Barron's Online -- November 10, 1997
Sweet Prospects?
Amylin's diabetes drug could be effective,
despite inconclusive test findings
By BILL ALPERT
Since she was 14, Heidi McNutt Paterson started each
day with a shot of insulin. Because she had diabetes,
she'd take more insulin before every meal, and
sometimes become disoriented because of her seesaw
blood-sugar levels. But two years ago, the Detroit
resident began testing a new diabetes drug from Amylin
Pharmaceuticals. Now, she doesn't need her morning
shot of insulin, she takes less insulin with meals, and on
days that she exercises, she needs no insulin at all. "My
blood sugar doesn't go up too fast or too high," says
Paterson, who is now in her mid-30s. "I feel more in
control."
Many others testing the Amylin drug were also able to
cut back on their insulin intake, a welcome change to
most diabetics, who frequently become dizzy and gain
weight from the aggressive insulin regime needed to slow
damage from the disease.
But changes in insulin usage among Amylin's test
subjects may be the reason the San Diego firm can't
convincingly prove that its drug, pramlintide, works. And
the inconclusive test results that Amylin made public in
August are why investors have dumped the company's
stock, pushing it down from $16 to around $8 recently
and halving the market capitalization of the company's
32 million shares to $250 million.
The failure to control insulin dosing looks like a goof in
Amylin's study design, blunting the effect that pramlintide
had on patients' diabetes. But Amylin went back and
improved controls in its ongoing pramlintide tests. Those
results won't be out until next year, but speculation
based on the muddled studies suggests that Amylin's
drug does work.
If the new drug pans out, it could be a half-billion-dollar
product for Amylin, which investors have shunned
because the firm is basically a one-product company.
Diabetes is a big
market for any drug
that works.
Although diseases
like AIDS and
cancer grab more
headlines, diabetes
is the fourth leading
cause of death in
America. It's the
leading cause of
adult blindness and
kidney failure, and
produces nerve
damage that can
lead to leg amputations. The disease and its
complications absorb 25% of Medicare dollars,
according to the American Diabetes Association, and
about 15% of total healthcare spending. Kaiser
Permanente, the large California HMO, concluded in a
recent study that it spends about $3,500 per year more
to care for its diabetic members than for its non-diabetic
ones.
The disease results from failures in the mechanisms that
regulate the body's level of sugars, like glucose. Insulin is
the main ingredient controlling glucose absorption in
human cells, and since insulin was identified in 1921,
injections of the drug have been the mainstay of
treatment and a big market for the likes of Eli Lilly.
In 1987, Oxford University researchers discovered a
second hormone that influenced sugar levels, and called
it amylin. This substance seems to regulate the release
of glucose from digestion and is normally secreted from
the pancreas with insulin.
Over 15 million Americans suffer from diabetes, only
half of whom realize they have the ailment. About a
million U.S. diabetics have juvenile-onset diabetes, in
which the pancreas loses the cells that make insulin and
amylin, sometimes before the patients are young adults.
Heidi Paterson has this type of diabetes, and already has
suffered slight damage to her eyes and kidneys.
But most diabetics have adult-onset diabetes, in which
the body becomes less sensitive to insulin, even though
the patient's pancreas may still produce the hormone. A
few new products have come on the market in recent
years for these diabetics, from such firms as Bayer AG,
Glaxo Wellcome and Warner-Lambert, but these
products don't work for people with juvenile-onset
diabetes. And last week, Warner-Lambert warned
doctors that its diabetes drug, Rezulin, may cause liver
problems.
The same year researchers reported discovery of
amylin, Ted Greene founded Amylin Pharmaceuticals
to develop medicines that would replace this partner
hormone to insulin. By 1992, the company had
concocted the synthetic analog to amylin that it calls
pramlintide, and had begun safety and effectiveness
studies on diabetes patients.
Early tests showed pramlintide was safe and that it
dispersed in the body like natural amylin. But the two
studies that Amylin reported in August of this year, and
that clobbered its stock, were the first of the so-called
Phase III trials that will form the basis for Amylin's
application for FDA approval. In these, patients took
pramlintide as well as insulin for a year, while control
groups took insulin and a placebo.
Researchers measured the drug's success by watching
patients' blood levels of glycated hemoglobin, a special
form of the oxygen-carrying molecule that rises as a
percent of total hemoglobin, when blood sugar levels are
high. Non-diabetics normally measure below 6%, while
many diabetics have glycated hemoglobin levels above
8% or 9%. Studies have shown that consistently high
readings in the hemoglobin test are strong predictors of
diabetes complications, such as blindness and kidney
disease.
But the patients who received pramlintide failed to show
impressive drops in their glycated hemoglobin measures,
on average. Juvenile-onset diabetics who were the
subject of one of the studies showed a decline of 0.3%
after a year on pramlintide, when compared to patients
on a placebo. Adult-onset diabetics showed declines of
roughly the same percent.
