Genetic Changes in Cancer: Second of Three Parts

by Eric P. Hoffman PhD
Source: Summer 2000 CCCF Newsletter

As described in an introductory article on cancer and genes in the last issue of this newsletter, the two major types of cancer genes are tumor suppressors, and oncogenes. Each one does what its name implies: tumor suppressors are proteins that are responsible for suppressing tumors, while oncogenes make proteins that promote oncogenesis (cancer).

Tumor suppressors are in all of us, and we have two genes for each tumor suppressor (one from our mother, and one from dad). A cell becomes a tumor when it loses the ability to produce the tumor suppressor. As explained in more detail in the previous article, one bad copy of the gene can be inherited in “cancer families”, while the second, good copy of the gene gets knocked out randomly in very rare cells in the body, and this single cell leads to the tumor. If we can restore the presence and function of the tumor suppressor in all the cancer cells of a patient, then this should stop the cancer.

Oncogenes are abnormal genes that cause cancer through the production of a toxic, oncogenic, protein. None of us are born with oncogenes. Instead, oncogenes are created randomly in very rare cells in rare patients, and this single cell gives rise to the cancer in the patient. In the case of oncogenes, ideal therapeutics would be directed towards eliminating the toxic protein from the cancer cells, or at least inhibiting its toxic nature. Indeed, this very directed approach of going after the specific abnormal, toxic protein in cancer patients has recently been shown to work. STI- 571 is a designer drug that inhibits the action of a specific toxic protein called “BCR-ABL fusion protein”. This oncogene is the result of two normal genes switching parts and creating a hybrid oncogene, which then is able to produce a hybrid toxic protein (Philadelphia chromosome; translocation of part of the ABL gene on chromosome 9 with part of the BCR gene on chromosome 22). We will explain more about the new STI- 571 breakthrough later in this article.


In some cancers, a single genetic change seems sufficient to produce a cancer cell. However, in other types of cancers, a series of genetic changes are needed before a metastatic tumor results. For example, neuroblastoma is not associated with any specific oncogene in early stage tumors, however late stage metastatic tumors show a large number of new copies of the n-myc gene being produced. This gene is not rearranged or changed, however it is “amplified” so that too much of it is made. Normal levels of n-myc are OK, but too much n-myc is toxic and creates the tumor. It is thought that many adult cancers are slow and insidious in nature, with the immune system recognizing the vast majority of new cancer cells as cells gone awry, and quickly destroying them. Thus, we may develop cancer each hour or each day, yet the immune system surveillance is good at removing the wayward cells. It is for this reason that patients with suppressed immune systems, such as AIDS patients, often show a high rate of cancer; their suppressed immune system does not have the same surveillance ability as a normal individual’s does.

To recap, cancer is a combination of loss of tumor suppressor proteins, gain of toxic proteins from new oncogenes, and immune surveillance. In this context, it is interesting to note that some of the most common cancers involve the immune cells themselves, with a single critical oncogene change creating the cancer. It seems that the immune system is at a particular disadvantage when it needs to screen its own cells for wayward citizens. Almost all the oncogenes in blood cancers involve combining two different recipes (genes), into one wayward gene. In the ‘Joy of Cooking’ examples used in the previous article, this is similar to taking the first half of the recipe for steamed hardshell crabs (the ingredients), and using the second half of a recipe for homemade ice cream (with blending and freezing of the crabs, shell and all). The frozen, homogenized crabs would not fulfill the function of either main course, or dessert, and would likely ruin the meal. Specific “fusion genes”, with the beginning of one gene added to the end of a second through a gene switch called a “translocation”, are associated with specific leukemias, and often linked with certain prognosis. For example, ALL patients often show a BCR-ABL fusion oncogene which results in too much signal to the cell to continue to grow unabated and this is considered a poor prognosis (although see STI-571 information below). Other ALL patients show a TEL-AML1 fusion oncogene, and these often have a better response to chemotherapy. Eighty-five percent of patients with Burkitt’s lymphoma show a chromosome 8 and chromosome 14 translocation, which plasters the first part of an immunoglobulin gene (IgH) onto the end of the MYC gene. While the MYC gene has a normal function in other cell types, it is told by the IgH gene to make far too much of the MYC protein in B cells, causing them to lose control of their growth and proliferation.

