Tests To Predict Cancer’s Aggressiveness

Dr. Michael K. Brawer talks about Prolaris, a molecular diagnostic test for prostate cancer

How does Prolaris fit into the spectrum of prostate cancer molecular biomarkers?

Dr. Brawer: Prolaris is a molecular biomarker that quantitates cell cycle progression genes. That is to say it measures 31 genes related to the process that tells a cell to divide into two cells and then the two cells to go to four cells, etc. (This process is called mitosis.)

Dr. Steve Stone, a very smart molecular biologist, decided to look at these cell cycle progression genes for prostate cancer, because these are the genes that provide the majority of meaningful information in breast cancer prognostic signatures.

What is important about these genes? If you ask a fourth grader, “What is cancer?” he mostly likely is going to tell you cancer is what happens when cells grow out of control. That is what makes it cancer: the control of the division of the cells, or the mitosis, has gone amuck and now you’ve got unfettered proliferation.

More importantly, in the process of dividing and multiplying cells, mutations that arise in one cell get passed on to the daughter cell. That is how cancer progresses down its path from an incipient cancer to a cancer that may kill the patient. This passing on of genetic information that accrues through mutations makes cancer progress into a lethal type of cancer.

Prolaris is a test done on tissue obtained from a diagnostic biopsy or a radical prostatectomy. The clinician sends the specimen to Myriad Genetics where we quantitate the expression of these genes and adjust for other genes, which are constantly expressed. We can tell whether a man has more of these 31 genes or less than these 31 genes expressed at a stable rate.

We’ve now published data on nine cohorts in several publications; we’ve looked at men treated with everything from watchful waiting to radical prostatectomy to external beam radiation therapy. We have shown in all the studies that Prolaris is either the dominant predictor or the co-dominant predictor of whether the man does well with his prostate cancer, dies, develops metastasis, or fails either radiation or radical prostatectomy by having PSA progression.

How are Prolaris’ results used in the clinic?

Dr. Brawer: You raise an important point. If the test makes no difference, why do the test?

We’ve published three utility studies. These utility studies ask, “How was the man going to be treated before the clinician got the Prolaris result” versus, “How was the patient actually treated after the clinician received the results?”

Overall, well over half the patients in all three studies received a different treatment after the Prolaris result then they would have before the test.

So then it does change treatment decisions?

Dr. Brawer: Right.

For example, say a 55-year-old man has a Gleason 3+3 prostate cancer, his PSA is 12, and he has T1C disease.

A urologist sees him. The urologist says, “Based on what I see here, I recommend a prostatectomy.” The patient could then say, “My golfing partner had a test called Prolaris. I’d like to get that, as well.” He gets a Prolaris test; there is a whole host of information provided in the result.

With Prolaris, we show the chance that a man will die of prostate cancer. We can do that because we have data on three cohorts of men initially treated conservatively who were then followed for a median of about a decade. We can see who ended up dying of the disease.

Let’s say this patient who was originally going to get a radical prostatectomy finds out through Prolaris that his prostate cancer is considered low aggressive.

Then his doctor can confidently recommend active surveillance because the chance that he is going to progress is extremely low. This particular patient may do fine for the rest of his life just having periodic examinations. He doesn’t need aggressive treatment. The opposite scenario is also possible.

To read Dr. Michael Brawer’s interview, fill out the form here to get a free copy of our June 2016 issue.

 

 

 

Advertisement

Gene Variants + Prostate Cancer

Dr. Stephen Thibodeau from the Mayo Clinic talks about hereditary prostate cancer.

What about the association between BRCA2 and prostate cancer?

There are some families that have BRCA2 mutations in prostate cancer. BRCA2 is one of the genes that has been identified in familial prostate cancer. But BRCA2 accounts for a very small percentage of prostate cancer cases, unlike what you might see in hereditary breast cancer. Some of the mismatch repair genes have also been shown to be involved in prostate cancer, but these also occur in a small fraction of the cases. There are really generally two types of genetic studies that have been done for prostate cancer. One is to look for these major genes, those genes that have high penetrance—that is, if you find a mutation in these genes, there is a high likelihood that the man will develop disease. An example of this is BRCA1 and BRCA2 for breast cancer or APC for colon cancer. As previously mentioned, finding these types of genes for prostate cancer has been very challenging.

What is the other type of study being done?

The other type of genetic study being done is called a genome-wide association study (or GWAS). These studies take advantage of the fact that there are a large number of common variants within individuals and within the population. This type of study, then, looks at these common changes in DNA in the population, or common polymorphisms, and tests whether or not some of the changes are found more frequently among affected men compared to normal men. Among the million common DNA changes known to be present in the population, about 100 of these have now been found to be associated with prostate cancer. Unlike highly penetrant genes, like BRCA1 and 2, these common variants increase the risk for prostate cancer by only a very tiny amount. This type of genetic study has actually been very productive for prostate cancer compared to other malignancies.

Very large studies have now been done. Some groups have looked at 40,000 cases with prostate and 40,000 controls, looking for differences in DNA among the cases compared to the controls. As I mentioned, over 100 DNA variants now have been identified, with each variant individually increasing prostate cancer risk. If you look at individuals who might have multiple variants present, then the risk increases even more. The data so far suggests that these risk variants may account for about 30% of what we expect the hereditary component of prostate cancer to be. If you have these variants, your risk increases, but it’s a very small risk in comparison to the general population.

Broadly speaking, those are the two main types of ongoing studies, with the idea that there is still a genetic component that it is not completely understood.

