United States Court of Appeals


                     FOR THE DISTRICT OF COLUMBIA CIRCUIT


  Argued March 13, 1998                             Decided April 14, 1998 


                                 No. 97-5082


                         Mova Pharmaceutical Corp., 

                                   Appellee


                                      v.


                              Donna E. Shalala, 

          Secretary, U.S. Department of Health & Human Services and

            Michael A. Friedman, Acting Commissioner, U.S. Food and 
                        Drug Administration, Appellees


                        Mylan Pharmaceuticals, Inc., 

                                  Appellant


                              Consolidated with

                                 No. 97-5111


                Appeals from the United States District Court 

                        for the District of Columbia 

                               (No. 96cv02861)

 


 




     Steven Lieberman argued the cause for appellant Mylan 
Pharmaceuticals, Inc., with whom E. Anthony Figg was on 
the briefs.

     Howard S. Scher, Attorney, United States Department of 
Justice, argued the cause for the federal appellants, with 
whom Frank W. Hunger, Assistant Attorney General, Mary 
Lou Leary, United States Attorney at the time the briefs 
were filed, and Douglas N. Letter, Litigation Counsel, were 
on the briefs.

     Steven J. Glassman argued the cause for appellant Phar- macia & Upjohn Company, with whom David O. Bickart was 
on the briefs.

     Ronald L. Grudziecki argued the cause for appellee, with 
whom James S. Rubin was on the brief.

     John F. Cooney filed the brief for amicus curiae Teva 
Pharmaceuticals, USA.

     Before:  Wald, Silberman and Tatel, Circuit Judges.

     Opinion for the Court filed by Circuit Judge Wald.

     Wald, Circuit Judge:  On December 19, 1996, the Food and 
Drug Administration ("FDA") approved an application by a 
drug manufacturer, Mylan Pharmaceuticals, Inc. ("Mylan") to 
market a generic version of micronized glyburide, a drug used 
to treat diabetes.  Another drug manufacturer, Mova Phar- maceutical Corp. ("Mova"), had filed an earlier application to 
market a generic version of the same drug;  however, Mova's 
application had not yet been approved, because of a patent 
infringement suit brought by Pharmacia & Upjohn Company 
("Upjohn"), in which Upjohn claimed that Mova's product 
infringed a patent belonging to Upjohn.

     When Mova learned that the FDA had approved Mylan's 
application, it brought suit in the United States District Court 
for the District of Columbia, relying on 21 U.S.C. 
s 355(j)(5)(B)(iv) (1994),1 to compel the FDA to delay the 

__________
     1 At the time that Mova brought its action, this section was 
designated 21 U.S.C. s 355(j)(4)(B)(iv).  On November 21, 1997, the 



effective date of this approval until 180 days after the earlier 
of the dates that Mova won its suit or began to market its 
product.  Because the statutory scheme governing the ap- proval of successive generic drug applications is quite com- plex, we will, for purposes of the introduction, describe the 
parties' contentions only in general terms.  Mova argued 
that, because it had filed a previous application to market a 
generic version of micronized glyburide, the applicable statu- tory provision, 21 U.S.C. s 355(j)(5)(B)(iv), granted a 180-day 
market exclusivity period to Mova running from the date 
Mova won its suit or began marketing its product, and the 
FDA was barred from approving Mylan's similar application 
until after the end of that 180-day period.  In response, the 
FDA cited a regulation that permitted the agency to approve 
Mylan's application immediately, because at the time Mylan 
submitted its application Mova had not yet "successfully 
defended" against (that is, prevailed in) Upjohn's patent 
infringement suit.  See 21 C.F.R. s 314.107(c)(1).  Mova in 
turn challenged the FDA's regulation as inconsistent with the 
plain language of s 355(j)(5)(B)(iv).  The district court agreed 
with Mova, and entered a preliminary injunction requiring 
the FDA to delay its approval of Mylan's application until 180 
days after Mova won its suit or began to market its product 
(whichever came first).  The FDA and Mylan have appealed 
this decision.

     While Mova's request for a preliminary injunction was 
pending, Upjohn submitted a motion to intervene in the 
litigation.  After granting the injunction, the district court 
denied Upjohn's motion to intervene, concluding that it was 
moot, and that in any case Upjohn did not have a legally 
protected interest in the subject matter of the litigation.  

__________ Food and Drug Administration Modernization Act of 1997 was 
enacted;  section 119(b)(1)(A) of that law inserted a new section 
355(j)(3), and redesignated former paragraphs 355(j)(3) to (8) as 
paragraphs (4) to (9).  See Pub. L. No. 105-115, 111 Stat. 2296 
(1997).  According to section 501 of that Act, the amendments "shall 
take effect 90 days after the date of enactment of this Act."  We 
will therefore use the section's new designation.



Upjohn has appealed this ruling and its appeal has been 
consolidated with that of the FDA and Mylan.

     On the merits of the preliminary injunction, we find that 
the district court was correct in finding that Mova was very 
likely to be able to show that the FDA's regulation exceeded 
its authority under the statute.  On Upjohn's motion to 
intervene, we find that the district court's reasons for denying 
the motion were erroneous, and that Upjohn is entitled to 
participate both in this appeal and in any further proceedings 
before the district court.

                                I. Background


     A.Statutory and Regulatory Framework

     We will first briefly outline the statutory and regulatory 
framework applicable to the marketing of generic drugs.  
Generic drugs are versions of brand-name prescription drugs 
that are often sold without a brand name and that contain the 
same active ingredients, but not necessarily the same inactive 
ingredients, as the original.  United States v. Generix Drug 
Corp., 460 U.S. 453, 455 (1983).  Ordinarily, an applicant to 
market a drug must complete a document called a New Drug 
Application, or NDA.  Preparing such applications can be a 
time-consuming and costly process, as they must include data 
from studies showing the drug's safety and effectiveness.  
Formerly, a firm that wished to make a generic version of a 
brand-name drug that had already been approved by the 
FDA was required to file a new NDA, complete with new 
studies showing the drug's safety and effectiveness.  See 
generally H.R. Rep. No. 98-857, Part I, at 16-17 (1984).

     In 1984, Congress enacted the Hatch-Waxman Amend- ments, which established a simplified procedure for FDA 
approval of generic drugs.  Under this procedure, the original 
applicant for FDA approval of a drug, called the "pioneer" 
applicant, must still complete a full NDA.  However, subse- quent applicants who wish to manufacture generic versions of 
the original have an alternative:  they may instead complete 
an Abbreviated New Drug Application, or ANDA, which 
relies on the FDA's previous determination that the drug is 



safe and effective, and thus avoid submitting new safety and 
effectiveness studies.

     The Hatch-Waxman Amendments specify the contents of 
an ANDA in detail.  One requirement is that, for each of the 
patents applicable to the pioneer drug, the ANDA applicant 
must certify whether the proposed generic drug would in- fringe that patent, and, if not, why not.  The statute provides 
ANDA applicants with four certification options:  they may 
certify (I) that the required patent information has not been 
filed;  (II) that the patent has expired;  (III) that the patent 
has not expired, but will expire on a particular date;  or (IV) 
that the patent is invalid or will not be infringed by the drug 
for which the ANDA applicant seeks approval.  21 U.S.C. 
s 355(j)(2)(A)(vii).  We will call these paragraph I, II, III, 
and IV certifications, respectively.

     If the applicant makes a certification under paragraphs I or 
II, the statute provides that the FDA may approve the 
ANDA effective immediately.  21 U.S.C. s 355(j)(5)(B)(i).  If 
the applicant makes a certification under paragraph III, the 
FDA may approve the ANDA effective on the date that the 
applicant certifies that the patent will expire.  21 U.S.C. 
s 355(j)(5)(B)(ii).

