IT’S THE ANTIGEN STUPID[1]:  A RISK/REWARD APPROACH TO THE PROBLEM OF ORPHAN DRUG 
ACT EXCLUSIVITY FOR MONOCLONAL ANTIBODY THERAPEUTICS


By Robert A. Bohrer[*]


The Orphan Drug Act of 1983 is an attempt at solving an important problem:  how 
to induce a market-driven pharmaceutical industry to develop new therapeutics 
for diseases affecting relatively small numbers of persons.  In the case of 
protein therapeutics produced by biotechnology, the FDA’s administration of the 
Act has repeatedly led to litigation and controversy.  The principal and 
recurrent problem has been determining when a second applicant for the approval 
of a drug to treat an orphan disease (or “indication”) is seeking approval for 
the “same” drug as a previously approved orphan drug.  The problem is how to 
provide sufficient protection for pioneering orphan drugs to stimulate the 
development of new drugs for smaller patient populations, while not so broadly 
protecting orphan drugs as to preclude additional therapeutics that make new 
contributions to the treatment of those diseases.   In the context of antibody 
therapeutics, that problem is somewhat different than for other protein 
therapeutics.  For antibodies, it is the antigen that will primarily determine 
the antibody’s efficacy, and it is the proof that an antigen is clinically 
useful in the treatment of a disease that provides the biggest risk in antibody 
development.   The approach to orphan drug exclusivity recommended here is that 
the FDA presume antibodies to the same antigen, of the same immunoglobin class 
and with the same mechanism of action to be the same drug unless the second 
antibody is shown to be clinically superior to the first.

[*2]
I. Introduction

      The Orphan Drug Act of 1983 (the “Act”)[2] is an attempt at solving an 
important problem:  how to induce a market-driven pharmaceutical industry to 
develop new therapeutics for diseases affecting relatively small numbers of 
persons.  Although some drugs for small patient populations have been 
exceedingly profitable,[3] the need for special incentives to spur pharmaceutical 
manufacturers to undertake the costly and risky process of drug development and 
research for less common diseases was clear to the Congressional sponsors of the 
Act and the Act has been largely judged a success.[4]  A key part of the solution 
that the Act provides is to reward manufacturer “sponsors” of orphan drugs with 
a seven year period of market exclusivity, during which the FDA will approve no 
other application to treat the same disease with the same orphan drug.  An 
orphan drug is defined as one treating a disease or condition which affects less 
than 200,000 persons in the United States or affects more than 200,000 persons 
but for which there is “no reasonable expectation that the cost of developing 
and making available in the United States a drug for such disease or condition 
will be recovered from sales in the United States of such drug.”[5]
      [*3]
	Despite the apparent simplicity and straightforward logic of the Act’s 
solution to the orphan disease problem, the FDA’s administration of the Act has 
repeatedly led to litigation and controversy.[6]  The principal and recurrent 
problem has been determining when a second applicant for the approval of a drug 
to treat an orphan disease (or “indication”) is seeking approval for the “same” 
drug as a previously approved orphan drug.   The FDA’s original approach to the 
“same drug” problem was to permit any chemical difference between a first and 
second orphan drug to be sufficient for the second to be approved.  This 
chemical identity test clearly undermined the basic purposes of the Act when 
applied to protein therapeutics.  In the new era of biotechnology, it proved all 
too easy for pharmaceutical manufacturers to make changes in a pioneering drug 
that would not affect its biological activity, and, if approved as a second 
entrant for an orphan indication, would greatly diminish the value of Orphan 
Drug Act exclusivity.[7]  The FDA has recognized the problem and repeatedly 
grappled with it.[8]  
      More recently, the FDA has revisited the problem of distinguishing same 
versus different under the Act for monoclonal antibody-based drugs, which are an 
important category of biotechnology-derived therapeutics.   Once again, the FDA 
appears to have gotten it wrong, providing a “solution” to the same versus 
different problem that seems almost certain to engender further uncertainty and 
litigation, and to undermine the important exclusivity incentives of the Act.  I 
use the word “problem” throughout this Article to refer to difficulty in 
determining the appropriate means by which to accomplish particular objectives.[9]  
In the case at hand, orphan drugs, the objective is to provide sufficient 
protection for pioneering orphan drugs to stimulate the development of new drugs 
for smaller patient populations, while not so broadly protecting orphan drugs as 
to preclude additional therapeutics which make new contributions to the 
treatment of those diseases.  In other words, the problem is to distinguish 
truly “different” second entries into orphan drug markets from would-be 
competitors that are more or less the “same” as the innovator first-entrant.    
      [*4]
A reasonable solution to the problem of orphan drug “sameness,” for 
monoclonal antibodies as well as other therapeutics, should satisfy three 
criteria.  First, it must not undermine the essential incentives of the Orphan 
Drug Act for undertaking the risky process of drug development for small patient 
populations.  Second, the criteria for determining sameness must provide 
relatively clear guidance to pioneer orphan drug developers as to the probable 
scope of their protection, at the earliest possible stages of drug development.   
Finally, the criteria must also provide would-be second entrants relatively 
clear guidance as to their chances of successful approval by the FDA as 
“different” and a reasonably clear roadmap to the approval of a second orphan 
antibody for the same indication.  The FDA’s efforts, which have focused on the 
chemical composition of monoclonal antibody therapeutics, have not been 
particularly successful at addressing these concerns, both with respect to not 
undermining the Act’s incentives and in providing clear guidance to pioneer 
orphan drug developers and second entrants.[10]   
      This Article will take a very different approach than that of the FDA to 
solving the problem of Orphan Drug Act exclusivity for monoclonal antibody 
products, one that attempts to relate orphan drug exclusivity rewards to the 
degree of orphan drug sponsor risk.   In providing the basic incentive of 
exclusivity, the Orphan Drug Act sponsors clearly intended to provide a 
sufficient reward for undergoing the risky (and costly) process of drug 
development.  Therefore, this Article suggests that by focusing on the factors 
that determine the degree of risk in orphan drug development, the FDA can 
implement guidelines for monoclonal antibodies that better serve the purposes of 
the Act and solve the “same drug” problem in a much more satisfactory manner.  
In other words, if the purpose of orphan drug incentives is to provide 
incentives for risky, expensive drug development, then orphan drug protection 
(and its converse, second-entry approval) should be correlated with the degree 
of risk faced by the applicant.  To use an example from a non-orphan drug 
context, the first developer of a selective serotonin reuptake inhibitor 
(Prozac) faced a good deal more uncertainty over the likelihood of proving 
efficacy in the clinic than subsequent developers of drugs in that class.  
Second entrants into a therapeutic class still face risk and may make important 
contributions (in increased efficacy or a better side effect profile) but do not 
face the same level of risk as the pioneer entrant.  
      Antibody therapeutics can be similarly thought of as belonging to 
therapeutic classes, in which the pioneer entrant faces the greatest risk.  In 
particular, I will argue that the biggest risk-creating uncertainty in 
developing a monoclonal antibody therapeutic for the treatment of a disease 
(orphan or otherwise) is whether an antibody directed to the particular disease-
related target (antigen, in antibody terms) will prove to have a sufficient 
positive therapeutic effect.  In other words, at the present state of antibody 
science, the biggest unknown is the value of the antigen selected as a target 
for particular disease.   Thus I will propose regulations under which the FDA 
would require sponsors of [*5]second antibody therapeutics aimed at the same antigen target for the same disease to demonstrate the clinical superiority[11] of their antibody.
      Part II of this Article provides a brief background to the field of 
monoclonal antibodies and the development of monoclonal antibodies as 
therapeutics.  Part III reviews the history of the FDA’s interpretation of the 
Orphan Drug Act and the FDA’s recent “Guidance for Industry: Interpreting 
Sameness of Monoclonal Antibody Products Under the Orphan Drug Regulations.”[12]  
Part IV provides the proposed framework for evaluating solutions to the problem 
of sameness for monoclonal antibody products—a risk-reduction-based model which 
rewards risk-taking pioneer orphan antibody developers by grounding orphan drug 
exclusivity in the magnitude of the risk faced by competing orphan drug sponsors 
for the same indication.  In Part V, the Article concludes with a summary of the 
proposed antigen-based framework for determining the sameness of antibody 
products in terms of the risk-reduction criteria of Part IV.
      
