“…we examined a huge number of genes, looking for the difference between normal tissue and benign tissue, so that we could identify which of these genes are unique to the benign tumor. This could potentially provide a target for therapy and tell us something about how these things come about and grow.”

The main difference is that leiomyomas are benign and leiomyosarcomas are malignant. Although most leiomyosarcomas don’t arise in uterine tissue, a fair percentage do. We have smooth muscle almost everywhere in the body, but uterine muscle is basically all smooth muscle.

Yes, we’re a team. She’s on the faculty in a different department, and we do a lot of research together.

We not only looked at differences in gene expression between leiomyomas and normal uterine muscle from which it arises, but we also looked at gene expression in a wide variety of other normal tissues. That made it a much more extensive study. We compared the leiomyoma gene expression with 18 types of normal tissue—for example, kidney, liver, lung, muscle, etc.—and each set contained a sizable number of samples. We were able to identify genes uniquely expressed in leiomyomas among these normal tissues.

It’s probably because it was, first of all, a big study. There were a lot of samples, more than 250. Second, at the time, it was probably the most extensive number of genes ever looked at in leiomyomas. A third point would be the fact that we were able to look at these genes in many different tumors and normal tissues.

That research was published in Cancer in 2003. We used some of the genes identified in the paper we’ve been talking about, and found that they are also expressed in leiomyosarcomas. As I said, we wouldn’t have found them otherwise. They’re only expressed in a subset of leiomyosarcomas. Doublecortin, for instance, is a gene expressed in leiomyosarcomas. We also found other genes over-expressed in leiomyosarcomas. One reason to do these assays is so we can then see if a gene is expressed more in cancer than in normal tissue. In the case of doublecortin, we found that it seems to be expressed in some uterine leiomyosarcomas, but not all of them.

That’s a difficult question to answer. But, however these leiomyosarcomas work, it’s being partly described by these data. Clearly what we found there is very important in how they work, but it doesn’t really tell us how to put it all together to fully understand it.

I think it’s kind of lucky that we did the leiomyoma paper first in terms of the leiomyosarcoma paper. If we hadn’t, we never would have found the doublecortin expression in leiomyosarcoma, for example.

We prefer not to publish everything in the same place. I think these days the choice of what journal you go to isn’t as important as it used to be. Now almost everybody finds papers of interest by searching on PubMed.

I’m surprised. I don’t really know how often my papers are cited. I don’t pay attention to that. Certainly, leiomyomas are a common problem, and most women get some, so maybe that contributes to the citation rate.

When you’re working with these DNA microarrays, each sample comes up with 40,000 to 60,000 numbers—that’s per sample. So to do the analysis, you have to rely on certain statistical approaches, and you can come up with false positives and negatives when you do that. How you analyze the data is very important, and it’s not always perfectly straightforward what the best way to do it is. There’s always more than one way, so figuring out which one is the best without being misled is a challenge.

We’re continuing to look at gene expression in different types of sarcoma along with some other malignancies. We’re not really following up on the leiomyomas per se. We’re looking more at the malignant aspects of tumors. We’re looking for potential targets in these leiomyosarcomas and other tumors that wouldn’t be present in normal tissue. If we find these, then we can exploit them. We’re also trying to identify subgroups that would have practical significance within some of the common tumor types.

The message that comes out of this is that we used a large number of samples of leiomyomas and normal tissue; we examined a huge number of genes, looking for differences between normal tissue and leiomyomas, so that we could identify which of these genes are unique to the benign tumor. We found several genes quite selectively expressed in leiomyomas. This could potentially provide a target for therapy and tell us something about how these tumors come about and grow.

That’s easy. We didn’t even have a glimmer that we could do this type of thing 10 years ago. Microarray technology was first described about 10 years ago, and commercial development came later. As soon as we learned about it and realized we could do it, that’s what we wanted to do. Ten years ago, it would have been mind-boggling to think that such a thing was possible. It’s still kind of amazing.

Keith M. Skubitz, M.D.
University of Minnesota Medical School
Minneapolis, MN, USA