1
|
Cosma G, McArdle SE, Foulds GA, Hood SP, Reeder S, Johnson C, Khan MA, Pockley AG. Prostate Cancer: Early Detection and Assessing Clinical Risk Using Deep Machine Learning of High Dimensional Peripheral Blood Flow Cytometric Phenotyping Data. Front Immunol 2021; 12:786828. [PMID: 34975879 PMCID: PMC8716718 DOI: 10.3389/fimmu.2021.786828] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/30/2021] [Accepted: 11/11/2021] [Indexed: 11/16/2022] Open
Abstract
Detecting the presence of prostate cancer (PCa) and distinguishing low- or intermediate-risk disease from high-risk disease early, and without the need for potentially unnecessary invasive biopsies remains a significant clinical challenge. The aim of this study is to determine whether the T and B cell phenotypic features which we have previously identified as being able to distinguish between benign prostate disease and PCa in asymptomatic men having Prostate-Specific Antigen (PSA) levels < 20 ng/ml can also be used to detect the presence and clinical risk of PCa in a larger cohort of patients whose PSA levels ranged between 3 and 2617 ng/ml. The peripheral blood of 130 asymptomatic men having elevated Prostate-Specific Antigen (PSA) levels was immune profiled using multiparametric whole blood flow cytometry. Of these men, 42 were subsequently diagnosed as having benign prostate disease and 88 as having PCa on biopsy-based evidence. We built a bidirectional Long Short-Term Memory Deep Neural Network (biLSTM) model for detecting the presence of PCa in men which combined the previously-identified phenotypic features (CD8+CD45RA-CD27-CD28- (CD8+ Effector Memory cells), CD4+CD45RA-CD27-CD28- (CD4+ Effector Memory cells), CD4+CD45RA+CD27-CD28- (CD4+ Terminally Differentiated Effector Memory Cells re-expressing CD45RA), CD3-CD19+ (B cells), CD3+CD56+CD8+CD4+ (NKT cells) with Age. The performance of the PCa presence ‘detection’ model was: Acc: 86.79 ( ± 0.10), Sensitivity: 82.78% (± 0.15); Specificity: 95.83% (± 0.11) on the test set (test set that was not used during training and validation); AUC: 89.31% (± 0.07), ORP-FPR: 7.50% (± 0.20), ORP-TPR: 84.44% (± 0.14). A second biLSTM ‘risk’ model combined the immunophenotypic features with PSA to predict whether a patient with PCa has high-risk disease (defined by the D’Amico Risk Classification) achieved the following: Acc: 94.90% (± 6.29), Sensitivity: 92% (± 21.39); Specificity: 96.11 (± 0.00); AUC: 94.06% (± 10.69), ORP-FPR: 3.89% (± 0.00), ORP-TPR: 92% (± 21.39). The ORP-FPR for predicting the presence of PCa when combining FC+PSA was lower than that of PSA alone. This study demonstrates that AI approaches based on peripheral blood phenotyping profiles can distinguish between benign prostate disease and PCa and predict clinical risk in asymptomatic men having elevated PSA levels.
Collapse
Affiliation(s)
- Georgina Cosma
- Department of Computer Science, Loughborough University, Loughborough, United Kingdom
- *Correspondence: Georgina Cosma, ; A. Graham Pockley,
| | - Stéphanie E. McArdle
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
- Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Gemma A. Foulds
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
- Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Simon P. Hood
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Stephen Reeder
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
- Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Catherine Johnson
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
- Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Masood A. Khan
- Department of Urology, University Hospitals of Leicester National Health Service (NHS) Trust, Leicester, United Kingdom
| | - A. Graham Pockley
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
- Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
- *Correspondence: Georgina Cosma, ; A. Graham Pockley,
| |
Collapse
|
2
|
Dale JM, Hood SP, Bowen O, Bright H, Cutler KL, Berry N, Almond N, Goldin R, Karayiannis P, Rose NJ. Development of hepatic pathology in GBV-B-infected red-bellied tamarins (Saguinus labiatus). J Med Virol 2020; 92:3584-3595. [PMID: 32181899 DOI: 10.1002/jmv.25769] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/16/2019] [Accepted: 03/14/2020] [Indexed: 01/08/2023]
Abstract
GB virus B (GBV-B) is a new world monkey-associated flavivirus used to model acute hepatitis C virus (HCV) infection. Critical for evaluation of antiviral or vaccine approaches is an understanding of the effect of HCV on the liver at different stages of infection. In the absence of longitudinal human tissue samples at defined time points, we have characterized changes in tamarins. As early as 2 weeks post-infection histological changes were noticeable, and these were established in all animals by 6 weeks. Despite high levels of liver-associated viral RNA, there was reversal of hepatic damage on clearance of peripheral virus though fibrosis was demonstrated in four tamarins. Notably, viral RNA burden in the liver dropped to near undetectable or background levels in all animals which underwent a second viral challenge, highlighting the efficacy of the immune response in removing foci of replication in the liver. These data add to the knowledge of GBV-B infection in New World primates which can offer attractive systems for the testing of prophylactic and therapeutic treatments and the evaluation of their utility in preventing or reversing liver pathology.
