The addition of PEGPH20 to standard nab-paclitaxel and gemcitabine improved progression-free survival by 4.9 months compared with the two agents alone in untreated patients with advanced pancreatic cancer who expressed high-levels of hyaluronan.
Sunil R. Hingorani, MD, PhD
The addition of the enzyme PEGPH20 to standard nab-paclitaxel and gemcitabine improved progression-free survival (PFS) by 4.9 months compared with the two agents alone in untreated patients with advanced pancreatic cancer with high expression levels of hyaluronan (HA), according to interim findings from a phase II study presented at the 2015 ASCO Annual Meeting.
In the ongoing study, patients with metastatic pancreatic cancer were randomized to nab-paclitaxel plus gemcitabine with or without PEGPH20, a pegylated version of recombinant human hyaluronidase. After enrolling 146 patients the study was placed on hold to address concerns regarding thromboembolic events in the investigational arm. At this point, the protocol was amended to exclude those at high risk for a thromboembolic event. Additionally, prophylaxis with enoxaparin was required in the ongoing study.
"There was improved efficacy in HA-high patients treated with PEGPH20 versus nab-paclitaxel and gemcitabine, specifically a doubling in PFS with statistical significance and an encouraging reduction in the hazard ratio to 0.39," lead investigator Sunil R. Hingorani, MD, PhD, medical oncologist with the Seattle Cancer Care Alliance, said during his presentation. "The thromboembolic events appear to be manageable, particularly in patients who received prophylaxes with 1 mg/kg per day of enoxaparin."
In the study, 74 patients in the PEGPH20 arm and 61 in the control arm received treatment prior to the hold placed on the study (Stage 1). Additionally, following the protocol adjustment, the study enrolled an additional 114 patients (Stage 2). Efficacy data presented at ASCO were from Stage 1, while safety data included limited data from Stage 2 and all of Stage 1.
Across both stages of the trial, PEGPH20 was administered at 3 µg/kg twice weekly for cycle 1 followed by weekly administration in subsequent cycles. Nab-paclitaxel and gemcitabine were administered at their standard FDA-approved doses. The dual primary endpoints of the study were PFS and thromboembolic events. Secondary endpoints included ORR, PFS by HA level, and overall survival (OS).
For patients treated in the overall population, the ORR was 41% with PEGPH20 versus 34% in the control arm (P = .48). The median duration of response (DOR) was 7.4 versus 4.2 months, in the investigational and control arms, respectively. The PFS in the full population was 5.7 months with PEGPH20 versus 5.2 months with nab-paclitaxel/gemcitabine (HR = 0.69; P = .11).
Tumor biopsy samples for the HA analysis were available for 61 patients treated with PEGPH20 and 45 treated in the control arm from Stage 1. HA testing was completed using immunohistochemistry. In the investigational arm, 38% were indicated as HA-high compared with 47% in the control arm.
In the full population of those with HA data available, the ORR was 43% in those treated with PEGPH20 with a median DOR of 8.1 months. In the control arm, the ORR was 31% and the DOR was 4.2 months. The PFS was 5.5 versus 4.8 months, with PEGPH20 and without, respectively (HR = 0.64; P = .09).
The median PFS in the HA-high group was 9.2 months with PEGPH20 versus 4.3 months in the control arm (HR = 0.39; P = 0.05). In the HA-low arm, the median PFS was 5.3 versus 5.6 months, with PEGPH20 and without, respectively (P = 0.74).
For those with HA-high tumors, the ORR was 52% with a DOR of 8.1 months compared with 24% and a DOR of 3.7 months, for PEGPH20 and the control, respectively. There was 1 complete response in the PEGPH20 arm, with a median DOR of 7.4 months. In the low group, the ORRs were similar between the two arms (37% vs 38%).
"If we restrict the evaluation of response rate to those patients evaluated prior to the clinical hold, this difference was even larger. Namely, 73% in the PEGPH20 arm and 27% in the control arm," Hingorani observed.
There was a trend toward improvement in OS with the triplet therapy; however, this was not deemed statistically significant. At the analysis, the median OS was 12 months with PEGPH20 versus 9 months without (HR = 0.62; CI 0.26-1.46).
"The trend in overall survival seen in the triplet arm does included data from 12 of 23 patients who discontinued the enzyme and received nab-paclitaxel and gemcitabine alone after the clinical hold," Hingorani noted.
At a median follow-up of 7 months, all but 4 patients had discontinued Stage 1 of the study, primary as a result of progression (44.3% in the PEGPH20 arm versus 52.5% in the comparator). Adverse events were the cause of discontinuation for 18.6% of patients in the PEGPH20 arm versus 16.4% in the control arm.
The most significantly increased adverse events of all grades with PEGPH20 versus without were peripheral edema (58.1% vs 31.1%), muscle spasms (55.4% vs 1.6%), and neutropenia (32.4% vs 18%). The most frequently observed grade ≥3 adverse event with PEGPH20 was neutropenia (24.3%).
"As for the increase in neutropenia, this was not reflected in an increase in episodes of febrile neutropenia in the PEGPH20 arm," Hingorani said.
A safety analysis was conducted at the time of the hold in April 2014, which revealed that 28.4% of patients in the PEGPH20 arm had experienced a thromboembolic event, which were comprised of pulmonary embolisms (12.2%), deep vein thrombosis (14.9%), and arterial events (6.8%). In the nab-paclitaxel/gemcitabine alone arm, 14.8% of patients experienced a thromboembolic event.
At a longer follow-up of patients enrolled in Stage 1 conducted in December 2014, the rate of thromboembolic events had increased to 42% in the PEGPH20 arm versus 25% in the doublet arm.
Patients enrolled in Stage 2 received 40 mg/day or 40-mg/day increased to 1 mg/kg/day of enoxaparin to counter thromboembolic events. In this stage of the trial, this adverse event was seen in 28% of patients with PEGPH20 versus 29% in the control arm. Additionally, a second group of patients enrolled in Stage 2 started enoxaparin at 1 mg/kg/day. In this group, no patients treated with PEGPH20 (n = 20) experienced a thromboembolic event compared with 10% in the comparator arm.
"Combining the dose groups suggests that the thromboembolic risk can in fact be managed," Hingorani said.
The phase III study will assess PEGPH20 in combination with nab-paclitaxel and gemcitabine for patients with metastatic pancreatic cancer with high HA levels. The primary endpoint of the study is PFS, with secondary outcomes measures focused on toxicity and OS. Halozyme hopes to use PFS data from the study for a potential new drug application, according to a statement.
The study is expected to begin enrolling in early 2016.
"We remain very encouraged by the interim data presented from our ongoing phase II study, including the positive trend in overall survival," Athena Countouriotis, MD, senior vice president and chief medical officer of Halozyme Therapeutics, the developer of PEGPH20, said in a statement. “We continue to plan for the first-quarter 2016 initiation of our phase III study.”
Hingorani SR, Harris WP, Hendifar AE, et al. High response rate and PFS with PEGPH20 added to nab-paclitaxel/gemcitabine in stage IV previously untreated pancreatic cancer patients with high-HA tumors: Interim results of a randomized phase II study. J Clin Oncol. 2015;33 (suppl; abstr 4006).