[Federal Register: December 19, 2005 (Volume 70, Number 242)]
[Notices]               
[Page 75180-75198]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr19de05-59]                         

-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Food and Drug Administration

[Docket No. 1980N-0208]

 
Biological Products; Bacterial Vaccines and Toxoids; 
Implementation of Efficacy Review; Anthrax Vaccine Adsorbed; Final 
Order

AGENCY: Food and Drug Administration, HHS.

ACTION: Notice.

-----------------------------------------------------------------------

SUMMARY: The Food and Drug Administration (FDA) proposed, among other 
things, to classify Anthrax Vaccine Adsorbed (AVA) on the basis of 
findings and recommendations of the Panel on Review of Bacterial 
Vaccines and Toxoids (the Panel) on December 13, 1985. The Panel 
reviewed the safety, efficacy, and labeling of bacterial vaccines and 
toxoids with standards of potency, bacterial antitoxins, and immune 
globulins. After the initial final rule and final order was vacated by 
the United States District Court for the District of Columbia on 
October 27, 2004, FDA published a new proposed rule and proposed order 
on December 29, 2004. The purpose of this final order is to categorize 
AVA according to the evidence of its safety and effectiveness, thereby 
determining if it may remain licensed and on the market; issue a final 
response to recommendations made in the Panel's report, and; respond to 
comments on the previously published proposed order. The final rule and 
final order concerning bacterial vaccines and toxoids other than AVA is 
published elsewhere in this issue of the Federal Register.

DATES: The final order on categorization of AVA is effective December 
19, 2005.

FOR FURTHER INFORMATION CONTACT: Kathleen Swisher, Center for Biologics 
Evaluation and Research (HFM-17), Food and Drug Administration, 1401 
Rockville Pike, Suite 200N, Rockville, MD 20852-1448, 301-827-6210.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Introduction
II. Background
    A. General Description of the ``Efficacy Review'' for Biological 
Products Licensed Before July 1972
    B. The December 1985 Proposal
    C. Additional Proceedings Following the December 1985 Proposal
III. Categorization of Anthrax Vaccine Adsorbed--Final Order
    A. Efficacy of Anthrax Vaccine Adsorbed
    B. Safety of Anthrax Vaccine Adsorbed
    C. The Panel's General Statement: Anthrax Vaccine, Adsorbed, 
Description of Product
    D. The Panel's Specific Product

[[Page 75181]]

Review: Anthrax Vaccine Adsorbed: Efficacy
    E. The Panel's Specific Product Review: Anthrax Vaccine Adsorbed: 
Labeling
IV. Comments on the December 2004 Anthrax Vaccine Adsorbed (AVA) 
Proposed Order and FDA's Responses
    A. Comments Supporting Placing AVA into Category I
    B. Comments on the Evidence of Safety and Effectiveness of AVA
1. Brachman Study
2. CDC Surveillance Data
3. CDC Open Label Safety Study
4. DoD Pilot Study and Safety Data
5. Long-Term Safety Monitoring and Additional Studies
    C. Comments Describing Adverse Events
1. Review of Adverse Event Reports Submitted to the Docket
2. Summary of Adverse Event Reports Submitted to the Docket
    D. Comments on the Vaccine Used in the Studies
    E. Comments about Allegedly Contaminated Vaccine and Inspectional 
Observations
    F. Comments on Labeling
    G. Additional Comments
    H. Comments on Matters Outside the Scope of this Proceeding
V. FDA's Responses to Additional Panel Recommendations
VI. References

I. Introduction

    Biological products licensed before July 1972 are subject to a 
review procedure described in Sec.  601.25 (21 CFR 601.25). AVA was 
licensed before July 1972. The purpose of this document is to: (1) 
Categorize AVA under Sec.  601.25 according to the evidence of its 
safety and effectiveness, thereby determining if it may remain licensed 
and on the market, (2) issue a final response to recommendations made 
in the Panel's report, and (3) respond to comments on the proposed 
order (69 FR 78281, December 29, 2004).

II. Background

A. General Description of the ``Efficacy Review'' for Biological 
Products Licensed Before July 1972

    In 1972, in an effort to assure that regulatory standards for drugs 
and biological products were harmonized, the National Institutes of 
Health (NIH) announced a review of all licensed biological products (37 
FR 5404, March 15, 1972). However, on July 1, 1972, NIH's Division of 
Biologics Standards, which had been charged with administering and 
enforcing the licensing provisions of the Public Health Service Act, 
was transferred to FDA (37 FR 12865, June 29, 1972). FDA then assumed 
responsibility for reviewing the previously licensed biological 
products. In the Federal Register of February 13, 1973 (38 FR 4319), 
FDA issued procedures for the review of the safety, effectiveness, and 
labeling of biological products licensed before July 1, 1972. This 
process was eventually codified in Sec.  601.25 (38 FR 32048 at 32052, 
November 20, 1973). Under the panel assignments published in the 
Federal Register of June 19, 1974 (39 FR 21176), FDA assigned each 
review of a biological product to one of the following groups: (1) 
Bacterial vaccines and bacterial antigens with ``no U.S. standard of 
potency,'' (2) bacterial vaccines and toxoids with standards of 
potency, (3) viral vaccines and rickettsial vaccines, (4) allergenic 
extracts, (5) skin test antigens, and (6) blood and blood derivatives.
    Under Sec.  601.25, FDA assigned the initial review of each of the 
six biological product categories to a separate independent advisory 
panel consisting of qualified experts. Each panel was charged with 
preparing for the Commissioner of Food and Drugs an advisory report 
which was to: (1) Evaluate the safety and effectiveness of the 
biological products for which a license had been issued, (2) review 
their labeling, and (3) identify the biological products that are safe, 
effective, and not misbranded. Each advisory panel report was also to 
include recommendations classifying the products reviewed into one of 
three categories.
     Category I, designating those biological products 
determined by the panel to be safe, effective, and not misbranded.
     Category II, designating those biological products 
determined by the panel to be unsafe, ineffective, or misbranded.
     Category III, designating those biological products 
determined by the panel not to fall within either Category I or 
Category II on the basis of the panel's conclusion that the available 
data were insufficient to classify such biological products, and for 
which further testing was therefore required. Category III products 
were assigned to one of two subcategories. Category IIIA products were 
those that would be permitted to remain on the market pending the 
completion of further studies. Category IIIB products were those for 
which the panel recommended license revocation on the basis of the 
panel's assessment of potential risks and benefits.
    In its report, the panel could also include recommendations 
concerning any condition relating to active components, labeling, tests 
appropriate before release of products, product standards, or other 
conditions necessary or appropriate for a biological product's safety 
and effectiveness.
    In accordance with Sec.  601.25, after reviewing the conclusions 
and recommendations of the review panels, FDA would publish in the 
Federal Register a proposed order containing: (1) A statement 
designating the biological products reviewed into Categories I, II, 
IIIA, or IIIB, (2) a description of the testing necessary for Category 
IIIA biological products, and (3) the complete panel report. Under the 
proposed order, FDA would propose to revoke the licenses of those 
products designated into Category II and Category IIIB. After reviewing 
public comments, FDA would publish a final order on the matters covered 
in the proposed order.

B. The December 1985 Proposal

    The Panel was convened in a July 12, 1973, organizational meeting, 
which was followed by multiple working meetings until February 2, 1979. 
The Panel completed its final report in August 1979. In that report, 
the Panel found that AVA, manufactured by Michigan Department of Public 
Health (MDPH, now BioPort), License No. 99,\1\ was safe and effective 
for its intended use and recommended that the vaccine be placed into 
Category I. The Panel based its evaluation of the safety and efficacy 
of AVA on two studies: The Brachman study, a well-controlled field 
study conducted in the 1950s (Ref. 1), and an open label safety study 
conducted by the National Center for Disease Control (CDC, now the 
Centers for Disease Control and Prevention) (50 FR 51002 at 51058, 
December 13, 1985). The Panel also considered surveillance data on the 
occurrence of anthrax disease in the United States in at-risk 
industrial settings as supportive of the effectiveness of the vaccine 
(50 FR 51002 at 51059, December 13, 1985).
---------------------------------------------------------------------------

    \1\On December 17, 1965, the company name was changed from the 
Division of Laboratories, Michigan Department of Health to the 
Bureau of Laboratories, Michigan Department of Public Health. On 
April 10, 1979, the name was changed to the Michigan Department of 
Public Health. On May 14, 1996, the name was changed to the Michigan 
Biologics Products Institute. On November 11, 1998, FDA accepted a 
name change to BioPort Corporation (BioPort) with an accompanying 
license number change to 1260.
---------------------------------------------------------------------------

    In the Federal Register of December 13, 1985 (50 FR 51002), FDA 
issued a proposed rule that contained the full Panel report on 
bacterial vaccines and toxoids with standards of potency,

[[Page 75182]]

including the anthrax vaccine,\2\ and FDA's response to the 
recommendations of the Panel (the December 1985 proposal). In the 
December 1985 proposal, FDA proposed regulatory categories (Category I, 
Category II, or Category IIIB as defined previously in this document) 
for each bacterial vaccine and toxoid reviewed by the Panel, and 
responded to other recommendations made by the Panel. FDA agreed with 
the Panel's recommendation and proposed to place AVA into Category I.
---------------------------------------------------------------------------

    \2\In addition to publication in the Federal Register of 
December 13, 1985 (50 FR 51002), the full Panel report is available 
on FDA's Web site at http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.fda.gov/ohrms/dockets/default.htm 

(Docket No. 1980N-0208). A copy of the Panel report is also 
available at the Division of Dockets Management, 5630 Fishers Lane, 
rm. 1061, Rockville, MD 20852.
---------------------------------------------------------------------------

    The public was provided 90 days to submit comments in response to 
the December 1985 proposal. FDA received four letters of comments in 
response to the December 1985 proposal, but none of those comments 
pertained to AVA. We discuss them in a final rule and final order 
concerning bacterial vaccines and toxoids other than AVA published 
elsewhere in this issue of the Federal Register.
    FDA addressed the review and reclassification of bacterial vaccines 
and toxoids classified into Category IIIA through a separate 
administrative procedure (see the Federal Register of May 15, 2000 (65 
FR 31003), and May 29, 2001 (66 FR 29148)).

C. Additional Proceedings Following the December 1985 Proposal

    On October 12, 2001, a group of individuals filed a citizen 
petition requesting that FDA find AVA, as currently manufactured by 
BioPort, ineffective for its intended use, classify the product as 
Category II, and revoke the license for the vaccine. The petitioners 
complained that the December 1985 proposal that placed AVA into 
Category I had not been finalized. FDA responded separately in a 
written response to the petitioners on August 28, 2002 (Docket No. 
2001P-0471).
    In March 2003, six plaintiffs, known as John and Jane Doe 1 through 
6, filed suit in the U.S. District Court for the District of Columbia 
(the Court) asking the Court to enjoin the Anthrax Vaccine Immunization 
Program (AVIP) of the Department of Defense (DoD), and to declare AVA 
an investigational drug when used for protection against inhalation 
anthrax. On December 22, 2003, the Court issued a preliminary 
injunction enjoining inoculations under the AVIP in the absence of 
informed consent or a Presidential waiver of informed consent (see 
Sec.  50.23 (21 CFR 50.23)). Doe v. Rumsfeld, 297 F.Supp. 2d 119 
(D.D.C. 2003).
    In the Federal Register of January 5, 2004 (69 FR 255), FDA 
published a final rule and final order amending the biologics 
regulations and categorizing certain biological products in response to 
the report and recommendations of the Panel. The final order placed AVA 
into Category I. Following FDA's issuance of the final rule and final 
order, on January 7, 2004, the Court lifted the preliminary injunction 
except as it applied to the six Doe plaintiffs. Doe v. Rumsfeld, 297 
F.Supp. 2d 200 (D.D.C. 2004).
    On October 27, 2004, the Court issued a memorandum opinion vacating 
and remanding the January 2004 final rule and final order to FDA for 
reconsideration, requiring an additional opportunity for comment. Doe 
v. Rumsfeld, 341 F.Supp. 2d 1 (D.D.C. 2004). On December 29, 2004 (69 
FR 78280), FDA published a withdrawal of the January 5, 2004, final 
rule and final order. Concurrently with the withdrawal of the final 
rule and final order, FDA published again a proposed rule and proposed 
order (69 FR 78281) (the December 2004 proposal) to provide notice and 
to give interested persons an opportunity to comment on FDA's proposals 
relating to bacterial vaccines and toxoids classified into Category I, 
Category II, and Category IIIB, including AVA. In the December 2004 
proposal, FDA reopened the comment period for 90 days on the entire 
Bacterial Vaccines and Toxoids efficacy review document.
    Most of the comments received in response to the December 2004 
proposal pertained to the anthrax vaccine (AVA). We provide a response 
to comments about AVA under section IV of this document. A discussion 
of comments to the December 2004 proposal concerning bacterial vaccines 
and toxoids other than AVA is provided in a final rule and final order 
published elsewhere in this issue of the Federal Register.

III. Categorization of Anthrax Vaccine Adsorbed--Final Order

    After review of the comments and finding no additional scientific 
evidence to alter the proposed categorization, FDA accepts the Panel's 
recommendation and adopts Category I as the final category for AVA and 
determines AVA to be safe and effective and not misbranded.
    In this section of this document, we describe the data supporting 
our conclusion that AVA is safe and effective for its labeled 
indication to protect individuals at high risk for anthrax disease. 
Anthrax disease can be fatal despite appropriate antibiotic therapy. We 
also discuss points of disagreement with certain statements in the 
Panel's report.
    In order to provide clarity to the reader, we use the following 
terms to refer to studies relevant to this final order. The versions of 
vaccine used in these studies reflect the optimization of anthrax 
vaccine during product and clinical development.
    1. Brachman study--The Brachman study was an adequate and well-
controlled clinical study conducted from 1954 to 1959 to evaluate the 
effectiveness of the anthrax vaccine. The vaccine used in the Brachman 
study (the DoD vaccine) was supplied by Dr. G. G. Wright and associates 
of the U.S. Army Chemical Corps., Fort Detrick, Frederick, MD.
    2. CDC open label safety study--The CDC open label safety study was 
conducted from 1966 to 1971. Merck Sharp & Dohme (MSD) manufactured 
anthrax vaccine (DoD/MSD vaccine) under contract to DoD in 1960 and 
1961. The Michigan Department of Public Health (MDPH) also manufactured 
anthrax vaccine (DoD/MDPH/AVA) under contract to DoD starting in the 
mid-1960s. CDC used one lot of DoD/MSD vaccine and one lot of DoD/MDPH/
AVA vaccine in the first year of the CDC open label safety study, but 
only DoD/MDPH/AVA vaccine was used for the remainder of that study. The 
vaccine manufactured by MDPH was licensed by the NIH, Bureau of 
Biologics, in November 1970 as AVA. MDPH subsequently underwent a name 
change to Michigan Biologic Products Institute (MBPI) and later, 
BioPort Corporation (BioPort).
    3. DoD pilot study--The DoD pilot study was conducted from 1996 to 
1999. The purpose of the study was to make an initial assessment of the 
effects that alternative immunization schedules and/or an alternative 
route of administration may have on the safety and immunogenicity of 
AVA. The DoD pilot study used the licensed DoD/MDPH/AVA vaccine.

