Michael Hofman: Thank you. It’s a pleasure to be here and present our evolving experience with PSMA theranostics. My disclosure and thank you is to Endocyte, now AAA Novartis, for supporting this research.
We’ve heard a lot about PSMA. I have to pinch myself every now and again and make sure we’re not at a nuclear medicine meeting because just about every talk has mentioned PSMA PET in some form. We’re using small molecules radiolabeled to either gallium or fluorine for the imaging side of things. In this talk we’re going to change track and talk about the therapy. So for the therapy we’re using a very similar molecule with some slight modification optimized for therapy and labeling it to a beta emitter such as Lutetium-177. This is a beta emitter with a 1 mm path link and a seven day half life. With that 1 mm path length we effectively target tumors nearby that don’t express PSMA. By doing that we limit the effects of tumor heterogeneity, but the short path length also means there’s not much damage to normal tissues.
My first interest was sparked by this research done in Heidelberg with MIP-1095, a small molecule originally labeled to radioactive iodine, as showing exceptional responses, and then this research from Richard Baum in Bad Berka using Lutetium-177 INT, a different molecule, which is still used quite widely in the world today. The molecule which has really been commercially developed is this molecule PSMA-617, in initial case report from the Heidelberg group in Germany showing a complete response on PSMA PET to two cycles of therapy back in 2015.
We’ve been doing theranostics for over 15 years at Peter Mac with Lutetium-177 for treating gastroenteropancreatic neuroendocrine tumors, a program led by Rod Hicks, and it was quite easy for us to transition across to PSMA because of the logistics and knowledge is really quite similar. What we did differently is much of the nuclear medicine world was doing this as a compassionate access retrospective type research and we elected to do all our PSMA research prospectively. That led to this publication in Lancet Oncology last year, which was the outcomes of a 30 patient cohort. We subsequently expanded the study to 50 patients and I’m pleased to share with you some of the results of the expanded cohort today.
This was a group of patients, 80% had failed or progressed after docetaxel and abi-enzalutamide. Almost 50% had also had cabazitaxel, so a group of really quite unwell patients. They had up to four cycles of lutetium-PSMA, and we now have 31 months follow up on these patients. This is the updated waterfall plot of PSA decline presented at ASCO GU earlier this year, and we see an over 50% decline in 64% of patients. What you’ll note is that there’s only two patients who did not get any drop in PSA, a marker of our patient selection in this group, and also that there’s a number of patients with very deep responses, almost 100% declines in PSA, which is quite remarkable for this group of patients. This is an image of the eight best responders in the series presented at our Society of Nuclear Medicine meeting last year. The best thing about this image is it keeps getting better and better as the screens get bigger and bigger at these meetings, so thanks to Aurelius and Silke for organizing such a great venue for this conference.
This is a really well-tolerated treatment. We see grade 1 or 2 dry mouth in around three quarters of patients, grade 3 or 4 thrombocytopenia in 10%, and we do see some drop in renal function in around 10%. This is some data that has not been presented before. After the trial was completed patients who subsequently progressed but who benefited were able to access some compassionate access expanded access lutetium off trial, and 30% of those 50 patients went on to receive additional cycles of lutetium. What we saw is that when they continued to express PSMA, the drop over 50% was 73%, so the patients continued to respond to the therapy. If you look at that spider plot of the PSAs you will note that the durability of the responses, however, decreases over time.
Now, here’s an example of a patient with an exceptional response. A 69-year-old who progressed after docetaxel, abi, enza, and cabazitaxel, really in trouble with severe back pain from this rapidly progressive retroperitoneal node. After the first dose of lutetium PSMA we did some fancy dosimetry and we could accurately determine that this lymph node determined 175 gray just from the first cycle. As expected, when we get those doses, we see an exceptional response and PSA actually declined to zero in this patient after three doses of therapy and the PSMA PET normalized.
I’ve put the CT scans up there for you because I know there are skeptics in the audience that still don’t believe PSMA PET. So, if you look at the CT scan that lymph node has now become 5 mm in size, so a complete response on anatomic imaging. This patient did recur, as did everyone in our series, but at the time of progression this patient had four additional doses of lutetium PSMA and he continues to respond to each cycle of therapy well tolerated without side effects, and he’s alive and well with almost 40 months follow up.