      
II. An Introduction to Monoclonal Antibodies

      Antibodies are proteins[13] made by the B cells of the immune systems of 
higher organisms (see Figure 1 for a detailed drawing of a typical antibody, or 
an Immunoglobin G or “IgG”).  The function of an antibody is to bind to an 
antigen, or substance which is recognized by the immune system as “foreign” and 
which therefore stimulates an immune response.  A three-dimensional feature of 
an antigen to which an antibody can bind is called an antigenic determinant, or 
epitope.  Large antigens can have many epitopes and even a single epitope may be 
the target for a number of different antibodies.[14]   Thus, stimulation of the 
immune system by a pathogen, such as a virus, results in many different B cells 
producing antibodies to that pathogen. [*6] [*7]
      An antibody binds to its antigen when the antigen fits into the complex 
three-dimensional space created by the variable regions of the antibody’s heavy 
and light chains (see Figure 1, areas labeled VH & VL on the upper right “arm” 
of the “Y”). The diversity of antibodies produced by antigen stimulation is the 
result of the rearrangement of the genes that produce the variable regions of 
the heavy and light chains and result in the specific three-dimensional 
configuration of the “complementarity determining regions” (hereinafter “CDR”) 
of the VH and VL (see the regions labeled “CDR” on the upper left hand “arm” of 
the “Y” in Figure 1).   
      The great leap forward in the ability to use antibodies as tools in 
biology and medicine came when Kohler and Milstein discovered that taking an 
individual B cell and fusing it with a tumor cell would result in an “immortal” 
hybrid cell, called a hybridoma, that would produce a cell line which could be 
maintained indefinitely in cell culture.[15]   Each daughter cell, or clone in the 
colony begun by a specific hybridoma, produces the same antibody as the original 
B cell fusion partner.  Such immortalized B cell hybridoma cell colonies, or 
“clones,” that produce antibodies of a single type are referred to as 
monoclonal, and the antibodies they produce are known as monoclonal antibodies.
      The great advantage of Kohler and Milstein’s breakthrough development of 
the technique for creating monoclonal antibodies is the ability to select a 
monoclonal antibody with desired characteristics and then to produce large 
quantities of the desired antibody by growing, dividing and growing still more 
colonies derived from the original hybridoma.  Among the significant 
characteristics of an antibody are its specificity[16] for a particular antigen 
and its affinity[17] for that antigen.   A desirable antibody for therapeutic 
purposes must be relatively specific for the target antigen and of a reasonably 
high affinity for the antigen.  Of course the binding of the antibody to the 
particular antigen chosen must also have the desired therapeutic effect.  
      [*8] In addition to specificity and affinity, antibodies also are characterized by their immunoglobin class or subclass, such as IgG1, IgG2, IgA1, etc.[18]  Antibodies of different immunoglobin types or subtypes may have naturally differing biological functions (such as the induction of cell-killing activity or the stimulation of the production of cytotoxic proteins).[19]  In addition, developers of antibody therapeutics may change or add to the antibody’s mechanism of action by linking an active agent, such as a radioactive isotope, to the antibody.[20]
      Since Kohler and Milstein made possible the unlimited production of 
specific antibodies, researchers and biotechnology companies have seen the 
potential of monoclonal antibodies as “magic bullets”—biological molecules that 
can be specifically targeted to hone in on disease agents or diseased cells 
without unwanted effects on normal tissue or organs.[21]  Nonetheless, the first 
generation of monoclonal antibody therapeutics turned out to be largely a flop, 
due to an unanticipated problem with human patients’ own immune systems 
responding to mouse-derived antibodies as foreign protein, a response known as 
HAMA, or Human Anti-Mouse Antibodies.[22]   The second generation of antibody 
therapeutics has attempted to overcome this in a number of ways, largely by 
“humanizing” the portions of the antibody that are not crucial to the antibody 
binding to its antigen.[23] [*9]
      A number of approaches have been developed to produce antibodies that do 
not produce a significant immune response directed at the antibody, resulting in 
successful antibody-based therapeutics, such as Centocor’s Reopro and 
Remicade,[24] IDEC’s Rituxan[25] and Genentech’s Herceptin.[26]  Each of these 
successful antibodies has had to resolve two problems.  The first is HAMA, but 
the second, identifying a relevant antigen, is more challenging and far more 
diverse.  While there are multiple solutions to the HAMA problem and each viable 
solution may work for a large number of different antibody therapeutics, the 
challenge of finding a clinically relevant antigen must be solved anew for every 
disease for which an antibody therapeutic is sought.  At the same time, 
different approaches to solving HAMA may or may not make a difference in the 
efficacy of the antibody therapeutic.  To date, the answer to whether the 
different major approaches to preventing HAMA make a difference in the efficacy 
of the antibodies has not been resolved.[27]
      The identification of clinically relevant targets for antibody 
therapeutics has progressed rapidly in recent years, both in cancer and in 
inflammatory disease.  In cancer, the success of the lymphoma antibody Rituxan, 
an unconjugated antibody to the CD20 marker on B-cells, was followed by the 
development of two subsequent radio-isotope-linked anti-CD20 antibodies, Zevalin 
and Bexxar.[28]  Similarly, the success of an antibody to Tumor Necrosis Factor 
(TNF) in the treatment of Crohn’s disease and juvenile rheumatoid arthritis 
(Remicade) was followed by the approval of a second antibody to TNF (Humira).[29]  
Tumor Necrosis Factor is also the target for another [*10]arthritis disease, the non-antibody-based Enbrel.[30]  The evidence that the Epidermal Growth Factor Receptor (EGFR) is a potentially useful target for cancer drugs provides another example of competition to develop antibodies against a particular antigen.  According to one report, five different EGFR-blocking antibodies have been developed so far.[31]  As these examples illustrate, antibody development is driven by the identification of potentially useful antigens and multiple antibodies can be developed to a clinically useful antigen.
      Rituximab and infliximab also represent two of the main types of antibody 
therapeutics:  rituximab is an anti-cancer antibody that works by stimulating an 
immune response to the target cells, thus providing a reduction in tumor burden, 
while infliximab is an anti-inflammatory antibody that works by binding to a 
pro-inflammatory molecule (TNF-alpha) and blocking it from binding to its 
natural target.  Blocking and cytotoxic antibodies represent the majority of 
antibody therapeutics to date, and each depends on the identification of a 
clinically effective target, or antigen.  Cytotoxic antibodies require finding 
an antigen target for the diseased cells that will not cause unwanted targeting 
of normal cells while blocking antibodies require identifying an antigen target 
that plays a sufficiently important role in the disease process. 
      