A. Efficacy of Anthrax Vaccine Adsorbed

    The Brachman study was conducted in four textile mills where, prior 
to initiation of the study, the yearly average number of human anthrax 
cases was 1.2 cases per 100 mill employees. These textile mills were 
located in the northeastern United States and processed imported goat 
hair. The study included 1,249 workers from these

[[Page 75183]]

mills. Of these 1,249 workers, 379 received anthrax vaccine, 414 
received placebo, 116 received incomplete inoculations of either 
anthrax vaccine or placebo, and 340 received no treatment but were 
monitored for the occurrence of anthrax disease as an observational 
group. The Brachman study used DoD vaccine administered subcutaneously 
at 0, 2, and 4 weeks and 6, 12, and 18 months. During the study, 26 
cases of anthrax were reported across the four mills: 5 inhalation and 
21 cutaneous anthrax cases. Of the five inhalation anthrax cases (four 
of which were fatal), two received placebo, three were in the 
observational group, and none received anthrax vaccine. Of the 21 
cutaneous anthrax cases, 15 received placebo, 3 were in the 
observational group, and 3 received anthrax vaccine. Of the three cases 
in the vaccine group, one case occurred just prior to administration of 
the third dose, one case occurred 13 months after the individual 
received the third of the six doses (but no subsequent doses), and one 
case occurred prior to receiving the fourth dose of vaccine.
    In its report, the Panel stated that the Brachman study results 
demonstrate ``a 93 percent (lower 95 percent confidence limit = 65 
percent) protection against cutaneous anthrax'' (emphasis supplied) and 
that ``inhalation anthrax occurred too infrequently to assess the 
protective effect of vaccine against this form of the disease'' (50 FR 
51002 at 51058, December 13, 1985). We do not agree with the Panel's 
statement that the protection was limited to cutaneous anthrax cases. 
The Brachman study's comparison between anthrax cases in the placebo 
and vaccine groups included both inhalation and cutaneous anthrax 
cases. Accordingly, the calculated effectiveness of the vaccine to 
prevent both types of anthrax disease combined was 92.5 percent (lower 
95 percent confidence interval = 65 percent) as described in the 
Brachman, et al. report (Ref. 1). We agree that the cases of inhalation 
anthrax reported in the course of the Brachman study, if analyzed 
separately, are too few to support a meaningful statistical conclusion. 
However, the Brachman study's analysis of the effectiveness of the 
vaccine appropriately included all cases of anthrax disease that 
occurred in individuals who received at least three doses of vaccine or 
placebo and were on schedule for the remaining doses of the six-dose 
schedule regardless of the route of exposure or manifestation of 
disease, and was not limited to cutaneous cases. Thus, the study 
supports AVA's indication for active immunization against Bacillus 
anthracis, independent of the route of exposure.
    As stated previously in this document, the Panel also considered 
epidemiological data--which we refer to as the CDC surveillance data--
on the occurrence of anthrax disease in at-risk industrial settings 
collected by the CDC and summarized for the years 1962 to 1974, as 
supportive of the effectiveness of AVA. In that time period, 
individuals received either DoD/MDPH/AVA vaccine or an earlier version 
of anthrax vaccine. The Panel explained,
    Twenty-seven cases of anthrax disease were identified. Three 
cases were not mill employees but worked in or near mills; none of 
these cases had been vaccinated. Twenty-four cases were mill 
employees; three were partially immunized (one with 1 dose, two with 
2 doses); the remainder (89 percent) were unvaccinated. Therefore, 
no cases have occurred in fully vaccinated subjects while the risk 
of infection has continued. These observations lend further support 
to the effectiveness of this product.
(50 FR 51002 at 51058, December 13, 1985).
    In 1998, the DoD initiated the Anthrax Vaccine Immunization 
Program, calling for mandatory vaccination of service members. 
Thereafter, questions about the vaccine caused the U.S. Congress to 
direct DoD to support an independent examination of AVA by the 
Institute of Medicine (IOM).\3\ The IOM committee was charged with 
reviewing data regarding the efficacy and safety of the currently 
licensed anthrax vaccine--Anthrax Vaccine Adsorbed (AVA)--and assessing 
the efforts to resolve manufacturing issues and resume production and 
distribution of vaccine. The committee in its published report 
concluded that AVA, as licensed, is an effective vaccine to protect 
humans against anthrax, including inhalation anthrax (Ref. 2). FDA 
agrees with the report's finding that certain studies in humans and 
animal models support the conclusion that AVA is effective against B. 
anthracis strains that are dependent upon the anthrax toxin as a 
mechanism of virulence, regardless of the route of exposure.\4\ 
However, our review of AVA, is independent of the IOM's review. We 
discuss later in this document comments that we received related to the 
IOM review.
---------------------------------------------------------------------------

    \3\In October 2000, the Institute of Medicine (IOM) convened the 
Committee to Assess the Safety and Efficacy of the Anthrax Vaccine. 
In March 2002, the Committee issued its report: The Anthrax Vaccine: 
Is It Safe? Does It Work? (Ref. 2). The report concluded that the 
vaccine is acceptably safe and effective in protecting humans 
against anthrax.
    \4\For example: The Brachman study (Ref. 1); the CDC 
surveillance data described in the December 1985 proposal; Fellows 
(2001) (Ref. 3); Ivins (1996) (Ref. 4); and Ivins (1998) (Ref. 5).
---------------------------------------------------------------------------

B. Safety of Anthrax Vaccine Adsorbed

    CDC conducted the CDC open label safety study under an 
investigational new drug application (IND) between 1966 and 1971 in 
which approximately 7,000 persons, including textile employees, 
laboratory workers, and other at-risk individuals, were vaccinated with 
DoD/MDPH/AVA vaccine\5\ and monitored for adverse reactions to 
vaccination. The vaccine was administered in 0.5-mL doses according to 
a 0-, 2-, and 4-week initial dose schedule followed by additional doses 
at 6, 12, and 18 months, with annual boosters thereafter. Several lots 
(approximately 15,000 doses) of DoD/MDPH/AVA vaccine were used in this 
study period. In its report, the Panel found that the CDC data 
``suggests that this product is fairly well tolerated with the majority 
of reactions consisting of local erythema and edema. Severe local 
reactions and systemic reactions are relatively rare'' (50 FR 51002 at 
51059).
---------------------------------------------------------------------------

    \5\In addition, one lot of the DoD/MSD vaccine was used during 
the CDC open label safety study.
---------------------------------------------------------------------------

    Subsequent to the publication of the Panel's recommendations, from 
1996 to 1999, DoD conducted the DoD pilot study, a small, randomized 
clinical study of AVA, administered by alternative route and schedules, 
compared to the vaccine administered according to the approved 
labeling. Safety data from the group that received the vaccine 
according to the labeling as well as post-licensure adverse event 
surveillance data available from the Vaccine Adverse Event Reporting 
System (VAERS), which FDA regularly reviews, further support the safety 
of AVA. These data provided the basis for labeling revisions approved 
by FDA in January 2002 (Ref. 6) to better describe the types and 
severities of adverse events associated with administration of AVA.

C. The Panel's General Statement: Anthrax Vaccine, Adsorbed, 
Description of Product

    The Panel report states:
    Anthrax vaccine is an aluminum hydroxide adsorbed, protective, 
proteinaceous, antigenic fraction prepared from a nonproteolytic, 
nonencapsulated mutant of the Vollum strain of Bacillus anthracis. 
(50 FR 51002 at 51058).
    The Panel's description of the anthrax vaccine has an inaccuracy. 
While the B. anthracis strain used in the manufacture of AVA is the 
nonproteolytic, nonencapsulated strain identified in the Panel report, 
it is not a mutant of the Vollum strain but was derived from a B. 
anthracis culture originally isolated from a case of bovine anthrax in 
Florida.

[[Page 75184]]

D. The Panel's Specific Product Review: Anthrax Vaccine Adsorbed: 
Efficacy

    The Panel report states:
    3. Analysis--a. Efficacy--(2) Human. The vaccine manufactured by 
the Michigan Department of Public Health has not been employed in a 
controlled field trial. A similar vaccine prepared by Merck Sharp & 
Dohme for Fort Detrick was employed by Brachman * * * in a placebo-
controlled field trial in mills processing imported goat hair * * *. 
The Michigan Department of Public Health vaccine is patterned after 
that of Merck Sharp & Dohme with various minor production changes.
(50 FR 51002 at 51059, December 13, 1985).
    FDA found that contrary to the Panel's statement, the vaccine used 
in the Brachman study was not manufactured by MSD, but instead this 
vaccine was manufactured by DoD and provided to Dr. Brachman by Dr. G. 
G. Wright of Fort Detrick, U.S. Army, DoD (Ref. 1). The DoD vaccine 
used in the Brachman study was manufactured using an aerobic culture 
method (Ref. 7). Subsequent to the Brachman study, DoD modified the 
vaccine's manufacturing process to, among other things, optimize 
production of a stable and immunogenic formulation of vaccine antigen 
and increase the scale of manufacture. In the early 1960s (after the 
Brachman study), DoD entered into a contract with MSD to standardize 
the manufacturing process for large-scale production of the anthrax 
vaccine and to produce anthrax vaccine using an anaerobic method.
    Thereafter, in the 1960s, DoD entered into a similar contract with 
MDPH to further standardize the manufacturing process and to scale up 
production for further clinical testing and immunization of persons at 
risk of exposure to anthrax. This DoD-MDPH contract resulted in the 
production of the anthrax vaccine that CDC used in the CDC open label 
safety study and that was licensed in 1970.
    We have reviewed the historical development of AVA and conclude 
that DoD directed the development of the vaccine, including its 
formulation and manufacturing process, from the vaccine used in the 
Brachman study (DoD vaccine) to the vaccine that was ultimately 
licensed and manufactured by BioPort (DoD/MDPH/AVA vaccine). All three 
versions of anthrax vaccine, DoD vaccine, DoD/MSD vaccine, and DoD/
MDPH/AVA vaccine, were tested in animals and demonstrated to protect 
test animals (e.g., guinea pigs, rabbits) against challenge with 
virulent B. anthracis spores. In addition, there are clinical data 
comparing the safety and immunogenicity of DoD/MDPH/AVA vaccine with 
DoD vaccine. These data, while limited in the number of vaccinees and 
samples evaluated, reveal that the serological responses to DoD/MDPH/
AVA vaccine and DoD vaccine were similar with respect to peak antibody 
response and seropositivity.
    Under FDA's long-standing approach to comparability, a manufacturer 
may make manufacturing changes in a product without performing 
additional clinical studies to demonstrate the safety and effectiveness 
of the similar product if data regarding the manufacturing changes 
support the conclusion that the versions are comparable. Put another 
way, after a manufacturing change, a manufacturer may use data gathered 
with a previous version of its product to support the effectiveness of 
a comparable version of the same product. These principles are further 
reflected in FDA's ``Guidance Concerning Demonstration of Comparability 
of Human Biological Products, Including Therapeutic Biotechnology-
derived Products'' (1996) (Ref. 8). As discussed previously in this 
document, DoD vaccine and DoD/MDPH/AVA vaccine are comparable in their 
ability to protect test animals against challenge with virulent strains 
of B. anthracis and to elicit similar immune responses in humans.

E. The Panel's Specific Product Review: Anthrax Vaccine Adsorbed: 
Labeling

    The Panel report states:
    3. Analysis--d. Labeling: The labeling seems generally adequate. 
There is a conflict, however, with additional standards for anthrax 
vaccine. Section 620.24 (a) (21 CFR 620.24(a)) defines a total 
primary immunizing dose as 3 single doses of 0.5 mL. The labeling 
defines primary immunization as 6 doses (0, 2, and 4 weeks plus 6, 
12, and 18 months).
(50 FR 51002 at 51059, December 13, 1985).
    The Panel was concerned with whether the vaccination schedule 
conformed to a standard set out in former Sec.  620.24(a), a rule that 
FDA revoked in 1996 with certain other biologics regulations because 
they were obsolete or no longer necessary (Ref. 9). The dosing schedule 
for AVA has always consisted of three doses of 0.5 mL administered in 
short succession at 0, 2, and 4 weeks, and three additional doses at 6, 
12, and 18 months, with additional doses at 1-year intervals to 
maintain immunity. However, the use of certain terminology has varied 
as discussed in this section of this document. Pre-licensure labeling 
(submitted to the license application with a letter dated January 25, 
1968) described the vaccination schedule as three initial doses, 
followed by three additional doses, and yearly subsequent doses. This 
schedule is consistent with the additional standards of AVA that were 
originally published on October 27, 1970 (35 FR 16631), immediately 
before the licensure of AVA. The 1979 labeling referred to ``primary 
immunization'' as consisting of six injections, with recommended yearly 
subsequent injections. The 1987 labeling of AVA, approved after the 
publication of the Panel's report, described the vaccination schedule 
as a ``primary immunization'' consisting of three doses followed by 
three additional doses (for a total of six doses), followed by annual 
injections. While the labeling has variously used the term ``primary'' 
to describe the AVA vaccination schedule, the licensed schedule itself 
has always consisted of three initial doses administered at 2-week 
intervals, followed by three additional doses at 6, 12, and 18 months, 
with additional annual doses to maintain immunity.

