Please wait a minute...

Frontiers in Biology

Front Biol    2011, Vol. 6 Issue (4) : 263-273
Advances in medical decision support systems for diagnosis of acute graft-versus-host disease: molecular and computational intelligence joint approaches
Maurizio FIASCHé1,2(), Maria CUZZOLA2, Giuseppe IRRERA2, Pasquale IACOPINO3, Francesco Carlo MORABITO1
1. DIMET, University “Mediterranea” of Reggio Calabria, Italy; 2. Transplant Regional Center of Stem Cells and Cellular Therapy, “A. Neri” , Reggio Calabria, Italy; 3. School of Hematology, University of Messina, Italy
Download: PDF(320 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Acute graft-versus-host disease (aGVHD) is a serious systemic complication of allogeneic hematopoietic stem cell transplantation (HSCT) causing considerable morbidity and mortality. Acute GVHD occurs when alloreactive donor-derived T cells recognize host-recipient antigens as foreign. These trigger a complex multiphase process that ultimately results in apoptotic injury in target organs. The early events leading to GVHD seem to occur very soon, presumably within hours from the graft infusion. Therefore, when the first signs of aGVHD clinically manifest, the disease has been ongoing for several days at the cellular level, and the inflammatory cytokine cascade is fully activated. So, it comes as no surprise that progress in treatment based on clinical diagnosis of aGVHD has been limited in the past 30 years. It is likely that a pre-emptive strategy using systemic high-dose corticosteroids as early as possible could improve the outcome of aGVHD. Due to the deleterious effects of such treatment particularly in terms of infection risk posed by systemic steroid administration in a population that is already immune-suppressed, it is critical to identify biomarker signatures for approaching this very complex task. Some research groups have begun addressing this issue through molecular and proteomic analyses, combining these approaches with computational intelligence techniques, with the specific aim of facilitating the identification of diagnostic biomarkers in aGVHD. In this review, we focus on the aGVHD scenario and on the more recent state-of-the-art. We also attempt to give an overview of the classical and novel techniques proposed as medical decision support system for the diagnosis of GVHD.

Keywords computational intelligence      gene selection      GVHD      machine learning      personalized modelling      wrapper     
Corresponding Author(s): FIASCHé Maurizio,   
Issue Date: 01 August 2011
 Cite this article:   
Maurizio FIASCHé,Maria CUZZOLA,Giuseppe IRRERA, et al. Advances in medical decision support systems for diagnosis of acute graft-versus-host disease: molecular and computational intelligence joint approaches[J]. Front Biol, 2011, 6(4): 263-273.
E-mail this article
E-mail Alert
Articles by authors
Maurizio FIASCHé
Giuseppe IRRERA
Francesco Carlo MORABITO
Fig.1  A diagram of personalized modeling-based gene selection method (PMGS).
Fig.2  Three phases of GVHD immuno-pathogenesis.
Gene nameOfficial full nameImmune function
BCL2A1BCL2-related protein A1Anti- and pro-apoptotic regulator.
CASP1°*Caspase 1, apoptosis-related cysteine peptidaseCentral role in the execution-phase of cell apoptosis.
CCL7Chemokine (C-C motif) ligand 7Substrate of matrix metalloproteinase 2
CD83CD83 moleculeDendritic cells regulation.
CXCL10°Chemokine (C-X-C motif) ligand 10Pleiotropic effects, including stimulation of monocytes, natural killer and T cell migration, and modulation of adhesion molecule expression.
EGR2°Early growth response 2Transcription factor with three tandem C2H2-type zinc fingers.
FASTNF receptor superfamily, member 6Central role in the physiologic regulation of programmed cell death.
ICOS°*Inducible T cell co-stimulatorPlays an important role in cell-cell signaling, immune responses, and regulation of cell proliferation.
IL4Interleukin 4Immune regulation.
IL10°*Interleukin 10Immune regulation.
SELPSelectin PCorrelation with endothelial cells.
SLPI°Stomatin (EPB72)-like 1Elemental activities such as catalysis.
STAT6Transducer and activator of transcription 6, interleukin-4 inducedRegulation of IL4- mediated biological responses.
Foxp-3*Forkhead box P3Regulatory T cells play important roles in the maintenance control of transplantation tolerance.
CD52°*CD52 antigenB cell activation.
Tab.1  The 13 genes selected from CFS (correlation-based feature selection) with their names and meaning; the 7 genes selected through the wrapper-Na?ve Bayes method () are marked with °; the 5 genes selected with SVM are marked with *
Fig.3  The profile of sample 7 in GVHD data set.
