Siddhartha Mukherjee: So this goes back a little bit to the cancer stem cell theory. Scientists have discovered that some forms of cancer such as leukemia have their own stem cells and may be able to regenerate themselves and this might be one reason that they relapse after chemotherapy. Tell us about the cancer stem cell theory and what is known and unknown about it. Anyone?
Lewis Cantley: I'll start on it and then you can correct whatever I get wrong. The cancer stem cell really I think emerged from observations of liquid tumors, leukemia, lymphomas where it was clear that the cancer was emerging at a partially differentiated stage of the tumor, and in tracing back what antigens are on the surface, one could determine that there was a subset of those tumors that had antigens that looked like a normal lymphocyte stem cell and that that subpopulation could continually self renew while the somewhat differentiated descendants of those cells would ultimately quit dividing, so the bulk of the total number of leukemic cells were not actually capable of indefinitely propagating. I think the data for that is extremely strong, and so some of the therapies that look like they’re working beautifully because you reduce the total number of cells by 98% or so, but you’re not killing the ones that are actually keeping this tumor going, so that is why there has been so much excitement about this. So we need to get therapies that hit the cancer stem cell, not just the ones that are descendents of it.
As we look at other types of cancers like melanomas on the other hand it looks like almost every single cell is a stem cell, that they are all capable of self renewing, so the stem cell idea is very important to understand and certain- some therapies and other therapies it’s probably not really relevant.
Siddhartha Mukherjee: So once again we have a heterogeneity problem.
Harold Varmus: One of the reasons there has been so much controversy here in my view is that stem cells of course ignites public excitement and even excitement among scientists. The real issue and I think Lou described this fairly is we know that cancers can recur. The question is can any cancer cell lead to a recurrence or are there subsets and if there are subsets that is obviously of great therapeutic importance because you really have to eliminate the cells that will generate the cancer; it’s not enough to measure the total bulk of the cancer cells. And I think the evidence, as Lou has said, is that some cancers virtually any cell going into the right environment, the environment make a big difference here, can regenerate and the tests are very hard for doing this. You have to know that when you’re introducing cancers back into an animal host, is that the appropriate animal host for testing? Frequently these are human cancer cells going into animals. What is the immune system of the animal like? What is it respecting and not respecting? So I think these are...the concept is very important because we need to know if cancers are heterogeneous—which goes back to their ability to recapitulate a cancer—we’ve got to know how to address the cells that are responsible for regeneration.
And the other thing, just to introduce a concept we haven’t talked about, drug resistance. Frequently it’s thought that those are cells that are resistant to therapies and in other words, one facet of so called drug resistance, which is a huge problem in cancer therapy today.