(Editor's note: Full-length version of story.)
Better understanding of the interaction between cancer genes and the
environment is the goal of a research project launched by the
Maine Institute for Human Genetics and Health and two of the leading
research institutions in the state - the University of Maine and
Jackson Laboratory. At the heart of the initiative is what researchers
hope is a new cancer-fighting tool, the Biobank of Maine, to
explore the relationship between the disease and its geographic,
demographic, health and environmental variables.
"We will ask questions such as why breast cancer rates are low in
Aroostook County while all rates of cancers are highest in Washington
County," says Janet Hock, director of the Maine Institute for Human
Genetics and Health, headquartered in Brewer. "It's not simply that
you live in an area with high radon or lead in the water. It's far more
complex. It's the interplay between genes and the environment.
We're going beyond asking whether variations of the genes make people
more susceptible, to asking what is their interaction with the
In 2007, the Maine Institute for Human Genetics and Health launched the
Biobank of Maine with a $1.8 million award from the
Department of Defense. The goal is to investigate the effects of rural
environments on human genetics and health to find solutions to
reducing the high rates of cancer in Maine.
In Maine, lung, breast, prostate and colorectal cancers predominate, and
melanoma rates are rising.
The biobank has two core components - a cancer tissue repository managed
by the Maine Institute for Human Genetics and Health,
and a complex geographic information system (GIS) database developed by
the University of Maine and Jackson Laboratory.
Both adhere to the strictest of confidentiality guidelines as set by the
National Cancer Institute and the International Society for
Biological and Environmental Repositories.
Nov. 1, the first of the tissue samples will be deposited in the biobank.
Patients consenting to have their postsurgical blood and tissue
samples archived in the repository will complete a confidential
geospatial questionnaire. On it they will be asked to provide
about their behavioral, occupational, residential and healthcare
The demographic information will be entered in the secure database known
as the Maine Cancer Geographic Information System or
MeCancerGIS. The database will integrate four basic types of data:
geographic, demographic, health and environmental. Publicly
available databases on geography, environment and occupational exposures
will be included, as will geospatial and temporal information
on environmental exposures in Maine, such as radon and arsenic in water
and bedrock, and herbicide and pesticide applications.
The goal is to provide a spatiotemporal perspective on the state's
incidence of cancer. Such detection and monitoring of cancer
incidence and mortality in space and time in Maine could help identify
regions where there are higher-than-expected cancer rates and
could aid investigations of the connections between specific
environmental variables and cancer.
The spatiotemporal database of risk factors for cancer also has the
potential to be a model for mapping patterns in other diseases.
"Maine Cancer GIS will map incidences of cancer and provide tools to
investigate interactions with environmental, socioeconomic and
demographic factors," says University of Maine Professor of Spatial
Information Science and Engineering Kate Beard-Tisdale, who is
collaborating with Jackson Laboratory bioinformatics scientist Carol
Bult to develop the database. "Through representation of the data,
we can generate many different ways to look at patterns and subsets.
Does where people live or have lived make any difference? Are
there spatial structures that can help diagnose how disease evolves?"
This summer, Beard-Tisdale and Bult focused on the intricate protocols
for the database. Graduate student Paul Smitherman, a research
assistant in the Department of Spatial Information Science and
Engineering, focused on the hardware configurations.
"There's a huge set of parameters to plug in to the statistical engine
in order to take a lot of inputs to output specific results," says
Smitherman, whose 2006-07 study of radon in more than 1,000 wells in the
Augusta area will be included in thedatabase. "The creation of the logical model of how to store data
spatially and temporally is a huge challenge. That actual, logical
structure of the database will be a groundbreaking research area."
The incorporation of GIS as an analytical tool for the Biobank of Maine
is "a unique marriage," says Hock. Other biobanks across the
country record some geocode information like the patient's last known
address, but Maine will be the first to comprehensively compile
the gene-environment interactions that change the risk of cancer.
The research initiative is being watched closely by other states.
Already, Hock is discussing partnerships to share resources with other
biobanks, including one in New Brunswick, the province with the highest
cancer rate in Canada.
The hope is to find out how the environment influences the genes, ultimately informing the development of treatments to target the risk
factors. Physicians could one day treat cancer patients by looking at
both their medical and environmental histories. Clinics could be
established in areas with populations at higher risk, helping overcome
socioeconomic boundaries in rural states like Maine.
Such a multifactor inquiry could shed light on questions like why some breast cancer patients go into remission and others do not, or
why the incidence of lung cancer in nonsmoking women is rising.
"There's potential richness in combining genetics with socio-economic,
environmental and geographic factors," says Beard-Tisdale, who
directs the National Center for Geographic Information and Analysis at
UMaine. "This data gives us the possibility of isolating which
factors pose greatest risk."
"The hope is that by building a baseline that is perpetuated, we can
create a model to tease apart the complexity of the different
potential risk factors to make a diagnosis about cancers. We hope the database helps us address some of the questions we have about
whether risk factors are working in isolation or in combinations of ways
we don't yet understand."
by Margaret Nagle
for more stories from the current issue of UMaine Today Magazine.