Comparing the efficacy of selective HER2 kinase inhibitors in HER2-positive metastatic breast cancer, it was shown that tucatinib may have been approved in the second-line setting due to it's unique impact on the central nervous system. Another HER2 inhibitor, trastuzumab deruxtecan has demonstrated promise in a phase 2 study and is now being confirmed in a phase 3 study.
Comparing the efficacy of selective HER2 kinase inhibitors in HER2-positive metastatic breast cancer, it was shown that tucatinib may have been approved in the second-line setting due to it's unique impact on the central nervous system. Another HER2 inhibitor, trastuzumab deruxtecan have demonstrated promise in a phase 2 study and is now being confirmed in a phase 3 study.
The comparison of these agents as HER2-positive metastatic breast cancer was the topic of discussion during a Targeted Oncology Case Based Peer Perspective event led by Sara A. Hurvitz, MD, associate professor, Department of Medicine, Hematology/Oncology, director, Hematology/Oncology, Breast Cancer Program, codirector, Santa Monica-UCLA, Outpatient Hematology/Oncology Practice, and medical director, Jonsson Comprehensive Cancer Center in Santa Monica, CA.
Targeted Oncology™: What do the National Comprehensive Cancer Network (NCCN) guidelines suggest for this setting?
HURVITZ: The guidelines indicate the preferred regimen is THP plus either docetaxel or paclitaxel in the frontline setting, and that’s clearly spelled out. The other regimens in the guidelines are less organized.1 T-DM1 is the approved second-line treatment for patients in this setting. However, we now have the approval of tucatinib [Tukysa]/trastuzumab/capecitabine, which we’re allowed to use in the second-line setting, even though the phase 3 trial didn’t go head-to-head against T-DM1.
Owing to its activity in the brain, I think that’s why this [was approved]. Everyone is excited about fam-trastuzumab deruxtecan-nxki [Enhertu] and the phenomenal efficacy data we see with this. And then we have the underdog, neratinib [Nerlynx] and capecitabine, which also was approved by the FDA a few months ago.2 I’m wondering if this combination is being used in metastatic disease, given its toxicity profile and the fact that we now have a HER2-selective tyrosine kinase inhibitor [TKI], tucatinib, in this setting.
How do the selective HER2 kinase inhibitors compare with each other in terms of their toxicity profiles?
Comparing the kinome map of tucatinib with neratinib, for example, we see that neratinib affects many more kinases than tucatinib. Neratinib is a more promiscuous TKI. All those enzymes are being inhibited by neratinib, and these off-target effects are responsible for the toxicity profile.
On the other hand, tucatinib is quite selective for HER2 and has much less of an impact on the enzymes that we don’t want to hit, including EGFR.
When comparing the kinase assays and the IC50 [inhibitory concentration] per nanomolar for each of the HER2-targeted TKIs, we see that neratinib, a pan-HER inhibitor, has an IC50 against HER2 of 5.6 nM. The problem is that its IC50 against EGFR is quite low, 1.8, the lowest of all the TKIs. That is why diarrhea is such a big challenge when we [treat] patients with this drug.
Lapatinib [Tykerb] has less of an impact on HER2 and a lower impact on EGFR, which means it’s a little less effective, we would expect, and less toxic. Findings from the NALA study [NCT01808573] indicate that clinically, neratinib is better than lapatinib but certainly more toxic.3 And then tucatinib, this elegant HER2 selective inhibitor, hits HER2 quite nicely, but the EGFR IC50 is quite high, 449, which is why we don’t see much diarrhea.
What was unique about the HER2CLIMB study?
This study started out as a phase 2 randomized clinical trial [NCT02614794].4 It was placebo controlled all along and started as a smaller signal finding study and then grew to be this registrational strategy pivotal trial with a lot more patients.
It was uniquely designed because it allowed patients who had brain metastases, either previously untreated or treated and stable, or treated and progressing, to enroll. About half of the patients enrolled in this study had some level of brain metastases.
Reviewing the baseline demographics, about a third of patients had de novo metastatic breast cancer when they originally received diagnoses. The median prior number of lines [of therapy] in the metastatic setting was 3 in this clinical trial; so, heavily pretreated.
Patients had to have received trastuzumab/pertuzumab and T-DM1, and 46% to 48% of patients had a history of brain metastases, either treated and stable, untreated, or progressing.
What were the primary and secondary end points for the trial?
