Phase II Study of Neratinib in Older Adults with HER2 Amplified or HER2/3 Mutated Metastatic Breast Cancer

Abstract

Objective:

The tolerability and efficacy of targeted therapy in older adults with cancer has not been adequately studied. Neratinib is a novel HER1, HER2, HER4 tyrosine kinase inhibitor that has recently been granted FDA approval for treatment of breast cancer. The major toxicity of neratinib is diarrhea, which affects up to 90% of patients. This phase II trial evaluates the safety and tolerability of neratinib in adults ≥ 60.

Methods:

Patients aged 60 or older with histologically proven metastatic breast cancer and HER2 amplification (defined by ASCO/CAP guideline) or HER2/HER3 activating mutation were enrolled to receive neratinib at 240 mg daily in 28-day cycles. The association between tolerability, defined as dose reduction and number of completed courses, and log₂ Cancer and Aging Research Group (CARG) toxicity risk score was assessed using a Student’s t-test and linear regression, respectively. Response rate, progression free survival, and overall survival were also evaluated.

Results:

25 patients were enrolled with median age of 66 (range 60–79). Seventy-six percent of patients were white, 16% Asian, and 8% African-American. Seventy-six percent were patients with hormone receptor (HR) positive metastatic breast cancer (MBC) and 24% were patients with HR negative MBC. Median number of prior lines of metastatic therapy were 3 (range 0–11). 20/25 (80%) had worst grade toxicities ≥ 2. A total of 9/25 (36%) had grade 3 toxicities including 5/20 (20%) diarrhea, 2/20 (8%) vomiting, and 2/20 (8%) abdominal pain. There were no grade 4 or 5 toxicities. A total of 9/25 (36%) had dose reduction, and 2/25 (8%) discontinued therapy due to toxicity. The association between dose reductions and CARG toxicity score reached borderline statistical significance suggesting a trend with participants with higher CARG toxicity risk scores being more likely to require a dose modification (p=0.054). 1/25 (4%) had a partial response, 11/25 (44%) had stable disease, 12/25 (48%) had progression of disease, and 1/25 (4%) was not assessed. Median progression free survival (PFS) was 2.6 months (95% CI [2.56–5.26]), and median overall survival (OS) was 17.4 months (95% CI [10.3, NA]).

Conclusions:

Neratinib was safe in this population of older adults with HER2 amplified or HER2/3 mutated metastatic breast cancer (BC). Higher CARG toxicity risk score may be associated with greater need for dose adjustments. Future studies are needed to confirm this finding.

INTRODUCTION

Human epidermal growth factor receptor 2 positive (HER2⁺) breast cancer (BC) accounts for 15–20% of total breast cancers. The introduction of HER2-targeted therapies for BC patients with HER2 amplification and overexpression has led to significant improvements in oncologic outcomes^1–3. Second generation HER2-targeted tyrosine kinase inhibitors (TKI) including neratinib and tucatinib have shown promising efficacy in recent clinical trials^4–7. Neratinib is a potent oral small molecule TKI that targets HER1, HER2, and HER4 receptors at the intracellular tyrosine kinase domain through irreversible binding at a targeted cysteine residue on the receptor^8,9. Somatic HER2 activating mutations, identified in 2% of BCs, could lead to constitutive HER2 activation in the absence of gene amplification^10,11. A phase II trial of neratinib for HER2 mutated, non-amplified metastatic BC showed a clinical benefit rate (CBR) of 36% in a heavily pretreated population¹². In the recent multicenter plasmaMATCH study, neratinib had a response rate (RR) of 25%, with 20% grade ≥3 diarrhea in patients with HER2 mutated BC using ctDNA testing¹³.

Older adults with BC are underrepresented in clinical trials^14,15. There is significant interest in studying neratinib among older adults, as it is a targeted therapy which has become the standard of care for HER2⁺ BC^6,16. However, there is a significant amount of gastrointestinal (GI) toxicity, including grades 1–4 diarrhea in 92% of patients. Grades 3–4 diarrhea occurred in over 30% of patients, and dose reductions secondary to diarrhea ranged from 20–53% across nine trials of neratinib alone or in combination with other therapies^17–25. This side effect is likely to be significant in older adults who are particularly vulnerable to diarrhea. Therefore, there is a significant gap of knowledge regarding the toxicity profile of neratinib in older adults, as well as optimal supportive care medications to minimize side effects.

