This blog will be both naive, incorrect, and overly simple, so please forgive me for errors of omission and commission, especially if you are the sophisticated science type. I had the real privilege of being at the PCF meeting last weekend, and the progress in understanding the question is simply outstanding. Generally, we think that cancer starts as a mutation in a single cell. Mutations can be changes in a single nucleotide in DNA, such as changing ATTCCG to AGTCCG, or they can be the result of insertions like this ATTCGCGCGCCG or deletions like this A..CG. An example is the TMPRSS-ERG fusion story.
In addition, the chromosomes can become scrambled so that there is a portion of one chromosome that becomes connected to another. Maps of all of these phenomena have been created that are fascinating, beautiful, and (when you realize the complexity of trying to “correct cancer”) terrifying. You can read about this in depth here.
At the meeting, we learned that all of these types of things are going on. I would say, however, that the one unifying theme was how often the androgen receptor is at the heart of what is going wrong. For example, one scientist demonstrated that the AR is responsible for “opening the chromatin” (i.e. unwinding the DNA) to start gene expression, and that the other proteins that are “attracted” to this region of the DNA create breaks that allow the types of recombination illustrated in the figure. There is intense work going on to attack the androgen receptor, ranging from further inhibition of androgen synthesis by the cancer cells (such as abiraterone and TAK 700) to better blockade of the binding of testosterone (or DHT) to the receptor (for example MDV-3100 or ARN-509). In point of fact, we already know that we may be getting way more mileage out of androgen suppression than we previously thought as in this article describing a study I am involved with.
Beyond these insights, there are stories like the XL-184 (cabozantinib) wherein the targets we know about are working, but we aren’t even that sure we know why. (How important is the Met oncolgene and its ligand HGF, versus the VEGFR2 pathway, versus a number of the other pathways that are inhibited but to a lesser degree ??) So I came away with optimism that we are really getting somewhere now, but also with the humility to share with you that there will be no “cure” for PCa other than the removal of the disease before it spreads. What we can hope for is putting the genie back in the bottle for longer and longer periods of time. This is the first real progress we have seen since the initial observations of  “simple” castration and the ability to do it “better” with chemistry when we introduced leuprolide to the field.
Mike,
I am curious about the implications of the AR impacting unwinding of chromatin sections. Unwinding of course permits expression and the amount or protein synthesized in addition to occasioning insult. Are the researchers looking at protein expression issues in addition to mutagens? Perhaps we are generating a wrong amount of a protein and disrupting a critical pathway? This might imply a protein basis for disease perhaps in addition to mutations. Similarly, can we identify more accurately unwound sections permitted by AR? If so can we identify genetic variation and hence affected proteins and pathways?
Thanks
b
As I understood the data, (with admittedly rapid note taking and it is not in my wheelhouse for expertise…) the findings were that chromosome breakpoints more often occur near AR binding domains than would be expected. They weren’t looking at critical downstream proteins, although there is lots of that type of investigation going on. Some of the nicest data presented was from the Michigan SPORE where they are looking at DNA,RNA, and protein expression all in the same tumors and comparing to the normal situation for that patient. An immense task, and likely to lead to interesting insights for “reality testing” the promise and pitfalls of “personalized medicine”.
Pingback: MDV 3100 | prost8blog