Is it a matter of looking more in-depth at the samples and the data that you already have, or is it that as the pace of technological innovation accelerates, we’ll learn more? (It’s only been a short time since we’ve developed the capability to quickly sequence the entire genome…)

Dr. Thibodeau: I think it’s a little bit of both. We’re really taking advantage of the sequencing capability that we haven’t had in the past. A lot of the methods that we’d used in the past to identify the genes were really indirect. BRCA1 and BRCA2 and the mismatch repair genes were all identified by a technique called linkage analysis, which is a way to globally look at all genes. Indirectly, you take a family and you look at markers scattered across the genome. You’re looking to answer the question, “Do these markers follow disease in the family?” If they do follow disease in the family, then you look for genes around that marker. It doesn’t really tell you what the gene is; it just tells you a location on the chromosome.

It is really only within the past five years with the introduction of next-generation sequencing capability that we can now sequence every gene. Rather than searching for genes indirectly, which is really very difficult to do, we can now look at genes directly. The idea is that we’ll take every gene and sequence them for all of the individuals in the family and then look for differences. What are the differences in each gene in men with prostate cancer compared to what we might see in men who don’t have prostate cancer. If you sequence enough individuals, you can begin to appreciate what is different in the groups that have cancer compared to the groups that don’t.

This is really the first time we’ve been able to use this technology. That is why we’re more hopeful. As newer technologies emerge, we have a better chance of identifying what might be present in familial prostate cancer. The technology is definitely having an impact and increases our chances.

The other important issue is that the biology of prostate cancer is very difficult. The disease is complex. Since prostate cancer is so common, it might be that some of the prostate cancers that we see in our families are not due to genetic causes; they’re due to something else.

What do you mean?

Like many diseases, there can be multiple causes. First, there are a variety of environmental factors that increase risk. Second, it is often the case that the disease is not caused be a single gene. There are often several genes that come into play. Finally, all of these factors play a role, so in the end you have multiple genes and then all of the environmental factors that come together to result in a disease process. All of these factors are still being sorted out for prostate cancer.

You may have a mixture within a family: some individuals have prostate cancer due to a genetic cause and some have prostate cancer due to another cause. The technology allows us to look at things that we weren’t able to look at before, but the biology is still very difficult and that complicates our ability to identify these genes. We’re still optimistic that over time we’ll identify more genes.

How does all of this play out once you’ve identified a gene? There are some companies that offer BRCA2 testing. How useful to patients are those commercial tests?

They’re very useful for the disease they target. For BRCA1 and BRCA2, clinicians now routinely test women at risk for breast and/ or ovarian cancer. The tests are a very important part of the toolkit that physicians have for at-risk individuals. (At-risk individuals are typically individuals who are very young when they develop a disease, or who have a family history.)

For some of these cancers— for colon, breast, and prostate, for example—one can do prophylactic, or preventative, surgery. Importantly, if you identify high-risk individuals, then these individuals can be monitored more carefully and more frequently with the goal of early detection and early treatment

To read Dr. Stephen Thiodeau’s interview, fill out the form here to get a free copy of our June 2016 issue.

Hereditary Prostate Cancer

Dr. Kathleen A. Cooney of the University of Utah talks about prostate cancer genetics.

What do we know about the genetics of prostate cancer—and what don’t we know?

In the 1990s, we were just starting to recognize that prostate cancer clusters in families. Soon it was clear that family history is a strong risk factor for prostate cancer. In fact, the three most important prostate cancer risk factors are age, race, and family history. Despite extensive research on other risk factors, these three remain the most significant.

The fact that prostate cancer runs in families could be chance or it could be the result of shared environment or another type of exposure. But genetics also seems a likely cause for some of the familial clustering of prostate cancer. This hypothesis led many investigative teams, including my own, to try and identify families with multiple cases of prostate cancer. We used those families to conduct linkage studies looking for segments of DNA that harbor genes that might be relevant to hereditary prostate cancer. This research, however, has proved much more challenging than initially anticipated.

Some of the family studies in prostate cancer were based on research on other common cancers like colon and breast cancer. In these cancers, linkage studies were quite successful in helping us identify key genes that are not only important for risk assessment in families, but also for providing knowledge about the basic biology of cancer.

Finding prostate cancer genes has been difficult for a number of reasons. First, prostate cancer is a late-onset disease. Often, by the time a man is diagnosed with the disease, other family members who may have also had the disease are already deceased. Obtaining access to these older medical records is often challenging. Another problem is that we have had variable ways of diagnosing prostate cancer over the past 30 years. With the introduction of PSA testing to find early-stage or asymptomatic disease, there was a dramatic increase in the number of prostate cancer cases in the US. Men with prostate cancer were encouraged to have their male siblings screened for prostate cancer and this made it more difficult to identify the hereditary cancers. This over-diagnosis of prostate cancer in family members complicated our field for a long time since we cannot differentiate between genetic cancers and sporadic or nongenetic cancers in terms of clinical factors.

The National Cancer Institute (NCI) supports groups working in prostate cancer genetics, including the International Consortium for Prostate Cancer Genetics (ICPCG), which has been funded for 15-20 years to collaborate on studies of hereditary prostate cancer. Our laboratory participates in this consortium to identify prostate cancer genes.

In 2012, our laboratory, in collaboration with Dr. Bill Isaacs at Johns Hopkins University and others, successfully used family-based approaches to identify a recurrent mutation in the HOXB13 gene. Additional studies by us, and others, suggest that this recurrent mutation accounts for about 5% of hereditary prostate cancer families worldwide. The protein encoded by this gene is important in prostate development. It’s still not clear what the mutations in the HOXB13 gene do to increase prostate cancer risk, but it is probably one of the more important findings in hereditary prostate cancer because it demonstrated our ability to use families and linkage studies to isolate a prostate cancer gene.

Subscribe or purchase the current issue to read the interview.