CANCER COALITION

Genetic Testing in Cancer

The National Cancer Institute exists to reduce the burden of cancer to our society and to alleviate the terrible toll it takes. The weapon of the NCI is knowledge about the causes and fundamental nature of cancer. It is this knowledge that will ultimately allow us to prevent, diagnose, and successfully treat these diseases.

We have a long way to go, but a cautious optimism is beginning to ripple through the scientific community--the result of an enormous increase in our understanding of just what happens to transform a normal cell to a cancer cell. We now know that cancer, all cancers, are genetic diseases. Each specific cancer is a result of changes in a relatively small number of genes--each gene being a specific unit of genetic information encoding the instructions that directs the production of protein products with the cell.

It is extremely important to emphasize that while cancer is a disease of genetic changes, it is generally not an inherited disease like cystic fibrosis or sickle cell anemia. Rather, most cancers arise within a cell of the body bgcolor="#FFFFFF" that,through its like time, accumulates the genetic changes peculiar to each cancer. For some cancers, we now know that the gradual and sequential change in perhaps half a dozen genes signals the transformation form a normal, well-behaved cell to a growing and spreading cancer.

While the vast majority of cancers are the result of these acquired, or somatic, mutation, 5 to 10 percent of cancers are in fact inherited as what we would call a simple genetic trait, like cystic fibrosis.

In these cases, we observe cancer in multiple family members and across multiple generations in the affected family. The cancers in these inherited cancer syndromes tend to occur at an earlier age than in their sporadic counterparts and tend to arise in a single individual multiple times. Inherited cancer syndromes account for up to 10 percent of many types od cancer including breast, ovarian, colon, melanoma, kidney, and prostate. Within these cancer prone families, individuals that inherit a single defective copy of a single gene are at a greatly increased risk of developing cancer. For some cancers, and some genes, this risk may be 90 percent or higher over the course of an individual's lifetime.

One of the great triumphs of current research is the identification of cancer genes that underlie a growing number of recognized inherited cancer syndromes. These cancer genes are often called tumor suppressor genes because the normal function of the normal version of these genes appears to protect cells from cancerous transformation. As cancer is a disease of genetic changes or something we refer to as a disease of genomic instability, the normal functions of these tumor suppressor genes in many cases are to maintain the integrity of the genome throughout the like of a cell. While we are here today to talk about genetic testing which is being made possible by the identification of cancer susceptibility genes, I cannot overstate the importance of the identification of these genes and the determination of how these genes function normally, and of how the loss of function of these genes predisposes to cancer. Discoveries in this area are profoundly and fundamentally changing our knowledge, not only if inherited cancers, but of their much more common sporadic counterparts. Few areas of cancer research are giving us as clear a set of windows into the nature of cancer than these cancer susceptibility tumor suppressor genes.

In families that suffer from inherited cancer syndromes, it is mutations or changes in specific tumor suppressor genes, resulting in a change or loss of function of the protein product that that gene instructs the cell to make, that explains the enormous increase in the predisposition to cancer seen in these families. The revolution in human molecular genetics that Francis Collins will describe to you is making these gene identifications possible. Over the past two years along, scientists have identified genes responsible for inherited forms of breast and ovarian cancers, colon cancer, melanoma, and kidney cancer, to name but a few. One of the major goals of cancer research is to predict who will get a particular cancer. With the ability to identify individuals within these cancer prone families who do and who do not carry the mutated gene, we can predict who in those families carries the particularly high predisposition to cancer and those who do not.

While these inherited cancer syndromes only explain the minority of cancers, the number of affected individuals is large--perhaps one million Americans carry a breast cancer predisposition gene mutation and another one million Americans carry mutations in a colon cancer predisposition gene. n these inherited cancer syndromes, the mutated gene which results in the cancer predisposition is inherited and that mean that the defective gene is present in the DNA carried in each and every of the trillions of cells of the individual. it is present in the DNA of blood cells and it is present from birth, long before cancer develops. It is this fact that allows the possibility of genetic testing to identify those individuals who carry the mutation.

This, however, is easier said than done So far, I have spoken only of diagnosing individuals within high risk families and only for families in whom the responsible gene has been discovered. While the past few years have seen the rapid discovery of some tumor suppressor genes responsible for inherited cancer syndromes, more await discovery. There is another reason that I have only spoken of family members. Each tumor suppressor gene is made up of hundreds to many thousands of letters of the genetic code. A defect in spelling anywhere in these enormous genetic words can, theoretically, be the culprit. Even when the cancer gene is discovered, such as the first breast cancer predisposition gene, BRCA-1, which accounts for about 50 percent of inherited breast cancer and about 75 percent of inherited breast plus ovarian cancer, nearly every affected family has its own misspelling. The result of this enormous genetic heterogeneity stretches the technical and financial feasibility of screening for mutation outside of families in which the painstaking work of mutation identification has already been done. Because the mutation found in each family is, by in large different, it is currently not feasible to screen populations searching for that unknown misspelling.

With this technical limitation in mind, I want to point out that Francis Collins and I announced yesterday the remarkable discovery of a single misspelling in the BRCA-1 gene that is found in as many as 1 percent of Ashkenazi Jews, or Jews of Central or Eastern European origin. This group represents 90 percent of the 6-7 million Jews in the United States. For the first time, the technical ability to actually screen a population for a cancer predisposition gene is feasible. This discovery signals a fundamental change in the many issues we must come to grips with and, because of the pace of scientific discovery--because of the success of the NIH--we must be prepared for the challenge of this changing landscape.

