s
`
`PUBLIC HEALTH
`1a d w
`Official Monthly Publication of the American Public Health Association, Inc.
`
`w
`
`Volume 46
`
`January, 1956
`
`Number I
`
`Poliomyelitis Vaccine in the Fall of 1955
`JONAS E. SALK, M.D., F.A.P.H.A.
`
`This paper reflects Dr. Salk's con-
`clusions growing out of the 1955
`polio-
`practical
`with
`experience
`myelitis vaccine.
`
`I
`
`The results of the experiences with
`poliomyelitis vaccine in 1954, and again
`in 1955, take on much more meaning
`when they are viewed in the light of the
`underlying theory and the basic experi-
`mentation upon which the practices of
`each year were based.
`It is understandable that, as a result
`of the episode of last spring,1' 2 a promi-
`nent doubt developed and still continues
`to
`exist concerning the fundamental
`safety of a killed-virus vaccine.
`This is
`true even in view of the earlier work 3 4
`and of the field trial evidence,5 and even
`though there has accumulated from the
`experience in three countries6-8
`evi-
`dence for the fundamental safety
`of
`In view of
`properly prepared vaccine.
`the doubt, then, "How does one ascer-
`tain vaccine safety?"
`
`Determination of Safety
`The principal objective is to elim-
`inate, for all practical purposes, the
`
`1
`
`presence of infectious virus from each
`single dose of each batch of vaccine re-
`leased for use.
`This reduces the prob-
`lem to each individual inoculated.
`This
`degree of safety cannot be ascertained
`merely on the basis of a test made on a
`sample of a "finished" batch-no matter
`how large the sample-unless, of course,
`the entire batch is tested and none re-
`mains to be used. Therefore, to provide
`the kind of assurance that is required
`for a vaccine that is made from an agent
`that possesses the potential hazard of
`any strain of poliomyelitis virus, more
`than the usual kind of care in processing
`and in testing is required " than had
`heretofore been applied to other less
`hazardous antigens.
`This particular problem requires that
`tests be applied-not only on samples
`of the finished material, but on samples
`
`Dr. Salk is professor of preventive medi-
`cine, Virus Research Laboratory, School of
`Medicine,
`University
`of
`Pittsburgh,
`Pitts-
`burgh, Pa.
`This manuscript is substantially Dr. Salk's
`contribution to the Special Session on Prac-
`tical Experience with Poliomyelitis Vaccine,
`presented at the Eighty-Third Annual Meeting
`of the American Public Health Association
`in Kansas City, Mo., November 17, 1955.
`The studies referred to were aided by a
`grant from the National Foundation for In-
`fantile Paralysis.
`
`CEV Exhibit 1022_001
`
`

`

`2
`
`JANUARY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH
`
`PROOUCTION
`SEQUENCE OF
`STRAIN POOLS
`( pool no.)
`
`CUMULATIVE VOLUME OF
`STRAIN POOLS PRODUCED
`(I pool 400 liters)
`
`CUMJLATIVE
`VOLUME
`TESTED IN TISSUE CULTURE
`(1000 mr/lot)
`
`RESULT OF TEST
`FOR LIVE VIRUS
`ExarplbS
`Exa0mpbA
`
`400,000 ml
`800,000
`1,200,000
`1,600,000
`2,000,000
`2,400,000
`2,800,000
`3,200,000
`3,600,000
`4,000,000
`4,400,000
`4 ,800,009
`
`23 456 789
`
`10
`II
`'a
`
`0 0 0 0 0 0 0 0 0 0 0
`
`0 0 t0 t t00 0 0
`
`1,000 ml
`2,000
`3,000
`4,000
`5,000
`6,000
`7,000
`8,000
`9,000
`10,000
`11,000
`
`12,000
`
`extended
`
`Ot
`Negotive tests
`Negotive tests
`ore ess signi-
`ore of inmeasmng
`ficant in pro-
`significatce
`as
`portion to Ithe
`nunber of
`experience is
`positives
`Figure I-Significance of a Consistent Negative Record in Production and Testing of
`Monovalent Strain Pools (Assuming Sensitive Testing and Representative Sampling)
`
`removed at several stages in the course
`of processing-to demonstrate the ex-
`tent to which a calculated amount of
`overtreatment has been applied to each
`batch of vaccine, to create what is re-
`ferred to as a "margin of safety." 9 This
`means a degree of treatment beyond
`that calculated to be necessary to care
`for the lingering virus that is expected
`to be the last to be acted upon.
`It was recognized early in this work
`that a single negative test, of itself, was
`inadequate
`establish
`safety
`with
`to
`
`sufficient
`significance, but that a se-
`quence of negative production lots is of
`increasing
`significance.10
`Therefore,
`any single negative test, in an erratic
`experience, is
`of diminishing signifi-
`cance in proportion to the frequency
`with which positives are interspersed
`when, theoretically,
`there
`should be
`negatives.
`This concept is illustrated
`in Examples A and B, in Figures 1 and
`2, which show the influence upon the
`degree of confidence for safety that can
`be expected from a process that results
`
`PRODUCTION
`SEQUENCE OF
`TRIVALENT VACCINE
`(lot no.)
`
`CUMULATIVE VOLUME OF
`TRIVALENT VACCINE
`(I lot = 1200 liters)
`
`CUMULATIVE VOLUME
`TESTED IN TISSUE CULTURE
`(1500
`ld/bt1)
`
`RESULT OF TEST
`FOR LIVE VIRUS
`Example A
`Example B
`
`1,200,000 ml
`
`1,500 ml
`
`2
`
`2,400,000
`
`3,600,000
`
`4,500
`
`0
`
`+
`
`0
`
`0
`
`0
`
`0
`
`4,800,000
`
`6,000
`
`O t
`tive tests
`Ne
`Negative tests
`s signi-
`are
`are oficreaosng
`ficont in pro-
`xigniflcm,ceas
`portion to the
`curniaive
`number of
`experience is
`positives
`exended
`Figure 2-Significance of a Sequence of Negative Tests on Trivalent Vaccine Prepared
`from "Negative" or Presumably Monovalent Pools
`
`CEV Exhibit 1022_002
`
`

