department of molecular oncology at the John Wayne Institute For Cancer Treatment and Research
Name: Dave Hoon
Position: Director of the department of molecular oncology at the John Wayne Institute for Cancer Treatment and Research, 1997-present.
Background: Associate member of the John Wayne Institute for Cancer Treatment and Research, 1991-1996; assistant professor, University of California, Los Angeles, John Wayne Cancer Clinic, division of surgical oncology, department of surgery, 1988-1991; postdoc at UCLA 1983-1986.
Dave Hoon is the lead author of a study published in the Dec. 15 issue of Cancer Research identifying two proteins that may be associated with the spread of breast cancer to nearby lymph nodes.
According to the study, the over-expression of protein peaks at 4,871 daltons and under-expression of a protein peak at 8,596 daltons were highly predictive of lymph node metastasis.
The study’s authors believe that the over-expressed protein is thymosin beta-10, which other studies have associated with out-of-control growth and cell differentiation, and the under-expressed protein is a ubiquitin protein that suggests a good prognosis in node-negative breast cancer.
ProteoMonitorspoke with Hoon recently about the study.
Summarize what you did in the study.
Basically, what we were doing, using ProteinChips, [and the] SELDI system, we were looking at primary breast cancer where we were looking at patients with node metastasis to the lymph nodes and those without, and then trying to identify signatures that could separate the two groups. What we found was several proteins, but two were significant in that they could predict from the primary tumor if there was nodal metastasis. And when I say nodal metastasis, I mean early stage like sentinel nodes that are not clinically definable.
The uniqueness of the study is we looked at tissue and then defined it, but we micro-dissected the tissue so it was very carefully identified actually to be tumor tissue in the primary, which other groups haven’t really done that much of. And then at the same time, [we took] the proteins’ ‘omic’ profiles, and then identified these proteins, and then validated them. The advantage is you can identify patients, at the time of the primary tumor, who have a likelihood of disease spread, which wouldn’t be identified just by clinical or imaging or anything like that.
So it’s a test that looks at very early stage disease spread.
There’s a lot of biomarker work being done on breast cancer. How is what you’re doing different?
Well, one [difference] is most of the other groups are doing work on serum. That’s SELDI-based work. We’re doing it on tissue. Secondly, we did it on frozen tissue, but we micro-dissected the actual tumor rather than having normal end tumor, so it’s carefully dissected. That’s the other difference.
The third difference is the most critical. We identified the primary utility. Like I said, we had primary tumors, 65 patients and divided those with micro-mass in the lymph nodes and those with no metastasis in the sentinel node. What we wanted to do was [see if we] could predict the patients who are at high risk at the time their primary tumor was removed. And this way will allow [us to determine] who should have a more aggressive lymph node dissection versus those who may not have any tumor in their nodes so they don’t need any lymph node dissection.
So it has more clinical meanings. And more and more patients come with early-stage disease now.
You have not been able to completely identify the proteins.
No, we’re working on characterizing and validating them.
How close are you to positively identifying them?
Pretty close. We’re going to sequence them and … further validate them in a larger study.
You’ve tentatively identified them as thymosin beta-10 and a ubiquitin protein.
Right, those are potential [matches], but they may not be. They have similar sizes.
We already know that the thymosin beta-10 is associated with out-of-control growth and cell differentiation and the ubiquitin protein is associated with a good prognosis for breast cancer.
Right, they kind of fit the character.
If they are, in fact, the proteins found in your research, would that change the value of your study because it would then be validating something we know already?
Well, no, [these proteins] haven’t been identified in breast [cancer]. We’re actually identifying them in early-stage aggressive disease that has spread. So it takes it a lot further in identifying early disease spreading. Right now in breast cancer, there’s absolutely no marker … out there [with which] you can take a breast biopsy and say, ‘Your chances of having lymph node metastasis are this.’
This kind of moves the field forward.
What happens if they turn out not to be these two proteins?
Whatever protein they are, the marker, they’ll stand up, and we’ll just further validate them in a larger study.
What if these are two proteins that have never been identified?
Well, then it becomes more unique. But these days everybody is always chasing something that’s never been identified as new, but you know there are a lot of proteins that are known, but they’ve never been identified for their clinical utility. We often tend to go ‘Let’s try something new.’ But there are others that are never fully evaluated as relevant. It’s not always totally [about] finding a new protein.
Even if you find a new protein, you have to validate it and most of them never make the cut when you start doing clinical analysis.
We’ve done the clinical analysis already, so instead of saying, ‘We’ve found a new protein, let’s see if we can validate it,’ we’ve already done that in this blinded study where we divided [the patients] into node-positive and node-negative.
Do these two proteins have to act simultaneously — the underexpression of one and the overexpression of the other — in order for cancer to occur?
No, independent [of each other], they can predict, so they weren’t dependent on each other. At this point, we have not looked at them together with other prognostic factors.
You said you saw other proteins. Did you see any patterns that could be predictors of cancer?
They were of borderline significance. We’re currently validating them. We’ve also looked between normal and tumor also, so we have candidates for that too. Those studies haven’t been published but they will soon [be].
If these two proteins are validated as biomarkers for breast cancer, what would be the next step?
Then it would be to develop a test, so that we can use it … in a clinical lab. And then, if possible, convert it into a blood test, which would be another phase, just to see if you can do it with a blood test.