Diagnostics & Imaging Week Washington Editor

GAITHERSBURG, Maryland — In a vote of 5-2, an FDA panel on Friday recommended against approval of a test kit from Nymox Pharmaceutical (Maywood, New Jersey), offered by the company as an aid in the diagnosis of Alzheimer's disease (AD).

Clinical trial results of the urine test, called AlzheimAlert, failed to convince panel members that it could provide an additional level of diagnostic reliability for the disease.

Panelists voting against approval cited too few patients enrolled, the number of patients who dropped out of the study, lack of sufficient sensitivity, and other study design flaws. One panelist, in particular, cited data "squishiness."

The test is designed to detect levels of neural thread protein (NTP), a brain protein found in high levels in Alzheimer's patients. NTP is found in patients' brain tissue, cerebrospinal fluid, and urine.

AlzheimAlert already is certified as a laboratory testing service in the U.S. through a CLIA waiver. However, at present, the only way to obtain a 100% accurate diagnosis of Alzheimer's is after a patient dies and an autopsy is performed.

The company's premarket approval application (PMA) would have extended the diagnostic tool's use to hospitals, primary care physician (PCP) practices, and other laboratories to perform their own analysis.

Physicians currently have few options for making even a non-definitive diagnosis of the disease. Those available include cognitive tests, use of medical and family history and brain scans.

Nymox officials argued that their test would give clinicians one more tool to aid in the diagnostic process, and provide more data physicians could use to decide whether more in-depth and costly diagnostic procedures would be necessary — or not.

Responding to the panel's concern about limited patient data, the company cited the not unexpected difficulties in gathering data from patients suffering from dementia.

Nymox has been down this part of the regulatory road before.

In 2004, the FDA sent the company a letter raising issues regarding its clinical study methodology and ruling the PMA not approvable.

The panel's not-approvable recommendation does not restrict the company's ability to market and sell the current version of its test under CLIA, or in Europe where it received CE-marking last year.

Over the last several years Nymox has processed more than 1,500 tests, according to data presented by company representatives at the meeting.

"This is not intended as a stand-alone diagnostic," Paul Averback, CEO of Nymox, told the panel. "There is no downside for anyone. There are no questions of safety, and this can provide very useful information for the primary care physician."

More than one panel member, however, cited concerns about effectiveness and voiced concerns about false positives or other inaccurate results that could leave patients open to "misdirected management of care."

Asked by Diagnostics & Imaging Week to respond following the vote, Averback would not comment on the panel decision.

"We are very pleased to have had the opportunity to present the results of the blinded nine-center trial, which showed that 89.5% of clinical cases of probable AD had elevated NTP and 91% of definite non-AD had normal NTP levels," he said in a statement released by the company on Monday. "The panel, however, felt that more data was required for approval, including, for example, post-mortem verification and clarification of how to use the test results."

The company, he said, would continue to work with the FDA toward "fulfilling the further data requirements that the agency may have in conjunction with the panel suggestions."

Some panelists pulled no punches in their criticism of the company's study design.

"It's like I am in the television show 'The Twilight Zone' when I listen to the sponsor's presentation," said Brent Blumenstein, a statistician with Tri Arc Consulting (Seattle), and an outspoken member of many FDA panels. "I find the analysis does not meet the study design and the data does not match the proposed indication. There is a great deal of squishiness in the data."

He added: "I don't think PCP's can handle these kinds of patients."

Company data seemed to disagree. Roughly 65% of all Alzheimer's patients are handled by PCPs, according to Averback.

"There are an infinite number of choices in the practice of medicine," Averback told the panel. "We're simply saying that this test adds useful information. The standard of approval is whether this device will influence diagnostic decision making."

Susanne Gollin, a panelist from the University of Pittsburgh, said: "I don't think the test is useful in refining the process of physicians' decision-making."

"I think the issue is where NTP really serves as a reliable biomarker of the disease," said Joseph Parisi, MD, a neuropathologist with the Mayo Clinic (Rochester, Minnesota).

Panel members supportive of the device were equally assertive.

"It sounds like we're trying to say that the results are black or white," said William Duffell Jr., PhD, head of government and regulatory affairs for Gambro BCT (Lakewood, Colorado) and a non-voting industry representative on the panel. "This is another piece of data in the medical bag, so to speak, in making a diagnosis. We're not supposed to draw a line on one side or the other. But I see a higher probability that physicians would come to a higher overall accurate conclusion with this test. There is value to this test."

"As a neurosurgeon who manages these patients, I think that the test can provide some valuable initial information and time will tell," said panel member Terrance Lichtor, MD, with Rush University (Chicago). "The downside is really nothing, and there are no risks that I see. But I think that there is a possibility that it would [be helpful], and I don't see any reason it can't be added to the number of other tests we order for these patients."

About 4.5 million Americans are suspected of suffering from Alzheimer's, which usually sets in after age 60 and gets worse with age. Symptoms include confusion and memory problems. Patients can also have difficulty reading, writing and speaking, and they also can become aggressive or anxious.

The FDA is not bound by the recommendations of its panels but usually follows them.