MethodTraining setTest set
PMGS-na?ve Bayes27(29)29(30)
Tab.2  Experimental results of a CFS with ANN classifier and a wrapper method combined with SVM and of the PMGS with Na?ve Bayes and with WKNN
MethodTest set (Dtst)
PMGS-Na?ve Bayes3(7)
(i)PMGS- Na?ve Bayes6(7)
(i)PMGS- WKNN6(7)
Tab.3  Experimental results of a CFS with ANN classifier and a wrapper method combined with SVM with PMGS
1 Ayuk F, Diyachenko G, Zabelina T, Panse J, Wolschke C, Eiermann T, Binder T, Fehse B, Erttmann R, Kabisch H, Bacher U, Kr?ger N, Zander A R (2008). Anti-thymocyte globulin overcomes the negative impact of HLA mismatching in transplantation from unrelated donors. Exp Hematol , 36(8): 1047–1054
doi: 10.1016/j.exphem.2008.03.011 pmid:18456390
2 Bacigalupo A, Lamparelli T, Milone G, Sormani M P, Ciceri F, Peccatori J, Locasciulli A, Majolino I, Di Bartolomeo P, Mazza F, Sacchi N, Pollicheni S, Pinto V, Van Lint M T, the Gruppo Italiano Trapianto Midollo Osseo (GITMO) (2010). Pre-emptive treatment of acute GVHD: a randomized multicenter trial of rabbit anti-thymocyte globulin, given on day+7 after alternative donor transplants. Bone Marrow Transplant , 45(2): 385–391
doi: 10.1038/bmt.2009.151 pmid:19584823
3 Beilhack A, Schulz S, Baker J, Beilhack G F, Wieland C B, Herman E I, Baker E M, Cao Y A, Contag C H, Negrin R S (2005). In vivo analyses of early events in acute graft-versus-host disease reveal sequential infiltration of T-cell subsets. Blood , 106(3): 1113–1122
doi: 10.1182/blood-2005-02-0509 pmid:15855275
4 Brodoefel H, Bethge W, Vogel M, Fenchel M, Faul C, Wehrmann M, Claussen C, Horger M (2010). Early and late-onset acute GVHD following hematopoietic cell transplantation: CT features of gastrointestinal involvement with clinical and pathological correlation. Eur J Radiol , 73(3): 594–600
doi: 10.1016/j.ejrad.2009.01.011 pmid:19201121
5 Buzzeo M P, Yang J, Casella G, Reddy V (2008). A preliminary gene expression profile of acute graft-versus-host disease. Cell Transplant , 17(5): 489–494
doi: 10.3727/096368908785096042 pmid:18714668
6 Chaussabel D, Pascual V, Banchereau J (2010). Assessing the human immune system through blood transcriptomics. BMC Biol , 8(1): 84–98
doi: 10.1186/1741-7007-8-84 pmid:20619006
7 Ferrara J L (2008). Advances in the clinical management of GVHD. Best Pract Res Clin Haematol , 21(4): 677–682
doi: 10.1016/j.beha.2008.07.003 pmid:19041607
8 Ferrara J L, Levine J E, Reddy P, Holler E (2009). Graft-versus-host disease. Lancet , 373(9674): 1550–1561
doi: 10.1016/S0140-6736(09)60237-3 pmid:19282026
9 Ferrara J L, Reddy P (2006). Pathophysiology of graft-versus-host disease. Semin Hematol , 43(1): 3–10
doi: 10.1053/j.seminhematol.2005.09.001 pmid:16412784
10 Fiasché M (2010). Implementations of Evolving Integrated Multimodel Systems, Algorithms and Applications in Biomedical Field. DissertationTip. DIMET, University “Mediterranea” of Reggio Calabria
11 Fiasché M, Cuzzola M, Fedele R, Iacopino P, Morabito F C (2010a). Machine Learning and Personalized Modelling based Gene Selection for acute GvHD Gene Expression Data Analysis. In: Artificial Neural Networks — proceedings of ICANN 2010 Part I, LNCS 6352
12 Fiasché M, Cuzzola M, Iacopino P, Kasabov N, Morabito F C (2010b). Personalized Modelling based Gene Selection for acute GvHD Gene Expression Data Analysis: a Computational Framework Proposed. Australian Journal of Intellignet Information Processing Systems, 12(4): Machine Iearning Applications (Part II)
13 Fiasché M, Verma A, Cuzzola M, Iacopino P, Kasabov N, Morabito F C (2009). Discovering Diagnostic Gene Targets and Early Diagnosis of Acute GVHD Using Methods of Computational Intelligence over Gene Expression Data. In: Artificial Neural Networks — ICANN 2009. Part II, LNCS 5769/2009 , Berlin: Springer Heidelberg, pp. 10–19