`

`in negative tests consistently (Example
`B) as compared with one in which
`positives are encountered from time-to-
`time (Example A).
`The summary in Figure 3 emphasizes
`the elements of a negative safety test:
`(1)
`the reliability of the test system
`itself, and (2)
`the significance
`of a
`negative test for judging safety.
`The
`degree of reliability of any one negative
`test is dependent (a) upon the sensi-
`tivity of the testing system itself, which
`includes volume of sample as well as
`the demonstrated capacity of the system
`to detect trace amounts of infectious
`virus;, and (b) upon physical homoge-
`neity, since the latter can affect the
`validity of sampling, especially if virus
`were present in more than one physical
`phase, i. e., a fluid phase as well as a
`solid phase that could settle out of sus-
`pension. The degree of significance of
`a negative test, for an interpretation of
`safety, depends to a great extent, there-
`fore, upon consistency of performance
`of the manufacturing process.
`Each of the elements that contribute
`to reliability of testing and to signifi-
`cance for safety can be influenced and
`controlled by attention to certain prac-
`tical details that are considered in charts
`that will follow: for example, the time
`at which samples for safety tests are
`
`POLIOMYELITIS VACCINE VOL. 46
`
`3
`
`taken, in relation to the rate at which
`virus is inactivated, as is illustrated in
`Figure
`simplified
`diagram
`This
`4.
`shows that it is possible, routinely, to
`demonstrate the rate at which virus in-
`fectivity is destroyed in the early stages
`of inactivation; and it illustrates the
`way in which the reaction continues, as
`is demonstrable under experimental cir-
`cumstances. On the basis of such ob-
`servations a theoretical projection can
`be made which can be further verified
`by experience accumulated in the course
`testing
`of
`samples
`from
`successive
`batches of vaccine.
`If the samples re-
`moved at any of the intervals indicated
`(i. e., 6, 9, 12 days) are negative for
`infectious virus, and if such were the
`case in a succession of experiences, this
`would tend to validate the conception
`of the theoretical projection.
`This was,
`in fact, demonstrated in the course of
`preparing material for use in the field
`trial."
`The results of the field trial experi-
`ence are illustrated in Figures 5 and 6.
`Figure 5 contains a comparison of the
`original and present interpretations of
`the available facts.
`The difference is
`due to the interpretation of the signifi-
`cance of lesions found in histologic sec-
`tions of monkey CNS tissue. Three lots
`were suspect and, therefore, not used,
`
`DEPENDS UPON
`
`CAN BE
`WILUENCED AND CONTROLLED BY
`
`RELIABILITY
`
`HOMOGENEITY
`AND
`REPRESENTIVE SAMPLNG
`
`*
`
`UNFORMITY OF
`SENSITIVITY OF TESTS
`EMPLOYED
`
`SPECFED PROCESSING PROCEDURES
`
`SPECIFIED TESTING PROCEDURES
`
`SIGNIFICANCE
`
`SEQUENTIAL CONSISTENCY
`OF
`NEGATIVITY
`[
`IN
`PRODUCTION-TESTIG RECORD
`Figure 3-Reliability and Significance of a Negative Safety Test
`
`ATTENTION TO
`THEORETICAL AND PRACTICAL DETAILS
`OF
`PROCESSING AND TESTING
`
`CEV Exhibit 1022_003
`
`

