GATiB

http://www.univie.ac.at/LSG/gatib/

Genome Austria Tissue Bank - GATiB

One of the largest collections of diseased and normal tissue worldwide will be developed into a biological resource center specifically designed to support the needs of systems biology approaches to human diseases and of personalized medicine.

Biobanks containing human biological samples, such as tissues, blood or body fluids as well as related data are essential resources for the establishment of the function and medical relevance of human genes. Biobanks containing both high quality normal and diseased human tissues are particularly valuable since they contain information on the genetic and epigenetic alterations as well as on modification of gene products that caused a disease and influenced its outcome. Large tissue collections provide insights into the great variability of human diseases and of responses to medical treatment and, therefore, provide an essential basis for the advancement of personalized medicine. To date, biobank development worldwide has focused on blood samples whereas tissue collections have been established only in a fragmented manner, resulting in tissue banks of variable size, composition, standards and with different goals. In recognition of the limitations of the current stand-alone biobank model, the establishment of international networks of bio(tissue)banks have been recently assigned a very high strategic priority, not only to cover the emerging demands for such resources but also to increase efficacy in medical genomics and to reduce research costs.
We plan to develop, in an internationally oriented context, one of the largest collections of diseased human tissues already established at the Medical University of Graz to qualify as an OECD Biological Resource Centre specifically designed to support the needs of systems biology approaches to human diseases, of drug discovery, and public health. Since human tissues are a very limited resource and to minimise the international distribution of human samples, special emphasis will be placed on well-coordinated analysis of samples, allowing in the future the distribution of high quality tissue-derived data rather than original tissues themselves. Key components of GATiB will comprise (i) archival tissue samples associated with long-term follow up and medical data, (ii) prospectively collected tissue and blood samples associated with standardized information on the patient's disease and environmental exposure, (iii) animal models molecularly validated for their human disease relevance, and (iv) IT-tools supporting sample tracking, data storage, data mining and protecting sample donor privacy. The development of GATiB will be guided by experiences obtained from research projects performed in the field of breast cancer and metabolic liver diseases and by investigation of major biobank projects in Europe, USA, Asia, and Australia. Furthermore we will carefully consider the social, ethical and political issues related to GATiB to ensure proper embedding of this important resource at the national and international levels in society.

 


 

The basic design of GATiB

Name Institution Main contribution Funding 
T. Bauernhofer

Dept. of Internal
Medicine, Division of
Oncology, MedUG 

Medical data for cancers
GEN-AU II 
A. Berghold Inst. of Med.
Informatics, Statistics
and Documentation,
MedUG
Medical data  
G. Casari  ORIDIS Biomed TMA technology GEN-AU II 
H. Denk Inst of Pathology,
MedUG 
Tissues, metabolic liver
diseases, animal models  
GEN-AU II 
J. Eder Department of
Knowledge and Business
Engineering, University
of Vienna  
IT tools (Data MART) GEN-AU II  
H. GottweisDept. of Political
Science, Univ. of
Vienna 
Ethical, legal and social
guidance 
GEN-AU II 
F. Iberer Clinical Dept. of
Surgery, Division for
Transplantation,
MedUG 
  
HJ Mischinger Dept. of Surgery,
MedUG 
Various diseases of
abdominal organs 
 
H. Müller Inst. of Pathology,
MedUG 
Data visualization FWF VIPEM
FFG GENOPTIKUM
H. Samonigg
Dept. of Internal
Medicine, Division of
Oncology, MedUG
Medical data for cancers
GEN-AU II  
W. Schippinger Dept. of Internal
Medicine, Division of
Oncology, MedUG 
Medical data for cancers GEN-AU II  
D. SchmalstiegInst. of Computer
Graphics and Vision,
TU Graz
 
Data visualization
FWF VIPEM
FFG GENOPTIKUM
K. Sargsyan,
A. Tiran
ZMFCore Facility Biobank
 
Z. Trajanoski
Inst. of Genomics and
Bioinformatics,
TU Graz
Data bases
CD-Laboratory
M Trauner
Dept. of Medicine,
MedUG
Liver diseases, medical
data, animal models
GEN-AU I
GEN-AU II
K. Zatloukal
Inst. of Pathology,
MedUG
Strategic design and
coordination
GEN-AU I
GEN-AU II
EU BBMRI


Principal investigators contributing to BioResource-Med. MedUG (Medical University of Graz), TU Graz (Graz University of Technology)

 

 

 Key components of BioResource-Med

Since human tissues are a very limited resource and to minimise the international distribution of human samples, special emphasis will be placed on well-coordinated analysis of samples, allowing in the future the distribution of high quality tissue-derived data rather than original tissues themselves.
The development of BioResource-Med will be guided by experiences obtained from investigation of major biobanks in Europe, USA, Asia and Australia and experiences gained from specific research projects relying on utilization of the resource performed in the fields of metabolic liver diseases and cancer. Our research will systematically take into account the complex national and international socio-ethical contexts in which BioRecource-Med operates.