14 Firoz B F, Lee S J, Nghiem P, Qureshi A A (2006). Role of skin biopsy to confirm suspected acute graft-vs-host disease: results of decision analysis. Arch Dermatol , 142(2): 175–182
doi: 10.1001/archderm.142.2.175 pmid:16490845
15 Foley J E, Mariotti J, Ryan K, Eckhaus M, Fowler D H (2008). Th2 cell therapy of established acute graft-versus-host disease requires IL-4 and IL-10 and is abrogated by IL-2 or host-type antigen-presenting cells. Biol Blood Marrow Transplant , 14(9): 959–972
doi: 10.1016/j.bbmt.2008.06.007 pmid:18721759
16 Fujimura J, Takeda K, Kaduka Y, Saito M, Akiba H, Yagita H, Yamashiro Y, Shimizu T, Okumura K (2010). Contribution of B7RP-1/ICOS co-stimulation to lethal acute GVHD. Pediatr Transplant , 14(4): 540–548
doi: 10.1111/j.1399-3046.2009.01279.x pmid:20136724
17 Guyon I, Elisseeff A (2003). An introduction to variable and feature selection. J Mach Learn Res , 3(7-8): 1157–1182
doi: 10.1162/153244303322753616
18 Harik G R, Lobo F G, Goldberg D E (1999). The compact genetic algorithm. IEEE Trans Evol Comput , 3(4): 287–297
doi: 10.1109/4235.797971
19 Ho V T, Cutler C (2008). Current and novel therapies in acute GVHD. Best Pract Res Clin Haematol , 21(2): 223–237
doi: 10.1016/j.beha.2008.02.009 pmid:18503988
20 Hu Y, Song Q, Kasabov N (2009). Personalized modelling based gene selection for microarray data analysis. In: K?ppen M, Kasabov N, Coghill G (eds.). ICONIP 2008, Part I. LNCS. Springer, Heidelberg , vol. 5506, pp. 1221–1228
21 Jacobsohn D A, Vogelsang G B (2007). Acute graft versus host disease. Orphanet J Rare Dis , 2(1): 35–44
doi: 10.1186/1750-1172-2-35 pmid:17784964
22 Lord J D, Hackman R C, Gooley T A, Wood B L, Moklebust A C, Hockenbery D M, Steinbach G, Ziegler S F, McDonald G B (2011). Blood and gastric FOXP3+ T cells are not decreased in human gastric graft versus host disease. Biol Blood Marrow Transplant , 17(4): 486–496
23 Kasabov N (2007a). Evolving Connectionist Systems: The Knowledge Engineering Approach, 2nd ed. Springer, London
24 Kasabov N (2007b). Global, local and personalized modelling and profile discovery in bioinformatics: An integrated approach. Pattern Recognit Lett , 28(6): 673–685
doi: 10.1016/j.patrec.2006.08.007
25 Kasabov N (2010). To spike or not to spike: A probabilistic spiking neural model. Neural Networks , 23(1): 16–19
26 Kato K, Cui S, Kuick R, Mineishi S, Hexner E, Ferrara J L M, Emerson S G, Zhang Y (2010). Identification of stem cell transcriptional programs normally expressed in embryonic and neural stem cells in alloreactive CD8+ T cells mediating graft-versus-host disease. Biol Blood Marrow Transplant , 16(6): 751–771
doi: 10.1016/j.bbmt.2010.01.012 pmid:20116439
27 Langley P (1994). Selection of relevant features in machine learning, In: Proceedings of AAAI Fall Symposium on Relevance , 140–144
28 Magenau J M, Qin X, Tawara I, Rogers C E, Kitko C, Schlough M, Bickley D, Braun T M, Jang P S, Lowler K P, Jones D M, Choi S W, Reddy P, Mineishi S, Levine J E, Ferrara J L, Paczesny S (2010). Frequency of CD4+CD25(hi)FOXP3+ regulatory T cells has diagnostic and prognostic value as a biomarker for acute graft-versus-host-disease. Biol Blood Marrow Transplant , 16(7): 907–914
doi: 10.1016/j.bbmt.2010.02.026 pmid:20302964
29 Miura Y, Thoburn C J, Bright E C, Phelps M L, Shin T, Matsui E C, Matsui W H, Arai S, Fuchs E J, Vogelsang G B, Jones R J, Hess A D (2004). Association of Foxp3 regulatory gene expression with graft-versus-host disease. Blood , 104(7): 2187–2193
doi: 10.1182/blood-2004-03-1040 pmid:15172973
30 Mohty M, Gaugler B, Faucher C, Sainty D, Lafage-Pochitaloff M, Vey N, Bouabdallah R, Arnoulet C, Gastaut JA, Viret F, Wolfers J, Maraninchi D, Blaise D, Olive D (2002). Recovery of lymphocyte and dendritic cell subsets following reduced intensity allogeneic bone marrow transplantation. Hematology , 7(3): 157–64
