Read the Full Video Transcript

Anwar Padhani: Good afternoon, everybody. I too would like to thank Silke and Aurelius for inviting me along today. Here are my disclosures, and as you might imagine, it's imaging companies.

We've all heard consistently I think over the last two days that bone scans and CT scans lead to poor confidence in knowing what is the true clinical state of patients not only before you start a treatment, but also during the treatment. This can have direct precision oncology implications.

One of the biggest problems with using bone scans is this new lesion that appears on the first scan, so this is the unconfirmed first follow-up study. What is this? Well, this could be a progression. It could be tumor momentum, in other words, temporary worsening of the disease before the ADT, which is a cytostatic, kicks in. Or it could be what we think it might be, which is a bone scan flare. In other words, a favorable response manifested because of an osteoblastic reaction.

And because there are three possibilities, the two by two rule was introduced. If you saw two lesions, then on the next scan you had to confirm that by showing two new lesions provided the two lesions were on the first scan and provided there was no clinical progression. Okay, so we applied that rule here. Nina, in fact, applied this rule here and you can see that over time this patient does respond to treatment.

This flare phenomenon, representing a response has been now well documented. Both for abiraterone as well as for enzalutamide. And here is the data published by Mike Morris showing that if you have unconfirmed new bone lesions, you actually live longer than if you confirm. If you have two plus two then you're clearly progressing and you will live less long. That's quite nice validation of this phenomenon.

There's some nuance to this as you might imagine, and it turns out that if you've had no prior docetaxel, if you've had no prior docetaxel, so for example, the PREVAIL study, that when you saw this unconfirmed first two new lesions, then it probably represented a pseudo-progression phenomenon. In other words, something like a bone scan flare. And importantly the same lesson that you should not prematurely stop your treatment. But if you saw, if you did the same treatment, enzalutamide post docetaxel. In other words, the AFFIRM study, if you saw those two lesions, they were probably progression. And that's the important thing.

What strategies can you use to talk about the bone scan flare? Well, you can apply a two plus two rule, but there's a problem, you have to wait for a followup scan, which may be 12 weeks later. You can apply your clinical acumen and we'll talk about that in a second. Or you can do something sexy like a functional imaging.

How good are you guys at assessing how well a patient is actually doing? And there was a hint of this this morning. In hormone-sensitive prostate cancer, it turns out that there is a discordance between what you think and what the imaging is showing in about 30 to 45% of patients. Just imagine that. That should come as a shock. But you know this is one of your studies, and in castrate-resistant prostate cancer, and we saw an example of this this morning, the discordance between clinical which includes cross-sectional imaging and PSA, which is what you guys are using, that discordance is between 20 and 30% of patients. There is, you guys are not as great as you think, imaging ought to be used.

Okay. This flare phenomenon is not only an exclusive bone scan phenomenon. In fact, it's been described with multiple traces such as sodium fluoride PET and as well as for just ordinary CT scans also. You can also see a flare phenomenon using PSMA. All the PSMAs, the new kid on the block? There is a flare phenomenon with the use of ADT believe it or not. But there doesn't seem to be a flare phenomenon with another technology which we have not really touched on, which is MRI, particularly diffusion rated MRI.

Just to show you what the flare phenomenon and how you might deal with it. For example, here is a patient who was screened with a PSA 45 and at week 25 on enzalutamide, you can see the PSA is nicely suppressed, but the MRI has shown a new lesion. How do I know this red dot at week 25 is a metastasis? Well the good thing about an MRI is you can do some imaging at that particular spot and show that this is a new lesion, wasn't there on the prior two scans. But of course, the bone scan is negative. The patient continues on enzalutamide. And then on the next scan, you see two lesions. Now, this is outside the flare period but still needs to be confirmed. And of course, at week 49 the patient changes treatment.

Okay, so what happened here? Well, the date of oligo-progressive disease is, in fact, week 25. The date at which we change treatment was week 50 and of course what gets recorded in the notes, and this is the fudge factor, is the middle scale, in fact, as it turns out. This need to wait to confirm progression doesn't serve you well when you are trying to do precision oncology. Which is why the images and we, in the UK, have been pushing whole-body MRI as a technique.

Now whole-body MRI is a way of displaying the cellularity of the tissue and also telling you about cellular health. And we can color code it in a traffic light system. Now we use a traffic light system because Johann de Bono wanted three colors, red, yellow and green. Because I think he drives to work and he wanted to know how to treat patients using a traffic light system. And we have color-coded, alive, dying and dead tissues. And you can, in fact, spin these around because these are really sexy. There we are, spinning images. Now you can see how in this particular patient, after four cycles of docetaxel, there are still red areas. These represent areas of treatment resistance and we can quantify those. You can see before treatment, 95% of the tumor was alive. After four cycles, 35% of the tumor is alive.

Now, this is nicely validated. Again, Johann de Bono's group and Nina Tunariu who is sitting in the front here did biopsies and correlated the number of cells in those biopsies at that site with imaging features. And showed that if you have a low ADC, water moves less or if a high signal on the diffusion rate sequence, it correlates with cellularity. Now not only that, if you see areas of resistance and you drill the bone in that area, you can find mutations that are actionable. One of these mutations is, of course, the BRCA2 mutation, which you can treat with olaparib and you can show a response. And when the patient relapses, if you drill again, in a different site on the right side of the image, the BRCA2 mutation is gone. In other words, you can start using this information to promote precision oncology in your patients. And if you keep doing this in patients, so here's the same patient looked at seven times in this particular instance, you can map the amount of tumor that's alive, dying and dead over time with different interventions. This is precision oncology in practice.