INFORMATION AND COMMUNICATION TECHNOLOGIES

(ICTS) FOR BIOBANKING AND ONCOLOGY RESEARCH

Analysis, Support Scenarios and a Case Study

Elena Sini

, Michele Torresani

, Silvia Veneroni

ICT,

Dept.of Experimental Oncology and Molecular Medicine

Fondazione IRCCS, Istituto Nazionale dei Tumori, Via Venezian 1, Milan, Italy

Paolo Locatelli, Nicola Restifo

Fondazione Politecnico di Milano, Piazza Leonardo da Vinci 32, Milan, Italy

Keywords: Tissue bank, Traceability, Radio frequency identification, Oncology research, Process reengineering.

Abstract: Human Tissue Banks are key for Oncological research and practice. Biobanking processes cross many care

departments and have a number of stakeholders, often carrying different objectives: quality assurance and

process efficiency are hard to garrison. Key issues in a biobanking project are: dedicated organization,

process control, completeness of clinical information on samples, integrated Information and

Communication Technologies. Fondazione IRCCS Istituto Nazionale dei Tumori is an oncologic research

and treatment institution in Milan (Italy). Our project started in October 2007 aiming at revising the whole

tissue collection process (from Surgery to Anatomical pathology assessment, to analysis and storage in the

Biobank), developing a clinical biobank management system collecting structured data on cases, and

designing an RFId-based system able to track the time- and temperature-sensitive specimens’ flow. Now

that go-live has begun, technological and - above all - organizational challenges of the project can be

discussed in detail. We hope other organizations will appreciate our efforts and are willing to apply a

biobanking network as soon as possible.

1 INTRODUCTION

Founded in 1925, the Fondazione IRCCS Istituto

Nazionale dei Tumori in Milan (henceforth: INT) is

recognized as a Scientific Research and Treatment

Institution in Oncology. Over 176 research projects

are currently under way, publishing nearly 400

scientific papers each year (IF 2272.32). In 2009

INT cared for about 14,000 inpatients, 10,000 day-

hospital admissions, 1 million outpatient treatments,

11,500 surgical operations (including 28 liver

transplants). It also inspired the Lombardy Oncology

Pathology Network (ROL).

This paper will describe our experience as

regards the development of a organizational and ICT

solutions for quality assurance, traceability and

operation support to biobanking of surgical samples

for experimental oncology research, sharing

challenges and results of our efforts.

2 EXPECTATIONS

AND CHALLENGES

IN BIOBANKING

Biosamples – e.g. pathologic and normal tissues,

blood, serum, nucleic acids – can play a vital role in

research that seeks to find new means of preventing,

diagnosing or treating cancer, mainly feeding in

vitro studies at in vivo conditions (e.g. on

biomarkers, molecular targets, biomolecular

characterization, genomics and proteomics),

minimizing research costs for future studies. Quality

assurance of sampling and processing activities

always worries researchers, e.g. because of possible

unknown biases on gene expression after tissue

devascularization (Spruessel, 2004). In order to

produce clinically valid and comparable results,

adequate case records, high-tech machinery,

standard sampling and testing procedures (SOPs) are

274

Sini E., Torresani M., Veneroni S., Locatelli P. and Restifo N..

INFORMATION AND COMMUNICATION TECHNOLOGIES (ICTS) FOR BIOBANKING AND ONCOLOGY RESEARCH - Analysis, Support Scenarios

and a Case Study.

DOI: 10.5220/0003161802740279

In Proceedings of the International Conference on Health Informatics (HEALTHINF-2011), pages 274-279

ISBN: 978-989-8425-34-8

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

paramount (Teodorovic, 2003, Bloom 2003).

We started analyzing the Italian and international

biobanking scenario (in Oncology above all), trying

to find organizational models and ICT solutions to

inspire our project. In general, we observed that

biobanks in other countries seem to be more

recognized as a valuable asset and institutionalized

within organizations. Biobanks are often connected

to research centers or universities, e.g. the IARC in

Lyon – France (www.iarc.fr), the Emory University

Hospital’s Winship Cancer Institute in Atlanta –

GA, USA (http://winshipcancer.emory.edu).

