The irreversible pan-HER tyrosine kinase inhibitor neratinib showed single-agent activity across cohorts of patients with <em>HER2</em>-mutant advanced cancers, according to findings from the phase II SUMMIT study presented at the 2017 AACR Annual Meeting.
David Hyman, MD
David Hyman, MD
The irreversible pan-HER tyrosine kinase inhibitor neratinib showed single-agent activity across cohorts of patients withHER2-mutant advanced cancers, according to findings from the phase II SUMMIT study presented at the 2017 AACR Annual Meeting.1
The greatest efficacy was observed in those with breast cancer, with further early activity demonstrated in a group of patients with cervical cancer and biliary cancer. No responses were seen for neratinib monotherapy in patients with bladder cancer, colorectal cancer (CRC), and for those withHER3mutations. A low level of enduring activity was seen in those with lung cancer, and the study continues to assess patients across other cancer types.
"A number of the breast cancer patients responded, and some quite durably," noted lead investigator David Hyman, MD, medical oncologist, Memorial Sloan Kettering Cancer Center. "The activity of neratinib was influenced by both mutation and tumor type."
One hundred and forty-one patients were enrolled across cohorts in the phase II "basket" trial. All patients hadHER2orHER3mutations and received neratinib at 240 mg daily with high-dose loperamide diarrhea prophylaxis during the first cycle of therapy. Patients with HR-positive breast cancer also received hormonal therapy and those with bladder cancer received paclitaxel. The primary endpoint was objective response rate at week 8 (ORR8). ORR at 6 months was also assessed.
Patients in the study had lung cancer (n = 26), breast cancer (n = 25), bladder/urinary tract cancer (n = 16), not specified solid tumors (n = 15), CRC (n = 12), biliary tract cancer (n = 9), endometrial cancer (n = 7), cervical cancer (n = 5), gastroesophageal cancer (n = 5), and ovarian cancer (n = 4). In a separate group, 17 patients hadHER3-mutated unspecified tumors.
In general,HER2mutations are mutually exclusive from HER2 amplifications, representing a unique group of patients, Hyman noted. Within the study population, the investigators identified 124 qualifyingHER2alterations, of which 92 were missense mutations, 30 were inframe insertions, 1 frameshift, and 1 structural variant. Testing was completed using both tumor and plasma-derived cell-free DNA, with 93% concordance, Hyman noted.
"This really illustrates the complexityHER2mutations seen in our patients. There were 35 uniqueHER2mutations in this cohort of patients," he said. "As anticipated, these tend to cluster in the extracellular domain, the transmembrane domain, and the kinase domain."
In the breast cancer cohort, the ORR8 was 32% and the ORR was 24%. The clinical benefit rate (CBR; ORR plus stable disease) was 40%. The median progression-free survival (PFS) was 3.5 months.
In cervical cancer, the ORR8 and the ORR were 20% and the median PFS was 20.1 months. The CBR was 60%. For those with biliary cancer, the ORR8 was 22.2% but the ORR was 0%. The median PFS was 2.8 months.
"Interestingly, when you look at the PFS data, these stable disease patients have quite stable disease," noted Hyman. "These are in patients with quite heavily pretreated disease, with 1 to 2 or more than 3 prior lines of chemotherapy."
There was no evidence of greater sensitivity to neratinib based on type of mutation, as both cancer type and mutation played a role. In some instances, missense mutations appeared more sensitive to neratinib compared with exon 20 insertions; however, this analysis was partially confounded by responses association with cancer type.
"We know thatHER2is a credentialed target in breast cancerI think this is another mechanism of action for breast cancer," said Hyman. "Does this mean thatHER2is not a driver for the other cancers or are we not drugging it effectively? It's probably both, but more the latter. There are differences in how these tumors are programmed to respond to inhibitors."
Findings presented at the 2016 San Antonio Breast Cancer Symposium (SABCS) focused on the group of patients treated with fulvestrant and neratinib.2In this portion of the study, the ORR was 41.7% in 7 evaluable patients treated with the combination, which included 2 complete responses and 3 partial responses. The median PFS was 3.7 months.
"Is neratinib monotherapy going to get us there? Probably not. But it might in combination with other agents," noted Hyman. "In other HER2 scenarios, the targeted therapy has been combined with agents like chemotherapy. Combinations may be needed to improve activity and durability."
Across the full study, the most common adverse events (AEs) of any grade were diarrhea (73.8%), nausea (43.3%), vomiting (41.1%), constipation (34.8%), and fatigue (31.9%). The most frequent grade ≥3 AEs were diarrhea (22%), anemia (7.1%), dehydration (5.7%), and abdominal pain (5%).
Diarrhea events were anticipated in the study, Hyman noted. Overall, there were no grade 4 or 5 diarrhea events and only 4 patients discontinued due to this AE. The median number of grade 3 diarrhea episodes was 1 with a median duration of 2 days. This typically occurred during the first 2 weeks of the first cycle of treatment.
"What you see, is a patient comes on and approximately 1 in 5 will have an episode of grade 3 diarrhea, typically during cycle 1 and lasting 2 days. This patient will subsequently be dose-reduced and it will be unlikely that they will experience additional high grade diarrhea events," Hyman noted. "Even though it is a high-grade AE, it's not one where they ultimately discontinue therapy or find it intolerable."
Several studies continue to assess neratinib across cancer types. A phase II study is looking at neratinib with or without fulvestrant for patients with HER2 non-amplified,HER2-mutant breast cancer (NCT01670877). Additionally, a phase I study is planned to assess neratinib with everolimus, palbociclib, or trametinib for patients withEGFRmutation/amplification,HER2mutation/amplification, orHER3/4mutation (NCT03065387).
In the lung cancer group, the ORR8 and ORR were both 3.8%. The clinical benefit rate was 42.3%, which translated into longer durations of PFS, Hyman hypothesized. In this group, the median PFS was 5.5 months. Two patients remained on therapy, with one remaining progression-free for 24.9+ months and the other for 17.8+ months.