      
III. The FDA’s Long Struggle To Interpret the Orphan Drug Act
      
      When the Orphan Drug Act was signed into law in 1983, it put in place a 
variety of incentives for the development of new drugs for “orphan” disease 
populations of less than 200,000 patients in the United States.[32]  It also 
created a major regulatory problem for the FDA:  how to administer its authority 
over the seven-year period of market exclusivity for pioneer sponsors of a new 
treatment for an orphan disease indication.  Under the statute, there are two 
stages in the orphan drug development process: designation and approval.  
Designation simply requires an FDA determination that the disease indication 
proposed by the applicant is in fact an orphan disease within the terms [*11]of the statute.[33]  Designation can be sought at any time during the drug development process and is really without significance unless the drug development process is successful and leads to marketing approval.  It is possible for more than one applicant to receive designation for the same or similar drugs and such competitions have recurred repeatedly.[34]  The more significant problem for the FDA occurs after the first such sponsor receives approval to market an orphan drug. The Act provides:
      
[T]he Secretary may not approve another application under section 505 [21 U.S.C. 
§ 355] or issue another license under section 351 of the Public Health Service 
Act [42 U.S.C. § 262] for such drug for such disease or condition for a person 
who is not the holder of such approved application, of such certification, or of 
such license until the expiration of seven years from the date of the approval 
of the approved application, the issuance of the certification, or the issuance 
of the license.[35] 

      The problem has been and continues to be for the Secretary, or in reality, 
the FDA, to determine when another application is for “such drug for such 
disease.”  The FDA’s initial response was to treat the new drugs produced by 
biotechnology, primarily proteins, which are very large biomolecules containing 
a hundred or more amino acids, in much the same way as more traditional drugs. 
Before the development of genetic engineering, drugs were commonly small 
molecules that were perhaps one-hundredth or even one three-hundredth the size 
of the average protein therapeutic.  For such small molecule drugs, even small 
changes in the molecular structure were likely to result in appreciable 
differences in the drug’s biological effect.  Thus the FDA took the view that 
for protein drugs, like small molecules, “such drug” was to be defined in terms 
of more or less absolute chemical identity.   On this basis, the FDA proceeded 
to approve as orphan drugs two very similar versions of recombinant human growth 
hormone, the first by Genentech, the second by Eli Lilly.[36]
      The continued use of a strict definition of “same” for macromolecules 
would have had a disastrous effect on the exclusivity incentives of the Orphan 
Drug Act for biotechnology companies developing protein-based therapeutics.  
Unlike the traditional small molecules, it is generally quite easy to make small 
changes in a protein, such as substituting similar amino acids, without 
affecting its activity.  The FDA responded by issuing new regulations, now 
codified in 21 C.F.R. § 316.3 (b)(13)(ii)(A)-(D) redefining “same drug” for 
large molecules.[37]  In short, the FDA almost completely reversed the [*12] presumption 
made for traditional drugs:  for small molecule drugs, only drugs with 
chemically identical active moieties (the principal active ingredient, 
regardless of its salt or ester formulation) are presumed to be the same; 
however, two proteins would be presumed to be the same even with minor 
differences in amino acid sequence, glycosylation,[38] or any other changes that 
occur after the gene encoding the protein has been translated into the primary 
amino acid sequence.  
      The FDA’s revised regulations were a significant attempt to solve the 
problem of providing real protection for pioneer developers of protein 
therapeutics:  no longer would the pioneer developer of a therapeutic protein 
for an orphan indication (such as for the long-term treatment of children who 
have growth failure due to a lack of adequate endogenous growth hormone 
secretion)[39] fear that a competitor could get approval simply by substituting 
one isoleucine for a leucine (or other very similar amino acid substitutes) in 
some region of the protein far from its active site.  Instead, any competitor of 
a substantially similar protein would have to prove that their product provided 
a real additional benefit to patients, or, as the FDA termed it, “clinical 
superiority.”  Thus, whatever the degree of chemical difference between two 
similar protein-based drugs, if the second drug actually worked better, or 
caused fewer side effects, the demonstration of clinical superiority would 
conclusively establish that the second orphan drug applicant was different from 
the first one approved.  
      The FDA’s effort to create a real barrier to “me-too” orphan drugs was a 
laudable one, but unfortunately, was soon shown to be porous.  In Berlex 
Laboratories, Inc. v. FDA,[40] Berlex had received approval for interferon-ß 1b 
for the treatment of multiple sclerosis.  The “1b” designation refers to 
interferon-ß produced in genetically engineered [*13] bacteria.  Avonex then applied 
to the FDA for approval of its interferon-ß 1a (produced in genetically 
engineered mammalian cells) for the treatment of multiple sclerosis.  Berlex had 
tested its drug at a high dose delivered by subcutaneous injection every other 
day.  Avonex had tested its drug at a much lower dose delivered once a week by 
intramuscular injection.  Despite the lower dose, the drug still worked.  
Perhaps because of the lower, less frequent dose and different route of 
administration, the drug produced less injection site reactions and problems.  
Because of this better safety, or side effect profile, the FDA ruled that Avonex 
product was clinically superior and approved it.  Avonex did not need to show 
that Berlex’s pioneering orphan drug would not work similarly at the lower dose 
and route of administration.  The District Court affirmed the FDA’s discretion 
in its interpreting the “same drug” provision of the Orphan Drug Act.[41]  
      On the heels of its victory in Berlex Labs, the FDA realized that the 
approach it had taken to the “same drug” problem for proteins generally would 
almost certainly lead to repeated controversies in the rapidly emerging field of 
antibody therapeutics.  The FDA’s solution to the problem of determining “same 
drug” for antibody therapeutics was published as a Guidance for Industry in July 
of 1999.[42]  The Guidance is remarkable in its simplicity.  The Guidance begins 
with a brief review of its interpretation of the Act for proteins the science of 
monoclonal antibodies and concludes that: 

The proposed interpretation of sameness for two monoclonal antibodies is that 
two monoclonal antibody drugs would be considered the same if the amino acid 
sequences of the complementarity determining regions were the same or if there 
were only minor amino acid differences between them. Other potentially important 
amino acid differences outside the complementarity determining regions, or 
differences due to glycosylation patterns or post translational modifications 
would not per se cause the products to be considered different unless the 
subsequent drug was shown to be clinically superior.[43]

In other words, two antibody-based drugs would be considered different if there 
were substantial amino acid differences in their CDRs and the same if there were 
not such substantial differences.  This is despite the Guidance’s acknowledgment 
that “[b]ecause of the many processes involved in generating antibody diversity, 
it is unlikely that independently derived monoclonal antibodies with the same 
antigen specificity will have identical amino acid sequences.”[44]  The FDA based 
its decision to determine “same drug” [*14] for antibodies on the amino acid sequence 
of the CDR region in order to harmonize its application in this context with its 
general approach to macromolecules:

The definition of sameness for a macromolecule is based on its principal 
molecular structure. For the purpose of determining sameness of naked monoclonal 
antibodies under the Orphan Drug Act and its implementing regulations, the 
complementarity determining regions of the heavy and light chain variable 
regions will be viewed by the FDA as the principal molecular structural feature 
of a monoclonal antibody product.[45]

While the FDA decision has the virtue of consistency, it ignores the difference 
between monoclonal antibodies and other macromolecules, particular proteins.  
While proteins are indeed generally defined by their principal molecular 
structure, antibodies are generally defined by their antigen.  It is this 
decision of the FDA, to focus on the amino acid sequence of the CDR (antigen 
binding) regions of the monoclonal antibody therapeutic, that is likely to 
substantially undermine the incentives of the Orphan Drug Act.  It provides the 
pioneer inventor with a very narrow scope of exclusivity and provides follow-on 
developers relatively easy entry into the market with a “me-too” orphan antibody 
therapeutic.  It is very likely, as the FDA itself acknowledges, that two 
antibodies of similar specificity and affinity for the same antigenic 
determinant will have different amino acid sequences, even in their CDR.[46]  