Moreover, in the majority of Italian cases,

biobanking is not an institutionalized process within

organizations and conceived as non-core by clinical

units. Sample collection is not systematic, most are

pathology-oriented, and often considered as own

property by physicians.

Further on, we looked for biobank information

systems and traceability technologies. The first thing

we found, is that DNA, semen, or blood banks, for

example, are more used in implementing

technologies for sample identification and biobank

management than tissue biobanks are. This gap is

even more relevant in private-run institutions.

Relevant cases of tissue banks show dedicated

management systems, integrated with local

laboratory equipment, which cover not only filing,

but also clinical information of patients and

specimens. For example, the biobank management

system developed by IBM at Karolinska Institutet in

Solna – Sweden (http://ki.se) is an excellence case

among those studied, also as regards the

implementation of syntactic and semantic standards

to gather data on clinical cases from the hospital

information system to the biobank. In Italy, we

found mainly local biobanks, where some cancer

centers run their own collections and poor support

systems (standalone software, or FileMaker®

archive..), which generally handle only storage

positions and basic clinical data. As regards

traceability, we found that, apart from hand-written

information on tubes, there is a basic use of barcode

labeling and only few interesting cases of Radio

Frequency Identification (RFId). In fact, RFId is

being recognized also in the healthcare sector as a

useful means to improve process safety and control.

Despite the growing number of implementation in

patient identification, internal logistics, clinical

operations (e.g. transfusion/drug administration

traceability – Vilamovska, 2009; School of

Management of the Politecnico di Milano 2005-

2009), there are still few implementations as regards

biobanking: on one hand this process requires quite

complex functionalities, on the other hand, a local

use of this technology only in the biobank may not

prove sustainable compared to barcoding. Major

issues are seamless integration within the sampling

process from surgery to anatomical pathology and to

the biobank, as well as operations at very low

temperatures. Istituto Ortopedico Rizzoli in Bologna

- Italy (www.btm.ior.it) traces bone tissues from the

biobank to the operatory theatre, where they get

updated with information on implanted patients.

Moreover, Paoli-Calmettes Institut in Marseilles –

France (www.institutpaolicalmettes.fr) and Mayo

Clinic in Rochester – MN, USA

(http://cancercenter.mayo.edu/mayo/research/bioba

nk) experimented in pilots that the use of high tech

RFId tags for freeze reading is still a capability to be

further developed.

Networking is another main issue, on one hand

to support the development of biobanks in smaller

institutions, on the other hand to exploit the value of

local biological assets joining international research

pipelines. In Italy focus is still on the setting up of

providers’ awareness, while foreign initiatives are

getting extensive: Tubafrost (www.tubafrost.org),

the Spanish National Tumor Bank Network

(www.cnio.es), the Wales Cancer Bank

(www.walescancerbank.com), EuroBioBank

(www.eurobiobank.org), the American NCI CaBIG

project (https://cabig.nci.nih.gov) and the Canadian

Tumor Repository Network (https://www.ctrnet.ca).

These projects follow different models, from the

creation of virtual collections to the centralization of

biobanking facilities.

Summing up, five key challenges can be

identified for present and future biobanking projects:

 Organization: institutionalization of the

biobanking process within organizations  process-

driven view, internal communication and

commitment.

 Quality assurance and process monitoring 

SOPs and traceability technologies. In particular,

RFId has innovative features like dynamic memory,

distance accessibility, bulk identification, embedded

sensors.

 Completeness of information on case profiles

merging data from clinical subsystems  data

coding and system integration implementing

semantic and syntactic standards (SNOMED, ICD9-

CM, ICD-10, EHR HL7, XML).

 Time-lasting preservation and stability 

adequate investments in laboratory instrumentation

and storage sites.

