Read the Full Video Transcript

Oliver Sartor: Hi. I'm Dr. Oliver Sartor here with UroToday, and a special guest, Neeraj Agarwal is going to be presenting some interesting data from CONTACT-02, updated at ESMO 2024 in Barcelona. Neeraj is Professor of Oncology at the Huntsman Cancer Center, Director of GU cancers there. Welcome, Neeraj.

Neeraj Agarwal: Thank you very much for having me.

Oliver Sartor: Good. I'm looking forward to hearing more about CONTACT-02.

Neeraj Agarwal: Thank you, Oliver, for having me. I'll be discussing the data from the CONTACT-02 trial, which was cabozantinib plus atezolizumab versus second ARPI in patients with metastatic castrate-resistant prostate cancer. This is the final overall survival results which I presented at the 2024 ESMO meeting. As a background, we know that patients who have progressed on ARPI and have mCRPC, they have limited treatment options, and ARPI followed by ARPI is often practiced in the community, especially when the option is available. We also know that cabozantinib is a multi-tyrosine kinase inhibitor, has been approved in kidney cancer, and has promising preliminary data from a Phase 1 trial in patients with mCRPC. Atezolizumab is a well-established anti-PD-L1, and we did evaluate cabozantinib plus atezolizumab in the Phase 1b COSMIC trial, which we presented a few years ago. This was the Phase 3 CONTACT-02 study, where patients who had to have measurable extrapelvic soft tissue metastasis, and had to have disease progression, and one ARPI were randomized to cabozantinib plus atezolizumab versus a second ARPI.

This is the schema in front of us right now. Again, I'd like to bring your attention to the fact that these patients had to have measurable extrapelvic soft tissue metastasis. PFS and OS were dual primary endpoints, and randomization was stratified by presence or absence of liver metastasis, prior docetaxel therapy, and in the setting, ARPI or prior ARPI was given. If you look at the patient disposition, out of 989 patients, about 400+ patients were ineligible. They failed screening, and the number one reason for screen failure was absence of extrapelvic measurable soft tissue disease during the screening scans. Ultimately, 575 patients were randomized. At the time of the presentation, or the data cutoff for this presentation was in April 2024, and the median follow-up for overall survival analysis was 24 months. The baseline characteristics were evenly distributed, and the fact that one of the inclusion criteria was extrapelvic measurable soft tissue metastasis, there were 23% of patients in each arm who had liver metastasis. We didn't plan for it, but it turned out by the time we finished accrual, we ended up accruing a relatively large number of patients with liver metastasis.

Beyond that, the patient population was fairly representative of what we expect from an mCRPC patient population. The PFS data have already been presented by us in the GU ASCO 2024 meeting. At that time, we showed that PFS was significantly improved with the hazard ratio of 0.65 with cabo+atezo versus second NHT, and the PFS benefit was more pronounced in patients with liver metastasis and bone metastasis. Also, bone metastasis patient population also had PFS benefit. Again, this all was measured... The primary endpoint was measured by RECIST 1.1, and the bone scans were evaluated by the PCWG3. Now, let's look at the overall survival data, which was the center point or point of discussion during the ESMO meeting. Despite early separation of the curve, the median PFS was similar in both arms, about 15 months. The hazard ratio was 0.89. It favored cabo+atezo but did not meet statistical significance.

If you look at the subgroup analysis, this forest plot shows that patients who had liver metastasis and who had bone metastasis seemed to have survival advantage. I'm pointing out these groups because these are of clinical interest. We often see patients with bone metastasis, predominant patient population, in the mCRPC setting, and of course, liver metastasis patients do not have many options. If you look at the final overall survival in these clinical subgroups of interest, patients with liver metastasis definitely seem to have advantage. Obviously, the trial was not powered for these subgroups, so definitely hypothesis-generating. Hazard ratio was 0.68 favoring cabo+atezo in liver metastasis patients. If you look at bone metastasis patients, hazard ratio was 0.79 favoring cabo+atezo again. Other clinically relevant endpoints, such as time to chemotherapy, time to symptomatic skeletal events seemed to favor cabo+atezo, and time to deterioration of quality of life was similar, suggesting that cabo+atezo combination did not seem to adversely affect the quality of life more than the second ARPI.

If you look at the adverse events, we see adverse event profile, which is very consistent with what we see. Cabozantinib is widely used in the community for RCC patients. We see that diarrhea, hypertension, and fatigue, these are the most common grade three-four side effects, again, in single digits. Despite frequent dose modifications with cabozantinib, we saw that median dose intensity of cabozantinib was quite high at 83%, and only 5% of patients discontinued both cabozantinib and atezolizumab in the cabo+atezo arm. If you look at subsequent therapies, 73% of patients received docetaxel. 38% of patients received cabazitaxel in the second NHT arm. In the cabo+atezo arm, slightly lower number of patients receiving chemotherapy, but still, majority of patients received chemotherapy, suggesting that cabo+atezo treatment did not preclude patients from receiving chemotherapy or other subsequent therapies. To conclude, we showed that CONTACT-02 met one of the primary endpoints of PFS with the hazard ratio of 0.65.

The PFS benefit was more pronounced in patients with liver metastasis. The second primary endpoint, not the secondary, favored cabo+atezo but did not meet statistical significance although we did see strong survival advantage in patients with liver metastasis with a hazard ratio of 0.68, and it's possible that we saw the data from the other trial where cabozantinib showed significant improvement in patients with neuroendocrine differentiation. We have a feeling that because liver metastasis have that AR indifferent phenotype, that could be one of the reasons why cabo+atezo combination seemed to be benefiting more in this context. We didn't see any new safety data, any new safety signal. Other secondary or clinically meaningful endpoints, such as time to chemotherapy or time to skeletal event, seemed to favor cabo+atezo. We concluded by saying that we have a combination with a novel mechanism of action, which may be useful for patients with mCRPC who have liver metastasis or who have bone metastasis. Of course, this can be one of the reasonable options if available in the clinic.