`

`4
`
`JANUARY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH
`
`SIZES OF BATCHES
`PROCESSED FOR
`INACTIVATION
`20,O-400P00 ml
`
`SIZE OF SAMPLE
`SAFETY TESTED N
`TISuIEw
`liuInK
`
`5 ml
`50 ml
`
`BiI
`
`I
`
`10-U
`
`Sam*e
`
`A 1
`
`04e
`
`TC ID
`per 0.5 5.
`IxlO 6
`
`IlxO a
`
`lxO 0
`
`x 10-3
`
`I x 10-6
`x lo-Q I
`lx I0-
`
`EEffect
`
`I'l
`
`E
`n
`Verifed Effect-
`
`X\
`
`Theoreticd
`Projection
`
`10-6
`
`0
`
`3
`
`6
`9
`12
`Time in days at 37 e C
`1:4000 Formnoln ond pH 7.0
`Figure 4-Combination of Processing and Safety Control
`Relationship Between the Measured Rate of Virus Inactivation
`and the Time of Removal of Samples for Safety Test
`
`15
`
`even though the monkeys showed no
`signs of paralysis nor was live polio
`virus found in the tissue culture testL
`The findings in the CNS were incidental
`to the histologic examination of sur-
`viving animals; that the lesions were not
`due to poliomyelitis virus infection is
`the
`opinion
`Dr.
`of
`David
`present
`Bodian, who reviewed the histologic
`sections.
`Although the cumulative experience
`in Figures 5 and 6 suggests that vaccine
`can be made in a consistently reliable
`
`manner and that the theoretical consid-
`erations
`(that were under test in the
`1954 application) * were valid,
`it
`is
`* Reference has been made 1 to the expe-
`rience of a third manufacturer that did not,
`at first, achieve a similar degree of consistency
`during preparation
`of material that might
`have been used in field trial.
`It was recog-
`nized at the time this occurred that certain
`technical factors could explain the early dif-
`ference
`in
`performance
`between manufac-
`turers. The production of this manufacturer
`was not considered for use in the field trial
`because of inconsistency in achieving a nega-
`tive test after treatment for the inactivation
`of virus.
`
`Sequence
`
`1
`
`5
`
`10
`
`15
`
`.20
`
`LabB
`
`Lab A
`
`* O O @
`
`00 0 0 0 0 0 0 0 0 0 0 0 0 0 0
`
`0 O O O O O O O O O O O T
`
`CRIGINAL
`JA33'I;
`
`ON
`
`0O O O O O O O O O O O O O O O O O O
`*
`
`* OO OO O O OO O
`
`l
`
`|Lab B |
`
`LabA
`
`*REVISW
`JO 0Tis
`ION
`*Based on Re-examinAtion
`M = Positive or Suspect
`O m Negative
`of CNS Sections
`Figure 5-Experience of Two Laboratories in Inactivation of Virus for
`1954 Field Trial Vaceine
`
`CEV Exhibit 1022_004
`
`