The median PFS [progression-free survival] was just over 2 months better for the treatment arm versus the control arm [7.8 months vs 5.6 months; HR, 0.54; P < .00001]. The hazard ratio was 0.54, and this was highly statistically significant in favor of tucatinib. The median overall survival [OS] also was statistically significantly improved at the first readout, and so it was truly a home run to see that benefit.
For secondary end points, for patients with a history of brain metastases, investigators observed a statistically significant improvement in PFS for patients [7.6 months in the treatment arm vs 5.4 months in the control arm; HR, 0.48; P < .00001]. So, regardless of whether patients had brain metastases or did not, this randomized study demonstrated benefit with tucatinib, and the overall response rate was 41%.
Please summarize the important grade 3/4 adverse effects (AEs) in HER2CLIMB.
For grade 3/4 AEs that are important to clinicians, they observed diarrhea in 13% of patients who received tucatinib versus 9% of patients who received placebo [placebo plus trastuzumab plus capecitabine]. Keep in mind, patients didn’t take loperamide up front in this study, and they received standard-dose capecitabine in this trial, which is in contrast to the NALA study where patients getting neratinib needed up-front loperamide and had to be on a lower dose of capecitabine.
Grade 3/4 liver function tests (ie, aspartate transaminase and alanine transaminase) were a bit higher in the patients [treated with tucatinib], and hand-foot syndrome is a bit higher. But for the most part, tucatinib isn’t blowing out the toxicity profile when it’s added to trastuzumab and capecitabine. And AEs leading to discontinuation of tucatinib were a little higher than discontinuation with placebo, 6% versus 3%.
Now let’s look at the intracranial response rate. So these are patients who had a history of brain metastases. At the time they had measurable brain metastases by RANO [response assessment in neuro-oncology] criteria. The confirmed intracranial objective response rate was 47% with tucatinib, which is more than double that seen with placebo, and the duration of response is also more than double by adding tucatinib.
For patients who have had brain metastases or they were untreated or they were treated previously with SRS [stereotactic radiosurgery], and they’re now progressing, OS was improved with the use of tucatinib.
What was the reason for tucatinib's approval?
The FDA approved tucatinib plus capecitabine and trastuzumab in the second-line setting, but HER2CLIMB-02 was a study in third line and beyond. So why did the FDA approve it there? Probably due to the unique activity that was observed in central nervous system [CNS] metastases.
The ongoing clinical trial...HER2CLIMB-02 compared T-DM1 in combination with either tucatinib or placebo [FIGURE]. Similar to HER2CLIMB-01, they are allowing patients on this study who have active, untreated, or progressing, or stable CNS metastases.
I think we’ll have an answer as to whether adding tucatinib to T-DM1 might benefit patients. We might be seeing this doublet regimen in the future. It should also be noted in HER2CLIMB-02, those patients who had isolated CNS metastases that progressed on study were allowed to receive SRS, local regional therapy to the brain, and continue on clinical trial. So, in spite of that, they were able to continue to benefit more from the tucatinib-based therapy—patients in both arms in this blinded study.
How active will tucatinib be in patients who have received prior capecitabine?
That is an outstanding question. Those patients were specifically excluded from the HER2CLIMB-02 clinical trial. They did allow patients who had received lapatinib as long as more than a year had gone by since their last lapatinib treatment, and it was the minority of patients.
They excluded anyone who had received neratinib. So for this reason, if I have a patient who’s received capecitabine, or neratinib, or recently lapatinib, I’m less compelled to use tucatinib in those types of patients.
But let’s say your patient has progressed, and you want to use tucatinib. Right now, I think the data support primarily using it with capecitabine and trastuzumab. But, if this is a patient who hasn’t seen T-DM1, I would consider enrolling them in the HER2CLIMB-02 trial.
There’s also phase 1B evidence of the combination of T-DM1 with tucatinib, so you have safety data showing they can be combined safely. You could consider doing that outside of the clinical trial. Single-agent activity for tucatinib has not been described in a big clinical trial setting, so we don’t know how active it would be. I would want to give it with chemotherapy and trastuzumab. I think that’s where we’re seeing the most efficacy.
Please discuss the NALA study.
The NALA study evaluated neratinib plus capecitabine versus lapatinib plus capecitabine. Something to point out is capecitabine was given at 1500 mg/m2 daily, so split dose in the neratinib arm, whereas, it was 2000 mg/m2 in the lapatinib arm, and patients received loperamide in the neratinib arm. So they were trying to mitigate the AE of diarrhea in the neratinib arm. It was a heavily pretreated patient population; a third had had prior trastuzumab plus pertuzumab and T-DM1. The majority of patients were hormone receptor positive, 80% visceral metastases. It was a patient population in the later-line setting, more heavy disease burden.