Investors had expected the hemoglobin measure to drop
by at least 1%, and some analysts pronounced
pramlintide ineffective. But Amylin had an explanation
for its drug's weak showing: The studies weren't
well-designed. In both studies, patients and their doctors
were free to vary their insulin dosage, and that's just
what they did. Patients on pramlintide decreased their
daily doses of insulin, while patients on the placebo
increased theirs.
Heidi Paterson was able to take less insulin, after getting
the Amylin drug in the study of juvenile-onset diabetes.
So was Evelyn Hawkins, a retired Detroit teacher who
used the drug as a participant in an adult-onset diabetes
study. In one branch of this study, patients on
pramlintide cut their insulin and those on the placebo
boosted insulin; in fact, the people in the placebo group
were using 12% more insulin by the end of the trial than
they had been at the beginning, compared to the group
that got Amylin's drug.
The changes in insulin doses made the patients better off,
but Amylin says it compromised the studies as tools for
isolating pramlintide's effect on glycated hemoglobin
levels. Researchers who weren't involved in Amylin's
study concur that varying insulin levels can throw off
measurement of a new drug's effect on diabetes. "That's
always true of a diabetes trial," says Dr. James E.
Everhart, chief of the clinical trials branch of the National
Institute of Diabetes and Digestive and Kidney Diseases.
"Changing insulin dosage throws a monkey wrench into
it. But it's a monkey wrench that can be planned for."
As
the
two
Amylin
studies
stand,
the
0.3%
glycated
hemoglobin
effect
leaves
doctors
as
disappointed
as
investors
-- although the studies haven't been fully reported in
professional journals. "As I look at these numbers, I'm
very disappointed," says Dr. P.J. Palumbo, an
endocrinologist at the Mayo Clinic in Scottsdale,
Arizona. A good diabetes treatment should bring
glycated hemoglobin levels down one or two percentage
points to get patients as close as possible to a normal
6%, says Palumbo, who concludes: "I'm not convinced
that small a change is going to have a long-term impact
on the patients' complication rate."
Reviewing the data from the first two studies, in an effort
to tease out pramlintide's effect on blood sugar, Amylin
looked at patients who had maintained steady insulin
levels throughout the study and who had started with
high levels of blood sugar, as indicated by glycated
hemoglobin levels of at least 8%.
In these subgroups, pramlintide seems to have had a
bigger impact on blood sugar levels, reducing the
glycated hemoglobin measure after one year by 0.66%
in juvenile-onset diabetics and by as much as 0.85% in
adult-onset diabetics, in comparison to patients who just
got a placebo.
Unfortunately, this analysis was performed after the
study ended, so the results lacked the reliability they
would have possessed had Amylin controlled insulin
levels in every randomly selected patient enrolled in the
studies. The company tacitly acknowledged its design
booboo by revising the design of its remaining four
Phase III trials so that participants maintain stable levels
of insulin.
Although reduced glucose levels were Amylin's primary
target in the studies, patients who got the Amylin drug
benefited in other important ways. Adult-onset diabetics
lost four pounds after a year on pramlintide, while
placebo patients gained three pounds. Juvenile-onset
diabetics receiving the drug also lost weight and saw
their blood serum ratios of "good cholesterol" to "bad
cholesterol" improve by 9%, compared with the ratios
for the people in a placebo group. There were no issues
with safety or low blood sugar.
Of course, the new and improved trials could prove to
be busts. But Johnson & Johnson, Amylin's partner in
developing pramlintide, has said it remains committed to
the drug. So far, J&J has invested $155 million in
Amylin's drug program, under an agreement to split
development costs and any future profits.
Amylin Chairman Ted Greene says that results of the
well-controlled Phase III trials will be reported next
year. And, he asserts, he is still pretty confident that the
drug will win FDA approval by the end of 1999. If
Amylin can sell its drug for $1,000 per year to a small
percent of the seven million insulin users in North
America and Europe, the little company's half-share of
profits could easily be hundreds of millions of dollars. If
Amylin maintains expenses at current dollar levels, it
could net $2-$3 a share from pramlintide. Those future
earnings would merit an Amylin share price well above
the recent $8.
Testimonials aren't proof of a drug's effectiveness, but at
the clinic level, pramlintide users already have given the
drug a vote of approval. Over 75% percent of the
people in the two studies have elected to continue using
pramlintide. "I can't pry the drug out of their hands," says
Davida Kruger, a nurse practitioner at Detroit's Henry
Ford Health Systems. Besides taking less insulin, Evelyn
Hawkins says she has lost weight and now can walk
three miles several times a week.
Heidi Paterson says her glycated hemoglobin now
averages under 7.5% and she has seen her rate of
diabetes complications slowing down. The kidney
damage she has suffered as a result of the disease has
stabilized, and she has only needed one laser treatment
for eye damage since starting on pramlintide, compared
with two in the comparable prior period.
"I'm sure diabetes will be the reason I die," she says
matter-of-factly, "but I'd like to live as healthy a life as I
can until then." |
