SPECIAL
ARTICLE
PRIMARY
IMMUNODEFICIENCIES
Report
of
a
World
Health
Organization
Committee
H. Fudenberg, Section of hematology nun Imnunology, School of Medicine, Uiiitersitj of California, San Francisco, California, USA
R. A. Good, Department of Pediatrics and Z’sficrobiologt,, Unicersity of Minnesota, Minneapoln,
Miii-ne.sota, USA
H. C. Goodman, Chief Medical Officer, Immunology, World Health Organization, Geneca, Switzer-land
w.
Hitzig, Kinderspital, Zurich, SwitzerlandH. G. Kunkel, Department of Immunochemistry, The Rockefeller Unicer.sity, New York, New York, USA (Chairman)
I. M. Roitt, Department of Immunology, Middlesex Hospital Medical School, London, England F. S. Rosen, Laboratory of Immunology, Department of Pediatrics, Harvard Medical School, Boston,
SIa.ssaclzusetts, USA
D. S. Rowe, Director, WHO International Reference Centre for Immunoglobulins, Lausamie, Switzer-land
M. Seligmann, Irnmunochernical Laboratory, Institute for Research on Blood Disorders, Hopital St. Louis, Paris, France
J. R. Soothill, Department of Immunology, Institute of Child Health, University of London, London, England
EDIToR’s Nori: This paper was prepared
by the participants (listed above) in an
in-formal discussion arranged by tile World
Health Organization, and held June 8-13, in Geneva, Switzerland. A French version
will be published in tile W7orld Health Or-ganization Bulletin.
Because of its unusual importance the
report is here printed as received, without
change of style in citation or listing of ref-erences to that customary in this journal.
I. INTRODUCTION
I
NCREASED understanding of the natureand variety of immunodeficiency states in man is rapidly accumulating both from
studies of human patients and from
experi-mental work on the immune response in an-imals. Progress is evident in the
develop-flWflt of diagnostic tests for deficiencies in
1)0th Immoral and cellular mechanisms of
imniunity, and in the introduction of new
forms of therapy; for example, the grafting
of lymphoid cells. Studies of
immunodefici-ency provide the most direct evidence
con-cerning the nature of the immune response
in man. and hence are of wide general in-terest. In this paper, current knowledge and
concepts are summarized, a logical
classifi-cation is presented, and recommendations
are made for the investigation and
treat-nient of these disorders.
II. THE CELLULAR BASIS OF
IMMUNE RESPONSES
Immunological responses are classically divided into those mediated by humoral
an-tibody and those by cells ( Table I ). Both
depend upon the activity of small
lympho-cytes ( Gowans and McGregor, 1965 ) which
are themselves ultimately derived from stem cell precursors which in postnatal life reside in the bone marrow ( Miller and Mitchell,
1969) . Neonatal thymectomy prevents the
development of cell-niediated immunolog-ical reactions, such as homograft rejection and delayed-type hypersensitivity and also the humoral antibody response to certain
antigens ( Miller and Osoba, 1967 ). In
iurds, antibody production-but not
cell-mediated reactions-is dependent upon tile
bursa of Fabricius, and it has been
postu-lated that gut-associated lymphoid tissue
provides the mammalian analogue
(Cooper, et a!., 1966 ); however, no single
or muitifocal organ has yet been identified
that can unequivocally be shown to exert a
51)ecific bursa-like function.
This suggests that stem cells originating
in the l)Ofle marrow differentiate to form at
928 PRIs’IARY IMMUNODEFICIENCIES
( Gabrielsen, et a!., 1969; Meuwissen,
Stut-man & Good, 1969 ): one (T-lymphocytes)
dependent on the presence of the thymus
( Roitt, et a!., 1969 ), and the other (
B-lvm-phocytes ) independent of the thymus.
The T-lyrnphocytes constitute the greater
part of the recirculating pool of small lym-phocvtes, whereas the B-cells appear to be more restricted to lymphoid tissue. Both pO1)ulatioflS contain antigen-sensitive cells,
probably with specific antii)ody ( or
anti-I)ody-hke molecules ) on their surface.
The B-lymphocytes can differentiate,
proliferate, and mature into plasma cells
which synthesize humoral antibody. Any
defect vhich limits the llunll)er and differ-entiatioli of the B-cell system vill lead to a
deficiency in immunoglobulin synthesis.
The T-cells can he nonspecificaily
stimu-lated in culture by mitogens, such as
phyto-hemagglutinin . T-iymphocytes from a
sen-sitized individual can also be stimulated by
contact with specific antigen ( which may
have to 1)e macrophage processed ) in vivo
and in vitro. These stimulated T-cells which do not have intracellular imnitinoglobulin
serve many functions ( Table I)
1. They divide further to form an
cx-panded population of ri1nedl
antigen-seii-sitive cells which make a niajor
contnihu-tion to irnhlluIlOlOgical memory because of
their long life-span.
2. They may be “killer” cells which are
cytotoxic for graft target cells.
3. They release a number of soluble
fac-tors which are chemotactic for mononuclear
cells, inhibit the migration of macrophages
and probably “activate” them, are
mito-genic for other lymphocytes, and increase
vascular permeability. These
charactenis-tics, together with No. 2 above, form the basis for the phenomena of specific cell-mediated immunity.
4. They may cooperate during the im-mune response to certain (
“thvmus-depend-ent” ) antigens by stimulating the B-lyrn-phocytes to produce antibody.
Failure to generate an effective T-cell
system will lead to defects in cell-mediated immunity and also in the humoral antibody
response, in those circumstances where
cooperation with T-lymphocytes is
irnpor-tant. In such a case, the B-cell system may l)e iiitact yet the individual max’ shov poor
antibody responses.
III. GENETICS OF IMMUNOGLOBULIN
SYNTHESIS
In some I)rotein systems, the al)sence of a
specific protein in certain individuals has l)een attributed to mutations that appear to
involve structural genes, although mutation
at the regulatory or initiator level might
have similar effects. Usually both parents
are heterozygous and have approximately one-half of the normal level of proteiii.
Afi-hrinogeneniia is a typical example. A
simi-Jar sittiation would be expected to appear in the inlmunoglobuhin system. Here, how-ever, the problem is much more compli-cated. vIu1tiple genetic loci are involved for tile independent chains making up the
basic four-chain immunoglobulin unit. As
regards the constant regions of the
poly-l)ePtide chains, at least 10 loci appear to op-crate for the different classes and subclasses
of heavy cilains (Natvig and Kunkel, 1968);
the kappa and lambda light chains involve at least two more loci. An undetermined
number of loci seem to be responsil)le for
the synthesis of the variable, or \, regions.
Independent \7-region subgroups have been
defined for kappa, lambda, and heavy chains (WHO, 1969; Hood and Talmage,
1970 ). The same ‘II region subgroup can
appear in combination with the C-region of
any immunoglobulin heavy chain (\Vang,
et a!., 1970) . However, the subgroups of \7,
chains are not interchangeable I)etween
kappa and lambda light chains. A relation-ship between the V-region subgroups and alterations of immunodeficiency states
re-mains to be shown.