     When an applicant makes a certification under paragraph 
IV, things become more complicated.  In such cases, the 
statute begins by providing a forty-five-day window during 
which the patent-holder may bring suit against the applicant.  
If the patent-holder brings suit during that forty-five-day 
period, the statute says that the FDA's approval of the 
ANDA must be delayed for thirty months, a provision that is 
presumably intended to allow the patent-holder time to vindi- cate its patent in court before the generic competitor is 
allowed entry into the market.  21 U.S.C. s 355(j)(5)(B)(iii).  
The statute permits the court to lengthen or shorten this 
period if it finds that either party has failed to "reasonably 
cooperate in expediting the action."  Id.  If the court finds 
that the patent is invalid or is not infringed, the FDA's 
approval becomes effective as of the date of that ruling.  21 



U.S.C. s 355(j)(5)(B)(iii)(I), (III).2

     It is the succeeding provision of the statute, however, that 
has occasioned the dispute involved in this suit (and many 
others).  That provision says: __________
     2 The full text of section 355(j)(5)(B)(iii) reads:

     If the applicant made a certification described in subclause (IV) 
     of paragraph (2)(A)(vii), the approval shall be made effective 
     immediately unless an action is brought for infringement of a 
     patent which is the subject of the certification before the 
     expiration of forty-five days from the date the notice provided 
     under paragraph (2)(B)(i) is received.  If such an action is 
     brought before the expiration of such days, the approval shall 
     be made effective upon the expiration of the thirty-month 
     period beginning on the date of the receipt of the notice 
     provided under paragraph (2)(B)(i) or such shorter or longer 
     period as the court may order because either party to the 
     action failed to reasonably cooperate in expediting the action, 
     except that--

          (I) if before the expiration of such period the court decides 
     that such patent is invalid or not infringed, the approval shall 
     be made effective on the date of the court decision,

          (II) if before the expiration of such period the court decides 
     that such patent has been infringed, the approval shall be made 
     effective on such date as the court orders under section 
     271(e)(4)(A) of Title 35 or

          (III) if before the expiration of such period the court grants 
     a preliminary injunction prohibiting the applicant from engag-
     ing in the commercial manufacture or sale of the drug until the 
     court decides the issues of patent validity and infringement and 
     if the court decides that such patent is invalid or not infringed, 
     the approval shall be made effective on the date of such court 
     decision.

     In such an action, each of the parties shall reasonably cooper-
     ate in expediting the action.  Until the expiration of forty-five 
     days from the date the notice made under paragraph (2)(B)(i) is 
     received, no action may be brought under section 2201 of Title 
     28 for a declaratory judgment with respect to the patent.  Any 
     action brought under section 2201 shall be brought in the 
     judicial district where the defendant has its principal place of 
     business or a regular and established place of business.



     If the application contains a certification described in 
     subclause (IV) of paragraph (2)(A)(vii) and is for a drug 
     for which a previous application has been submitted 
     under this subsection continuing [sic] 3 such a certifica-
     tion, the application shall be made effective not earlier 
     than one hundred and eighty days after--

          (I) the date the Secretary receives notice from the 
     applicant under the previous application of the first 
     commercial marketing of the drug under the previous 
     application, or

          (II) the date of a decision of a court in an action 
     described in clause (iii) holding the patent which is the 
     subject of the certification to be invalid or not infringed, 
     whichever is earlier. 21 U.S.C. s 355(j)(5)(B)(iv).  This provision on its face ap- pears to provide an advantage to the first party who files a 
paragraph IV ANDA (henceforth, the "first applicant"), by 
granting him a 180-day period in which to market his generic 
drug without competition from other ANDA applicants.4  We 
will call this Edenic moment of freedom from the pressures of 
the marketplace the statute's "exclusivity period."  Section 
355(j)(5)(B)(iv) has two sub-clauses, each of which can trigger 
the start of the 180-day exclusivity period;  we will call sub- clause (I), which turns on the first commercial marketing of 
the drug, the "commercial-marketing trigger," and sub-clause 
(II), which turns on a court decision finding the patent to be 
invalid or not infringed, the "court-decision trigger."

     The FDA, however, concluded, for reasons discussed below, 
that Congress could not have intended for this provision to be 
read literally.  Thus, in drafting the regulations implementing 

__________
     3 This should probably read "containing."

     4 The statute actually says that the exclusivity period applies 
whenever there is a "previous" application;  thus, the statute might 
conceivably be read to confer this 180-day period on a second or 
third applicant in some situations.  For purposes of this litigation, 
however, the only previous application is Mova's, which is also the 
first.  We will therefore refer only to the first applicant.



section 355(j)(5)(B)(iv), the FDA added its own requirement 
that the first applicant must have "successfully defended 
against a suit for patent infringement" before the exclusivity 
period can begin to run.  (We will refer to this as the 
"successful defense" requirement.)  The relevant regulation 
states:

     If an abbreviated new drug application contains a certifi-
     cation that a relevant patent is invalid, unenforceable, or 
     will not be infringed and the application is for a generic 
     copy of the same listed drug for which one or more 
     substantially complete abbreviated new drug applications 
     were previously submitted containing a certification that 
     the same patent was invalid, unenforceable, or would not 
     be infringed and the applicant submitting the first appli-
     cation has successfully defended against a suit for pat-
     ent infringement brought within 45 days of the patent 
     owner's receipt of notice submitted under s 314.95, ap-
     proval of the subsequent abbreviated new drug applica-
     tion will be made effective no sooner than 180 days from 
     whichever of the following dates is earlier:

          (i) The date the applicant submitting the first applica-
          tion first commences commercial marketing of its drug 
          product;  or

          (ii) The date of a decision of the court holding the 
          relevant patent invalid, unenforceable, or not infringed. 21 C.F.R. s 314.107(c)(1) (emphasis added).

     B.The Factual Scenario of This Case 

     Glyburide is a drug used in treating diabetes, and micron- ized glyburide is one form of that drug.  Upjohn holds a 
patent for a particular formulation of micronized glyburide 
entitled "Spray-Dried Lactose Formulation of Micronized 
Glyburide."

     In December, 1994, Mova filed an ANDA for a generic 
form of micronized glyburide, which included a paragraph IV 
certification.  Mova gave the required notice to the patent- holder, Upjohn;  within 45 days of receiving this notice, 



Upjohn filed a patent infringement suit against Mova in the 
District of Puerto Rico.

     While this litigation was underway, in November, 1995, 
Mylan, too, filed an ANDA for a generic form of micronized 
glyburide.  Mylan's initial filing contained a paragraph III 
certification;  this meant that Mylan conceded patent infringe- ment, so that its ANDA could not receive FDA approval until 
Upjohn's patent expired.  In August 1996, however, Mylan 
amended its ANDA to contain a paragraph IV certification.

     As Mova had, Mylan gave the required notice to Upjohn;  
but this time Upjohn, for reasons unclear, failed to sue within 
the prescribed 45-day period.5  The thirty-month suspension 
of FDA approval provided for in section 355(j)(5)(B)(iii) only 
applies if the patent-holder sues the ANDA applicant within 
45 days;  thus, this waiting period did not apply to Mylan.  
And, because Mova had not yet "successfully defended" 
against Upjohn's patent infringement suit, 21 C.F.R. 
s 314.107(c)(1), which incorporates the FDA's interpretation 
of section 355(j)(5)(B)(iv), did not require the FDA to delay its 
approval of Mylan's ANDA.  Thus, the FDA approved My- lan's ANDA effective immediately, as of December 19, 1996.

