Adjuvant Neratinib Misses Primary End Point of OS in Glioblastoma, But Significantly Prolongs PFS in EGFR+ Subset

Isabel Arrillaga-Romany, MD, PhD

A subset of patients with EGFR-positive glioblastoma experienced significantly longer progression-free survival (PFS) when treated with adjuvant neratinib (Nerlynx) following chemoradiation vs adjuvant temozolomide alone, although no overall PFS benefit or any overall survival (OS) improvement was observed in the overall patient population, according to data from the phase 2 INSIGhT trial (NCT02977780).¹

Data from the trial, which were presented during the 2021 Society for Neuro-Oncology Annual Meeting, showed that in the EGFR-positive subgroup (n = 73), the median PFS achieved with neratinib (n = 43) was 6.3 months vs 4.6 months with the control regimen (n = 30;HR, 0.58; log-rank test P = .04). However, in this group, neratinib did not significantly improve median OS over the control, at 14.4 months and 15.3 months, respectively (HR, 0.97; P = .94).

No statistically significant improvement in PFS or OS was observed with neratinib vs the control in the overall patient population (n = 149). The median PFS in this population was 6.0 months with neratinib and 4.7 months with the control regimen (HR, 0.75; P = .12). The median OS with neratinib and the control regimen were 13.8 months and 14.7 months, respectively (HR, 1.01; P = .75).

No significant improvement in PFS or OS was achieved with neratinib in the EGFR VIII–mutant subgroup (n = 28), either. The median PFS with neratinib (n = 16) was 6.2 months vs 5.1 months with the control (n = 12; HR, 0.88; P = .77). The median OS with neratinib was 16.9 months vs 12.7 months (HR, 0.44; P = .09).

“Although preliminary results did not achieve the primary end point, subgroup analyses demonstrated improved PFS in patients with EGFR activation and a nonsignificant trend toward improved OS in those with EGFR VIII mutations, which could warrant further investigation,” Isabel Arrillaga-Romany, MD, PhD, director of Neuro-Oncology Clinical Trials at Massachusetts General Cancer Center, stated in a press release.² “Additionally, we are very pleased that this trial reinforced feasibility of randomized Bayesian adaptive platform trials for newly diagnosed glioblastoma.”

Although EGFR is an intriguing target in glioblastoma, several trials that have been conducted in patients with this disease have demonstrated limited efficacy with EGFR TKIs. INSIGhT is the first biomarker-driven, prospective controlled trial to explore an EGFR small molecule inhibitor in patients with newly diagnosed glioblastoma.

Notably, the trial utilized a Bayesian adaptive approach, which means that investigators utilized an adaptive algorithm to randomize patients to treatment arms and updated randomization probabilities by biomarker grouping each month. Specifically, individual biomarker subsets and the PFS of enrolled patients were leveraged to identify randomization probabilities.

The use of this algorithm was meant to accelerate and provide a competitive advantage to investigative arms linked with promising data early in the study. Notably, new arms can be added, and existing arms can also be dropped with this approach.

To be eligible for enrollment, patients needed to have histologically confirmed intracranial glioblastoma or gliosarcoma, be at least 18 years of age, have a Karnofsky performance status (KPS) of at least 60, and normal organ and marrow function. Patients also needed to have plans to initiate radiation treatment within 14 to 42 days after surgical resection and be negative for IDH1 R132H mutation on immunohistochemistry. Standard-of-care assays also needed to detect evidence that the MGMT promoter is unmethylated.

Those who received prior therapy apart from surgery, had significant intratumoral hemorrhage, were taking EIAED, received greater than 4 mg of decadron, had uncontrolled intercurrent illness, or impairment of gastrointestinal function, were excluded. Moreover, patients with a history of a different malignancy, unless disease free for at least 2 years and determined by study investigator to be at low risk for recurrence of that malignancy, were also excluded.

After patients with newly diagnosed unmethylated glioblastoma underwent genotyping for biomarker subgrouping, they underwent randomization. In the control arm, patients underwent radiation at 6000 cGy and temozolomide at a daily dose of 75 mg/m² for 42 days followed by a 4-week break, and adjuvant temozolomide at a once daily dose ranging from 150 mg/m² to 200 mg/m² for 5 days and 6 treatment cycles.

There were 3 investigative arms to the trial. In 1 arm, patients received radiation plus temozolomide followed by adjuvant abemaciclib (Verzenio). In another arm, patients were given radiation plus CC-115 followed by adjuvant CC-115. In the last investigative arm, patients were administered radiation plus temozolomide followed by adjuvant neratinib given at a daily dose of 240 mg in 28-day cycles until disease progression or unacceptable toxicity.

The primary objective of the research was to identify whether any of the experimental approaches improved OS in patients with glioblastoma whose tumors harbored unmethylated MGMT promoters vs standard therapy with temozolomide.

Key secondary objectives included determining whether specific a priori defined biomarkers could predict benefit from investigative therapy, assessing the toxicity in investigative arms, evaluating PFS in investigative arms and biomarker subsets, assessing OS in investigative arms and biomarker subsets, and determining the link between PFS and OS effects in the investigative arms.

Data from the neratinib arm were shared during the meeting. The median age of those in the neratinib arm was 60 years (range, 24-78) vs 59 years (range, 24-75) in the control arm. In the investigative and control arms, 46% and 41% of patients, respectively, were male; 74% and 63% were Caucasian; 79% and 65% were non-Hispanic, 31% and 40% had a KPS of 90; 43% and 30% had EGFR positivity; 64% and 49% had CDK positivity; and 53% and 39% had PI3K positivity.

Safety data showed that neratinib was generally well tolerated, with reported toxicities mirroring previously reported data. Notably, no new toxicity signals were identified with the agent, and no patients experienced a grade 4 adverse effect (AE).

Nineteen grade 3 AEs associated with the study drug were reported, and they included colitis (n = 1), diarrhea (n = 6), fatigue (n = 2), sepsis (n = 1), urinary tract infection (n = 1), alanine aminotransferase increased (n = 1), platelet count decreased (n = 1), anorexia (n = 1), dehydration (n = 1), hypokalemia (n = 1), generalized muscle weakness (n = 1), hypertension (n = 1), and surgical and medical procedures (n = 1).

“[These are] the first data demonstrating an effect of neratinib in EGFR-amplified or -mutated glioblastoma,” Alan H. Auerbach, chief executive officer and president of Puma Biotechnology, added in the press release.² “While we are not looking to pursue further clinical investigations of neratinib in this indication, we are evaluating the potential to develop a backup compound, HKI-357, which has preclinically demonstrated better EGFR activity, in this indication.”

References

1. Arrillaga-Romany I, Trippa L, Fell G, et al. Preliminary results of the neratinib arm in the individualized screening trial of innovative glioblastoma therapy (INSIGhT): a phase II platform trial using Bayesian adaptive randomization. Presented at Society for Neuro-Oncology Annual Meeting; November 17-21, 2021; Boston, MA. Accessed March 2, 2022.
2. Puma Biotechnology presents data from the neratinib arm of the INSIGhT Trial at the 2021 SNO annual meeting. News release. Puma Biotechnology; November 20, 2021. Accessed March 2, 2022. https://bit.ly/3IEv9Dk