Studies iii patiellts with
hypoganimaglo-bulinemia have not as yet shown isolated
deficiencies that can be ascribed to any of the structural genes mentioned above. IgA
may be undetectable without recognized abnormalities in the other
sib-lings have not revealed consistent
altera-tions and 110 clear genetic pattern has
emerged (Clanian et a!., 1970 ). In nlost
other instances, multiple classes of
immuno-globulins are involved and it has not been
possible to implicate a single gene defect.
Studies with Gm markers have indicated
tile absence of certain markers in a few
par-ents of affected children, and the possibility
of gene deletions has been raised ( Yount, et
a!., 1970 ). However, affected individuals
have also shown abnormalities in other
im-munoglobulms which are difficult to relate
to the presumed gene deletions.
It has been postulated for a number of
years that regulatory gene defects may
pro-vide an explanation for partial deficiencies of specific immunoglobulins in certain
pa-tients ( Parker and Beam, 1963; Fudenberg
and Franklin, 1963; Fudenberg and
Hirsch-horn, 1965 ). Support for this hypothesis was
recently obtained from measurements of
genetically determined antigens where
sin-gle gene products could be determined.
Marked divergence in the amounts of
pro-teins coded for by allelic genes has been
ob-served in some individuals ( Rivat, Burtin
and Ropartz, 1969; Yount, et a!., 1970). Very
little is known about regulatory genes in
mamnialian systems; the difficulties are
in-creased ill the imniunoglobulin system
where unique closely linked structural genes
are involved. In addition, the phenomenon
of allelic exclusion probably results in each
immunoglobulin being produced by an
in-dependent cell line, making it even more
difficult to interpret regulatory and
struc-tural gene relationships.
One form of hypogammaglobulinemia is X-linked, indicating that the X chromosome
)lays some role in the regulation of the
B-cell system.
IV. TEST FOR ASSESSING IMMUNE STATUS
ANTIBODIES AND IMMUNOGLOBULINS
MEASUREMENTS OF SERUM
IMMUNOGLOBULIN CONCENTRATIONS
A variety of methods is at present
avail-able for measuring serum concentrations of
immunoglobulins. \\ith tile exception of
electroplioresis, all de1x’nd upon the uSC of
s1)ecific aiitisera. They include smgk radial
diffusion and double diffusion in agar gel,
inimunoelectrodiffusion, and radioi iii
run-noassay. By such techniques, the
irnrnullo-globulin conceiitration of the test seruiiii is
compared with a standard solution of
le-fiuied concciitratioii. No test has ieeii
P(r-fected to the 1)Oiflt where the
interpreta-tion of tile results is entirely straightforward.
The single radial diffusion assay of
\Ian-cmi is widely used and various
modifica-tiosis of the method have lxen (lescribedi
( sIancini, Carbonara and Herenlans, 1965;
Fahey aild ‘dcKeivey, 1965 ). Each test
should include at least three standard
solu-tions On the saifle plate. The error of the
method is represented by a constant
cOeffi-cient of variation which under oI)tinlal
con-ditions may he less than 10i, except at
extremes of concentration ; however, such
1)recisioil may he difficult to achieve in
rou-tine lal)oratories. The limit of detectability
of protein, using low concentrations of
aiiti-serum, is about 10 i.g/ml. The volume of
sc-rum required to perform the test is about
0.01 iii! for each deteriiiination. Results can
l)e O1)tained after 24 hours of diffusion in
the case of normal sera, I)ut further time
nay l)e required for the assessment of very
high or very low levels.
Immunoelectrodif-fusion is a neWer technique which may
have greater pr’cision and greater
sensitiv-ity than single radial diffusion (Freeman
and Smith, 1970 ) but is at pr(’sellt confined to speci(uiized laboratories.
Gel diffusion methods are sensitive to
dif-ferences ill diffusion constants ; sl)(’cial
P1k’-cautions should he taken to ensure that
im-niunogiobulins in the standards and scra
under test are not split or aggregated. Also,
reliable measurements cannot 1)0 niadle of
such Iroteins as low molecular weight 1gM
and secretory IgA, unless a standard
immu-noglobulin Irelaratioil is available in the
same form. Greater sensitivity in the
esti-mation of immunoglobulins CdIl 1)0
achieved by the use of radioactive isotopes.
PRIMARY IMMUNODEFICIENCIES
ZOIlO of preciritate is treated with
radioac-tive anti-immunoglobulin, and is visualized
1)y autoradiography ( Rowe, 1969) . By
us-ing this method, protein can be detected at
concentrations of the order of 100 i.g/ml; serum IgE and immunoglobulins in
se-cretions have been measured in this way.
Radioimmunoassay methods possess even
greater sensitivity (\Vide and Porath, 1966).
They depend upon the capacity of the
im-munoglobulin in the test serum to inhibit
the interaction of antiserum with purified
labelled immunoglobulin.
Standards and Antisera
Discrepancies in results have arisen from
the use of different standards by different lai)oratories. Ideally, a standard should con-sist of a stable preparation of immunoglob-ulin of known concentration, in a form identical with that in the fluid under test. The World Health Organization now makes available reference preparations for the five classes of human serum immunoglobulins
( Rowe, Anderson and Grab, 1970; Rowe,
Anderson and Tackett, 1970; Rowe, et a!.,
1970 ) , and it is recommended that these be
used as standards for immunoglobulin
de-terminations . #{176}Working standards for
gen-eral use can be prepared and related to
these standards.
Antisera to human immunoglobulins have
been prepared in a variety of species using
preferably normal or, failing that,
mono-clonal proteins for imnmnization. Although it appears desirable in principle to use pooled monoclonal proteins rather than a
single protein for immunization, the use of
single monoclonal proteins in species, such
as tile goat, sheep, or rabbit, does not
ap-pear to introduce important discrepancies.
oill(’se preparations can be obtained from tile Who International Reference Centre for
Inirnuno-glol)uiins, Institut de Biochimie, rue du Bugnon 21,
Lausanne, Switzerland; the WHO Regional
Refer-ence Centre for Immunoglobuhns, National Cancer
Institute, Bethesda, Maryland 20014, USA; and tile
De1mrtment of Biological Standards, National
In-stitute of Medical Research, London, N. \V. 7,
Engiand.
All antisera, including those from
commer-cial sources, must be shown to be specific in
the test for which they are being used. In a
recent study ( Rowe, Anderson and Grab,
1970 ) , a number of laboratories obtained
reasonable agreement in the measurement of imnmnogiobulin in seruni samples when
the same WHO reference preparation was
used. The use of different standards is likely
to yield much wider variations in results. Antisera and standards obtained from corn-rnercial sources have on occasion been found to be unsatisfactory.
