National Cancer Institute

at the National Institutes of Health

Understanding Cancer Series

Posted: 10/20/2009

Slide 1

Single-Gene Tests: Local Search

Single-gene tests focus on a specific location in a patient's genome. Using this approach, scientists have looked for single genes linked to cancer. This research has revealed some important discoveries such as gene changes called mutations located within the BRCA1 or BRCA2 genes that may confer a significantly increased risk of breast and ovarian cancer. And some single-gene tests continue to inform treatment decisions. For example, colon cancer patients may be tested for a K-ras mutation, or breast cancer patients for HER2 gene activity, before treatment is planned.

A geographical analogy helps depict the scale of the search involved in single-gene tests.

Graphic shows a globe and a continent and a country and a U.S., all appear left to right in order of their decreasing size. Below this row appears a cell , a chromosome, a gene, and then a few chemical bases, also in order of decreasing size. Next a magnifying glass shows that a single gene test is like looking at only the smallest end of the scale, the chemical bases.

Slide 2

Genome-Wide Profiling: Global Search

Single-gene tests alone are unable to completely unravel the complexity of cancer. Because cancer involves the simultaneous interaction of many different mutated genes and proteins within malignant cells and their surrounding tissues, a global approach is needed to capture all the activity. Scientists have developed new methods to cast a genome-wide search and look globally for all the changes in DNA, RNA, or proteins* that contribute to cancer's progression. These new approaches are collectively called genome-wide profiling.

Again, a geographical analogy shows the global scale of the search involved in genome-wide profiling.

(*To learn more basic information about DNA, RNA, and proteins in relation to cancer, please visit http://www.nci.nih.gov/cancertopics/understandingcancer/moleculardiagnostics/)

Graphic again shows a globe and a continent and a country and a U.S., all geographic items are of decreasing size, below this row appears a cell, a chromosome, a gene, and then a few chemical bases, also all of decreasing size. Next a magnifying glass shows that genome-wide profiling is like looking at only the largest end of the scale, the genome.

Slide 3

Starts With Proper Collection and Storage

Genome-wide profiling requires proper sample collection and storage. This is why biopsy specimens must be collected, stored, tracked, and used according to the highest scientific and ethical standards.

Informed consent for use of cancerous biopsy specimens (or a waiver of this consent) must be in place at the time of collection.

Samples must be given an electronic, unique identifier that links the specimen to a patient, to a treatment protocol, and to the informed consent (or waiver).

Samples must be stored and retrieved following best clinical practices.

When all this is in order, a researcher can use the biopsy sample to get a genome-wide profile of the cancer.

Graphic shows that proper care of a tissue sample. It shows a clinician getting informed consent from the patient before removing a tissue sample, removing sample, adding barcode identifier to the container holding the sample, storing sample carefully in liquid nitrogen, carefully thawing the sample before analyzing it, and carefully refreezing any remaining tissue sample.

Slide 4

Genome-Wide Profiling: Many Ways

Genome-wide profiling refers to the collection of large-scale information about the sequence and expression of the genes in normal or tumor cells. Techniques include:

Comparative Genomic Hybridization and Spectral Karyotyping
looking at global changes in chromosome structure when cancer cells are compared to normal ones from the same patient
Polymorphism Analysis
looking at genetic variations called polymorphisms in individuals and in populations (These specific altered regions or sequences are spread across the entire genome.)
Gene Expression Profiling
capturing total gene activities, both increases and decreases, across a genome as patterns of gene expression

Most of these techniques use high-throughput technologies, which are automated systems that can perform rapid analysis of large numbers of samples. Data generated through genome-wide profiling are sometimes called genomic signatures or genomic profiles.

Graphic shows four clinicians all holding different technology used for genome-wide profiling of cancer tissue. First holds a set of chromosomes each with a different color attached. This lets the researchers see two colors where genetic material has been transferred between chromosomes. The second holds markers for genetic variants called polymorphisms, the third holds a microarray used to capture changes in gene expression, and the fourth holds a slide with test DNA attached and with labeled probes for both normal and tumor DNA being poured over the test sample.

Slide 5

Genome-Wide Profiling: Comparative Genomic Hybridization

Comparative genomic hybridization (CGH) is a method that allows researchers to detect large-scale changes in chromosomes. They can visualize where there is extra genetic material (repeating copies), deleted material, or regions where large portions of genetic material exchange between chromosomes. Many cancers exhibit these types of chromosomal abnormalities.