Consensus statements now recommend including a geriatric assessment (GA) as part of the evaluation of an older patient with cancer^26–29. GA tools can be used to assess an older adult’s risk of significant toxicity resulting from chemotherapy, and to assist with the discussion of treatment options between the oncologist and the patient. The Cancer and Aging Research Group (CARG) developed and validated a chemotherapy toxicity score for older adults with cancer³⁰. However, the CARG toxicity risk score has not been validated in older patients receiving most targeted therapies. Our group has recently presented a phase II trial evaluating tolerability of lapatinib and trastuzumab in older patients (≥ 60) with HER2⁺ metastatic BC³¹. The current phase II trial was designed to evaluate the tolerability of neratinib in patients ≥ 60 years of age with HER2 amplified and HER2/3 mutated metastatic breast cancer. In addition, the association of neratinib tolerability and CARG toxicity risk score was studied.

MATERIALS AND METHODS

Patients:

Patients with histologically proven metastatic breast cancer; HER2 amplification (defined by American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guideline) or HER2/HER3 activating mutation; 60 and older; Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≤ 2; life expectancy of > 12 weeks; RECIST 1.1 measurable or non-measurable disease; no limitation on the number of prior HER2-targeted therapy; left ventricular ejection fraction ≥ 50%; QTc interval ≤ 450 msec (men) or ≤ 470 msec (women); ≥ 3 week wash-out from radiotherapy or systemic therapy; and adequate organ function were eligible for enrollment. Patients with treated and stable brain metastasis were also eligible. Somatic HER2 or HER3 mutations were identified by Clinical Laboratory Improvement Amendments (CLIA)-certified laboratories. HER2 mutations included but were not limited to the following: missense substitutions (G309A, G309E, S310F, S310Y, S653C, V659E, R678Q, V697L, T733I, L755S, L755P, D769H, D769Y, D769N, G776V, G776C, V777L, L841V, V842I, R849W, L869R); insertions/deletions (A775_G776insYVMA aka Y772_A755dup, G776VinsC, G776AinsVGC, G778_S779insCPG, P780_781insGSP aka G778_P780dup, L755_T759del); and/or HER3 activating mutations.

This study was conducted in accordance with the Declaration of Helsinki and the principles of Good Clinical Practice and was approved by City of Hope’s regulatory and ethics committees. All participants provided written informed consent.

Study design and treatment:

The primary objective was to estimate the rate of grade ≥ 2 toxicities attributed to neratinib. Secondary objectives were to estimate the rate of dose reduction and hospitalizations; pharmacokinetic parameters; overall response rate (ORR) defined by RECIST 1.1; progression free survival (PFS); and overall survival (OS). In addition, the role of a cancer-specific GA tool in predicting tolerability, defined as dose modifications and completed courses, was evaluated. Geriatric assessment (GA) was performed to determine CARG toxicity risk scores including measures of function, comorbidity, cognition, nutrition, and psychosocial status at baseline, cycle 4, and end of study. The activity based on response rate (RR), progression free survival (PFS), and overall survival (OS) was determined.

Patients were started with a neratinib dose of 240 mg oral daily in a 28-day cycle. Diarrhea prophylaxis with loperamide was mandatory during the first cycle of treatment. Loperamide was administered at an initial dose of 4 mg every eight hours for the first fourteen days, followed by 4 mg twice a day through the first cycle of therapy (day 28) from start of neratinib dosing. Thereafter, loperamide was administered as needed, with the goal of 1–2 bowel movements a day. A total of three dosing levels of neratinib (240 mg, 160 mg, and 120 mg) were tested. Adverse events (AEs) were assessed by NCI Common Terminology Criteria for Adverse Events (CTCAE) 4.0 weekly during the first cycle, and day 1 of subsequent cycles. Patients underwent tumor evaluation by RECIST 1.1 every twelve weeks, and echocardiograms every three months. Study treatment was continued until disease progression or unacceptable toxicity. Patients who permanently discontinued treatment entered a long-term follow-up phase until death or withdrawal of consent.

