Tarlatamab Shows Encouraging Activity in Phase 1 Trial in DLL3-Positive Neuroendocrine Prostate Cancer - Rahul Aggarwal
June 17, 2024
Rahul Aggarwal presents findings on targeting DLL3 in neuroendocrine prostate cancer. He explains that DLL3 is highly expressed in high-risk neuroendocrine prostate cancer, particularly in the treatment-emergent form seen after androgen deprivation therapy. This Phase 1b study expanded on prior research in small cell lung cancer and included 40 patients with broad inclusion criteria for DLL3 positivity. The trial aimed to evaluate safety and showed a manageable safety profile with low-grade cytokine release and neurologic events. Encouragingly, the study found that patients with strong DLL3 expression, especially those with small cell histology, responded well to the treatment, with some experiencing durable responses. Moving forward, Dr. Aggarwal emphasizes refining patient selection based on DLL3 positivity using PET imaging and biopsies, aiming to enhance treatment options for this high-risk patient group.
Biographies:
Rahul Aggarwal, MD, Medical Oncologist, Associate Professor, University of California, San Francisco, UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Biographies:
Rahul Aggarwal, MD, Medical Oncologist, Associate Professor, University of California, San Francisco, UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Read the Full Video Transcript
Alicia Morgans: Hi, I'm so excited to be here today with Dr. Rahul Aggarwal, who is joining me from UCSF, and also joining me after ASCO 2024, where he presented a rapid oral abstract looking into DLL3 as a target for prostate cancer. Thank you so much for being here with me today.
Rahul Aggarwal: It's my pleasure, Alicia.
Alicia Morgans: Wonderful. It is always so good to talk with you. And I love all of the work that you and your team at UCSF and associated sites do in drug development. Can you tell me a little bit about DLL3, why that's an important target, and then, of course, the study that you presented?
Rahul Aggarwal: Of course, yeah. This is really a target that we think is highly enriched in neuroendocrine prostate cancer, which we know is a very high-risk subset of the disease, both the de novo form that presents at the time of diagnosis, but even more commonly, the treatment-emergent neuroendocrine form that develops after prior androgen deprivation therapy that may be prevalent in up to 15, 20% of all patients. So it's not a tiny population. In that subset, DLL3 is really one of the top surface antigens that we think is a good target to go after in this setting.
Alicia Morgans: Wonderful. And certainly not only expressed in prostate cancer, from what I understand. Expressed in other solid tumors?
Rahul Aggarwal: Yeah, it's really an aberrant expression of an atypical Notch ligand that gets expressed in really high-grade neuroendocrine cancers of any site. So certainly in small cell lung cancer. Tarlatamab, the subject of our discussion today, just got FDA approval, accelerated approval in that setting. In high-grade GI neuroendocrine cancers. Really across the board.
Alicia Morgans: So tell me a little bit about your study. This was a Phase 1, right?
Rahul Aggarwal: It was a Phase 1b, really piggybacking on the Phase 1 small cell lung cancer trial. And so we really designed a study that was predominantly dose expansion, so it could be thought of as almost like a large dose expansion. It ended up being 40 patients, a single arm. We used pretty broad inclusion criteria for NEPC, just recognizing it's not always the easiest diagnosis to detect, whether it be via metastatic biopsy or genomic classification. So we used either histologic evidence of small cell, IHC evidence of neuroendocrine differentiation, or genomic selection criteria, so if you had two or more loss of tumor suppressors, RB1, P53, PTEN. Any of those criteria could be used. We didn't prospectively require DLL3 expression, but we did retrospectively analyze outcomes by expression status.
Alicia Morgans: Okay. Well, before we get into that, in terms of the outcomes, can you just sort of help me think through... Neuroendocrine prostate cancer, even with those selection criteria, can be a bit of a grab bag, right?
Rahul Aggarwal: It can be. I think that we need to still develop more standardized pathologic criteria. We see a lot of heterogeneity in terms of patients that undergo metastatic biopsies. Is this called small cell? Is it called neuroendocrine differentiation? Is it called high-grade carcinoma? So that's why we use a really more pragmatic composite definition.
Alicia Morgans: Okay, that makes sense. Well, tell me, what did you find?
Rahul Aggarwal: Yeah, so the study was really designed to look at safety as the primary endpoint. And we used at the time, it's really a one-step dose, 1 milligram on cycle one, day one, and then really up to the target dose. At the time we designed this study, 100 milligrams was really the target dose that had been found to be safe in the small cell lung cancer trial. We know subsequently that 10 milligrams might be the more optimal dose from a risk/benefit standpoint. But basically, everyone was treated with a 1 milligram step followed by a 100 milligram target dose every two weeks really until progression. And we looked at safety.
