At A Glance
Name: Sharat Singh
Title: Incoming chief scientific officer in oncology at ViroLogic
Background: Tech fellow at Syntex/Dade-Behring — 1987-1997; Aclara — 1997-2004
Education: Post-doctoral fellowship at Columbia University, New York — 1987; PhD in Organic Chemistry from The Indian Institute Science in Bangalore, India — 1985; MS in chemistry from The University of Hyderabad, India — 1980
It’s true that Aclara had numerous attractive technologies, but as ViroLogic acquired the company, its focus was fixed on the patient-identifying possibilities of Sharat Singh’s eTag assay system.
Singh is joining ViroLogic as its new chief scientific officer for oncology as the company tries to take eTag into the world of personalized cancer treatment.
Before coming to his recent position at Aclara, Singh contributed to a number Syntex patents, including that for the company’s Loci technology. At Aclara, Singh was the key inventor on a number of patents in addition to eTag’s, including the Arteas microfluidic chip technology and the Plurex multiplexing system.
How long was the eTag technology in development?
Approximately five years.
What do you see as its advantages in identifying responders to cancer drugs?
The tests which are right now being used are immunohistochemistry and FISH. So, IHC is used for Herceptin, a drug for her2-expressing breast cancer patients. For that they use IHC to look at just her2 expression. This is a test from Dako and Ventana.
For the same drug, there is an other test called FISH, which is again supplied by Vysys, Dako, and Ventana. So that’s just gene amplification which is observed there. And then, the other tests which have been approved are Bcr-Abl targeting IHC for Gleevec, C-kit for Gleevec. [There is a test] approved for Erbitux, which was approved for colorectoral cancer — that test was really not a very good test.
The methodologies which are being developed by people, and where a huge amount of effort has gone into, is looking at downstream footprints, as it is called.
So, just a little bit about cancer, to start with — cancer is rapidly proliferating cells; abberrant signaling pathways which are activated, which lead to rapid cell proliferation.
These aberrant signaling pathways are typically activated because the receptor tyrosine kinases, which are actually on the membrane of the cell — her2 is an example, for which Herceptin is a target. EGFR is an example for which Erbitux, Terceva, Iressa are the drugs.
So these actually become non-functional. They start misbehaving due to mutations or other things. This leads to rapid proliferation.
So, the techniques like gene expression, where you’re looking at downstream footprints of these aberrant receptor tyrosine kinases — people have tried long and hard trying to use gene expression technology, using primarily Affymetrix and other technologies, and have not been successful at all.
This has not led to identification of any clear footprints of these receptor tyrosine kinases which have become dysfunctional.
The effort — the work — goes on. People are still working on it — they’ve been working on it for the last … so many years. And the companies which are involved in it are Roche, Celera Diagnostics, Abbott Diagnostics, a lot of companies are involved in it.
Then the other approach has been to do mass spec profiling — profiling of the proteome using mass spectrometry. You can think of a number of companies trying to do that.
When you’re trying to select patients for targeted therapies, until now, these approaches have not been successful. There has been a lot of effort which has been put into it.
The third approach is to detect phosphoproteins, which also identifies signaling pathways. There are two ways of looking at signaling pathways — one is looking at phosphoproteins, another one is looking at protein-protein interactions. There has been a lot of effort, a lot of companies which have come up, which have developed antibodies for specific phosphoproteins. This huge effort has not yielded anything.
The other way of looking at signaling pathways is to look at protein-protein interactions. There was no technology out there which could do that. So eTag technology is unique — it is the only technology which can look at protein-protein interactions in formalin-fixed, paraffin-embedded tissues, which you can access from tissue biopsies from cancer patients. That’s the basic advantage.
We’re directly looking at activated targets, we’re not looking at any downstream footprints. If you say that the target is attacking an apparent signaling pathway, which is initiated by dysregulation of EGFR, then we look at EGFR dimers as an activated marker for selecting patients who would respond to anti-EGFR therapy, like Iressa, Terceva, Erbitux, Panitumumab.
