Over 30 years of reporting on Congress, federal agencies and the White House for corporate America as well as national trade and professional associations.

OPENING THE DOOR TO FOLLOW ON PROTEINS

October 2006 issue of Biotechnology Healthcare

When the FDA approved the follow-on protein Omnitrope in May, it gave generic drug
makers the wedge they were hoping for. With pressure building in Washington, did
Omnitrope push the door open – or was it just an anomaly? BY STEPHEN BARLAS


When the U.S. Food and Drug Administration last May approved Sandoz’s Omnitrope — a follow-on protein to Pfizer’s Genotropin, the leading biotech human growth hormone
— hope sprung in the generic drug community that perhaps the agency had finally seen it their way, setting a precedent that would allow for the production and sale of follow-on proteins in the United States.
Omnitrope (somatropin) is the first follow-on protein from a generic pharmaceuticals company that the FDA has ever approved. Equally significant — and troubling
to the biotechnology indus-try — is the FDA’s near-withering 52-page reply to two biotech manufacturers and the Biotechnology Industry Organization, whose citizen
petitions had marshaled a phalanx of legal and regulatory arguments intended to persuade the
agency not to approve Omnitrope. Essentially, the three petitions said the FDA could not approve Omni-trope nor any other follow-on protein submitted via the 505(b)(2)
pathway, which otherwise permits a sponsor to rely on published studies or the agency’s finding of safety and effectiveness for an approved drug to support approval. The BIO
petition, for instance, cited significant differences between therapeutic protein products and chemical drugs, in terms of complexity and heterogeneity, and thus argued that the use of other companies’ data is no assurance of safety.
Perhaps more important than the approval of Omnitrope itself were the questions it raised. Was this a one-time shot, or did it create a defacto path for others to follow —even before a formal regulatory road map for follow-on biologics is developed and approved? Would
any future approvals be on a caseby-case basis? And how will Congress follow up?
Even though the FDA’s approval was epochal, the agency bent over backwards to refute that impression. In a Questions and Answers document posted on its Web site when it approved Omnitrope, the agency downplayed the precedent value of its approval by citing other
follow-on proteins it has approved under 505(b)(2): GlucaGen (glucagon recombinant for injection), Hylenex (hyaluronidase recombinant human), Hydase and Amphadase (hyaluronidase), and Fortical (calcitonin-salmon recombinant) nasal spray.
Tom Newton, PhD, pharmaceuticalmarket analyst for visiongain, a United Kingdom consulting company that recently published the exhaustive report Biogenerics 2006:
Challenges Ahead for an Emerging Market, says companies such as Novo Nordisk (GlucaGen), Halozyme (Hylenex) and PrimaPharm (Hydase) cannot really be thought of as “generic” manufacturers in the sense that Teva and Sandoz are. “In my opinion, that makes the application for Omnitrope significantly different from these examples,” he says. More importantly, he adds, all of the follow-on proteins alluded to by the FDA in its Q&A “appear to
be modifications of treatments already on the market, whereas Omnitrope is presented as a pure biogeneric.” Adds Newton, “This case has definitely raised the profile of biogenerics.
The authorities will come under increasing pressure to do something about it.”

