Our trial is a Canadian Cancer Trials Group PR.21 study and we conducted the study to compare two treatments – Lutetium PSMA-617 and docetaxel. The background to this is that lutetium has proven efficacy in patients with metastatic castration resistant prostate cancer that’s progressing after an AR pathway inhibitor. However, also a standard of care in this setting is docetaxel chemotherapy and the two have never been compared. So, to inform clinical practice we conducted the study.
The study design was that patients had to have metastatic castration resistant prostate cancer that was chemotherapy naïve and they had to have been progressing after an AR pathway inhibitor. They also had to have PSMA-PET positive disease, so all patients had a PSMA-PET scan at screening, and they had to have adequate kidney, liver and bone marrow function. They also had an ECOG performance status of 0 through 2.
Patients were randomised 1:1 to receive either docetaxel at standard dosing schedule or lutetium, again at a standard dosing schedule, and they were followed for radiographic progression. Perhaps unique to the study, not only comparing an active treatment versus an active treatment, but also on protocol at progression patients could cross over to the opposite treatment. Then they were followed for second progression. The primary endpoint was radiographic progression free survival with the first treatment and then we had other endpoints including overall survival, tolerability and quality of life.
What we found is that we had 200 patients that were randomised, actually 199 patients that were randomised. What we found, interestingly, is that an equal number of patients received the treatment, however, about 15 patients on the docetaxel arm came off early because of treatment-related adverse events while only one patient on the lutetium arm came off early because of a treatment-related adverse event.
What we also found that was interesting, and will have a little bit of impact later as I talk about the results, is that there were more patients that crossed over from docetaxel to lutetium, 56 patients, versus 38 patients that crossed over from lutetium to docetaxel. So there’s a bit of a difference there. Post-protocol therapy, ten patients in each arm received subsequent life-prolonging therapy, so that was the same. So the big difference there is how many patients crossed over.
For the primary endpoint of radiographic progression free survival there was no difference, the treatments were the same. Median time to radiographic progression was about 8-10 months, the hazard ratio was 1.01. Now, on subgroup analysis there was no evidence of a differential treatment effect on various prognostic subgroups and we even looked at PSMA-PET SUVmean. So this is how much uptake there is on the PSMA PET scan for the PSMA radio tracer. We split the group along the median, so either a low PSMA SUVmean or a higher PSMA SUVmean, and what we expected, I’ll be honest, is that we thought patients with a higher SUVmean would benefit more from lutetium and a lower SUVmean would benefit more from docetaxel. But in fact we didn’t see that at all. It was prognostic, meaning that patients that had a lower SUVmean did worse but there was no differential treatment effect; both treatments worked equally whether they had a low or a high SUVmean.
The treatment-related adverse events were as expected for both but certainly there was a lot more toxicity in the docetaxel arm – 34% of patients had a grade 3 or 4 or higher treatment-related toxicity – while that was about half the amount in the lutetium arm. As well, there were two deaths that were considered treatment related and they were both from chemotherapy, from the docetaxel chemotherapy.
So all of that was not a surprise, we thought that there was probably not a big difference but that the tolerability of lutetium would be better. But what did come as a surprise is that overall survival was different. The hazard ratio was 1.64 and this was statistically significant, p-value 0.02 and it was in favour of docetaxel, for the patients randomised to initial docetaxel. Median survival was about 18 months versus 14 months for the patients receiving lutetium. Now, this could be because possibly of the differential crossover but this was an intention to treat analysis.
So overall we conclude that there is no difference in radiographic progression free survival between lutetium and docetaxel, although lutetium seems to be better tolerated, as well as the response rates were higher. Better tolerated, less patients were coming off for side effects. However, there is this overall survival difference, from a differential crossover effect possibly, but we really need to look at more data to see why this is.
What are the next steps?
We’re going to be doing further analysis, we’ll have longer follow-up on survival. We also have a lot of correlative studies, quality of life studies, and as well we looked at serial circulating tumour DNA, so we’ll look at some of the biological correlates to see if there are any other biomarkers that we could find that are prognostic or predictive.
What is the take home message?
Outcomes with lutetium are very similar to docetaxel in terms of radiographic progression free survival but it is better tolerated. There is an overall survival advantage if you get docetaxel first, so we have to really look at to why patients weren’t crossing over from lutetium to docetaxel. Was it because they didn’t want to – in fact a lot of patients don’t want to get chemotherapy – or was it because they couldn’t? Docetaxel can be hard to deliver after lutetium therapy potentially. So we really need to delve into this.
Regardless, I think we have to recognise that docetaxel chemotherapy is a standard of care, does improve outcomes and people should consider it more.