     After learning of this approval, Mova, on December 26, 
1996, filed suit in the United States District Court for the 
District of Columbia, seeking a temporary restraining order 
compelling the FDA to postpone the effective date of Mylan's 
approval.  Although declining to issue a TRO, the district 
court, on January 23, 1997, granted a preliminary injunction 
requiring that the FDA render its approval of Mylan's ANDA 
effective no earlier than 180 days after the earlier of (1) 
Mova's first commercial marketing of its micronized glyburide 
product, or (2) the date of Mova's victory in the Puerto Rico 
litigation.  (This was precisely the relief Mova had sought.)  
In an accompanying memorandum, the district court ex- plained that the successful-defense requirement in the FDA's 
regulations was inconsistent with the plain language of sec- tion 355(j)(5)(B)(iv), and therefore unenforceable.  The FDA 

__________
     5 Upjohn did eventually sue Mylan, on February 17, 1997.  On 
March 31, 1998, the court ruled that Upjohn's patent was invalid 
and not infringed.



and Mylan (which had intervened in the proceedings) appeal- ed.

     One further matter remained to be resolved.  Before the 
preliminary injunction was issued, Upjohn had filed a motion 
seeking to intervene in the proceedings before the district 
court.  The district court had not adverted to this motion in 
granting the preliminary injunction.  A few days later, how- ever, on February 10, it denied Upjohn's motion to intervene, 
stating that the motion was moot (because the district court 
had already issued the preliminary injunction), and also that 
Upjohn did not have a "cognizable interest" in the litigation, 
and was therefore not entitled to intervene.  Upjohn has 
appealed this order, and argues that it should be allowed to 
participate in this appeal and in any further proceedings in 
the district court.

     While these appeals were pending, there have been subse- quent developments in Upjohn's patent infringement suit 
against Mova.  On December 2, 1997, a jury found that 
Upjohn's patent was invalid, unenforceable, and had not been 
infringed.  Mova received final approval from the FDA to 
market its product on December 22, and began to sell its 
product shortly afterwards.  By its terms, the district court's 
preliminary injunction will therefore expire 180 days after 
December 2, 1997, on May 31, 1998.

                                 II. Analysis


     A.The Preliminary Injunction

     To demonstrate entitlement to a preliminary injunction, a 
litigant must show "1) a substantial likelihood of success on 
the merits, 2) that it would suffer irreparable injury if the 
injunction is not granted, 3) that an injunction would not 
substantially injure other interested parties, and 4) that the 
public interest would be furthered by the injunction."  City- fed Financial Corp. v. Office of Thrift Supervision, 58 F.3d 
738, 746 (D.C. Cir. 1995).  The district court balances the 
litigant's showings in these four areas in deciding whether to 
grant an injunction.  Id. at 747.  "We review a district court's 
decision regarding a preliminary injunction for abuse of dis-



cretion, and any underlying legal conclusions de novo."  Id. at 
746.

     Balancing these factors, the district court found, as to the 
first, that Mova had a "very high" likelihood of success on the 
merits, because Mova would probably be able to show that 
the FDA's successful-defense requirement was contrary to 
the plain language of section 355(j)(5)(B)(iv) and therefore 
unenforceable.  As to the second, the district court found that 
"the earliest generic drug manufacturer in a specific market 
has a distinct advantage over later entrants," and that Mova, 
a small company, would find it extremely difficult to compete 
against the much larger Mylan if Mylan got its product to 
market first.  As to the third prong, the district court found 
that any harm to Mylan was small, because Mylan was so 
likely to lose on the merits.  And, as to the fourth, the district 
court found that the public was the principal other interested 
party, and that the public's interest in the "faithful application 
of the laws" outweighed its interest in immediate access to 
Mylan's generic product.

     The FDA and Mylan have not seriously contested the 
district court's findings as to the second, third and fourth 
factors.6  However, they both vigorously contest the district 
court's conclusion that the successful-defense requirement is 

__________
     6 The FDA notes that "the mere existence of competition is not 
irreparable harm, in the absence of substantiation of severe eco- nomic impact."  WMATA v. Holiday Tours, 559 F.2d 841, 843 n.3 
(D.C. Cir. 1977).  Here, however, the district court found that Mova 
would be harmed by the loss of its "officially sanctioned head start," 
and that Mova's small size put it at a particular disadvantage.  This 
suffices to show a severe economic impact to Mova.

     Both the FDA and Mylan also contend that the district court 
should have declined to issue a preliminary injunction in order to 
further the public's interest in the rapid movement of generic drugs 
into the marketplace.  Supposing that they are right in their 
assessment of the public's interest, however, this factor alone 
cannot support denying an injunction.  Our polity would be very 
different indeed if the courts could decline to enforce clear laws 
merely because they thought them contrary to the public interest;  
we decline to embark upon that path.



inconsistent with section 355(j)(5)(B)(iv).  We will focus on 
the FDA's arguments, because it is on an agency's own 
justifications that the validity of its regulations must stand or 
fall.  See SEC v. Chenery Corp., 318 U.S. 80 (1943).

     The FDA concedes that the text of section 355(j)(5)(B)(iv) 
makes no provision for a successful-defense requirement.  It 
asserts, however, that a literal reading of that statutory 
provision would produce consequences of a kind Congress 
could not have intended when it wrote the law, and that its 
interpolation of a successful-defense requirement is an appro- priate way of implementing Congress's underlying intent.

     The FDA points to two principal situations in which a 
literal reading of the statute would produce bizarre results:  
(1) cases in which the first applicant is never sued, and (2) 
cases in which the first applicant loses its suit.  If the first 
applicant is never sued, the FDA claims, then the court- decision trigger will never be satisfied.  Later ANDA appli- cants will be unable to market their products until the first 
applicant decides to put its product on the market, thereby 
satisfying the commercial-marketing trigger.  But the first 
applicant could in theory wait indefinitely to begin selling its 
product, and thereby block all sales by later applicants.  This 
unfortunate scenario could happen, for instance, if the first 
applicant colludes with the pioneer drug company to eliminate 
generic competition, or if the first applicant is simply unable 
to obtain FDA approval of its production facilities and so 
cannot put its product on the market.

     If the first applicant loses its infringement suit, the delay 
problem could be even more serious.  The first applicant 
would then be able to satisfy neither the court-decision trig- ger nor the commercial-marketing trigger (because, having 
lost a patent-infringement suit, it would be unable to sell its 
product).  Thus, the FDA claims, no generic drugs could 
enter the market until after the pioneer company's patent 
expired.

     The successful-defense requirement, according to the FDA, 
is calculated to eliminate both occurrences.  An applicant that 
is never sued or that loses its suit will not have "successfully 



defended against a suit for patent infringement," 21 C.F.R. 
s 314.107(c)(1), and so the exclusivity period will not apply.  
Such applicants will therefore not interfere with the orderly 
movement by successive applicants of generic drugs into the 
marketplace.

     1. Applicable Principles of Judicial Review

     In assessing the validity of an agency's interpretation of a 
statute, we begin by asking whether "Congress has directly 
spoken to the precise question at issue";  if so, "the court, as 
well as the agency, must give effect to the unambiguously 
expressed intent of Congress."  Chevron U.S.A. Inc. v. Natu- ral Resources Defense Council, Inc., 467 U.S. 837, 842-43 
(1984).  If we find that "the statute is silent or ambiguous 
with respect to the specific issue, the question for the court is 
whether the agency's answer is based on a permissible con- struction of the statute."  Id. at 843.

     In the district court's judgment, the FDA's pragmatic 
reading of the statute could not survive the first prong of 
Chevron.  Concluding that the language of the statute "may 
be complex, and even cumbersome, but it is plain and unam- biguous," and that the statute "does not include a 'successful 
defense' requirement, and indeed it does not even require the 
institution of patent litigation," the district court found that, 
under the plain language of the statute, if a paragraph IV 
ANDA has been filed by a prior applicant, the FDA must 
delay approval of all subsequent ANDAs until either the 
court-decision trigger or the commercial-marketing trigger is 
satisfied.