Geriatric Assessment:

Geriatric assessment surveys were administered at baseline, cycle 4 day 1 and at the end of the study. The measures in the geriatric assessment are outlined in prior publications and listed in Supplemental Table 1^32,33 and the variables used to generate the CARG toxicity risk score were listed previously³⁰. These include: tumor type and stage, pretreatment laboratory data (WBC count, hemoglobin, blood urea nitrogen, creatinine, albumin, and liver function tests), and treatment characteristics (chemotherapy regimen, dosing, line of therapy, the use of WBC or RBC growth factors, and the timing of initiation of WBC growth factors). The CARG toxicity risk score calculator was accessed at the CARG website: http://www.mycarg.org/Chemo_Toxicity_Calculator.

Pharmacokinetics:

Pharmacokinetics (PK) samples were collected at the following time points: Cycle 1 Day 15 pre-dose, 6 hours post-dose, Day 1 Cycle 3 and Day 1 Cycle 4 pre-dose. Plasma neratinib concentrations were measured using a validated liquid chromatography/tandem mass spectrometry method³⁴.

Statistical methods:

The original sample size of 40 subjects was chosen so the maximum half-width of the 95% confidence limits for the rate of grade 2 or higher toxicities was 0.16. With a sample size of 40, we would expect that toxicities with a true rate of occurrence of 0.05 in 87 of 100 trials. Rates and associated 95% exact Clopper and Pearson binomial confidence limits were estimated for 1) grade 2 or higher toxicities attributed to neratinib, 2) dose reductions and hospitalizations, 3) the objective response (CR+PR), and 4) PFS and OS estimated using the product limit method of Kaplan and Meier. The association between tolerability (dose reductions and number of completed courses) and log₂ chemotherapy toxicity risk score was assessed using t-test and linear regression respectively. The study was stopped early (N = 25) due to sponsor’s decision to halt further accrual due to lack of funding.

RESULTS

Patients:

A total of 25 patients were accrued between December 2016 and March 2019 (Table 1). The median age was 66 (60–79), including 16/25 (64%) aged 60–69, 5/25 (20%) aged 70–74, and 4/25 (16%) aged ≥75. A total of 76% of patients were Caucasian, 16% Asian, and 8% African-American. Seventy-six percent were patients with hormone receptor positive metastatic breast cancer. The median lines of therapy for metastatic disease received prior to study drug was 3 (range 0–11). Thirty-six percent of patients had lung metastases, 36% had liver metastases, and 12% had brain metastases. Seventy-six percent received trastuzumab, pertuzumab containing regimen, and/or trastuzumab-emtansine (TDM-1) therapy prior to this trial.

Table 1:

Demographic and baseline clinical characteristics of patients

Characteristic	Patients (N = 25)
Age (years)
Median (range)	66 (60–79)
60–69	16 (64%)
70–74	5 (20%)
≥75	4 (16%)
Gender
Female	24 (96%)
Male	1 (4%)
Race
Caucasian	19 (76%)
Asian	4 (16%)
African-American	2 (8%)
ECOG performance status
0	8 (32%)
1	17 (68%)
Hormone receptor status
Positive	19 (76%)
Negative	6 (24%)
Sites of metastasis1
Bone	16 (64%)
Lung	9 (36%)
Local	9 (36%)
Liver	9 (36%)
Distant	5 (20%)
Chest wall/sternum	4 (16%)
Skin	3 (12%)
Brain	3 (12%)
Peritoneum	2 (8%)
Breast	2 (8%)
Other2	5 (20%)
Median number of prior lines of metastatic therapy (range)	3 (0–11)
Prior systemic therapy for metastatic disease3
Trastuzumab	17 (68%)
Trastuzumab emtansine (TDM-1)	8 (32%)
Pertuzumab	13 (52%)
Hormone therapy	7 (28%)
Other systemic therapy	11 (44%)

¹18/25 (72%) had multiple metastases

²Other metastases include: left carotid artery, spleen, CNS, pleura, and clavicle

³All 25 patients had multiple therapies

Adverse Events:

