At A Glance
Name: Cy Stein
Position: Professor, medicine/urology/molecular pharmacology, Albert Einstein College of Medicine
Background: Assistant professor, pharmacology/clinical medicine, Columbia University Medical Center — 1990-2003; Staff physician, National Cancer Institute — 1989-1990; Clinical fellow, NCI — 1985-1989; Resident, internal medicine, New York Hospital-Cornell Medical Center — 1982-1985; MD, Albert Einstein College of Medicine — 1982 PhD, chemistry, Stanford University — 1978; BS, chemistry, Brown University — 1974
Cy Stein has spent almost two decades in the antisense field, work that has led him to RNAi. He recently spoke with RNAi News about his research efforts, RNAi, and Genta’s recent setback with its antisense cancer drug Genasense.
How did you get involved with RNA interference?
It was sort of a natural for me, because I had been in the antisense oligonucleotide business since 1987. I had done a great deal of lab work and worked with a number of companies on the development of antisense oligonucleotides, which of course are DNA molecules. So, it was a natural thing to do — to slide over into the RNAi world.
So, maybe I should step back and inquire about how you got started in antisense.
That started in 1987. When you get to a certain point in your fellowship you’re expected to start working in the laboratory. There was a laboratory at the NIH that was one of the very few laboratories in the world that was working with phosphorothioates, and I had a lot of experience with sulfur chemistry from my PhD. That was also sort of a natural thing for me to go back to.
Can you talk a little bit about what sort of projects you’re doing in the RNAi field right now?
There are obviously so many things, but what I’m most interested in is the idea as to whether you can develop RNAi [molecules] or siRNAs into useful pharmaceutical candidates. My particular interest is in the cancer-related field — there are obviously lots of other possibilities for these things — but it’s cancer where I’m most familiar.
Are you actually conducting projects in your lab at this point?
Yeah. For us, RNAi is extremely useful. I still have a great deal of interest in the mechanisms of action of various antisense oligonucleotides, and by themselves it’s almost impossible to figure out how these things are working — they’re very, very complicated. In the laboratory, the use of RNAi allows us a lot of flexibility in dissecting mechanisms that we ordinarily wouldn’t have. So, it’s particularly useful to me in that way.
We’re also very interested … the potential use of these kinds of molecules as drugs. While they have certain advantages over the corresponding antisense oligonucleotides, there are also certain disadvantages that heretofore people may not have considered very intensely. In my opinion, RNAi, it’s not a done deal by any means, that these things can be turned into drugs.
There are some big questions. For example, when you think about it, the technology, at least for siRNAs, has always been based on lipid transfection. That’s where you get most of your results from — virtually all of them, at least with siRNAs. It’s not true in that way with RNAi, but certainly with siRNAs. Many of the results you see when you lipotransfect, if you look very closely, are not necessarily due to siRNA effects at the level of the RISC. There are other effects; there are other signatures that you can see when you look at a microarray analysis. So, we’ve been involved with Dharmacon in trying to dissect out what some of these other effects are, and I think we have a fairly good understanding of what they may be and how you can see a great deal of transfection artifact that has to be accounted for.
The bottom line is, you can’t just throw siRNAs at everything and necessarily ascribe everything that you see purely to an effect of the siRNAs — it’s more complicated than that. And that’s not really surprising.
Another thing that we’re interested in is this whole idea of how you are going to get naked siRNAs, if you want to use them as therapeutic agents, into cells. You have the same problem here that you have with antisense oligonucleotides, which is [that] these molecules are charged and they’ll enter cells via absorptive endocytosis or fluid-phase pinocytosis. But, the problem then is: How do they get across the endosomal membrane barrier without some kind of delivery vehicle? That’s a real problem, a real tough problem, and it’s never been solved for antisense oligonucleotides in vivo. I think you’re going to have the same problem with siRNAs.
The other problem you’re going to have is [that] these molecules don’t adsorb to the cell surface all that well when they’re naked — that is, when there’s no carrier. Far and large, they don’t seem to bind to the heparin-binding proteins quite as well as, for example, phosphorothioate oligonucleotides do. That’s going to affect the uptake. That effect may be very, very sequence dependent in ways that are, at the moment, unknown and unpredictable.
These are all issues one has to think of in terms of the design of siRNAs for potential clinical therapeutic work. It’s not a minor thing to get it to work in a test tube, but once you try to get it out of a test tube and try to make a drug out of it, then you’re faced with a whole new set of problems.
