Matthias Preusser, MD, PHD, reviews TUXEDO-3 findings demonstrating promising CNS activity with HER3-DXd in patients with brain metastases and leptomeningeal disease. He highlights encouraging survival outcomes, neurological stability during treatment, and the need for randomized trials to guide future treatment selection and sequencing.

Transcript:

Can you please talk about the rationale for TUXEDO-3, and what makes HER3 a promising target for treating central nervous system metastases?

Dr. Preusser:  Yeah, so in TUXEDO-3, we investigated an ADC, an antibody-drug conjugate called patritumab deruxtecan, or HER3-DXd, which targets HER3. We’ve seen in prior work in the lab that HER3 is overexpressed specifically in brain metastases of lung cancer and breast cancer. That, together with the observation that ADCs in principle can be active even in the CNS, provided the rationale to perform TUXEDO-3.

So in this trial, we investigated HER3-DXd in patients with breast cancer and active brain metastases. That was one cohort. Patients with non-small cell lung cancer with active brain metastases were the second cohort, and patients with LMD, so leptomeningeal disease of any solid tumor, were in cohort three.

What are the most clinically meaningful findings from the final analysis, and how did activity differ across the breast cancer, NSCLC, and leptomeningeal disease cohorts?

Dr. Preusser: We were very happy to see that in all three cohorts, the primary endpoints were met. We saw with HER3-DXd that intracranial responses could be achieved both in cohort one, the breast cancer cohort, and cohort two. In cohort three, the LMD cohort, we saw that the overall survival rate at three months, which was the primary endpoint in that cohort, was also higher than the predefined threshold.
So we can show once more with TUXEDO-3 that an ADC, in this case specifically HER3-DXd, can be active in the CNS. That’s one important take-home message. The most striking and surprising finding to us was the high overall survival rate seen in the LMD cohort because it is well known that LMD is associated with a very poor prognosis and very poor survival.

We saw a 65% overall survival rate at three months, which was well beyond the predefined efficacy threshold. When we looked at median overall survival time, and that is what we’re presenting now here at ESMO, we found a 10.5-month median OS in this cohort, which is higher than what we would expect from this patient population.
Although the limitation here is that it’s a single-arm study and prone to some bias, this still seems higher than would be expected. Plus, in the clinical setting in this trial, we saw patients who did well, who improved clinically, and who benefited when we take into account our experience with this disease.

For patients with leptomeningeal disease, how should clinicians interpret the survival findings alongside changes in neurologic symptoms and quality of life?

Dr. Preusser: Well, these data show, and we can interpret this in a broader sense than just specifically for HER3-DXd, together with data from other small studies and case series, that the accumulating evidence seems to indicate that ADCs, despite the fact that these are large, complex compounds, can have activity also in the leptomeningeal compartment against LMD.

So that’s probably the takeaway that we can deduce from these data. HER3-DXd is not available in routine clinical practice outside of clinical trials at this point in time, so we cannot base concrete treatment recommendations on these data at this point. However, it opens the avenue for future research, more clinical trials, and more controlled clinical trials, because I think that’s an important next step to build on the data that we have generated here.

What did the study show about tolerability, and which adverse events are most important when treating patients who already have neurologic symptoms or functional limitations?

Dr. Preusser: So the safety profile in TUXEDO-3 with this drug was similar to that seen with HER3-DXd in other clinical trials. This was not the first clinical trial performed with this agent, so we were not surprised here. What we saw is that the neurological symptoms, neurocognitive function, and quality of life remained overall stable during the active treatment phase, which is important information, as LMD and also active CNS metastases are oftentimes associated with severe neurological symptoms.

We feel that in this palliative disease setting that we’re in here, this is very important information and should be followed up. At this point in time, we don’t really know whether it’s better than other treatment approaches, for example, radiotherapy, because we have not formally performed a comparative study.

But from clinical experience and from what we saw in these patients in the clinical trial, we were happy and also, to some extent, surprised to see this disease stabilization and even improvement in some patients.

Given the study’s size and single-arm design, what needs to happen next before these findings can inform routine treatment decisions, particularly the sequencing of systemic therapy and radiation?

Dr. Preusser: Yeah, so what we’re missing at this point in time are randomized trials. We’re working on that at a European level with other ADCs and bispecific antibodies. At this moment, we don’t know what’s better, to start with SRS, for example, in patients with active brain metastases, or with an ADC, how to sequence or combine these treatments, and for which patients.

So that’s certainly something that needs to be followed up on, ideally in randomized controlled trials. What we’ve shown with TUXEDO-3 and other trials, TUXEDO-1, TUXEDO-2, TUXEDO-4, that have been completed, but also others in the meantime, is that we can enroll these patients quickly into clinical trials.

That’s also important information because for decades, patients with brain metastases, and especially active brain metastases, have simply been excluded from clinical trials due to several concerns. One of them is that accrual could be difficult. But on the contrary, we saw in this multicenter trial and the others that I mentioned that accrual can go quite quickly because there are no established treatment options and there is a high unmet clinical need.