PUBLISHED

UNITED STATES COURT OF APPEALS
                FOR THE FOURTH CIRCUIT


SIGMA-TAU PHARMACEUTICALS,            
INCORPORATED,
               Plaintiff-Appellant,
                v.
BERNARD A. SCHWETZ, Acting
Principal Deputy Commissioner,
Food and Drugs; TOMMY G.
THOMPSON, Secretary, Department of            No. 01-2206
Health and Human Services,
              Defendants-Appellees,
                and
GENSIA SICOR PHARMACEUTICALS,
INCORPORATED,
               Intervenor-Appellee.
                                      
           Appeal from the United States District Court
            for the District of Maryland, at Greenbelt.
                Catherine C. Blake, District Judge.
                        (CA-01-1377-CCB)

                      Argued: April 3, 2002

                      Decided: May 2, 2002

  Before WILKINSON, Chief Judge, WIDENER, Circuit Judge,
    and Walter K. STAPLETON, Senior Circuit Judge of the
      United States Court of Appeals for the Third Circuit,
                     sitting by designation.



Affirmed by published opinion. Chief Judge Wilkinson wrote the
opinion, in which Judge Widener and Senior Judge Stapleton joined.
2              SIGMA-TAU PHARMACEUTICALS v. SCHWETZ
                              COUNSEL

ARGUED: Mark D. Gately, HOGAN & HARTSON, L.L.P., Balti-
more, Maryland, for Appellant. Barbara Jeanne Stradling, Office of
Consumer Litigation, UNITED STATES DEPARTMENT OF JUS-
TICE, Washington, D.C., for Federal Appellees; David G. Adams,
VENABLE, BAETJER, HOWARD & CIVILETTI, L.L.P., Washing-
ton, D.C., for Appellee Gensia Sicor. ON BRIEF: Steven F. Barley,
HOGAN & HARTSON, L.L.P., Baltimore, Maryland; Catherine E.
Stetson, HOGAN & HARTSON, L.L.P., Washington, D.C., for
Appellant. Robert D. McCallum, Jr., Assistant Attorney General,
Larry D. Adams, Assistant United States Attorney, Office of Con-
sumer Litigation, UNITED STATES DEPARTMENT OF JUSTICE,
Washington, D.C.; Daniel E. Troy, Chief Counsel, Carl I. Turner,
Associate Chief Counsel, UNITED STATES FOOD AND DRUG
ADMINISTRATION, Washington, D.C., for Federal Appellees.


                              OPINION

WILKINSON, Chief Judge:

   Sigma-Tau Pharmaceuticals, Inc. claims that the Food and Drug
Administration acted contrary to law in approving generic versions of
its levocarnitine drug because the generics infringed on the seven-
year period of orphan exclusivity that its drug currently enjoys under
the Orphan Drug Act, 21 U.S.C. §§ 360aa-ee. The district court dis-
agreed, concluding that the FDA did not act unlawfully in approving
the generics for an indication that was no longer protected by market
exclusivity under the Act. Because the district court correctly inter-
preted the governing statute’s clear language, we affirm.

                                   I.

   Sigma-Tau Pharmaceuticals developed a drug to treat a rare condi-
tion known as carnitine deficiency in people with inborn metabolic dis-
orders.1 The FDA designated Sigma-Tau’s levocarnitine drug an
    1
   Carnitine is a naturally occurring amino acid derivative produced by
the liver and kidneys and found in red meat and dairy products. It trans-
               SIGMA-TAU PHARMACEUTICALS v. SCHWETZ                     3
"orphan drug" — one designed to treat a rare disease or condition —
and approved Sigma-Tau’s application to market it. Under the Orphan
Drug Act ("ODA"), 21 U.S.C. §§ 360aa-ee, Sigma-Tau was entitled
to seven years of market exclusivity to sell its drug, known as Carni-
tor, for that orphan indication. Its exclusivity for inborn metabolic dis-
orders expired in 1999.

   Sigma-Tau later received FDA approval for use of its levocarnitine
drug for the prevention and treatment of a second rare condition —
carnitine deficiency in patients with end-stage renal disease ("ESRD")
who are undergoing dialysis. Sigma-Tau’s exclusivity for treating car-
nitine deficiency in ESRD patients expires in 2006.