We looked at cytokine release and ICANS, was really the two side effects that we're certainly most interested in with the bispecifics. And the good news there is most CRS was low grade. There was one Grade 3 event that was transient. We had five ICANS events, but four of those were low grade. There was one patient that had high-grade neurologic tox that had to come off. So overall, a manageable safety profile. I think we subsequently learned in small cell lung cancer that a 10 milligram target dose might even be better in terms of a toxicity profile.
Alicia Morgans: Okay. Well, tell me, how were these patients dosed? Is this an outpatient regimen, an inpatient regimen? Where are they getting treatment?
Rahul Aggarwal: Yeah, this is one of the big challenges with the bispecifics. Because of that risk of cytokine release and neurologic toxicity, early in the first cycle, it does require inpatient monitoring for the step dose as well as the first two target doses as part of this trial. So it is pretty intensive inpatient monitoring for at least 48 hours for those doses. After that, it transitions to outpatient every two weeks.
Alicia Morgans: Okay. Well, that's certainly encouraging, I think, for our patient population. Did you find any early signals of efficacy? Obviously, Phase 1b, as you said, the goal is safety, but sometimes we can see some prolonged responses.
Rahul Aggarwal: Absolutely. There definitely were some encouraging responses, particularly in the DLL3-positive subset. So in the all comers broad inclusion, the response rate was 10%, but all of those four confirmed responses really were in patients that were DLL3 positive with small cell histology. So when you really drill down to those that really have true small cell transformation with DLL3 strong, positive scores, that's where we saw the most activity. And actually in that subset, four out of 11 patients had a durable objective response. The median duration response was over seven months, and we have one patient who's actually still on it for over two years with an ongoing response. I think it's encouraging, for sure.
Alicia Morgans: Wow. So where do you go from here?
Rahul Aggarwal: I think we go forward in the areas further studies planned, but we need to clearly take a better look at the selection criteria. So really focusing on the target rather not getting so much hung up on the histologic classification, knowing that that can be a challenge. So really selecting based on the DLL3 positivity is where we're going next.
Alicia Morgans: Okay. And is that a biopsy? Is there a PET scan that can target that? How are you doing that?
Rahul Aggarwal: We're going to do both, and we're going to leverage Memorial Sloan Kettering's ability to do DLL3 PET now in the clinic, as well as tissue biopsies and knowing that there's a CLIA-approved IHC assay that we can use really to help guide patient selection.
Alicia Morgans: Great. So tell me, what is your final message to listeners as they're thinking about this data? And congratulations on this study. And what do they tell patients?
Rahul Aggarwal: I think there is encouraging activity with DLL3 targeting with bispecifics, and it is a high-risk patient population with a huge unmet need. So I really want to continue to push forward and, hopefully, ultimately, get a new treatment option for patients in this setting.
Alicia Morgans: Absolutely, and especially if we're talking about something that can help patients with small cell prostate cancer, which is a clear unmet need and has been sort of something that I think you and others have been working on for many years, it would be transformative. So thank you so much for your efforts in this space and in all others and your time and expertise today.
Rahul Aggarwal: Thanks, Alicia. Appreciate it.
Alicia Morgans: Hi, I'm so excited to be here today with Dr. Rahul Aggarwal, who is joining me from UCSF, and also joining me after ASCO 2024, where he presented a rapid oral abstract looking into DLL3 as a target for prostate cancer. Thank you so much for being here with me today.
Rahul Aggarwal: It's my pleasure, Alicia.
Alicia Morgans: Wonderful. It is always so good to talk with you. And I love all of the work that you and your team at UCSF and associated sites do in drug development. Can you tell me a little bit about DLL3, why that's an important target, and then, of course, the study that you presented?
Rahul Aggarwal: Of course, yeah. This is really a target that we think is highly enriched in neuroendocrine prostate cancer, which we know is a very high-risk subset of the disease, both the de novo form that presents at the time of diagnosis, but even more commonly, the treatment-emergent neuroendocrine form that develops after prior androgen deprivation therapy that may be prevalent in up to 15, 20% of all patients. So it's not a tiny population. In that subset, DLL3 is really one of the top surface antigens that we think is a good target to go after in this setting.
Alicia Morgans: Wonderful. And certainly not only expressed in prostate cancer, from what I understand. Expressed in other solid tumors?
Rahul Aggarwal: Yeah, it's really an aberrant expression of an atypical Notch ligand that gets expressed in really high-grade neuroendocrine cancers of any site. So certainly in small cell lung cancer. Tarlatamab, the subject of our discussion today, just got FDA approval, accelerated approval in that setting. In high-grade GI neuroendocrine cancers. Really across the board.
Alicia Morgans: So tell me a little bit about your study. This was a Phase 1, right?