Subclasses
Levels of each of the four subclasses of IgG inlrnunoglobulins can be determined
by the Mancini procedure. Precipitating
antisera for IgG1 are the most difficult to
obtain but many antisera are suitable for
hemagglutination studies. Inhibition of
hemagglutination has been found to be a
useful technique for the gross measure-ment of the extreme changes found in some sera from patients with immunodeficiency
(Yount, et a!., 1970).
General Recommendations
For routine purposes, the following
pro-cedures are recommended:
1. A preliminary assessment by
electro-phoresis and/or immunoelectrophoresis.
2. Specific measurement of
immuno-globulin by one of the above methods, at a laboratory experienced in such techniques.
Even gross deficiencies of certain im-munoglobulins cannot be detected by
elec-trophoresis. Immunoelectrophoretic
tech-niques are not quantitative, and also require
experience in interpretation. Therefore,
ab-normaiities cannot be excluded by either of
these methods, which should be regarded
as preliminary screening procedures.
Antibodies and lmmunoglobulins in
Immunodeficiency
Measurement of immunogiobulin
con-centration by the techniques outlined above
is of value for diagnosis and analysis of the
concen-trations of immunoglobulins in adult serum
vary from individual to individual, and
there is at present no conclusive evidence
of breaks in distribution curves to justify
thresholds for the diagnosis of
immuno-glol)ulin deficiencies. However, 200 mg/
100 ml or less of IgG may constitute a
prac-tical threshold value if associated with
symptoms. IgA is undetectable in
approxi-niately 0. 1% of the normal population
( Bachman, 1965) . Age in childhood and
prematurity have been reported to have an
effect on immunoglobulin concentrations.
Values are available of immunoglobulin
concentrations in adult populations related to age, environment, and sex. The relevance
of these variations to disease is not clearly established, except in the case of gross
defi-ciencies.
In patients losing immunoglobulins
through exudative enteropathy, nephritis,
nephrosis, and so forth, normal levels of 1gM are usually found; a low albumin con-centration reflects loss of albumin in paral-lel with loss of IgG. X-linked recessive
agammaglobulinemia is characterized by
very low levels of all five classes. However,
when sufficiently sensitive tests are used,
even these patients can regularly be shown to possess small amounts of
immunoglobu-lins of each class. Patients with other forms
of immunodeficiency may have
concentra-tions of the immunoglobulins as low as
those found in the X-linked recessive form;
thus, concentrations of immunoglobulins
cannot be used as the sole criteria for
diag-nosis of X-linked agammaglobulinemia.
Concentrations of IgM, IgG, IgA, IgD, and
IgE may vary in patients with other forms
of immunodeficiency. Variations occur
be-tween members of the same family, and
from time to time in the same individual.
Immunoglobulin levels should be measured
in each patient suspected of primary
immu-nodeficiency not only for diagnostic
pur-poses but also for future analysis.
Knowl-edge of immunoglobulin concentrations
may be of help in the diagnosis of other
dis-eases; for example, 1gM may be low in tile
\Viskott-Aldrich syndrome (Cooper, et a!.,
1968) and IgA and IgE may l)e low in
atax-ia-telangiectasia ( Amann, et a!., 1969 ). In
general, syndrome classifications based on
inimunoglohulin patterns; e.g., so-called
dysgamnmgiobulinemia, have not been
found useful (Sehigmann, Fudenherg and
Good, 1968).
One notal)le exception so far has been
IgA deficiency. In this frequent condition,
IgA is usually deficient in secretions also.
Occasionally, patients have been
encoun-tered who have normal levels of circulating
IgA and low levels of IgA in secretions;
con-versely, patients having low levels of
cir-culating IgA may have normal levels of IgA
in secretions (Goldberg, Douglas and
Fu-denberg, 1969) . The two subclasses, IgA1
and IgA are usually reduced in equal
pro-portions. These patients should always be
studied for the presence of antibodies
against IgA, which are found in a significant
number of cases (Vyas, et a!., 1969).
Low levels of IgM associated with both
meningitis and disseminated
nonprogres-sive vaccinia have been reported and
re-quire further analysis ( Hobbs, Mimer and
Watt, 1967 ). Low levels of circulating IgE
associated with recurrent respiratory
dis-ease have also been described. Subjects
with IgE concentrations of less than 10 i.g/
iiil and otherwise normal immunoglobuiin
levels, who are free from infection have,
however, been noted in several centres
( Cain, et a!., 1969).
Imbalances of IgG subclasses have
fre-quently 1)een encountered in patients with
immunity deficiency but no such
imbal-ances have been found in patients with
X-linked agammaglobulinemia ( Yount, et a!.,
1969 ). Since antibodies to certain
carl)ohy-drate antigens in man are mainly of the
subclass IgG2, the possibility of a
relation-ship Ijetiveen subclass imbalances and
lective increase or deficiency of antibody
responses to certain antigens must he
con-sidered. Alternatively, inordinate exposure
to stimulation with certain kinds of antigen
should be examined as a i)asis for the
im-balances observed.
PRIMARY IMMUNODEFICIENCIES
homogeneous immuiioglobuiins, abnormal
kappa-lambda light chain ratios, and
abnor-ITliuhities of function; e.g. , complement
fixa-tion; are frequently encountered in patients
\\Ti tli immunodeficiency syndromes (Hong
audI Good, 1967; Pickering, Hong and Good, 1967; Seligmann, Meshaka and Da-llOfl, 1967 ). Homogeneous
iiiimunoglobu-lins are also encountered in very young
l)remat1re infants (Hitzig, unpublished
re-suits).
Failure to respond to antigenic
stimula-tioll can sometimes ie observed in patients
with nornial or high levels of all
immuno-globulins, and ill I)1tients with isolated
im-munoglobulin deficiencies. This may be
general or restricted to some antigens. Thitis, iiormal imniunogiobulin
concentra-tioii does not exclude antibody deficiency,
and responses to antigenic stimulation are
crucial ill studies of these patients.
Assessment of Antibody Formation Following Immunization
Humoral immunity function may be
stud-ied by tests for existing antibodies to
anti-gens to which the population is commonly
exposed, or by tests for antibody formation
following active immunization.
Antigens
Antigens used may be either proteins or polysaccharides. Live vaccines (BCG and
attenuated viruses ) should never be given.
Alum-precipitated antigens may he used.
Antigen Dose
Antigen dose should be increased with
continuing antigenic stiniulation, and
nlul-tiple doses should be given at 3 to 6 weekly
intervals. A complete schedule usually
corn-1)nis(’s three to four injections.
Antibody Determination
Antil)ody determination should be
car-ned out 011 sertlIfl samples taken
approxi-mately 3 weeks after the last injection. Only
gross increases in titre are significant.
I Iighl reproducible methods are not
es-sential, therefore, and the widely used haemagglutination inhibition methods are
acceptable. For precise comparison of
re-sponses of patients with immunodeficiency
and other diseases, techniques that deter-mine antigen-binding capacity are prefera-l)le.
Recommended Procedures
1. “Natural” antibodies: A and B
isohe-magglutinms, heteroagglutinins, and
het-erolysins; e.g., against sheep or ral)l)it red
cells; antistreptolysin, and hactericidins
against E. co!i, when present, are useful for
screening.