 Contributing to research  sample use protocols,

INFORMATION AND COMMUNICATION TECHNOLOGIES (ICTS) FOR BIOBANKING AND ONCOLOGY

RESEARCH - Analysis, Support Scenarios and a Case Study

275

syntactic/semantics, biobanking/pathology networks.

Web Rich Internet Applications (RIA) enable

sophisticated interfaces as in client/server systems,

but with a natural capability of supporting

networking and cooperation with external

organizations.

These subjects will now be addressed presenting the

real case of the Oncologic Tissue Bank at INT.

3 TISSUE BIOBANKING AT INT:

KEY ISSUES AND RESULTS

The project at the Italian National Cancer Institute in

Milan started in 2007 when the Experimental

Oncology Dept. discussed with the Chief

Information Officer (CIO) the opportunities of ICT

support for the local tissue biobank. The ICT Unit

and partner Fondazione Politecnico di Milano

started investigating processes, staff organization

and existing tools, highlighting critical points and

needs. Analysis highlighted a situation coherent with

the scenario discussed in Section 2: highly

fragmented way of working with poor support for

activities and low overall effectiveness (uncertain

quality assurance, high number of uncollected

cases…). Thus, a new targeted project was set up in

order to:

 Establish a real biobanking infrastructure;

 Investigate quality control procedures for tissue

processing and storing and review the organization

implementing a clearly structured biobanking

process;

 Ensure the exact identification of specimens;

 Guarantee traceability of operations and

transport lead times, monitoring specimens’

environmental conditions;

 Automate information flows between Depts.;

 Develop a shared scientific data-base integrated

with the hospital information system, where to

gather significant clinical information on patients

and data on specimens, to support diagnosis and

research.

A first help was in 2009 the opening of the

Amadeo-Lab, a new seat to concentrate research

units and labs, centralizing storage in a dedicated

infrastructure. Before, freezers were dispersed

among the 9 floors of the INT main building.

The solution we designed is mainly based on a

new Tissue Bank information system and the

extension of the INT RFId platform to support the

entire biobanking workflow (from surgery to freeze).

This is represented in Figure 1 (see numbers in

brackets). Inpatients at INT are always assigned an

RFId wristband at admission, where an

accompanying nurse stores information for its

unique identification [ref. 1]. This as INT is running

an own RFId traceability platform developed for

patient/staff identification, access management,

treatments safety and so on (Locatelli, 2010).

Operating Rooms daily schedule and real time status

are broadcasted via HL7 messages, so that

technicians in the biobank can plan activities and

signal surgeons which cases have a higher relevance.

Staff in the O.R. [1] checks-in the patient, verifying

its wristband vs. the room planning and their sheets

in the clinical system. During surgery [2], samples

of tumor or sane tissue are taken for diagnosis. Staff

can register and RFId-label sampled specimens on

the surgery system (e.g. site of sampling,

information for pathologists, notes for the Tissue

Bank..) while filling the digital surgery report on the

O.R. laptop. Samples are now ready to be sent to the

Anatomical Pathology labs: a clerk takes them out

and checks them in on the “Tissue Express” [3, 4], a

trolley with a plain RFId antenna and tablet PC

running a touch-screen software, where to declare

which new samples loaded and by whom (each staff

member has an RFId badge to log in to systems).

Dedicated staff deliver the trolley to the pathology

labs and check them in, while it tracks lead times

and surrounding temperatures thanks to a RFId

semi-active tag installed in the box containing the

samples. The software on the tablet PC collects its

transmission and joins them with data on carried

samples. These are checked in by pathologists

reading their labels on their laboratory system [5].

Once they have assessed a sample (macro-, micro-,

diagnosis), parts of it are given to a Tissue Bank

technician, who identifies it again reading the

original RFId label printed in the O.R [5]. The

received sample is recorded into the new Tissue

Bank information system. Alerts are shown in case

processing times or temperatures crossed preset

thresholds. The technician proceeds subdividing the

sample into aliquots and stocking them into vials,

each one labeled with a unique identifier and key

data printed into a Datamatrix 2D barcode. After

examinations and documentation on the system,

aliquots are frozen in the laboratory and periodically

transferred to the new biolab [6, 7].

