Read the Full Video Transcript

Alastair Lamb: If we look at the molecular basis for cancer behavior, as Matt said, "Where might we draw that line?" Very briefly, we've used spatial transcriptomics to create a high-resolution clonal map of two prostates. The beauty of this approach is that it allows us to observe the whole transcriptome at near cellular level to infer genomic copy number status, and because the tissue is intact, all of this can be underpinned by very detailed consensus pathology. I'm going to focus on one of these prostates, and you'll see why because it displays some of the typical heterogeneity that we see in prostate cancer.

First, let's zoom in on the square over on the left there, that's H2_1, which includes in one five-millimeter square benign grade group two and grade group four disease. We get several thousand whole transcriptomes, these spots each covering about 10 cells. Now, here's the pathology on the left here. Blue is benign, red is cancer, thanks, and we allocate each spot to a clone based on copy number status, and then translate these back to their tissue location.

In this case, one of our most striking findings was that many of the mutational events commonly associated with aggressive prostate cancer, such as c-MYC amplification on chromosome 8, PTEN loss on chromosome 10, are already evident in areas of benign non-transformed tissue. Most notably, the yellow clone, C, which you'll remember as you look on the left, is in that blue benign area. We're then able to construct these clone trees plotting the branching lineage from benign to cancer. Where do we draw the line? We'll go back to our organ-wide map, and this time, let's focus on the section at the bottom there, H1_2, which is purely grade group one disease.

With spot level clone tracking, we see a striking difference this time. Taking, for example, benign clone E, which gives rise to grade group one clone F, that's the orange one. These pattern 3 areas lack the events we previously saw in chromosome 8 and 10. At least, in this patient, grade group one is a molecularly different cancer. Let's average all these spots together using a Manhattan-style plot separated by grade, perhaps more familiar to some of you, and there is a clear difference between grade group one and high-grade disease. Indeed, if we now include the non-malignant histology, that altered benign region we saw, and the pin, actually, they look more concerning than grade group one. This leads to an important question, Matt, you asked me to finish with a question. Is there something about the mutations that lead to the development of grade group one, which is sufficient for cellular transformation, loss of basal cells, but which effectively drives these glands to indolence? Perhaps this makes it impossible for these cells to invade and subsequently spread. Thank you.