Rahul Aggarwal: It was a Phase 1b, really piggybacking on the Phase 1 small cell lung cancer trial. And so we really designed a study that was predominantly dose expansion, so it could be thought of as almost like a large dose expansion. It ended up being 40 patients, a single arm. We used pretty broad inclusion criteria for NEPC, just recognizing it's not always the easiest diagnosis to detect, whether it be via metastatic biopsy or genomic classification. So we used either histologic evidence of small cell, IHC evidence of neuroendocrine differentiation, or genomic selection criteria, so if you had two or more loss of tumor suppressors, RB1, P53, PTEN. Any of those criteria could be used. We didn't prospectively require DLL3 expression, but we did retrospectively analyze outcomes by expression status.
Alicia Morgans: Okay. Well, before we get into that, in terms of the outcomes, can you just sort of help me think through... Neuroendocrine prostate cancer, even with those selection criteria, can be a bit of a grab bag, right?
Rahul Aggarwal: It can be. I think that we need to still develop more standardized pathologic criteria. We see a lot of heterogeneity in terms of patients that undergo metastatic biopsies. Is this called small cell? Is it called neuroendocrine differentiation? Is it called high-grade carcinoma? So that's why we use a really more pragmatic composite definition.
Alicia Morgans: Okay, that makes sense. Well, tell me, what did you find?
Rahul Aggarwal: Yeah, so the study was really designed to look at safety as the primary endpoint. And we used at the time, it's really a one-step dose, 1 milligram on cycle one, day one, and then really up to the target dose. At the time we designed this study, 100 milligrams was really the target dose that had been found to be safe in the small cell lung cancer trial. We know subsequently that 10 milligrams might be the more optimal dose from a risk/benefit standpoint. But basically, everyone was treated with a 1 milligram step followed by a 100 milligram target dose every two weeks really until progression. And we looked at safety.
We looked at cytokine release and ICANS, was really the two side effects that we're certainly most interested in with the bispecifics. And the good news there is most CRS was low grade. There was one Grade 3 event that was transient. We had five ICANS events, but four of those were low grade. There was one patient that had high-grade neurologic tox that had to come off. So overall, a manageable safety profile. I think we subsequently learned in small cell lung cancer that a 10 milligram target dose might even be better in terms of a toxicity profile.
Alicia Morgans: Okay. Well, tell me, how were these patients dosed? Is this an outpatient regimen, an inpatient regimen? Where are they getting treatment?
Rahul Aggarwal: Yeah, this is one of the big challenges with the bispecifics. Because of that risk of cytokine release and neurologic toxicity, early in the first cycle, it does require inpatient monitoring for the step dose as well as the first two target doses as part of this trial. So it is pretty intensive inpatient monitoring for at least 48 hours for those doses. After that, it transitions to outpatient every two weeks.
Alicia Morgans: Okay. Well, that's certainly encouraging, I think, for our patient population. Did you find any early signals of efficacy? Obviously, Phase 1b, as you said, the goal is safety, but sometimes we can see some prolonged responses.
Rahul Aggarwal: Absolutely. There definitely were some encouraging responses, particularly in the DLL3-positive subset. So in the all comers broad inclusion, the response rate was 10%, but all of those four confirmed responses really were in patients that were DLL3 positive with small cell histology. So when you really drill down to those that really have true small cell transformation with DLL3 strong, positive scores, that's where we saw the most activity. And actually in that subset, four out of 11 patients had a durable objective response. The median duration response was over seven months, and we have one patient who's actually still on it for over two years with an ongoing response. I think it's encouraging, for sure.
Alicia Morgans: Wow. So where do you go from here?
Rahul Aggarwal: I think we go forward in the areas further studies planned, but we need to clearly take a better look at the selection criteria. So really focusing on the target rather not getting so much hung up on the histologic classification, knowing that that can be a challenge. So really selecting based on the DLL3 positivity is where we're going next.
Alicia Morgans: Okay. And is that a biopsy? Is there a PET scan that can target that? How are you doing that?
Rahul Aggarwal: We're going to do both, and we're going to leverage Memorial Sloan Kettering's ability to do DLL3 PET now in the clinic, as well as tissue biopsies and knowing that there's a CLIA-approved IHC assay that we can use really to help guide patient selection.
Alicia Morgans: Great. So tell me, what is your final message to listeners as they're thinking about this data? And congratulations on this study. And what do they tell patients?
Rahul Aggarwal: I think there is encouraging activity with DLL3 targeting with bispecifics, and it is a high-risk patient population with a huge unmet need. So I really want to continue to push forward and, hopefully, ultimately, get a new treatment option for patients in this setting.
Alicia Morgans: Absolutely, and especially if we're talking about something that can help patients with small cell prostate cancer, which is a clear unmet need and has been sort of something that I think you and others have been working on for many years, it would be transformative. So thank you so much for your efforts in this space and in all others and your time and expertise today.
Rahul Aggarwal: Thanks, Alicia. Appreciate it.