The coprimary end points show that the PFS for patients was in favor of neratinib plus capecitabine. Because there was no difference at the median OS, investigators followed the survival curves out and did this restricted means analysis and calculated the mean PFS, and the mean PFS was just over 2 months better with neratinib.
The time to intervention for CNS metastases was better for patients [treated with neratinib]; 23% of patients had to have CNS intervention for metastases in the brain in the neratinib arm versus 29% with lapatinib arm.
The objective response rate was 33%, which is similar to that seen with lapatinib. If you’re doing cross trial comparisons, these data don’t look as good as what we saw with tucatinib/trastuzumab/capecitabine. We’re not supposed to do cross-trial comparisons, but the objective response rate in HER2CLIMB-02 was 41%.
I think this is why none of us are really using [neratinib] in the metastatic setting, given that we have tucatinib to use. Diarrhea at grade 3/4 was seen in 24% of patients, which was close to double that seen with lapatinib, and this is in spite of a lower dose of capecitabine and the use of loperamide with cycle 1, day 1. Also, the treatment discontinuation due to AEs was similar with the 2 arms, and other AEs were fairly similar, although there was a little more nausea and vomiting with neratinib.
To characterize the incidence and duration of diarrhea, it occurs quickly with neratinib. You have to really forewarn your patients and have them set up.
Are there other ways to mitigate the AE of diarrhea that’s observed with neratinib?
For clinicians using neratinib in the extended adjuvant setting or who are using it in the metastatic setting, the CONTROL study [NCT02400476] can inform other ways to mitigate the AE of diarrhea.5
A poster presented most recently at ASCO [American Society of Clinical Oncology annual meeting] last year looked at using loperamide alone or loperamide with a gastrointestinal-acting steroid called budesonide [TABLE on page 101]. The investigators noted that the lowest rates of grade 3 diarrhea occurred in the escalation approach or with the use of loperamide and colestipol. That’s something you might refer to in case you’re going to be using these agents in your practice.
What’s unique about trastuzumab deruxtecan?
It’s a novel ADC [antibody-drug conjugate] that is a trastuzumabtype HER2-targeted antibody linked to a topoisomerase I inhibitor, which is a potent payload. It’s an exatecan derivative. This is a drug that’s characterized by a high drug-to-antibody ratio of 7 to 8. If you contrast that with T-DM1, the amount of maytansine per trastuzumab molecule in T-DM1 is only 3 to 4. So this is potent. It’s delivering a big payload, and the payload is membrane permeable, so it can leave the HER2+ cancer cell after it’s been released and kill neighboring tumor cells. This is known as the bystander effect, which is why it’s shown some early evidence of activity even in HER2-low expressing disease.
What are the details some other studies in this setting?
The DESTINY-Breast01 trial [NCT03248492] was a single-arm study.6 The data presented relating to this study apply to the patients who were treated with the 5.4-mg/kg dose who had T-DM1 resistant or refractory disease.
There were 184 patients who received 5.4 mg/kg either in the dose-finding stage or the continuation stage.
Roughly half the patients were hormone receptor positive, but the presence of visceral disease was 92%, [which is noteworthy]. Thirteen percent of patients had brain metastases in this clinical trial. The median number of prior lines of therapy in the metastatic setting was 6, with a range of up to 27. A patient enrolled in this study had previously been treated with 27 prior lines of therapy in the metastatic setting. Everybody received trastuzumab and T-DM1; two-thirds received prior pertuzumab, and then a smattering of other therapies.
What were the findings of this trial for patients with metastatic breast cancer?
The objective response rate was 61%, and I’ve heard uniform enthusiasm about the result, with 6% complete response. The disease control rate was over 97%, and the duration of response was over a year at 14.8 months. The time to response was fast, 1.6 months, which is essentially the first time the scan is seen.
The forest plot indicates that responses were seen across different subgroups. Keep in mind, it’s a single-arm study with fewer than 200 patients enrolled, so I want to underscore the importance of us following out the randomized trials.
The PFS was [more than] 16 months in this trial, which is phenomenal when you’re considering how heavily pretreated patients were. The OS hadn’t yet been reached, and this was with a median follow-up of only 11 months. Interestingly, we usually see reports that PFS and OS for patients who have a history of brain metastases is worse. But, in this case the patients with brain metastases had a median PFS of 18 months. This is at least as good as that seen in the overall population.
How was the agent’s safety profile?
