Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Noel, it sounds as if there is more practice-changing data from the STAMPEDE trial; this volume effect seems to be important.

Noel Clarke, MBBS, FRCS, ChM: Yes, I think it is important, and I think the first thing to say is the critical effect of having large data, or a large body of data, which you can pull in and analyze, is a very powerful tool. That has enabled us to do a lot of work, which has been a collective team effort.

Firstly, we were skeptical about the charted volume data because, as Nick has said, we didn’t see heterogeneity of effects. In the STAMPEDE data, in M0 and M1, what we saw was that the hazard ratios didn’t shift, suggesting that you got the treatment and the effect; we saw this also in abiraterone acetate.

We started with a basic hypothesis, which is that if you pull in all the scans and analyze them, we can see what the effect is on a prespecified analysis. In abiraterone, we’d already published the data, and we knew that the M1RT [multidimensional item repo component] to STAMPEDE was not yet analyzed.

In the United Kingdom, we weren’t able to pull in the Swiss components of these scans for logistic reasons; however, we have a very good system with which we were able to circumnavigate the bureaucratic hurdles and pull in all the scans and analyze them according to specific criteria. We did a very detailed analysis: We mapped all the bone scans and the CT scans, looking at nodal metastases, bone metastases, MET numbers, and so on. Two very talented research fellows, who were training medical oncologists, did much of the hard work here. It was coordinated by the STAMPEDE team.

What we found, first of all, looking at the STAMPEDE M1RT data, is that there is a critical threshold based on 4 metastases and above. The interesting thing was that this was on classic imaging with a predictive bone scan CT. That tells us quite a lot about the biology. As Nick had alluded to, it doesn’t matter what the PSMA [prostate-specific membrane antigen] scan will show or whether they have microscopic lymph nodes. You get the effect and the big hazard, and you get the benefit.

That’s fantastic because it does really set the tone for how to manage oligometastatic patients. You might remember that, in the St. Gallen Conference, there was no APCCC [Advanced Prostate Cancer Consensus Conference] agreement as to what oligometasasis meant—was it bone, lymph node, or volume? We’ve got the standard benchmark now, which allows us to go forward.

On the abiraterone acetate side of things, we had a preplanned analysis in the M1RT, which was statistically valid and predictive. We were able to do a post hoc analysis using the same rules. As we did with M1RT, we looked at charted criteria and latitude criteria. What we found was that the effect was the same. We also found that with abiraterone acetate, if you use charted criteria and not latitude, and vice versa, there was an 18% difference in how you classified risk. Overall, they overlapped to the extent that there was only a 2% difference. In a population sense, it doesn’t make a difference how you classify it. In an individual sense, it can make a bit of a difference. In other words, you might exclude treatment using different criteria.

We found again that there was a homogenous effect when considering abiraterone acetate. The classification is from the mentally different. As a clinical community internationally, we have to think about how we classify volume, risk, and disease burden because there is no standard classification.

Nicholas James, MD: They’re all prognostic, aren’t they? The key thing here is that the charted classification is predictive of response to chemotherapy. Is the failure-free survival [FFS] effect the same even in M0, that limited their volume? As you rightly pointed out, this is a predictive response as well as a prognostic one. In order to reproduce the results you have to do what you do in the trial. If people start PSMA-PET scanning if they’re D4 staging, they’re going to have to do a bone scan as well in order to apply these criteria. What they’ll find, I’m sure, is that they have more than 3 metastases. It doesn’t matter because the data clearly show that if you’ve got up to 4 METs, you’ll benefit.

Until such a time when we can refine METs by doing further trials, it’s important when people do imaging trials that they don’t focus on positive and negative predictive values of imaging, and that they focus on whether altering the imaging changes the outcome, which is what we’ve done here. You shouldn’t change your practice; you should do what you did in the trial.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: They miss out on more curative, or at least longer-term treatments.

Nicholas James, MD: Depriving them of the treatment, which we know works.

Noel Clarke, MBBS, FRCS, ChM: That’s important to point out because there’s a classic Will Rogers effect about to happen, which is the restaging of patients who don’t necessarily need restaging, because, by all the criteria that we’ve used in trials, which is based on classical imaging, they’re going to shift, and that will have an adverse effect. The international community is going to have a problem if it doesn’t deal with it, which is that we don’t know the natural history of microscopic lymph node disease, which shows up for positive PSMA scanning. We don’t know what the natural history of that is.

Nicholas James, MD: The bone METs as well.

Noel Clarke, MBBS, FRCS, ChM: Of those 2 large-scaled trials, SBCG 7 and CRCP 3, which were high-risk clinically localized disease and where there’s a clear benefit to ADT [androgen deprivation therapy] and local treatment, a number of those had PSMA positivity and a high proportion, at least 20% to 30%, would have microscopic node positivity.

Nicholas James, MD: Absolutely. It wasn’t mandated to do a CT scan in PRO 7. They may have had loads of lymph node METs. We simply don’t know.

Noel Clarke, MBBS, FRCS, ChM: And PSA above 100 were allowed in that study as well.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: The idea that doing radiotherapy to the local disease in metastatic setting will yield the same effect across the spectrum, irrespective of imaging, is very interesting.

Transcript Edited for Clarity