Oliver Sartor: Great. Thank you, Neeraj. Gosh, there are a lot of questions and comments I have that I need your perspective on. I think there's still a question in the mind of many about why the combination? The cabozantinib did not succeed in the Phase 3. We don't really have much in the way of the PD-1, PD-L1 inhibition. Why put the two drugs together? Subsequently, I'd like you to address is the activity in liver you think cabo, or the combination, or how would we know?

Neeraj Agarwal: Great question, Oliver. We will show some more data on the combination versus single-agent cabo in the near future. Obviously, the combination was tested together in the COSMIC trial, in the Phase 1 trial. We had a negative Phase 3 trial in the post-chemotherapy setting, as you know, the COMET trial, in about a decade ago. We thought it would be difficult to have a trial again with a single-agent cabozantinib monotherapy and randomize patients to that trial knowing that single-agent monotherapy or single-agent data were not favorable in the COMET trial. I think those were the reasons why we decided to go with cabozantinib plus atezolizumab.

Oliver Sartor: Do we need to rethink the liver issue? I'm looking at the data that you showed and I think I have it right. Overall survival, 12.2 versus 7.1 months, and a lot of these patients had chemotherapy following the treatment. That hazard ratio for the liver was 0.68. That's overall survival. I thought that was quite interesting.

Neeraj Agarwal: Absolutely. One question came up about did we plan to have that many liver metastasis patients in the trial. There was no plan. I think this population self-selected itself to be one of the common patient population because we required extrapelvic measurable disease. Coming back to your question, Oliver, this is definitely intriguing. If you just look at the number of liver metastasis patients, like 125 patients in the overall trial, and if we just ignore the rest of the data for a second, just for the sake of discussion, obviously not a preplanned analysis, this 125 patients seems to be sufficient enough, as far as number is concerned, to show positive results in overall survival. Again, as you said, a lot of these patients got chemotherapy after they had disease progression on the ARPI. Definitely, we are intrigued by these data, and I'm really hoping that we have this option available someday for our patients in the clinic.

Oliver Sartor: What do you think a control group might look like if you were to propose a liver-metastasis-only trial? I'm a little bit curious, because I think we all struggle with the treatment of liver metastatic disease. But I'm just curious, what would you believe would be standard of care? Would it be docetaxel? Would it be a docetaxel with platinum? Or what's standard of care in your routine practice for those with liver metastasis?

Neeraj Agarwal: Docetaxel chemotherapy, definitely chemotherapy. It's the standard of care. For most of the patients with liver metastasis, sometimes we see sub-centimeter lesions in the liver. We have not done the biopsy and many of these patients get other therapies, radioligand therapies, even alternate ARPI. They are very reluctant to receive chemotherapy in the past, and PSA slowly rising, which was the case with many of these patients. But if you ask me, there's a biopsy-proven liver metastasis patient in my clinic, definitely, and with good performance status obviously, cabazitaxel with carboplatin or docetaxel with carboplatin is often used in my clinic.

I also use docetaxel chemotherapy because we have sometimes difficulty in obtaining pre-authorization for cabazitaxel with carboplatin, if we have not used docetaxel, and we have many patients we have used docetaxel first followed by cabazitaxel with carboplatin. Because of lack of level one evidence for cabazitaxel with carboplatin, we have been using docetaxel more often first as a first therapy. If we have a trial, say just for the sake of discussion, if we have designed a trial with patients with mCRPC who have prior disease progression on ARPI, I think the control arm will be docetaxel chemotherapy.

Oliver Sartor: Good. Yeah, thank you very much. One more comment and then we'll have a wrap-up. The overall survival here was only 15 months on median. That really speaks to the fact that these patients have a very poor prognosis. In some of the other trials, it turns out that they've been better. But it was interesting, a lot of the patients appear to be kind of garden variety. I mean, they were lymph node positive, bone scan positive. I mean, these are not totally atypical patients outside of the liver mets. I'm just curious in your perspective, why do you think the 15 months, is it really brought down by the higher percentage of liver or is there more to the story?

Neeraj Agarwal: It's possible that unexpectedly higher number of patients with liver metastasis, what quarter of these patients, may have brought down the overall survival to slightly lower than what we expected. We did look at the data or overall survival data, actually published those data recently on overall survival after failure of one ARPI. It is about less than two years, for sure, regardless of whether ARPI is used in the metastatic hormone-sensitive setting or castration-resistant setting. It was around 18 to 20 months, so definitely, it is less than what we expected, and I think it's driven by disproportionately higher number of patients with liver metastasis for sure.

Oliver Sartor: Thank you. Before we wrap up, are there any additional or final comments you'd like to add for our listeners' benefit?

Neeraj Agarwal: Definitely, there is a patient population with huge unmet need in mCRPC. I think overall patient population with mCRPC, there's a huge unmet need, but patients with liver metastasis, they have a higher unmet need. There are very few treatment options for them. By default, many of the trials have not included them, so many options which are available in the clinic such as Radium-223, they are not applicable to this patient population. I really hope that we have more options available for these patients with novel mechanisms of action so that we have more choices for our patients.

Oliver Sartor: Thank you so much, Neeraj Agarwal, Professor of Oncology in Utah. Thank you so much for being here today, Neeraj.

Neeraj Agarwal: Thank you for having me, Oliver.