All 25 patients were evaluable for adverse events (AEs). Worst grade per adverse event type per subject were summarized. 20/25 patients (80%, 95% CI [59%, 93%]) experience a worst grade toxicities ≥ grade 2 attributed to neratinib. They are listed in Table 2. 9/25 patients (36%, 95% CI [18%, 57%]) had grade 3 toxicities attributed to neratinib (Figure 1 and Table 2). Grade 3 GI toxicities were: diarrhea (N =5, 20%), vomiting (N = 2, 8%), abdominal pain (N = 2, 8%), and nausea (N = 1, 4%). Grade 3 hematological toxicities were: anemia, leukopenia, neutropenia, and lymphopenia (N = 1, 4% each). Other grade 3 toxicities were fatigue, anorexia, dehydration, weight loss, acute kidney injury, and hypertension (N = 1, 4% each). There were no grade 4 or 5 toxicities. Four patients were hospitalized including two for neratinib-induced gastrointestinal toxicities, one for spinal cord compression, and one for appendicitis.

Table 2:

Grade 2 and 3 adverse events by CTCAE 4.0

Adverse Events (CTCAE 4.0)	Grade 2, N (%)	Grade 3, N (%)
Blood and lymphatic system disorders:
Anemia	3 (12%)	1 (4%)
Lymphocyte count decreased	5 (20%)	1 (4%)
Neutrophil count decreased	0 (0%)	1 (4%)
Platelet count decreased	1 (4%)	0 (4%)
White blood cell count decreased	3 (12%)	1 (4%)
Gastrointestinal disorders:
Abdominal pain	2 (8%)	2 (8%)
Diarrhea	11 (44%)	5 (20%)
Nausea	3 (12%)	1 (4%)
Vomiting	0 (0%)	2 (8%)
General:
Anorexia	2 (8%)	1 (4%)
Dehydration	2 (8%)	1 (4%)
Fatigue	2 (8%)	1 (4%)
Generalized muscle weakness	1 (4%)	0 (0%)
Weight loss	1 (4%)	1 (4%)
Laboratory abnormalities:
Acute kidney injury	0 (0%)	1 (4%)
Aspartate aminotransferase increased	1 (4%)	0 (0%)
Creatinine increased	1 (4%)	0 (0%)
Hypoalbuminemia	1 (4%)	0 (0%)
Skin:
Rash maculo-papular	1 (4%)	0 (0%)
Skin laceration	1 (4%)	0 (0%)
Cardiovascular disorders:
Ejection fraction decreased	1 (4%)	0 (0%)
Hypertension	0 (0%)	1 (4%)
Syncope	0 (0%)	1 (4%)

Figure 1.

Worst grade toxicities attributed to treatment at grade 2 and above.

Dose hold and dose reduction:

Of 25 patients, the median number of cycles completed was 3 (0–12). A total of 9/25 patients (36%, 95% CI [18%, 57%]) had dose modifications in at least 1 cycle. The dose reductions were due to diarrhea (N = 6), rash (N = 1), vomiting (N = 1), and trouble breathing (N = 1).

Response and survival:

24/25 patients had restaging imaging for RECIST 1.1 measure. 1/25 (4%, 95% CI [0%, 20%]) had a partial response, 11/25 (44%, 95% CI [24%, 65%]) achieved stable disease, 12/25 (48%, 95% CI [28%, 69%]) had disease progression, and 1/25 (4%, 95% CI [0%, 20%]) was not assessed. 22/25 (88%) went off treatment due to progression. 1/25 (4%) went off treatment for toxicity but progressed before being given further treatment; therefore, there was a total of 23 events in the PFS analysis. Median PFS was 2.6 months (95% CI [2.56–5.26]). At the time of this analysis, 16/25 were reported to have died (64%, 95% CI [43%, 82%]), including two that were within 30 days of treatment. The median overall survival was 17.4 months (95% CI [10.3, NA]). At time of data cut-off, one patient was on treatment and had been receiving treatment for ten months. 24 patients discontinued therapy including 4/24 (17%) for clinical progression, 18/24 (75%) for progression by RECIST, and 2/24 (8%) for neratinib GI toxicity (grade 3 diarrhea and nausea).

CARG toxicity risk scores (N = 22):

The average CARG toxicity risk score was 0.34 (min=0.19, max=0.59). There was a trend in the difference in CARG toxicity risk scores by whether a patient had a dose reduction (mean difference in log₂ risk = no dose modification - dose modification = −0.41, 95% CI [−0.82, 0.008], p = 0.054). Patients with higher risk scores were more likely to require a dose reduction (Figure 2). The lack of significance is most likely due to the small sample size. Using linear regression, we did not find log₂ CARG toxicity risk score to be a significant predictor of courses completed (slope = −1.29, se =1.44, p = 0.39).

