Pierre Blanchard: We'll discuss testosterone recovery. Those are my disclosures, which are not relevant. And I'll start with a patient. And I'm sure you have all seen a patient like this in your practice. So I saw him more than 10 years ago. He was 72 at that age, no comorbidity, had high-risk—we'd say on the lower part of the high risk—localized disease, potent with no urinary symptoms at that time, and was treated with external beam and a brachy boost, and 18 months of GnRH agonist that ended in September 2014.

When I saw him in 2023, almost 10 years after that, his PSA was undetectable, but his testosterone level was 0.7, which is far below the normal value. It's supracastrate, but one could question whether it's really—it's clearly not normal. So we'll discuss testosterone recovery, we'll discuss the impact of ARPI (or abiraterone), predictive factors of time to testosterone recovery, and the impact of a longer time to recovery, and then touch upon very briefly on testosterone replacement therapy.

So first, agonists. I've selected just a few studies of those. The first one on the left is a retrospective study of two institutions in the US. And they looked at testosterone recovery as defined as the normal value. And what you see here between long-term and short-term ADT is that the recovery after the last injection of ADT—the median was between 17 and 24 months after the last injection.

And on the right, it's an analysis of a randomized trial where they counted the recovery starting at the first injection of ADT. So that's differences in how they count it. But to go to a supracastrate level, it took 15 months and full recovery 18 months. So that's much, much longer than the six-month theoretical duration of castration.

With antagonists, of course, it's much faster with degarelix. That's just one example of the use of two injections of degarelix for two months for cytoreduction prior to brachytherapy. And it took between three and six months to recover testosterone. And as you have seen before in the HERO trial with relugolix, the testosterone level went back to normal one to two months after the end of relugolix. Of course, it's much, much faster with antagonists.

Abi or other ARPIs do not have an impact on the time to testosterone recovery. And that's one example on the left of a study that combined preoperatively degarelix or degarelix plus abiraterone. And you see that the curves, the green and red curves. And on the right side, it's the EMBARK trial where you see that the mean testosterone levels are very similar between the two groups of leuprolide and leuprolide plus ENZA. So it's an indirect way to say that probably that does not impact testosterone recovery.

One thing that you may not know is that radiotherapy in itself has an impact on testosterone. So radiotherapy to the testes can lead to azoospermia. But it can also lead to a decrease in testosterone in patients who have received radiotherapy alone. You can see a median decrease in testosterone level by 30%, and 13% of the patients have a decrease in over 50% of their testosterone level. And the recovery takes up to six months as a median value.

And we've shown that the dose to the testicle is linked to the testosterone level. And it's impacted by the technique that you use in radiotherapy and by the volumes that you treat. If you treat pelvic lymph nodes, you give more dose to the testicles.

So, summary of this first part:

Testosterone recovery on average for agonists is 6 to 24 months after the end of theoretical efficacy, and some patients will never recover.Antagonists: 1 to 6 months after the end of theoretical efficacy.No impact of concurrent Abi or ARPI.Radiotherapy by itself can lead to a decrease in testosterone.So what are the risk factors? Just three studies because all of them are very consistent. First one is a Canadian analysis of two randomized trials. And as you can see here, the factors associated with longer testosterone recovery were lower baseline testosterone value, older age, diabetes, and longer duration of ADT. And that's the plot by ADT duration. The blue line is 0 months, green is 6 months, violet is 18 months, and red is 36 months. And as you can see, many of the patients do not recover testosterone when treated with ADT—more than 50% when given 36 months of ADT.

Similar predictive factors in the Spiegel study that I presented earlier: age, duration of ADT, baseline testosterone, BMI, and race. And in a study, the senior author being Dr. Morgans, as you can see here, the non-testosterone recovery patients were older, had lower testosterone levels at baseline, and were more likely to have diabetes, hyperlipidemia, and hypertension.

So what's the impact of a longer time to testosterone recovery? It has impact on quality of life, of course. And that's been discussed previously. That's the study that I just quoted. People who recovered that testosterone had a lower risk of diabetes, had a lower risk of depression, and had probably better sexual function. At least they were seeking treatment for sexual dysfunction.

But it can also—and it's a question mark—have an oncological impact. And that's a publication by D'Amico from patients included in a randomized trial in the late '90s. And what they showed is that people who had no comorbidities and a long time to recovery had better survival than the others. And that also shows that there is an impact of testosterone on the disease. That's not been shown consistently in all studies. But that's something that we need to keep in mind when moving from agonist to antagonist, for example, because the duration is not the prescribed duration but the actual duration of castration.

So in summary of those two last parts:

Time to testosterone recovery is associated with patient age, baseline testosterone, ADT duration, BMI, and comorbidities.Longer time to recovery impacts general health but may also improve DFS in patients without comorbidities, so that's something we need to keep in mind.And finally, testosterone replacement therapy—so I'm not going to go into the details, but please know that it's based on very small cohorts, low level of evidence. Those cohorts do not show an indication for an increased risk of recurrence. Of course, it's only in patients in whom we stopped ADT—so like locally advanced patients, for example. And it should be done only in men with a proven testosterone deficiency and bothersome symptoms of hypogonadism. And the shared decision-making with the patient is paramount because there is a theoretical risk of increased disease recurrence. This review is very well done, and it provides help and guidance towards the implementation of testosterone replacement therapy.

So going back to our patient, the patient had symptoms of hypogonadism. He was referred to endocrinology in 2023. He was started on testosterone replacement therapy in spring 2023. And it's a happy story. His sexual function and general well-being improved, and his PSA remained undetectable. But of course, his follow-up needs to be continued.

So as a conclusion, just remember that concomitant and adjuvant ADT improves survival in high-risk localized prostate cancer, and that those studies were done with GnRH agonists mostly. That testosterone recovery is long and uncertain with GnRH agonists, and it's faster with antagonists. Time to recovery is associated with patient age, baseline testosterone, ADT duration, BMI, and comorbidities. And the risk/benefit ratio needs to be discussed when planning testosterone replacement therapy, and of course, careful PSA monitoring.