The Tissue Bank information system routinely

collects relevant process and clinical data via HL7

and Web Services integration from the O.R. system,

the Anatomical Pathology, the slides digitalizer, the

HEALTHINF 2011 - International Conference on Health Informatics

276

RFId platform, the Enterprise Patient Registry, the

Enterprise Clinical and Reports Repository, and

from external documentation. All relevant data is

linked to the tissue collection and organized

enabling the user to browse collections per sample-,

patient-, or clinical data. The system manages

storage and sample retrieval, offering research and

browsing capabilities through rooms, freezers and

boxes containing samples, supporting technicians in

withdrawing cases and aliquots for research.

In our context, as probably in many other

research institutions, the biggest challenge we had to

face was to standardize sample-related data collected

in two heterogeneous historical databases: an

experimental stand-alone application for breast and

ovarian cancer cases and a MSExcel® archive for

other kind of samples. Basic data on samples was

written down on paper during handling and then

entered manually. No integrated system could ever

have been implemented with such a data structure.

The review took almost a year and required a tight

collaboration between the ICT Unit and clinical

staff. As results we obtained a common data

structure to be implemented, thus recovering all

historical collections. The new database was

designed to be more flexible and general as possible,

thus being able in the future to extend the

institutional biobank to other types of biological

material (e.g. blood, RNA, tissue microarray), and

also to join networks. The Tissue Bank information

system has been designed as a web-based Rich

Internet Application, allowing a very complete and

friendly user interface. Moreover, the three-tier

application architecture will enable us to easily scale

the system, simplifying maintenance and evolution

(data and application server are centralized, while

users access the latest version of the application via

web browser).

4 DISCUSSION:

TECHNOLOGY-RELATED

AND ORGANIZATIONAL OPEN

ISSUES

The solution has been available to users in

Experimental Oncology Labs since May 2010 for

user testing. Go live was October 2010, starting

from the Urology and Melanoma-Sarcoma

Operating Room, reaching full coverage of 10 ORs

at INT by Spring 2011.

To ease organizational change, we involved key

actors and final users in analysis and design phases:

since the beginning we engaged Operating Theater,

Anatomical Pathology and Tissue Bank referees in

order to approach the process from a global point of

view. Decision makers and operative referees were

committed to analyze processes, IT support,

documentation and information flows, critical

aspects, considering all points of view. Once As-Is

analyses were shared in focused meetings, discus-

Figure 1: Schema of the process and ICT solution supporting the new Tissue Bank at INT.

INFORMATION AND COMMUNICATION TECHNOLOGIES (ICTS) FOR BIOBANKING AND ONCOLOGY

RESEARCH - Analysis, Support Scenarios and a Case Study

277

sion followed on hypothesis about new ICT

solutions and process reengineering. Several

meetings were necessary to address issues like: what

type and how to use RFId, how to integrate the new

Tissue Bank system to the different Hospital

Information System (HIS) modules involved in the

biobanking process, which would be the tracking

steps in delivering samples, how to modify surgical

workflow in taking and recording tissue samples,

and so on. After defining specifications, we had to

coordinate all technology partners, one for each

system involved in the biobanking process. This

required a huge effort, to reconcile views and

garrison system integration.

Change management and implementation

activities have been running for almost a year, while

consensus building efforts accompanied the project

from the beginning. In fact, change management

issues were challenging, because of some

peculiarities common to many healthcare

projects._First of all, a process like biobanking

crosses at least three different care departments and

has a number of other stakeholders (Anesthetists,

Auxiliary Personnel, the Scientific Directorate, the

Ethics Office...), often carrying different priorities

and views of the process. Common to contexts like

public institutions were a certain resistance to

change while introducing a process-driven way of

thinking (instead of focusing on own clinical areas),

and a general low computer literacy. Internal project

management at INT was undertaken by the CIO (as

usually happens here for ICT projects, the ICT

Office takes leadership), with strong internal

commitment by INT top management. The CIO was

supported by a direct delegate and colleagues from

Fondazione Politecnico di Milano. Strong support

was required from the clinical area, so that clinical-

scientific issues could be taken into consideration

during design. The existence of previous successful

projects (e.g. RFId transfusion traceability, new

Surgery management system...) involving the same

roles and their referees helped to boost cooperation.