IV. Comments on the December 2004 Anthrax Vaccine Adsorbed (AVA) 
Proposed Order and FDA's Responses

    We received about 350 comments on the December 2004 proposal. Most 
comments related to AVA. To provide clarity to readers, we separated 
the AVA final order from the final rule and final order for other 
bacterial vaccines and toxoids. We are describing and responding to 
comments about AVA in this section of this document. Comments relating 
to other portions of the December 2004 proposal are discussed in a 
final rule and final order concerning bacterial vaccines and toxoids 
other than AVA published elsewhere in this issue of the Federal 
Register.
    We carefully reviewed all comments submitted to the Docket, 
including those attaching copies of articles and other references. 
However, a number of comments submitted to the Docket simply referred 
to articles or other publications, or to Web site materials, without 
providing copies of the materials. FDA regulations governing 
submissions to the Docket expressly provide that ``information referred 
to or relied upon in a submission is to be included in full and may not 
be incorporated by reference unless previously submitted in the same 
proceeding.'' (Sec.  10.20(c) (21 CFR 10.20(c)). Without a copy to 
review, we were unable to review all references cited but not included 
in the comments. We obtained and reviewed readily available recognized 
medical or scientific textbooks (see Sec.  10.20(c)(1)(iv)). The 
provision of Web site addresses, without substantive material, posed an 
additional problem. Since Web sites change continually, we were unable 
to review material at the Web site addresses provided with any

[[Page 75185]]

degree of certainty that the comment intended to incorporate the 
material we found. Also, many Web sites we checked contained irrelevant 
information. It was often difficult to determine a connection between 
the Web site and the comment's submission. FDA regulations require that 
only relevant information is to be submitted (Sec.  10.20(c)(3)) and 
failure to comply with these requirements results in exclusion from 
consideration of any portion of the comment that fails to comply (Sec.  
10.20(c)(6)).
    Many comments agreed with the Panel's recommendation that AVA is 
safe and effective and supported licensure of the vaccine; other 
comments advocated a need for a panel of experts to review in depth the 
data on AVA. Many of the comments did not support placing AVA into 
Category I as recommended by the Panel. Many comments described adverse 
events and suggested a relationship between the administration of AVA 
and the adverse events. Other comments recommended further testing of 
AVA through the conduct of clinical studies or other means. Numerous 
miscellaneous comments were received, some of which are not relevant to 
the proposed order. Many of the comments expressed an opinion about the 
conduct of vaccination administration programs, the need for 
compensation from public funds to individuals suffering injury from 
vaccinations, or other activities that are outside of FDA's 
jurisdiction, authority, and control.
    To make it easier to identify comments and our responses, the word 
``Comment,'' in parentheses, will appear before the description of 
comments, and the word ``Response,'' in parentheses, will appear before 
our response. We numbered the comments to help distinguish between 
different types of comments. The number assigned to a comment is purely 
for organizational purposes and does not signify the comment's value or 
importance or the order in which the comment was received.

A. Comments Supporting Placing AVA into Category I

    (Comment 1) We received a number of comments expressing support for 
the safety and effectiveness of AVA, and for FDA's proposal to accept 
the Panel's recommendation to place AVA into Category I. Some of these 
comments were specific in their support of the Brachman study as 
evidence of effectiveness against anthrax regardless of route of 
exposure; others discussed or described results of animal studies that 
they regarded as providing additional supporting evidence that AVA is 
effective in preventing inhalation anthrax. Some were from vaccine 
recipients and medical personnel who expressed support for the DoD 
vaccination program in its effort to protect military personnel from 
anthrax used as a biological weapon. Others were supportive of the work 
conducted by DoD to document and evaluate adverse events experienced by 
military personnel enrolled in the vaccination program.
    One comment was from a former director of the Division of 
Biological Standards (DBS) of the NIH and subsequently within the FDA, 
who stated his recollection that AVA had been subject to a careful 
review by DBS staff prior to approval in 1970. He stated that there 
have been three detailed, unbiased, and scientifically sound reviews, 
including the initial review by DBS, the expert Panel review in the 
1970s (published in the December 1985 Proposal), and the IOM review 
more recently; and all three reviews concluded that the vaccine is safe 
and effective. Two comments were submitted by scientists who had been 
clinical investigators in the Brachman study. One stated that during 
the study he was blinded to group assignment when evaluating the 
reactions; i.e., he did not know whether the subject had received the 
placebo or the vaccine. He also stated that the pathophysiology of 
human anthrax, regardless of where the organism gains entrance to the 
body, is a result of the toxin released by the organism. Thus, it is 
appropriate to combine inhalation and cutaneous disease in the 
analysis. The other scientist stated that the vaccine has demonstrated 
effectiveness in animal and human studies, as described in published 
scientific literature articles.
    We received comments from Army research scientists in support of 
placing AVA into Category I. One of these included tables of data from 
anthrax spore inhalation challenge studies in non-human primates and 
rabbits evaluating the effectiveness of AVA in prevention of death from 
disease. The comment noted that a high degree of protection was 
observed in these animals following only one or two doses of AVA, and 
that the IOM committee concluded that these animal models are 
representative of the human form of inhalation anthrax. Another 
research scientist also noted that, in addition to the Brachman study, 
inhalation anthrax challenge studies in non-human primates provide 
evidence of AVA's effectiveness in preventing disease caused by anthrax 
spores. Further, he noted that current knowledge of the pathogenesis of 
anthrax would indicate that, regardless of the route by which spores 
enter the body, toxins produced after those spores germinate into 
growing bacilli are essential for the anthrax organism to cause 
disease. Current scientific understanding of how the toxins work 
indicates that antibodies induced by AVA block the activities of 
anthrax toxins such that they would be effective in preventing any form 
of the disease regardless of the route of exposure to B. anthracis 
spores. Another researcher discussed further and in more detail how the 
pathology of cutaneous and inhalation anthrax at the cellular level is 
fundamentally the same, i.e., dependent upon the actions of anthrax 
toxin, such that cytotoxic activities are blocked by antibodies 
produced in response to AVA in the same manner despite the route of 
exposure.
    Military personnel involved in the vaccine's administration under 
the DoD vaccination program also filed comments in support of 
classifying AVA into Category I, reasoning that the vaccine is 
important for soldiers entering potentially dangerous areas; however, 
one comment stated that long-term use of the vaccine should be studied 
further. Another comment was submitted by a physician who thought that 
there was evidence that AVA protects against inhalation anthrax and 
that the side effects of vaccination were comparable to other adult 
vaccines. Comments supportive of placing AVA into Category I were also 
submitted by a representative of the Armed Forces Epidemiological Board 
(AFEB), a civilian advisory body to the Assistant Secretary of Defense 
for Health Affairs and the military Surgeons General. This comment 
described the AFEB deliberations on the use of anthrax vaccine by the 
military and the recommendations made by the AFEB to the DoD supporting 
use of AVA as an appropriate force protection measure. A representative 
of the Partnership for Anthrax Vaccine Education, a coalition of public 
and private organizations, also submitted comments reflecting that 
organization's support for placing AVA into Category I.
    (Response) We agree with those comments that provided support for 
placing AVA into Category I.

B. Comments on the Evidence of Safety and Effectiveness of AVA

    (Comment 2) Some comments were concerned about the safety of AVA.
    (Response) With regard to safety, FDA finds that AVA is safe for 
its indicated use as noted in the 2002 package insert:

[[Page 75186]]

    BioThrax [the Tradename for AVA] is indicated for the active 
immunization against Bacillus anthracis of individuals between 18 
and 65 years of age who come in contact with animal products such as 
hides, hair or bones that come from anthrax endemic areas, and that 
may be contaminated with Bacillus anthracis spores. BioThrax is also 
indicated for individuals at high risk of exposure to Bacillus 
anthracis spores such as veterinarians, laboratory workers and 
others whose occupation may involve handling potentially infected 
animals or other contaminated materials. (Ref. 6)
    The adverse reactions observed after administration of AVA in 
clinical study settings are described in the product labeling approved 
in 2002. At that time, FDA conducted an extensive review of the 
clinical study data from the DoD pilot study, reports from DoD safety 
surveys conducted as part of their Anthrax Vaccine Immunization 
Program, and reports submitted to the Vaccine Adverse Event Reporting 
System (VAERS). Since approval of the revised labeling in 2002, FDA has 
conducted periodic evaluations of the reports in the VAERS database, 
and, as discussed elsewhere in this document, continues to find AVA to 
be safe for its intended use: To protect individuals at high risk for 
anthrax disease. Anthrax disease can be fatal despite appropriate 
antibiotic therapy.
1. Brachman Study
    (Comment 3) Some comments expressed criticisms of the design and 
conduct of the Brachman study (Ref. 1).
    (Response) The Brachman study was an adequate and well-controlled 
clinical study that involved workers in four textile mills that 
processed imported goat hair in the northeastern United States. This 
selected population was at risk because the mill workers routinely 
handled anthrax-infected animal materials. Prior to vaccination, the 
yearly average number of human anthrax infections among workers in 
these mills was 1.2 cases per every 100 employees.
    The Brachman study design permitted a valid comparison of the 
vaccine group with the placebo control group to provide a quantitative 
assessment of effectiveness. For this study, employees with no known 
history of anthrax disease were assigned to one of two groups, 
treatment and placebo. The groups were balanced with regard to the 
individual's age, length of employment, department and job; both men 
and women were enrolled into the study. Voluntary cooperation was 
solicited and those who refused did not receive inoculations but were 
monitored for anthrax disease as part of the observational group. The 
subjects who chose to receive inoculations were not told whether they 
received anthrax vaccine or placebo. The published report of the 
Brachman study (Ref. 1) described all anthrax cases that occurred in 
the study, including ones in the vaccine, placebo, and observational 
groups. The Brachman study's efficacy analysis included only the cases 
that occurred in the treatment and placebo groups in completely 
vaccinated subjects (i.e., those receiving at least three inoculations 
and on schedule to receive the remaining three doses of the six-dose 
series), an approach that remains typical of vaccine analyses to date. 
We determine that the original statistical analysis presented in the 
report from the Brachman study was correct in its estimation of vaccine 
effectiveness. Some of the specific criticisms of the Brachman study 
included in the submitted comments claimed that the sample size was too 
small and that it was inappropriate to combine data from all four mills 
in the efficacy analysis.
    Clinical studies are designed with a sample size sufficient to 
assure with high probability that, if there is a true effect of the 
intervention under study, that effect will be ``detected;'' that is, a 
comparison of outcomes in the treatment and control groups will show a 
``statistically significant'' difference. To obtain the required sample 
size, investigators often have to implement the study at multiple sites 
(i.e., a multicenter study). The number of patients enrolled at any 
given site may be small, relative to the total number, and may not 
afford a high probability of achieving statistical significance at each 
individual site independently. Thus, when analyzing a multicenter 
clinical study, it is not reasonable to expect a statistically 
significant result at each site. Instead, consistent effects among 
individual study sites are the standard for multicenter studies (Ref. 
10).
    The Brachman study, a multicenter study, was based on an adequate 
sample size and appropriately combined the data from all mills in its 
analysis of vaccine efficacy. The site-specific data for the Brachman 
study are quite consistent in that at all sites, the vaccine group had 
fewer cases of anthrax than the placebo group. The strength of the 
overall finding of vaccine efficacy is such that, even with small 
numbers at each site, differences in outcome between the treatment and 
control groups are clearly statistically significant in one site and 
marginally significant in another. Thus, the site-specific data are 
fully supportive of the overall result, which showed a large reduction 
in risk of anthrax among those receiving vaccine.
    (Comment 4) One comment noted that a 1960 publication by Brachman 
et al. stated ``The efficacy of the anthrax cell-free antigen as a 
vaccine was not fairly tested in this epidemic. Although none of the 9 
cutaneous plus inhalation cases occurred in vaccinated individuals, 
only approximately one fourth of the employees had received the 
vaccine. There was an apparent difference in attack rates between 
workers who received placebo inoculations and those who received 
vaccine, but analysis of their job categories suggested that the 
vaccinated group was not at as high a risk as the placebo or 
uninoculated control groups.'' The comment makes several critical 
statements, based upon this 1960 publication, about FDA's reliance upon 
the Brachman study as evidence of vaccine effectiveness, claiming that 
the placebo group was at a greater risk of anthrax disease than the 
vaccine group.
    (Response) Prior to publication of the complete study report in 
1962, Brachman et al. published two papers (Refs. 11 and 12) describing 
the clinical features and epidemiology of an outbreak of inhalation and 
cutaneous anthrax cases that occurred in the Manchester, New Hampshire 
mill, one of the four mills included in the field study. The 
publication describing the epidemiology of that outbreak does include 
the statement quoted previously; however, the statement is specifically 
in reference to one study site and not to the field study as a whole, 
across the four woolen mills. The subsequent 1962 publication (Ref. 1) 
of the complete study across all four sites includes a table depicting 
participation of employees from all four mills included in the study. 
The table shows whether employees worked in high or low risk work areas 
and whether they received vaccine, placebo, or refused to participate 
in the study (Ref. 1 at Table 2). Of note, the totals for recipients of 
vaccine, placebo, incomplete inoculation and refusals in high risk work 
areas were 209, 226, 65 and 89, respectively. The same totals in low 
risk work areas were 170, 188, 51 and 251, respectively.
    The distribution of vaccine recipients, placebo recipients, and 
incompletely inoculated subjects was similar for both the high and low 
risk work areas, which means that the vaccine and placebo groups were 
balanced with regard to the exposure risk factor. A larger number of 
persons who did not participate in the study (observation group) were 
in the low risk work areas than in the high risk areas, but the 
efficacy analysis did not

[[Page 75187]]

include cases that occurred in the observational group. The 
effectiveness calculation described in the 1962 publication included 
the anthrax cases that occurred in participants who received at least 
three doses of either vaccine or placebo and remained on schedule for 
the remainder of the six doses for all four mills, not just the 
Manchester, New Hampshire mill described in the 1960 publications. 
Thus, FDA's consideration of the Brachman study as evidence of 
effectiveness is based upon the complete analysis across all four study 
sites.
    (Comment 5) One comment stated that it was inappropriate for the 
Brachman study to include both cutaneous and inhalation cases in the 
efficacy analysis.
    (Response) The efficacy analysis presented in the Brachman study 
includes both cutaneous and inhalation anthrax cases that occurred in 
individuals who received at least three doses of vaccine or placebo and 
were on schedule for the remaining doses of the six-dose schedule. It 
did not include cases that occurred in the observation group. Based on 
this analysis, the calculated effectiveness level against all reported 
cases of anthrax combined in those subjects was 92.5 percent (lower 95 
percent confidence interval = 65 percent). The efficacy analysis 
included the combined outcome of cutaneous and inhalation anthrax cases 
and thus included anthrax cases regardless of the route of exposure or 
manifestation of the disease.
    The inclusion of both cutaneous and inhalation cases of anthrax in 
the analysis of the Brachman study was appropriate because it was not 
possible to predict the route of exposure (cutaneous versus inhalation) 
that would occur within the environmental setting of the woolen mills. 
With regard to the known pathophysiology of anthrax, the signs and 
symptoms of disease arise due to the production of toxins by anthrax 
bacteria growing within the infected individual. The toxins produced by 
anthrax bacteria do not vary based on the route of exposure. The 
antibodies produced in response to vaccination contribute to the 
protection of the vaccinated individual by neutralizing the activities 
of those toxins. Thus, AVA elicits an antibody response to disrupt the 
cytotoxic effects of toxins produced by anthrax bacteria, regardless of 
the route of infection.
    (Comment 6) One comment stated that any decision by FDA to license 
AVA must provide a scientifically valid explanation of how FDA has 
assessed this vaccine's effectiveness against anthrax infection by 
inhalation in humans in the absence of an adequate and well-controlled 
clinical study specifically studying its effectiveness against anthrax 
infection by inhalation. The comment contends that in the absence of 
such data, or unless FDA uses the ``animal efficacy rule,'' FDA should 
not license AVA as a Category I biological product.
    (Response) AVA has been licensed since 1970. The Panel, as 
reflected in its report published in the December 1985 proposal, and 
the FDA, as reflected in this final order, have determined that AVA is 
safe and effective for its labeled indication, decisions based in part 
on the Brachman study, which was an adequate and well-controlled study. 
Even if the referenced ``animal efficacy rule''\6\ had been in effect 
at the time of AVA licensure, it would not have been applicable because 
there are sufficient data from adequate, well-controlled clinical 
studies to assess the safety and effectiveness of AVA as a vaccine 
against anthrax infection regardless of route of exposure. The ``animal 
efficacy rule'' does not apply to products that can be approved based 
on efficacy standards described in other regulations (Sec.  601.90 (21 
CFR 601.90)).
---------------------------------------------------------------------------