     We think that the district court achieved the right result, 
but we are not quite as sanguine as the district court that, in 
applying the first prong of Chevron, it suffices to look only at 
the plain language of the statute.  "[I]n expounding a statute, 
we must not be guided by a single sentence or member of a 
sentence, but look to the provisions of the whole law, and to 
its object and policy."  Pilot Life Insurance Co. v. Dedeaux, 
481 U.S. 41, 51 (1987);  see also McCarthy v. Bronson, 500 
U.S. 136, 139 (1991).  Here, the FDA cannot point to any 
particular ambiguity in the words of section 355(j)(5)(B)(iv) 



that permits it to interpolate its "successful defense" require- ment.  Instead, the FDA's argument is that a literal reading 
of the statute would thwart Congress's central goal, in enact- ing the Hatch-Waxman Amendments, to bring generic drugs 
onto the market as rapidly as possible.

     In effect, the FDA seeks to invoke the long-standing rule 
that a statute should not be construed to produce an absurd 
result.  "It is a familiar rule, that a thing may be within the 
letter of the statute and yet not within the statute, because 
not within its spirit nor within the intention of its makers....  
If a literal construction of the words of a statute be absurd, 
the act must be so construed as to avoid the absurdity."  
Holy Trinity Church v. United States, 143 U.S. 457, 459-60 
(1892);  see also United States v. X-Citement Video, Inc., 513 
U.S. 64, 68-69 (1994) (rejecting the "most natural grammati- cal reading" of a statute to avoid "absurd" results);  Green v. 
Bock Laundry Machine Company, 490 U.S. 504, 527, 527-29 
(1989) (Scalia, J., concurring);  In re Nofziger, 925 F.2d 428, 
434 (D.C. Cir. 1991);  Veronica M. Dougherty, Absurdity and 
the Limits of Literalism, 44 Am. U.L. Rev. 127 (1994).  Over a 
hundred years ago, the Court explained this rule thus:

     The common sense of man approves the judgment men-
     tioned by Puffendorf, that the Bolognian law which en-
     acted "that whoever drew blood on the streets should be 
     punished with the utmost severity," did not extend to the 
     surgeon who opened the vein of a person who fell down 
     on the street in a fit.  The same common sense accepts 
     the ruling, cited by Plowden, that the statute of 1st 
     Edward II, which enacts that a prisoner who breaks 
     prison shall be guilty of a felony, does not extend to a 
     prisoner who breaks out when the prison is on fire--"for 
     he is not to be hanged because he would not stay to be 
     burnt." United States v. Kirby, 74 U.S. (7 Wall.) 482, 487 (1868) 
(citations omitted).

     In deciding whether a result is absurd, we consider not 
only whether that result is contrary to common sense, but 
also whether it is inconsistent with the clear intentions of the 



statute's drafters--that is, whether the result is absurd when 
considered in the particular statutory context.  If " 'the literal 
application of a statute will produce a result demonstrably at 
odds with the intentions of its drafters,' ... the intention of 
the drafters, rather than the strict language, controls."  Unit- ed States v. Ron Pair Enterprises, 489 U.S. 235, 242 (1989) 
(quoting Griffin v. Oceanic Contractors, Inc., 458 U.S. 564, 
571 (1982));  see also Environmental Defense Fund, Inc. v. 
EPA, 82 F.3d 451, 468 (D.C. Cir. 1996) (applying Ron Pair ).

     The rule that statutes are to be read to avoid absurd 
results allows an agency to establish that seemingly clear 
statutory language does not reflect the "unambiguously ex- pressed intent of Congress," Chevron, 467 U.S. at 842, and 
thus to overcome the first step of the Chevron analysis.  But 
the agency does not thereby obtain a license to rewrite the 
statute.  When the agency concludes that a literal reading of 
a statute would thwart the purposes of Congress, it may 
deviate no further from the statute than is needed to protect 
congressional intent.  Of course, the agency might be able to 
show that there are multiple ways of avoiding a statutory 
anomaly, all equally consistent with the intentions of the 
statute's drafters (and equally inconsistent with the statute's 
text).  In such a case, we would move to the second stage of 
the Chevron analysis, and ask whether the agency's choice 
between these options was "based on a permissible construc- tion of the statute."  Id. at 843.  Otherwise, however, our 
review of the agency's deviation from the statutory text will 
occur under the first step of the Chevron analysis, in which 
we do not defer to the agency's interpretation of the statute.

     Here, we think that the FDA's interpretation cannot sur- vive analysis under the first step of Chevron.  The FDA's 
"successful defense" requirement achieves the FDA's stated 
goal of preventing first applicants who are not sued or who 
lose their suits from benefiting from the exclusivity period.  
But it also does more.  The FDA had two routes by which it 
could have addressed the problem of first applicants who lose 
their suits, which we will call the "wait-and-see" approach and 
the "win-first" approach.  Under the wait-and-see option, 
later applicants would need to wait to see whether the first 



applicant won or lost its patent infringement suit.  If the first 
applicant lost, the exclusivity period would not apply;  if he 
won, it would.7  The FDA chose the other option, the win- first approach.  Under this approach, later applicants do not 
need to wait to see whether the first applicant wins or loses.  
Instead, while the first applicant's litigation is underway, the 
FDA may approve the applications of later ANDA applicants 
effective immediately, if they are otherwise eligible for ap- proval.8  Thus, the first applicant must win its lawsuit before 
section 355(j)(5)(B)(iv) has any effect on subsequent appli- cants at all.

     It is the FDA's decision to adopt the win-first approach 
that led to the present litigation.  If the FDA had instead 
chosen the wait-and-see approach, the FDA could not have 
approved Mylan's application when it did;  instead, Mylan 
would have needed to wait for the end of Mova's patent 
infringement suit.  We will therefore focus on this aspect of 
the successful-defense requirement.

     We conclude that the FDA's successful-defense require- ment is inconsistent with the unambiguously expressed intent 
of Congress.  The rule is gravely inconsistent with the text 
and structure of the statute.  Nor can the FDA show that the 
successful-defense requirement is needed to avoid "a result 
demonstrably at odds with the intentions of [section 
355(j)(5)(B)(iv)'s] drafters."  Ron Pair Enterprises, 489 U.S. 
at 242.  The FDA could have adopted a more narrow solution 
to the problem of first applicants who are never sued or who 
lose their suits.  It instead adopted the broad win-first rule, 
which it cannot show is needed to implement congressional 
intent.  In effect, the FDA has embarked upon an adventur- ous transplant operation in response to blemishes in the 

__________
     7 We do not mean to foreclose a third possibility, which is that 
some lawsuit other than that against the first applicant might 
satisfy the court-decision trigger before the first applicant's suit is 
over.  We discuss this possibility below.

     8 That is, if they have not been sued by the patent-holder, and are 
therefore not subject to the 30-month waiting period, or if the 30- month period has expired.



statute that could have been alleviated with more modest 
corrective surgery.

     2. The Statute's Text and Structure

     Section 355(j)(5)(B)(iv) is far from a model of legislative 
draftsmanship.  The district court in this case called the 
provision "cumbersome";  another district court described it 
as "very confusing and ambiguous."  Mylan Pharmaceuti- cals, Inc. v. Sullivan, No. 89-36-C(K), slip op. at 6 (N.D.W.V. 
May 5, 1989).  But, to the extent that the statute is clear 
about anything, it clearly forecloses the FDA's successful- defense requirement.9

     The successful-defense requirement is inconsistent with the 
literal language of the statute.  Section 355(j)(5)(B)(iv) says 
that, if an applicant has already filed a paragraph IV ANDA, 
later applications shall be approved "not earlier than one 
hundred and eighty days after" the commercial-marketing 
trigger or the court-decision trigger is satisfied.  The FDA's 
successful-defense requirement, by contrast, permits later 
applications to be approved even though neither trigger has 
been satisfied, simply because the first applicant's litigation 
has not yet come to a successful conclusion.

