Metastatic Prostate Cancer: A Practical Review of Therapy - Episode 4

The Impact of DNA Repair Defects on Prostate Cancer


Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Just looking at the last section on biology now, Johann, I’m going to turn to you, because you’re an expert in this field. What do we know about the role of DNA repair defects—for example, BRCA1, BRCA2, ATM, etc—and the development of prostate cancer?

Johann de Bono, PhD, MB, ChB: The first thing to say is that our DNA is being damaged continually by hundreds of thousands of assaults every day, by lots of different carcinogens: smoking, alcohol, ultraviolet light. The reason we don’t get cancer is that we have DNA repair processes that are amazingly intricate, highly complex, and very precise. Cancer essentially happens largely because our DNA repair processes have not coped with an insult in some say, and those DNA repair defects have led to an accumulation of DNA damage that generates cancer. This is quite pertinent in prostate cancer. We now know for aggressive lethal disease that 10% to 15% of patients have inherited germline DNA repair defects, and these families have a higher risk for many cancers. For the BRCA type, we know that it’s breast, ovary, pancreas, leukemia, and other cancers. So, what we know now for sure, incontrovertibly, is that a high proportion of lethal prostate cancers have DNA repair defects.

The proportion is much, much lower for the more benign cancers. There is a much lower chance of heritable defects in the more benign cancers. This confused us for a long time. If you have a family history you have a much higher risk of a germline defect, but what’s interesting is that 7% of our patients at the Royal Marsden—a cohort of 150 with no family history—still have germline defects. What’s also fascinating is that many of the men that were diagnosed with lethal prostate cancer in their 70s with no family history still had germline defects. So, actually, our assumptions on age of diagnosis and family history were incorrect.

I think the biology is fascinating, because if you have a defect in these genes, you change the cells’ balance. The cell usually has balanced repair—you have a balance between high-fidelity repair and low-fidelity repair—and that balance is key. If you now tip the scale so that you have more non-homologous end joining repair of double-strand DNA breaks, that generates rearrangements. Prostate cancer is a disease of rearrangements, gene copy number gains, and genome mutations. And there’s clear evidence that what you have in prostate cancer is an imbalance with high-NHEJ (non-homologous end joining) repair/low-fidelity repair and low-high fidelity repair, not only in these cancers but actually these germline defects. In fact, we now know that a third of somatic prostate cancers have these DNA repair defects.

There’s actually other evidence emerging that there are other genes that are lost in prostate cancer, such as CHD1. We have just published a paper in Annals of Oncology that a CHD1 loss also impacts DNA repair. It often associates with SPOP mutations, hyperactive AR, and AKT signaling. So, you can imagine that if you have a DNA repair defect and you have hyperactive androgen receptor signaling, you’re driving transcription. That transcription usually needs to repair the DNA damage caused by transcription, it’s just a part of transcription. And if you’ve got an imbalance, that repair is done through non-homologous end joining, you get rearrangements at those androgen receptor response elements, and that drives prostate cancer generation.

Critically, today, these data will impact our focused targeting screening on populations at high risk of prostate cancer with these gene aberrations in their blood or in saliva. This is going to impact therapy with not only late-stage disease, but also adjuvant, neoadjuvant, PARP inhibitor, carboplatin, and maybe even radiation-type therapies with radium. We’ll have to get more data. At this ESMO meeting, we’ve seen some really interesting data confirming the results of our data published in the New England Journal of Medicine by Pritchard et al, from the group in Spain, showing even in Spain that 9% or 10% of patients with prostate cancer and advanced disease have these repair defects. We now have data from many groups, so I think there is no doubt this is real.

Transcript Edited for Clarity