   The FDA recently approved the applications of two drug manufac-
turers, private intervenor Gensia Sicor Pharmaceuticals, Inc. and Bed-
ford Laboratories, to market and sell generic forms of Sigma-Tau’s
levocarnitine drug. The agency approved the generics for the treat-
ment of patients with inborn metabolic disorders, the unprotected
indication. The generics compete with Carnitor.

   As a result of these generic drug approvals, Sigma-Tau brought suit
against the FDA on May 10, 2001. Sigma-Tau sought to have the
approvals rescinded, or, in the alternative, to have the FDA change
the generics’ labeling to protect Sigma-Tau’s orphan exclusivity.
Sigma-Tau claimed that the FDA had violated the ODA Amendments
to the Federal Food, Drug, and Cosmetic Act ("FDCA"), 21 U.S.C.
§§ 360aa-ee, the ODA’s implementing regulations, and the Adminis-
trative Procedure Act ("APA"), 5 U.S.C. § 706(2)(A). In particular,
Sigma-Tau alleged that the FDA ignored substantial evidence that the
generics were intended for use in an orphan-protected market, and
that the agency’s approvals were arbitrary and capricious because the
generics infringed on the seven-year period of orphan exclusivity that
Carnitor currently enjoys under the ODA.

ports long-chain fatty acids into the mitochondria, where they are oxi-
dized and release energy. It also functions as a waste remover. Carnitine
deficiency can manifest itself in many ways, including the failure to
thrive in infants, cardiomyopathy, recurrent infections, muscle weakness,
and liver dysfunction.
4               SIGMA-TAU PHARMACEUTICALS v. SCHWETZ
   After two hearings, the district court ruled against Sigma-Tau. In
so ruling, the district court applied the well-settled principles of Chev-
ron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 

(1984). Under the first step of the Chevron analysis, 

the court concluded that Congress had spoken directly to the issue,
and that the FDA’s approvals of the generic manufacturers’ products
were consistent with the clear language of the governing statute,
§ 360cc(a) of the ODA. Noting the statute’s directive that the FDA
"may not approve another application . . . for such drug for such dis-
ease or condition . . . until the expiration of seven years," 

reasoned that the FDA had not approved another drug application "for
such disease or condition," but rather had done so for a disease or
condition no longer subject to exclusivity.

   Alternatively, the court held that even if the statute was not clear,
the FDA’s permissible construction of it was entitled to deference
under the second step of the Chevron inquiry. 

.
Further, the court concluded that the agency was entitled to substan-
tial deference in interpreting its own regulations, especially on a com-
plex and highly technical issue.

   The court thus held that the FDA’s approvals were "not arbitrary
or capricious, an abuse of discretion, or otherwise a violation of law."
It accordingly entered judgment in favor of the agency. Sigma-Tau
appeals.

                                    II.

   In dispute here are provisions of the FDCA that govern orphan
drugs. See 21 U.S.C. §§ 360aa-360ee. These sections were added to
the FDCA by the Orphan Drug Act of 1983 ("ODA"), Pub. L. No. 97-
414, 96 Stat. 2049. The ODA was enacted in order to provide drug
manufacturers with incentives to develop "orphan" drugs — that is,
drugs for the treatment of rare diseases or disorders that affect only
small patient populations. See Genentech, Inc. v. Bowen, 

, 302-303 (D.D.C. 1987). In pursuit of this objective, Congress
offered research assistance, grants, and tax incentives to companies
that undertake development of orphan drugs. 

In addition,
Congress provided for seven years of market exclusivity for approved
orphan drugs. 21 U.S.C. § 360cc(a). As noted above, this provision of
                SIGMA-TAU PHARMACEUTICALS v. SCHWETZ                      5
the ODA states that the FDA "may not approve another application
. . . for such drug for such disease or condition . . . until the expiration
of seven years." 

the FDA’s approvals of generic versions of
Carnitor. Sigma-Tau submits that the generics were in fact intended
for use in patients with ESRD who are undergoing dialysis, and that
they thereby infringed on the seven-year period of orphan exclusivity
that Carnitor currently enjoys under the ODA.

                                    III.

                                    A.

   Reviewing the district court’s grant of summary judgment de novo,
see Higgins v. E.I. DuPont de Nemours & Co., 

, 1167
(4th Cir. 1988), we agree that the plain language of the ODA is unam-
biguous, and that the FDA’s approvals of the generics in this case
comported with the clear wording of the statute. It is apparent that the
FDA did not "approve another application . . . for such drug for such
disease or condition" here, § 360cc(a), but rather approved "another
application . . . for such drug" for a different disease or condition, one
that was no longer subject to exclusivity. That is, the agency approved
generic versions of Sigma-Tau’s levocarnitine drug for people with
inborn metabolic disorders, for which the period of orphan exclusivity
had expired. The FDA did not approve the generics for the treatment
of ESRD patients.

