Here is the story from the Pittsburgh Post-Gazette as reported in the dual BICO-VITK thread:
12:57 Bicosend: Biocontrol's Diabetes Monitor Still Faces Hurdles
Pittsburgh Post-Gazette
By Patricia Sabatini
Copyright 1996
Anxious shareholders and diabetics waiting for Biocontrol Technology Inc.'s
experimental blood sugar meter to win government approval any time soon appear
headed for another disappointment.
Disregarding unequivocal advice from a Food and Drug Administration panel
that rejected the machine in February, Biocontrol last month submitted a new
application that isn't backed by the large-scale clinical testing that panel
members and FDA officials said they wanted to see.
Biocontrol had been told it would need the new test results to prove the
device worked, after the advisory panel concluded the data it reviewed showed
the machine was potentially dangerous.
Panel members also had stressed that the Diasensor, which was developed to
measure glucose levels by shining an infrared beam through the forearm instead
of drawing blood, needed to be tested on patients where it's intended to be
used - in the home.
Biocontrol, however, hasn't performed the recommended clinical studies or
tested the Diasensor in any homes, Chief Executive Officer Fred Cooper told a
congressional subcommittee in September. He indicated the only new data
submitted was collected from a handful of patients the company tested at its
own Indiana, Pa., research labs.
"That clearly falls short of what my expectations would be and the panel's
expectations in general," Dr. Jose Cara of Henry Ford Hospital in Detroit, one
of the panelists at February's review, said in a recent telephone interview.
"I would expect to see a (patient sample) that is going to give
statistically significant data ... data on 50 to 100 individuals with
diabetes, who are using the monitor pretty much on a daily basis, either in a
home or office setting."
Unfortunately, he said, the panel's recommendations "seem to have fallen on
deaf ears."
Biocontrol's public relations office did not respond to questions regarding
the new submission last week, and Cooper turned down a request for an
interview.
FDA officials also declined to provide details on the new submission, saying
the agency honors companies' requests for confidentiality.
But in a press release this summer, Biocontrol said it was planning only to
test seven patients, for a total of 60 days - even fewer subjects than the
company presented the first time.
At the congressional hearing, convened to review the FDA's role in
regulating home-testing devices, Cooper hinted at one explanation for
Biocontrol's apparent snub.
"Frankly, we feel our data is very good with our (previous) submission,"
Cooper said when asked by Rep. Ron Klink, D-Murrysville, if the company's new
application would address any of the advisory panel's concerns.
"`We hope this time we will get a fair assessment."
At the hearing, Cooper alleged the company had been treated unfairly by the
FDA and called on Congress to investigate. Afterward, lawmakers said they saw
no basis for such a probe.
"We told them to do a better job of showing their product worked," Klink
said, noting that the panel's decision had been unanimous.
Cooper had charged that panel members didn't approve the Diasensor because
they lacked the background necessary to understand its "complex technology,"
not because there was anything wrong with the device.
But Dr. Arthur Karmen, the physician who chaired the advisory panel,
disagreed.
He said the blood glucose readings Biocontrol submitted for the eight
patients it considered a success "really didn't have any correlation" with the
patients' actual glucose levels.
"You could have low blood sugar and (the Diasensor) would read the same as
if you had a high blood sugar," Karmen, professor at the Albert Einstein
College of Medicine in the Bronx, said at the hearing.
"As far as I was concerned, I thought the company ought to go back to the
drawing board."
He said he hoped the Diasensor's technology "is explored, but explored
critically and improved. Because the way it is now, I don't think it would be
safe to use."
Cara agreed.
"I thought the (idea behind) the technology was wonderful," he said in an
interview.
"The reason why I had problems with the Diasensor wasn't because of the
technology, but just because it was a lousy meter."
The company contends it has improved the Diasensor since the panel review,
mainly increasing how often it's able to produce a response. (Previously, the
Diasensor only gave a reading, accurate or inaccurate, about half the time it
was asked for one.)
Still, Cooper indicated at the congressional hearing that the test results
in the company's new application wouldn't be any better than the last time,
saying the Diasensor is unchanged from the one reviewed in February.
"Sir, it's the same device. It just has some lettering on it and it's a
little smaller. It's the same device," Cooper told Rep. Joe Barton, R-Texas.
Cooper contended advisory panel members rejected the Diasensor because they
didn't understand what they were looking at.
But panel members said they knew exactly what they were looking at, and
didn't like what they saw.
They noted, among other problems, that seven of the eight patients who the
company said were successfully monitored showed wide variations in glucose
readings taken during a single test session.
One patient, for instance, who was tested on the Diasensor three times
within 35 minutes, received readings ranging from 172 to 478 milligrams.
"What's the patient supposed to believe?" one panel member asked
incredulously.
Another "successful" patient essentially received constant readings from the
Diasensor over the entire month-long trial, despite the fact the actual
glucose level "was moving all around."
Panel members said they also were disturbed that the company didn't provide
any data in very low blood sugar ranges, levels at which an inaccurate reading
could be fatal.
And the panel stressed that long-term studies were essential to show that
the Diasensor, which the company attempts to program - or calibrate - for each
patient, continues to function for a reasonable length of time.