     The win-first rule also infringes on the statutory scheme in 
a second, subtler way:  its practical effect is to write the 
commercial-marketing trigger out of the statute.  The com- mercial-marketing trigger seems intended to ensure that, if a 
first ANDA applicant chooses to begin marketing its product 
before it has won its patent-infringement suit, the 180-day 
exclusivity period will begin to run immediately.  Under the 
FDA's regulation, however, the 180-day exclusivity period is 
only available to an applicant who has already "successfully 
defended against a suit for patent infringement."  Thus, if the 
first applicant begins marketing its product before it wins its 

__________
     9 We note that the Fourth Circuit recently came to the same 
conclusion in an unpublished opinion.  See Granutec, Inc. v. Shala- la, Nos. 97-1873, 97-1874, slip op. at 13-14 (4th Cir. Apr. 3, 1998).  
Because the rules of the Fourth Circuit disfavor (but do not 
prohibit) citation of unpublished opinions, we will not discuss the 
reasoning of Granutec further.



infringement suit, the 180 days of exclusivity do not begin to 
run;  other applicants remain eligible for FDA approval to 
begin marketing their products, at least up to the date that 
the first applicant wins the infringement action.

     If the first applicant eventually wins its lawsuit, the exclu- sivity period is counted as though it had begun to run when 
the applicant started commercial marketing.  Thus, an appli- cant who begins commercial marketing 120 days before win- ning its lawsuit receives only 60 days of exclusivity;  an 
applicant who begins commercial marketing 180 days (or 
more) before winning its suit receives no exclusivity period at 
all.  The FDA thus construes the commercial-marketing trig- ger to potentially hurt, but never benefit, the first ANDA 
applicant.

     There is no indication in the text or history of section 
355(j)(5)(B)(iv) that the commercial-marketing trigger is sup- posed to function in that one-sided manner.  The FDA itself 
provided a more plausible explanation of how it should work 
in an initial notice of proposed rulemaking for the ANDA 
regulations.  As the FDA then explained the statutory 
scheme,

     Congress's decision to begin the 180-day period under 
     section 505(j)(4)(B)(iv)(I) of the act from "the first com-
     mercial marketing of the drug," rather than from the 
     effective date of the ANDA, serves a rational policy only 
     if Congress contemplated a situation in which an approv-
     al of an ANDA is in effect but the applicant's decision not 
     to market the drug deserves to be protected because a 
     delay in marketing serves the public interest.

          Such a situation occurs where, under the terms of 
     section 505(j)(4)(B)(iii) of the act, an ANDA goes into 
     effect 30 months after a lawsuit is filed, but the lawsuit is 
     still pending.  It serves the public interest to permit a 
     prudent ANDA holder in that situation to stay off the 
     market until the litigation is resolved, thereby minimiz-
     ing potential damages.

          As drafted, sections 505(j)(4)(B)(iv)(I) and (II) of the 
     act carefully avoid providing an incentive for immediate 
     marketing;  the 180-day reward of exclusive marketing 
     begins when the applicant wins the lawsuit or when the 



     applicant actually begins marketing, "whichever is earli-
     er."  The applicant thus does not lose any of the 180-day 
     period by electing to stay off the market until the lawsuit 
     is over. Abbreviated New Drug Application Regulations, 54 Fed. Reg. 
28,872, 28,894 (1989).  In other words, the 180-day exclusivity 
period should begin to run as soon as the first applicant 
begins commercial marketing.  In adopting the successful- defense requirement in its final rulemaking, the FDA neither 
rejected the foregoing analysis, nor explained how the suc- cessful-defense requirement would be consistent with it.  See 
Abbreviated New Drug Application Regulations, 59 Fed. Reg. 
50,338, 50,353 (1994).10

     3. Does a Literal Reading Produce Absurd Results? 

     The FDA contends that the statute should not be read 
literally, because such a reading would produce results that 
are clearly inconsistent with the intent of Congress in enact- ing the statute.  We do not think that it is sufficiently clear 
that Congress intended the "win-first" reading of the statute 
to justify disregarding the most natural reading of the statu- tory text.

     We begin by setting aside the problems of the first appli- cant who is never sued or who loses his lawsuit.  The FDA 
may or may not be right that a literal reading of the statute 
to permit a first applicant to receive an exclusivity period in 
these situations would be inconsistent with the statutory 
scheme;  we do not decide this question.  As we have already 
pointed out, the fatal flaw in the FDA's "absurd results" 
argument is that the agency could have addressed these two 
(supposedly) problematic situations without imposing the 
broad win-first rule by creating narrower exceptions to sec- __________
     10 The FDA said only that "[o]ne comment said the rule, as 
drafted, created an incentive for frivolous claims of patent invalidity 
or noninfringement because it would give ANDA applicants exclu- sivity even if the applicant was unsuccessful in defending against 
the patent owner's lawsuit.  The comment would replace the phrase 
'to be sued within 45 days' with 'and to successfully defend a suit 
brought within 45 days.'  The FDA agrees and has amended 
s 314.107(c) accordingly."  Abbreviated New Drug Application 
Regulations, 59 Fed. Reg. 50,338, 50,353 (1994).



tion 355(j)(5)(B)(iv) for the "no suit" and "lost suit" cases.  
Indeed, the FDA has come close to doing so already.  As for 
first applicants who are never sued, the initial draft of 21 
C.F.R. s 314.107(c)(1) contained, instead of the successful- defense requirement, only a requirement that the first appli- cant have been "sued for patent infringement within 45 days 
of the patent owner's receipt of notice" in order to be eligible 
for the statutory period.  Abbreviated New Drug Application 
Regulations, 54 Fed. Reg. 28,872, 28,929 (1989).  This lan- guage would have corrected the problem of first applicants 
who are never sued (and only that problem).11

     As to first applicants who lose their suits, Mova observed at 
oral argument that one of the FDA's current regulations 
suggests a possible way of addressing this problem (and 
indeed may already have solved it).  That regulation provides 
that, if an ANDA applicant who makes a certification under 
paragraph IV later loses its patent-infringement suit, it must 
amend its ANDA to make a new certification under para- graph III, and provides that the ANDA will then "no longer 
be considered to be one containing a certification under 
paragraph [IV]." 12  The FDA claims that the regulation does 

__________
     11 Even this may not be the narrowest way of resolving the 
underlying problem.  After all, Congress may have intended to 
reward the first ANDA applicant for his enterprise whether or not 
he is later sued;  the statutory scheme only runs into problems if 
the first applicant never starts selling his product.  An alternative 
might be to prescribe a period within which a first applicant who 
has not been sued must bring his product to market in order to 
benefit from the exclusivity period.

     12 The relevant regulation provides:

     An applicant who has submitted a certification under para-
     graph (a)(12)(i)(A)(4) of this section and is sued for patent 
     infringement within 45 days of the receipt of notice sent under 
     s 314.95 shall amend the certification if a final judgment in the 
     action against the applicant is entered finding the patent to be 
     infringed.  In the amended certification, the applicant shall 
     certify under paragraph (a)(12)(i)(A)(3) of this section that the 
     patent will expire on a specific date.  Once an amendment or 
     letter for the change has been submitted, the application will no 



not have the effect of rendering the exclusivity period inappli- cable after such an amendment, and of course we owe defer- ence to the agency on the interpretation of its own regula- tions.13  But even if the present version of the regulation does 
not accomplish the desired end, the FDA could presumably 
draft a regulation that did so.

