The Relapse Prediction in Childhood Acute Leukemia
Data Analysis and Modeling the Relapse Predict for Children with Acute Leukemias Based on Prognostic Risk Factors to Enforce Efficacy of Therapy
Tech Area / Field
- INF-SOF/Software/Information and Communications
3 Approved without Funding
Belarusian Research Center for Pediatric Oncology and Hematology, Belarus, Minsk
- National Academy of Sciences of the Republic of Belarus / Institute of Informatics Problems, Belarus, Minsk
- University of Calgary / Department of Electrical and Computer Engineering, Canada, AB, Calgary\nMedizinische Hochschule Hannover, Germany, Hannover\nKyushu University / Faculty of Medicine / School of Health Sciences, Japan, Fukuoka\nVU Medisch Centrum, The Netherlands, Amsterdam
Project summaryThe aim of the proposed project is to prevent relapses in childhood acute leukemia due to developing an information analytical system of modeling relapse prediction based on the analysis of patient-related and tumor-related risk factors in combination with minimal residual disease detection.
Problem description: Acute leukemia (AL) is the most common form of childhood cancer (Steliarova-Foucher E, et al, 2004). Despite the progress advances in treatment of childhood acute leukemia there still remains a 30 percent risk of relapse in acute lymphoblastic leukemia (ALL) and more than half of patients have relapses in acute myeloid leukemia (AML) (Harms D., 2000, Pui C.H., 2004, Hoelzer D., et al., 2004, Crutzig U., et al., 2005, Kaspers G., et al., 2005). In contrast to primary disease results of relapse treatment are very poor (Aleinikova O., et al., 2004, Kaspers G., et al., 2005). Therefore the treatment of primary ALL and AML should be effective at most to prevent relapses. Recent systems for risk group stratification in childhood ALL and AML include mostly patient-related prognostic risk factors. The relapse-associated prognostic significance of many biological features of leukemic cells are still obscure because many studies have been either provided with retrospective univariate analyses or have involved a small number of patients. One of the ways to increase the treatment effectiveness is to reveal the new relapse-associated risk factors both at the time of diagnosis and during the antileukemic treatment so an adequate patient stratification can be applied to enforce the therapy.
This proposal is necessitated for the following reasons: (i) the modern therapy of acute leukemia in children is very effective in induction and maintenance of remission, but the risk of relapses still remains; (ii) conventionally used prognostic risk factors are not good enough at predicting the relapse development; (iii) the significance of minimal residual disease and most of biological features of leukemic cells in relapse prognosis is still obscure despite the fact that a number of new methods for their research have been recently developed; (iv) inpidualization of the treatment is still under consideration; (v) there is no conventional mathematical models at the moment that could be used for relapse prediction in acute leukemia in children.
The traditional uni- and multivariate analysis are not enough for modeling relapse prediction (Tom I., 2002; Novoselova N., Krasko O. et al, 2003) because of dynamics of combinations prognostic risk factors: some of them appear only on various treatment stages. Besides the well-know statistical methods like multivariate Cox-regression used for data analysis in leukemia, the intelligent soft computing methods will be developed and used for elaboration of the range and significance of prognostic risk factors combinations for treatment result and relapse rate. The basis of intelligent soft computing tools will be the neuro-fuzzy models (Jang J.S., Sun C.T., Mizutani, 1997; Nauck D., Kruse R., 1999), genetic algorithms (Freitas A., 2004; Kwedlo W., Kretowski M., 1999) and fuzzy clustering techniques (Abonyi J. and Szeifert F., 2003; Vytchenin D., 2004; Verbruggen H.B., Babuska R., 1998) They will be used to develop new efficient models for prediction and classification relapse for children with AL. These models are preferred for their possibility to obtain interpretable results with reasonable tradeoff between simplicity and accuracy.
The effect of the project will be of high international value for its new approach to increase the treatment effectiveness in children with AL based on relapse prediction. Created information analytical system and criteria for risk group stratification, obtained after project fulfillment, will both enable to enforce therapy and decrease the relapse rate in childhood acute leukemia, and be the base for developing an inpidualized AL treatment. Moreover, current knowledge in the biology of tumor cells will be improved.
Project participants: The team from Belarusian Research Center for Pediatric Oncology & Hematology (BRCPOH) consists of 18 highly qualified medical doctors and doctors of bilogical sciences including 9 persons with PhD degree and two – with Professor, D.Med.Sc. degree. The team from United Institute of Informatics Problems of the National Academy of Sciences of Belarus (UIIP NASB) consists of 10 experienced informatic engeneers including 2 persons with PhD degree and one – with Professor, D.Sc. degree. BRCPOH and UIIP NASB have already gained experience in organizing and accomplishing of international projects related to medical data analyses (ISTC B-522, B-736, B-517, and B-323).
The proposed project is referred to the technology area – BIO-PAB/INF-SOF and category of technology development – Applied Research/Technology Development according the ISTC classifier.
- Extended database of the following data for prognostic risk factors determination in children with acute leukemia: minimal residual disease during and after treatment; patient-referred and leukemia cells related parameters at time of diagnosis.
- Prospective definition of the prognostic significance of patient-related and tumor-related features, and minimal residual disease at different time points of therapy protocol at large number of patients receiving similar treatment for the purposes of relapse prediction in childhood ALL and AML based on newly developed technique of definition.