    \6\New Drug and Biological Drug Products; Evidence Needed to 
Demonstrate Effectiveness of New Drugs When Human Efficacy Studies 
Are Not Ethical or Feasible; Final Rule (21 CFR 601.90 through 
601.95) (67 FR 37988, May 31, 2002).
---------------------------------------------------------------------------

    (Comment 7) One comment pointed out that the route of exposure to 
an infectious agent can be a critical factor influencing vaccine 
effectiveness.
    (Response) We agree that the route of exposure to an infectious 
agent may potentially have an impact on the effectiveness of a vaccine. 
The impact likely depends on the nature of the infectious agent in 
terms of its mechanism of virulence and the pathophysiology of 
infection and disease, and the mechanism of protection afforded by the 
vaccine. The Brachman study showed the anthrax vaccine to be effective 
in preventing anthrax disease regardless of route of exposure (Ref. 1). 
This finding is consistent with our current knowledge of the critical 
role played by anthrax toxins in the pathophysiology of cutaneous and 
inhalation anthrax and how antibodies generated in response to 
vaccination with AVA disrupt cytotoxic activities of those toxins. 
Furthermore, aerosolized anthrax spore challenge studies in both 
rabbits and nonhuman primates do demonstrate the ability of AVA to 
protect the test animals against inhalation anthrax (Refs. 3, 4, and 
5).
    (Comment 8) One comment proposed that a vaccine would have to be 
inhaled in order to protect against inhalation anthrax, noting that the 
lungs are susceptible to anthrax.
    (Response) Vaccines generally do not need to be administered by the 
same route of exposure as the infectious agent uses to infect humans. 
In fact, there are numerous examples to the contrary. For example, 
vaccines against pertussis, pneumococcus, Hemophilus influenzae type b, 
meningococcus, measles, varicella, and influenza are administered by 
injection, although the infectious agents gain entry into humans by the 
respiratory route. The inactivated poliovirus vaccine is administered 
by injection, although the poliovirus infects humans by way of the 
intestinal tract. Although these vaccines are administered by a route 
that differs from the route of exposure, clinical trials have 
demonstrated their effectiveness against the targeted infectious 
disease. The same is true of anthrax vaccine. The vaccine is 
administered by injection, but has been shown to be effective against 
anthrax in a study that included both cutaneous and inhalation cases 
(Ref. 1). Furthermore, animal studies in which injected AVA protected 
animals from inhalation anthrax challenge are consistent with the 
finding of effectiveness in the clinical study. (Refs. 3, 4, and 5)
    (Comment 9) One comment stated that FDA has deviated from the 1985 
Panel recommendations (i.e., ``No meaningful assessment of its value 
against inhalation anthrax is possible due to its low incidence.'' 50 
FR 51002 at 51059) and that FDA should not dispute its advisory 
committee's analysis of the safety and effectiveness data.
    (Response) A critical component of the efficacy review process is 
FDA's consideration of the Panel's recommendations (Sec.  601.25(f)). 
Such consideration, by necessity, provides for the possibility that FDA 
might disagree with the Panel's recommendations. Indeed, in the 
preamble to Sec.  601.25, FDA stated that ``the report of each panel is 
advisory to the Commissioner, who has the final authority either to 
accept or to reject the conclusions and recommendations of the panel.'' 
(38 FR 4319 at 4321, February 13, 1973). As noted in section III.A of 
this document, and as stated in the December 2004 proposal, we do not 
agree with the Panel's assessment that the vaccine is 93 percent 
efficacious against cutaneous anthrax only. In fact, the calculation of 
effectiveness presented in the published report of the Brachman study 
pertains to both cutaneous and inhalation anthrax. The Brachman study 
included in the effectiveness calculation both the

[[Page 75188]]

cutaneous and inhalation cases that occurred in vaccine and placebo 
recipients who received at least three doses and remained on schedule 
to receive the rest of the six-dose series.
2. CDC Surveillance Data
    (Comment 10) One comment stated that the CDC surveillance data do 
not provide a reliable basis for an assessment of effectiveness 
because: (1) They represent the use of at least two earlier versions of 
anthrax vaccine, which are not the same vaccine currently produced by 
BioPort; (2) they are not statistically significant; and (3) these data 
may not be accurate and complete. Other comments asked why the CDC 
surveillance data for the years 1962 to 1974 are not regarded as 
supportive of safety of anthrax vaccine.
    (Response) During the time these surveillance data were collected 
by CDC, both DoD/MSD vaccine and DoD/MDPH/AVA vaccine were available 
for use. The DoD/MDPH/AVA vaccine was licensed in 1970 and is the same 
vaccine currently manufactured and distributed by BioPort. An 
additional response to comments regarding different versions of the 
anthrax vaccine is addressed later in this document.
    Although we do not consider the CDC surveillance data to be 
statistically significant, we regard the data as indicative that, 
during this time period, workers continued to be at risk of exposure, 
because anthrax cases were identified in unvaccinated and partially 
vaccinated individuals employed at woolen mills. The data are 
supportive of the effectiveness evidenced by the Brachman study, in 
that no anthrax cases were reported in fully vaccinated individuals 
during that time period. We do not regard the CDC surveillance data as 
contributing to an assessment of safety because the data do not 
describe adverse events occurring after vaccination.
    The comment provides no support for the conclusion that the CDC 
surveillance data were unreliable. The comment described an anecdotal 
report of an additional anthrax case that occurred in an unspecified 
year and apparently was not included in the CDC surveillance data. We 
recognize that there is a potential for underreporting in disease 
surveillance systems. However, this one report does not provide a basis 
for concluding that the CDC surveillance data were unreliable for the 
purposes of supporting the effectiveness of the vaccine.
3. CDC Open Label Safety Study
    (Comment 11) Some of the comments questioned the reliability of the 
CDC open label safety study, alleging that the open label safety study 
conducted by CDC ``made no attempt to identify, quantify or follow 
systemic adverse vaccine reactions'' and thus would be of no value in 
establishing vaccine safety, or that the study did not use consistent 
standards to identify and grade adverse events occurring at different 
study sites.
    (Response) As described previously in this document, FDA believes 
that there are adequate data to demonstrate the safety and 
effectiveness of AVA. Moreover, the CDC open label safety study 
appropriately collected and analyzed adverse event reports. The IND 
protocol for the CDC open label safety study included specific criteria 
to be used to categorize mild, moderate and severe local reactions 
reported in the course of the study. In addition, the annual study 
reports submitted to the IND included information regarding systemic 
reactions reported during the respective reporting periods, and those 
data are described in the current product labeling for AVA: ``In the 
same open label safety study, four cases of systemic reactions were 
reported during a five-year reporting period (< 0.06% of doses 
administered). These reactions, which were reported to have been 
transient, included fever, chills, nausea and general body aches.'' 
(Ref. 6)
    (Comment 12) One comment claimed that one annual safety report for 
the CDC open label safety study might have underreported adverse 
reaction rates for that period, alleging that arithmetic 
miscalculations caused underreporting in one May 1967 reactogenicity 
table.
    (Response) The commenter refers to the May 1967 table included in 
an appendix to one of the annual reports to the CDC trial; the appendix 
describes a protocol and the results of a small safety and 
immunogenicity study comparing DoD vaccine and DoD/MDPH/AVA vaccine. 
The safety data from this small study were reported separately from the 
CDC open label safety study due to differences in protocol design, such 
as the administration of one-half volume booster doses to some subjects 
instead of the full 0.5 mL human dose. Inclusion of safety data from 
the small ancillary safety study with a different protocol design does 
not support the inference that the annual safety report for the CDC 
open label safety study might have underreported adverse reaction rates 
for that period.
    (Comment 13) One comment stated that in the course of the CDC open 
label safety study, Ft. Detrick and mill employees were required to be 
vaccinated as a condition of employment and therefore, they may have 
underreported adverse reactions to the vaccine from fear of losing 
their jobs. The comment also states that the employees did not provide 
free informed consent to participate in the study because they were 
compelled to be vaccinated, and no informed consent documents were 
signed by Ft. Detrick employees. Thus, the study did not comply with 
FDA requirements for informed consent.
    (Response) The comment provides no support for the assumption that 
subjects in the CDC open label safety study may have underreported 
adverse reactions to the vaccine. With regard to the statements that 
mill workers in the CDC open label safety study were compelled to be 
vaccinated, and therefore did not provide informed consent, and that 
the Ft. Detrick subjects in the study did not sign informed consent 
documents, we note that the CDC open label safety study was conducted 
under IND 180 from 1966 through 1971. The NIH was responsible for 
reviewing IND 180 and the subsequent marketing application for AVA 
under the regulations then in effect. Significantly, the NIH did not 
reject the study, or place it on hold. Moreover, the comment does not 
identify a legal basis for requiring FDA to reject the study for this 
reason.
    FDA is committed to assuring the protection of human subjects in 
clinical trials, as evidenced by the comprehensive regulations now in 
place (see FDA's current informed consent regulations, 21 CFR part 50, 
in effect since 1981, and IND regulations, 21 CFR part 312, in effect 
since 1987). Other data and studies, such as the DoD pilot study, 
conducted subsequent to the CDC open label safety study and under 
current informed consent regulations, provide additional safety 
evidence that corroborate the CDC open label safety study findings. We 
decline to reject the findings of the CDC open label safety study and 
we continue to view them as supportive of safety.
4. DoD Pilot Study and Safety Data
    (Comment 14) One comment inquired whether the results of the DoD 
pilot study relating to the vaccine's safety required changes to AVA 
labeling in 2002, and whether additional data were considered in 
support of the new labeling. Other comments asked whether the DoD pilot 
study was also regarded as supportive of effectiveness.
    (Response) BioPort voluntarily submitted to FDA proposed revised 
labeling for AVA for review and comment as part of an ongoing process 
of updating product and manufacturing information. In the course of 
FDA's review, revisions were made to the proposed labeling. Following 
our

[[Page 75189]]

review, in 2002 we approved revised product labeling that incorporated 
more recently acquired safety information from the DoD pilot study and 
FDA's ongoing review of reports to VAERS. The DoD pilot study was not 
intended to assess effectiveness; rather its purpose was to make an 
initial assessment of the effects that alternative immunization 
schedules and/or an alternative route of administration may have on the 
safety and immunogenicity of AVA.
    (Comment 15) One comment claimed that the 1996 to 1999 DoD pilot 
study as reported is entirely inadequate to determine the safety of 
AVA, noting that the study was ``uncontrolled'' and that a quarantined 
lot was used in the study.
    (Response) As discussed previously in this document, the CDC open 
label safety study, involving approximately 7,000 subjects who received 
DoD/MDPH/AVA vaccine,\7\ demonstrated the safety of AVA. The DoD pilot 
study, which included 28 subjects randomized to receive the licensed 
vaccine according to the labeling, was conducted subsequent to 
licensure and provided additional data in support of the safety of AVA. 
The DoD pilot study was a controlled clinical study; the group 
receiving AVA according to the licensed schedule and route of 
administration served as the control group for the other groups 
receiving the vaccine under alternative vaccination schedules and/or 
route of administration. The purpose of the DoD pilot study was to make 
an initial assessment of the effects that alternative immunization 
schedules and/or an alternative route of administration may have on the 
safety and immunogenicity of AVA. The alternative schedules were 
alterations of the 0-2-4 week initial series of the licensed six-dose 
schedule (i.e., 0-4 weeks, 0-2 weeks). These alternative schedules were 
administered intramuscularly and subcutaneously. However, because one 
of the arms of the study included individuals vaccinated according to 
the labeling, we appropriately took such information into account as we 
continued to assess the safety of AVA. In this arm of the study, 
volunteers received subcutaneous doses of AVA according to the licensed 
schedule. Each volunteer was scheduled for follow-up evaluations at 1 
to 3 days, 1 week, and 1 month after vaccination, and reactions were 
reported up to 30 days after each dose. For subjects who received the 
vaccine according to the licensed route and schedule, the latest 
follow-up occurred 30 days after the 18-month dose (Ref. 13).
---------------------------------------------------------------------------

    \7\In addition, one lot of the DoD/MSD vaccine was used during 
the CDC open label safety study.
---------------------------------------------------------------------------

