Newlink Genetics Corp. said Friday its phase III IMPRESS trial of algenpantucel-L in patients with surgically resected pancreatic cancer will continue without modification on the recommendation of the independent data safety monitoring board (DSMB) following the first of two planned interim analyses.
The DSMB’s initial review of IMPRESS (Immunotherapy for Pancreatic Resectable Cancer Survival Study) was scheduled to occur following 222 patient events in the trial, which completed enrollment of 722 patients in September 2013. A second interim analysis is planned at 333 patient events and, if needed, a final analysis at 444 events.
The phase III of lead candidate algenpantucel-L, also known as HyperAcute pancreas, is being conducted under a special protocol assessment (SPA) with the FDA. The study – the largest pivotal immunotherapy trial ever conducted in the U.S. in resected pancreatic cancer, according to Newlink chairman and CEO Charles Link – began enrolling in April 2010 with the goal of assessing overall survival (OS) after treatment with a regimen of adjuvant standard of care (gemcitabine alone or with 5-FU chemoradiation) with or without algenpantucel-L.
HyperAcute product candidates consist of human, tumor-specific – but not patient-specific – cancer cell lines that are modified to express alpha-gal, a carbohydrate to which humans have preexisting immunity. The alpha-gal-modified cancer cells stimulate an immune response that trains the body’s natural defenses to seek out and destroy similar cancer cells.
By enrolling patients with pancreatic cancer who had undergone surgery but remained at high risk of disease progression, the study design is intended to demonstrate that treatment with HyperAcute pancreas immunotherapeutic increases median OS by 22 percent. Disease-free survival is a secondary endpoint. (See BioWorld Today, May 28, 2010.)
The trial is randomized 1:1 between the algenpantucel-L and control arms, with patients on treatment receiving up to 18 vaccinations of 300 million vaccine cells every two weeks over a six-month period, followed by six monthly injections.
Despite passing the first major safety hurdle, with no concerns raised by the DSMB, investors were hankering for more. The company’s shares (NASDAQ:NLNK) dropped 16.2 percent on Friday, closing at $31.60, for a loss of $6.11. The Nasdaq biotech index also was off, for the second straight day, but at a more modest 2 percent.
INTERIM ANALYSIS 'HASN'T' REDUCED OUR ENTHUSIASM’
Like many biotech stocks, Newlink has been on a tear the last 12 months, more than tripling in value through Feb. 25, when the stock hit a 52-week high of $53.48. Since then, the stock has slumped 40 percent.
Much of that decrease could be pinned on investor expectations. In a Feb. 24 note following Leerink Partners’ Global Healthcare Conference, analyst Howard Liang suggested that, for Newlink, “there appeared to be high interest in the timing of its interim readout,” noting that the company was still waiting for the number of events to be reached for the first phase III interim analysis. The company estimated that blended median survival was in “the high 20s months range, which compares favorably with historical estimate of about 19 months for the control group,” implying the vaccine treatment group was performing better in phase III than in phase II studies.
Newlink did not expect an early halt to the phase III study for efficacy, according to Link, but some investors apparently did.
“We’ve done a lot of thinking about this, and we’ve tried to guide the market to the fact that we didn’t think the first interim could be positive,” Link told BioWorld Today.
Because the company is conducting a log rank analysis, he explained, the issue with the first look is a matter of time on therapy. The second half of IMPRESS patients enrolled from June 2012 to September 2013, so a large portion of those patients only had about a year of follow-up.
“When you project out from our phase II trials, in the first six months there wasn’t much difference,” Link said. “At a year, there was about a 30 percent difference. By two years, there was a 60 percent difference. By three years, there was a 100 percent difference.” In short, the “pattern of survival” in immunotherapy shows much greater benefit as a trial progresses.
“We didn’t think there would be enough time for the second half of the population, which was going to be averaged into the first half of the population in the log rank analysis, to create enough curve separation to hit that statistic,” he said. “We’re not looking at medians here. We’re looking at the aggregation of every single event in the trial.”
Although Link conceded that “it’s hard to explain” the analytics to the market, he said the company remains confident about the direction of the trial and the ultimate outcome.
“We know for a fact that the overall median in the trial is significantly beyond 25 months and moving into the high 20s,” he said. “If the control arm is anchored somewhere within a few months of 18 months, which the last national study showed, we could still see quite a remarkable separation – a 40 percent or 50 percent difference – that wouldn’t be reflected by log rank analysis when half of the patient population had not had very long follow-up.”
Newlink expects the second interim analysis to occur in seven to 10 months, or toward year-end. Although that estimate “depends on the performance of the control arm,” by that time every patient will have been enrolled for at least one year and most for at least two years of follow-up, “which would be enough, in our estimation, to see pretty significant curve separation,” Link said.
“This is a very robust study that allows us to discern a 20 percent difference in the final look,” he added, and the interim analysis “hasn’t reduced our enthusiasm.” Part of the issue with assessing cancer immunotherapies is that the technologies are “fundamentally different” from each other, so it’s “a hard story to tell when there have been five or six or 10 different peptide vaccines that have failed.”
The technology underpinning Newlink’s vaccines was developed and published for more than a decade before the first program went into the clinic. When tested against peptide vaccines, “we found fairly dramatic superiority in all of the clinical work in this category of vaccines,” including settings beyond pancreatic cancer, Link said.
Newlink, of Ames, Iowa, also is testing algenpantucel-L in a second phase III study, dubbed PILLAR, (Pancreatic Immunotherapy with algenpantucel-L for Locally Advanced non-Resectable), in patients with locally advanced pancreatic cancer. That study, undertaken at the behest of IMPRESS investigators, is enrolling well.
A second HyperAcute product, tergenpumatucel-L, is in an adaptive design, randomized phase IIb/III trial currently accruing up to 240 patients with non-small-cell lung cancer. The company reported promising data on the therapy at last year’s American Society of Clinical Oncology annual meeting in Chicago. (See BioWorld Today, June 3, 2013.)
In addition, the company continues to advance its indoleamine 2,3-dioxygenase pathway inhibitor platform and is enrolling phase II studies of indoximod, the D isomer of 1-methyl-tryptophan, in phase II studies in metastatic breast and prostate cancer.
Newlink reported cash and equivalents of $52 million as of Sept. 30, 2013. Also in September, the company filed a prospectus supplement for an at-the-market offering of up to $60 million in common stock. The stock’s run “gave us some breathing room to raise additional funds,” Link said, noting the company would offer more color in its year-end 10-K financials, which could be filed as early as this week.
The company plans to keep the U.S. market for algenpantucel-L in pancreatic cancer to itself, reckoning that IMPRESS, at some 70 major U.S. centers, is prepping more than two-thirds of the surgeons who treat pancreatic cancer.
For the other, larger programs, Link said the company is already in discussions with potential partners.