You mentioned Dharmacon. Are there any other companies you’re collaborating with?
In terms of siRNA, it’s Dharmacon right now and, as you know, I’m on the SAB of Benitec.
There seem to be two groups out there: The people looking to turn siRNA into therapeutics and people looking to express shRNA. Do you have a take on the two different approaches and which you think is most viable?
You’re absolutely right — there’s two ways of thinking about it, two ways of doing it.
The problem that you’re going to have with the shRNA approach is the same problem people have been having with gene therapy over the years. It’s still a delivery problem — can you get this stuff into enough cells for long enough periods of time to make a difference? If you have to, can you treat more than once? What’re going to be the toxic effects of your delivery system? The gene therapy effort got into a lot of trouble about that. There’re weren’t too many, but there were a few very well-publicized toxic events that occurred because of, probably, some sub-optimal administration of gene therapy and its vector. I’m not saying that’s going to happen in this particular case, either, but one always has to be aware of that possibility.
So is the idea that, at this point, they both have their hurdles to overcome?
Yes, I think so. One of the things that strikes me very favorably about — I guess I can say this — about the Benitec approach in general, without giving away any specifics, is that they seem to be taking an ex vivo approach. So, you get away from the problems of systemic delivery, which have really hampered people for the entire lifetime of this business, which is now approaching 20 years.
Are there any lessons you think people should take from what’s happened in antisense?
Oh, yes. Just let me say that if they don’t take the lessons away, they deserve whatever’s going to happen.
I think the major lesson is that you can’t only approach the problem from one side. What I mean by that is: Just because the technology is very interesting and very useful and it works well in tissue culture doesn’t mean you can make a drug out of it. It doesn’t mean you can’t, but that doesn’t by itself necessarily mean that there’s a drug there. You have to have under your control not only the chemistry, but the cell biology and the medicine. This happens a lot of times at companies — they have one or the other. You have to have, in my opinion, experts at all levels. You have to understand the medicine, you have to understand the disease that you’re working in, you have to understand the business opportunities or the business problems of the disease you’re working in, and you have to understand very well what the regulatory climate is. If you don’t have experts at all levels in that process, you can be sure that you’re going to get tripped up somewhere.
A perfect segue opportunity. You played a major role in the development of Genta’s Genasense. Genta recently pulled the NDA for the drug [in malignant melanoma] after an FDA panel recommended against approval. What went wrong and, more broadly, what does this mean for the antisense and RNAi fields?
To be perfectly honest, as a member of [Genta’s] SAB, I cannot comment on anything about that, other than to say that the data is going to be re-reviewed over all the patients. My feeling is that [the company] will be back. I think there are some incredibly tantalizing results there that just can’t be explained away by statistical manipulation.
Of course, it’s very painful for everybody who’s been dealing in the antisense field. We have hoped, we had anticipated that we would see an approval in that respect. It didn’t happen the way we had hoped. But, there are some things there that make me as a clinician think that there is activity of this material. That’s my reasoning for why I think it’s going to be back for that indication.
In addition, one also has to recall that there are still two phase III trials with Genasense that will come out some time this year.
A lot of times it seems that the way people talk about RNAi, it’s the next step beyond antisense. Is this an accurate way to look at it?
I’ll give you an analogy — you may not want to print this. It’s sort of like some triumphalist Christians saying that Christianity is one step beyond Judaism. I’m not sure I’d view it that way. They’re really two parallel [branches] that stem from the same root.
So there’s room for them both to grow and exist?
Absolutely. They do different things. I am quite certain — I am 100 percent certain — that there are going to be cases in which you can’t use siRNAs or RNAi to achieve certain therapeutic benefits. There are other cases where you clearly can’t use antisense oligonucleotides, but there may be other instances in which you can.
When you’re talking about cancer therapeutics, people ask: Where is there an unmet need in cancer therapeutics? The answer is: Where isn’t there an unmet need? Yeah, there’re a few places, but there’s unmet need all over the place, and the trick is taking advantage of what we have and what nature has given us. The problem is finding out where these things can be slotted in. To do that, one has to have a very, very good understanding of the disease, and then there has to be an iterative process between the clinician and the scientist. The problem is those two folks rarely talk to each other.
Why is that?
They don’t speak the same language, and there’s biases on either side. Sometimes the clinicians think that the basic scientists have their heads in the clouds. Sometimes the basic scientists think that the clinicians are stupid. They’re both right and they’re both wrong.