Graphic shows a closeup of a slide holding all the chromosome pairs present in a cell that is dividing. Probes from normal DNA are tagged red and those from tumor DNA are tagged green. Where the probes bind equally, a chromosome on the slide appears yellow. Where there is loss of tumor DNA the chromosome appears red because more probe from normal DNA binds. Finally, where the chromosome gains tumor DNA, the color is green because more probe from tumor DNA binds.

Slide 6

Genome-Wide Profiling: Karyotyping

Spectral karyotyping is a method that allows researchers to use probes that assign a different color spectrum to each chromosome, making it possible to see where genetic material has been added or exchanged within a cancer genome.

Graphic shows that human chromosomes are first labeled so that each one has a separate color, then, after 72 hours of interacting with one another on a slide, a camera is used to see if any genetic material has exchanged. A chromosome appears that bears two colors, so genetic material has been exchanged.

Slide 7

Genome-Wide Profiling: Polymorphism Analysis

Another approach is called polymorphism analysis. Polymorphism analysis is used to find variation in genetic sequence across the genome; some tests look for variation in single base pairs (single-nucleotide polymorphisms), whereas others look for tandem repeats (simple sequence repeats) or insertions of several telltale bases (restriction-fragment length polymorphisms).

Once sequence polymorphisms that are potentially associated with cancer are identified, these must be validated through further research in independent populations. Polymorphism data will be used in conjunction with other genetic, genomic, and clinical information.

Graphic shows little flags projecting from coiled DNA and a zoom from the flags shows that the different color of flag represents a different change that can occurred in the genetic material. A red flag means that a chemical base has switched from the common one to a less common one at a particular spot in the DNA. A blue flag means that some chemical bases are missing from their usual spot in the DNA. And a purple flag means that extra bases have been inserted into the DNA.

Slide 8

Genome-Wide Profiling: Gene Expression Profiles

Researchers use gene expression profiling to study the activity of genes in a patient's tumor sample and in surrounding tissue. This approach measures levels of messenger RNA (mRNA), which is a good indicator of the proteins about to be made by the cell.

An individual's messenger RNA is processed and labeled and then applied to a specially designed chip, often referred to as a microarray, which includes complementary sequences for hundreds of thousands of genes. A chip reader measures levels of the expressed genes in the sample.

Other techniques used to assess gene activity monitor levels of genome-wide DNA methylation, which can silence gene activities, or levels of proteins already made in the cell, or the chemical properties of some proteins (e.g., phosphorylation status).

Graphic shows a clinician looking at a pathology slide and two slides look same from two different patients, but when he looks at the two microarrays, there are differences. He is shown wondering what the differences mean. Could they be important to diagnosis? prognosis? or treatment?

Slide 9

Why Is It Important To Collect Genome-Wide Profiles?

Genomic profiling provides a way to collect extensive information about individuals, analyze the profiles from many patients and their tumor(s), and use informatics to discover important differences. Researchers hope to correlate genomic profiles with known clinical factors and outcomes.

Graphic shows genetic material coming from a general population and being analyzed for the presence of a genetic variant suspected of being linked to cancer risk. An analysis team separates the population on the basis of having or lacking the suspected variant and tries to see if this variant can predict cancer risk.

Slide 10

Can Genome-Wide Profiles Predict Cancer Risk?

Not yet. Cancer is a complex disease that is driven by genetic, lifestyle, and environmental factors. Furthermore, each individual and each tumor are unique. Thus, it has proven difficult to identify a predictive genomic profile that is valid for multiple populations. Claims that a particular genomic profile can predict cancer risk should be interpreted with caution. It is likely that genomic profiles will need to be used in conjunction with other genetic and clinical information to accurately assess a person's cancer risk.

Graphic shows three persons with same genetic variant yet with three different health results based upon environments and behaviors. One is in a healthful environment, one in a healthful environment but person is a smoker, and third is a nonsmoker in a carcinogenic factory environment.

Slide 11

Can Genome-Wide Profiling Discover Polymorphisms That Influence Cancer Risk?

Possibly. Studies of polymorphisms in individual patients and in populations are yielding interesting hints. Some (but not all) sequence polymorphisms appear to influence cancer risk. Some sequence polymorphisms may not themselves be associated with cancer risk but may serve as "markers" for nearby genes that do exert an influence.