     The FDA did not choose to adopt these narrower ap- proaches;  instead, it adopted the "win-first" reading of the 
statute, which deviated from the literal language of the 
statute by allowing an application to be approved while the 
first applicant's lawsuit was pending and before either statu- tory trigger had been satisfied.  In analyzing the successful- defense requirement, then, we must ask whether the win-first 
reading is needed to avoid "a result demonstrably at odds 
with the intentions of [section 355(j)(5)(B)(iv)'s] drafters."  
Ron Pair Enterprises, 489 U.S. at 242.

     The FDA did not explain its decision to adopt the win-first 
approach (instead of a narrower approach) in issuing its 
regulation, see 59 Fed. Reg. 50,338, 50,353 (1994), and it has 
not presented any argument for that approach in this litiga- tion.  There is, however, a compelling argument for the win- first approach, which is advanced by Mylan.  What if the first 

__________
     longer be considered to be one containing a certification under 
     paragraph (a)(12)(i)(A)(4) of this section.  If a final judgment 
     finds the patent to be invalid and infringed, an amended 
     certification is not required. 21 C.F.R. s 314.94(a)(12)(viii)(A).

     13 The FDA said at oral argument that its regulation is intended 
only for "housekeeping" purposes, and that it should not be read to 
affect the application of section 355(j)(5)(B)(iv).  We owe "substan- tial deference" to an interpretation by the FDA of its own regula- tions, which has "controlling weight unless it is plainly erroneous or 
inconsistent with the regulation."  S.G. Loewendick & Sons, Inc. v. 
Reich, 70 F.3d 1291, 1294 (D.C. Cir. 1995) (quoting Thomas Jeffer- son Univ. v. Shalala, 512 U.S. 504, 512 (1994)).  We confess to not 
understanding how the FDA can reconcile its reading with the 
language of its own regulation, but stress that this issue has not 
been briefed and is not necessary to the decision in this case.



applicant does a poor job of designing its product to avoid 
infringing the patent-holder's patent, and the second appli- cant does a much better job?  The first applicant would then 
be sued for infringement by the patent-holder, but the second 
applicant would not.  Indeed, this is exactly what Mylan 
claims happened in the present case (although Mova vigorous- ly contests this claim).

     In such a situation, a literal reading of the statute admit- tedly produces a strange result.  The second applicant, even 
though it has designed its product well and avoided suit, is 
barred from selling its product until the first applicant's 
lawsuit finishes (maybe years later).  The ingenious second 
applicant is thus harmed, and the public is deprived of the 
fruits of its ingenuity--a result seemingly at odds with Con- gress's apparent purposes, in enacting section 355(j)(5)(B)(iv), 
of rewarding innovation and bringing generic drugs to market 
quickly.  Indeed, the first applicant could even collude with 
the original patent-holder to prolong their litigation, and 
thereby keep the second applicant's drug off the market 
indefinitely.14

     Yet we are not persuaded that this third anomaly suffices 
to show that a literal reading of the statute leads to results 
manifestly inconsistent with the intent of Congress.  The 
legislative history of section 355(j)(5)(B)(iv) is limited, and 
fails utterly to specify or even provide any signals as to 
whether Congress intended that a second ANDA applicant 
who was not sued for patent infringement would have to wait 
until one of the statutory triggers was satisfied, or instead be 

__________
     14 An amicus brief filed by Biovail Corporation International 
dramatically illustrates an analogous risk, not necessarily involving 
collusion.  Biovail was the second applicant to file a paragraph IV 
ANDA for a generic version of a heart medication.  Biovail was not 
sued by the pioneer drug company.  The first applicant and the 
pioneer drug company are now in litigation, and, Biovail claims, the 
pioneer is paying the first applicant some $10 million per quarter in 
exchange for the first applicant's agreement not to sell its product 
after the 30-month waiting period expires.  Under these circum- stances, neither party would seem to have maximum incentive to 
bring the litigation to a close.



able to immediately market its product.  Congress may very 
well never even have thought about this question.  But it is 
not inconceivable that Congress meant what the statute says, 
i.e., that the second applicant would have to wait for the first 
lawsuit to finish.  The fact that a patent-holder fails to sue an 
ANDA applicant does not necessarily mean that it has con- cluded that the applicant did a good job of designing around 
its patent.  The patent-holder might have simply made a 
mistake, and negligently failed to file suit (or filed a few days 
after the end of the 45-day window).  If a second ANDA 
applicant who is not sued by the patent-holder is allowed to 
immediately market its product, then the patent-holder's 
error will have unfairly deprived the first applicant of the 
benefits of the exclusivity period.  Moreover, even if the 
second applicant is sued, the successful-defense requirement 
will allow him to receive FDA approval immediately once the 
30-month waiting period expires.  Given the nature of litiga- tion, the first applicant's patent-infringement suit could easily 
take longer than thirty months.  The successful-defense re- quirement may therefore have the effect of allowing many 
ANDA applicants to sell their products without regard to the 
exclusivity period, a result that Congress might not have 
intended.15

     Additionally, there may be other ways in which a second 
applicant with a better product can bring that product to 
market before the first lawsuit terminates.  Amicus curiae, 
Teva Pharmaceutical Inc. ("Teva"), has pointed to one possi- bility.  Teva observes that the court-decision trigger, by its 
terms, can be satisfied by any "decision of a court in an action 
described in clause (iii) holding the patent which is the 
subject of the certification to be invalid or not infringed."  
Teva claims that the actions "described in clause (iii)" are not 

__________
     15 Under the FDA's regulation, once a later applicant's drug has 
been approved, it will apparently remain on the market even if the 
exclusivity period later begins to run.  The regulation only applies 
the 180-day exclusivity period to ANDAs that are "subsequent" to a 
successful defense by the first applicant, 21 C.F.R. s 314.107(c)(1), 
and an ANDA that has already been approved does not fall in this 
category.



limited to infringement suits by the patent-holder, because 
the last two sentences of clause (iii) say:

     Until the expiration of forty-five days from the date the 
     notice made under paragraph (2)(B)(i) is received, no 
     action may be brought under section 2201 of Title 28 for 
     a declaratory judgment with respect to the patent.  Any 
     action brought under section 2201 shall be brought in the 
     judicial district where the defendant has its principal 
     place of business or a regular and established place of 
     business. 21 U.S.C. s 355(j)(5)(B)(iii).  Thus, Teva says, a declaratory 
judgment action provides an alternative way of satisfying the 
court-decision trigger.  An ANDA applicant who doesn't want 
to wait for the first applicant's patent infringement litigation 
to finish can bring its own declaratory judgment action 
against the patent-holder, and, if the second applicant pre- vails, the court-decision trigger will be satisfied, and it will be 
allowed to market its product.

     Teva's argument is elegant and textually persuasive.  It 
also provides a particularly appropriate solution in cases in 
which the second applicant has done a better job of designing 
around the pioneer drug manufacturer's patent than the first 
did:  in such cases, the second applicant should find it (rela- tively) easy to win a declaratory judgment action against the 
patent-holder.  Teva's reading thus rewards those applicants 
(and only those applicants) who have built a better mouse- trap.