The AE profile was well tolerated, although nearly 80% of patients will have nausea, and [that means] using antiemesis agents up front, primary prophylaxis on the day of treatment, and sending patients home with prescriptions because nausea can be bothersome to patients. Patients do feel tired. Their is hair thinning. I haven’t seen full hair loss yet, but definitely patients notice hair thinning. Neutropenia does occur. I’m not needing to use GCSF [granulocyte colony-stimulating factor] up front, but it is something that mandates watching closely and sometimes holding drug and using GCSF.
Because this is the era of coronavirus disease 2019 (COVID-19), how would you characterize the interstitial lung disease (ILD) or pneumonitis?
In the era of COVID-19, important AEs to keep an eye on are ILD or pneumonitis. It was observed that 13.6% of patients had some level of pneumonitis. Four patients died on this trial from pneumonitis. Early detection is key. If you see a patient develop ground glass opacities or interstitial infiltrates on the CT following their disease response, and they are saturating 100%, not coughing, and fully functional, you should still hold therapy. Grade 1 ILD mandates a treatment hold. We have to be aggressive at picking this up early. For those [patients who] are more symptomatic, you need to get the pulmonary team involved, starting immune suppression, doing bronchoscopy, and ruling out COVID-19.
What’s next for trastuzumab deruxtecan?
Investigators are looking at trastuzumab deruxtecan head-to-head against T-DM1 in the ongoing DESTINY-Breast03 clinical trial [NCT03529110] in which patients who had received trastuzumab/taxane previously are randomized to those 2 arms equally. Hopefully, we’ll be seeing the results from that soon.
We may have a new standard approach in the second-line setting if trastuzumab deruxtecan beats T-DM1. And, based on the kinds of response rate and PFS we’re seeing, I wouldn’t be surprised if it beats T-DM1. I still think people would use T-DM1 after that because they do have distinct cytotoxic payloads, so it would make sense that there wouldn’t be cross-resistance to those agents.
Then we have the DESTINY-Breast02 trial [NCT03523585], which is aiming to secure trastuzumab deruxtecan’s place in the third-line setting because the DESTINY-Breast01 was a single-arm trial, not a comparative trial. This is trastuzumab deruxtecan versus investigator’s choice, either trastuzumab or lapatinib plus capecitabine. It’s a 2:1 randomization. I believe the enrollment for this is still ongoing, so, hopefully, we’ll have some information here as well. Then given the activity of trastuzumab deruxtecan in HER2-low breast cancer, 1+ and 2+, there was the J2101 phase IB trial (NCT004265560) that reported an objective response rate in the range of 44% for HER2, 1+, 2+ nonamplified breast cancer, and that included both hormone receptor—positive and triple-negative disease. They now have a confirmatory randomized trial called DESTINY-Breast04 [NCT03734029] comparing trastuzumab deruxtecan versus physician's choice of chemotherapy as a single agent. And, if the patients had hormone receptor–positive disease, they have to have exhausted all hormonally targeted therapies, and they allow prior line of chemotherapy as well.
Tucatinib is also being evaluated in combination with T-DM1. So it may be that we have 2 positive studies, T-DM1 plus tucatinib in 1 setting and then trastuzumab deruxtecan being the new second-line setting. There are also planned studies for both of these agents in the early stage setting. So I think in the next 5 to 6 years, we’re going to see this field continue to rapidly evolve.
What is your general approach to brain metastases management in second-line HER2 positive breast cancer and in the third-line setting?
I am excited to see the results from HER2CLIMB-02 and see whether adding the tucatinib to T-DM1 will delay the progression or the recurrence of CNS metastases in that second-line setting. Most clinicians are eager to see that data before the use of T-DM1 with tucatinib, even in patients with brain metastases. So it seems [as though] the right trial is being done.
Please comment on other novel agents being investigated in late-stage clinical trials for this setting.
One of them involves margetuximab. This is a HER2-targeted monoclonal antibody against HER2, similar to trastuzumab where the Fc portion is engineered to bind more tightly to the Fc receptors on our immune effector cells. The thought being that that will induce antibody-dependent cell-mediated cytotoxicity, [and allow it] or the immune response to be generated against tumors.
The SOPHIA phase 3 trial [NCT02492711] showed an improvement in median PFS on the order of 5 weeks or so.7 But that improvement was more substantial in patients who have a genotype that predicts for a less tight binding of their Fc receptors to Fc portions of antibodies, so those patients who are F carriers. That is 80% of 85% of the population.