Figure 2.

The association of CARG toxicity risk score and neratinib dose. Boxplots showing the difference in log₂ risk score between patients with no dose reduction vs. patients with reduction.

Pharmacokinetics:

Neratinib concentrations at hour 0 measure at cycle 1 day 15 were used to define the concentration at steady state based on a 240 mg daily dose. Twenty-one participants had a concentration value at this time point; one participant’s dose had been modified so was removed. Neratinib concentration was skewed right, so we transformed it using a log base 2 transformation. Least squares regression was used to assess the relationship between steady state neratinib concentration and age (Figure 3). There was a linear relationship with a regression coefficient of −0.093 (se = 0.0398, p = 0.03). The antilog steady state values for participants aged 65 and 75 were 31.9 ng/mL and 16.7 ng/mL, respectively. The least squares regression was used to determine if CARG toxicity risk score at baseline was predictive of steady state neratinib concentration. The results showed CARG toxicity risk score was not predictive (1.40, se = 2.39, p = 0.57).

Figure 3.

The association between neratinib steady state concentrations and age. Graph presenting the least squares regression line, 95% confidence interval and data.

Mutation Analysis:

5/25 (20%) had activating HER2 or HER3 mutations: patient 1 (ERBB2 V777L and ERBB2 amplification); patient 2 (ERBB2 L755S and ERBB3 E928G); patient 3 (ERBB2 R784C); patient 4 (ERBB3 S846I); and patient 5 (ERBB2 V777L) (Supplemental Table 2). Responses in the HER2/3^mut tumors were: 1 PR (20%), 3 SD (60%), and 1 PD (20%). CBR was 80% and 4/5 (80%, 95% CI [28%, 99%]) were alive at six months. Responses in the HER2^amp group were: 8 SD (40%), 11 PD (55%), and 1 not assessed (5%). 17/20 (85%, 95% CI [62%, 97%]) were alive at six months.

DISCUSSION

With promising results from multiple clinical trials and recent FDA approval, neratinib has emerged as a standard of care therapy for patients with HER2⁺ breast cancer. The current study of single agent neratinib in older adults with breast cancer demonstrates the feasibility of neratinib therapy in patients ≥ 60. Grade ≥ 3 diarrhea was observed in 20% of patients with prophylactic use of Imodium, in comparison with 30.7% observed in the CONTROL trial. Only 2/25 (8%) discontinued therapy due to neratinib-induced GI toxicity. The efficacy of single agent neratinib in this heavily pretreated cohort of patients (median 3 lines of therapy; range 0–11) with mixed HER2 amplified or HER 2/3 mutated breast cancer appears to be modest.

Neratinib was initially FDA approved for extended adjuvant therapy in the phase III ExteNET trial which assessed the efficacy of neratinib as extended adjuvant therapy in patients with early-stage HER2⁺ breast cancer who had completed adjuvant therapy with trastuzumab. The neratinib treated patients had significantly higher rates of invasive DFS than placebo, corresponding to a relative reduction in the risk of invasive disease recurrence of 34% and 27% at 2 and 5 years, respectively. HER2-targeted therapies for metastatic BC continue to evolve over time⁴. The current standard of care treatment for patients with HER2⁺ metastatic breast cancer is first-line trastuzumab plus pertuzumab and a taxane, followed by second-line trastuzumab-emtansine³⁵. After disease progression during treatment with the second-line metastatic therapy trastuzumab-emtansine, there are multiple treatment options available. These include third-line therapy drug-antibody conjugate fam-trastuzumab deruxtecan³⁶ and the combination of neratinib plus capecitabine. In addition, the novel HER2-targeted TKI tucatinib has shown great efficacy in the phase III HER2CLIMB trial⁷. Burstein et al. reported neratinib as monotherapy in HER2-positive MBC with overall response rate of 24% in trastuzumab-refractory patients and 56% in trastuzumab-naive patients²⁰. The modest response observed in our trial is comparable. In the Burstein trial, grades 3 to 4 diarrhea occurred in 30% of patients with prior trastuzumab treatment. Although older adults with mean age of 50 (range 31 – 83) were included, there is no specific toxicity data reported for the older adults. HER2-targeted therapy has been studied in older adults with breast cancer. O’Connor et al. reported the safety and toxicity of first-generation TKI lapatinib with trastuzumab in older adults with breast cancer³¹. To our knowledge, the current trial is the only study focusing on the safety and tolerability of neratinib in older adults.