     Teva's reading is not, however, flawless.  One difficulty is 
that the 180-day exclusivity period will seemingly always go 
to the first applicant, no matter whose suit satisfies the court- decision trigger;  the statute provides that any applications 
after the first one "shall be made effective not earlier than 
one hundred and eighty days after" the court-decision trigger 
is satisfied.  21 U.S.C. s 355(j)(5)(B)(iv).16  It seems odd to 
reward the first applicant if some later applicant was the 

__________
     16 This is the most natural reading of the statute, but we do not 
necessarily find that it is the only permissible reading.



party that actually prevailed in the patent-infringement litiga- tion.17

     Mylan has also noted what may be a more serious fly in the 
(patented) ointment.  In order to satisfy the Constitution's 
case or controversy requirement, a party filing a declaratory 
judgment action must show that there is a controversy of 
"sufficient immediacy and reality to warrant the issuance of a 
declaratory judgment."  Federal Express Corp. v. Air Line 
Pilots Ass'n, 67 F.3d 961, 964 (D.C. Cir. 1995) (quoting 
Maryland Casualty Co. v. Pacific Coal & Oil Co., 312 U.S. 
270 (1941)).  To employ Teva's declaratory-judgment device, 
a party would therefore need to demonstrate a "reasonable 
apprehension of facing a lawsuit."  Federal Express Corp., 67 
F.3d at 964.  The Federal Circuit has exclusive jurisdiction 
over appeals in actions for patent infringement.  28 U.S.C. 
s 1292(c).  Under the Federal Circuit's caselaw, a declarato- ry judgment plaintiff must be able to point to some conduct 
by the patent-holder that suggests that the plaintiff is at risk 
of being sued;  the fact that the patent-holder has sued others 
is "pertinent," but "not always conclusive."  See West Interac- tive Corp. v. First Data Resources, 972 F.2d 1295, 1297-98 
(Fed. Cir. 1992).  An ANDA applicant seeking to bring a 
declaratory judgment action might have difficulty in meeting 
this test, especially if the patent-holder disclaims any inten- tion of bringing suit.18 __________
     17 Indeed, the first applicant may still be enmeshed in patent- infringement litigation when the 180-day period begins, and there- fore be unable to take advantage of the exclusivity period.

     18 One way of eliminating strategic behavior of this kind might be 
for the FDA to provide by regulation that a court decision ruling 
that an ANDA applicant cannot reasonably anticipate suit by a 
patent-holder is equivalent, for purposes of section 355(j)(5)(B)(iv), 
to a ruling that the patent is invalid or not infringed.  After all, the 
purpose of the scheme set up by section 355(j)(5)(B) is to allow the 
patent-holder an opportunity to defend its patent.  If the patent- holder declines even to create enough adversity to support a 
declaratory judgment action, it might well be fair to deem the 
patent-holder to have conceded noninfringement, at least for pur- poses of the statutory scheme.  Certainly, there would be no danger 



     The problem of the meritorious second applicant is a real 
one, but the successful-defense requirement is too blunt an 
instrument to solve it.  The requirement cannot be reconciled 
with the literal language of the statute, and alters the statuto- ry scheme in a number of ways that do not clearly serve 
congressional intent.  We do not, of course, foreclose the 
FDA from attempting to address the problem of the meritori- ous second applicant in some narrower way, as long as that 
solution conforms to the statute.  For now, however, we are 
presented with the successful-defense requirement, and we 
uphold the district court's decision to enjoin the FDA's en- forcement of that requirement.

     B.Upjohn's Motion to Intervene

     We now turn to Upjohn's appeal of the district court's 
denial of its motion to intervene.  The district court denied 
Upjohn's motion on two grounds:  first, that Upjohn's motion 
was mooted by the grant of the preliminary injunction, and 
second, that s 355(j)(5)(B)(iv) "does not provide a cognizable 
interest upon which a pioneer patent owner or an NDA owner 
can challenge the approval of an ANDA."  We find that the 
district court was in error on both grounds.

     A motion to intervene as of right turns on four factors:  (1) 
the timeliness of the motion;  (2) whether the applicant 
"claims an interest relating to the property or transaction 
which is the subject of the action," Fed. R. Civ. P. 24(a);  (3) 
whether "the applicant is so situated that the disposition of 

__________ in such a case that the patent-holder's failure to enforce its patent is 
attributable to a mistake.

     Moreover, the Federal Circuit has had no occasion to decide 
whether there is "a controversy of sufficient immediacy and reality" 
to support a declaratory judgment action, Federal Express Corp., 67 
F.3d at 964, when the plaintiff requires a judgment under section 
355(j)(5)(B) in order to bring its product to market.  It is possible 
that such a statutorily-created bottleneck, coupled with the statute's 
express reference to declaratory judgment actions as a means of 
relieving that bottleneck, might suffice to allow a plaintiff to show 
the existence of a "case or controversy" without demonstrating an 
immediate risk of being sued.



the action may as a practical matter impair or impede the 
applicant's ability to protect that interest," id.;  and (4) wheth- er "the applicant's interest is adequately represented by 
existing parties."  Id.  To the extent that a district court's 
ruling on a motion to intervene as of right is based on 
questions of law, it is reviewed de novo;  to the extent that it 
is based on questions of fact, it is ordinarily reviewed for 
abuse of discretion.  See Massachusetts School of Law at 
Andover, Inc. v. United States, 118 F.3d 776, 779-80 (D.C. 
Cir. 1997) (noting, however, that application of the abuse-of- discretion standard seems anomalous in some circumstances).  
The issues that Upjohn raises on its appeal are all pure 
questions of law, so we apply de novo review.

     The district court erred in finding that Upjohn's motion to 
intervene was moot.  The district court had entered only a 
preliminary injunction, not a permanent injunction.  The 
district court presumably would have considered new evi- dence or new arguments in future proceedings, had Upjohn 
(or some other party) wished to present them.  And interven- ing even after the injunction had been issued would have 
allowed Upjohn to participate in the appeal of the injunction.   
See Massachusetts School of Law at Andover, 118 F.3d at 779 
(discussing intervention before the district court for purposes 
of appeal).

     The district court was also in error in finding that Upjohn 
did not "claim[ ] an interest relating to the property or the 
transaction which is the subject of the action," as is required 
by Federal Rule of Civil Procedure 24(a)(2).  Rule 24(a) 
"impliedly refers not to any interest the applicant can put 
forward, but only to a legally protectable one."  Southern 
Christian Leadership Conference v. Kelley, 747 F.2d 777, 779 
(D.C. Cir. 1984).  Thus, a party that seeks to intervene as of 
right must demonstrate that it has standing to participate in 
the action.  See id.  There is no dispute that Upjohn has 
constitutional standing;  numerous cases have found that a 
firm has constitutional standing to challenge a competitor's 
entry into its market.  See, e.g., Association of Data Process- ing Serv. Orgs., Inc. v. Camp, 397 U.S. 150, 152 (1970) ("Data 
Processing ").



     Mylan argues, however, that Upjohn is not within the "zone 
of interests" of section 355(j)(5)(B)(iv), and that it therefore 
lacks prudential standing.  We do not agree.  The first step 
in the prudential standing analysis is to identify the interests 
protected by the statute.  To do so, we consider the purposes 
of the specific statutory provision that is at issue (here, 
section 355(j)(5)(B)(iv)), read in the context of the statutory 
scheme as a whole.  See Bennett v. Spear, 117 S. Ct. 1154, 
1167 (1997);  Clarke v. Securities Industry Assoc., 479 U.S. 
388, 401 (1987) (stating that a court is "not limited to consid- ering the statute under which respondents sued, but may 
consider any provision that helps us to understand 
Congress's overall purposes....").  The purpose of section 
355(j)(5)(B)(iv) is to provide a reward, in the form of an 
exclusivity period, to generic drug companies that are the 
first to file paragraph IV ANDAs.  Section 355(j)(5)(B)(iv) is 
not intended to benefit pioneer drug companies directly.  
Indeed, quite the opposite is true:  the provision is intended 
to reward generic drug manufacturers who challenge pioneer 
drug companies' patents.  Thus, in the nomenclature of this 
circuit's caselaw, Upjohn cannot show that it is an "intended 
beneficiary" of section 355(j)(5)(B)(iv).  Scheduled Airlines 
Traffic Offices, Inc. v. Dept. of Defense, 87 F.3d 1356, 1359 
(D.C. Cir. 1996) ("Scheduled Airlines").