WILL OTHERS FOLLOW?
Officials at top biotech companies concede that Omnitrope is likely to be succeeded by other
follow-on products, which is what has happened in Europe where follow-on proteins are referred to as “biosimilars.” David Beier, senior vice president for global government affairs at
Amgen, points to guidance documents published by the European Medicines Evaluation Agency
(EMEA). EMEA has developed both clinical and nonclinical guidances for recombinant products containing human insulin, somatropin(human growth hormone), granulocyte
colony stimulating factor (GCSF), and erythropoietin (EPO), and future guidance is expected on ainterferon and immunogenicity. Those publications, says Beier, led to the EMEA’s approval of Omnitrope and Valtropin, a recombinant human growth hormone, and its rejection
of a hepatitis C product. “Those European regulatory requirements in terms of safety, efficacy,
and pharmacokinetics are very, very similar to innovator products, but not identical,” Beier says. “In the main, although not in every detail, EMEA guidance has effectively
protected patient safety.”
Given the EMEA conditions for approval, and because in many cases newer innovative biotechnology medications are available that offer advantages over the older medicines that follow-on versions attempt to imitate, Beier feels that biologic follow-on products could
play a limited role in the marketplace by offering alternative products. But, he adds, the price advantage for biosimilars in Europe is substantially more modest than for the differential between brandname conventional drugs and generic copycat products. When they
first come on the market, European biosimilars offer a 10 to 20 percent price advantage over innovator drugs, according to Beier. Moreover, he says, there are fewer innovator
drugs coming off patent in the next five years than the generic industry “overestimates,” and biosimilars, which will have a less-robust safety profile, will offer no therapeutic advantage.
Ajaz Hussain, vice president and global head for biopharmaceutical development at Sandoz, agrees that the price differential in Europe for biosimilars will be smaller compared with traditional generic drugs. But, he argues, a follow-on product that costs that much less
than an innovator’s drug — whose annual cost to the patient may be in the $20,000 to $100,000 range —still amounts to an immense savings. Hussain declines to comment on how Sandoz will price Omnitrope in the United States.
Sandoz is already looking beyond Omnitrope, based on what Hussain describes as the FDA’s
precedent-setting decision, to other drugs the company could submit via the 505(b)(2) process. “We certainly plan to use it when we have a product that would fit the criteria,” he says, though he declines to be specific about possible candidates.

FDA: WHERE NEXT?
Omnitrope’s approval and the EMEA guidance documents would appear to point the FDA toward its next step: publication of guidance for approval of follow- on proteins submitted via
the 505(b)(2) pathway, which was developed before the emergence of complex biopharmaceuticals.
The U.S.agency has had some public workshops over the past two years, and has progressed on a guidance document in fits and starts, but it may be years before a guidance document is
drafted, vetted through public hearings, revised, and implemented. Established by the 1984 Drug Price Competition and Patent Term Restoration Act (commonly known as Hatch-Waxman), the 505(b)(2) pathway may be available for follow-on versions of drugs approved under section 505 of the Food, Drug, and Cosmetic Act, such as hGH and insulin, which are small-protein drugs containing few sugars —making them easier to duplicate.
Follow-on proteins in these categories are not exact copies, owing to the inexactitude of reproducing a drug via biotechnology — and, to be sure, Omnitrope itself is not rated as
therapeutically (AB) equivalent to any other human growth hormone and, therefore, is not substitutable for one. The 505(b)(2) pathway may be used, however, for a follow-on protein product that is sufficiently similar to an approved drug product to permit reliance, where scientifically justified, on certain existing information (including the FDA’s findings
of safety and effectiveness for an approved drug product) and may relieve the applicant from having to submit full-scale supporting data.
But GlucaGen, Hylenex, Hydase and Amphadase, and Fortical, all approved under 505(b)(2), are relatively simple molecules. The majority of biopharmaceuticals, infinitely more complex, are not approved as drugs under the Food, Drug, and Cosmetic Act, but licensed as biological products under section 351 of the Public Health ServiceAct — the “gold mine” route for
the generic drug industry. There is no approval pathway analogous to 505(b)(2) for products licensed under section 351.
That isn’t stopping generic drug makers such as Sicor, LG Chemicals, GeneMedix, Cangene, Rhein Biotech, Dr. Reddy’s Laboratories, Wockhardt, and Dragon Biotech from already supplying interferons, erythropoietin, and other biopharmaceutical products to Lithuania,
Mexico, China, Korea, India, Argentina, Egypt, Peru, and Brazil. These companies lick their chops at the prospect of getting FDA approval for those drugs, which are infinitely more complicated than human growth hormone and insulin.
Most observers, even those in the generic industry, agree that Congress would have to give the FDA new authority before it could approve generic versions of section 351 biopharmaceuticals. Amgen’s Beier notes that any new approval process for products regulated under section 351 must be constructed in a way that ensures patient safety and respects the intellectual property rights of the innovators. As Beier points out, a would-be sponsor of a followon biologic would be using a different cell line and different growth media to produce the
protein, and would likely use different fermentation methods, purification processes, and specifications. “Because of the inherent differences in these materials and processes, a generic sponsor cannot produce the same product as the pioneer,” he argues.
Genentech made a similar point in its April 2004 citizen petition, which asks the FDA to refrain from establishing standards for “similarity” of biotechnology-derived products under 505(b)(2). Alluding to safety concerns, the company maintained that “Current science
[does not] allow for reliance on analytical data and information generated from one biotechnologyderived product to support approval of a product manufactured through a different process.” Varied manufacturing processes, it pointed out, affect product purity
and can lead to immunogenicity. Genentech’s petition also gets to Beier’s point about respect for intellectual property rights, contending that any company that relies on another’s data to make safety and efficacy claims for a product that is not an exact copy of the innovator drug has, in essence, unfair access to the innovator’s trade secrets. The petition refers to a 505(b)(2) application from Dr. Reddy’s, an Indian company, for amlodipine maleate, based on the FDA’s approval of Pfizer’s amlodipine besylate (Norvasc), a calcium-channel blocker. The FDA stayed the effective date of the approval of Dr. Reddy’s product “because questions [were] raised
about the source of the data the [FDA] relied on in approving” the application, according to the petition. “We are similarly concerned about the protection of our confidential commercial information.”
Congress is notably slow-footed, so whether legislation pertaining to section 351 passes in the near future is unlikely. What is more likely over the next six months to a year is the submittal of additional 505(b)(2) applications by major generic drug companies.