31 Paczesny S, Braun T M, Levine J E, Hogan J, Crawford J, Coffing B, Olsen S, Choi S W, Wang H, Faca V, Pitteri S, Zhang Q, Chin A, Kitko C, Mineishi S, Yanik G, Peres E, Hanauer D, Wang Y, Reddy P, Hanash S, Ferrara J L M (2010a). Elafin is a biomarker of graft-versus-host disease of the skin. Sci Transl Med , 2(13): ra2
doi: 10.1126/scitranslmed.3000406 pmid:20371463
32 Paczesny S, Hanauer D, Sun Y, Reddy P (2010b). New perspectives on the biology of acute GVHD. Bone Marrow Transplant , 45(1): 1–11
doi: 10.1038/bmt.2009.328 pmid:19946340
33 Paczesny S, Krijanovski O I, Braun T M, Choi S W, Clouthier S G, Kuick R, Misek D E, Cooke K R, Kitko C L, Weyand A, Bickley D, Jones D, Whitfield J, Reddy P, Levine J E, Hanash S M, Ferrara J L M (2009a). A biomarker panel for acute graft-versus-host disease. Blood , 113(2): 273–278
doi: 10.1182/blood-2008-07-167098 pmid:18832652
34 Paczesny S, Levine J E, Braun T M, Ferrara J L M (2009b). Plasma biomarkers in graft-versus-host disease: a new era? Biol Blood Marrow Transplant , 15(Suppl): 33–38
doi: 10.1016/j.bbmt.2008.10.027 pmid:19147075
35 Payne D K, Sullivan M D, Massie M J (1996). Women’s psychological reactions to breast cancer. Semin Oncol , 23(Suppl 2): 89–97
36 Piper K P, Horlock C, Curnow S J, Arrazi J, Nicholls S, Mahendra P, Craddock C, Moss P A H (2007). CXCL10-CXCR3 interactions play an important role in the pathogenesis of acute graft-versus-host disease in the skin following allogeneic stem-cell transplantation. Blood , 110(12): 3827–3832
doi: 10.1182/blood-2006-12-061408 pmid:17766680
37 Platt J (1998). Fast training of support vector machines using sequential minimal optimization. Advances in Kernel Methods–Support Vector Learning. MIT Press
38 Schub N, Günther A, Schrauder A, Claviez Aβ, Ehlert C,βGramatzki M,βRepp R (2011). Therapy of steroid-refractory acute GVHD with CD52 antibody alemtuzumab is effective. Bone Marrow Tranplant , 46: 143–147
39 Socié G (2009). Graft-versus-host disease: proteomics comes of age. Blood , 113(2): 271–272
doi: 10.1182/blood-2008-10-185991 pmid:19131558
40 Socié G, Blazar B R (2009). Acute graft-versus-host disease: from the bench to the bedside. Blood , 114(20): 4327–4336
doi: 10.1182/blood-2009-06-204669 pmid:19713461
41 Song Q, Kasabov N (2006). TWNFI — Transductive weighted neuro-fuzzy inference system and applications for personalized modelling. Neural Netw , 19(10): 159–596
doi: 10.1016/j.neunet.2006.05.028
42 van Bekkum D W, Vries De M J (1967). Radiation Chimaeras, London: Logos Press
43 Varadi G, Or R, Slavin S, Nagler A (1996). In vivo CAMPATH-1 monoclonal antibodies: a novel mode of therapy for acute graft-versus-host disease. Am J Hematol , 52(3): 236–237
doi: 10.1002/(SICI)1096-8652(199607)52:3<236::AID-AJH23>3.0.CO;2-C pmid:8756101
44 Verma A, Fiasché M, Cuzzola M, Iacopino P, Morabito F C, Kasabov N (2009). Ontology Based Personalized Modelling for Type 2 Diabetes Risk Analysis: An Integrated Approach. In: Leung C S, Lee M, Chan J H (Eds.): ICONIP 2009, Part II, LNCS 5864, Springer, Heidelberg , pp. 360–366
45 Weissinger E M, Schiffer E, Hertenstein B, Ferrara J L, Holler E, Stadler M, Kolb H J, Zander A, Zürbig P, Kellmann M, Ganser A (2007). Proteomic patterns predict acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation. Blood , 109(12): 5511–5519
doi: 10.1182/blood-2007-01-069757 pmid:17339419
46 Wolf D, Wolf A M, Fong D, Rumpold H, Strasak A, Clausen J, Nachbaur D (2007). Regulatory T-cells in the graft and the risk of acute graft-versus-host disease after allogeneic stem cell transplantation. Transplantation , 83(8): 1107–1113
doi: 10.1097/ pmid:17452902
47 Yu X Z, Liang Y, Nurieva R I, Guo F, Anasetti C, Dong C (2006). Opposing effects of ICOS on graft-versus-host disease mediated by CD4 and CD8 T cells. J Immunol , 176(12): 7394–7401
Full text