IV.  Matching Reward to Risk:  An Antigenic Determination of Same Drug for 
Monoclonal Antibody Orphan Drugs

      This section will address the problem of monoclonal antibody orphan drug 
“same” drug determinations in the general context of antibody drug discovery and 
development.  By looking at orphan drug discovery in this context, one can 
better understand how the standard used to determine the scope of orphan drug 
exclusivity will impact the actual process of discovering and developing new 
antibody-based drugs for orphan disease indications.  In particular, the thesis 
presented here is that the scope of orphan drug exclusivity should be related to 
the risk of orphan drug discovery and [*15] development.  The protection should be 
tailored to prevent second orphan drug applicants from a low-risk development 
effort that largely capitalizes on the discovery and development of the first 
approved orphan sponsor.  In this way, the reward of orphan drug exclusivity is 
commensurate with the risk undertaken by the orphan drug developer.  Any 
standard of exclusivity that allows second entrants to free-ride upon the 
discovery and development of first entrants would substantially undermine the 
value of orphan drug exclusivity and reduce the investment in developing 
treatments for orphan disease indications.[47]
      Drug discovery generally begins in one of two ways:  either it begins with 
the serendipitous finding that a particular substance has an apparently 
medically useful property, as was the case with Alexander Fleming’s discovery of 
the mold that produced penicillin,[48] or it begins with the discovery of a 
pathway or biological process that provides a target for drug discovery and 
development, as is increasingly the case in the era of genomics and 
proteomics.[49]  Of course, even when a drug is found serendipitously, it 
immediately triggers a search for an understanding of its mechanism and its 
structure-function relationship, so that improved versions, second generation 
drugs, or competitive drugs can be developed.   
      Once a candidate drug that operates on a new pathway or target has been 
identified, it undergoes a variety of pre-clinical tests to determine its 
toxicology profile and its efficacy in animal models, followed by several 
increasingly large-scale tests for safety and efficacy in the appropriate human 
patient population.[50]  If it is shown to be safe and effective for a major 
disease indication, it is likely to stimulate other companies to develop drugs 
aimed at the same pathway, perhaps even before the first drug is approved by the 
FDA.  These second entrants rush to identify new chemical entities that act on 
the same target because the work of the earlier “pioneer” company, demonstrating 
the clinical utility of the target, substantially reduces the risk of their own 
development [*16]efforts.[51]  In the biggest therapeutic categories, such as anti-
cholesterol agents, anti-depressants, and anti-hypertensives, there are numerous 
agents competing in a category and several in each disease category are aimed at 
the same target: statins for cholesterol;[52] selective serotonin-reuptake 
inhibitors for depression;[53] and multiple beta blockers or ACE inhibitors for 
hypertension.[54]  These therapeutic categories are large enough to induce the 
development of multiple drugs, with sometimes small chemical and pharmacological 
differences, each generating a profit.  For these large patient population 
indications, pioneer drug developers are not deterred by the knowledge that any 
new target developed with clinical utility will quickly be pursued by other drug 
companies.  The pioneer developer’s additional risk in developing a drug for a 
previously unproven target is generally rewarded by the brand recognition and 
market share that accrue to the first entrant into a new therapeutic category 
with a very large potential patient population.[55]  
      Of course, the rationale for the Orphan Drug Act’s exclusivity provisions 
was to provide an inducement for the development of drugs for smaller markets, 
where multiple entrants would make the market unprofitable for all.  Because of 
this, it is particularly [*17] important to provide pioneers with protection against 
“me-too” entrants into the market.[56]  In the context of the Orphan Drug Act, 
this protection against “me-too” entrants takes the form of a seven-year period 
of exclusivity where the FDA will not approve the “same” drug for the same 
indication.  As discussed above,[57] the FDA has recognized for some time that 
protein-based therapeutics present a different challenge for Orphan Drug Act 
exclusivity determinations than small-molecule chemically synthesized 
pharmaceuticals.  The FDA’s approach to the problem has been to continue to 
focus on chemical identity as the first principle by which exclusivity 
determinations should be made, while attempting to accommodate the challenges of 
protein therapeutics by loosening the standard from chemical identity to “minor 
amino acid differences” in the CDR.[58]  The FDA’s use of “minor amino acid 
differences” is an attempt to define a structure-based standard that bars minor 
changes in proteins that do not alter the three-dimensional structure of the 
protein to an extent that would affect the protein’s biological activity.   Such 
a chemical-identity approach is workable for protein drugs that are copies of 
naturally-occurring, secreted proteins that operate by binding to receptors.  It 
would indeed be difficult for a would-be orphan drug copyist to make substantial 
changes to the amino acid sequence of erythropoietin, human growth hormone, or 
interferon-ß[59] without losing the desired biological activity.  In the antibody 
field, the FDA Guidance uses this same concept of amino acid sequence similarity 
and specifically applies it to the portion of the antibody that binds to its 
target antigen, which is known as its CDR.[60]
      However, antibody drugs are different.  While a would-be copyist who tried 
to make changes to a successful pioneer antibody would indeed find it difficult 
to significantly alter the antibody, in the case of monoclonal antibodies to the 
same antigen produced by different laboratories, it is quite possible for there 
to be significant differences in the amino acid sequence of two, independently-
developed antibodies, even in the CDR regions that are the focus of the FDA’s 
proposed guidance for “same drug” determinations for orphan drug monoclonal 
antibodies.[61]  As discussed in Part II of this [*18] Article, it is not only 
scientifically likely that different antibodies to the same antigen will both be 
clinically effective, but it has already been the case that multiple antibodies 
will be developed against a particular clinically relevant antigen.[62]  The 
problem for orphan drug exclusivity is adequately protecting the first developer 
of an antibody that successfully targets a particular antigen for the treatment 
of an orphan disease.  To achieve this, the FDA should define antibody 
“sameness” by the antigen to which the antibody joins, not by amino acid 
sequence.[63]  While developers of additional antibodies to an antigen that has 
been shown to be clinically significant certainly face some risk, they will 
generally face much less risk than did the pioneer entrant who demonstrated the 
utility of the target.  
      What then would be the definition of “same drug” for antibody 
therapeutics?  Two antibodies with the same mechanism of action[64] would be 
considered the same drug for Orphan Drug Act exclusivity if they both targeted 
the same antigen for the treatment of the same orphan indication, unless the 
second antibody therapeutic was demonstrated to be clinically superior to the 
prior-approved antibody therapeutic.   How would an antigen-focused presumption 
of sameness compare with the FDA’s CDR-focused sameness standard in satisfying 
the three criteria for a reasonable solution to the orphan drug sameness problem 
for antibodies?[65]  With respect to the first criteria, insuring that there is a 
sufficient incentive for undertaking the risky process of drug development, the 
antigen presumption clearly broadens the scope of protection to pioneer orphan 
antibody developers, thus providing a greater incentive than the FDA’s standard.  
Second, the antigen presumption provides pioneer developers with a relatively 
clear understanding of the scope of their protection.  Third, the antigen 
presumption similarly provides would-be second orphan antibody developers a 
clear understanding of their chances of approval [*19] and a roadmap to second 
antibody approval.  If the second antibody is aimed at the same antigen, it must 
either have a different mechanism of action (e.g. radioactive versus non-
radioactive) or demonstrate in clinical trials that it is superior to the 
pioneer antibody either in efficacy or safety.  
      The FDA’s administration of the Orphan Drug Act has always recognized that 
while the exclusivity protection against competition from the same drug was 
central to the Act, the exclusivity provisions should not be used as a barrier 
to the development of superior therapeutics.  The FDA Orphan Drug Regulations 
were intended “to ensure that improved therapies will always be marketable, and 
that orphan drug exclusive approval does not preclude significant improvements 
in treating rare diseases.”[66]  The FDA introduced the concept of “clinical 
superiority” to accommodate the need to allow significant improvements in the 
treatment of orphan indications which might otherwise be barred from approval by 
similarity to a previously approved orphan drug.  In the context of antibody 
therapeutics for orphan drugs, under this approach clinical superiority would be 
the means by which developers of second antibody therapeutics to the same 
antigen could prove their antibody sufficiently different to warrant approval.  
A second antibody developer that has chosen a different method of reducing the 
immunogenicity of an originally murine antibody may well be able to establish 
that its antibody produces a longer lasting biological effect, or otherwise 
improves patient care.[67]  Conversely, the first company to demonstrate that an 
antibody directed against a particular antigen is safe and effective for the 
treatment of an orphan indication would be entitled to market exclusivity 
against antibodies targeting the same antigen until one of those antibodies was 
shown to be safer or more effective.  Thus both first and second orphan antibody 
developers would have a much clearer understanding of when the demonstration of 
clinical superiority would be required.
      At this point, we can look at a hypothetical conflict between two 
currently marketed antibodies, infliximab, marketed as Remicade by Centocor,[68] 
and Adalimumab, marketed as Humira by Abbott Laboratories.[69]  Both are 
antibodies to the same antigen, tumor necrosis factor alpha, a cytokine (or 
chemical messenger) that plays a key role in producing inflammation.[70]   
Remicade was approved first and received orphan drug exclusivity for moderately 
to severely active Crohn’s disease resistant to [*20] conventional therapy[71] as well 
as non-orphan marketing approval for various indications in rheumatoid 
arthritis.[72]  Humira also received approval for use in various arthritis 
indications.  There is no actual orphan exclusivity battle between Humira and 
Remicade, since Abbott has not sought to have the FDA approve its antibody for 
use against Crohn’s disease, Centocor’s orphan indication.  However, we can 
hypothesize such an attempt and compare the results of applying FDA’s current 
policy and our proposed policy.  Under the FDA’s current policy, to determine 
approval of Abbott’s orphan Crohn’s indication application, the FDA would 
evaluate whether the two antibodies had the same or very similar amino acid 
sequences in their CDR or antibody-binding regions.  The fact that Humira is a 
recombinant human antibody while Remicade is a chimeric antibody (having a 
murine antibody-binding domain engrafted into a human IgG constant region)[73] 
certainly raises doubts as to whether the two antibodies share substantial 
homology in the antibody-binding regions.  There would not have been much 
certainty, either from Centocor’s perspective or Abbott’s, about the scope of 
Humira’s protection prior to the introduction of both drugs, though either one 
could use available laboratory techniques to determine the relevant amino acid 
sequences to predict the FDA’s response ex post.  Conversely, it is clear that 
under the test proposed here, Abbott would know that the only way to have their 
antibody approved in light of Centocor’s prior approval would be to demonstrate 
that Humira was clinically superior, that is, offers a significant benefit to 
patients over Remicade. 
      
