Read the Full Video Transcript

Amar Kishan: Thanks very much for the kind invitation to talk today. I'll be going over the role of metastasis-directed therapy in oligometastatic disease, primarily in the BCR setting. It's not all exclusively based on PSMA, but more focused on the data that we have so far. Here are my disclosures. Sorry. There we go. There we go. Apologies for that.

Okay, so what is oligometastatic prostate cancer? Well, it's a disease state that's characterized by the presence of a limited number of clinically detectable mets, generally one to five. It can either be synchronous (de novo) or metachronous (recurrent after definitive therapy) and conceptually can encompass a number of different factors. It can be truly indolent disease biology with limited polymetastatic potential, aggressive biology identified early in the course, or subclinical disease that has been identified by increasingly sensitive imaging.

What's the goal of MDT? Most patients with prostate cancer recurrences after primary treatment tend to relapse with three or fewer identifiable lesions. Ablating all sites of disease in these patients might delay the initiation of systemic therapy and its attendant adverse effects, or if they're already on systemic therapy, it might prolong the interval to the next line of therapy. If a patient truly has a limited amount of disease, it is conceptually plausible that MDT could be curative in the same fashion that salvage radiation of the prostate fossa is curative after radical prostatectomy.

I'm going to go over the randomized data for MDT in a number of different disease settings and then some prospective trials that focus just on PSMA.

The STOMP trial we're all probably very familiar with is a randomized trial, 62 men randomized to MDT or surveillance in the setting of having three or fewer lesions detected on choline PETs. Most of these patients got SBRT, but some got surgery. Importantly, about 58% of patients on the MDT arm had solitary lesions, 42% had node-positive recurrences, and 45% had M1b disease. The median PSA at enrollment was actually quite high, 5.3 in the MDT arm and 3.8 in the surveillance arm. The primary endpoint was ADT-free survival, and the initiation of ADT was symptomatic progression, progression to more than three lesions, or local progression.

You can see from the primary endpoint results published in 2018, median ADT-free survival was prolonged from 13 months to 21 months with the addition of MDT. No symptomatic or local progressions, no grade 2 or 3 toxicity. Polymetastatic progression was the main mode of failure. About one-third of patients did undergo repeat MDT.

A five-year update that was presented about four years ago found that there was a consistent benefit in five-year ADT-free survival, so 34% versus 8%, but this did not translate into other outcomes like castrate-resistant disease.

The ORIOLE trial was a second randomized trial. Here, patients had three or fewer lesions on conventional imaging and were randomized to either SBRT or surveillance in a 2:1 fashion. All patients got a PSMA PET, but this was not known to the treating physician at the time of treatment. 58 to 67% had node-positive disease. There's a range there because it wasn't entirely clear in the tables who was N1 versus M1a. The median PSA at enrollment also quite high, six on the MDT arm, seven on the surveillance arm.

The primary endpoint was disease progression at six months, and this was a composite outcome triggered by PSA rise, radiographic progression, symptomatic progression, or initiation of ADT.

You can see here the composite PFS on the left and then the biochemical PFS on the right. Both improved with the addition of MDT. The six-month progression rate was about 19% with SBRT versus 61% with surveillance. The median PFS was not reached versus about six months with surveillance. Interestingly, in patients who had MDT, about 16 of the patients had non-treated lesions on the PSMA. If you looked at the six-month progression rate among patients who had no untreated lesions, it was only 5%, whereas it was 38% for those who had untreated lesions. There appeared to be a benefit to consolidation in patients who had oligometastatic disease by PSMA as well and all of it was treated. The six-month distant metastasis progression rate was also significantly different, about 15.8% if there were no untreated lesions versus 62.5% if there were untreated lesions.

There was also some evidence, though provocative but not confirmatory, that there was some immunogenicity elicited by the SBRT. There was a differential clonal type abundance, more pronounced with SBRT, significantly more expanded clones at the 90-day mark. In the SBRT arm only, the baseline T cell clonality was associated with less progression at six months.

Recently, they pooled data from both trials and published this meta-analysis of about 116 patients that are eligible. You can see the pooled PFS is improved with MDT from about 5.9 months to 11.9 months. Other endpoints were not significantly different, including radiographic progression-free survival. Remember, PFS could have been triggered by a PSA rise or initiation of ADT for any reason, and that might explain that difference. Notably, 70 patients across both trials did have tissue for next-generation sequencing to evaluate for the presence of a high-risk mutational phenotype. That was defined by pathologic confirmation of mutations in ATM, BRCA1/2, retinoblastoma, or TP53, and having any of those portended a worse prognosis regardless of treatment arms. If you look at the left there, the high-risk mutational phenotype, they're failing very early. However, the hazard ratio for adding MDT was 0.05. That's a huge hazard ratio compared to the low risk where there's less failure. The hazard ratio for MDT benefit is more muted at 0.4. It's still an impressive hazard ratio but not nearly as large.

Summary of these trials: They compared against observation, MDT nearly doubles PFS, median increase from six to 12 months. If PSMA is available for all lesions, MDT further improves progression rates and potentially MFS, but long-term responses are rare. There may be some predictive biomarkers out there.