The toxicity profile of this drug is quite good, with a slightly increased risk of infusion-related reactions, but it otherwise looks similar to trastuzumab. You still have to give it with single-agent [chemotherapy]. OS was not significantly improved, but there’s a strong trend in those patients who are F carriers. So it will be interesting to see if the FDA approves this, and if they do, if they selectively approve it only for patients who have a genotype of the F carriers for the Fcγ3A.
Eighty to 85% of patients are F carriers based on the SOPHIA study, so either FF genotype or VF genotype. It’s a minority of patients, only 15% to 20% of patients who are VV genotype whose immune system is already set up to tightly bind to the Fc receptor. They don’t appear to benefit from margetuximab. It was a study that was run in the third-line setting and beyond. So, if you’re expecting the immune system to mount a good response, this drug may have been more effective in the earlier-line settings when a patient’s immune system is less beat up.
The phase 3 PHOEBE trial [NCT03080805] was looking at pyrotinib, another pan-HER potent inhibitor, and was conducted primarily in Asia.8 Unfortunately, the phase 3 trial looked at pyrotinib/capecitabine versus capecitabine alone. Nobody in the United States would not get a HER2-targeted therapy. Everybody gets HER2-targeted therapy, so it wasn’t a fair comparison. I don’t think that we’re going to see a big uptake of this agent. There was a fair amount of diarrhea. It’s somewhat similar to neratinib, so I’m not sure that there’s a whole lot of data to support the use of this drug. We’ll have to see a different design trial to address that for our patients.
And then there is the PRECIOUS study [UMIN000018202],9 which is looking at pertuzumab retreatment in patients who’ve already received pertuzumab/trastuzumab and chemotherapy, so [it] addresses whether we can get any added responses and benefits for patients whose disease has already been exposed to pertuzumab.
1. NCCN. Clinical practice guidelines in oncology. Breast cancer, version 6.2020. September 8, 2020. Accessed October 16, 2020. https://bit.ly/3dAbtll
2. FDA approves neratinib for metastatic HER2-positive breast cancer. FDA. February 25, 2020. Accessed October 16, 2020. https://bit.ly/2H5qJe9
3. Saura C, Oliveira M, Feng YH, et al. Neratinib + capecitabine vs lapatinib + capecitabine in patients with HER2+ metastatic breast cancer previously treated with ≥ 2 HER2-directed regimens: findings from the multinational, randomized, phase III NALA trial. Presented at: 2019 ASCO Annual Meeting; May 31-June 4, 2019; Chicago, IL. Accessed October 16, 2020. https://bit.ly/379Ppgd
4. Murthy RK, Loi S, Okines A, et al. Tucatinib, trastuzumab, and capecitabine for HER2-positive metastatic breast Cancer. N Engl J Med. 2020;382(7):597-609. doi:10.1056/NEJMoa1914609
5. Barcenas CH, Hurvitz SA, Di Palma JA, et al. Effect of prophylaxis on neratinib- associated diarrhea and tolerability in patients with HER2+ early-stage breast cancer: phase II CONTROL trial. J Clin Oncol. 2019;37(suppl 15):548. doi:10.1200/JCO.2019.37.15_suppl.548
6. Modi S, Saura C, Yamashita T, et al. Trastuzumab deruxtecan in previously treated HER2-positive breast cancer. N Engl J Med. 2020;382(7):610-621. doi:10.1056/NEJMoa1914510
7. Rugo HS, Im SA, Shaw Wright GL, et al. SOPHIA primary analysis: a phase 3 (P3) study of margetuximab (M) + chemotherapy (C) versus trastuzumab (T) + C in patients (pts) with HER2+ metastatic (met) breast cancer (MBC) after prior anti-HER2 therapies (Tx). J Clin Oncol. 2019;37(suppl 15):1000. doi:10.1200/JCO.2019.37.15_suppl.1000
8. Xu B, Yan M, Ma F, et al. Pyrotinib or lapatinib plus capecitabine for HER2+ metastatic breast cancer (PHOEBE): a randomized phase III trial. J Clin Oncol. 2020;38(suppl 15):1003. doi:10.1200/JCO.2020.38.15_suppl.1003
9. Yamamoto Y, Iwata H, Ueno T, et al. A randomized, open-label, phase III trial of pertuzumab retreatment in HER2-positive locally advanced/metastatic breast cancer patients previously treated with pertuzumab, trastuzumab and chemotherapy: the Japan Breast Cancer Research Group-M05 PRECIOUS study. Jpn J Clin Oncol. 2018;48(9):855-859. doi:10.1093/jjco/hyy097