Following a series of meetings with the FDA, Puma Biotechnology announced that it plans to delay the submission of a new drug application (NDA) until mid-2016 for neratinib.
The NDA will be based on data from the phase III ExteNET trial and a phase II study exploring diarrhea prophylaxis with loperamide. The delay followed a request from the FDA for a new statistical analysis of data from the ExteNET trial that involved the censoring of data for patients who missed 2 or more scheduled disease assessments prior to recurrence or death.
With the new statistical model, the 2-year invasive disease-free survival (iDFS) rates were 94.2% and 91.9% for neratinib and placebo, respectively. This represented a 34% reduction in the risk of disease recurrence or death (HR, 0.66; CI, 0.49-0.90;P= .004). At in the original primary analysis, the 2-year iDFS rate was 93.9% in the neratinib arm versus 91.6% with placebo, representing a 33% reduction in the risk of recurrence or death (HR, 0.67; 95% CI, 0.50-0.91; P= .009).1
“The primary analysis results of the trial do not appear to be materially changed due to this change in event and censoring rule,” Alan H. Auerbach, chief executive officer, president, and chairman of the board of Puma, said during a webcast. “In order to accommodate this change that was requested by the FDA, Puma will need to rewrite the clinical sections that are currently in its NDA filing. In order to accomplish this, Puma intends to delays its proposed timeline for filing its NDA until mid-2016.”
In the phase III study, 2840 patients who remained disease-free following 1 year of treatment with adjuvant trastuzumab and chemotherapy were randomized to neratinib (n = 1420) or placebo (n = 1420). The interval between receiving trastuzumab and entering the trial was approximately 4.5 months. Neratinib was administered for 12 months at 240 mg per day.
The median age of patients in the study was 52 years and approximately 23.8% had node negative disease, with 46.6% of patients having 1 to 3 positive nodes and 29.6% had ≥4 positive nodes. Anthracyclines were administered as adjuvant chemotherapy in the majority of patients (77%). Appropriate endocrine therapy was administered to 94% of patients with hormone receptor (HR)-positive breast cancer.
According to the primary assessment and original statistical model, in patients with HR-positive breast cancer (n = 1631) the 2-year iDFS rate with neratinib was 95.4% compared with 91.2% with placebo (HR, 0.51;P= .001). In the HR-negative group (n = 1209) the 2-year invasive DFS rate was 92% with neratinib and 92.2% with placebo (HR, 0.93; P= .735).
In an exploratory 3-year analysis,2the iDFS rate was 92% with neratinib versus 89.9% for placebo (HR, 0.74; 95% CI, 0.56-0.96; P= .023). In those specifically with centrally confirmed HER2-positive/HR-positive breast cancer, the 3-year DFS rate was 94.4% versus 88% for neratinib and placebo, respectively (HR, 0.43; 95% CI, 0.26-0.70; P<.001).
With longer follow-up the 2-year iDFS rate was 94.1% with neratinib versus 91.6% with placebo (HR, 0.68; 95% CI, 0.50-0.90;P= .004). When applying the FDA mandated censoring rule to the longer follow-up data, the iDFS rates were 94.4% and 91.8% in the neratinib and placebo arms, respectively (HR, 0.67; 95% CI, 0.49-0.90; P= .004).
“The FDA did not raise any objections to Puma filing with the 2-year invasive disease-free survival data,” said Auerbach. “The FDA further stated that Puma did not need to submit the results of the 5-year disease-free survival data or the overall survival data in its NDA filing.”
Overall, 95.4% of patients treated with neratinib experienced all-grade diarrhea (39.9% was grade 3/4). The trial design did not mandate antidiarrhea prophylaxis. Other gastrointestinal-related adverse events (AEs) included nausea (43%), fatigue (27%), vomiting (26.2%), and abdominal pain (24.1%). In the placebo arm, 35.4% of patients had all-grade diarrhea, with a grade 3/4 incidence of just 1.6%.
A phase II study was launched to quantify the benefits of prophylactic loperamide for reducing neratinib-related diarrhea. Data from this study, which will be included in the NDA, were presented during a webcast by Puma, and showed that grade 3 neratinib-related diarrhea was reduced to 16% with loperamide.
“The FDA requested that Puma include the results of this phase II trial that incorporated the loperamide prophylaxis with the NDA filing and stated that the potential to include this data in the label for neratinib would be a review issue and part of the labeling negotiation,” noted Auerbach.
In the phase II open-label study, patients were enrolled within 1 year of completing adjuvant treatment with trastuzumab. Neratinib was administered at 240 mg daily for 12 months. In the original protocol, 27 patients received loperamide at 16 mg on day 1 followed by 12 mg per day on days 2 and 3 and 6 to 8 mg on days 4 to 56. After an amendment to the protocol to simplify treatment, 23 patients received loperamide at 12 mg per day for the first two weeks followed by 8 mg per day until day 56.
In the original protocol group, 18.5% of patients experienced grade 3 diarrhea (95% CI, 6.3-38.1). Of those who experienced a grade 3 event (n = 5), 60% were non-compliant to the loperamide regimen. In the amended group, the grade 3 diarrhea rate was 13% (95% CI, 2.8-33.6), and 67% of patients were non-compliant.
The rates of all-grade diarrhea were 74.1% and 43.5% in the original and amended protocol groups, respectively. Grade 4 diarrhea did not occur and hospitalizations were not required. Most treatment-emergent diarrhea occurred within the first 4 weeks.