`

`POLIOMYELITIS VACCINE VOL. 46
`
`5
`
`Lab.
`
`Lots
`No.
`
`Trivalent
`Vaccine
`Volume
`
`Tested in T.C.
`and Monkeys
`Volume
`
`Monkeys
`Inoculated
`No.
`
`A
`
`B
`
`12
`
`20
`
`1, 400,000 ml
`
`3, 312 ml
`
`648
`
`3,000,000 ml
`
`5,520 ml,
`
`Total
`
`32
`
`4',4oo,00
`Figure 6-Accumulated Consecutive Experience of Two Laboratories
`in Preparation of Vaccine for Field Trial Use
`
`ml
`
`8,832 ml
`
`1,728
`
`also known that in the beginning certain
`manufacturers experienced inconsisten-
`This fact has been interpreted
`cies.1
`One interpretation was
`several ways.
`that the theoretical considerations in
`processing and testing were not wholly
`that fundamentally it
`sound 1;
`is not
`possible to prepare a killed vaccine
`safely12; another interpretation is that
`technical factors were responsible for
`the inconsistencies that were observed,
`but that the fundamentals were sound.10
`
`Factors that Could Cause
`Inconsistency in Production
`The fact that not all manufacturers
`experienced the same degree of incon-
`sistency, and that the experiences of
`some did coincide with what was ex-
`pected theoretically tended to support
`the view that the difficulties were due to
`technical
`Further support 13
`factors.
`for this view was derived (1) from an
`analysis of manufacturing practices and
`experiences, (2) from tests upon fluids
`from
`manufacturing
`the
`certain
`of
`laboratories, and (3) from -the effects
`upon the record of consistency, when
`specific attention was directed to certain
`This has been more
`technical details.
`fully discussed by the Public Health
`Service Technical Committee on Polio
`Vaccine in a report 13 to the Surgeon
`General. The discussion that follows is
`-drawn, in part, from the substance of
`this report.
`
`An analysis 13
`information
`of
`on
`manufacturing practices and experiences
`available to the U. S. Public Health
`Service's Technical Committee on Polio-
`myelitis Vaccine suggested the likeli-
`hood that virus incorporated in a solid
`phase, and out of contact with formalin,
`could be responsible for incompleteness
`of
`this
`inactivation;
`surmise
`was
`confirmed by direct test upon fluids
`obtained from different
`A
`sources.
`schematic presentation of these test ob-
`servations, and of their interpretation,
`is shown in Figure 7 in which is illus-
`trated the effect of the existence of virus-
`containing
`sediment
`prevented
`that
`contact between virus and formalin.
`It
`is clear that under such circumstances
`could not be expected
`virus
`to be
`destroyed by the formalin with which
`it had had no contact.
`It was for
`reasons, such as this, that the original
`requirements stated
`that filtration be
`done before inactivation;
`the purpose
`being the removal of particles of such
`size as might incorporate and protect
`virus from interaction with formalin.9
`Still further investigation 13 revealed
`that filtration was not always carried
`out shortly before the inactivation step;
`at times, weeks or months elapsed be-
`tween filtration and formalin treatment.
`During such periods of storage precipi-
`tates tend to form.
`Failure to remove
`such
`precipitates,
`incorporated
`with
`virus, could result in incomplete inter-
`action between formaldehyde and all of
`
`CEV Exhibit 1022_005
`
`

`

`6
`
`JANUARY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH
`
`oI
`
`a-a11
`
`12
`15
`9
`6
`Days at 37 C
`1:4000 Formalin pH 7.0
`Figure 7-Effect of Virus-Formaldehyd(
`Virus Infectivity
`
`0
`
`3
`
`6
`12
`9
`Days at 37 C
`
`5
`
`Contact Upon Rate of Destruction of
`
`the virus and could readily account for
`incompleteness of inactivation.
`Other facts that emerged from these
`inquiries were that there existed differ-
`ences in the kind of filter system used,
`as well as differences in filtration prac-
`tices, so far as elapsed time of storage
`between the harvesting of fluids and
`filtration.
`These
`details
`are
`to
`be
`described by the Technical Committee
`whose study
`correlated
`these
`obser-
`vations.
`
`Essentials for Consistency in Production
`A great deal was learned from the
`different processes in effect, and as a
`result
`certain
`practices were recom-
`mended and others were required to
`obviate such variation.
`Those required
`to accomplish basic uniformity are sum-
`marized graphically in Figure 8.
`It
`may be seen here that a filtration is
`required within 72 hours of beginning
`inactivation.
`Another filtration
`step
`has been added as a further requirement
`against the contingency of the formation
`of precipitates, or of protected virus
`particles, or aggregates, before inactiva-
`tion
`completed.
`is
`This
`additional
`
`filtration is to be carried out at a point
`in the process when virus infectivity
`has been reduced to levels that are not
`readily detectable, which is sometime
`between the third and sixth days. The
`continuation of treatment for the pre-
`determined time should be as effective
`in destroying infectivity as is suggested
`by the "theoretical projection," and this
`is verified by tests upon samples as is
`indicated in Figures 7 and 8.
`The required tests in tissue culture
`are made during the first three days of
`treatment to establish the degree of in-
`activating effect of formaldehyde upon
`virus, and, again, during the last three
`days of treatment to verify the complete-
`ness with which the inactivating effect
`has been accomplished.
`Thus, there is
`built into the manufacturing and testing
`process a measure of safety that far
`exceeds in significance merely the iso-
`lated results of a set of tests upon a
`particular lot that is being considered
`for use in man.
`This is accomplished
`by relating treatment time and the time
`of selection of safety-test samples to the
`rate of virus inactivation and by con-
`sidering
`the
`consistency with which
`
`CEV Exhibit 1022_006
`
`