      
V.  Conclusion

      The Orphan Drug Act provision for seven years of exclusivity against 
competition by the same drug for the same indication was intended to provide a 
significant incentive for undertaking the costly and risky process of drug 
development where the size of the patient population would limit ordinary 
commercial incentives.  For larger patient populations, drug developers know 
that they will likely face competition by similar drugs and knowingly bear the 
risks of drug development to capture the rewards that accrue to the first to 
establish a new therapeutic category.  The Orphan Drug Act, however, was 
predicated on the assumption that smaller patient populations needed an assured 
market size, free from “me-too” competition, to induce development and provide 
adequate returns to the pioneer developer.  
      The FDA’s experience with the first major biotechnology-produced products, 
proteins, convinced the FDA that protein therapeutics required a relaxed 
standard that would presume two proteins to be the same if there were no more 
than “minor differences in amino acid sequence.”  The focus on amino acid 
sequence for non-antibody proteins [*21] works because the general amino acid sequence 
determines the biological function of the protein.  For antibodies, however, the 
amino acid sequence in the antigen-binding region, or CDR, that is the focus of 
the FDA’s proposed Guidance for Industry, does not determine the antibody’s 
function in the same way that amino acid sequence determines an enzyme’s 
function.  Rather, it is the antigen that will primarily determine the efficacy 
of a particular antibody, and it is the proof that an antigen is clinically 
useful in the treatment of a disease that provides the biggest risk in antibody 
development.   The approach to orphan drug exclusivity recommended here is that 
the FDA presumes antibodies of the same immunoglobin class and with the same 
mechanism of action[74] to be the same drug unless the second antibody is shown to 
be clinically superior to the first.  This focus on the antigen would provide a 
substantial amount of certainty to the developers of antibody therapeutics for 
orphan indications, provide real protection against “me-too” antibodies, yet 
leave sufficient room for innovative antibody development to provide evidence of 
significant therapeutic advantage.

---

   [1] “It’s the Economy Stupid” was the central theme of Bill Clinton’s 1992 run 
for the presidency.  It is used here as a reminder that a successful approach to 
important problems requires keeping a clear focus on the central issue. See 
David S. Broder and Dan Balz, Gore Fails to Cash In on Prosperity; Swing Voters 
Give Vice President Little Economic Credit, Wash. Post, July 23, 2000, at 
Section A, page A01 for a reference to the Clinton theme.
  
   [*] Professor of Law at California Western School of Law.

   [2] Pub. L. No. 97-414, 96 Stat. 2049 (1983) (codified as amended in scattered 
sections of 21 U.S.C. and 42 U.S.C.).  See also Orphan Drug Regulations, 56 Fed. 
Reg. 3338, 3338 (1991) (to be codified at 21 C.F.R. pt. 316) (proposed Jan. 24, 
1991) (discussing legislative intent to provide economic incentives) 
(hereinafter “Proposed Regulations”).  This Article is specifically about the 
problem of monoclonal antibodies under the Orphan Drug Act and is not intended 
to be a complete introduction to the Orphan Drug Act and its applications.  
Those interested in an expanded introduction to the Act and its history should 
see Robert A. Bohrer & John T. Prince, A Tale of Two Proteins: The FDA’s 
Uncertain Administration of the Orphan Drug Act, 12 Harv. J.L. & Tech. 365 
(1999) (arguing that the FDA’s approach to determining orphan drug sameness for 
protein therapeutics is overly generous to second applicants for orphan drug 
approval) (hereinafter “Bohrer & Prince”); see also David B. Clissold, 
Prescription for the Orphan Drug Act:  The Impact of the FDA’s 1992 Regulations 
and the Latest Congressional Proposals for Reform, 50 Food & Drug L.J. 125 
(1995) (providing an analysis of the U.S. Orphan Drug Act and proposed 
legislative changes to it); see also Gina M. Cavalier, Note, Pushing Parentless 
Pharmaceuticals:  Toward an International Home for “Orphan Drugs” and a Cure for 
“Zebra” Diseases, 27 Law & Pol’y Int’l Bus. 447 (1996) (providing a general 
overview of orphan drug development in the U.S. and Japan, as well as 
recommendations for an EU orphan drug policy). 