    In the December 2004 proposal, FDA discussed the safety data 
collected under this study for subjects receiving the vaccine according 
to the labeling. Similarly, descriptive information regarding adverse 
reactions reported in individuals receiving the vaccine according to 
the licensed schedule under this study was included in the 2002 
labeling. Thus, the December 2004 proposal and the 2002 labeling 
reported this recently acquired safety information, which had been 
collected in a planned and prospective manner.
    In addition, we believe no subjects in the study received 
quarantined doses of lot FAV 016, the lot mentioned in the comment. We 
understand that some subjects received lot FAV 032 while the voluntary 
quarantine of that lot was being implemented. However, this information 
does not provide an adequate basis for us to refuse to consider the 
data derived from the study. It is important to note that one of the 
chief uses of the study was as one of the bases for the expanded 
description of adverse events included in the 2002 labeling. Thus, the 
study results provided additional information for individuals 
administering and receiving AVA. We believe that this limited use of 
lot FAV 032 did not cause the results of the entire study to be 
unreliable, particularly in light of the purposes for which we use the 
data derived from this arm of the study. We will continue to monitor 
all available sources of information relating to the safety of AVA.
    (Comment 16) One comment was critical of the fact that the results 
of the DoD pilot study were included in the 2002 labeling when the data 
were not peer reviewed or available to the public.
    (Response) FDA performs its own review of data that are submitted 
in support of labeling changes. There is no requirement for peer review 
of data submitted to FDA in support of a labeling change. The DoD pilot 
study was intended to serve as a pilot study of alternative vaccination 
schedules and an alternative route of administration (intramuscular) to 
provide information for the design of a larger, more statistically 
robust study of promising alternative vaccination schedules and route 
of administration. The investigators published their report of this 
study in a peer-reviewed journal (Ref. 13).
    5. Long-Term Safety Monitoring and Additional Studies
    (Comment 17) A number of comments discussed the absence of a long-
term safety study using AVA and the absence of studies of the potential 
effects of vaccination on vaccine recipients' children.
    (Response) The pre-licensure safety evaluation of a new vaccine may 
include clinical studies that extend several months to several years 
after administration of the first dose. For example, the CDC open label 
safety study spanned from 1966 through 1971. Pre-licensure safety 
studies focus on those adverse reactions closely associated with the 
time of vaccine administration such as local injection site reactions 
and systemic reactions such as fever, malaise and allergic reactions. 
However, all serious adverse events that are reported during the 
conduct of the study are evaluated regardless of when they occur 
relative to vaccination. Longer-term controlled clinical trials (i.e., 
those extending more than several years after vaccination) are not 
generally conducted prior to approval of any medical product, including 
vaccine products.
    The attribution to a vaccine or other drug product of adverse 
events or health conditions that develop long after administration is 
difficult to make with confidence because other factors such as 
environmental exposures, general health, genetic predisposition, etc., 
may also contribute to the development of health problems, symptoms or 
diseases. Elsewhere in this document, we provide a more detailed 
discussion of FDA's approach to post-licensure safety monitoring of 
AVA.
    With regard to the potential effects of vaccination on offspring, 
the current approved labeling for AVA addresses administration of AVA 
to pregnant women. The labeling describes a preliminary assessment of 
the possibility that an increase in the rate of birth defects may be 
associated with AVA vaccination during pregnancy. Based upon the 
limited information available, the vaccine was assigned a Pregnancy 
Category D designation. The labeling states that ``Although these data 
are unconfirmed, pregnant women should not be vaccinated against 
anthrax unless the potential benefits of vaccination have been 
determined to outweigh the potential risk to the fetus.'' (Ref. 6)
    DoD has undertaken to verify these preliminary results. We will 
review those results, when available, and we will continue to review 
adverse events.
    (Comment 18) Many comments expressed concern about FDA's process of 
monitoring the safety of AVA.
    (Response) For any drug or biological product, rare adverse events 
not observed during pre-licensure clinical studies may occur post-
licensure. The

[[Page 75190]]

need to understand the relationship between vaccination and adverse 
events that occur after licensure, and the limitations of clinical 
trials, have led to the use of other methods to detect and evaluate the 
link between vaccination and rare events. Post-marketing monitoring of 
vaccine safety involves the identification of possible adverse effects 
of vaccination, followed in some cases by evaluation of these 
``signals'' for a possible causal link to the vaccine.
    The most common method of signal generation is through the 
evaluation of spontaneous reports of cases of adverse events reported 
to manufacturers or government-sponsored systems such as the Vaccine 
Adverse Event Reporting System (VAERS). The identification of 
``signals'' and their prioritization for evaluation involves 
qualitative and quantitative aspects, along with medical and 
epidemiological judgment. Evaluation of signals can involve literature 
review and clinical, laboratory, and epidemiological studies.
    Surveillance for adverse events after vaccination is undertaken 
using VAERS, which is jointly managed by FDA and CDC. Uses of VAERS 
include detecting unrecognized adverse events, monitoring known 
reactions, identifying possible risk factors, and vaccine lot 
surveillance. Established in 1990, VAERS receives approximately 15,000 
adverse event reports annually. Reports are submitted by vaccine 
manufacturers, vaccine providers, other health care givers, vaccine 
recipients and their relatives, attorneys, and other interested 
parties. While vaccine manufacturers are responsible for investigating 
and evaluating reports made to them, FDA and CDC also follow up reports 
from other parties of deaths and adverse events resulting in life-
threatening illness, hospitalization, prolongation of hospitalization, 
persistent or significant disability, or congenital anomaly/birth 
defect, by telephone to obtain additional information about the event 
and the patient's prior medical history.
    Passive surveillance systems such as VAERS are subject to 
limitations. Vaccine-associated adverse events will inevitably be 
underreported to an unknown extent. Moreover, adverse events reported 
in association with vaccination may or may not be caused by 
vaccination. For example, some adverse events might be expected to 
occur by coincidence after vaccination. Temporal associations often are 
reported with little data to evaluate whether any causal connection 
with the vaccine exists. Given these limitations, while safety signals 
may be detected, incidence rates cannot be determined from VAERS data. 
A particularly important limitation on the usefulness of VAERS reports 
as a means of investigating the possible causal relationship between an 
event and a vaccination generally is the lack of a direct, concurrent 
and unbiased comparison group from which to determine the incidence of 
the same type of adverse events among people who have not been 
vaccinated.
    Another important limitation is the lack of standardization of 
diagnoses in VAERS reports. Reporting of unconfirmed diagnoses is 
common with VAERS reports. On follow-up, initially reported diagnoses 
are sometimes found to be inaccurate. Reports are coded by non-
physicians, without the benefit of standardized case definitions, using 
the Coding Symbols for Thesaurus of Adverse Reaction Terms (COSTART) to 
describe the adverse event in a computerized database. Report coding 
depends on the reporter's use of certain words or phrases. This results 
in the use of the same COSTART term for reports with different degrees 
of diagnostic precision. For example, a report may simply say, ``I 
developed arthritis after I received the vaccine,'' without any other 
supporting medical information. Such a report would likely be coded as 
``arthritis,'' as would a report that included a complete medical 
record in which a physician documents joint swelling and tenderness. As 
a result, coding terms must be interpreted very cautiously.
    Because of the limitations of passive surveillance data, it is 
usually not possible to assess whether a vaccine caused the reported 
adverse event, except for conditions such as injection site reactions, 
some hypersensitivity conditions (e.g., anaphylaxis occurring shortly 
after vaccination), and illnesses consistent with the naturally 
occurring disease where vaccine components can be recovered from tissue 
specimens (e.g., recovery of live attenuated vaccine virus from 
vaccine-associated paralytic polio).
    Analysis of VAERS data focuses on describing clinical and 
demographic characteristics of reports and looking for patterns to 
detect ``signals'' of adverse events plausibly linked to a vaccine. In 
FDA's guidance document on ``Good Pharmacovigilance Practices and 
Pharmacoepidemiologic Assessment'' (Ref. 14), we define a safety signal 
as a concern about an excess of adverse events compared to what would 
be expected to be associated with a product's use. This guidance 
document also details approaches for signal evaluation. Evidence of a 
signal in case reports and in case series of spontaneous reports 
includes unexpected patterns in clinical conditions by such factors as 
age, gender, time to onset, and dose. Three reports of an event can be 
used as the minimum number for case series analysis of rare conditions. 
Positive rechallenge is defined as the same event occurring after more 
than one dose of the same vaccine in the same subject and may also be 
considered evidence of a signal. Signals detected through analysis of 
VAERS data do not necessarily represent a causal relationship with the 
vaccine and almost always require confirmation through additional 
study.
    In addition to the approach combining descriptive epidemiology with 
medical judgment, described above, several quantitative approaches, 
sometimes referred to as ``data mining'' methods, have been proposed. A 
common feature of data mining methods is that they identify patterns in 
the data that consist of a condition or group of conditions that are 
reported as a higher proportion of all adverse events after a 
particular vaccine or combination of vaccines than after other 
vaccines.
    Calculations of reporting rates (number of adverse events reported/
number of doses of vaccine distributed) and reporting rate ratios 
(ratio of reporting rate in the vaccine of interest to the reporting 
rate in the comparison vaccine(s)) of adverse events have been used to 
generate signals. Comparison of reporting rates with background 
incidence rates for an adverse event is also sometimes advocated. 
Biases in reporting, inadequate denominator data, uncertainty of the 
risk interval (the interval after vaccination during which a person 
might be at risk for the adverse event under study) and lack of 
background incidence rates from an appropriate comparison population 
for some conditions limit the utility of the reporting rate approach.
    Regardless of the method used, interpretation of vaccine-adverse 
event combinations that are identified as possible signals with any 
quantitative method must use medical knowledge about the disorders and 
take into account biases in reporting, misclassification of reports 
that occur with adverse event coding systems, and other limitations of 
passive surveillance systems previously discussed. Signals generated 
through such quantitative analysis need to be subject to the same 
clinical, descriptive epidemiological, and other analysis as for case 
reports and case series of spontaneous reports. Elevated reporting rate 
ratios or proportional reporting ratios or similar scores from data 
mining should not by themselves be interpreted as

[[Page 75191]]

establishing a causal relationship between an adverse event and a 
vaccine, but almost always require independent confirmation through 
additional study.
    In spite of these limitations, use of VAERS data has provided 
initial reports that upon further evaluation have raised suspicions, 
later confirmed, about rare reactions to vaccines (e.g., 
intussusception after rotavirus vaccine). VAERS data also have 
suggested the need for further study of other adverse events (e.g., 
myopericarditis after smallpox vaccine).
    Many possible signals\8\ can be generated with these methods and 
prioritization for further evaluation is required. Because information 
submitted to VAERS is often incomplete, it is sometimes necessary to do 
enhanced follow-up of reports to systematically collect information as 
the first stage in the signal evaluation process. Objective factors 
such as seriousness and ``newness'' of the adverse event, size of the 
population potentially affected, ability to prevent the adverse event, 
and ability to study the question, influence priority for further 
evaluation.
---------------------------------------------------------------------------

    \8\Safety signals that may warrant further investigation may 
include, but are not limited to, the following: (1) new unlabeled 
adverse events, especially if serious; (2) an apparent increase in 
the severity of a labeled event; (3) occurrence of serious events 
thought to be extremely rare in the general population; (4) new 
product-product, product-device, product-food, or product-dietary 
supplement interactions; (5) identification of a previously 
unrecognized at-risk population (e.g., populations with specific 
racial or genetic predispositions or co-morbidities); (6) confusion 
about a product's name, labeling, packaging, or use; (7) concerns 
arising from the way a product is used (e.g., adverse events seen at 
higher than labeled doses or in populations not recommended for 
treatment); (8) concerns arising from potential inadequacies of a 
currently implemented risk minimization action plan (e.g., reports 
of serious adverse events that appear to reflect failure of a risk 
minimization action plan goal); and (9) other concerns identified by 
the sponsor or FDA. (``Guidance for Industry: Good Pharmacovigilance 
Practices and Pharmacoepidemiologic Assessment,'' March 2005.)
---------------------------------------------------------------------------

    VAERS reports are not the only source of information used to 
evaluate the safety of a vaccine. Evaluation of signals usually 
requires a literature review followed by epidemiological studies, 
sometimes combined with clinical and laboratory analysis. To evaluate 
specific hypotheses it is sometimes necessary to conduct cohort, 
population-based case series, case-control or other epidemiological 
studies using large administrative databases with medical record 
review.
    If a clinical trial with sufficient statistical power to evaluate 
the adverse event of interest has not been conducted, assessing the 
potential causal link between a vaccine and an adverse event often 
requires integration of different types and quality of information 
(e.g., laboratory studies, case reports, epidemiological studies, and 
clinical studies). Causal inference criteria, patterned after those 
proposed by A. Bradford Hill in 1965 and adapted by others, and formal 
risk assessment have been applied to vaccine safety assessments. In a 
study of pertussis and rubella vaccines in the early 1990s, the IOM 
used the strength of association, the nature of the dose-response 
relation, the existence of a temporally correct association, 
consistency of association, specificity of the association, and the 
biological plausibility of the association for assessing whether 
evidence indicates a causal relationship between an adverse event and 
vaccine exposure (Ref. 15). These criteria were also used in other more 
recent vaccine safety reviews performed by the IOM in 2001 through 2004 
(Ref. 16).
    (Comment 19) Many comments questioned the role of VAERS.
    (Response) Data from VAERS cannot generally be used to determine if 
a vaccine causes an adverse event, but VAERS data can be useful for 
hypothesis generation. As noted in the AVA labeling, a report of an 
adverse event is not proof that the vaccine caused the event.
    From 1990 through March 31, 2005, approximately 1.3 million 
military personnel received 5.3 million doses of AVA. We evaluated the 
4,370 VAERS reports of adverse events following administration of AVA 
submitted to VAERS from 1990 through August 15, 2005, (4,279 through 
March 31, 2005) using a combination of the techniques described 
previously in this section of this document (e.g., pattern assessment 
using frequency calculations, identification and descriptive analysis 
of case series, assessment of reporting rates for certain clinical 
conditions in the context of available information about background 
incidence rates and risk intervals, and data mining). Based on our 
review, we cannot conclude that there is a causal relationship between 
serious adverse events (other than some injection site reactions and 
some reports of allergic reactions) or deaths and AVA (Ref. 17). 
However, as with any medical product, FDA cannot rule out that some 
rare adverse events could be caused by AVA. As described in our 
response to Comment 21, VAERS data were used, along with other data, to 
develop a list of certain adverse events that were considered for 
further study by the Vaccine Analytic Unit. The Vaccine Analytic Unit 
has selected five topics for initial study to determine whether AVA has 
a causal role in certain serious adverse events. FDA continues to 
perform surveillance and periodic evaluations of adverse event reports, 
and will review post-marketing data from any studies that become 
available to FDA.
    (Comment 20) Some comments on the December 2004 proposal seemed to 
interpret the spontaneously reported adverse events that are listed in 
the AVA labeling as being caused by the vaccine.
    (Response) To make physicians and others aware of what is being 
reported, adverse events are sometimes included in the vaccine labeling 
even though it has not been shown that the vaccine actually caused the 
adverse event. Thus, for AVA, that section of the labeling is preceded 
by the statement, ``The following four paragraphs describe spontaneous 
reports of adverse events, without regard to causality'' to indicate 
that the relationship to the vaccine cannot be determined from the 
information provided in the reports for those events.
    (Comment 21) One comment asked if FDA has required BioPort or DoD 
to conduct focused studies of any safety signals.
    (Response) We encourage and support the expeditious conduct of 
well-designed studies evaluating the relationship between AVA and 
adverse events. The Vaccine Analytic Unit (VAU) was formed as a 
collaboration between DoD and CDC to conduct vaccine post-marketing 
surveillance investigations of AVA and other vaccines using data 
collected by the Defense Medical Surveillance System, which holds 
information on vaccinations, hospitalizations, outpatient visits, 
occupational variables, and demographics for all U.S. military 
personnel. FDA worked with the VAU to develop a list of adverse events 
for further study based on VAERS and other data sources. In 2004, VAU 
participants and a workgroup of the National Vaccine Advisory Committee 
(NVAC) agreed that the VAU's research agenda would include five topics 
for initial study: Systemic lupus erythematosus, optic neuritis, 
arthritis, erythema multiforme, and multiple, near-concurrent 
vaccinations.\9\
---------------------------------------------------------------------------

    \9\Description of the VAU and the topic selection process are 
available at http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.cdc.gov/nip/webutil/about/annual-rpts/ar2005/2005annual-rpt.htm#online
 (click on ``Leadership in Vaccine 

Safety'') and http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://cdc.confex.com/cdc/nic2004/techprogram/session_787.htm
.