Ninety percent of patients with Ewings sarcoma show a translocation between chromosomes 11 and 22 (notated as t (11:22) ). This creates a novel fusion oncogene between the EWS and FL11 genes. Rhabdomyosarcomas of the alveolar type show a translocation between chromosomes 2 and 13, involving the PAX3 and FKHR genes. In this case, the PAX3 gene tells the cell when it should differentiate into some specific type of cell, but it’s sending the wrong signals to the wrong (FKHR) gene, causing the cell to become tumorgenic.

Tumor suppressor genes are lost in cancers, they do not “gain (toxic) function” in cancers as do oncogenes. This is a critical point, as we all have two copies of each tumor suppressor, and a cell needs to lose both of the copies before it has “lost function” of this protein, and become a cancer cell. This is in contrast to oncogenes, where only one fusion gene needs to be made to create the toxic protein. Loss of tumor suppressor genes are often the critical event in certain types of childhood cancer, and these typically involve “deletions”, where the entire gene is removed from the chromosome and lost. For example, retinoblastoma invariably involves the loss of both copies of the RB1 gene on chromosome 13. In familial cases, one chromosome 13 with a lost RB1 gene is passed on from one parent, while the second RB1 gene is lost just in the cells which develop into a tumor. Neuroblastoma patients often lose tumor suppressors on chromosome 1, while Wilms’ tumor patients lose critical tumor suppressors on chromosome 11.


With all the many genes now known as critical events in the development of specific tumors, some therapeutic strategies seem obvious; put the right tumor suppressor proteins back into retinoblastoma, neuroblastoma, and Wilm’s tumor cells, and get rid of the toxic oncogene fusion proteins in ALL, Burkitt’s, Ewings, and rhabdomyosarcoma tumors.

Putting missing genes back into cells involves “gene delivery”, or what is commonly referred to (in a forward-thinking way) as “gene therapy”. While many of the normal genes are currently in hand for specific tumors, the challenge lies in getting the right genes to the right place (eg. the normal RB1 gene to the retinoblastoma cancer cells). While technology is evolving rapidly, it is still a major technical hurdle to get each cancer cell a complete copy of a good gene. More promising strides have been made in the destruction (or inhibition) of toxic oncogene proteins. A recent dramatic example is with regards to the design of a drug to inhibit the action of the BCRABL fusion protein that is responsible for many cases of adult CML, and many child/ teenage cases of ALL and CML. A website gives a comprehensive history of the discovery of this drug (www. newcmldrug.com). Novartis Pharmaceuticals developed a series of inhibitors of the activity of tyrosine kinases (regulatory proteins which provide signals to cells to tell them, among other things, when to start and stop growing). Investigators at Novartis and Oregon State Health University showed that one of these, STI-571, was particularly good at knocking out the activity of the toxic BCR-ABL fusion, created by the translocation between chromosomes 9 and 22 (Philadelphia chromosome). Importantly, the drug seems targeted for only the BCRABL toxic protein, and hence does not seem to have any significant side effects. It is also taken orally. Over 200 patients are currently enrolled in STI-571 trials, with new trials scheduled to open for both adult and pediatric CML and AML patients (including COG studies; see front page article in this issue). Patients with Philadelphia chromosomepositive CML or ALL can contact Novartis at its toll free number 1-800- 340-6843 to inquire about eligibility and location of clinical trials.

In the next issue of the newsletter, emerging high-tech genetic technologies will be described, including DNA microchips, and gene delivery using viral parasites of other viruses. Dr. Hoffman has had a long-term interest in genetics and its implications for health and disease. As a post-doctoral fellow, he participated in the identification of the first ‘positionally cloned’ human gene and the identification of its protein product. He is currently Director of the Research Center for Genetic Medicine at the Children’s National Medical Center in Washington DC, Professor of Pediatrics at George Washington University (and is married to wife Ruth.)