   By using the words "such drug for such disease or condition," Con-
gress made clear its intention that § 360cc(a) was to be disease-
specific, not drug-specific. In other words, the statute as written pro-
tects uses, not drugs for any and all uses. Congress could have written
§ 360cc(a) more broadly by prescribing that the FDA "may not
approve another application . . . for such drug," but it chose not to
draft the statute in that way. Because Congress has spoken directly to
the dispositive question before us, our inquiry is at an end. 

; see also Hillman v. IRS, 

, 342 (4th
Cir. 2001) (citing Caminetti v. United States, 

, 485
(1917)).
6              SIGMA-TAU PHARMACEUTICALS v. SCHWETZ
   Understanding the implications of this analysis under the first step
of Chevron, Sigma-Tau argues that the plain language of § 360cc(a)
is ambiguous. Specifically, Sigma-Tau submits that the phrase "such
disease or condition" may refer either to the disease or condition for
which the drug is labeled, or to the disease or condition for which it
is intended to be used.

   But Sigma-Tau cannot create a genuine ambiguity in § 360cc(a)
under Chevron by raising the evidentiary question of labeled use ver-
sus intended use. Section 360cc(a) simply provides that the FDA
"may not approve" generics for a protected indication. Thus, the stat-
ute is clearly directed at FDA approved-use, not generic competitor
intended-use. And in view of this textual emphasis on approved-use,
the evidentiary basis for the agency’s approvals must be the use for
which the approvals are sought — that is, the use for which the gener-
ics are labeled. Thus, the FDA does not violate § 360cc(a) by relying
upon the generic manufacturers’ proposed labeling as opposed to the
alleged evidence of intended use discussed below. This statute is not
ambiguous.

                                  B.

   Sigma-Tau nevertheless urges us to look beyond the face of the
ODA to the FDA’s regulations. In particular, Sigma-Tau contends
that if the agency had properly applied its intended-use regulation, 21
C.F.R. § 201.128 (2001), it would have concluded that the generics
at issue were intended for treatment of ESRD patients. Sigma-Tau
asserts that the FDA should have considered "compelling, readily
available, objective evidence of the generics’ intended use," such as
market data for Carnitor, dosage forms, and federal drug reimburse-
ment policies, which would have resulted in the generics not being
approved based on 21 C.F.R. § 316.3(b)(13) (2001).2
    2
   Specifically, Sigma-Tau claims: (1) data show that 80% of the market
for Carnitor is for treating ESRD; (2) Gensia Sicor and Bedford Labs
sought approval only for the injectable form of the drug, which is the
only form approved for treating ESRD; and (3) the Centers for Medicare
and Medicaid Services, the federal agency reimbursing 93% of ESRD
treatments in the United States, does not distinguish between orphan and
generic drugs in making payments.
               SIGMA-TAU PHARMACEUTICALS v. SCHWETZ                    7
   To reiterate, the statute has foreclosed this line of argument. And
even if we were to consider the regulations, application of § 201.128
does not entitle Sigma-Tau to relief. Sigma-Tau does not attack the
validity of the FDA regulations themselves. Because Sigma-Tau
raises no challenge to the validity of the regulations themselves, we
have no occasion to consider the deference due the FDA’s interpreta-
tion of the ODA under United States v. Mead Corp., 

(2001). Mead accords Chevron deference to agency interpretations
that take the form of rules promulgated pursuant to an express grant
of "authority to the agency generally to make rules carrying the force
of law." 