FIGHTING BACK
Of course, Pfizer is worried about Omnitrope. In its May 2004 citizen petition, Pfizer argued that the FDA could not legally approve Omnitrope via 505(b)(2) because the agency would have to access confidential Pfizer manufacturing and clinical data to do so. The Genentech and BIO petitions made a different point: The FDA could not issue guidance for generic drug companies on how to navigate the 505(b)(2) process, though they cited reasons similar to the ones Pfizer
used in its anti-Omnitrope petition.
When it approved Omnitrope, the FDA sent a 52-page explanation to each of the three petitioners, carefully knocking down every one of the industry’s objections like pins at the end of a bowling lane — and seemingly with enough force that those pins could fly across lanes.
Briefly, the agency said that the active ingredient in Omnitrope, somatropin, is highly similar to the
active ingredient somatropin in
Genotropin. Sandoz was able to
demonstrate that Omnitrope was
“sufficiently similar” to Genotropin
to warrant reliance on FDA’s finding
of safety and effectiveness for
that drug. Somatropin is the active
ingredient that has been part of
seven different recombinant hGHs
the agency has approved since somatrem
(Protropin) in 1985. Sandoz
also provided extensive independent
evidence of Omnitrope’s
safety and effectiveness for use in
pediatric patients with growth hormone
deficiency through three sequential,
multicenter, phase 3 pivotal
trials over a 15-month period,
along with other supportive data.
Moreover, the FDA said it did not
depend on any trade-secret information
submitted by Pfizer to approve
Omnitrope.
So the FDA seemed to be arguing
that small follow-on proteins, as
opposed to big molecules — at
least in a couple of therapeutic categories
— could be approved without
reliance on brand-name trade
data, and the FDA will continue to
do just that. The unstated message
was: stop complaining and live
with it.
Whether Pfizer accepts that message
is another matter. Paul Fitzhenry,
a spokesman for Pfizer, says
his company has not yet decided
whether to take the FDA to court
over its approval of Omnitrope.
The Genentech and BIO petitions
have even broader implications, because
they argued that the FDA does
not have the authority to either approve
follow-on proteins or guidance
for generic drug companies.
Walter Moore, vice president of government
affairs for Genentech, says
his company has made no decision
yet on whether to take the FDA to
court. But he adds, “We do not consider
the FDA letter a full response to
our citizen petition by any means.”
Does the FDA even need to issue
guidance at this point? Its answer
to the petitions may be guidance
enough for companies making follow-
on protein products that fit an
Omnitrope-type profile. An FDA
spokesperson, Karen Mahoney says
that, “The Agency will not comment
on the status of unpublished
draft guidance.”
WHAT WILL CONGRESS DO?
Some members of Congress believe
that FDA guidance on how
follow-on protein products can
clear through 505(b)(2) would be a
useful political exclamation point,
such as Republican Sen. Orrin
Hatch of Utah and California Democratic
Rep. Henry Waxman,
authors of the landmark 1984 law
that opened the door to expedited
FDA approval of abbreviated applications
from generic drug companies.
Earlier this year, Hatch and Waxman sent a letter to the FDA
asking it to publish guidance on
hGH and insulin, two categories
amenable to follow-on proteins.
On Sept. 29, Waxman, along with
New York Democratic Sen. Charles
Schumer, introduced the “Access to
Life-Saving Medicines Act,” which
would amend section 351 to approve
abbreviated applications for
biologics that are “comparable” to
previously approved products. The
bill defines comparable as demonstrating
