Above median Below median Fig. 3 Overall survival analysed by grouping patients above and below the median for (a) BDCA-2+ DC % of total PBMC at baseline, and (b) BDCA-3+ DC % of total PBMC at cycle 2 day 8 or cycle 3 day 8. Statistical significance calculated using Mantel Cox log rank test. Solid and dotted lines show data above and below the median values, respectively McDonnell et al. BMC Cancer (2017) 17:417 Page 9 of 15 Fig. 4 a Representative flow cytometry data showing gating strategy on whole blood samples used to obtain absolute volumetric cell count data. Lymphocytes were identified on the basis of FSC vs. SSC, with CD4+ or CD8+ T cells subsequently identified from within the CD3+ subset. b-c Longitudinal empirical data, linear mixed models and estimated means (left, centre and right-hand panels respectively) for: (b) absolute volumetric cell count data; (c) ratio of CD8+ T cells to CD4+ T cells (P-values: * <0.05, ** <0.01, *** <0.001, **** < 0.0001) Table 2. The majority of analyses showed no significant differences in OS above vs below the median, however those patients who had an ICOS+ % of CD8 + T cells above the median at baseline achieved better OS (Fig. 7). Both positive predictors of better OS in baseline samples, BDCA2+ DC% and ICOS + % of CD8+ T cells, were seen to correlate (r = 0.563, p = 0.029). Discussion It is becoming increasingly important to identify immune biomarkers in patients treated with the variety of McDonnell et al. BMC Cancer (2017) 17:417 Page 10 of 15 Fig. 5 a Representative flow cytometry data demonstrating the gating strategy used on PBMC for CD8+ T cells. FSC-area vs. FSC-height was used for doublet discrimination. A “dump” channel was then used to gate out dead cells (LIVE/DEAD fixable viability stain), CD14+ monocytes, and CD19+ B cells, and lymphocytes were selected by FSC vs. SSC. CD8+ T cells were subsequently selected on the basis of CD8 vs. CD3 staining, followed by the identification of proliferating (Ki67+) or activated (ICOS+) cells, and “effector CD8” cells as HLA-DR+CD38+. b Longitudinal empirical data, linear mixed models and estimated means (left, centre and right-hand panels respectively) for Ki67+ and ICOS+ CD8+ T cells, and HLA-DR+CD38+ effector CD8+ T cells (P-values: * <0.05, ** <0.01, *** <0.001, **** < 0.0001) immunotherapies currently being developed; both to assist with clinical decision making and to help understand the immunological basis of response, or lack thereof. Given that activating anti-CD40 has strong preclinical and early clinical evidence of efficacy, we undertook this study of systemic immune parameters to establish the pattern of changes induced by this agent in the context of chemotherapy, and to undertake an exploratory analysis of any correlation between these changes and patient outcomes. We recently published the results from a phase Ib clinical trial combining the anti-CD40 agonist antibody CP-870,893 with cisplatin/pemetrexed chemotherapy in patients with mesothelioma [9]. We