4.1 RFId Maturity – Can Healthcare

Organizations Face this Alone?

Positive experiences on using HF13,56 MHz RFId

technology for patient, operator and item

identification with near field applications (e.g. in the

transfusion chain) led us to extend the use of this

technology also to biobanking. We searched the

market for RFId solutions ready for biobanking, but

the only one meeting our needs was focused on vials

identification, too expensive, and from outside Italy

(with potential difficulties in customization and

integration activities). So, we evaluated how to

develop the extension on our own.

First of all, we learned RFId is not at all an “on-

the-shelf” technology. Even being supported by a

high-profile partner, many solutions had to be found

in an experimental way. Variety in implementation

of interoperability and communication standards by

producers of tags and devices, hardness to find

mature RFId handheld readers, unexpected

behaviour of devices and drivers instability, are

some of the main challenges to be faced. In fact, also

because of low experienced suppliers, we had to

work by a trial-and-error approach, often re-

designing integration components. This slowed up

system developments substantially.

Two key examples will help understanding this

issue. First, the trolley had to be designed and

produced with craftsmanship, while unexpected

interactions of the electromagnetic field with

samples required many modifications to obtain a

field with required characteristics such as shape and

strength. The second example comes from a request

by researchers to prove that RFId would not damage

samples. Once we started assessing literature

(among which: ICNIRP, 1998; Ahlbom, 2004;

Jauchem, 2008), we discovered that RFId

technology isn’t supported by a consolidated

environment due to lack of specific laws and

implementation guidelines for the healthcare sector:

studies on long-term consequences of biological

interactions connected to HF RFId fields have not

led to conclusive result yet. What we concluded after

scrutinizing a large number of papers, is that, given

the physical characteristics of tissues and the type of

RFId emissions used (short impulses, frequency,

power of few dozen mW), both short- and long-term

effects on tissues can be negligible. Besides, we

verified through several tests that the use of RFId

would not interfere with ordinary clinical activities

(Radiology, Radiotherapy..) and medical equipment.

Only tests done on infusion pumps led to a 5-10 cm

minimum distance requirement in RFId operations,

due to slight alteration in measured volumes in case

of repeated read/write activities.

Another issue was transport temperature

monitoring via semi-active or active tags: scouting

to find the right device with proper reliability and

battery duration was hard.

But the main critical point in using RFId for

biobanking is represented by extreme low storage

temperatures. Starting from -80°C of mechanic

freezers, tissues can be stocked in liquid nitrogen at -

196°C: standard RFId tags are not readable under -

30/-40°C. Specific extremely expensive RFId

solutions (vials with little ad-hoc button-size tags)

can improve reliability when sample is defrosted,

HEALTHINF 2011 - International Conference on Health Informatics

278

but with no significant enhancements for readability.

This were the main reasons why we decided – for

now- to switch to barcode for supporting sample

identification at storage and retrieval, relying on 2D

Datamatrix labels to code more data on each vial.

4.2 Expected Improvements

Technicians experience a remarkable rise in number

and quality of surgical specimens provided by the

operating theatre to the bank: before the bank was

able to collect specimens from less than 40% of

surgeries, often not knowing certain data about how

they have been processed before. State-of-the-art

facilities and tools will allow a more accurate

screening of incoming tissue samples, better support

high level research activities, and even to offer its

services (both storage space and sample provision)

to external organizations.