2. The usual active immunizations with
diphtheria, tetanus, and pertussis (DPT)
vaccines are very useful. Polio vaccine
should only be used in the inactivated form.
3. Additional active immunizations that
may he recommended are with
polysaccha-rides derived from pneumococci. H. in
flu-enzae and N. meningitides antigens are
po-tent, pure, and harmless antigens and may
prove to be of prophylactic value (
Gotsch-iich, Goldschneider and Artenstein, 1969).
Their use is recommended as soon as they
become more widely available.
4. Vi antigen and flagellin are also potent and harmless antigens. Their use should be considered but they are not known to l)e of prophylactic value.
5. Addititional active immunizations to
be used only in special studies include those with typhoid H, keyhole limpet
he-mocyanin, phage 4i X 174, and blood-group
antigens. All have been used to analyse
an-tibody responses of patients with immunity
deficiency syndromes. Possible side effects and insufficient experience suggest that
these antigens should only be used
cati-tiousiy under carefully controlled
circum-stances.
6. All live virus vaccines, including
vac-cinia, I)OliO, measles, rubella, as well as
BCG and tularemia, should he avoided.
Recommended Regimens
1. Commercial DPT vaccine, given in a
i-cell system
‘I’liyiiius-depeiident T-lymplioeytes
‘Iliumus-dependent T-lymphocytes
‘FABLE I
‘IJIF: (‘EILuLs1 BAsis OF IMMUNE RF:spoxsF:s
j)hytoheluaggiut in in
(eii-iuediated iiiiiioiiiitv
+antigeii
---
-3 Plasma cell lineBlast cells Killer cells (e.g., graft rejection)
+antigen
Tliyiuus-depeiident I’-lyinpliocytes SoIuI)le lymphocyte factors
Il-cell system:
‘I’liymus-independeut “hursa
equivalent” B-lymphocytes
‘l’here is evi(lenee that I-cells plus aiitigeii (811 (0-operate with 11-cells to prltiee antil)ody.
3 successive weeks. Blood is taken 14 days
after the last injection. Antibodies may be
determined by hemagglutination
tech-fli(lues.
2. Killed polio vaccines (1.0 ml) injected
intramuscularly at 2-week intervals for
three doses. Blood is taken 2 weeks after the last injection. Antibody is determined by virus neutralization.
3. Pneumococcal polysaccharide (0.1
mg ) intramuscularly for three successive
weekly injections. Blood is taken 2 weeks
after the last injection. Antibody is esti-mated by the precipitin technique or
pref-eral)ly by an antigen binding technique.
4. H. influenzae polysaccharide (0.05
mg ) is injected subcutaneously, and blood
is taken 2 weeks later. Hernagglutination
or antigen binding techniques may he used for antibody analysis.
5. N. meningitides polysaccharide (0.05
mg ) is injected subcutaneously, and blood
is taken 2 weeks later ( Gotschlich,
Gold-schneider and Artenstein, 1969 ). Antibody
is determined by precipitin analysis.
6. Vi antigen : Immunization with Vi
an-tigen from E. coli may be used. In this
as-say, 100 ig of antigen is given subcutane-ously in a volume of 1 ml in saline. Blood is
taken 2 weeks later. Antibody is determined
by passive hemagglutination assay using
human group 0 cells, as described by Ceppelhini and Landy, 1963. Flagelhin may
also be used and has been found to he
with-out danger; 5 sg ill 0.1 ml of
phosphate-buffered saline, pH 7.0, is injected
subcu-taneously into the forearm ( Rowley,
Mer-nh and Mackay, 1969).
Cell-Mediated Immunity (CMI)
Three tests are commonly employed for
assessing cell-mediated immunity: ( a )
de-layed-type skin reactions; ( b ) in vitro
stim-ulation of lymphocytes to divide and form blast cells; and ( c ) release from
lympho-cytes of macrophage migration inhibition
factor. The relative sensitivity of these
methods has not been adequately evai-uated. Specific lymphocyte-mediated cy-totoxic tests ( Mauel, et a!., 1970 ) may
be-come available in the future.
Pre-existing Immunity
This may he studied by all three
tech-niques. Five different antigens are generally
used: mumps, trichophytin, purified protein
derivative, Candida, and
streptokinase-streptodornase.
For skiii testing, it is recommended that
tlie following materials I)e injected
intra-dermally:
1. Tuberculin, 0.1 ml of a 1:10,000
dilu-tion. If negative, repeat at 1 : 1,000.
2. Candida, 0.1 ml of a 1:10 dilution for
infants or of a 1:1,000 dilution for older
children and adults.
:3. Tnichophytin, as for Candida.
Strep-934 PRIMARY IMMUNODEFICIENCIES
tokinase-streptodornase, at a concentration
of 5 units of streptokinase per 0. 1 ml. If negative, repeat at a concentration of 40
units per 0.1 ml.
5. Mumps skin testing antigen, 0.1 mg.
Reactions should 1)e read at 4 hours to
as-sess any Arthus reactions, and at 24 and 48 hours for delayed hypersensitivity. The di-ameter of induration and erythema should be recorded.
W7here delayed hypersensitivity to a fur-ther antigen is prevalent; e.g., coccidioidin
in California; this could usefully be tested for.
Pre-existing cell-mediated immunity to these antigens can also be investigated by
in vitro culture methods. \\Then all these tests are negative, the one-way mixed lyrn-phocyte reaction in vitro ( see below ) may
provide additional information.
Active Sensitization
Only sensitization to 2
:4-dinitrochloro-benzene is recommended (Dupuy and
Preud’homme, 1968 ) , and even then it
should be borne in mind that unpleasant
burn reactions may be produced,
partidu-larly in young children. The
2:4-dinitrofluo-robenzene should 1)e avoided. Keyhole
him-pet hemocyanin produces both humoral and cellular immunity, and could
theoreti-caliy sensitize against shellfish; it is not
rec-ommended for general use.
A 30% solution of 2:4-dinitrochloroben-zene (DNCB ) in acetone (10% for
infants ) is used, 0.05 1111 being applied to
the volar surface of the forearm on a filter paper 1 cm in diameter. A test dose ( see
below ) is applied at the same time to serve
as a presensitization control. The filter
pa-pers are removed after 12 to 24 hours and
the control is read 48 hours after applica-tion.
Patients are skin tested 14 to 21 days later with 0.05 ml of a 0.10 or 0.05 solu-tion of 2:4-diiiitrochlorobenzene (DNCB) in acetone on filter paper. The paper is re-moved 12 to 24 hours later and the reaction
assessed 48 hours after application. Subjects
who remain negative after this regimen
should be skin tested again 30 to 35 days after attempted sensitization.
Skin reactions are assessed as follows: 0-no reaction
+-erythema only
+ + -erythema and induration + + + -vesiculation
± + ± + -bullae and ulceration
Only + + or greater reactions are accepted
as evidence of sensitization.