The PK of TKIs in older adults with cancer has not been well studied. Lapatinib is a HER2 and EGFR TKI with variable bioavailability due to its low solubility and first-pass metabolism by CYP3A4/5³⁷. There are no dosing recommendations provided for older adults with breast cancer because no data is available on the age-related PK of lapatinib³⁷. Available data indicate that advanced age has no relevant influence on the PK of TKIs such as lapatinib, imatinib, dasatinib, nilotinib, pazopanib and sunitinib³⁸. In this study, the steady state neratinib concentrations for a patient aged 65 was twice that of a patient aged 75 based on least squares regression. The etiology is unclear and poor adsorption in older adults could contribute to the difference.

The epidermal growth factor receptor (EGFR) family of transmembrane receptor tyrosine kinases activates signaling pathways regulating cellular proliferation and survival³⁹. The HER2 receptor is a non-ligand binding member of the EGFR family, and exerts its activity through heterodimerization with other EGFR family members. HER2 functional activation promotes oncogenesis, leading to the investigation of HER2-directed agents in cancers with HER2 alterations¹⁰. Somatic HER2 and HER3 mutations can lead to constitutive HER2 activation in the absence of gene amplification. Activating HER2 mutations occur in the HER2 receptor extracellular and transmembrane domains, and at multiple hotspots in the kinase domain with no single mutation predominating¹⁰. The rate of HER2/HER3 mutations in a variety of solid tumors has an incidence rate not exceeding 5–10% in any tumor type. HER2-activating mutations in primary breast cancer is approximately 2–3%, which translates to 4,000 to 6,000 patients with HER2-mutated MBC^40–43. Neratinib is active in HER2 mutated, non-amplified MBC. In a phase II trial by Ma et al., neratinib demonstrated a CBR of 31% (90% CI [13%–55%]) in a heavily pretreated patient population with HER2 mutated non-amplified MBC¹². The phase II basket SUMMIT trial is evaluating the efficacy of neratinib in HER2 mutated solid tumor regardless of tumor types (NCT01953926). Although neratinib can inhibit growth of HER2 and HER3 mutant tumors in preclinical models, the clinical activity was only observed in HER2 mutated, not HER3 mutated tumors. Clinical responses were observed in tumors harboring HER2 S310, L755, V777, P780_Y781insGSP, and A775_G776insYVMA mutations across different tumor types including breast, cervical, biliary, salivary, and non-small-cell lung cancers. More patients had stable disease than objective response. No objective responses were observed in bladder or colorectal cancers, or in the HER3 mutant. In the subgroup analysis of eleven patients with ER⁺ HER2 mutant MBC treated with neratinib plus fulvestrant, an overall RR of 18.2% (2/11, 95% CI [2.3 to 51.8]) and a CBR 54.5% (6/11, 95% CI [23.4 to 83.3]) was observed. SUMMIT may provide the largest body of clinical data to date on the use of a pan-HER inhibitor in solid tumors with somatic HER2/HER3 mutations, although variable responses in HER2 mutants varied by type of tumor and the specific HER2 mutation were observed. Importantly, not all HER2 mutations generate the same level of HER2 hyperactivity or oncogenic dependence⁴⁴.