---------------------------------------------------------------------------

    (Comment 22) Some comments suggested that new clinical studies be 
conducted using anthrax spores milled to a fine powder or using all 60 
strains of anthrax. Others asked why it would

[[Page 75192]]

be unethical to conduct additional human efficacy studies.
    (Response) It is generally accepted that due to the significant 
health risks associated with exposure to anthrax spores, it would not 
be ethical to actively expose human study subjects to B. anthracis 
spores in order to assess the effectiveness of an anthrax vaccine in a 
controlled clinical trial. Furthermore, naturally occurring anthrax is 
now so rare that a field study of vaccine effectiveness is no longer 
feasible in the United States. For any future effectiveness studies, it 
is likely that the efficacy studies will need to be conducted in well-
characterized animal models with an appropriate bridge to human 
immunogenicity data as described under the ``animal efficacy rule''\10\ 
where human efficacy studies are not feasible or ethical (Sec. Sec.  
601.90 and 601.91(a)).
---------------------------------------------------------------------------

    \10\New Drug and Biological Drug Products; Evidence Needed to 
Demonstrate Effectiveness of New Drugs When Human Efficacy Studies 
Are Not Ethical or Feasible; Final Rule (21 CFR 601.90 through 
601.95) (67 FR 37988, May 31, 2002).
---------------------------------------------------------------------------

C. Comments Describing Adverse Events

1. Review of Adverse Event Reports Submitted to the Docket\11\
---------------------------------------------------------------------------

    \11\Docket Number 1980N-0208.
---------------------------------------------------------------------------

    (Comment 23) Many comments to the docket described adverse events 
stated to have occurred following administration of AVA. For 
approximately 111 individuals, information was provided to the docket 
about specific adverse events experienced by the person filing the 
comment, a family member, or another person. Several comments indicated 
that a report about the adverse event had been submitted previously to 
VAERS. However, most of these comments did not mention whether a report 
to VAERS had been submitted.
    (Response) The comments submitted to the docket for the December 
2004 proposal described adverse events after administration of AVA in 
approximately 111 individuals. Multiple submissions were received for 
some individuals. To facilitate analysis of this information and to 
compare the comment reports with other VAERS reports, we entered into 
VAERS the adverse events reported in comments to the extent possible 
based on the information provided. Comments to the docket that reported 
only non-specific adverse events such as became ``ill'' or had a ``bad 
reaction'' were not entered into VAERS because of the lack of adequate 
specificity. Also, submissions that described groups of persons, 
adverse event statistics, or otherwise lacked key individual-level 
details used in VAERS, were not entered into VAERS, but were reviewed 
and considered.
    More than one source (e.g., health care provider, patient, and 
manufacturer) might submit to VAERS information concerning a single 
individual's adverse events following a particular vaccination date, 
resulting in multiple reports. Routine report processing in VAERS 
includes steps aimed at identifying and linking such related reports. 
Using these processes, we found that 48 (43 percent) of the individuals 
described in adverse event reports submitted in comments to the docket 
were the subjects of reports previously entered into VAERS.
    We categorized 106 of the 111 reports as serious, including 6 
deaths. Most described one or more chronic symptoms or illnesses, 
though the duration was not always evident. VAERS reports had 
previously been received for two of the persons who died.
2. Summary of Adverse Event Reports Submitted to the Docket
    The adverse event reports submitted to the docket did not provide 
substantially different information about possible new safety signals 
than the previous reports to VAERS. The previous reports to VAERS, 
together with the reports to the docket, do not establish a causal 
relationship between death or serious adverse events (other than some 
injection site reactions and some reports of allergic reactions) and 
AVA (Ref. 17). We entered into the VAERS database the conditions 
described in comments to the docket. These conditions will be 
considered along with all other adverse event reports received through 
continuing surveillance and incorporated into the periodic evaluations 
of these reports.

D. Comments on the Vaccine Used in the Studies

    (Comment 24) Several comments raised issues about the versions of 
vaccine used in the Brachman study, the CDC open label safety study, 
and the vaccine made by MDPH at the time of licensure.
    (Response) While the December 2004 proposal discussed the 
historical development of AVA, in light of the comments received, we 
believe that additional clarification of the historical development is 
warranted. In the 1950s, Brachman, et al., conducted a well-controlled 
field study in four woolen mills in the United States using DoD vaccine 
provided by Dr. G. G. Wright of Fort Detrick, U.S. Army (Ref. 1). This 
vaccine was produced from the growth of a nonencapsulated, 
nonproteolytic mutant (R1-NP) of the Vollum strain of B. anthracis 
using an aerobic culture method and evaluated for potency (i.e., 
ability to protect test animals against challenge with virulent B. 
anthracis spores) (Ref. 7).
    In the early 1960s, subsequent to completion of the Brachman study, 
DoD modified the vaccine manufacturing process to, among other things, 
optimize production of a stable and immunogenic formulation of vaccine 
antigen and to increase the scale of production. These changes included 
a change in the mutant B. anthracis strain (V770-NP1-R) used to produce 
the vaccine and use of an anaerobic culture method (Refs. 18 and 19). 
These changes coincided with initiation of a contractual agreement 
between DoD and Merck Sharp & Dohme (MSD) to standardize the 
manufacturing process for large-scale production of anthrax vaccine and 
to produce anthrax vaccine using an anaerobic method. Vaccine lots 
manufactured by MSD under this contract were evaluated for potency 
(i.e., ability to protect test animals against challenge with virulent 
B. anthracis spores). One lot of vaccine manufactured by MSD (Merck-9) 
was also used during the first year of the CDC open label safety study.
    In the mid-1960s, DoD entered into a similar contract with MDPH to 
further standardize the manufacturing process and to scale up 
production for further clinical testing and immunization of persons at 
risk of exposure to anthrax spores. This DoD/MDPH/AVA vaccine was made 
using the same strain of B. anthracis as that used under the DoD 
contract with MSD (DoD/MSD vaccine) and similar culture conditions. 
Vaccine lots manufactured by MDPH under this contract with DoD were 
evaluated for potency (i.e., ability to protect test animals against 
challenge with virulent B. anthracis spores). DoD/MDPH/AVA vaccine lots 
were used in the CDC open label safety study. Under the contract with 
DoD, MDPH pursued pre-market approval of the vaccine. The DoD-MDPH 
contract resulted in the production of AVA, which the NIH Bureau of 
Biologics licensed in 1970, FDA now regulates, and BioPort presently 
manufactures.
    The safety and immunogenicity of the three generations of the 
anthrax vaccine were evaluated in three groups of vaccinees, one 
receiving DoD vaccine, another receiving DoD/MSD vaccine, and the third 
group receiving DoD/MDPH/AVA vaccine. Vaccine recipients were monitored 
for local and systemic adverse events. Antibody responses, expressed as 
Geometric Mean Titers and

[[Page 75193]]

percent seropositives, were measured in blood samples collected at 
regular intervals following administration of the third vaccine dose 
utilizing an agar-gel precipitin-inhibition (AGPI) test. These data, 
while limited in the number of vaccinees and samples evaluated, reveal 
that the serological responses to DoD/MDPH/AVA vaccine and DoD vaccine 
were similar with respect to peak antibody response and percent 
seropositives and support our conclusion that data generated by 
administration of DoD and DoD/MSD generations of the vaccines support 
licensure of DoD/MDPH/AVA vaccine.
    (Comment 25) Some comments mentioned that, in the 1985 report, the 
Panel noted that DoD/MDPH/AVA vaccine had not been employed in a 
controlled field study.
    (Response) Although the Panel Report included the statement 
described in Comment 25, the Panel immediately followed with a 
statement that a ``similar'' vaccine was employed in a placebo-
controlled field trial. The Panel then concluded that DoD/MDPH/AVA 
vaccine was ``patterned after'' the vaccine used in that trial (which 
the Panel mistakenly referred to as DoD/MSD vaccine, rather than DoD 
vaccine) ``with various minor production changes.'' (50 FR 51002 at 
51059, December 13, 1985). Thus, the Panel concluded that the Brachman 
study, which used DoD vaccine, supported a finding of safety and 
effectiveness of DoD/MDPH/AVA vaccine. It is common practice for a 
product to undergo manufacturing changes as it moves from initial 
development to product approval. If an earlier generation is 
comparable, then studies using that earlier-produced product are 
relevant to the later product. As we discuss elsewhere in this section 
of this document, the controlled field study using DoD vaccine was 
relevant to DoD/MDPH/AVA vaccine, since the two vaccines were 
comparable in terms of their ability to protect test animals against 
challenge with B. anthracis and to elicit an immune response in humans.
    (Comment 26) One comment stated that FDA is using potency data 
``that it knows are unreliable to assert comparability of two different 
anthrax vaccines [DoD and DoD/MDPH/AVA vaccines]'' and if reliable 
``would only establish comparable animal efficacy for the two vaccines, 
and fail to establish human efficacy, human safety and the 
comparability of the vaccines for humans.''
    (Response) We note here that the comment did not provide evidence 
to support the statement that the potency data are ``unreliable.'' The 
potency data described in the response to Comment 24 demonstrated that 
the products are comparable. In addition, the clinical data described 
in response to Comment 30 demonstrated clinical comparability between 
the vaccines with regard to Geometric Mean Titer and seropositivity 
rates.
    (Comment 27) One comment inquired about whether the differences in 
the versions of AVA resulted in differences in their safety.
    (Response) There are ample clinical data and information from the 
CDC open label safety study, conducted under IND in the 1960s, which 
demonstrate that the DoD/MDPH/AVA vaccine is safe.
    FDA's assessment of vaccine safety considered the data collected 
under the CDC open label safety study (1966 through 1971). During the 
first year of this study, CDC used one lot of DoD/MSD vaccine and one 
lot of DoD/MDPH/AVA vaccine, but only DoD/MDPH/AVA vaccine was used 
during the remainder of the safety study. Thus, the majority of the 
safety data accumulated in that study was from the use of vaccine 
manufactured by MDPH. Information pertaining to the incidence and 
severity of adverse reactions associated with administration of DoD/
MDPH/AVA vaccine was collected for approximately 7,000 individuals 
participating in the CDC open label safety study. In addition, the 
safety of the vaccine is evaluated on an ongoing basis through review 
of new studies, such as the DoD pilot study, and periodic assessments 
of VAERS data.
    (Comment 28) One comment stated that the differences in reported 
systemic reaction rates for the Brachman study and the later DoD pilot 
study indicate that DoD vaccine and DoD/MDPH/AVA vaccine are distinctly 
different such that the effectiveness associated with DoD vaccine 
cannot be regarded as evidence of effectiveness of DoD/MDPH/AVA 
vaccine.
    (Response) We agree that the rates of reported systemic reactions 
associated with administration of anthrax vaccine in the Brachman study 
are lower than the rates reported in the DoD pilot study. However, we 
believe that the Brachman study provided evidence of effectiveness of 
the licensed vaccine. Differences between the Brachman study and the 
DoD pilot study in reported systemic reactions are attributable to a 
number of factors. The latter study was specifically designed to 
closely monitor and solicit subjects' information pertaining to adverse 
reactions associated with administration of the vaccine in accordance 
with the licensed schedule and route of administration so that 
comparisons of adverse reaction rates could be made between the 
licensed schedule and route and the alternative schedules and route 
also under investigation in that study. Differences in methodologies 
and design as well as a heightened awareness and sensitivity toward 
adverse reactions on the part of both study investigators and study 
subjects has resulted in a more comprehensive description of adverse 
reactions experienced in association with vaccination in the more 
recent DoD pilot study.
    As discussed more fully previously in this document, DoD/MDPH/AVA 
vaccine was used in the CDC open label safety study; the production 
strain and culture methods were the same as those currently used by 
BioPort. To provide a more current picture of the types and severities 
of reactions associated with DoD/MDPH/AVA vaccine, the product labeling 
now includes descriptions of adverse events reported in association 
with administration of AVA in the DoD pilot study. Although the 
reporting rates for certain reactions are greater in the DoD pilot 
study, we continue to regard AVA to be safe for its intended use: To 
protect individuals at high risk for anthrax disease. Anthrax disease 
can be fatal despite appropriate antibiotic therapy.
    (Comment 29) One comment stated that the anthrax vaccine produced 
in Michigan has undergone a series of manufacturing changes since it 
was licensed, resulting in a materially altered product that is much 
more concentrated than the original MDPH vaccine.
    (Response) We note that the comment did not provide evidence to 
support the claim that DoD/MDPH/AVA vaccine is ``more concentrated'' 
now than when originally licensed. The DoD/MDPH/AVA vaccine currently 
manufactured by BioPort was licensed in 1970. Since then, the strain of 
B. anthracis used to produce the vaccine has not changed and the 
vaccine formulation has not changed. Changes in the manufacturing 
process (including equipment changes) have been reviewed and approved 
by FDA. Each lot of final vaccine product must pass certain criteria, 
including potency testing, as described subsequently in this document 
in the response to Comment 33.
    (Comment 30) Some comments inquired about whether the change in 
vaccine during the 1962 to 1974 surveillance period altered the 
vaccine's effectiveness. One comment was critical of FDA's assessment 
that both the DoD generation and the DoD/MDPH/AVA