`

`POLIOMYELITIS VACCINE VOL. 46
`
`7
`
`REFILRATION
`
`el
`Neg.
`
`eg
`Neg.
`W5ZMNOLENT I
`
`',
`
`I1
`
`500
`TESTS ON TWO SAMPLES
`3 DAYS APART.
`st AT LEAST 3X
`BASE LINE INTERCEPTION
`
`TRIVALENT POOL
`TISSUE CULTURE TEST
`ON
`1500 ml
`/I
`NEGATIVE
`IMULTIPLE FILLING LOTS
`5
`5
`5
`5
`MONKEYS PER FILLING LOT
`(Minimum of 25 Monkeys for
`Each TRIVALENT POOL)
`
`FLTRAT1ON
`
`IxlO 6 t
`
`Ix 10 3
`
`IX lO 0
`
`IX 10-3
`
`lO-6
`
`0 C U<
`
`)
`
`-3-2 -1 0
`
`3
`
`6
`9
`12
`Time in days at 37 e C
`1:4000 Formolin and pH 7.0
`Figure 8 -Guarding Safety by a
`Combination of Requirements for
`Processing and Testing
`
`15
`
`negative tests are obtained in successive
`lots.
`In addition, further tests are required
`on the trivalent mixture.
`The report
`that a mixture made of three negative
`monovalent pools sometimes has re-
`vealed live virus has led some to specu-
`late about the possibility of reactivation
`of virus believed to be inactivated. This
`could occur readily where virus is con-
`tained in sediment in the monovalent
`pool, only to be revealed in samples
`removed shortly after stirring of sedi-
`ment upon admixture of the three strain
`pools.
`After a negative tissue culture test
`of the trivalent pool, still another safety
`test is performed-this one in monkeys.
`This is done on samplings of the differ-
`ent filling lots.
`It has been recognized
`not only that the monkey test is less
`sensitive than the tissue culture test for
`detection of small amounts of virus, but
`that difficulties can occur as a result
`of so-called "false-positives" (Figure 5).
`This problem has now been reduced by
`changes in the manner of performance
`
`of the monkey test and in the criteria
`for
`establishing
`the
`significance
`of
`histologic
`changes,
`especially
`when
`these are not accompanied by signs of
`paralysis. The greater sensitivity of the
`tissue culture test does not obviate the
`need for a test in monkeys at this stage
`of experience.
`Since a number of months are nor-
`required
`for
`completing
`the
`mally
`testing of a lot of vaccine, after com-
`pletion
`of
`the
`inactivation
`process,
`information is available on consistency,
`not only for the period preceding the
`start of manufacture of any one lot, but
`for a period of several months that fol-
`low, and before the lot is ready for
`release.
`Thus, the value of the consecu-
`tive record, as a safety guide, is en-
`hanced by the availability of information
`on inactivation after preparation of the
`lot of vaccine that is considered for
`release, as well as before its manufac-
`ture was begun.
`Although the signifi-
`cance and importance of this was not
`as thoroughly appreciated before, as it
`is now, this concept was the basis for
`
`CEV Exhibit 1022_007
`
`