In the current study, the following diarrhea prophylaxis was used: Loperamide 4 mg three times a day for the first 14 days followed by 4 mg twice a day for days 14–28 on cycle 1. Beyond cycle 1, loperamide was used as needed. In the ExteNET trial, where antidiarrheal prophylaxis was not protocol-mandated, grade 3–4 diarrhea rate was 39.9%, median duration of grade ≥ 3 diarrhea was 5 days; and neratinib dose reductions and dose holds due to diarrhea occurred in 26.4% and 33.9% of patients, respectively. In this study, we observed 20% grade ≥ 3 diarrhea. In addition to Imodium, effects of adding budesonide or colestipol to loperamide prophylaxis on neratinib-associated diarrhea in patients with HER2⁺ early-stage breast cancer was evaluated through the CONTROL trial⁴⁵. Incidence of grade ≥ 3 diarrhea was 30.7% (95% CI [23.1–39.1]) with loperamide prophylaxis (loperamide cohort), 23.4% (95% CI [13.8–35.7]) with loperamide prophylaxis plus budesonide (budesonide cohort), and 11.5% (95% CI [2.4–30.2]) with loperamide prophylaxis plus colestipol (colestipol cohort) vs. 39.9% without protocol-mandated loperamide prophylaxis in the ExteNET trial. Based on the CONTROL trial result, it is reasonable to recommend addition of colestipol to loperamide as prophylaxis of diarrhea for older adults receiving neratinib.

The American Society of Clinical Oncology (ASCO) and the Institute of Medicine have identified therapeutic phase II trials as a key research priority to increase the evidence base for older adults with cancer^14,46. While targeted therapies may represent a less toxic option for older patients, few trials have studied their tolerability and efficacy in older adults^14,47. Older patients with breast cancer receiving adjuvant chemotherapy are at increased risk of chemotherapy toxicities, and currently there are limited BC-specific tools^33,48,49,50. The Cancer and Aging Research Group (CARG) developed and validated a chemotherapy toxicity score for older adults with cancer³⁰. The CARG toxicity risk score was associated with dose delay/reduction, chemotherapy discontinuation, hospitalization, and RDI < 85% (p<0.001) in patients with breast cancer⁵¹. Extending the validation of the CARG toxicity risk score to targeted therapies and immunotherapies would be a meaningful development in the era of precision medicine with broader application of targeted therapies. In a recent study using the CARG BC score in older adults with HER2 amplified breast cancer receiving the combination of lapatinib and trastuzumab, CARG-Basa was not associated with the toxicity profiles⁵². Our current study also showed a trend of increased CARG toxicity risk score with poor tolerance of neratinib, which calls for specific toxicity-predicting tools in the era of targeted therapy, including TKIs, immunotherapy, and CART therapy^50,53.

The current study was limited by its small sample size and limited number of patients 70 and older. While on this trial, mandated management guideline using loperamide were used; however, other agents such budesonide and colestipol were not systemically addressed since the evidence of these supportive agents were not available at time of initial trial design in 2015. A shift in management guideline (dose titrating up from 160 mg to 240 mg if tolerated)⁵⁴ was not used. In addition, FDA approved the combination of neratinib plus capecitabine based on the efficacy data observed from the NALA study⁵⁵. Our single agent neratinib study may not be directly translated to combination use in older adults. A future study specifically assessing the combination of capecitabine and neratinib in older adults with cancer is needed. In addition, the oral bioavailability of HER-directed TKI, specifically lapatinib and erlotinib, is reduced in the presence of gastric acid-reducing agents. It is possible that neratinib absorption may also be decreased by reduced gastric acid secretion^56,57. Similarly, several studies suggest age-related decreases in gastric acid secretion and reduced oral chemotherapy absorption consequently^58–60. Lastly, the current study did not include quality of life assessment (QOL), which is a critical endpoint in older adults with advanced stage cancer.

CONCLUSION

Neratinib was safe in this population of older adults with HER2 amplified or HER2/3 mutated metastatic BC. Higher CARG toxicity risk score showed a trend towards a need for dose modification. Future studies with larger sample sizes are needed to confirm this finding.

List of abbreviations

AE

Adverse event

ANC

Absolute neutrophil count

ALT

Alanine transferase

AST

Aspartate transferase

BC

Breast cancer

CLIA

Clinical Laboratory Improvement Amendments

CT

computed tomography

DLTs

Dose-limiting toxicities

ECHO

echocardiogram

ECOG

Eastern Oncology Group

FISH

Fluorescence In-situ Hybridization

HER2

Human epidermal growth factor receptor amplification

IHC

Immunohistochemistry

IRB

Institutional review board

MUGA

multiple-gated acquisition scan

MTD

Maximum tolerated dose

MBC

Metastatic BC

ORR

Overall response rate

OS

Overall survival

PFS

progression free survival

PHI

Protected health information

PK

Pharmacokinetics

RR

response rate

TEQR

Toxicity equivalence range

UNL

Upper limit of normal