Genetic testing for cancer predisposition s becoming a reality but just how will it be used and, most importantly, how will its use benefit people? What types of screening will become available? Will we look for mutations in particular genes or in sets of genes in the general population? Will we look for mutations in particular genes in selected populations such as what I just described for BRCA-1 in Ashkenazi Jews? Or will we limit our screening for mutations to individuals who are deemed to be at high risk because of a particular or compelling family history?

To answer how we will apply gene testing to clinical practice will require more knowledge than we now have. We need of Course to continue to identify additional cancer susceptibility genes. Second, we need to focus on developments that will address the technical feasibility, cost, and cost benefit of screening for particular genetic defects. Finally, we need to generate data that addresses the ultimate issues in genetic testing for cancer--how the information gained from genetic testing either helps of does not help the individual affected.

The responsibility of the biomedical community at this point must be aimed at providing information that addresses these issues so that individuals can make informed decisions about whether or not to seek such genetic testing.

It is important to point out that testing negative for a particular cancer susceptibility gene defect tells an individual that they do not carry the risks of a particular cancer or cancers associated with that specific gene defect but does not change the significant risk that this individual, like any individual, has of getting cancer due to causes other than that particular predisposition gene.

On the other hand, what do we have to offer people that do test positive? Here is the central problem. It is attempting to answer this question that takes us to the limits of our current knowledge and tells us what types of information we will need to gather.

For a particular mutation in a particular cancer susceptibility gene:

a) What is the risk of developing cancer and when? Remember, these are cancer susceptibility genes and even when they confer an 80 to 90 percent lifetime risk of developing cancer, we need to know what other environmental, behavioral, and genetic factors determine when, and if, an individual who carries a particular mutation develops cancer.

b) How should "at-risk" individuals be followed to monitor for the development of cancer?

c) Finally, how should "at-risk" individuals be counseled in terms of treatment and prevention options?

To answer all these questions requires careful clinical studies and patients and health care providers must have knowledge about and access to studies aimed at answering questions about risks, surveillance, screening, prevention, and treatment.

The identification of genetically high risk individuals provides an extraordinary opportunity to more rapidly and effectively accomplish clinical trials in cancer prevention through dietary, drug, immunologic, or other interventions. It also provides the opportunity to establish trials aimed at developing and evaluating early detection using genetic or other biomarkers as well as imaging technologies.

To accomplish all of these things requires that we address several needs and challenges:

1) Basic research

We need to continue to discover and characterize cancer predisposition genes. So far, we have only talked about cancer predisposition genes that are inherited as what geneticists refer to as simple traits, in which the inheritance of one specifically altered gene is alone responsible for the increased cancer susceptibility. As I have described, such simple genetic predispositions already provide us with enormous scientific and technical challenges. However, it is fair to say that these simple genetic predispositions are likely to only be the tip of the iceberg of the influence of heredity on cancer predisposition. We will need also to turn our attention in basic research to develop the ability to identify genetic predisposition in families where it results from inheritance of more than one genetic locus. We need also to be able to identify modifier genes and other modifying factors that affect what we call the penetrance of a cancer predisposition gene--in other words, genes that modify the risk of getting cancer in individuals with the inherited predisposition. Finally, we need to establish the non-genetic factors, such as environmental and dietary exposures, behavior and lifestyle, infectious agents, and others that may influence the penetrance of cancer susceptibility gene.

2) Technology Development

We need to develop new technologies to make the identification of mutations and other alterations in specific genes both technically and financially feasible. these technologies must be developed with respect to validation, reliability, automation, and cost. We need to establish databases to catalogue mutations and to correlate the specific mutations in an individual gene, such as BRCA-1, with the clinical consequences of that individual mutation including age of onset, aggressiveness, responsiveness of the tumor to therapy, and efficacy of surveillance and diagnostic studies. Coordinated databases will be needed for family registries and epidemiologic data. Finally, we will need to develop centralized tissue and DNA banks for future studies. These tissue and DNA banks must themselves be linked to excellent clinical databases and made widely available to the research community.

3) Human Resource Development

Genetics is changing the landscape of biomedical research and it will change the landscape of clinical practice. To be prepared for these changes will require attention to human resource development. Here I will just touch upon one issue--the need for genetic counseling in oncology. there is a real need to train genetic counselors and for physicians, other health care providers, patients and communities to have access to effective educational materials and guideline for all the issues surrounding the use and interpretation of test aimed at addressing genetic susceptibility to cancer. We must include training in genetics, risk assessment, and the ethical, legal, social, and behavioral aspects of genetics for health care providers.

Conclusion

It has long been observed that cancer runs in families. We are here today, all of us, as participants in a revolution in medicine, in science, and indeed a revolution in our very conceptualization of individual identity and of predicting the type of future an individual may face in terms of his or her health. The discoveries we have talked about today, as with all discoveries, raise opportunities and very serious challenges. We must address ourselves to both the new opportunities raised by these discoveries, opportunities for the early detection, for the possibility of prevention and ultimately for the development of new therapies for cancer. Equally, we must be aware of the challenges. I have limited my remarks to some of the scientific, technical, and human resource challenges, but the challenges do not end there. The potential power of reading ones own genetic script raises societal and personal issues about insurance, employability, privacy, and personal choice that we cannot ignore and that my colleague, Francis Collins, will address.

I thank you for your attention, for this opportunity, and I would be delighted to answer any questions.