no clinically meaningful
differences. Introduced a
week before Congress recessed
for elections, the bill got its
sponsors some publicity back
home but won’t see any action
before the the 109th Congress
officially closes next month.
A more likely congressional
response to the FDA’s consideration
of Omnitrope would
be legislation forcing the FDA
to make decisions promptly on
individual 505(b)(2) applications,
even when citizen petitions
are filed. Innovator biotech
companies have filed a couple dozen
citizen petitions in the past few years
in an effort to prevent FDA approval
of both follow-on proteins and conventional
generics.
At hearings the Senate Select
Committee on Aging on July 20,
Gary Buehler, director of the FDA’s
Office of Generic Drugs, said that
an agency evaluation of 42 petitions
answered between 2001 and 2005
showed that 33 had been denied in
full, 3 denied in part, and 6 granted.
“While the citizen petition process
is a valuable mechanism for the
agency to receive information from
the public, it is noteworthy that very
few of these petitions on generic
drug matters have presented datatherapeutically (AB) equivalent to
any other human growth hormone
and, therefore, is not substitutable
for one. The 505(b)(2) pathway may
be used, however, for a follow-on
protein product that is sufficiently
similar to an approved drug product
to permit reliance, where scientifically
justified, on certain existing information
(including the FDA’s findings
of safety and effectiveness for an
approved drug product) and may relieve
the applicant from having to
submit full-scale supporting data.
But GlucaGen, Hylenex, Hydase
and Amphadase, and Fortical, allapproved under 505(b)(2), are relatively
simple molecules. The majority
of biopharmaceuticals, infinitely
more complex, are not
approved as drugs under the Food,
Drug, and Cosmetic Act, but licensed
as biological products under
section 351 of the Public Health Service
Act — the “gold mine” route for
the generic drug industry. There is
no approval pathway analogous to
505(b)(2) for products licensed
under section 351.
That isn’t stopping generic drug
makers such as Sicor, LG Chemicals,
GeneMedix, Cangene, Rhein
Biotech, Dr. Reddy’s Laboratories,
Wockhardt, and Dragon Biotech
Pressure is likely to mount on
Congress and the FDA to address
the absence of section 351 followon
protein products because of the
cost of drugs like Epogen, Procrit
and Eprex — the three biggest selling
versions of epoetin alpha, a recombinant
form of erythropoietin,
which is a hormone that stimulates
the production of red blood cells.
Epoetin alpha brands and their derivatives
are the most successful
biotech drugs on the pharmaceutical
market. The first process
patent to expire for epoetin
alpha lapsed in 2001 in Europe
and 2004 in the U.S., according
to visiongain. Most epoetin
products are already off patent
— approximately 70 percent
of the total market has lost
patent protection.
Former Deputy FDA Commissioner
William Schultz,
now a Washington lawyer
who has represented the
Generic Pharmaceutical Association
and generic companies,
says a coalition of business and
health groups is pushing Congress
to give the FDA the authority it
needs to approve more follow-on
protein products. “Cost of biopharmaceuticals
is pushing this issue,”
says Schultz, “not just from generic
companies and consumers, but
from payers in the private sector
and the federal government.”
Congress may get its first chance
to amend section 351 in 2007 when
Congress must reauthorize the Prescription
Drug Users Fee Act. It will
be an event worth watching. BH
Stephen Barlas has covered the FDA and
drug issues since 1981 when he became
a full-time freelance Washington journalist
for business and trade publications.