The biology is that EGFR expression does not code for activation of a signaling pathway. In this instance, what you have is EGFR needs to bind to a growth factor, and there are multiple growth factors. There are about 15 to 20 different growth factors — so you can’t detect all of them.
The other part is that you could have dysregulation of this pathway because of overexpression of the growth factor, overexpression of the receptor, mutations in the receptor, mutations in the receptors which prevent endocytosis — downregulation of the receptor from the cell membrane.
All these lead to a highly activated pathway coming from EGFR, and the best way of detecting it is going out and looking at dimers. You’re not looking at protein expression, but protein activation, and to my knowledge, we’re the only company which does that. The rest of the people are all going after downstream footprints using either mass spec or using gene expression arrays. Or using phosphoproteins, which, again, all these do not yield any good results.
Do you expect any of these techniques to have the potential to yield useful results?
When we initially started the project, we looked at everything. When we did that, one of the things which we noticed was, in tumor cells, you have a number of pathways which get activated — it’s not just one pathway. Very rarely, a tumor is driven by just activation of one, single pathway.
So when you have these pathways activated, each pathway upregulates or downregulates a set of genes. And now, you’re going so much downstream from the cell membrane into the nucleus that the information — because you have started multiple [steps] from the cell membrane — has gotten so complicated that it’s very difficult to figure it out.
The reason why I say that is, just think about the amount of money which has been spent on gene expression and mass spec over the last few years. It’s like NCI funding it, Affymetrix funding it, Celera Diagnostics funding it, Roche funding it, Abbott funding it, including all the pharmaceutical companies and biotech companies. Nothing has come out of it. So if you look at the amount of effort which has gone in, and then you look at the results, it’s not very satisfying.
When people have used gene expression, they have done it in a very smart way. For example, there was a test for breast cancer which got a lot of publicity recently, but there, they’re looking at ER.
For breast cancer, people who’ll respond to chemotherapy, or not respond to chemotherapy, or who need chemotherapy, or not need chemotherapy — most of the genes which are selected are very well-known proteins which have been used in IHC. It’s a multiplex version of that.
So, then, will eTag actually produce results — with less money and effort?
Yes. The advantage is — pharmaceutical companies have developed these drugs to target a particular target, which is based on years of research. That target is somehow aberrant — it’s overexpressed in these cancers, or it got dysregulated by multiple mechanisms. What we are basically going in and saying is, ‘We’ll use that information, and develop assays where we can look at an activated form of that pathway.’
And using protein-protein interactions to do that is a fairly novel way of doing it, especially in formalin-fixed or [fresh frozen plasma] samples. We have filed a lot of patents on that. We kind of believe we are in a position to do something which nobody has done before.
Why focus on EGFR right now?
There is a need. Every year about 175,000 patients are diagnosed with lung cancer. Out of that 140,000 have non-small cell lung cancer. And if you look at the percentage which gets a targeted therapy, most of the targeted therapies are targeted at stage two and stage three, which is approximately 40,000 patients.
The important aspects about these targeted therapies are: they’re not very toxic; and if the patient has the target, these therapies can make a substantial difference to the patient. So, if you can select the right patients, this is going to have a huge effect.
Based on some of the clinical data which has come out, you have approximately 10 percent of the patients who are responding, and these are all stage two and stage three patients. As soon as they are diagnosed, they are given chemotherapy.
When they are not responding to chemotherapy, they are given these targeted therapies.
If you can identify these patients as soon as they’re diagnosed, and treat them with targeted therapies, the probability of response in these patients increases dramatically. Very good for the patients.
It’s very good for the pharmaceutical companies because you have a patient population which is going to take the drug much longer, they’re going to survive longer.
Ultimately, we need to cure this dreadful disease. This would be one step forward towards that.