Collapse
Affiliation(s)
- Jessica M Dale
- Division of Virology, National Institute for Biological Standards and Control, Medicine and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire, UK
| | - Simon P Hood
- Division of Virology, National Institute for Biological Standards and Control, Medicine and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire, UK
| | - Ori Bowen
- Division of Virology, National Institute for Biological Standards and Control, Medicine and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire, UK
| | - Helen Bright
- Internal Medicine Research Unit, Pfizer Research and Development, Sandwich, Kent, UK
| | - Keith L Cutler
- Division of Virology, National Institute for Biological Standards and Control, Medicine and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire, UK
| | - Neil Berry
- Division of Virology, National Institute for Biological Standards and Control, Medicine and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire, UK
| | - Neil Almond
- Division of Virology, National Institute for Biological Standards and Control, Medicine and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire, UK
| | - Robert Goldin
- Department of Cellular Pathology, Imperial College London, St. Mary's Campus, Norfolk Place, London, UK
| | - Peter Karayiannis
- Department of Medicine, Imperial College London, St. Mary's Campus, Norfolk Place, London, UK
| | - Nicola J Rose
- Division of Virology, National Institute for Biological Standards and Control, Medicine and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire, UK
| |
Collapse
|
3
|
Hood SP, Cosma G, Foulds GA, Johnson C, Reeder S, McArdle SE, Khan MA, Pockley AG. Identifying prostate cancer and its clinical risk in asymptomatic men using machine learning of high dimensional peripheral blood flow cytometric natural killer cell subset phenotyping data. eLife 2020; 9:50936. [PMID: 32717179 PMCID: PMC7386909 DOI: 10.7554/elife.50936] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/08/2019] [Accepted: 06/25/2020] [Indexed: 02/06/2023] Open
Abstract
We demonstrate that prostate cancer can be identified by flow cytometric profiling of blood immune cell subsets. Herein, we profiled natural killer (NK) cell subsets in the blood of 72 asymptomatic men with Prostate-Specific Antigen (PSA) levels < 20 ng ml-1, of whom 31 had benign disease (no cancer) and 41 had prostate cancer. Statistical and computational methods identified a panel of eight phenotypic features ([Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text]) that, when incorporated into an Ensemble machine learning prediction model, distinguished between the presence of benign prostate disease and prostate cancer. The machine learning model was then adapted to predict the D'Amico Risk Classification using data from 54 patients with prostate cancer and was shown to accurately differentiate between the presence of low-/intermediate-risk disease and high-risk disease without the need for additional clinical data. This simple blood test has the potential to transform prostate cancer diagnostics.
Collapse
Affiliation(s)
- Simon P Hood
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Georgina Cosma
- Department of Computer Science, Loughborough University, Loughborough, United Kingdom
| | - Gemma A Foulds
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom.,Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Catherine Johnson
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom.,Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Stephen Reeder
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom.,Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Stéphanie E McArdle
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom.,Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Masood A Khan
- Department of Urology, University Hospitals of Leicester NHS Trust, Leicester, United Kingdom
| | - A Graham Pockley
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom.,Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| |
Collapse
|
4
|
Sabry M, Zubiak A, Hood SP, Simmonds P, Arellano-Ballestero H, Cournoyer E, Mashar M, Pockley AG, Lowdell MW. Tumor- and cytokine-primed human natural killer cells exhibit distinct phenotypic and transcriptional signatures. PLoS One 2019; 14:e0218674. [PMID: 31242243 PMCID: PMC6594622 DOI: 10.1371/journal.pone.0218674] [Citation(s) in RCA: 21] [Impact Index Per Article: 4.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/30/2019] [Accepted: 06/06/2019] [Indexed: 11/19/2022] Open
Abstract
