ONCAlert | Upfront Therapy for mRCC

# R/R HER2+ mBC: Interpreting Findings From the NALA Trial

Targeted Oncology

**Published Online:**12:42 PM, Fri July 19, 2019

**Adam M. Brufsky, MD, PhD:**The NALA trial is a large randomized phase 3 registration trial that compares neratinib and capecitabine to lapatinib and capecitabine. The idea is that neratinib, unlike lapatinib, is an irreversible binder to HER1 [human epidermal growth factor receptor 1], which is the EGF receptor; HER2, which is the HER2 receptor; and HER4. And because it’s irreversible, the idea is that it would bind to the HER2 and permanently inactivate it until the receptors regenerate. So it has a longer term efficacy theoretically than lapatinib. We know that lapatinib/capecitabine is the current standard of care for a lot of people in the later lines, third, fourth, or fifth line. So the idea is potentially, is this any better? Is this irreversible binder any better?

What happened in the NALA trial is that—I’m part of the NALA group—we randomized 621 women who had at least 2 anti-HER2 regimens for metastatic disease. A third of those women actually did have pertuzumab and TDM1 [trastuzumab emtansine] together. About 25% or so just had trastuzumab and pertuzumab. Another 30% or 40% just had trastuzumab-based regimens. And about I think 10% or 15% had regimens that had included TDM1 at some point.

There was a global trial, and these women, of which about 30% to 40% had de novo metastatic disease, were randomized to lapatinib, at the standard dose, with capecitabine at 2,000 mg/m

^{2}versus neratinib at 240 mg a day; with capecitabine at 1,500 mg/m

^{2}per day. And the reason it was a lower dose, that had been established in prior trials.

The study had 2 co-primary endpoints. It had the hazard ratio for progression-free survival [PFS], and the hazard ratio for overall survival. And the hazard ratio for progression-free survival was met. The hazard ratio was 0.076 with a

*P*value that was .00053, or something like that; it was a significant

*P*value. That’s number 1. In terms of overall survival, the hazard ratio was about 0.84 or 0.88, and the

*P*value was not significant, it was around .02.

In terms of the absolute numbers, these trials were very interesting in that in terms of progression-free survival the separation of the survival curves occurred late, really almost after the median. And so the trials actually had a prespecified what’s called restricted means analysis. What that meant is that we looked at the mean survival as opposed to the median because when a curve separates late after the median, the median really isn’t going to tell you what happened. So you’re going to look at the survival of the entire curve.

And we restricted it to the mean survival at 24 months because most of that’s occurred before 24 months of follow-up. When we did this restricted means analysis, it was prespecified, we didn’t go, “Oh, we didn’t see this median so we changed it.” Now it’s a very important thing to understand. This was a prespecified analysis before we even looked at the data.

And it turns out that the difference in means was about 2.2 months. So it was 8.8 months in the neratinib arm versus 6.6 in the lapatinib arm. That was statistically significant. It was around .004 or something like that. So we had that. The restricted mean survival was again about 0.8, again about a 1.8-month difference, but it was not statistically significant. Clearly it met its endpoints, it met the primary endpoint of PFS, so clearly is a positive trial.

The other secondary endpoints from the duration of response was 8.5 months in the neratinib arm versus I think about 5.5 months in the lapatinib arm, that was statistically significant. But most importantly again, getting to the initial thing that we had talked about, the incidence of treatment for symptomatic brain metastasis in this trial was 22% of the patients on the neratinib arm versus I think 29% on the lapatinib arm. And what that tells us is that it reduced the incidence of symptomatic brain metastasis, so theoretically symptomatic brain progression, by about 6%, 7%. In absolute terms that’s by about 25% in relevant terms. That that to me is extremely significant, so I think it’s a big deal.

It’s not so much a synergistic effect, it is that you’re getting the capecitabine in and you’re getting the anti-HER2 therapy in with neratinib, and it’s staying there because it’s an irreversible binder. That’s the best way to think about it. It binds to the HER2 tyrosine kinase a little bit, the tyrosine kinase domain, and doesn’t get off, whereas the lapatinib does, it just bounces off. And so, again, if the plasma levels are low, if the CSF [cerebral spinal fluid] levels are low, it’s simply going to come off and it can’t stick. And it can’t irreversibly damage the receptor. So it’s more permanent, it gets into brain and stays is the best way to think about it.