Let's shift gears to the ADT setting, the EXTEND trial published late last year. 87 men with five or fewer lesions randomized to MDT plus intermittent hormone therapy versus intermittent hormone therapy alone. All men were already on ADT for two months, and the plan was to continue this for six months. Most had conventional imaging, a handful had fluciclovine, 72% had recurrent disease, a handful had de novo disease.

Here most of this was M1 disease, and most patients had more than one lesion. You can see the PSA is actually quite low, however, less than two compared to the other trials that I just presented. Some patients had castrate-resistant disease and again, a composite outcome of disease progression at six months.

Looking at the PFS curves here, you can see on the left the median PFS was not reached in patients who got MDT, whereas it was 15.8 months in patients who got intermittent ADT, and generally that was six months of duration. Importantly, look at the eugonadal PFS. This is men who have achieved testosterone recovery to 150 or higher. You can see in those who didn't receive MDT, it was about 6.1 months, very consistent with that pooled estimate from STOMP and ORIOLE. On the surveillance arms, it was 5.9 months, but it was not reached in patients who got MDT. It's an impressive result.

If you look at the subgroup analysis, probably the most important is where we're looking at whether patients got a second-generation ARPI or not. Even if they did, there was still a benefit to MDT. Again, they did find an immunologic component to the SBRT treatment or the MDT treatment as well, in this case, corroborating the findings from the ORIOLE trial.

The last randomized study I'll mention is the ARTO trial, which just came out. This was a trial in castrate-resistant prostate cancer. 157 men with three or fewer non-visceral lesions were randomized to MDT plus abiraterone versus abiraterone alone. Most of these patients were staged with choline PETs. A handful had fluciclovine or PSMA. 50% had node-only disease, and 37% had one lesion. The median PSA was about 3.4.

The primary endpoint here was a biochemical response with a 50% reduction in baseline PSA, but PFS was a secondary endpoint. If you look at the biochemical response, there's a 92% response in the MDT arm versus 68%, which is still a good response in the control arm. They did a univariate analysis to look for other predictors, and the only thing that was significant other than treatment arm was the basal PSA and on an MVA, it was really the treatment arm that was predictive of improved biochemical response. If you look at the PFS curves, I would say these are pretty impressive. In the MDT arm, the median was not reached, whereas it was 17 months in the abiraterone-only arm. There was similarly a kind of almost a mirror reflection kind of prolonged time to progression onto the next line of systemic therapy.

In summary for these settings, MDT prolongs PFS in the context of intermittent hormone therapy in men with de novo or recurrent disease and also prolongs PFS in men with castrate-resistant prostate cancer in the context of second-generation therapy, namely abiraterone.

Now, I'll wrap up with just a few lines on some trials that focused on just PSMA. These are Phase II trials. The first I think was alluded to earlier. The MRgRT trial from Canada, a prospective study of 72 patients with rising PSA after prior therapy, generally prostatectomy, and had negative conventional imaging. All patients had a PSMA PET, and those with five or fewer lesions were candidates for MDT. It turned out that about 37.5% of patients met that specific criteria. Overall, 49% of patients had one lesion detected, and only regional nodal disease was seen in the majority of these patients. The primary endpoint was biochemical response.

You can see here the PFS curves for the patients who received MDT. The biochemical control initially was 60%, but the biochemical no evidence of disease at longer-term follow-up was only 22%. Even with PSMA consolidation, it's a little bit different than the very promising early results in the ORIOLE trial. At long-term follow-up, it's a minority of patients that are disease-free.

The SATURN trial was a trial run out of UCLA. Dr. Rettig, Dr. Nickols mentioned the counterpart of this trial and Dr. Calais, Dr. Reiter. This was a prospective trial of 28 patients who had one to five lesions, M1a or M1b based on PSMA PET in the context of having a recurrence after prior radical prostatectomy. Second-line recurrences were allowed as well. All patients got SBRT and then six months of abiraterone, apalutamide, and leuprolide, so this was androgen annihilation. The median number of lesions was one, and the median SUVmax of these lesions was seven. Most of these had M1b disease, but a slight minority had M1a.

The primary endpoint was the percentage of patients who had undetectable PSA with ultra-sensitive detection six months after testosterone recovery. The median PFS overall was 19.3 months. The median eugonadal PFS was 11.4 months. 50% of men maintained an undetectable PSA six months after T recovery with a median time to T recovery of about 3.8 months.

As an overall summary, we have multiple randomized trials that show MDT improves progression-free survival in various settings: the recurrent setting, a mix of de novo and recurrent with intermittent hormone therapy, and castrate-resistant disease. Long-term cures, unfortunately, are rare, but they're seen in a minority of patients, and improved tools are needed to identify patients who might benefit from treatment intensification.

A cheap plug for a trial that we just completed run by myself and Dr. Calais at UCLA where we randomized men coming for MDT based on PSMA. They had one to five lesions in the oligo recurrent setting to receive two cycles of lutetium-PNT2002 prior to SBRT versus SBRT alone. That just finished accruing a couple of months ago, and we'll look forward to results within the next year. Thank you very much, and sorry for the difficulties with the pointer.