[[Page 75194]]

generation of the vaccine stimulated similar peak antibody responses 
and seropositivity rates since there was not an ELISA assay available 
at the time the antibody responses were measured. The comment argued 
that antibody levels cannot be used as a surrogate marker for 
effectiveness.
    (Response) The antibody responses were measured by agar-gel 
precipitin-inhibition test, which was an acceptable assay. The 
immunogenicity data resulting from this testing showed that the DoD and 
the DoD/MDPH/AVA generations of the vaccine were both immunogenic. 
After the third dose, the peak Geometric Mean Titer for antibodies to 
anthrax was 1.30 (60 percent seropositivity of samples tested) for DoD/
MDPH/AVA vaccine, and 1.4 (60 percent seropositivity of samples tested) 
for DoD vaccine. Thus, while limited in the number of vaccinees and the 
number of samples analyzed, the results do indicate comparable immune 
responses with regards to seropositivity rates and peak antibody titer 
levels (GMT). Rather than representing a surrogate for effectiveness, 
these results are a means of bridging the immunogenicity of these 
generations of the vaccine. In any event, the CDC surveillance data, 
which were gathered when the DoD/MDPH/AVA and DoD/MSD generations of 
the vaccine were in use, corroborate the efficacy data provided by the 
Brachman study.
    (Comment 31) Some comments inquired whether the DoD pilot study or 
a larger ongoing CDC study are intended to provide data to reduce the 
vaccine dose level. Another comment asked how FDA has validated the 
current dose and inoculation schedule.
    (Response) The DoD pilot study was followed by a larger, more 
statistically robust and significant CDC study in order to obtain 
safety and immunogenicity data to support a reduction in the total 
number of doses to be administered in a complete vaccination schedule. 
The new CDC study is a double-blind, randomized, placebo controlled 
trial conducted under IND to compare the licensed AVA schedule and 
route of administration (subcutaneous) to regimens with a different 
route of administration (intramuscular) and/or reduced number of doses. 
Safety and immunogenicity are assessed. The study started in May 2002 
and is currently ongoing. The clinical studies referenced in the 
comment were not intended to seek a change in the amount of vaccine 
administered with each dose. The current dosage for AVA is 0.5 mL per 
inoculation and has been used for anthrax vaccine since before the 
Brachman study was conducted in the 1950s. The current 0.5 mL dosage 
and 6-dose regimen and schedule are based on the dosage, regimen, and 
schedule used in the Brachman study.
    (Comment 32) One comment noted that there would have been no need 
to continue to develop newer and different anthrax vaccines had 
Brachman's vaccine produced acceptable safety and efficacy.
    (Response) On the contrary, DoD (in particular, the Army, Dr. G. G. 
Wright and his colleagues) pursued improvements in the manufacturing 
process, formulation, and other aspects of anthrax vaccine precisely 
because it had been shown to be safe and effective in the Brachman 
study. The changes implemented with the transfer of production to MSD 
and then to MDPH were with the intent of increasing ease of production 
and yield to support further study and ultimately licensure of the 
vaccine. FDA encourages license holders to embrace continuous 
improvement.

E. Comments about Allegedly Contaminated Vaccine and Inspectional 
Observations

    (Comment 33) Some comments asserted that AVA is contaminated or 
adulterated, citing FDA inspections of the Michigan Biologic Products 
Institute (MBPI, and then BioPort) facility. Some comments expressed 
concerns about particular lots of AVA received by soldiers in the U.S. 
military, stating that they were not made under current good 
manufacturing practice (cGMP) or were contaminated.
    (Response) FDA has a lot release program to determine whether lots 
of the AVA licensed vaccine meet criteria for release, which include 
sterility, general safety, potency, and specified levels of 
benzethonium chloride, aluminum, and formaldehyde. All lots released 
from the manufacturer for administration to military personnel and 
other individuals met these criteria.
    Additionally, FDA performs inspections of all biological product 
license holders biennially and at additional times when FDA deems that 
more regulatory oversight is warranted. On the basis of such 
inspections, FDA issued to AVA's manufacturer a Warning Letter in 1995, 
and a Notice of Intent to Revoke the license to manufacture all 
products, including AVA, in 1997. FDA did not initiate license 
revocation proceedings because BioPort committed to and implemented 
appropriate corrective and preventive actions to address the issues 
identified by FDA and demonstrated over time its commitment to comply 
with all applicable FDA requirements. BioPort did this by, among other 
things, renovating its AVA manufacturing facility, discontinuing the 
manufacture and distribution of all non-AVA products, closing its 
aseptic filling facility, and moving the AVA filling operations to a 
contract manufacturer. We believe that the manufacture of AVA is 
currently in compliance with regulatory requirements. We continue to 
evaluate the production of AVA to assure compliance with applicable 
federal standards and regulations.
    (Comment 34) A number of comments alleged that squalene had been 
added to AVA and questioned how AVA could be approved when it contains 
squalene. Others claimed that health problems reported by some 
recipients of AVA were caused by squalene. Another comment noted the 
finding of small amounts of squalene in samples of AVA tested by FDA 
and advocated the testing of all lots of AVA for the presence of 
squalene. One comment claims that squalene ``overcharges'' the immune 
system when injected into the body even in tiny amounts.
    (Response) Squalene is a naturally occurring biodegradable oil 
found in plants, animals, and humans. Squalene is an intermediate in 
the cholesterol biosynthetic pathway and is a natural constituent of 
dietary products including both vegetable and fish oils. Squalene is 
synthesized in the liver and circulates in the bloodstream and is 
present in human serum at 250 parts per billion (250 nanograms per 
milliliter) (Ref. 20). Antibodies to squalene occur naturally in 
humans, have an increased prevalence in females, are not correlated 
with vaccination with AVA, and appear to increase in prevalence with 
age (Ref. 21). Squalene is not used in the AVA manufacturing process 
and is not a component of the vaccine.
    In 1999, FDA performed testing to determine whether squalene was 
added to AVA as an adjuvant. FDA believes that the testing was adequate 
for the intended purpose of determining whether squalene had been added 
to AVA as an adjuvant, and demonstrated that this was not the case. The 
values reported from FDA's testing of certain lots were minute (10 to 
83 parts per billion, which is below the low levels normally detected 
in human serum (Ref. 20)) and at the low end of the analytical 
sensitivity of the test method. Given the extremely low level detected, 
more extensive testing and validation would be needed to ascertain 
whether any squalene was actually present.
    At DoD's request, Stanford Research International (SRI) conducted 
testing designed to detect low levels of impurities (including 
squalene), in a

[[Page 75195]]

quantitative manner. SRI detected squalene at up to 9 parts per billion 
in 1 lot only of the 33 lots of AVA tested. This value can be 
contrasted with the amount of squalene added as a component of MF59 
adjuvant included in FLUAD, an influenza vaccine which is marketed in 
many European countries and whose safety has been evaluated by European 
regulatory authorities. (The current version of this adjuvant is 
technically named MF59C.1.) According to the ``Summary of Product 
Characteristics,'' the amount of squalene contained in FLUAD is 9.75 mg 
per dose of 0.5 mL (about 2 parts per hundred or 20 million parts per 
billion), which is greater than 2 million times more than that detected 
by SRI in one lot of AVA.
    We do not believe that additional testing of AVA is warranted 
because squalene is not used in the manufacturing process and is not a 
component of the vaccine. Moreover, at this time, we reviewed the 
evidence and conclude that such minuscule amounts of squalene, if even 
present in AVA, would not alter our view of the safety of AVA. The 
comment claiming that squalene overcharges the immune system did not 
provide any data in support of this assertion.
    (Comment 35) Some comments noted that AVA contains formaldehyde.
    (Response) The comments are correct in that formaldehyde, at a 
concentration of 100 microgram/mL, is included in AVA as a 
preservative. We note that formaldehyde has been used in the 
manufacture and formulation of AVA since MDPH started manufacturing AVA 
in the 1960s. Formaldehyde was present in the vaccine lots used in the 
CDC open label safety study and, in similarly small amounts, is a 
component of numerous other injectable products. The presence of 
formaldehyde in these small amounts does not alter our view of the 
safety of AVA.
    (Comment 36) One comment was critical of the CDC open label safety 
study claiming that activities described in a program report for work 
conducted under contract with DoD indicated that some lots of anthrax 
vaccine used in the CDC open label safety study were adulterated with 
formaldehyde because additional formaldehyde was added.
    (Response) The report referenced by this comment was written by 
Merck Sharp & Dohme (MSD). It noted that additional formaldehyde was 
added to DoD/MSD vaccine Lots 5 and 7, which were not used in the CDC 
open label safety study. One lot of DoD/MSD vaccine (Lot 9) was used in 
that study. It was used during the first year of the CDC open label 
safety study, along with one lot of DoD/MDPH/AVA vaccine; thereafter, 
only DoD/MDPH/AVA vaccine lots were used. Accordingly, the CDC open 
label safety study was unaffected by the lots that the comment cites.

F. Comments on Labeling

    (Comment 37) Some comments noted the Panel statement regarding an 
apparent discrepancy between the labeling and a now rescinded section 
of the Code of Federal Regulations with regards to the number of doses 
to be administered.
    (Response) We addressed this issue in section III.E of this 
document. The dosing schedule for AVA, from the time of the Brachman 
study to the present, has always consisted of six doses; a 0.5 mL dose 
at 0, 2, 4 weeks and then at 6, 12 and 18 months, followed by a 
subsequent 0.5 mL dose at 1-year intervals to maintain immunity. In any 
event, perceived variances to a rescinded regulation are not relevant 
to this final order under Sec.  601.25, where we determine that AVA is 
appropriately placed into Category I, as a vaccine that is safe, 
effective, and not misbranded.
    (Comment 38) One comment questioned the need for a six-dose 
immunization schedule referencing studies in animals where two doses of 
vaccine administered 2 weeks apart protected non-human primates from 
inhalation challenge with anthrax spores up to 104 weeks later.
    (Response) The current immunization schedule described in the AVA 
labeling was demonstrated to be effective in the Brachman study. That 
schedule consists of a total of six doses of 0.5 mL administered 
subcutaneously at 0, 2, 4 weeks, 6, 12 and 18 months with annual 
boosters thereafter to maintain immunity. Changes to this vaccination 
schedule may be reviewed and considered for approval by FDA based upon 
the submission of scientific data to support changes to the product 
labeling.

G. Additional Comments

    (Comment 39) Several comments were critical of FDA for ``relying'' 
upon the IOM report as the scientific basis for placing AVA into 
Category I and were critical of the IOM report with respect to its 
consideration of studies conducted by DoD as supportive of vaccine 
safety or its consideration of animal studies as evidence of 
effectiveness against inhalation anthrax. However, other comments 
stated that FDA was ``somewhat indirect'' regarding the IOM report and 
suggested that FDA ``accord the IOM report significant weight as expert 
scientific judgment.''
    (Response) In the December 2004 proposal, we agreed with the IOM 
committee's general conclusion that AVA, as licensed, is an effective 
vaccine for protection of humans against anthrax infection, including 
inhalation anthrax and that certain studies in humans and animals 
support the conclusion that AVA is effective against B. anthracis 
strains that are dependent upon the anthrax toxin as a mechanism of 
virulence, regardless of the route of exposure. In response to the 
comments submitted regarding the IOM committee report, we wish to 
clarify that the general conclusions of the report are consistent with 
FDA's own independent assessment of the available data regarding the 
safety and effectiveness of AVA.
    In response to public concerns expressed about the use of AVA in 
the DoD's Anthrax Vaccine Immunization Program, Congress called for DoD 
to support an independent examination of AVA by the IOM. The IOM 
committee was charged with reviewing data regarding the effectiveness 
and safety of the currently licensed anthrax vaccine and assessing the 
manufacturer's efforts to resolve manufacturing issues and resume 
production and distribution of vaccine.
    While the IOM committee did invite FDA scientists to participate in 
their open meetings and comment on portions of the draft report, FDA 
was not a participant in their closed review sessions, nor did FDA 
participate in the writing or finalization of the IOM report. 
Similarly, FDA has conducted its review under Sec.  601.25, culminating 
in this final order, independently of the activities of the IOM 
committee. FDA did not actively seek input or comment from the IOM 
committee during its review process.
    (Comment 40) Some comments questioned the utility of animal data 
with one comment stating that animal testing is ``absolutely not at all 
relevant to the study of safety for humans.'' Another comment noted 
that AVA provided protection in guinea pigs against spores of some 
strains of B. anthracis but not others.
    (Response) We wish to clarify that animal studies have not been 
relied upon for a determination of the safety of AVA for human use. The 
safety database is comprised of data from the CDC open label safety 
study in the late 1960s to early 1970s during which approximately 
15,000 doses manufactured at MDPH were administered to approximately 
7,000 subjects. In addition, safety data from the DoD pilot study (Ref. 
13) and adverse reactions reported to VAERS as associated with 
administration of AVA were considered as part of FDA's continual 
process for assessing the