Some of Biocontrol's data suggested the Diasensor essentially shut down in
the latter half of the one-month study the company performed.
Biocontrol had proposed avoiding the issue by requiring diabetics to bring
the personal computer-sized Diasensor, which it expects to sell for about
$8,000, back to the company's research labs each month for a checkup. If the
machine was malfunctioning, it would then be "recalibrated."
"Presumably a company would not intentionally show that kind of poor (test
data) and expect to get a favorable review," Karmen said in a recent
interview.
"That part is still somewhat mystifying to me."
Biocontrol shareholders apparently have been wondering the same thing.
The company is facing a number of lawsuits, filed following the panel review
and seeking class action status, alleging the company intentionally misled
stockholders about the progress and effectiveness of the Diasensor.
Biocontrol has denied the allegations.
Cooper maintained at the congressional hearing that the company was forced
into the panel review before it had enough data, and without sufficient time
to prepare.
Presumably, though, Biocontrol should have been ready with its data more
than two years earlier, back in January 1994, when the company first asked for
clearance to sell the Diasensor.
And at the same time Cooper was accusing the FDA of rushing the company, he
blasted the agency for moving too slowly and allowing the company's
application to "languish" in an "FDA black hole."
According to agency officials, the FDA offered a panel review in the summer
of 1995 and again in the fall, but Biocontrol balked, saying it wasn't ready.
And soon after February's review, the FDA said it offered to work with
Biocontrol to design a "mutually acceptable" study protocol to further
investigate the Diasensor and allow the company to recover development and
production costs by charging for the machines to be used in the studies.
"Regrettably, the company did not avail itself of this option," the FDA's
Dr. Bruce Burlington testified at the September hearing. "We've had no further
substantive response from the firm in the intervening months" since the
review, Burlington, director of the agency's Center for Devices, said.
Biocontrol hasn't said why it turned down the offer.
In the months following the review, some observers have questioned why the
Diasensor was even brought before a panel of experts from across the country,
given the sparse nature of the test data.
Karmen said he believes the reason involved public pressure.
The FDA likely was going to reject the Diasensor, he testified, "and the
decision was probably not going to be one that was going to be popular, and
(the FDA) wanted some support from outside experts."
The FDA's Burlington said that although the agency may have "serious
questions" about whether a product under review is ready to market, a panel of
outside experts may "see things differently and change our minds."
"Or, if we haven't fully formed a conclusion (the review) helps us form
one," he said.
But Biocontrol's Cooper, leveling vague accusations of bias at the agency,
alleged something more sinister was at work.
"I've believed for some time that the FDA has another agenda, and that
agenda is not the agenda we're on," he said at the hearing.
He repeated his assertions that several members of the 19-member panel had
conflicts of interest (not Karmen or Cara, incidentally) and that those
panelists "influenced many, many of the (other) panel members."
Cara, for one, scoffed at such charges.
"They are people, including myself, that gave their honest opinion about the
validity of the data and the validity of the machine," he said in an
interview.
He added that he was "excited" about the prospect of such a machine and the
relief it might bring to diabetics who endure a painful daily ritual of
pricking their fingers.
If anything, "I was looking for an excuse to approve it," Cara said.
Ironically, Biocontrol had acknowledged at the February review that it did
indeed need to perform large-scale studies with the Diasensor.
The company proposed testing its "first 60 patients" at such hospitals as
Johns Hopkins in Baltimore and Mercy in Pittsburgh - if the FDA would approve
the Diasensor first.
The company also said it would test the Diasensor on 20 insulin-dependent
diabetics in their homes.
"We don't think (the Diasensor) is perfect. We don't like the way it is,"
the company's chairman, David Purdy, told the panel.
But he said Biocontrol needed marketing approval "to make the money that we
need to do the testing that you would like us to do."
Still, Cooper has sought repeatedly since then to assure anxious investors
that Biocontrol has the cash to cover the additional testing required for
approval.
Whether Biocontrol's latest attempt to bring the Diasensor to market goes up
in smoke or not, one thing is clear - it hardly would be the first time the
company's pronouncements have exceeded the firm's ability to deliver.
Over the years, the company has repeatedly claimed a saleable product was
just around the corner.
As far back as 1989, for instance, the company was telling investors to
expect a sensor the size of a cereal box to be ready for use in doctors'
offices within the year, and a home-use model the size of a portable phone
within two years.
But perhaps the most glaring example of hyperbole came in January 1994, when
Biocontrol first requested FDA approval to sell the Diasensor.
The company based that initial submission on six patients it studied using a
prototype machine "that didn't analyze patient data, or produce actual
readings," according to testimony at the congressional hearing by the FDA's
Burlington.
The machine stored information on blood sugar levels, which was downloaded
and analyzed only by the company at its research center, he said.
Questions of accuracy aside, since the device didn't provide a readout, it
couldn't be used by diabetics to adjust their diet or insulin injections - the
very purpose the machine was intended for.
Nevertheless, Cooper issued a statement at the time asserting FDA approval
was imminent.
"We are commencing with our manufacturing start-up," he declared, in
anticipation of approval in "90 to 120 days."
(END) 11-11-96
12 57 EST
(At this writing BICO is up 1/8 at 1 29/32 on volume 529,300) |