Lymphocyte Transformation
Lymphocyte stiniulation by antigen
in-voives a relatively small number of
spe-cificahly precommitted cells which undergo
mitosis and blast cell transformation (
Op-penheim, 1968 ) and are thought to release
nonspecific mitogenic factors that further
recruit nonsensitized cells ( Lawrence and
Landy, 1969 ). Although this largely reflects
tile response of T-lymphocytes, and thereby
indirectly their ability to generate specific
cell-mediated immunity, there is a
signifi-cant, but probably minor, contribution from B-cells. However, the one-way mixed
lym-phocyte reaction, in which the lymphocytes
are stimulated by mitomycin-blocked allo-geneic cells; i.e., cells from the same species that are not genetically identical; does ap-pear to reflect an exclusive activity of a T-cell population ( Johnson, 1970).
In addition to stimulation by antigen, the
T-lymphocytes, or a major population
thereof, normally respond in a nonspecific
manner to mitogens, such as phytohemag-glutinin and concanavalin, which activate the cells via receptors distinct from those involved in antigen recognition. The
re-sponse to nonspecific mitogens provides
some measure of the overall number of the
relevant T-cells present hut need not
neces-sarily reflect the ability to respond to
spe-cific antigens.
Our Present limited experience suggests that phytohemagglutinin provides the
greatest discrimination when assessed on
the third day of culture, although where
very low responses are encountered,
cul-hires can profitably be studied up to 7 days
(LAS$IFI(ATION OF Puii.uvr IMMUNODFFI(lEscY I)JsounF:ns
Type
11-cells T-cells Stem (eli.?
+ +
+ +
+ +
(soinet inies)
+ +
+ +
+ +
+ +
+ + +
+ + +
+ + +
+ + +
+ +
The response to antigenic stimulation is normally maximal by the fifth day.
Lyniphocyte stimulation may be
evalu-ated in terms of morphological change, or
by the incorporation of 3H thymidine
meas-ured either by direct counting of radioac-tivity or by autoradiography on smear
prep-arations. It is usual to set up cultures in
which the lymphocytes are adjusted to give
a concentration of 10 cells/mi; in each
case, it is desirable to allow for differences in cell survival in culture by expressing the results in terms of a fixed number of viable cells. The degree of lymphopenia also clearly has to be taken into account in the final interpretation of the response. The morphological changes ( “transformation”)
do not always occur in parallel with the thymidine uptake; this has been shown to be the case in variable immunodeficiency
(Fudenberg, et a!., 1967 ). For clinical
diag-nosis, the niost convenient and useful test
would l)e tritium u1)take, i)ut it is desirable
to obtain data i)y all three methods.
How-ever, the volumes of blood available are of-ten limited, and it seems likely that the ati-toradiographic method could most readily be adapted to small numbers of cells.
Control values of unstimulated cultures
are often high in certain deficiency states, and it is essential to give data on both un-stimulated and stimulated cultures. Because of other variations in the technique and in
the specific activity of different thymidine preparations, it is essential to determine the
phytohemagglutinin response of
lympho-cytes simultaneously from at least two
donors who are known to be normal. Since human sera contain variable amounts of factors that may enhance or inhibit these
responses, it is important that a standard
human or fetal calf serum be used in all
cul-‘FABLE II
Infantile X-linked aganiniagiohulinein in +
Selective i!nn)unOglObulifl deficiency (IgA) +
‘l’ransient hypogammaglol)ulinenua of infancy
X-linked immunodehciency with hyper-IgM
Thyniic hypopiasia (pharyngeal pouch syndrome, l)i( eorge’s syndronie)
Episodic lymphopenia with -mphocytoxin
lininunodeficiency with normal or without liyperiinmunoglohulineniia
(Faulk, Tomsouk and Fudenherg, 1970)
llllmunodehciency with ataxia-teiangieetasia
Iniinunodeficiency with thronhoeytopenia :111(1 eczeiiia
(Wiskott-Aldnich syndrome) Iininunodeficiency with thymonia
Immunodeficiency with short-limlwd (Iwarfisni
(Gatti, et a!., 1969; Lux, et a!., 1970)
Ininiunodeficiency vith generalized hPInatOl)oiet ic hy)oplasia
Severe (olnl)ined illlluUnodeficiellcy
(a) autosornal recessive (h) X-Iinked
(() sporadic
Variable iininunodeficiency (conunon, largely unclassihe(l)
SUggeste(l (‘elinlar Defect
(sonic)
(soiiiitiiiies)
936 PRIMARY IMMUNODEFICIENCIES
tt.res. Inhibitory substances may be present
in serum as, for example, in certain
immu-nodeficiency states, such as ataxia telan-giectasia; these may l)e demonstrated by
setting up additional cultures with
autolo-gous serum.
Migration Inhibition Factor
The interaction of sensitized T-lympho-cytes with antigen releases a numi)er of
fac-tors, olle of which inhibits the migration of
guinea pig mlcro1)hages from capillary
tui)(’s. Release of this migration inhibition
factor appears to be a good in vitro indica-tor of cell-mediated immunity ( Lawrence and Landy, 1969 ). The technique that is
preferred at present involved culturing
l)iood lymphocytes with antigen for periods
of tip to 3 days, and adding the
concentra-ted cell-free supernatant to chambers
con-taming guinea pig macrophages ( Thor, et
a!., 1968; Rockhin, Meyers and David,
1970).
Human leukocvte niigration inhibition
(Soborg, 1967 ) , although technically more convenient, has not i)een generally
ac-ce1)ted as a concomitant of cell-mediated
immunity. Lymphocyte migration is now
being studied and appears to reflect
cell-nlediated iilinhilnity \Vllen particulate, but
not soluble, antigens are em1)loyed (
Zabri-skie, et (ii., 1970; Zabriskie and Falk, 1970).
Ill. MORPHOLOGY AND HEMATOLOGY
Examination of the histopathoiogical
changes ill lymphoid tissue is a useful
ad-junct to the diagnosis of immunodeficiency. Bone marrow, rectal mucosa, and lymph nodes are tile tissues most accessible for study.
Lymph Nodes
Biopsies of iiodes should be performed 5 to 7 days after local antigenic stimulation.
In children, it is convenient to inject anti-gens, such as di1)htlleria and tetanus toxoids, into the medial aspect of the thigh and sub-se(uentiy to remove an ipsilateral inguinal lymph node for histological examination
(Gitlin, et a!., 1959; Good, 1955).
When the diagnosis of combined
immu-nodeficiency is readily apparent from other
criteria, lymph node biopsy is unnecessary; it may even be detrimental, as lmphoid
tis-sue is difficult to find in affected infants and
surgical incisions provide a portal of entry
for serious infection.
Lymph nodes should be examined to
as-sess the thymic-dependent paracortical
areas, the numbers of plasma cells in the
medullary portioi, and the cortical
germi-nal centres. Plasma cells and germinal
cell-tres are absent, for example, in the most
complete isolated B-cell deficiencies, and deficiencies of lymphocytes in the deep
cor-tical regions indicate defects of the T-ci’ll
system.