An emerging cellular immunotherapy for cancer is based on the cytolytic activity of natural killer (NK) cells against a wide range of tumors. Although in vitro activation, or “priming,” of NK cells by exposure to pro-inflammatory cytokines, such as interleukin (IL)-2, has been extensively studied, the biological consequences of NK cell activation in response to target cell interactions have not been thoroughly characterized. We investigated the consequences of co-incubation with K562, CTV-1, Daudi RPMI-8226, and MCF-7 tumor cell lines on the phenotype, cytokine expression profile, and transcriptome of human NK cells. We observe the downregulation of several activation receptors including CD16, CD62L, C-X-C chemokine receptor (CXCR)-4, natural killer group 2 member D (NKG2D), DNAX accessory molecule (DNAM)-1, and NKp46 following tumor-priming. Although this NK cell phenotype is typically associated with NK cell dysfunction in cancer, we reveal the upregulation of NK cell activation markers, such as CD69 and CD25; secretion of pro-inflammatory cytokines, including macrophage inflammatory proteins (MIP-1) α /β and IL-1β/6/8; and overexpression of numerous genes associated with enhanced NK cell cytotoxicity and immunomodulatory functions, such as FAS, TNFSF10, MAPK11, TNF, and IFNG. Thus, it appears that tumor-mediated ligation of receptors on NK cells may induce a primed state which may or may not lead to full triggering of the lytic or cytokine secreting machinery. Key signaling molecules exclusively affected by tumor-priming include MAP2K3, MARCKSL1, STAT5A, and TNFAIP3, which are specifically associated with NK cell cytotoxicity against tumor targets. Collectively, these findings help define the phenotypic and transcriptional signature of NK cells following their encounters with tumor cells, independent of cytokine stimulation, and provide insight into tumor-specific NK cell responses to inform the transition toward harnessing the therapeutic potential of NK cells in cancer.
Collapse
Affiliation(s)
- May Sabry
- Department of Haematology, University College London, London, United Kingdom
| | - Agnieszka Zubiak
- Department of Haematology, University College London, London, United Kingdom
| | - Simon P. Hood
- John van Geest Cancer Research Centre, Nottingham Trent University, Nottingham, United Kingdom
| | - Poppy Simmonds
- Department of Haematology, University College London, London, United Kingdom
| | | | - Eily Cournoyer
- Department of Haematology, University College London, London, United Kingdom
| | - Meghavi Mashar
- Department of Haematology, University College London, London, United Kingdom
| | - A. Graham Pockley
- John van Geest Cancer Research Centre, Nottingham Trent University, Nottingham, United Kingdom
| | - Mark W. Lowdell
- Department of Haematology, University College London, London, United Kingdom
- InmuneBio Inc., La Jolla, California, United States of America
- * E-mail:
| |
Collapse
|
5
|
Hood SP, Foulds GA, Imrie H, Reeder S, McArdle SEB, Khan M, Pockley AG. Phenotype and Function of Activated Natural Killer Cells From Patients With Prostate Cancer: Patient-Dependent Responses to Priming and IL-2 Activation. Front Immunol 2019; 9:3169. [PMID: 30761160 PMCID: PMC6362408 DOI: 10.3389/fimmu.2018.03169] [Citation(s) in RCA: 16] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/29/2018] [Accepted: 12/24/2018] [Indexed: 12/27/2022] Open
Abstract
Background: Although immunotherapy has emerged as the “next generation” of cancer treatments, it has not yet been shown to be successful in the treatment of patients with prostate cancer, for whom therapeutic options remain limited to radiotherapy and androgen (hormone) deprivation therapy. Previous studies have shown that priming natural killer (NK) cells isolated from healthy individuals via co-incubation with CTV-1 cells derived from an acute lymphoblastic leukemia (ALL) enhances their cytotoxicity against human DU145 (metastatic) prostate cancer cells, but it remains unknown to what extent NK cells from patients with prostate cancer can be triggered to kill. Herein, we explore the phenotype of peripheral blood NK cells in patients with prostate cancer and compare the capacity of CTV-1 cell-mediated priming and IL-2 stimulation to trigger NK cell-mediated killing of the human PC3 (metastatic) prostate cancer cell line. Methods: The phenotype of resting, primed (co-incubation with CTV-1 cells for 17 h) and IL-2 activated (100 IU/ml IL-2 for 17 h) NK cells isolated from frozen-thawed peripheral blood mononuclear cell (PBMC) preparations from patients with benign disease (n = 6) and prostate cancer (n = 18) and their cytotoxicity against PC3 and K562 cells was determined by flow cytometry. Relationship(s) between NK cell phenotypic features and cytotoxic potential were interrogated using Spearman Rank correlation matrices. Results and Conclusions: NK cell priming and IL-2 activation of patient-derived NK cells resulted in similar levels of cytotoxicity, but distinct NK cell phenotypes. Importantly, the capacity of priming and IL-2 stimulation to trigger cytotoxicity was patient-dependent and mutually exclusive, in that NK cells from ~50% of patients preferentially responded to priming whereas NK cells from the remaining patients preferentially responded to cytokine stimulation. In addition to providing more insight into the biology of primed and cytokine-stimulated NK cells, this study supports the use of autologous NK cell-based immunotherapies for the treatment of prostate cancer. However, our findings also indicate that patients will need to be stratified according to their potential responsiveness to individual therapeutic approaches.
