Amy Peterson
Analyst · Andy Hsieh with William Blair. Your line is open
Okay. Now I understand your question much better. Thank you. So with regard to the PFS delta, one thing, take a step back, and we do look at many parameters of clinical benefit, not just an improvement in a single point in time. In general, the optimal way to view the treatment effect is really based on the hazard ratio, which looks at the difference between experimental and control arm across the entire PFS analysis, not at one point estimate, i.e., the median here that you're referring to. There are multiple examples of no benefit at the median, but a positive hazard ratio resulting in approval going all the way back to ipilimumab in melanoma and really most recently, belzutifan in RCC, where the curves separate after the median, yet, clinical benefit was identified and the drugs have been approved. Going into what we have observed in our study, we have very robust PFS findings seen across all subgroups, similar in the ITT population and according to the PCWG Working Group criteria. We believe that these are clinically meaningful PFS benefits, especially in this patient population in CONTACT, which represents overall a worse prognosis than other studies conducted in this space. We have a lot of - this - 100% of patients had measurable disease. 40% had visceral disease. In the ITT, our hazard ratio of 0.65 translates to a 50% improvement over the control arm, which is exactly what we saw at the median 4.2 versus 6.3 months. In the liver MET subgroup, the hazard ratio of 0.43 translates to a more than doubling the improvement over control arm. And in fact, at the median, we observed a trebling of the effect from 2.1 to 6.2 months. In the patients with prior docetaxel, the hazard ratio was 0.57, which translates to a 75% improvement over the control arm. And in that median, we saw a doubling of median PFS from 4.1 to 8.8 months. So this is part of the totality of data, but it is not all of the totality of data. And in CONTACT-02, we have to also consider the patients that were enrolled and how they are unique and distinct from other studies that have evaluated drugs against second NHT. We have to consider the unmet need that is existent in this space and other important outcomes that we looked at. For example, I mentioned time to chemotherapy. We delayed time to chemotherapy with Cabo-Atezo. We actually -- Cabo-Atezo resulted in an increase in time to symptomatic skeletal events. These are attainable events for patients, so there are other outcomes that we look at. And what we also look at is the adverse event profile. And the adverse event profile that we showed with Cabo-Atezo versus second NHT. These are patients who enrolled in the study having tolerated their first NHT quite well. Median time on prior NHT was 12 months, right? So these are patients who know the toxicities of NHT and they tolerate them. And so when you juxtapose the toxicities of Cabo-Atezo against that, it looks different. I'll grant you that. But when you look at that compared to the toxicity profile of chemotherapy, it's very differentiated. We do not have cytopenias. We do not have febrile neutropenia. We do not have alopecia, and we do not have peripheral neuropathy. So it's a differentiated toxicity profile from otherwise other available therapies to this patient. And furthermore, when you look at this toxicity profile and compare it to cabo IO in other diseases where this doublet is used like kidney cancer by the same oncologists that treat prostate cancers, the toxicity profile is nearly identical. So going back to what I said at the beginning, when we talk about clinical benefit, we really need to consider a variety of parameters that we believe here we've met and support a robust and clinically meaningful benefit in a very unique patient population. Now I'll go on to the zanza Arcus question, okay? I know I went long there, but - okay. So you asked about zanza have given the data with cabo bel and not being real clear that belzutifan might be bringing in anything to the table. Let me just take a step back and remind everybody what we're talking about with zanza is what we believe to be a best-in-class VEGFR TKI drug. It has a similar target profile to cabo, but the tolerability profile is differentiated. Monty Pal was true at R&D Day, we presented at IKCS, patients that actually responded to zanza to progress on cabo. So it's different. We think we have an opportunity for a best-in-class. Arcus is evaluating monotherapy AB-521. We'll see whether or not that emerges as a best-in-class agent. But there's a reason to bring two potentially active agents together in this disease, and we're really excited about STELLAR-009. We are - it's co-funded by Arcus and us, and in that study, we will test the combination in patients with RCC to figure out whether and how we might move this combination forward.