Bone Marrow
Enumeration of cell ty)es in the bone
marrow of all cases of immunodeficiency is (lesirai)le, and such data should i)e gathered and collated. In young infants, enumeration of hone marrov lymphoid cells may not he helpful
u-i
diagnosis of inlillunodeficiency( Steiner and Pearson, 1966).
Rectal Biopsy
Rectal tissue is readily accessible to biopsy and does not require Prior antigenic
stimulation for roper study. Exaniination
of rectal tissue for plasma cells by ordinary
histology or by immunofluorescence is useful in patients who have markedly diminished serum immunoglobuhin
concen-trations due to excessive intestinal loss of
protein or possibly to failure of
imniuno-gloI)uhin secretion from plasma cells. Ab-sence of plasma cells from the lamina
propria in the rectal biopsy performed
re-fleets deficiency of the local IgA
immuno-globulin system ( Crabbe and Heremans, 1967).
Peripheral Blood
In severe combined immunodeficiency, the blood lymphocyte count is usually
markedly depressed. However, normal h’m-phocyte counts do not exclude this
Chimerism
Chimerism is the simultaneous
occur-rence of two genetically different cell lines
( Dunsford, et a!., 195:3; Woodruff and
Len-1TLOX, 1959 ). In immunodeficiency diseases,
it has been observed in two different forms:
CONGENITAL. This is due to intra-uterine
implanation of niaternal cells into the fetus.
It has been documented only twice, in a
male infant and a male fetus; in 1)0th cases
some of the cells were of female karyotype. The difficulty of diagnosis arising from the
extremely small nuniber of lymphocytes in
the blood may be overcome by preparing
S(1uaSh preparations of unstimulated bone
marrow cells. The condition is thought to be
caused by a take of transplacentally trans-mitted cells in a fetus having
immunodefi-ciency. The establishment of an
immuno-competent population of foreign cells may
lead to iii tra-uterine graft-versus-host
dis-ease. witil all its Consequences.
ACQUIRED. The successful implantation of
any kind of foreign cells into subjects with
immunodeficiency leads to chimaerism
( Beilby, et a!., 1960; Mathe, et a!., 1963;
Bronson, McGinniss and Morse, 1964 ). This
has i)cen Ol)Served particularly after
recoii-stitution of infants with combined
immuno-deficiency disease using bone marrow from
other than fetal donors. It can be produced by simple blood transfusion when it has
often been overlooked. The danger for the patient lies in the production of
graft-ver-stis-host disease prior to the establishment
of true chimaerism, unless perfectly matched
donor cells are transferred.
Tests for Early Diagnosis in Infants
These may be a1)1)hed to neonates in
families that have previously had children
vith immunodeficiency.
Severe Combined Immunodeficiency
Early diagnosis is iniportant l)ccause
treatment is possible and should be
insti-ttited i)efOre severe infection has occurred.
The serum IgG level is not useful for
diag-nosis. The lymphocyte count may be low,
hut this is not consistent in the newborn. Low 1gM concentrations after tile first few
days of life may help in diagnosis. Lympho-cyte uptake of tritiated thvmidine on
phyto-hemagglutinin stimulation is mature at birth
in normal infants. Such a normal response
in cord or venous l)lOOd provides a
satisfac-tory method for exclusion of severe
corn-I)ined immunodeficiency (Papiernik, 1969).
In cases with little or no response, the full
range of immunity function tests, listed
above, should be performed. Transient
by-poimmunoglobulinemia may occu r in
families in which one member has been
shown to have certain combined
inimuno-deficiency states-presumably as a
manifes-tation of heterozygosity for an abnormal
autosomal gene ( Soothill, 1968 ). Repeated
immunoglol)ulin esti niations in infants in
these families with a normal PHA response,
may anticipate a potentially dangerous, 1)ut
transient flfld treatai)ie, IgG deficiency.
X-Iinked Hypo-immunoglobulinemia
Low serum 1gM concentration in the see-ond and subsequent weeks of life provide an early, presymptoniatic prediction of tile affected boys in these families ( Soothill,
1968).
Thymic Hypolasia
Immunity function tests should be
per-formed in all cases of neonatal tetany.
Other Immunity Deficiency Diseases
No early or antenatal diagnosis has Ixell
reported.
V. DEFINITION AND CLASSIFICATION OF
IMMUNODEFICIENCY STATES
CLASSIFICATION OF PRIMARY SPECIFIC
I M M UNODEFICIENCY
Primary specific inimunodeficiencv re-stilts from a failure to produce tile effectors of the immune response, i.e., antil)odies and
sensitized lymphocytes. Excluded from tile
definition are ilypercatabolic states, imm
u-nodeficiency states due to exogenous
938 PRIMARY IMMUNODEFICIENCIES
and immunodeficiency states associated with lymphopenia due to intestinal lym-phangiectasis, with neoplasia (myelomatosis, leukaernia, and so forth ) , with complement
defects (C3 or C5 aI)normality ) , and with
phagocytic dysfunction syndrome .t The
lationship between immunodeficiency and
infection is complex, since each may lead to
the other.
The extraordinary variability of
immuno-logical findings preseilts a major difficulty
in classification. Often neither etiological,
nor functional, nor structural considerations.
taken aloiie or together, are satisfactory for
adequate classification of deficiency of im-munity in man. Well characterized diseases
associated with clear-cut immunodeficiency
are listed in Table II. However, the majority
of patients with immunodeficiency cannot
vet be unequivocally classified, and are
therefore grouped under the heading of
variable immunodeficiency.”
The variable inirnunodeficiency group presumably includes many syndromes he-cause of lack of information on definite pat-terns and causes. Included in this group are
cases previously classified as “congenital”
non-X-hinked ( or sporadic) hypogamma-glohulinemia, primary “dysgammaglobu-hinemia” of both childhood and adult life, and “acquired” primary hypergammaglo-huhinemias. It is hoped that careful
analy-sis of such patients, including tinie of onset
of the condition, immunoglobuim patterns,
family studies, and studies of associated
dis-Ordlers will result in delineation of several homogeneous syndromes based on
estab-hisil(’d hereditary mechanisms or other
eti-ological factors. In spite of interesting
cx-l)(’rim(ntai data on the effect of endocrine
deficiencies oii the development of the im-mune system, no clear-cut clinical sylldrOmes maii ifesting th is relationship have yet l)een reported
I Such causes of susceptibility to infection,
esl)eciallV thOS(’ of genetically determined etiology,
shoulti 1)e investigated l)\ appropriate tests in P’
tients ‘vitli chronic infection suspected of PrimtrY
he nnmunodeficiencv (Douglas an(l
Fuden-berg, 1969).