`

`8
`
`JANUARY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH
`
`the statement about consistency that has
`been retained through several amend-
`ments to the Public Health Service's
`Minimum Requirements for Prepara-
`tion of Poliomyelitis Vaccine.'4
`The purpose of the information and
`analyses
`just
`presented
`help
`is
`to
`answer the question "How is vaccine
`safety determined?"
`Each of the sev-
`eral elements that are part of the ques-
`tion of safety must be applied with
`equal vigilance, regardless of the patho-
`genic characteristics of the strains used
`for inclusion in the vaccine.
`
`Mahoney Strain
`The use of the Mahoney strain in the
`initial investigations in animals, and in
`the initial studies in man, was deter-
`mined by a number of
`deliberate
`considerations: (1) At first, of all of the
`Type I strains examined in the initial
`attempts to find viruses that would grow
`well in tissue culture, this strain sur-
`passed
`all
`others.'5
`With
`further
`developments in tissue culture technics
`differences
`between
`strains,
`this
`in
`have been
`considerably
`respect,
`re-
`duced.
`(2) Since the monkey was more
`sensitive for the detection of Mahoney
`virus than the earliest
`tissue culture
`systems, while for detecting MEF-1 and
`Saukett the tissue culture system was
`more sensitive than the monkey,3 it
`seemed wise to work with viruses of
`both types to permit the development
`of methods that would destroy virus
`more readily detectable in either system.
`Such methods could then be applied to
`viruses of any variety in accordance
`with the needs of the problem.
`Further
`developments have now made the tissue
`culture test more sensitive than the
`monkey test for all three strains.
`(3)
`In the absence of direct knowledge of
`the relationship between virulence and
`antigenicity it seemed desirable to in-
`vestigate the effectiveness of a vaccine
`made from a particularly virulent strain.
`
`The decision to orient the investigation
`this way was influenced by the
`in
`knowledge that a vaccine made from
`the virulent Phase I pertussis organisms
`is more effective than vaccine made
`from less virulent varieties.
`Moreover,
`there are other similar examples in the
`field of immunology16 and this consid-
`eration
`worthy
`seemed
`further
`of
`investigation.
`Also, there was the basic scientific
`question of proof that a killed-virus vac-
`cine can actually induce immunity. For
`proving this point the Mahoney strain
`was best suited, since this would then
`leave no doubt that monkeys vaccinated
`successfully, without the development of
`disease, would have, in fact, developed
`their immunity from the killed virus.
`The successful vaccination of human
`subjects,
`without
`the
`induction
`of
`disease
`either in the subject himself or
`in his familial associates or community
`contacts-provided the
`desired proof
`that antibody could, in fact, be induced
`by a nonliving form of the poliomyelitis
`virus and gave further support to the
`hypothesis that living virus was not
`needed for immunization against this
`disease.17
`Furthermore,
`it
`appeared
`that with due care, vaccine with this
`strain-a particularly
`virulent one-
`could be prepared safely.
`If this be so,
`then surely, the fundamental scientific
`question of safety and effectiveness of
`a killed-virus vaccine could be answered
`decisively.5
`This, I believe, has been
`done.6
`If the Mahoney strain had not been
`used, the significance of the scientific
`proof might have been questioned.
`In
`fact, even today, there are some who
`are of the opinion that the only reason
`antibody is formed after vaccination
`with a killed-virus vaccine is because
`of the presence of traces of live virus
`or of virus that becomes "reactivated"
`after injection.
`For these hypotheses
`no proof has been presented, and the
`weight of evidence is to the contrary.
`
`CEV Exhibit 1022_008
`
`

`

`The Question of Strain Change
`The wisdom of continuing the use
`of Mahoney strain for preparation of
`vaccine on a large scale has been ques-
`tioned 18 understandably. The existence
`of defective vaccine was first revealed
`because of the effect of this component
`in those who received such vaccine. The
`opinion that vaccine should not be made
`with the Mahoney strain still left unre-
`solved the reasons for the defections in
`the
`preparation and testing
`vaccine
`process that permitted this to occur. The
`solution of the basic problem would
`solve simultaneously the problem of the
`Mahoney strain, since this same solution
`would be applicable to any virus, re-
`gardless of its potential pathogenicity
`for man.
`If the basic problem has, in
`fact, been solved, then the question of
`change of strain is being considered for
`reasons other than that it is essential
`for the preparation of a safe vaccine;
`the hazards of any living polio virus
`are to be avoided.
`A change was contemplated over a
`year ago as a result of observations
`which revealed the lesser stability of
`Mahoney, in the presence of merthio-
`late-hence the poorer showing for
`Type I as compared to Types II and III
`in the field trial.4 When this observa-
`tion was made11 a study was instituted
`to select another Type I strain.
`The problem of the instability of the
`Mahoney strain-and
`of
`the
`other
`strains
`as well-in the presence
`of
`merthiolate was controlled promptly in
`a number of ways.
`These have all been
`put into practice; preservatives may be
`omitted entirely;
`merthiolate used in
`combination with versene,4 or any one
`of a number of other preservatives can
`be used.
`The practice in the United
`States is to use a preservative, while in
`other countries it is omitted.
`Even though the stability problem
`seemed to have been resolved, the study
`of strains was continued because of in-
`terest in the basic question of the inter-
`
`POLIOMYELITIS VACCINE VOL. 46
`
`9
`
`relationship, if any, between virulence
`and antigenic capacity.
`It is now evi-
`dent that degree of virulence and degree
`of antigenic capacity are independent
`variables;
`each
`be
`cultivated
`can
`separately or together.
`The problem
`has now been sufficiently far advanced
`to indicate that a group of strains exist,
`within each type, that are in the lower
`range of virulence for animals, and that
`are equal to or better,
`antigenically,
`when compared with strains of higher
`virulence.
`It has been established, also, in the
`of
`these
`by
`studies 19
`and
`course
`others 20, 21 that virus populations that
`are dominantly avirulent can give rise
`to offspring that are virulent.
`For this
`reason, the assurance of the absence of
`infectious virus is of first importance
`in preparing vaccine.
`The opinion has been expressed that
`vaccinations should cease pending the
`production of vaccine containing "aviru-
`lent" strains for all three types or that
`there be a halt until the Mahoney strain
`replaced.22
`is
`This
`opinion can be
`questioned in terms of the relative im-
`portance (1) of safeguards based upon
`rigid requirements for production and
`testing
`as compared with safeguards
`based upon strain composition and (2)
`in terms of the advantages that could
`result from the availability of a supply
`of properly produced and tested vaccine
`as compared with a lesser supply that
`could be expected as a result of an
`abrupt change-over.
`These two alterna-
`tives must be considered in proper per-
`spective to guide the judgment of those
`who have the responsibility for the de-
`cisions involved in vaccine manufacture,
`vaccine application, as well as those for
`safeguarding the public health.
`
`Factors Related to Effectiveness
`Evidence continues to accumulate in-
`dicating that the prevention of paralysis
`is mediated through the action of anti-
`
`CEV Exhibit 1022_009
`
`

