Using genomics in cancer, and for early diagnosis in infectious disease

In the previous post I discussed the field of pharmacogenomics.

Today I will focus on:

Nutragenomics
Disease classification
Disease prognostication
Early and rapid diagnosis
Prediction of diseases to develop later in life

Nutragenomics or the science of how our genes affect what we eat is a developing field.

Here is one example. Cholesterol is an important part of our physiology although too much in the wrong places is harmful. Cholesterol is carried in the body by lipoproteins; low density lipoproteins or LDL in excess are harmful and high density lipoproteins or HDL must be relatively high to be protective. The levels of these lipoproteins are controlled both by our genes and by our diet. We also have a set of genes that produce proteins called apoEs that are involved in cholesterol metabolism. Our genes specify if we have apo2, apo3, or apo4s and which one you have makes a big difference in how you metabolize the cholesterol you eat. Most of us have apo3s but about 30% have apo4s; different because of just a one letter change at codon #158, a change just enough to alter cholesterol metabolism. Those with apo4s tend to have higher blood levels of LDL but, to their advantage, adjusting their diet can have a major impact, much more than the same diet change in a person with apo3s. The person with apo3s is at less risk but it is also more difficult for them to control their LDL with diet.

Genomics is proving to be very valuable in disease classification, especially with cancer. A pathologist’s evaluation looking at a microscopic slide has been the basis for most cancer classification – separating out breast cancer from lung cancer but then sub classifying each such as small cell and non-small cell lung cancer or the various subcategories of lymphomas. To this was added some years ago histochemical analysis to learn if a breast cancer was high in estrogen or progesterone receptors and then molecular diagnosis to find, for example, if the tumor had a high complement of the receptor Her2neu – each being important markers for the approach to treatment. Now genomics is adding an ability to delve much more deeply and find what the DNA mutations are in the individual tumor and how they are similar or different from others. This in turn is leading to searches for new drugs, as discussed last time.

This same work allows for early prognostication. Consider 100 women with breast cancer that appear by all the usual criteria to be the same type and of the same early stage. We know that most of them will respond well to current therapy of surgery, radiation locally and, in certain circumstances, systemic chemotherapy or hormonal therapy. But a small percentage will have a relapse. The problem is that there has been no way to determine in advance who is at risk of relapse. Genomics has begun to answer this problem. Analyzing the genomics of the tumor at the time of diagnosis, it is possible to separate these women into a good prognosis group and a poor prognosis group. The former rarely relapse and one might even consider if they need the same level of aggressive therapy as they are now getting. And the latter group is at high risk of recurrence; they are obvious candidates for clinical trials of alternate approaches to determine if relapses can be reduced. One such genomic prognostic test has been approved by the FDA and others are in the works for multiple cancers.

Genomics can be used for early diagnosis, especially in the field of infectious diseases. Remember the gentleman who flew to Italy on his honeymoon but who had tuberculosis? It led to an international concern that he might have infected others and that his TB might be of the drug resistant variety. One of the problems was that it takes about six weeks to grow the TB bacteria in the laboratory and then, if present, another six weeks to test for antibiotic susceptibility. But genomic tests can speed that process up to just hours. The TB bacteria (Mycobacterium tuberculosis) has a characteristic genomic profile so, if present in a sample from the patient, it can be detected within hours. And since antibiotic susceptibility is driven by the bacteria’s genes, they can be analyzed at the same time. A huge improvement in time to diagnosis and getting the right drug started from the beginning.

We might want to know if we are predisposed to develop a certain disease later in life. It is possible that genomics can be of real assistance here; indeed this has been a major “promise.” It turns out that most of the common, important diseases such as diabetes and coronary artery disease have not one but vast numbers of genes that have some impact on their development. So we will not find a simple answer for many of these. But as more is learned it is very possible that each of us will be able to learn our relative risk to some important and common illnesses. If you knew, for example, that you were at increased risk of heart disease, it might be a stimulus to you to be more diligent in eating a Mediterranean style diet, exercising more often and looking for ways to control stress- and it would be an added inducement to stop smoking. Similarly, if you were at risk for early onset colon cancer, you might be more careful to eat a diet high in fiber and low in fat and begin having colonoscopies at an earlier age.

These are just some of the advances coming from genomics; expect to see many more because genomics represents a true revolution in medicine and we have only seen the beginning.

Stephen C. Schimpff is a retired CEO of the University of Maryland Medical Center in Baltimore and is the author of The Future of Medicine — Megatrends in Healthcare. He blogs at Medical Megatrends and the Future of Medicine.