Collapse
Affiliation(s)
- Simon P Hood
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Gemma A Foulds
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Heather Imrie
- School of Animal Rural and Environmental Sciences, Nottingham Trent University, Nottingham, United Kingdom
| | - Stephen Reeder
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Stéphanie E B McArdle
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| | - Masood Khan
- Department of Urology, University Hospitals of Leicester NHS Trust, Leicester, United Kingdom
| | - Alan Graham Pockley
- John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, United Kingdom
| |
Collapse
|
6
|
Hood SP, Mee ET, Perkins H, Bowen O, Dale JM, Almond NM, Karayiannis P, Bright H, Berry NJ, Rose NJ. Changes in immune cell populations in the periphery and liver of GBV-B-infected and convalescent tamarins (Saguinus labiatus). Virus Res 2013; 179:93-101. [PMID: 24246306 PMCID: PMC3969288 DOI: 10.1016/j.virusres.2013.11.006] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/05/2013] [Revised: 11/03/2013] [Accepted: 11/07/2013] [Indexed: 01/09/2023]
Abstract
GBV-B infection of tamarins is a valuable model for acute HCV infection. We observed distinct expression patterns of PD-1, a marker of T-cell activation, on peripheral and hepatic lymphocytes. Differential PD-1 expression is coincident with reduction in peripheral GBV-B. Liver-associated viral RNA in the absence of peripheral viraemia indicates maintenance of occult infection.
Flaviviruses related to hepatitis C virus (HCV) in suitable animal models may provide further insight into the role that cellular immunity contributes to spontaneous clearance of HCV. We characterised changes in lymphocyte populations in tamarins with an acute GBV-B infection, a hepatitis virus of the flaviviridae. Major immune cell populations were monitored in peripheral and intra-hepatic lymphocytes at high viraemia or following a period when peripheral virus was no longer detected. Limited changes in major lymphocyte populations were apparent during high viraemia; however, the proportions of CD3+ lymphocytes decreased and CD20+ lymphocytes increased once peripheral viraemia became undetectable. Intrahepatic lymphocyte populations increased at both time points post-infection. Distinct expression patterns of PD-1, a marker of T-cell activation, were observed on peripheral and hepatic lymphocytes; notably there was elevated PD-1 expression on hepatic CD4+ T-cells during high viraemia, suggesting an activated phenotype, which decreased following clearance of peripheral viraemia. At times when peripheral vRNA was not detected, suggesting viral clearance, we were able to readily detect GBV-B RNA in the liver, indicative of long-term virus replication. This study is the first description of changes in lymphocyte populations during GBV-B infection of tamarins and provides a foundation for more detailed investigations of the responses that contribute to the control of GBV-B infection.
Collapse
Affiliation(s)
- Simon P Hood
- Division of Virology, National Institute for Biological Standards and Control, Medicines and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK
| | - Edward T Mee
- Division of Virology, National Institute for Biological Standards and Control, Medicines and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK
| | - Hannah Perkins
- Internal Medicine Research Unit, Pfizer Research and Development, Sandwich, Kent CT13 9NJ, UK
| | - Ori Bowen
- Division of Virology, National Institute for Biological Standards and Control, Medicines and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK
| | - Jessica M Dale
- Division of Virology, National Institute for Biological Standards and Control, Medicines and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK
| | - Neil M Almond
- Division of Virology, National Institute for Biological Standards and Control, Medicines and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK
| | - Peter Karayiannis
- Hepatology and Gastroenterology Section, Department of Medicine, Imperial College London, Variety Wing Floor D, St. Mary's Campus, Norfolk Place, London W2 1PG, UK
| | - Helen Bright
- Internal Medicine Research Unit, Pfizer Research and Development, Sandwich, Kent CT13 9NJ, UK
| | - Neil J Berry
- Division of Virology, National Institute for Biological Standards and Control, Medicines and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK.
| | - Nicola J Rose
- Division of Virology, National Institute for Biological Standards and Control, Medicines and Healthcare products Regulatory Agency, Blanche Lane, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK.
| |
Collapse
|