Associated Disorders
MALIGNANCIES AND IMMuNoDrncnNcy:
\‘Ialignancies occur frequently in patients
with certain types of immunodeficiency
diseases. Malignancies are especially
fre-quent in ataxia-telangiectasia and the Wis-kott-Aldrich syndrome, about 10% of such
cases having died from this cause;
ma-hignancies appear to be even more fre-quent in the common variable form of
im-Illunodeficiency ( Gatti and Good, 1970).
Tile lymphoid system is frequently involved, l)ut the incideilce of epithelial tumours is also unexpectedly high. In X-linked agam-maglobulinemia, acute leukemia has been reported in a few cases; this may represent
a high frequency in view of the rarity of
this disorder (Page, Hansen and Good,
1963).
AUTOIMMUNE DISEASE: The high
mci-dence of auto-antibodies, with or without autoimmune disease, in patients with
im-niunodeficiency has been well documented,
i)Iit the reason for the association is not
known. The association may be related
di-rectly to tile genetic abnormality, or to
ef-fects of the immunodeficiency, such as latent
virus infecton (Fudenherg, 1966).
VI. TREATMENT OF SPECIFIC
IM M UNODEFICIENCY
The following are the main lines of
treat-ment of immunodeficiency. Although most
are of obvious clinical benefit in certain
situ-ations, oiliy immunoglobulin replacement
tilerapy has I)een shown to l)e of value by
controlled trial. Early diagnosis and
treat-illellt give the best results.
Prophylaxis of Infection
This can safely he attempted by hygiene arid external antibacterial measures, al-though there is no evidence tilat this method is effective. Prophylactic
antibiot-ics, although possibly of 5011W value in
spe-cml cases, are not generally recommended
as they may induce infection with fungi or
other resistant organisms. Individual
doses of appropriate antii)iotics. The
sensi-tivity of organisms to antibiotics should be tested, and if possible, drugs of narrow
SI)ectrulll should be used.
Inlmunodeficiency due to an identifiable
cause, such as pr0tei1lUril, intestinal
iro-tein, and lymphocyte loss, cytotoxic drugs, and so forth, may occasionally be radically
cured. Prevention of primary
immuno-deficiency by genetic Coullselhing is possible,
although caution should he exercised in
as-suining a nlOde of inheritance in any
partic-ular family. Immunodeficiency in congeni-tal rui)ella is also preventable.
Immunization with the antigens listed above is safe and is recommended. Live
antigens ( viral and BCG ) must be avoided
in all cases. A case has been reported of selective deficiency of humoral and cellular
immunity to a single organism (
staphylo-coccus ) associated with repeated infection
apparently due to tolerance. Immunization with a cross-reacting antigen ( modified
leucocidin ) was followed by a cellular and humoral response and freedom from infec-tion for a limited time ( David, Douglas and Fudenberg, 1969). This type of treatment deserves further study ill other selective deficiencies.
Blood transfusion carries a grave risk of graft-versus-host disease in patients with
immunodeficiency. Old blood is not safe;
however, fresh blood can be made safe by
irradiation when fully saturated with
oxy-gen, ‘hen 1000 r will kill all the lymphoid
cells. ( In nonoxygenated blood, 6000 r is not effective ). Blood is also safe when
pro-cessed by the “cytoagglomerator” (
Hug-gins, 1966 ) and frozen in glycerol at -60#{176}C. Washed red cells or unprocessed
plasma are dangerous.
Dialvzable transfer factor prepared from
lymphocytes from a selected donor was
given to one patient with the Wiskott-Al-drich syndrome. This was followed by
tamed restoration of delayed skin
hvpersen-sitivity, production of migration ininbitioll
factor, and marked clinical improvement
( Levin, et a!., 1970) . This important
ob-servation needs confirmation.
Replacement of Immunoglobulin
Intramuscular illjection of alcohol- or
etiler-fractionated immunogiol)ulin at a
minimum dose of 0.1 g/kg/montll, or 0.025 g/kg/week, has been used with
oh-vious benefit over the ‘ast 15 years. On this
reginlell, patients vitil X-iinked
hypo-im-nlunoglobulinemia have reached adult life
and rarely suffer from specific viral
cx-anthenla. A small but statistically significant
effect of therapy has 1)een shown in a
con-trohied trial of patients with
hypo-immuno-glohulinemia receiving either the above
dose or twice as much ( Medical Research
Council, 1969 and 1970). In many patients,
the smaller dose was obviously effective
and tile difference in effects produced by
the two doses was small, i.e., the lower dose was clearly adequate for many
pa-tieilts hovcver, tile trial confirmed
previ-ous impressions that some patients benefit
froni higiler doses.
The injections may l)e associated with
re-actions-hypotension, collapse, flank pain,
dyspnea, fever, rasiles, and even death-but these are rare in X-linked agammaglo-buhinemia. These reactions are thought
sometimes to be complement mediated.
Ag-gregates of immunoglobulin, with or
with-out antibody response to them ( Barandun,
et a!., 1962; Henney and Ellis, 1968 ) , and
antibodies to IgA (Vyas, Perkins and
Fu-denberg, 1968 ) , and to genetic
determi-nants (“allotypes” ) of immunoglobulins
lacking in the patient may be responsible
(Fudenberg, et a!., 1964; Vyas and Fuden-i)erg, 1969).
Intravenous Immunoglobulin
Preparations intended for intramuscular
use must not 1)e given intravenously
i)e-cause of frequent severe reactions. Tile
ill-cidence of such reactions can he reduced
by various treatnlents; e.g., acid and pepsii.
The in vivo half-life of molecules so treated
is usually Illtich reduced; e.g., to 24 hours;
but large quantities can be given quickly
With OillV a small risk of reaction. Such
940 PRIMARY IMMUNODEFICIENCIES
)ro1)hylaXis, bt may be useful in severe
in-fection. Never mctllOds of l)reParations,
e.g., with hunlan fibrinolysin, capr1ate,
kaolin, and hentonite, wilich result in
re-moval of aggregatcs hut which do not de-stroy tile Fe fragment, 1l)Pear promising
111d deserve further study. These
prepara-tions survive vehl in vico.
Tests for Aggregation of Immunoglobulin
Reactions to 1)0th intramuscular aild
in-travenous l)rel)arations may in part be
re-latedi to aggregation. Various tests for ag-gregation are available. Ultracentrifugation
is relatively insensitive, but anticomplement
activity and precipitin reactions witil the
complerneilt breakdown product C, or
rheumatoid factor are IllO sensitive. The
rheumatoid factor test ( Edelman, Kunkel
and Franklin, 1958 ) can l)e performed
sim-ply with whole rheumatoid arthritis serum,
and nligllt i)e a valuable screening test for
ampoules of immunoglobulin to he given to pati(’nts with immunodeficiency.