   [3] For example, Genzyme’s Ceredase is a therapeutic for Gaucher’s disease, 
which afflicts approximately 20,000 persons in the United States.  The only 
treatment for that very serious disease, Ceredase costs patients an average of 
$145,000 per year, providing Genzyme with revenues of $172 million in 1994.  
Alex Pham, Ceredase Report May Foretell Trouble for Genzyme, Boston Globe, Mar. 
2, 1995, at 36.  The problem of the high cost of orphan drugs is an important 
one, but is beyond the scope of this paper.  So long as market exclusivity and 
profits remain the principal incentive for drug discovery and development, the 
problem of drug cost and access to important therapeutics will be a problem that 
needs to be addressed.  However, since market exclusivity is in fact a principal 
incentive provided in the Orphan Drug Act, the purpose of this paper is to 
critique the FDA’s administration of that incentive in its own terms.

   [4] Abbey S. Meyers, Intellectual Property at the Public-Private Divide: A 
Response, 3 U. Chi. L. Sch. Roundtable 581 (1996) (attributing the success of 
the Orphan Drug Act to its exclusivity incentives while questioning the need for 
broad patents on gene sequences).    

   [5] 21 U.S.C. § 360ee(b)2 (2000).

   [6] See Berlex Labs v. FDA, 942 F. Supp. 19 (D.D.C.1996); Genentech, Inc. v. 
Bowen, 676 F. Supp. 301, 306 (D.D.C. 1987); Bohrer & Prince, supra note 2. 

   [7] Berlex Labs, 942 F. Supp. at 23.

   [8] See The Pink Sheet, Vol. 60; No. 19; Pg. 19  May 11, 1998: 

FDA is planning to update its definitions of orphan drug "sameness" and 
"difference" to address new issues related to biotechnology and biologics, FDA 
Office of Orphan Products Development Director Marlene Haffner said at a May 5 
workshop in Brussels on rare diseases and orphan drugs.
"Same versus different keeps challenging us," Haffner stated at the meeting, 
which was sponsored by the European Foundation for the Advancement of Science. 
"We just discovered some holes in our regulation that we didn't know about even 
last year, so we are going to be republishing some questions."

   [9] See generally Thomas D. Barton, Creative Problem Solving: Purpose, Meaning, 
and Values, 34 Cal. W. L. Rev. 273 (1998).  Barton views problems as a 
disjuncture between environment and desired objectives, with problem solving 
being an effort to get from the initial environment to the desired objective.  
Here the problem is providing adequate incentives and certainty for the 
developers of antibodies for orphan diseases.

   [10] Draft Guidance for Industry - Interpreting Sameness of Monoclonal Antibody 
Products Under the Orphan Drug Regulations, 
http://www.fda.gov/cber/gdlns/orphan.htm (hereinafter “Guidance”) (last visited 
April 23, 2003).

   [11] Clinical superiority is a concept currently employed by the FDA to allow 
sponsors of second orphan drugs that are substantially similar to first-approved 
orphan drugs to nevertheless gain approval.  See Berlex Labs, 942 F. Supp. at 
19.  Clinical superiority is determined by greater effectiveness or fewer side 
effects or, in some cases, by greater patient ease-of-use, such as an oral form 
instead of an injectable one.  See 21 C.F.R. 316.3 (b)(13) et. seq.

   [12] See Guidance, supra note 10.

   [13] Proteins, which may also be referred to as polypeptides, are long chains 
of amino acids, joined together by what chemists refer to as a peptide bond.  
Proteins may also have carbohydrate, or sugar, molecules attached to one or more 
of the amino acids of the protein.  Such sugar-coated proteins are known as 
glycoproteins.  See Bruce Alberts et al., The Molecular Biology of the Cell, 
107, 417 (2d ed. 1989).  

   [14]   See id. at 1007:
Most antigens have a variety of antigenic determinants that stimulate the 
production of antibodies or T cell responses . . . even a single antigenic 
determinant will, in general, activate many clones [different antibody-producing 
cells], each of which produces an antigen-binding site with its own 
characteristic affinity for the determinant . . . . Even . . . [a] . . . 
relatively simple structure . . . can be ‘looked at’ in many ways . . . 
[stimulating] . . . the production of hundreds of species of . . . antibodies, 
each of which is made by a different B cell clone.

   [15] See Johns Hopkins Univ. v. CellPro, 152 F.3d 1342, 1351 n.14 (Fed. Cir. 
1998). 
  
   [16] Specificity is the propensity of an antibody to bind to a particular 
target antigenic determinant and in so doing to distinguish between the target 
and non-targets.   See Alberts et al., supra note 13, at 178 (“[t]his [antigen-
binding site of a monoclonal antibody] will recognize, for example, a particular 
conformation of a defined sequence of five or six amino acid side chains on a 
protein.  Their uniform specificity alone makes monoclonal antibodies much more 
useful than conventional antisera ….”).
  While in Figure 1 the CDR regions appear to be flat, in fact they have a 
complicated three-dimensional structure that binds the target region, or 
epitope, of the antibody’s target antigen.  The amino acids that are linked 
together in the chain of the CDR region provide the three-dimensional structure 
of this antigen binding space and account for the specificity of the antibody’s 
binding to its target.  Two antibodies can bind the same antigen despite 
substantial differences in their CDR regions, so long as the antigen is large 
enough to provide multiple epitopes or antigenic determinants to which an 
antibody can bind.  If the target of an antibody is a protein, it is certain to 
have multiple epitopes to which an antigen can bind.  See Alberts et al., supra 
note 13, at 107.

   [17] Affinity is, more or less, the strength of the binding of the antibody to 
its target.  The higher the affinity the more “tightly” an antibody binds to its 
antigen and the lower the percentage of antibodies that would be released from 
the antigen under particular conditions.  See Alberts et al., supra note 13 at 
1016--17.

   [18] See Richard A. Goldsby et al., Chapter 5: Immunoglobulins: Structure and 
Function in Kuby Immunology 4e, at 
http://www.whfreeman.com/immunology/CH05/kuby05.htm (last visited on April 23, 
2003).

   [19] See id. “Antibody class or subclass is determined by the amino acid 
sequence of the constant region of the antibody, and determines the biological 
activity of that type of antibody.”  

   [20] E.g., IDEC Pharmaceuticals, 
http://www.idecpharm.com/site/science/zevalin.htm (last visited April 23, 2003): 
“Radioimmunotherapies like Zevalin are made by linking monoclonal antibodies – 
engineered in a laboratory to recognize and attach to substances on the surface 
of certain cells – to radioactive isotopes.”  For a blocking antibody such as 
infliximab, the mechanism of action is quite simple, by binding to a portion of 
the target that ordinarily would bind to the target’s natural biological target, 
the blocking antibody prevents that natural binding and the biological response 
that is ordinarily triggered.  For such an antibody, all that matters is that 
the CDR region bind to some portion of the target in a way that obstructs normal 
binding.  For a cytotoxic antibody, the mechanism of action is more complicated.  
Binding to an appropriate target is only the first step, the second step 
requires that the cell to which the target is attached be killed, which can 
occur through triggering an immune response to the target cell or by the 
antibody’s attachment to a cell-killing substance, such as a radio-isotope or 
toxin.