     But a plaintiff can be within the zone of interests of a 
statute even in the absence of "an indication of congressional 
purpose to benefit the would-be plaintiff."  Clarke, 479 U.S. 
at 399-400.  As the Court recently made clear in National 
Credit Union Administration v. First National Bank & 
Trust Co., 118 S. Ct. 927 (1998) ("NCUA "), a plaintiff only 
needs to show that its interest is among those "arguably ... 
to be protected" by the statute.  Id. at 935 (quoting Data 
Processing, 397 U.S. at 153) (emphasis added).  This analysis 
focuses, not on those who Congress intended to benefit, but 
on those who in practice can be expected to police the 
interests that the statute protects.  In NCUA, the provision 
before the Court imposed a rule called the "common bond 
requirement," which limits the membership of credit unions 
to "groups having a common bond of occupation or associa-



tion."  12 U.S.C. s 1759 (1994).  The question for the Court 
was whether a group of banks who had an interest in limiting 
the markets that credit unions could serve were within the 
zone of interests of this provision.  The Court found that the 
common bond requirement was intended to "reinforce the 
cooperative nature of credit unions, which in turn was be- lieved to promote their safety and soundness and allow access 
to credit by persons otherwise unable to borrow."  NCUA, 
118 S. Ct. at 935 n.6.  The Court reasoned that "by its very 
nature, a cooperative institution must serve a limited mar- ket," id., so that there is an "unmistakable" link between the 
statute and a "limitation on the markets that federal credit 
unions can serve."  Id. at 935 & n.6.  The Court concluded 
that limiting the markets served by credit unions is therefore 
an interest "arguably to be protected" by the statute, so that 
the banks had prudential standing.

     The test applied by NCUA is not far removed from this 
circuit's "suitable challenger" test.  See, e.g., Scheduled Air- lines, 87 F.3d at 1359-61 (applying this test).  Under the 
"suitable challenger" test, a plaintiff must demonstrate that 
its "interests are sufficiently congruent with those of the 
intended beneficiaries that the litigants are not 'more likely to 
frustrate than to further ... statutory objectives.' "  Id. at 
1359 (quoting First Nat'l Bank & Trust Co. v. National 
Credit Union Admin., 988 F.2d 1272, 1275 (D.C. Cir. 1993) 
(quoting Clarke, 479 U.S. at 397 n.12)).  NCUA allows a 
plaintiff to demonstrate that its interest and the interest 
served by the statute have, by their "very nature," an "unmis- takable" link.  NCUA, 118 S. Ct. at 935 & n.6;  in other 
words, the plaintiff may show an inevitable congruence be- tween the two interests.  The two standards are thus very 
similar.

     It seems clear under NCUA that Upjohn has prudential 
standing.  Here, Upjohn is seeking to enforce (its interpreta- tion of) section 355(j)(5)(B)(iv), a statute by which Congress 
sought to regulate the timing of generic drug manufacturers' 
entry into the market.  Although the statute speaks directly 
only to freeing the first generic drug company to file a 



paragraph IV ANDA from competition from other generic 
drug manufacturers, this necessarily entails freeing the pio- neer drug producer from such competition as well.  Thus, 
Upjohn's interest in limiting competition for its product is, 
"by its very nature," NCUA, 118 S. Ct. at 935 n.6, linked with 
the statute's goal of limiting competition between generic 
manufacturers.  See also MD Pharmaceutical, Inc. v. Drug 
Enforcement Admin., 133 F.3d 8, 12-13 (D.C. Cir. 1998) 
(finding that a drug company was within the zone of interests 
of an "entry-restricting" statute that regulated entry into its 
market);  Scheduled Aircraft, 87 F.3d at 1360-61 (finding that 
a party seeking to enforce a "statutory demarcation" is a 
suitable challenger) (quoting First Nat'l Bank & Trust, 988 
F.2d at 1278).

     Upjohn need not show anything more than that it has 
standing to sue in order to demonstrate the existence of a 
legally protected interest for purposes of Rule 24(a).  See 
Mausolf v. Babbitt, 85 F.3d 1295, 1299-1302 (8th Cir. 1996) 
(finding that a showing of standing suffices to demonstrate a 
legally protected interest for purposes of Rule 24(a));  but cf.  
United States v. 39.96 Acres of Land, 754 F.2d 855, 859 (7th 
Cir. 1985) (finding, on the peculiar facts of that case, that 
more than a showing of standing was required).  We there- fore reject the district court's contrary conclusion that Up- john did not have a sufficient interest in the action to inter- vene.

     The district court never reached the remaining elements of 
the Rule 24(a) analysis--timeliness, the risk that Upjohn's 
interests would be impaired, and whether Upjohn's interests 
were already adequately represented in the litigation.  Up- john has included in its brief on this appeal a number of 
arguments for the affirmance of the district court's injunction.  
In order to determine whether Upjohn is properly a party to 
the appeal of the injunction question, we must reach the 
remaining Rule 24(a) issues.  See Dimond v. District of 
Columbia, 792 F.2d 179, 193 (D.C. Cir. 1986) (similarly ad- dressing Rule 24(a) issues that the district court had, after 
making an erroneous legal ruling, failed to reach);  see also 



Mausolf v. Babbitt, 125 F.3d 661, 666-67 (8th Cir. 1997) 
(holding that if the court of appeals reverses the district 
court's denial of a party's motion to intervene, that party may 
participate in an appeal of a later ruling in the same litigation, 
if it has met the procedural requirements for doing so).  As 
to timeliness, Upjohn sought to intervene a few weeks after 
Mova initiated its action, and before the district court ruled 
on the preliminary injunction;  this cannot be regarded as 
untimely.  Upjohn's interests were also at risk;  Upjohn was 
in danger of losing market share to Mylan if the district court 
denied the injunction and allowed Mylan's product on the 
market.  Finally, as to adequacy of representation, Mova is a 
generic drug manufacturer, and therefore might have strate- gic reasons not to press certain arguments available to Up- john in anticipation of (perhaps) finding itself in Mylan's 
situation in a future case.  We thus conclude that Upjohn was 
entitled to intervene as of right, and that Upjohn is therefore 
a proper party to the appeal of the injunction order and in all 
further proceedings in the district court.19

                               III. Conclusion


     We find that the FDA exceeded its statutory authority in 
imposing the successful-defense requirement as a prerequisite 
to the invocation of the 180-day exclusivity rule by a first 
applicant under section 355(j)(5)(B)(iv).  The successful- defense requirement is inconsistent with the statutory text 
and structure, and is not justified by a need to protect the 
essential function of the statute or a clear congressional 
intent.  We therefore affirm the district court's decision to 
strike down the successful-defense requirement.

     As to Upjohn's motion to intervene, we conclude that the 
district court erred in finding that Upjohn's motion was moot 
and that Upjohn did not have a sufficient interest in the 

__________
     19 Upjohn also challenges the district court's denial of its motion 
for permissive intervention under Federal Rule of Civil Procedure 
24(b).  Because we find that Upjohn was entitled to intervene as of 
right, we do not reach this issue.



subject-matter of the litigation.  We also find that the other 
elements of Rule 24(a) have been met by Upjohn, and reverse 
the district court's denial of Upjohn's motion.  Upjohn is 
properly a party to this appeal, and to any proceedings on 
remand. So ordered.