- The information analytical system (including the software and the electronic register database for relapse prediction) applicated in pediatric oncology and hematology clinical centers. The methods of data analysis for modeling the relapse prediction.
- The treatment strategy based on the new stratification in childhood ALL and AML for patients’ risk group definition to decrease relapse rate.
The aim of the project will be reached due to enforcing and inpidualization of the treatment in childhood acute leukemia based on developed analytical system of relapse prediction.
The rights for intellectual property that are generated during the course of the project will be regulated by the laws of the Republic of Belarus and by the procedures, which have been developed by the ISTC.
Expected Results Application. Project’s results will be applied in Belarus-Russia Study Group for patients’ risk group definition in childhood ALL and AML treatment protocols, and recommended for other pediatric acute leukemia study groups. The information analytical system with newly developed methods for modeling the relapse prediction will be tested in clinics which are using other AL treatment protocols, and will be presented as -version of commercialized product. The book on prognostic significance of patient-referred and tumor-related risk factors in childhood acute leukemia will be published.
Meeting ISTC Goals and Objectives:
- Providing 12 weapon scientists and engineers who are among the main participators of the project (more than 55%) and were involved in fundamental exploring and designing of biological weapons (alternative employment in this project is to search for a new data in biology of childhood leukemia) or in researches of data collection and handling for missile guidance (alternative employment in this project is to develop and to apply information analytical system for medical purposes).
- Promoting integration of the participants of the project from BRCPOH and UIIP NASB into international scientific community: participation in the international meetings and workshops, joint publications, project work coordination, results verifications.
- Supporting basic and applied research and technology development for peaceful purposes in field of healthcare.
- Contributing to the solution of national (Belarus-Russia) and international problem – reduction of the relapse rate in childhood acute leukemia.
- Reinforcing the transition to market economies responsive to civil needs by commercializing main project results.
Scope of Activities. Project duration: 36 months. Participants: 28 specialists and 3 technical assistants (total effort – 10057 person*days). The proposed project includes four interrelated tasks 1, 2, 3, 4 performed as partly parallel stages: Task 1 – 4519 person*days; deliverables: confidential manual, nondisclosure agreements, report on registration form, user’s guide, article, protocol of the test results, publications, meetings; Task 2 - 2141 person*days; deliverables: report, publications, meeting, articles, technique; Task 3 - 2291 person*days; deliverables: report, user’s guide, publications, meetings; Task 4 - 1106 person*days; deliverables: conclusive report, book, articles, publications.
Role of Foreign Collaborators/Partners:
Prof., M.D., PhD Karl Heinrich Welte (Kinderklinik, Medizinische Hochschule Hannover Germany): to provide comments to the technical reports; to test the created database software and stratification system, developed in the course of the project; to conduct join seminars and workshops and also to coordinate the task of selection patient-related prognostic risk factors, including MRD detection.
Prof., M.D., Ph.D. Akinobu Matsuzaki (Division of Child Health, School of Health Sciences, Faculty of Medicine, Kyushu University, Japan): to conduct information exchange in the course of project implementation for ALL; to provide comments to the technical reports, to coordinate the task of selection relaps-related prognostic risk factors.
M.D., Ph.D., Assoc. Prof. Gertjian J.L. Kaspers (Department of Pediatric Oncology/Hematology VU University Medical Center, the Netherlands): to participate in monitoring of project activities for tasks 1 and 4; to conduct information exchange in AML; to provide comments to the technical reports. Methodological support for investigation of biological features of tumor cells.
Assoc. Prof., D.Sc. Svetlana Yanushkevich (Department of Electrical and Computer Engineering, University of Calgary, Canada): to participate in monitoring of project activities for tasks 2 and 3; to conduct information exchange in information technologies; to provide comments to the technical reports.
Technical Approach and Methodology. The participants have the scientific and technical resources as following: 1) in addition to ordinary used patient-related and tumor-associated parameters, BRCPOH provides the investigation and collection of complex of leukemic cells related parameters: immunophenotype, multidrug resistance, spontaneous apoptosis and proliferation in vitro, aberrant expression of tumor suppressor genes, expression of chimeric oncogenes. Data on immunophenotyping, leukemic cell proliferation, apoptosis and expression of MDR proteins will be estimated by flow cytometry. Oncogenes/antioncogenes will be detected by PCR–based methodology. Drug resistance will be evaluated with in vitro culturing of leukemic cells, MRD detection – by flow cytometry and real-time PCR. Minimal residual disease and its reduction rate will be studied during and after treatment of AL. 2) UIIP NASB provides the complex of methods and models for processing and intelligent analysis of data, including neuro-fuzzy modeling, fuzzy clustering and genetic algorithms for patient stratification and relapse prediction based on medical investigation at BRCPOH. Appropriate software will be developed. This software will be integrated in information-analytical system (IAS) on PC platform operating in Windows 2000/XP environment. IAS will be based on “client – Web-server – DB server” technology and will include the electronic register in the form of the relational database (DB); software for the intelligent data analysis of mixed data from the electronic register.
The International Science and Technology Center (ISTC) is an intergovernmental organization connecting scientists from Kazakhstan, Armenia, Tajikistan, Kyrgyzstan, and Georgia with their peers and research organizations in the EU, Japan, Republic of Korea, Norway and the United States.
ISTC facilitates international science projects and assists the global scientific and business community to source and engage with CIS and Georgian institutes that develop or possess an excellence of scientific know-how.