[[Page 75196]]

safety of AVA. In 2002, information from the DoD pilot study and VAERS 
were included in the sections of the labeling describing safety and 
adverse reactions. We continue to perform periodic evaluations of 
adverse events reported to VAERS.
    With regard to data suggesting that the vaccine protected guinea 
pigs against spores from some strains of B. anthracis but not others, 
we note that different animal species may exhibit different levels of 
susceptibility to an infectious organism. The course of infection and 
disease may depend greatly upon the strain of the infectious organism 
for some species but not so much for other species (Refs. 3, 4, and 5). 
Thus, based on the strain used or other factors, studies in some animal 
species are likely to produce different results than studies in other 
species.
    (Comment 41) One comment suggested that AVA had been administered 
to military personnel during Desert Storm/Desert Shield under an IND.
    (Response) NIH's Division of Biologics Standards originally 
licensed AVA under the Public Health Service Act in 1970. 
Administration of AVA, an approved product, to military personnel by 
DoD during Desert Storm/Desert Shield was not under an IND.
    (Comment 42) Many comments claimed that AVA was not properly 
licensed.
    (Response) We disagree. AVA has been legally licensed since 
November 1970.
    The purpose of the biologics efficacy review procedures is to 
determine whether biological products licensed before July 1, 1972, are 
safe and effective and not misbranded. In 1972, the Department of 
Health, Education, and Welfare redelegated from the NIH to FDA 
authority and responsibility to regulate biological products. FDA 
initiated a comprehensive review of the safety, effectiveness, and 
labeling of all licensed biologics, including AVA, shortly after the 
redelegation of authority. In keeping with Sec.  601.25, independent 
advisory panels made up of scientific experts from outside the Federal 
Government, reviewed biological products licensed prior to July 1, 
1972, in order to recommend to FDA how the agency should classify the 
products. One panel reviewed the safety, effectiveness, and labeling of 
AVA and recommended that FDA place the vaccine into Category I--safe, 
effective, and not misbranded. This recommendation was based on a 
review of the available data from the Brachman study and the CDC open 
label safety study, and the CDC surveillance data, as described 
elsewhere in this document. FDA followed the requirements of Sec.  
601.25(f), requiring publication of a proposed order for 
classification, and published a proposed rule in the Federal Register 
on December 13, 1985 (50 FR 51002). Since the publication of the 
December 1985 proposal, FDA has focused on removing Category II 
products-unsafe, ineffective, or misbranded, from the market and 
completing the final classification of the Category III products-
products with insufficient information to allow classification and 
further testing is required. The purpose of this final order, and the 
final rule and final order published elsewhere in this issue of the 
Federal Register, is to complete FDA's categorization of bacterial 
vaccines and toxoids licensed prior to July 1, 1972. As stated in 
section III of this document, FDA concludes that AVA is safe, 
effective, and not misbranded.
    (Comment 43) Some comments questioned why FDA did not reconvene an 
advisory review panel when it reopened the comment period in response 
to the Court order of October 27, 2004. The comments claim that FDA has 
attempted to avoid the normal approval process or circumvented its own 
rules by not convening an advisory review panel to review new data 
generated by DoD.
    (Response) Neither the applicable FDA regulation, Sec.  601.25, nor 
the Court's order of October 27, 2004, requires that an advisory review 
panel be convened at this time. FDA regulations at Sec.  601.25 
explicitly detail the procedures to be used to determine that 
biological products licensed prior to July 1, 1972, are safe, 
effective, and not misbranded. These regulations require FDA to submit 
a product to an advisory review panel at the initiation of the review. 
The panel then submits to the Commissioner of Food and Drugs a report 
containing the panel's conclusions and recommendations with respect to 
the biological product. The Commissioner, after reviewing the 
conclusions and recommendations, then publishes a proposed order 
categorizing the product as safe and effective (Category I), unsafe or 
ineffective (Category II), or determining that the available data are 
insufficient to classify such biological product (Category III). 
Thereafter, any interested person may within 90 days after publication 
of the proposed order, file written comments. After review of the 
comments, the Commissioner of Food and Drugs publishes a final order on 
the classification.
    In Doe v. Rumsfeld, 341 F.Supp.2d 1 (D.D.C. 2004), the Court 
examined the step in the process involving the opportunity for public 
comment on the agency's proposed order. The court noted that FDA had 
published the Panel report in its entirety as a proposed order. 
However, the Court concluded that the proposed order did not provide 
public notice that FDA considered the vaccine to be indicated for use 
against inhalation anthrax, a conclusion that FDA made in its January 
2004 final order. Accordingly, the Court remedied what it considered to 
be an Administrative Procedure Act violation, by vacating the January 
2004 final order, and remanding it to FDA to reconsider following an 
additional opportunity for comment. The Court did not find fault with 
the Panel report. FDA believes that, with the requirements of Sec.  
601.25 satisfied with respect to the advisory review panel report, it 
is not necessary to consult another advisory panel on these issues. In 
drafting this final order, FDA has been able to review and consider 
extensive comments on the December 2004 proposed order.
    (Comment 44) Some comments expressed concern that certain Panel 
members were also involved in developing AVA. They suggest that the 
members were biased, and their role in the review process self-serving. 
One comment specifically complained of the bias of Dr. Stanley Plotkin, 
who was a co-author on the Brachman study (Ref. 1).
    (Response) As provided in Sec.  601.25, the Commissioner appointed 
qualified experts to serve on the advisory review panel and the Panel 
included persons from lists submitted by organizations representing 
professional, consumer, and industry interests. A review of the Panel 
members appointed to review the data and information and to prepare a 
report on the safety, effectiveness, and labeling of bacterial 
vaccines, toxoids, related antitoxins, and immune globulins reveals 
that the list did not include the name of Dr. Stanley Plotkin or any 
other scientist who worked directly with the development of AVA. (50 FR 
51002 at 51003 (December 13, 1985)).
    (Comment 45) One comment alleged that FDA and DoD had a conflict of 
interest and that the agencies were working together to promote 
vaccinations.
    (Response) FDA is charged with implementing the Federal Food, Drug, 
and Cosmetic Act, as well as certain provisions of the Public Health 
Service Act. Under these authorities and applicable regulations, 
including Sec.  601.25, FDA is responsible for reviewing the safety and 
effectiveness of vaccines. In issuing this order, FDA is

[[Page 75197]]

fulfilling this responsibility, and is not working to promote, or 
discourage, vaccination for members of the armed forces. Rather, as 
described in this order, FDA has evaluated AVA and concluded that the 
product is safe, effective, and not misbranded.
    (Comment 46) Other comments expressed concern that FDA had not 
considered alternatives to vaccination such as the use of detection 
devices and antibiotics to protect individuals from anthrax infection, 
or expressed the opinion that antibiotics are a better means of 
protection against anthrax.
    (Response) Detection devices, if effective, would not prevent 
infections, but would simply detect the presence of anthrax spores in 
the environment. Moreover, a device would provide this information only 
for the particular location under observation by the device and only if 
the device was in use and functioning properly at the time.
    Moreover, although antibiotic therapies are safe and effective in 
the treatment of anthrax disease and in the prevention of anthrax 
disease when administered as part of a post-exposure prophylaxis 
regimen, the safety and effectiveness of long term use of such 
therapies in individuals at high risk for anthrax disease, potentially 
for a period of years, has not been studied. Moreover, the early stages 
of inhalation anthrax present with flu-like symptoms, and diagnosis may 
be delayed. The initiation of antibiotic therapy only after a 
definitive diagnosis of inhalation anthrax has a diminished success 
rate. Anthrax disease can be fatal despite the use of antibiotics. The 
fatality rate for inhalation anthrax in the United States is estimated 
to be approximately 45 percent to 90 percent. From 1900 to October 
2001, there were 18 identified cases of inhalation anthrax in the 
United States, the latest of which was reported in 1976, with an 89 
percent (16/18) mortality rate. Most of these exposures occurred in 
industrial settings, i.e., textile mills. From October 4, 2001, to 
December 5, 2001, a total of 11 cases of inhalation anthrax linked to 
intentional dissemination of B. anthracis spores were identified in the 
United States. Five of these cases were fatal (Ref. 6). These 
fatalities occurred despite aggressive medical care, including 
antibiotic therapy (Refs. 22 and 23).
    Thus, we have considered possible alternatives to AVA, and continue 
to conclude that AVA is safe, effective, and not misbranded.

H. Comments on Matters Outside the Scope of this Proceeding

    (Comment 47) We received numerous comments on the December 2004 
proposal that, although they relate to significant issues, are not 
relevant to the proposed order for placing AVA into Category I. These 
comments concerned: (1) The need for compensation programs for 
individuals injured by AVA, (2) statements that the vaccine should be 
optional for members of the armed forces, (3) statements that antidotes 
to anthrax should be developed, (4) concerns about DoD responsibilities 
and recordkeeping, and (5) requests for an investigation of BioPort 
stock ownership.
    (Response) These comments are on matters outside the scope of this 
final order and FDA's jurisdiction, authority, and control. 
Accordingly, we do not respond to them.

V. FDA's Responses to Additional Panel Recommendations

    In the December 1985 proposal, FDA responded to the Panel's general 
recommendations regarding the products under review and to the 
procedures involved in their manufacture and regulation, and to the 
Panel's general research recommendations. Published elsewhere in this 
issue of the Federal Register in a final rule and final order 
concerning bacterial vaccines and toxoids other than AVA, FDA responds 
in final to the Panel's general recommendations.

VI. References

    The following references have been placed on display in the 
Division of Dockets Management (HFA-305), Food and Drug Administration, 
5630 Fishers Lane, rm.1061, Rockville, MD 20852, and may be seen by 
interested persons between 9 a.m. and 4 p.m., Monday through Friday. 
(FDA has verified the Web site address, but we note subsequent changes 
to the Web site might have occurred after this document publishes in 
the Federal Register).
    1. Brachman, P. S., H. Gold, S. A. Plotkin, F. R. Fekety, M. 
Werrin, and N. R. Ingraham, ``Field Evaluation of a Human Anthrax 
Vaccine,'' American Journal of Public Health, 52:632-645, 1962.
    2. Institute of Medicine, ``The Anthrax Vaccine, Is It Safe? 
Does It Work?'' Committee to Assess the Safety and Efficacy of the 
Anthrax Vaccine, Medical Follow-Up Agency, Washington, DC: National 
Academy Press, 2002, http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.nap.edu/catalog/10310.html.

    3. Fellows, P. F., M. K. Linscott, B. E. Ivins, M. L. M. Pitt, 
C. A. Rossi, P. H. Gibbs and A. M. Friedlander, ``Efficacy of a 
Human Anthrax Vaccine in Guinea Pigs, Rabbits, and Rhesus Macaques 
Against Challenge by Bacillus Anthracis Isolates of Diverse 
Geographical Origin,'' Vaccine 19(23/24):3241-3247, 2001.
    4. Ivins, B. E., P. F. Fellows, M. L. M. Pitt, J. E. Estep, S. 
L. Welkos, P. L.Worsham, and A. M. Friedlander, ``Efficacy of a 
Standard Human Anthrax Vaccine Against Bacillus Anthracis Aerosol 
Spore Challenge in Rhesus Monkeys,'' Salisbury Medical Bulletin 
87(Suppl.):125-126, 1996.
    5. Ivins, B. E., M. L. M. Pitt, P. F. Fellows, J. W. Farchaus, 
G. E. Benner, D. M. Waag, S. F. Little, G. W. Anderson, Jr., P. H. 
Gibbs, and A. M. Friedlander, ``Comparative Efficacy of Experimental 
Anthrax Vaccine Candidates Against Inhalation Anthrax in Rhesus 
Macaques,'' Vaccine, 16(11/12):1141-1148, 1998.
    6. Anthrax Vaccine Adsorbed (BIOTHRAX) Package Insert (January 
31, 2002).
    7. Wright, G. G., T. W. Green, and R.G. Kanode, Jr., ``Studies 
on Immunity in Anthrax: V. Immunizing Activity of Alum-Precipitated 
Protective Antigen,'' Journal of Immunology, 73:387-391, 1954.
    8. ``FDA Guidance Concerning Demonstration of Comparability of 
Human Biological Products, Including Therapeutic Biotechnology-
derived Products,'' April 1996, http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.fda.gov/cber/gdlns/comptest.pdf
.

    9. ``Revocation of Certain Regulations; Biological Products,'' 
Final Rule; 61 FR 40153, August 1, 1996.
    10. ``International Conference on Harmonisation; Guidance on 
Statistical Principles for Clinical Trials,'' Notice of 
Availability; 63 FR 49583, September 16, 1998.
    11. Plotkin, S. A., P.S. Brachman, M. Utell, F. H. Bumford, and 
M. M. Atchinson, ``An Epidemic of Inhalation Anthrax, the First in 
the Twentieth Century, I. Clinical Features.'' American Journal of 
Medicine, 29:992-1001, 1960.
    12. Brachman, P.S., S. A. Plotkin, F. H. Bumford, and M. M. 
Atchinson, ``An Epidemic of Inhalation Anthrax: The First in the 
Twentieth Century, II. Epidemiology.'' American Journal of Hygiene; 
72:6-23, 1960.
    13. Pittman, P. R., G. Kim-Ahn, D. Y. Pifat, K. Coonan, P. 
Gibbs, S. Little, J. G. Pace-Templeton, R. Myers, G. W. Parker, and 
A. M. Friedlander, ``Anthrax Vaccine: Immunogenicity and Safety of a 
Dose-Reduction, Route-Change Comparison Study in Humans,'' Vaccine; 
20(9-10):1412-1420, 2002.
    14. ``Guidance for Industry: Good Pharmacovigilance Practices 
and Pharmacoepidemiologic Assessment,'' March 2005, http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.fda.gov/cber/gdlns/pharmacovig.htm
.

    15. Institute of Medicine, ``Adverse Effects of Pertussis and 
Rubella Vaccines,'' A Report of the Committee to Review the Adverse 
Consequences of Pertussis and Rubella Vaccines. Washington, DC, 
National Academy Press, 1991.
    16. Institute of Medicine. ``Immunization Safety Review.'' 
http://frwebgate.access.gpo.gov/cgi-bin/leaving.cgi?from=leavingFR.html&log=linklog&to=http://www.iom.edu/project.asp?id=4705.


[[Page 75198]]

    17. Review of VAERS Anthrax Vaccine Reports Received Through 8/
15/05, and Adverse Event Reports Submitted to Docket No. 1980N-0208; 
dated December 2005.
    18. Puziss, M., L. C. Manning, J. W. Lynch, E. Barclay, I. 
Abelow, and G. G. Wright, ``Large-Scale Production of Protective 
Antigen of Bacillus anthracis in Anaerobic Cultures,'' Applied 
Microbiology, 11(4):330-334, 1963.
    19. Wright, G. G., M. Puziss, and W. B. Neely, ``Studies on 
Immunity in Anthrax, IX. Effect of Variations in Cultural Conditions 
on Elaboration of Protective Antigen by Strains of Bacillus 
anthracis,'' Journal of Bacteriology, 83:515-522, 1962.
    20. Nikkila K., K. H[ouml]ckerstedt, and T. A. Miettinen, 
``Serum and Hepatic Cholestanol, Squalene and Noncholesterol Sterols 
in Man: A Study on Liver Transplantation,'' Hepatology, 15:863-70, 
1992.
    21. Matyas, G. R., M. Rao, P. R. Pittman, R. Burge, I. E. 
Robbins, N. M. Wassef, B. Thivierge, and C. R. Alving, ``Detection 
of Antibodies to Squalene III. Naturally Occurring Antibodies to 
Squalene in Humans and Mice,'' Journal of Immunological Methods, 
286: 47-67, 2004.
    22. Jernigan, J. A., et al., ``Bioterrorism-Related Inhalational 
Anthrax: The First 10 Cases Reported in the United States,'' 
Emerging Infectious Diseases, 7(6):933-944, 2001.
    23. Barakat, L. A., et al., ``Fatal Inhalational Anthrax in a 
94-Year-Old Connecticut Woman,'' Journal of the American Medical 
Association, 287(7):863-868, 2002.

    Dated: December 12, 2005.
Jeffrey Shuren,
Assistant Commissioner for Policy.
[FR Doc. 05-24223 Filed 12-15-05; 8:45 am]

BILLING CODE 4160-01-S