   Rather, Sigma-Tau challenges the agency’s application of the regu-
lations to the facts of this case. We owe "substantial deference" when
reviewing an agency’s interpretation of its own regulations. Thomas
Jefferson Univ. v. Shalala, 

, 512 (1994); see also Zeneca,
Inc. v. Shalala, 

, 168 (4th Cir. 2000); Friends of Iwo
Jima v. Nat’l Capital Planning Comm’n, 

, 775 (4th Cir.
1999). We must defer to the FDA’s interpretation of its regulations
unless it is "plainly erroneous or inconsistent with the regulation."
Auer v. Robbins, 

, 461 (1997) (internal quotations omit-
ted); see also Thomas Jefferson 

; 

. Indeed, "[o]ur review in such cases is more deferential
than that afforded under Chevron." Wyo. Outdoor Council v. U.S.
Forest Serv., 

, 52 (D.C. Cir. 1999) (internal quotation
omitted). The "broad deference" due the agency "is all the more war-
ranted when, as here, the regulation concerns ‘a complex and highly
technical regulatory program,’ in which the identification and classifi-
cation of relevant ‘criteria necessarily require significant expertise
and entail the exercise of judgment grounded in policy concerns.’"
Thomas Jefferson 

(quoting Pauley v.
BethEnergy Mines, Inc., 

, 697 (1991)).

   The FDA’s application of its regulations to the facts before it was
not erroneous. On the contrary, the agency’s approvals of the generics
were reasonable in view of the language of the regulations imple-
menting the orphan drug provisions of the FDCA. They state that
"[o]rphan-drug exclusive approval . . . means that . . . no approval will
be given to a subsequent sponsor of the same drug product for the
same indication for 7 years . . . ." 21 C.F.R. § 316.3(b)(12) (2001).
8              SIGMA-TAU PHARMACEUTICALS v. SCHWETZ
"Same drug" is defined as "a drug that . . . is intended for the same
use as the previously approved drug." 21 C.F.R. § 316.3(b)(13)(i).
And "intended use" is defined as "the objective intent of the persons
legally responsible for the labeling of drugs." 21 C.F.R. § 201.128.
Section 201.128 further provides that "intent is determined by such
persons’ expressions or may be shown by the circumstances sur-
rounding the distribution of the article."

   The FDA determined the intended use for Gensia Sicor’s and Bed-
ford Labs’ generic drugs by relying primarily upon the proposed
labeling provided by the companies. In so doing, the FDA did not
contravene § 201.128. The manufacturers’ labels certainly constitute
"such persons’ expressions" within the meaning of that section.
Indeed, § 201.128 specifically mentions "labeling claims" and "writ-
ten statements" by manufacturers. As we have previously affirmed,
"no court has ever found that a product is ‘intended for use’ or
‘intended to affect’ within the meaning of the [FDCA] absent manu-
facturer claims as to that product’s use." Brown & Williamson
Tobacco Corp. v. FDA, 

, 163 (4th Cir. 1998) (internal
quotation omitted), aff’d, 

(2000).

    Sigma-Tau contends that the FDA was obligated to look beyond
the labeling to what Sigma-Tau maintains is the reality of the situa-
tion, which is that most of the need for the generics — and thus most
of the money to be made — lies in treating patients with ESRD. But
this point is unavailing. Section 201.128 provides that "intent is deter-
mined by such persons’ expressions or may be shown by the circum-
stances surrounding the distribution of the article" (emphasis added).
The regulation is phrased in the disjunctive, not the conjunctive. And
it states that intent "may be shown" by the surrounding circumstances,
not that it must be so shown. The district court correctly found that
§ 201.128 grants the agency discretion to decide what evidence of
intent it will examine.

   The regulation does so for good reason. The FDA necessarily
approves the generics before their manufacturers engage in any actual
marketing. This is obvious enough, but the potential consequences of
following Sigma-Tau’s approach in view of this fact may not be. If
we were to ignore the deference due the FDA and impose exacting
evidentiary standards upon its generic drug approval process, the
               SIGMA-TAU PHARMACEUTICALS v. SCHWETZ                     9
agency would be faced with formidable problems. This is because
many of the sources of evidence and market data to which Sigma-Tau
points cannot be effectively analyzed in the pre-approval context. As
the FDA stresses, that is why § 201.128 provides the agency with
flexibility, and why it states that "[t]he intended uses of an article may
change after it has been introduced into interstate commerce by its
manufacturer." Thus, the intended-use inquiry Sigma-Tau urges upon
us might evolve into a foreseeable-use test. Then, once the FDA
approved an orphan drug for a protected indication, generic competi-
tors might be prohibited from entering the market for almost any use.

