The FORMULA-509 study was a multicentre randomised controlled study looking at patients who had a rising PSA after radical prostatectomy who had at least some kind of higher risk features. One of the standards for these patients right now, if they’re getting salvage radiation, is to give them six months of a hormone therapy, typically a GnRH agonist. What the FORMULA-509 study tried to find out was whether adding apalutamide and abiraterone to six months of a GnRH agonist could improve outcomes compared to adding six months of bicalutamide to a GnRH agonist. The two specific outcomes we were looking for were progression-free survival or metastasis-free survival.

To be eligible for this study patients had to have a PSA of at least 0.1 after a radical prostatectomy and then some kind of unfavourable factor. That could have been a Gleason 8-10 or node positive or a high Decipher score; it could also have been pathologic T3 or negative margins or a persistent PSA after radical prostatectomy or it could have been a short doubling time, less than 10 months, for example. So there were many ways where patients could get into the trial.

Once they entered the trial they were randomised 1:1 to receive… well, all of them received salvage radiation and six months of a GnRH agonist and the 1:1 randomisation was either to six months of abiraterone and apalutamide or the standard which was bicalutamide. 

The way the statistics were run in the trial, we powered it to find a hazard ratio of 0.5 for progression free survival using one-sided statistics with an α of 0.05. But, very importantly, we stratified the trial by two factors: one of them was a PSA greater than 0.5 versus less than 0.5 and the other was pathologic node positive versus node negative. The reason I emphasise this is that we pre-planned subgroup analyses based on these factors which turned out to be very important later and we used two-sided p-values for those subgroup analyses.

The key findings of the study, when we looked at the overall result of the study the primary endpoint was progression-free survival for the whole cohort. This came very close to meeting the endpoint but didn’t quite meet it. The hazard ratio was 0.71 and the p-value one-sided was 0.06, so we just missed our less than 0.05 cut-off. For metastasis-free survival something similar happened – the hazard ratio was 0.57 but we just missed the cut-off and the p-value was 0.05. So that was the overall trial – close but not quite statistically significant.

However, remember we discussed the subgroups. So, on the pre-planned subgroups, just looking at those patients with a PSA greater than 0.5 there was a very significant benefit in both the progression-free survival and metastasis-free survival outcomes. So for progression-free survival it cut the risk of progression in half. The hazard ratio was 0.5 and the p-value was 0.03. Even more importantly, metastasis-free survival for this subgroup of patients with a PSA less than 0.5 was very significant – the hazard ratio was 0.32 and the p-value was 0.02. I want to say again, these are two-sided p-values and this is a pre-planned analysis on a stratified subgroup.

So this is a very important result because we know that metastasis-free survival is a surrogate for overall survival. Just to stay on metastasis-free survival for a moment, at three years the difference was very large: there was an 18% absolute benefit in metastasis-free survival. That meant that we only needed to treat five patients with the drugs to prevent one metastasis at three years.

In terms of what to do with these results, I think that it provides a potentially nice alternative to lengthening the duration of hormones to 24 months. Why do I say that? The RADICALS-HD study was a very important study that looked at very similar patients and compared a similar control arm to ours which was just six months of standard hormone therapy. But instead of intensifying the hormone therapy they looked at a different strategy of lengthening the hormone therapy to 24 months and they found a significant benefit in metastasis-free survival. They found that if you go from 6 months to 24 months metastasis-free survival has a hazard ratio of 0.77. What we did instead was instead of lengthening the duration we intensified the therapy and we found in our primary endpoint a metastasis-free survival hazard ratio of 0.57. Now, I know it’s not correct to compare trials directly but, just to give everyone a sense of what’s the impact of intensifying treatment, is it similar to what the RADICALS-HD was achieving? Yes, our hazard ratio was lower. When we look at the patients with a PSA greater than 0.5 in the RADICALS-HD study compared to our study, I think in RADICALS-HD the hazard ratio was in the 0.5-something range for PSA greater than 0.5 but our hazard ratio was 0.32. 

So what it suggests is that I’m not saying that this is a head-to-head comparison where our hazard ratio is better but certainly our hazard ratio is at least comparable. So now if you’re a patient you have these two strategies in front of you. If you’ve got higher risk disease you can either lengthen your hormones to 24 months or you can stick with 6 months but intensify it, perhaps with apalutamide, maybe with abiraterone and apalutamide, like was done in this setting. It’s going to be more attractive to patients to intensify their therapy but keep the duration shorter to minimise the duration of side effects.

So this is going to be formally tested in a trial, the next trial is called the PROSTATE-IQ trial and it’s being led by Dr Karen Hoffman at MD Anderson and Dr Marisa Kollmeier at Memorial Sloan Kettering who were part of the FORMULA-509 trial. They’re going to test this two years versus six months intensified strategy.

The overall message for everyone is that for patients with aggressive features following a radical prostatectomy who are getting six months of GnRH agonist with their salvage radiation, we found that adding abiraterone and apalutamide for six months probably improves MFS and PFS but especially in that subgroup of patients with a PSA greater than 0.5 where a pre-planned subgroup analysis showed a statistically significant benefit to PFS and MFS. I do think that this strategy of intensifying the ADT for six months rather than lengthening it to 24 months is going to be an attractive strategy for patients and it’s something that’s going to be tested formally in an upcoming trial.