High Activity Pain Reduction and Low Toxicity with Lu177 PSMA617 Theranostics - Michael Hofman

Professor Michael Hofman presents details of a phase II prospective study on lutetium-PSMA-617. This study used the diagnostic companion of gallium-PSMA-11 to select patients with a high PSMA expression that were suitable for lutetium-PSMA-607 therapy. This was a prospective study. It was a 30 patient study and the results of that initial cohort were published in Lancet Oncology.  This was followed by a 20 patient expansion cohort which Professor Hofman shares updates on in this presentation.  


Michael Hofman, MBBS (Hons), FRACP, FAANMS, Professor Michael Hofman is a nuclear medicine physician and physician-scientist. Peter MacCallum Cancer Center, Victoria, Australia
Read the Full Video Transcript

Michael Hofman: Thank you. It's a pleasure to present our data from Peter Mac from the uro-oncology team on our Phase II prospective study on lutetium-PSMA-617. This study was possible through the generous supply of no carrier added mutation from ANSTO and also PSMA-617 from ABX, which is now owned by Endocyte. By way of introduction, I'd like to highlight some of the key research that led to this clinical protocol. This was the paper that got us interested in the area. Research from DKFZ, using a compound from John Babich, really the first inhuman use of a small molecule rather than an antibody targeting the PSMA receptor using Iodine-124 and Iodine-131 at the time showing quite impressive responses.

This was followed up by the first use of lutetium-177 by Richard Baum using a compound from Munich and presented at the World Theranostics World Congress in 2015 and then the compound of this study, first published in 2015 again from the DKFZ group showing a complete response on PSMA-PET after two cycles with PSMA-617. This study used the diagnostic companion of gallium-PSMA-11 to select patients with a high PSMA expression that were suitable for lutetium-PSMA-607 therapy. This was a prospective study. It was a 30 patient study and the results of that initial cohort were published a couple of months ago in Lancet Oncology. You can look up that manuscript to see our more detailed protocol, including how we dose-adjusted the administered activity from 4 to 8 gigabecquerels according to tumor burden, renal function, and body weight. This was followed by a 20 patient expansion cohort and I'm pleased to present some of the updated results today, which were also shown recently at ASCO.

Endpoints were toxicity activity and secondary endpoints of PFS, overall survival and dosimetry. Patients were suitable if they had metastatic disease, castration-resistant and progressed after docetaxel and abiraterone or enzalutamide unless they were contra-indicated or the patients refused. A high uptake on PSMA-PET was defined using an SUV criteria of more than 1.5 times the intensity of the liver. Patients had to have had adequate bloods. We also did FDG-PET in all patients to help identify whether they were PSMA expressing at all sites. So this is an example of a screen failure, extensive liver and bony metastatic disease, but uptake really not greater than the liver. Another patient with apparent high-intensity uptake within the bone, but a large anterior mediastinal mass with no PSMA expression and high FDG uptake. By virtue of a dominant site of a non-PSMA expressing disease, this patient was not treated.

This is the baseline characteristics of the 50 patient cohort. I'd like to highlight the PSA doubling time of 2.6 months. So a really poor prognostic group and a heavily pretreated. Most of the patients having had abiraterone and enzalutamide, 84% having progressed after docetaxel and half the cohort having failed cabazitaxel. The median number of cycles was 4 and the activity was 7.5 gigabecquerels. This was our trial profile, so we screened 75 patients to find the 50 patients that we treated. So 20% roughly were excluded on the basis of low PSMA expression at some sites of disease. We intended to give 4 cycles of lutetium, but that did not occur in all patients. In 15% this was because of an exceptional response. We suspended therapy if there was no or minimal uptake on post-therapy imaging. 20% of patients progressed during the treatment coming off the trial.

The primary endpoint of efficacy, I'd like to highlight two things; we saw dry mouth xerostomia in around 68%, but this was exclusively grade 1 and no grade 3 or 4 toxicity. Grade 1 xerostomia is relatively mild, not interfering with the patient's quality of life. We saw a grade 3 or 4 thrombocytopenia in 10% of patients. This was attributable to lutetium-PSMA and these patients had baseline abnormal blood function linked to either extensive marrow disease or effects of prior chemotherapy. This is a waterfall plot of the best PSA response in all the cohort. We saw a PSA reduction of more than 50% in 62% of patients. What's remarkable about this chart is you'll notice only two patients did not have any drop in PSA. I think this is attributable to our patient selection, including the use of FDG. A number of patients had rather deep responses. This graph, or rather image, shows the pre- and post-Gallium MIP images in eight patients with all disease with an SUV above 3 highlighted in red. You can see there were a number of patients with complete biochemical responses, undetectable PSA associated with complete responses on PSMA-PET.