   [21] See Jeffrey P. Kushdan, Protein Patents and the Doctrine of Equivalents: 
Limits on the Expansion of Patent Rights, 6 High Tech. L. J. 109, 115 n.20 
(1991). 

   [22] See S.K. Sharma et al., Human Immune Response to Monoclonal Antibody-
Enzyme Conjugates in ADEPT Pilot Clinical Trial, Cell Biophys. Aug--Dec 1992, at 
21(1-3):109--20; Edward G. Routledge et al., Reshaping Antibodies for Therapy, 
at http://www.path.cam.ac.uk/~mrc7/reshaping/index.html (last visited Apr. 23, 
2003) [hereinafter Routledge et al.).  The reason that murine antibodies produce 
an immune “HAMA” response in humans is that a relatively small number of amino 
acids that differ between the murine and mouse antibody proteins cause the human 
immune system to respond to the murine antibody as non-self, or foreign.  See M. 
Clark, Antibody Humanization for Therapeutic Applications, at 
http://www.path.cam.ac.uk/~mrc7/humanisation/index.html (last visited Apr. 23, 
2003). 

   [23] Routledge et al., supra note 22.

   [24] Reopro is approved for the prevention of restenosis in conjunction with 
angioplasty, while Remicade is approved for Crohn’s disease and arthritis.  See 
Reopro (Abciximab), at http://www.centocor.com/cgi-bin/site/products/prod_reopro.cgi 
(last visited Apr. 23, 2003).  Remicade has orphan exclusivity for Crohn’s disease 
and juvenile arthritis; FDA List of Orphan Designations and Approvals, at   
http://www.fda.gov/orphan/designat/list.xls (last visited July 23, 2003) 
[hereinafter “Orphan List”].

   [25] Approved for non-Hodgkin’s lymphoma.  See IDEC Pharmaceuticals, at 
http://www.idecpharm.com/site/science/rituxan.htm (last visited Apr. 22, 2003).  
Rituximab received orphan exclusivity for a subpopulation of non-Hodgkin’s 
lymphoma patients.  See Orphan List, supra note 24.

   [26] Herceptin is approved for metastatic breast cancer expressing the Her2 
receptor.  See Genentech: In Business for Life, at 
http://www.gene.com/gene/products/information/oncology/herceptin/index.jsp (last 
visited Apr. 23, 2003).  

   [27] See M. Clark, The Therapeutic Antibody Human Homology Project (TAHHP), at 
http://www.path.cam.ac.uk/~mrc7/humanisation/TAHHP.html, (last visited Apr. 23, 
2003).

   [28] See T.M. Illidge & M.C. Bayne, Antibody Therapy of Lymphoma, (Abstract) at 
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=11585011&dopt=Abstract 
(last visited Apr. 23, 2003) (discussing the development of Rituxan (riotuximab), 
Bexxar (tositumomab) and Zevalin (ibritumomab);  see also Brief, Washington Drug 
Letter, Jan. 6, 2003, at Vol. 35, No. 1.

   [29] See David Hamilton, “Abbott Enters Arthritis-Drug War with FDA Approval of 
Humira,” The Wall St. J., Jan. 2, 2003, at D4.  Hamilton notes that Abbot’s 
anti-TNF antibody will be competing against Centocor’s Remicade anti-TNF 
antibody and Immunex’s Enbrel, a soluble receptor that works to block TNF.

   [30] Id.

   [31] See F. Ciardiello & G. Tortora, A Novel Approach in the Treatment of 
Cancer: Targeting the Epidermal Growth Factor Receptor, Clinical Cancer Research 
29958-70, Oct. 7, 2001, available at 
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=11595683&dopt=Abstract 
(last visited Apr. 23, 2003) (“At least five blocking monoclonal antibodies have 
been developed against the EGFR.”).

   [32] Or populations greater than 200,000 for which commercial development would 
not be undertaken because there is no reasonable expectation that the costs of 
development would be recovered.  See supra note 5.  The incentives, in addition 
to the seven year period of exclusivity for an approved orphan drug in its 
orphan market, 21 U.S.C. § 360cc (2000), include possible grants to defray the 
costs of clinical testing, 21 U.S.C. § 360ee(a) (2000), tax credits, 26 U.S.C. § 
38(b) 12 (2000), and assistance with the design of clinical trials, 21 U.S.C. § 
360aa (2000).

   [33] 21 U.S.C. §360bb (2003).

   [34] See Bohrer & Prince, supra note 2.  The fact that the prize of exclusivity 
goes to the winner of the race to marketing approval for a particular protein or 
antibody is a major reason why second entrants seek approval for a competitive 
entry into a small market and why the determination of sameness is so important.

   [35] 21 U.S.C. §360cc.

   [36] See Genentech, Inc. v. Bowen, 676 F. Supp. 301, 306 (D.D.C. 1987).

   [37] Of particular relevance is the definition of “same drug” for proteins:  
(ii) If it is a drug composed of large molecules (macromolecules), a drug that 
contains the same principal molecular structural features (but not necessarily 
all of the same structural features) and is intended for the same use as a 
previously approved drug, except that, if the subsequent drug can be shown to be 
clinically superior, it will not be considered to be the same drug. This 
criterion will be applied as follows to different kinds of macromolecules: 

(A) Two protein drugs would be considered the same if the only differences in 
structure between them were due to post-translational events or infidelity of 
translation or transcription or were minor differences in amino acid sequence; 
other potentially important differences, such as different glycosylation 
patterns or different tertiary structures, would not cause the drugs to be 
considered different unless the differences were shown to be clinically 
superior.
21 C.F.R. § 316.3(b)(13)(ii)(A) (2003).

   [38] The attachment of carbohydrates to a protein is known as glycosylation and 
the resulting protein is a glycoprotein.

   [39] See Genentech, 676 F. Supp. at 306. Genentech’s orphan drug approval for 
recombinant human growth hormone was followed eight months later by the FDA’s 
approval of Eli Lilly’s version of recombinant human growth hormone, which 
differed from Genentech’s by a single amino acid.  Genentech’s attempt to have 
the FDA’s action overturned by the courts was unsuccessful, but the FDA did 
later revise their regulations, see 21 C.F.R. § 316.3, in an attempt to avoid a 
repetition of the Genentech-Lilly debacle. 

   [40] 942 F. Supp. 19 (D.D.C. 1996).

   [41] See id.; see also Bohrer & Prince, supra note 2, at 395--96.

   [42] See Guidance, supra note 10, §§ IV and V.

   [43] Id.  § IV(B).

   [44] See id.; see also Anthony S. Lubiniecki, Sc.D., Vice President of Smith 
Kline Beecham, Letter to the FDA (Oct. 22, 1999), available at 
http://www.fda.gov/ohrms/dockets/dailys/102999/c0004.pdf  (last visited Oct. 18, 
2002).  Dr. Lubiniecki expresses a number of concerns about the FDA’s position, 
including the concern central to this Article that:
  
Two monoclonal antibodies with very, different amino acid sequences within the 
CDRS could nonetheless bind to the same epitope with the same affinity, serve 
the same function, and provide the same therapeutic effect (see Kashmiri et al, 
1999, copy attached). If "sameness" was defined solely on the basis of the amino 
acid sequence within the CDRS, FDA would apparently be forced to approve both 
monoclonal antibodies. Thus, FDA's proposed approach could seriously undermine 
the primary incentive for orphan drug development intended by and provided by 
Congress in the Orphan Drug Act ­ exclusive marketing rights.
  
  It is the purpose of this Article to propose a solution to the problem created 
by the FDA’s Guidance and recognized by Dr. Lubiniecki.