`

`10
`
`JANUARY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH
`
`body. On the basis of this evidence the
`presence of antibody in the circulating
`blood could be regarded as synonymous
`with immunity.
`However, it seems that
`if antibody was once present, but has
`declined to nonmeasurable levels, that
`a state of immunity continues.
`The
`mechanism that appears to be operative
`a hyperreactive antibody
`is
`that
`of
`response.23
`Although it is known that
`the presence of a measurable level of
`antibody will prevent paralytic polio
`in the lower primates,17'24 and probably
`in man,25 it is not known what degree
`of hyperreactivity, in the absence of
`measurable circulating antibody, is suf-
`ficient for prevention of paralysis.
`It
`appears from studies
`reported
`else-
`where 23 that the degree of hyperreactiv-
`the immunologic mechanism
`ity
`of
`required is very little; more precise in-
`formation should be forthcoming from
`studies now in progress.
`Thus, the primary objective of an im-
`munization program is that of inducing
`the formation of a measurable level of
`circulating antibody in all vaccinated
`subjects-and in so doing to have stim-
`ulated the immune mechanism in such
`a way that the resulting hyperreactive
`state will persist throughout life, even
`though in the course of time antibody
`levels decline; this occurs whether anti-
`body is the result of vaccination or
`natural infection.4
`The basic goal is
`the initial production of antibody, since
`this would reflect the effectiveness of
`the primary stimulation
`that accom-
`panies the initiation of the hyperreactive
`state. The use of secondary, or booster
`stimulation, is to assure further the full
`development of the hyperreactive state
`and to serve as a third opportunity to
`induce primary stimulation if this may
`have failed on both previous occasions
`for any one of a number of reasons-
`relative insensibility of the vaccinated
`subject, or because of the use of vaccine
`of low potency.4
`The goal is to achieve the maximal
`
`effect with the minimal amount of treat-
`ment. The critical factors are (1) anti-
`genic mass,10 i.e., potency of vaccine
`and volume inoculated, and (2)
`the
`number of inoculations and the spacing
`between each.4 10 Vaccine potency and
`stability are the most critical factors in
`this regard.
`While the degree of antigenic potency
`of vaccine can be established technically
`in any of a number of ways the signifi-
`cance of any laboratory test can be
`established only by comparison between
`the vaccine under test and a reference
`vaccine,4 the performance of which has
`been standardized in fully susceptible
`children.10
`Such a reference vaccine
`has been established and standardized
`and is being used.
`Correlation between level of potency
`of lots of vaccine used in areas where
`epidemics occurred in 1955, and the de-
`gree of effectiveness observed after a
`single dose, will contribute considerably
`to the establishment of a rational limit
`of potency to be required for maximal
`effectiveness.
`From such correlations
`it should be possible,
`also,
`to gain
`further insight into the requirements
`for persistence of vaccine effectiveness
`in terms of level of antibody associated
`with immunity to paralysis.
`In view of certain
`details
`the
`in
`Francis Report 5 questions have been
`raised about the immune response in
`relation to age.
`The answers to these
`questions may be seen in Figures 9 and
`10 which show, respectively, antibody
`response to infection and to vaccina-
`tion in individuals of different ages.
`It
`would appear from these data that, over
`the age range studied, there are no
`significant differences in terms of anti-
`body response to the two different kinds
`of antigenic stimulation.
`No further developments have oc-
`curred to warrant changes in the recom-
`mendation for two doses, spaced not less
`than two weeks apart and, preferably,
`four to six weeks apart. A third dose
`
`CEV Exhibit 1022_010
`
`