Polymorphisms within protein coding regions may alter the function of the encoded protein, particularly if they change its structure or amino acid sequence. Polymorphisms in regions of the DNA that regulate gene expression may result in altered protein levels, alternative splicing, or other modifications.

Changes in a protein's sequence or its regulation can influence cancer risk by altering how a person responds to carcinogens, drugs, chemicals, or other environmental exposures.

Graphic shows two populations of men: The No Prostate Cancer group have 2 purple flags present that indicate a particular genetic variant while the Prostate Cancer group have 10. Variants that seem more prevalent in a cancer population may influence cancer risk indirectly, as a marker for other genes.

Slide 12

Can Genome-Wide Profiles Help Diagnose Cancer?

Yes. Genomic profiles can be used to subclassify certain cancers within pathologically defined groups. These profiles can further differentiate tumors that look similar under the microscope. Gene expression microarrays—the major tool used to measure the genome-wide changes in gene expression—have been used experimentally to subclassify acute myelogenous leukemias and to distinguish Burkitt’s from diffuse large B-cell lymphoma. And work is under way to build a genome-based classification system for glioma, the most common type of brain tumor.

Slide 13

Can Genome-Wide Profiles Inform Cancer Prognosis?

Yes. Certain gene changes have already been associated with cancer prognosis. For breast cancer, researchers have identified a set of 21 cancer-related genes whose expression is associated with risk of breast cancer recurrence among women with estrogen receptor-positive breast cancer treated surgically. The profile yields a "score" that predicts the likelihood of disease recurrence. It is not yet clear how each of the genes functions, whether separately or together, whether as "driver" or "passenger," in the pathophysiology of breast cancer.

Graphic shows how a breast cancer tissue sample can be probed with 21 markers for genes linked to recurrence. Based on the test, a report tells there are 34 chances out of 100 that breast cancer will recur.

Slide 14

Can Genome-Wide Profiles Inform Cancer Treatment Planning?

Yes. Certain gene changes have already been associated with cancer prognosis. If therapies are available that target genes, proteins, or pathways that appear, on the basis of genome-wide profiling, to be altered in tumors, these can be used to try to treat the patient. Researchers and clinicians are also analyzing large genomic profile databases to help match patients to the best possible treatments. The genomic profile of a newly diagnosed patient is compared with previous patients' genomic profiles, which are linked to information about how these patients were treated and responded to their treatment. This approach will allow newly diagnosed patients to be given treatments that were effective against cancers with similar genetic profiles.

Graphic shows how colon cancer patients are first sorted by their genome-wide profiles, and, based on the results, those with a mutated gene called K-ras are given a certain targeted therapy.

Slide 15

Can Genome-Wide Profiles Monitor the Response to Cancer Treatment?

Not yet. Once a molecular pathway and its activities that support a cancer are known, genomic analysis may eventually be used to monitor whether or not a pathway-specific targeted treatment has effectively disrupted this pathway, but this has not yet been demonstrated in the clinic.

Graphic shows a researcher trying to blow up the mTOR pathway in a cancer cell and his colleague is using a magnifying glass to peer into the cell and see if the pathway is destroyed.

Slide 16

Will Genomically Informed Cancer Care Be Better for Patients?

Yes. Genome-wide profiling-whether it involves comparative genomic hybridization, karyotyping, polymorphism analysis, or gene expression profiling-is an essential element of genomically informed cancer care. The goal of genome-based care is to tailor cancer risk assessment, diagnosis, prognosis, and treatment to each individual and malignancy. Researchers are using genome-wide profiling to lay groundwork that should enable this to become a reality. However, there is still much work to be done before this new approach can be translated into better care for patients.

Graphic shows a woman trying to navigate the results of her breast cancer diagnosis. There are two stations: One says Phenotype portraits where she is picking up her mammogram, and another says genotype portraits where results of the genome-wide profiling of the tumor are being explained. Patient asks what this means in terms of her care.

Graphic shows a woman trying to navigate the results of her breast cancer diagnosis. There are two stations: One says Phenotype portraits where she is picking up her mammogram, and another says genotype portraits where results of the genome-wide profiling of the tumor are being explained. Patient asks what this means in terms of her care.