**Transcript edited for clarity.****Adam M. Brufsky, MD, PhD:**The NALA trial is a large randomized phase 3 registration trial that compares neratinib and capecitabine to lapatinib and capecitabine. The idea is that neratinib, unlike lapatinib, is an irreversible binder to HER1 [human epidermal growth factor receptor 1], which is the EGF receptor; HER2, which is the HER2 receptor; and HER4. And because it’s irreversible, the idea is that it would bind to the HER2 and permanently inactivate it until the receptors regenerate. So it has a longer term efficacy theoretically than lapatinib. We know that lapatinib/capecitabine is the current standard of care for a lot of people in the later lines, third, fourth, or fifth line. So the idea is potentially, is this any better? Is this irreversible binder any better?

What happened in the NALA trial is that—I’m part of the NALA group—we randomized 621 women who had at least 2 anti-HER2 regimens for metastatic disease. A third of those women actually did have pertuzumab and TDM1 [trastuzumab emtansine] together. About 25% or so just had trastuzumab and pertuzumab. Another 30% or 40% just had trastuzumab-based regimens. And about I think 10% or 15% had regimens that had included TDM1 at some point.

There was a global trial, and these women, of which about 30% to 40% had de novo metastatic disease, were randomized to lapatinib, at the standard dose, with capecitabine at 2,000 mg/m

^{2}versus neratinib at 240 mg a day; with capecitabine at 1,500 mg/m

^{2}per day. And the reason it was a lower dose, that had been established in prior trials.

The study had 2 co-primary endpoints. It had the hazard ratio for progression-free survival [PFS], and the hazard ratio for overall survival. And the hazard ratio for progression-free survival was met. The hazard ratio was 0.076 with a

*P*value that was .00053, or something like that; it was a significant

*P*value. That’s number 1. In terms of overall survival, the hazard ratio was about 0.84 or 0.88, and the

*P*value was not significant, it was around .02.

In terms of the absolute numbers, these trials were very interesting in that in terms of progression-free survival the separation of the survival curves occurred late, really almost after the median. And so the trials actually had a prespecified what’s called restricted means analysis. What that meant is that we looked at the mean survival as opposed to the median because when a curve separates late after the median, the median really isn’t going to tell you what happened. So you’re going to look at the survival of the entire curve.

And we restricted it to the mean survival at 24 months because most of that’s occurred before 24 months of follow-up. When we did this restricted means analysis, it was prespecified, we didn’t go, “Oh, we didn’t see this median so we changed it.” Now it’s a very important thing to understand. This was a prespecified analysis before we even looked at the data.

And it turns out that the difference in means was about 2.2 months. So it was 8.8 months in the neratinib arm versus 6.6 in the lapatinib arm. That was statistically significant. It was around .004 or something like that. So we had that. The restricted mean survival was again about 0.8, again about a 1.8-month difference, but it was not statistically significant. Clearly it met its endpoints, it met the primary endpoint of PFS, so clearly is a positive trial.

The other secondary endpoints from the duration of response was 8.5 months in the neratinib arm versus I think about 5.5 months in the lapatinib arm, that was statistically significant. But most importantly again, getting to the initial thing that we had talked about, the incidence of treatment for symptomatic brain metastasis in this trial was 22% of the patients on the neratinib arm versus I think 29% on the lapatinib arm. And what that tells us is that it reduced the incidence of symptomatic brain metastasis, so theoretically symptomatic brain progression, by about 6%, 7%. In absolute terms that’s by about 25% in relevant terms. That that to me is extremely significant, so I think it’s a big deal.

It’s not so much a synergistic effect, it is that you’re getting the capecitabine in and you’re getting the anti-HER2 therapy in with neratinib, and it’s staying there because it’s an irreversible binder. That’s the best way to think about it. It binds to the HER2 tyrosine kinase a little bit, the tyrosine kinase domain, and doesn’t get off, whereas the lapatinib does, it just bounces off. And so, again, if the plasma levels are low, if the CSF [cerebral spinal fluid] levels are low, it’s simply going to come off and it can’t stick. And it can’t irreversibly damage the receptor. So it’s more permanent, it gets into brain and stays is the best way to think about it.

**Transcript edited for clarity.**