`

`POLIOMYELITIS VACCINE VOL. 46
`
`11
`
`Figure 9-Antibody Response in Relation to Age
`Type I Antibody Levels in a Group of 87 Recently
`Paralyzed Convalescents (1952-Pittsburgh)
`Age Group
`7-8-9 yr.
`4-5-6 yr.
`
`Antibody
`Titer
`
`1-2-3 yr.
`
`10tyr.
`
`2048
`
`1024
`
`512
`
`256
`1238
`
`32
`
`8
`
`0@@*
`0@@
`
`**.*.....
`
`so
`
`0
`
`*
`
`*eeoeo
`
`00
`
`*@000e0
`**eee,.e.e.
`
`0@@@
`
`s*e
`
`Deese
`
`0eo
`
`e
`
`Nou .
`Subjects
`
`21
`21
`
`27
`
`17
`
`22
`
`is required for producing the full effect;
`this can be approximated by administra-
`tion of the third dose at least seven
`months after the second.
`Whether or
`not reinoculation will be required pre-
`school, for maintenance of immunity,
`
`for children first vaccinated early in
`life, or additional inoculations to com-
`pensate for low potency-if found to
`be less than optimal-will be deter-
`mined by continued study; appropriate
`recommendations
`will
`be
`suggested.
`
`Figure 10-Antibody Response in Different Age Groups with No
`Demonstrable Preantibody for Any Type
`Two Doses of Vaccine Given 4 Weeks Apart (1 ML.-I.M.)
`TYPE I
`TYPE H
`TYPE m
`2-5 YRS
`6-e YRS.
`6-18Ye S
`2-5 YRS.
`6-1B YRS.
`2-5 YRS.
`
`0
`
`_ine
`
`Sin
`
`0
`
`000000
`
`o0000
`
`00000
`
`1024-
`
`2
`
`"@
`
`0
`
`*O@O
`
`000000
`
`§ 64
`
`0inO
`
`0000000
`
`mn"e
`
`0000000000
`
`816
`
`@00.0.
`
`0000000
`
`0_.
`
`000000
`
`0 0
`
`in
`
`00000
`
`0
`
`0
`
`19 LSBJWT5
`
`24 5U9ECTS
`
`CEV Exhibit 1022_011
`
`

`

`12
`
`JANUARY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH
`
`PRIMARY
`(1.0 ml. at 0,1,5 weeks)
`POST.
`PRE.
`riAg/5A)
`(5/12/54)
`
`POST- BOSTER
`(6-1-55)
`1.0 ml.
`0.1 ml.
`K.M.
`I.M. OR ID.
`
`00
`0000000
`000000000000
`000000000
`
`0 0
`
`000
`
`0 0
`
`0
`
`.0
`_
`see"
`*
`
`xx
`
`8t92t
`4096
`
`512
`
`2@
`P 32l
`216-
`8-
`24-
`
`s
`
`of 3 F
`ep
`types in 22 sbcts.
`Figure 11-Influence of Dose on "Booster" Response in 22 Children
`with No Demonstrable Preantibody for Any Type
`
`NQAla 304
`
`VOccine Lot NAl C-I
`
`Until such time, the schedule recom-
`mended can be regarded as optimal and
`the dose for all ages is 1 ml, given
`either
`intramuscularly
`subcutane-
`or
`ously.
`Although
`the
`intramuscular
`response is perceptibly better than the
`subcutaneous,
`is not sufficiently
`this
`great to be so significant as to warrant
`the use of one route to the exclusion of
`the other.
`The only limitation in the usefulness
`of the intradermal route is the amount
`of antigen that can be so administered.
`Any demonstrable immunologic advan-
`tage of the intradermal route, in terms
`of efficiency of antibody formation, is so
`slight as to be offset by the limitation
`in volume that can be given in this way
`as compared with the subcutaneous or
`intramuscular routes.
`This does not
`mean to say that administration
`of
`smaller volumes intradermally will not
`be effective-it merely says that there
`quantitatively
`demonstrable
`im-
`are
`munologic advantages to the use of
`larger doses which can be given more
`readily either intramuscularly or sub-
`cutaneously.
`The relative
`degree
`of
`response elicited by a small "booster"
`dose given intradermally or intramuscu-
`
`larly as compared with a
`intramuscularly
`given
`is
`Figure 11.
`
`larger dose
`shown
`in
`
`Future Developments
`Even though we have continued to
`