   [45] See Guidance, supra note 10, § IV(B).

   [46] Id.

   [47] See Bohrer & Prince, supra note 2, at 365.  In this earlier article, the 
difference between the scope of exclusivity for traditional small-molecule drugs 
and protein drugs was analyzed, however the problem of antibody therapeutics was 
not addressed.  

   [48] See A Science Odyssey:  People and Discoveries, at 
http://www.pbs.org/wgbh/aso/databank/entries/dm28pe.html (last visited Apr. 23, 
2003).   Fleming noticed that staph bacteria were not growing in the vicinity of 
a mold that had contaminated one of the cultures in his laboratory.  His 
observation led him to the breakthrough discovery of the first powerful 
antibiotic.

   [49] See Creating Medicines, at http://science.gsk.com/m2m/randd.pdf (last 
visited Apr. 23, 2003) (GlaxoSmith Kline document providing an overview of drug 
development).  An example of target-based drug discovery can be seen in the 
field of anti-hypertensives, for example in the number of ACE inhibitors that 
all are aimed at blocking the workings of a particular target “angiotensin 
converting enzyme” or ACE.  See D. MacAreavey & J. Robertson, Angiotensin 
Converting Enzyme Inhibitors and Moderate Hypertension 40 Drugs 326 (1990) 
(discussing the numerous different drugs aimed at that target as well as the 
advantages and disadvantages of particular ACE inhibitors).

   [50] See James Thuo Gathii, Rights, Patents, Markets and the Global Aids 
Pandemic, 14 Fla. J. Int'l L. 261, 334--40 (2002) (discussing the FDA testing 
and review process in the context of barriers to competition in the global 
pharmaceutical marketplace).

   [51] See U.S. Congress, Office of Technology Assessment, Pharmaceutical R&D: 
Costs, Risks and Rewards, OTA-H-522 (1993), available at 
http://www.wws.princeton.edu/~ota/ns20/alpha_f.html (last visited Apr. 3, 2003).  
In that study the OTA found that: 
  
The discovery and development of NCEs [new chemical entities] is the heart of 
pharmaceutical R&D, because the developers of follow-on or generic products 
build on the knowledge produced in the course of developing them.  The market 
for the compound and all its follow-on products or generic copies in future 
years rests on the R&D that led to its initial introduction to the 
market . . . .

. . . .

NCE’s comprise two poorly-defined subcategories:  pioneer drugs and “me-too” 
drugs.  Pioneer NCEs have molecular structures or mechanisms of action that are 
very different from all previously existing drugs in a therapeutic area.  The 
first compound to inhibit the action of a specific enzyme, for example, is a 
pioneer drug.  Me-too drugs are introduced after the pioneer and are similar but 
not identical to the pioneer compounds in molecular structure and mechanism of 
action.  Many me-too drugs are developed through deliberate imitation of the 
pioneer compound and have a shorter and more certain discovery period.  But, the 
R&D cost advantage gained by imitation is typically met by a reduction in 
potential dollar returns by being a late entrant to the market.
Id. at 6--7 [citations omitted].

   [52] American Hospital Formulary Service Drug Monographs 24:06.08 (1997) 
(discussing Simvastatin, Pravastatin, Lovastatin and Atorvastatin).

   [53] American Hospital Formulary Service Drug Monographs 28:16.04 (1997).

   [54] American Hospital Formulary Service Drug Monographs 24:24, 24:32.04 
(1997).

   [55] See U.S. Congress, Office of Technology Assessment, supra note 51.

   [56] See id.  As the OTA study notes, the distinction between pioneer and “me-
too” can be “fuzzy” when the development of the first entry based on a new 
target or pathway is competitive, with multiple companies in a race to be the 
first to win approval.  In these cases, the losers may have costs just as high 
as that of the successful pioneer.

   [57] See supra text accompanying notes 6--8.

   [58] See Guidance, supra note 10, § IV(B).

   [59] See Bohrer & Prince, supra note 2, at 395--401, which deals at length with 
the interferon-ß orphan drug controversy.  In that case, there was almost no 
amino acid variation between the two competitor orphan drug entrants, but 
differences in administration lead to adverse effect differences that produced a 
finding of “clinical superiority.”

   [60] See Guidance, supra note 10, § IV(A)--(B).

   [61] To summarize the distinction between other kinds of protein therapeutics 
and antibodies, the relationship between a traditional protein and its receptor 
requires conservation of amino acid sequence, while the relationship between 
antigen and antibody permits antibody diversity.  See Alberts et al., supra note 
13.  The FDA Guidance itself acknowledges that “[b]ecause of the many processes 
involved in generating antibody diversity, it is unlikely that independently 
derived monoclonal antibodies with the same antigen specificity will have 
identical amino acid sequences.” Guidance, supra note 13, § II. However, the FDA 
did not recognize the significance of this basic scientific fact for orphan drug 
exclusivity.

   [62] See sources cited supra notes 28--29 (discussing multiple antibodies to 
CD20 (lymphoma) and TNF (inflammatory disease)).

   [63] See Comment of Antisoma on the FDA’s Proposed Guidelines, at 
http://www.fda.gov/ohrms/dockets/dailys/092399/c00001.pdf  (last visited on 
February 21, 2003):
  
Based on the above discussion points our comment on the draft guidelines would 
be to allow the remit of Orphan Status Protection to include an element of 
functionality. That is when an antibody is against the same target antigen and 
has the same effecter it should be regarded as being the same for orphan drug 
purposes irrespective of the binding site sequence of the antibody.
  
  In the course of my research for this Article, I found that Antisoma’s brief 
comment on the guidelines supports the notion that biotechnology companies would 
support the policy argument made here.  But see Comment of Centocor, at 
http://www.fda.gov/ohrms/dockets/dailys/100899/c000002.pdf (last visited Feb. 
21, 2003) (“We have carefully reviewed the draft and have no suggestions for 
changes. The definition of “sameness” based on the complementarity determining 
regions as the principal structural feature of the antibody molecule provides 
reasonable protection of orphan drugs and has a sound scientific basis.”).

   [64] See supra text accompanying notes 16--19.

   [65] See supra text accompanying note 8.

   [66] 56 Fed. Reg. 3338, §1 para. 3.

   [67] See M. Clark, Antibody Humanization for Therapeutic Applications, at 
http://www.path.cam.ac.uk/~mrc7/humanisation/index.html (last visited on Apr. 
23, 2003) for a general discussion of antibody humanization techniques and 
issues.

   [68] See Description of Remicade Drug at 
http://www.centocor.com/cgi-bin/site/products/prod_remicade.cgi (last visited 
July 20, 2003).

   [69] See Description of Humira Drug at http://www.rxabbott.com/pdf/humira.pdf 
(last visited July 20, 2003).  

   [70] See id.

   [71] See List of All Orphan Products Approved at 
http://www.fda.gov/orphan/designat/allap.rtf (last visited July 17, 2003).

   [72] See supra note 68.

   [73] American Hospital Formulary Service Drug Monographs 92:00 (Remicade is 
listed under Infliximab, Humira is listed under Adalimumab).

   [74] See supra text accompanying note 16.  Two IgG antibodies targeting the 
same antigen without attached isotopes or toxins would be the same, while two 
radioactively tagged IgG antibodies would be considered the same, regardless of 
whether different isotopes were used, unless the second was shown to offer a 
significant therapeutic advantage warranting a finding of clinical superiority.

Vol. V  The Columbia Science and Technology Law Review  2004