At A Glance
Name: Kojo Elenitoba-Johnson
Position: Associate and assistant professor, division of anatomic pathology, University of Utah Health Sciences Center, since 1997.
Background: Fellowship, department of hematopathology, National Cancer Institute, 1995-1997.
Residency, department of anatomic and clinical pathology, Brown University School of Medicine, 1991-1995.
MD, Lagos University Teaching Hospital, Lagos, Nigeria, 1983-1988.
During this week’s Biomarker Pipeline conference in Boston, scientists discussed the transition from a biomarker’s discovery phase into its clinical development phase. ProteoMonitor caught up with a research scientist/medical doctor this week to discuss the interface between research and clinical work in developing biomarkers for lymphomas.
What is your background in terms of research and clinical training?
I’m a chematopathologist and a molecular pathologist. I guess that means that I’m board certified in hematology and molecular genetic pathology. Prior to that, I did my subspeciality training at NIH, and prior to that I did my residency at Brown University for specialty training in pathology. Pathology is my specialty.
Do you do mostly clinical work now, or mostly laboratory work?
It’s probably half and half, or maybe a little more research.
What drew you to the research side of it, from the clinical side?
Well, my laboratory is interested in three things: One, molecular genetics, or the molecular pathogenesis of lymphomas. We’re also interested in leveraging new technologies to improve the diagnosis of those lymphomas. The third thing is, by leveraging these technologies we hope to be able to find rational targets at which novel therapies may be directed.
When did you get into developing new technologies to improve diagnosis?
Perhaps as a third-year resident at Brown. One of my mentors — a guy by the name Thomas King — took me into his molecular lab, and at that time Southern Blots and PCR were just being introduced as adjunctive diagnostic information for determination of clonality and hence the diagnosis of a lymphoma. The molecular lab wasn’t routinely used. In the past they had been using mostly routine hematologic and maybe some immunophenotypic methods, but not molecular.
I figured that a lot of times we were able to provide the diagnostic pathologist with some information that helped him resolve controversies. It was interesting to me, so I learned more and more about it. Any time any new thing came up I would look into it as how can this help diagnostics, and so on.
What did you do after you left Brown?
After that, I went to NIH. Then I could use molecular techniques to ask questions that related to the molecular pathogenesis of lymphoma. When I came to the University of Utah, there’s a unique setting here, because the clinical division here is sort-of an organization called ARUP. They’re always looking to be as close to the forefront of new technologies as possible. So we did some work using some real-time fluorescence PCR, which also at that time was relatively new. We were able to incorporate them into some diagnostic work. Both my clinical interests and my research interests are related, and that was quite advantageous, because the disease entitites that we’re studying I’m quite familiar with. I think I know what the problems are that need to be resolved, and what would be a good way to resolve them.
What were some of the problems that you decided to work on first that needed to be resolved?
Well, first there was the technology we used called the LightCycler — and it was actually developed in the department of pathology at the University of Utah by one of the faculty here — we tried to use it to identify chromosomal translocations, which are often recurrent in hematologic malignancies, and they’re frequently characteristic. So if you identify a translocation that has a high association with a particular type of hematologic malignancy, that gave you additional support if you had a case that was sort of difficult to classify. That technique is real-time fluorescence PCR.
When did you start using proteomics techniques to study lymphomas?
So the simple thing is, most of the techniques that existed really addressed abnormalities at the DNA level. We considered that it would be fairly useful to have technologies that could interrogate at the protein level, especially if you wanted to screen for new markers of certain diseases, either in serum or in tissue — either way. So we looked at that platform as a way to evaluate potential candidates. That would’ve been probably in 2000.
What kind of proteomic techniques did you start off using?
First of all, we needed to decide what would be the appropriate platform for the kinds of questions we wanted to address. We decided that we wanted to utilize a technology that was versatile. Maybe it wouldn’t have the highest sensitivity on the market, but it would at least give us a very strong option to be able to identify the proteins. I talked to a number of people who were experts in the industry, and they recommended that I get an ion trap tandem mass spectrometer, because it would help facilitate the kinds of experiments I had described to them, that we would be interested in doing. So we got an ion trap, and it turns out it was a good decision, at least for our needs. It also turns out that no one mass spectrometer can do everything, and you have to tailor your needs to the instrument you purchase.
Were you doing quantitative comparisons?
We started by looking at qualitative comparisons, and then we started also into quantitative work.
What type of lymphomas did you work on?
We work on three different types of lymphomas at this point — one is called anaplastic large cell, the other is called follicular lymphoma, and the third is Hodgkin’s lymphoma.
In the case of anaplastic large cell, we wanted to figure out what the consequences of a recurrent chromosomal translocation are. What happens when this chromosomal translocation occurs in the cell? So we studied that.
In the case of follicular lymphoma, about 20 to 60 percent of patients undergo some kind of an event, whereby what is an indolent tumor characterized by survival in the order of several years worsens to a survival rate of somewhere on the order of weeks to months. We wanted to see if there was any type of protein signature that accompanied that.
So imagine I have a disease that seems like it’s just stable, and then later it just becomes dramatically worse. We wanted to figure out why it gets dramatically worse.
Can you disclose any results of your studies? What proteomic signatures did you find?
Well, we had interesting results. And we found that proteomics based methods can provide some novel insights into cancer pathogenesis. But the results haven’t been published yet, so I can’t disclose them.
You said before that one of your goals was to come up with new targets for drug therapy — did you accomplish that?
Yes, I think we’ve accomplished that. Well, I think preliminary results are promising.
Are you thinking of collaborating with any pharmaceutical companies?
Well, I think the most important thing is for us to validate that the target is a legitimate target and publish our work, and we’ll see what happens. I care more about the basic work. I guess if it “has legs”, as they say in the industry, mostly what we’d like to do is the target identification and validation. Perhaps somebody else would be interested in clinical trials and all that.
Are you planning on doing any clinical trials?
No. Not right away, anyway, but some other investigators would probably be interested.
Are there other projects that you’re working on?
There are several projects. As I said, there are three principal diseases our laboratory concentrates on. We’re also going to concentrate on some biomarker discovery for serum testing. I think that’s basically it.
Do you find that having a clinical background helps you significantly in your research?
Oh, absolutely. I know the diseases that I’m interested in. I think that research pathologists sit at the interface of being clinicians and being scientists. We can apply whatever basic technique we want to study our disease of interest, and I think that’s an important position to fill if you want to get new and relevant information about a disease.
For the validation studies, are you planning on doing a large retrospective study?
We have to think about that some. We might need to do several of those to be sure about some things. I think initially, what we hope to do is the discovery part of the work, and then somewhat simultaneously explore the validation to be sure that what we discovered is relevant.