Key components of BioResource-Med comprise:

  • archival tissue samples associated with long-term follow up and medical data,
  • prospectively collected tissue and blood samples associated with standardized information on the patient's disease and environmental exposure,
  • animal models molecularly validated for their human disease relevance, and
  • IT-tools supporting sample tracking, data storage, data mining and protecting sample donor privacy.

Represented population

BioResource-Med is a combination of a population-based and disease-focused collection of biological materials. It currently contains approx. 2.9 mio samples from 888,000 patients representing a non-selected patient group characteristic for central Europe. Since the Institute of Pathology was until 2003 the almost exclusive pathology service provider for major parts of the province of Styria including its capital Graz (population of approx. 1.2 mio people), samples from all human diseases, which were treated by surgery or diagnosed by biopsy, are included in the collection at their natural frequency of occurrence and thus represent cancers and non-cancerous diseases from all organs, and from children to aged people.
The scientific value of the already existing tissue collection is, therefore, not only characterized by its size and its technical homogeneity (all samples have been processed in one Institute under constant conditions since more than 20 years), but also by its population-based character. These features provide ideal opportunities for epidemiological studies and allow the validation of biomarkers for the identification of specific diseases and the response to treatment regimes.
Prospectively collected tissues, blood samples and clinical data will on the one hand comprise randomly selected samples from all diseases and patient groups to have sufficient numbers of samples for the evaluation of the disease-specificity of any gene or biomarker. On the other hand in selected diseases (such as breast, colon and liver cancers as well as metabolic diseases) all available samples will be collected.

 

 Population-based vs. disease-focused biobanks

Samples and data

 Paraffin-embedded
material

Cryo-conserved
material
 
Patients
 888.000 8.090
Organs
 1.750.000 11.100
Blocks/vials
 3.016.639 18.800


Overview on sample composition of BioRecource-Med (status June 2006)

Organ
Patients
stomach201.400 
colon 157.300 
breast 30.600 
ovary 28.800 
small intestine 27.800 
prostate
26.600 
thyroid gland
19.200 
bone marrow
19.700 
liver
17.600 
oesophagus
17.200 
lymph node
17.400 
bone
16.300 
lung
15.400 
muskel10.700 
kidney
7.700 
urinary bladder7.600 
CNS
6.200 
heart
4.800 
pancreas
2.100 
other organs
328.900 
total
963.300


Overview on organ representation in BioResource-Med (status June 2006)

Medical information generally associated with samples of BioResource-Med:

  • Disease diagnosis
  • Full histopathological report including classification, grading and staging of tumours
  • State of the art immunohistochemical characterisation
  • State of the art molecular genetic characterization

Survival Data

Available for 63,000 samples from 24,000 patients
In addition for 6,000 of these patients detailed autopsy data are available

Clinical Data

Full medical records are available for 1,100,000 samples from 470,000 patients, although information is required to be entered in a searchable database. Already electronically accessible are medical records of 2,000 patients.

IT-Infrastructure

Each sample is associated with many pieces of information and data. Some of them are direct characteristics of the biological material (organ, tissue, preparation, donor sex, donor age, tumour stage, immunohistochemical characterisation...) other data describe the medical relevance and history (diagnosis, treatment history, life style data) and others just the manipulation (storage place, performed analysis). Management of data and samples is enabled by an IT- infrastructure that supports collection, administration and retrieval. A data mart was designed to enable complex queries combining genetic data with detailed medical information, while protecting sample donor privacy by preventing re-identification of individual patients.

 

 

Schematic diagram of the IT infrastructure for BioResource-Med. Several original sources of relevant information (Sample DB, clinical DB) are accessed directly and all data stay at their original location. Knowledge about where to find and how to access it is kept in a MetaData repository and managed at the integration layer. The scientific workbench offers a single access interface to the anonymised information for researchers that need to search, access, and analyze data

Stark K, Eder J, and Zatloukal K. (2006)
Priority-Based k-Anonymity Accomplished by Weighted Generalisation Structures, Lecture Notes in Computer Science, Vol. 4081, 394 - 404.