   As the district court noted, not only might this course of events
result in extensions of exclusivity periods that Congress never
intended, but it also might frustrate the longstanding practice of Con-
gress, the FDA, and the courts not to interfere with physicians’ judg-
ments and their prescription of drugs for off-label uses. See, e.g.,
Bristol-Myers Squibb Co. v. Shalala, 

, 1496 (D.C. Cir.
1996); Rhone-Poulenc Rorer Pharm., Inc. v. Marion Merrell Dow,
Inc., 

, 514 n.3 (8th Cir. 1996). In light of the ensuing
effects on the delivery of health care and drug prices in this country,
such interference with off-label use is not something we would be
wise to welcome, let alone help to bring about. Even Sigma-Tau
appears to agree that the medical community’s foreseeable off-label
use of drugs does not violate the ODA.

   In arguing that the FDA cannot accept the generic competitors’
representations but rather must draw inferences from market forces,
Sigma-Tau is in effect campaigning for a regulatory regime in which
relatively few generics are approved. Though it does not couch its
contentions in these terms, Sigma-Tau in essence wants foreseeable
off-label use to bar the approval of generic drugs, even for unpro-
tected indications. But the consequences of adding a huge evidentiary
hurdle to the generic drug approval process would be profoundly anti-
competitive. And that is not all. Sigma-Tau’s approach also implicitly
frowns upon the practice of off-label use itself. But the Supreme
Court has not indicated that off-label use is illegitimate. On the con-
trary, it recently stated that "‘off-label’ usage of medical devices . . .
is an accepted and necessary corollary of the FDA’s mission to regu-
late in this area without directly interfering with the practice of medi-
10             SIGMA-TAU PHARMACEUTICALS v. SCHWETZ
cine." Buckman Co. v. Plaintiffs’ Legal Comm., 

, 350
(2001).

   In addition, the FDA persuasively argues that it must balance the
ODA’s incentive structure for the development of orphan drugs
against the goals of the Drug Price Competition and Patent Term Res-
toration Act of 1984, commonly known as the Hatch-Waxman
Amendments to the FDCA. Pub. L. No. 98-417, 98 Stat. 1585. This
statute seeks "to make available more low cost generic drugs" by
establishing an abbreviated generic drug approval procedure. H.R.
Rep. No. 98-857(I), at 14 (1984), reprinted in 1984 U.S.C.C.A.N.
2647, 2647. Rather than balancing the ODA and the Hatch-Waxman
Amendments, Sigma-Tau quite unapologetically puts all weight on
the orphan drug development end of the scale, as if no tension exists
between the two statutes that the FDA must negotiate.

   Thus, the FDA did not commit plain error or act inconsistently with
its regulations insofar as it declined to examine other evidence besides
the proposed labeling in approving the generic drugs at issue. See

. Accordingly, the agency’s approvals were not
"arbitrary, capricious, an abuse of discretion, or otherwise not in
accordance with law." 5 U.S.C. § 706(2)(A).3
  3
    Sigma-Tau contends that the FDA’s generic approvals also violated
the "same labeling" requirement of the FDCA, which mandates that the
labeling of a generic drug be exactly the same as that of the innovator,
except for certain specified differences. 21 U.S.C. § 355(j)(2)(A)(v). But
as the FDA correctly points out, one of the permissible discrepancies
concerns labeling "changes required . . . because the new [generic] drug
and the listed drug are produced or distributed by different manufactur-
ers." 

the generics in this case are not allowed to have the
same labeling as Carnitor while Carnitor is enjoying a second seven-year
period of exclusivity for the treatment of ESRD, Sigma-Tau’s argument
constitutes nothing more than another attempt to obtain market exclusiv-
ity for any and all uses of its drug, thereby preventing generic competi-
tors from entering the market for any indication. Indeed, the D.C. Circuit
rejected Sigma-Tau’s proposed interpretation of 21 U.S.C. § 355(j)(2)
(A)(v) for primarily this reason in the context of Hatch-Waxman pioneer-
drug exclusivity under 21 U.S.C. § 355(j). See 

.
               SIGMA-TAU PHARMACEUTICALS v. SCHWETZ                 11
                                 IV.

   The statute governing the outcome of this case is clear on its face.
And even if it were not, the FDA’s application of its regulations here
was not in error. If the underlying facts of this dispute are as Sigma-
Tau alleges — that the generic manufacturers said one thing to the
agency when they intended to do something else from the very start
— then the FDA may reconsider its approvals of their generics at a
later date. But the FDA is not obligated to assume bad faith on the
part of generic manufacturers at the beginning of the approval pro-
cess. And neither the controlling statute nor FDA regulations require
the agency to do more than it did in this case. For the foregoing rea-
sons, the judgment of the district court is

                                                         AFFIRMED.