5 CONCLUSIONS

In 2007-2010 INT redesigned biobanking processes

from the operating theatre to storage, developing a

new Biobank Management System conceived as a

collector of all information flows about patients and

samples coming from clinical subsystems (Surgery,

Anatomical pathology, Laboratory...). This was also

integrated with the enterprise RFId platform to

identify samples, register process and transport lead

times, and to monitor specimens’ processing thanks

to readers at key steps of the process and trays

tracing environmental conditions on samples leaving

the operatory theatre. Systems are supporting

information flows all over processes, creating links

between applications and units that were not

cooperating at best and enabling process control and

quality, with an expected impact also on research.

The biobank management system was designed as a

flexible structure, and a web-based application able

to manage different biobanks and collections into a

virtual repository, the main feature of a network. A

first initiative of shared infrastructure has just started

with partner hospitals from a Regional research

group on Colon-Recto Cancer. Moreover, INT is

now trying to develop an extranet module and align

its systems to international guidelines in order to

apply international biobank networks (BBMRI...)

and offer its services to other institutions.

Challenges to be faced are very high. Innovation

in Public Healthcare meets a lot of difficulties in

being adopted due to strong organizational habits

and rules, as well as low computer literacy; rising

staff commitment towards a project/process

involving many different departments is a daily

commitment. Thus, change management issues and

accurate process redesign activities are key to

success. Introduction of innovation in such contexts

must be gradual (pilot projects), proceeding by

further refining and upgrades of systems. Challenges

were also technology-related. We realized that RFId

technology is not at all an “on-the-shelf” technology

and its implementation is not supported by a mature

environment due to lack of specific laws and

guidelines for the healthcare sector; this required

experimenting solutions and developing technology

skills on our own together with partners.

REFERENCES

Ahlbom A., Green A., Kheifets A., Savitz D., Swerdlow

A., 2004. ICNIRP - International Commission on

Non-Ionizing Radiation Protection, Standing

Committee on Epidemiology. Epidemiology of health

effects of radio frequency exposure, Environ. Health

Perspect, 2004; 112:1741-54.

Bloom G., Brower J., Clancy N., Eiseman E., Olmsted S.,

2003. Case studies of existing human tissue

repositories: Best practices for a biospecimen

resource for the genomic and postgenomic era. Santa

Monica, CA. RAND Corp.; www.Rand.Org/

publications/mg/mg120

ICNIRP - International Commission on Non-Ionizing

Radiation Protection, 1998. Guidelines for limiting

exposure to time-varying electric, magnetic and

electromagnetic fields (up to 300GHz), Health

Physics; 1998; 74(4).

Locatelli P., Restifo N., Facchini R., Sini E., Torresani M.,

July 2010. Strategic Planning of Identification

Technologies in Healthcare Processes: a Model and a

Real Case, regular research paper and presentation at

WORLDCOMP 2010, Intl. Conference on

Bioinformatics and Computational Biology, Las

Vegas, USA.

School of Management of the Politecnico di Milano,

Osservatorio RFID / Osservatorio Mobile&Wireless

Business. Reports 2005-2009. Politecnico di Milano,

Dipartimento di Ingegneria Gestionale, Milano, Italy.

Spruessel A, Steimann G, Jung M et al., 2004. Tissue

ischemia time affects gene and protein expression

patterns within minutes following surgical tumor

excision. BioTechniques. 2004;36:1030-7.

Teodorovic I, Therasse P, Spatz A, Isabelle M, Oosterhuis

W, 2003. Human tissue research: EORTC

recommendations on its practical consequences. Eur J

Cancer. 2003; 39:2256-2263.

Vilamovska AM, Hatziandreu E, Schindler HR, Van

Oranje-Nassau C, De Vries H, Krapels J, 2009. Study

on the requirements and options for RFID application

in healthcare, RAND Corporation.

INFORMATION AND COMMUNICATION TECHNOLOGIES (ICTS) FOR BIOBANKING AND ONCOLOGY

RESEARCH - Analysis, Support Scenarios and a Case Study

279