at the single-cell level. So I’m most excited
about technologies that are addressing these
questions at a single-cell level—in a collection
of cells, but at a readout of the single cell.
I think that’s where the action’s going to be.
KP: To us, one of the biggest questions is
how heterogeneity is important for clinical
behavior of the tumor in terms of progression,
therapeutic responses, and also, trying
to understand what sustains heterogeneity.
I mean, what is the reason why you have
particular combinations of clones or mutations
and so on? So we have some data from
experiments where we’re putting in a heterogeneous
mixture of tumor cells with a known
driver, and it’s not becoming homogeneous.
So then the question we’re asking ourselves
is why is it that they don’t have to become
homogeneous to drive the tumor? Because
that has been the accepted view—the driver
mutation becomes the dominant clone; you
don’t have to care about the rest. And I think
the experiments tell us many cancers don’t
conform to that.
To get back to the original question about
technology, there are still things you cannot
do on single cells. Like whole-genome
sequencing; it’s not there yet in terms of having
mutation data. My dream experiment is
to do whole-genome sequencing in situ, if
we’re going to get there, but I think the way to
overcome it is to do whole-genome sequencing
on the bulk and then have technologies
that allow you to look back at the mutations
in single cells. Because we have these studies
coming out, they have hundreds of mutations
per tumor. Are they in the same cell? Or
what fraction of the mutations is? Because
everybody assumes that you need five or six
mutations to get a tumor. But then, why do
we have so many mutations? And then, what
is the composition in different cells and what
are the different combinations?
To me, those are some of the big questions
that we can address, technology-wise. We
cannot fully do everything yet, and of course
there’s the cost of sequencing as well, which
can still be high for single-cell work. But I
think that’s the way to go, to look at single
cells.
AJI: I think we have to develop a few additional
tools, especially the ability to analyze
whole genome sequences from single cells
to fully understand genetic heterogeneity.
Emerging techniques, such as immunohistochemical
analysis with mutation-specific
antibodies, will allow us to perform in situ
studies of heterogeneity of common driver
mutations. We also need multiple sets of