This patient here having lymph node only disease and a very durable response with 1.5 years of follow-up and no evidence of recurrence. But the vast majority of patients in this cohort did progress. The median time to PSA progression was 7 months and this was prolonged in patients who achieved a deeper PSA response. Overall survival, we had an immature followup of our 20 patient cohort expansion. So this is the data from the first 30 patients: median survival of 13.4 months. But you can see a marked difference in those who had a PSA response over 50% surviving for roughly 18 months. Imaging responses, these are three months after the last cycle of lutetium-PSMA therapy. I'd just like to highlight the middle column, which is the CT at RECIST response, a complete response in 29% and a partial response in 53%, so that's an objective response of 82%, which we saw in an earlier talk.

This somewhat overestimates the real response because some of these patients were progressing with a bone marrow disease, which was not visible by CT. This was evident on PSMA-PET or PSA progression. We looked at the quality of life. Their pain was assessed using the Brief Pain Index, mean severity score. In patients with pain at baseline by cycle two, 37% had a significant improvement and you can see that that was maintained through cycles three, four and up to the three months follow-up period. Other quality of life parameters were assessed using the EORTC-QC parameters and the global health status was maintained throughout therapy. There was a 10 point improvement in 37% by cycle two. This really correlated with the patients having a major drop in their PSA. We also saw improvements in role functioning, emotional functioning, and interestingly, cognitive functioning.

By way of discussion, we did do a quantitative SPECT/CT at three time points up to 96 hours in all our patients. These will be presented in more detail at EINM in a few months. We've used the novel method where we combined all these images together and we're able to show a dosimetry map. What we've done here is highlighted all the disease in red, like a metabolic tumor volume, and then we've looked at the main dose to all of the tumor rather than looking at sort of the three hottest lesions. And interestingly we see that the whole-body tumor dose correlates with the PSA response at 12 weeks. On the right, you can see the main dose to tumor around 14 Gray in patients that achieved a PSA response over 50%, to a much lower number in those that didn't. And interestingly, if we look at a threshold of approximately 10 Gray, if you got below that dose, there was only one responder and 10 non-responders, so there seems to be a clear association between dosimetry and response.

In conclusion, in men with metastatic castrate-resistant prostate cancer who progressed after standard therapies with PSMA have a disease, we see high response rates, limited toxicity and improvements in pain and wellbeing. This data has led to an Australian multicenter study called the TheraP trial, which is a 200 patient study running at 10 centers and randomizing lutetium-617 to cabazitaxel. This trial has already randomized about 25 patients and is progressing well in our center. We're also going to start two studies, one combining with anti-PD1 therapy and PARP inhibitor and obviously we have the Endocyte VISION, a phase III multicenter trial, which will hopefully lead to registration and widespread access to this therapy. So I'd like to thank all our funding partners that have made this possible and the team at Peter Mac.

Speaker 2: Beautiful work. I've got one question for you: are you looking at genetic markers to also predict response in those exceptional responders particularly?

Michael Hofman: Yeah, a lot of these patients have had that performed, and that analysis is ongoing. It wasn't part of the formal clinical trials protocol, but in a large cancer center, a lot of those patients have had that testing done.

Speaker 3: Thank you, Michael, for your presentation. You showed that you have about 20% of the patients excluded because of that FDG-PSMA imaging phenomenon. Do you believe these patients wouldn't have benefited from the therapy? Because from an ethical point of view and from our experience, we think that the majority of those patients could have benefited according the quality of life. This is a benefit which they would have, and they were excluded. So what do you think about that?

Michael Hofman: We don't know, because we didn't treat those patients. It's certainly plausible that they may have had an improvement in pain and other clinical parameters. We have actually followed up those patients that we didn't treat. And they have a very, very poor survival, actually in the order of only a few months. So it seems somewhat unlikely that they would have derived major benefit, but we don't know. And perhaps if you look at our waterfall plot, we've been a little too stringent. We don't have too many patients having primary progression of PSA. But as a prospective trial, we decided on a criteria and we followed it strictly.

Speaker 3: Okay. Thank you.

Speaker 2: Thank you very much.

Michael Hofman: Thank you.