As to how do our results compare to others? Our PSA decline over 50% being 64%, well, there’s no other large prospective data to date. There’s a 145-patient multicenter German collective study, and a 100-patient consecutive series from Munich. You can see that the PSA response in the Munich group is only 32%. So why do we have this difference? I think it’s about our patient selection. We have been very careful. So, is this patient suitable for therapy? The patient has extensive bony metastases with very high uptake on the PSMA PET. Many groups around the world doing lutetium therapy would say this patient is highly suitable and rush off to treat them. We have been performing FDG-PET in all these patients. What we see is that the patient has a liver full of metastasis which are FDG positive and PSMA negative. To make this easier we can color code the disease and in red is the PSMA disease that we can’t target. So we’re actually able to image this tumor heterogeneity, and these patients were excluded from our trial which means we enriched for patients likely to respond.
Not many groups around the world are doing the FDG-PET, so we’ve heard a lot of PSMA, but I would like to encourage you in the audience to consider in this castrate-resistant population to use a bit more FDG because we’ve identified these phenotypes. The two groups on the left, either low PSMA throughout or sites of FDG-discordant disease do very badly. Although we didn’t treat them on trial we followed them up and they had a median survival of only 2.5 months and so very poor. When you’re doing your questionnaire tomorrow to the panel members there might be a question on the role of FDG for lutetium selection, and you might not be doing lutetium therapy but read this paper tonight and don’t abstain because you’re not doing lutetium, just vote yes for FDG, and then we’ll incorporate it in trials and get some data.
In the patients who were treated the overall survival was at 13.3 months and patients who had a PSA decline over 50% there lived longer, but that’s difficult without a comparative group. Are we really prolonging survival? We’ve moved on and we’ve been performing for a year and a half now, the TheraP trial in Melbourne. This was started back in early 2016, worked up by Ian Davis and the ANZUP group. This was in a time when no one had really heard about lutetium theranostics, so it was a more high risk. There wasn’t even much industry involvement back then. In this trial we’re taking men who are post docetaxel who were suitable for cabazitaxel. They have an FDG and PSMA PET/CT centrally read, and then they’re randomized to either six cycles of lutetium-617 or cabazitaxel chemotherapy and no crossover. It’s been running at 11 sites around Australia, at 200 patients, and it’s been running really well. This is the recruitment curve and there’s one more patient to randomize, so I’m going to check my email at the end of the talk to see if we’ve achieved this, and we hope to present the outcome data hopefully in the middle of next year.
There’s also the Endocyte VISION trial, an industry-led registration level trial, a much bigger study, 750 patients, I think expanded to over 800 patients now. They’re running in multiple countries, randomizing lutetium-617 to best standard of care. This will give us very important data for registration.
We’ve also been focused on developing some phase 1, 2 trials in our center to answer some other high impact or critical indications. This is the first one, the PRINCE trial. This is a trial combining lutetium-617 at 2-pembrolizumab in an advanced group post docetaxel, enza-abi to see if we can improve outcomes even further. This is led by our medical oncologist, Shahneen Sandhu. There’s a very similar trial running by Tom Hope at UCSF combining these two agents, and the philosophy here is that we can enhance the effect of immunotherapy, which on its own is not very effective, by introducing radiation.
We’re also running the LuPARP trial, a phase one dose escalation trial of PSMA-617 combined with olaparib. This is not just in patients with DNA repair mutations but in all comers. We’re using the olaparib as a dose sensitizer and a shout out to Howard Soule from the PCF for supporting this through a challenge award. Thanks to Movember who’s injected about $6 million worth of funding in the last three months into Australia. We’re running two more trials which are in design and hopefully will commence in quarter four this year. This is the first one. The #UpFront PSMA trial. We’ll take patients with high volume de novo metastatic disease and randomize them to ADT plus docetaxel or upfront lutetium PSMA followed by the standard of care docetaxel, and we’ll run this at the therapy network of sites around Australia.
Declan‘s always worried about becoming obsolescence in his surgical practice, so he’s taken on this trial (#LuTectomy) looking at looking at lutectomy rather than prostatectomy in patients with high-risk localized prostate cancer with high PSMA expression will have one to two cycles of lutetium therapy and then go on to have prostatectomy and pelvic nodal dissection. We’ve set the primary endpoint as a dosimetry endpoint. What doses of radiation can we get into the prostate? We’ll also look at safety, PSA and pathology responses.
I hope I’ve taken you on a little bit of a journey about lutetium PSMA, which has been used really end of line, and our vision to move it to an earlier line of therapy and also in combinations with other therapies to see if we can further improve outcomes. There’s a lot of people involved in our multidisciplinary team. I’d like to thank them all and also our funding partners that have made this research possible. Thank you.