Infusion of Plasma
liltravenous infusion of l)lasllla, obtained 1)y plasmapheresis of a few donors ( the
“buddy systelil” ) and given to the patient after piasmapheresis ( to avoid
overload-nig ) ,permits repiacenlent of a wider range
of immunogloi)u liiis (Stiehm, Vaerman and
Fudenberg, 1966 ). The larger quantities
that cai l)e given ill this ‘ay permit
treat-ments to i)(’ nore widely spaced. It is
esseii-tial that tile plasma he frozen and thawed
i)efOre USe to destroy cells that may cause
graft-versus-host disease. This treatulent
has i)een of benefit in severe
hypogamma-gloi)ulnlemia, particularly in I)atiellts with
diarrhea \‘IlO are unresponsive to otiler
treatment, and in certain patients with
itaxia telangiectasia (Amniann, et (ii., 1969),
i)Iit slloUld Ilot be given to those with anti-IgA antibodies.
Indications for Immunoglobulin Treatment
in Immunoglobulin Deficiency
Current I)reparations are largely I gG
with about 5 IgA. The effect of treatment
is thus mainly confined to transferring IgG
antibodies to the blood and interstitial fluid.
The recommended doses usually restilt in
an increase ill SCUfll IgG of 100 to 200 mg/
100 ml. There is a prima facie case for this
treatment Iii Iatiellts with IgG deficiency.
The effectiveness of treatnlent will depend
on the nature of the associated immunity
deficiencies. It is perhaps most successful in
X-hinked aganlmagiobuiinemia, the IgG
deficiency of prematurity, and transient by-pogammagioi)ulinemia. It makes no great difference to the prognosis of patients with severe combined immunodeficiency, and it
is uncertain to what extent it can replace
such local functions as the secretory IgA sys-tern where this is deficient. There is evidence that IgG therapy may have prevented re-current menmgitis in patients with 1gM an-tibody deficiency ( Medical Research
Council, 1970 ). Patients with normal
iflinlu-noglohuiin levels, but with either a general
or restricted defect in antibody response,
may also i)eilefit from immunoglohulin
treatment. In spite of satisfactory
improve-Illeilts in other respects, smo-puirnonary
suppuratioii ulay still persist in some
pa-ticnts on immunoglobuhin therapy.
Hyperimmune Immunoglobulin
Immunoglobulin from donors
hyperim-triune to individual infections, e.g., tetanus,
measles, or pseudomonas may be of value
for patients ex1)osed to tilese infections.
This probal)ly applies not only to l)atiellts
vith prinlary immunodeficiency, i)ut also to
those with defects of Ilonspecific immunity
mecilanisms or with hypercatabohic imm
ii-Ilodeficiency.
Treatment of Immunodeficiency by
Graftingf
Severe Combined Immunodeficiency
Efforts to correct the functional
deficien-cies in tilese l);ltieilts by grafting of tissues
:$:Because of tile hazard both from ilospital
)ati1ogens and froni the host’s own microbial flora,
tile uS(’ of laminar flow isolation comi)ined with
elinlination of gut flora with antibiotics is to i)C
have included the following:
1. Injections of adult marrow, spleen,
lymph node, and peripheral blood cells
from ilonmatched immunologically mature
donors (Miller, 1967; Hong, Gatti and Good, 1968).
2. Thymus transpiailt from nonniatched
fetal or adult donors.
3. Combined fetal liver and tilymus
transplant from the same unmatched donor
(Hong, et a!., 1968).
4. Whole bone marrow from donor matched with recipient at the HL-A locus by cytotoxic typmg and nlixed leucocyte re-actions ( Gatti, et a!., 1968; Ammann, et a!.,
1970).
5. Implantation of presumed stem cell fraction from immunologically mature
matched donor, with or without thymus
graft (Dekoning, et a!., 1969).
6. Implantation of presumed stem cell fraction or whole marrow from parent
con-bined with prior administration of enhanc-ing antibody directed against HL-A
anti-gens of the recipient ( Soothihl, unpublished results; Buckley and Rowlands, 1970).
7. Implantation of presumed stein cell fraction from parent, combined with immu-nosuppressive regimen ( Good, 1969;
Men-wissen, Terasaki and Good, 1970).
Methods 4 and 5 have both completely corrected combined immunodeficiency.
In many instances, injection or
irnplanta-tioii of blood cells, or cells from
hemato-poietic tissues, of inimunologically
compe-tellt donors has reconstituted the immune
capacity of the recipient to a varying
de-gree, but has regularly led to a fatal
graft-versus-host reaction ( Miller, 1967 ). This
re-action appears 8 to 20 days after injection
of the foreign immunologically competent cells, and is usually manifested by fever,
Coombs’-positive hemolytic anaemia, a
characteristic erythematous maculopapular
skin rasil, bloody diarrhea,
hepatosple-nomegaly, aregenerative pancytopen ia,
and death. These severe graft-versus-host
reactions in man have not been treated
effectively by any regimen. Therefore,
ad-iiiinistration of s’hole blood or
transplanta-tion of inlliluilologicahl competent cells
without profound modification is strictly to
he avoided ill children \vitil severe coni-billed immuiiodeficiency (Hong, Gatti aixi
Good, 1968; Good, 1969).
Thymus traiisplantation from ci thier fetal
or nlature donors has also i)cen tried Oil
nu-merous occasions, and has never fully cor-rected the immunodeficiency (Hitzig,
1968 ). Establishment of ability to produce
an electrophoreticahly restricted i
nlnluno-globulin attributable to donor cells has
been reported ill Oil case ( Harboe, et a!.,
1966 ) , but this was not associated with
res-toration of imnlune competence. It is
coil-eluded that this approach has i)ecn given
sufficient trial and has provel ineffective
( Meuwissen, et a!., 1969).
Combined fetal liver and fetal thymus
transplants have been tried repeatedly as
an approach to correction of severe
corn-buied iflllllunodeficiencv. These transplants
sometimes seem to have given temporary
and minimal imnlunoiogical reconstitution,
but no lasting or clinically sigilificant
cOllstitiltloflS have been reported. It is
con-eluded that tilis method has l)eell given a
thorough trial and has not fully corrected
the inlmune deficiency in any instance.
Treatmen t with uiifractionated l)one
narrow from a matciled sibling donor has
l)eei1 attenlpted, and in every instance has
estal)lishled an immuiie capacity in the
re-cipiellt cilild. At least five examples of
Ill-munological reconstitution are known, and!
three of these five cilildren have remainedl
immunologically normal up to tile pr’seiit
time for as long as 2 years following the
transplant. Tue two other children, aithougil
dramatically restored immunological iy, died
of pre-existing pneuniocystis carinii
Pull-monary infection . Graft-versus-host
reac-tions have often been observed in such
cases; some ‘ere acute and severe, others
I)r5i5tel as a low-grade reaction for many
months, hut (‘veil without treatment iioie
\V15 fatal.
These findings parallel those of
reconsti-tuition experiments in mice and rats, in
se-942 PRIMARY IMMUNODEFICIENCIES
verity according to tile degree of genetic
disparity between the donor and host
tis-sues.
Another cllild has been completely
re-constituted immunologically by treatnlent
Witil d SteIll cell fraction ( albumin gradient
technique ) from a matched sibling donor.
No graft-versus-host reaction was oI)served.
This child also was given a transplant of
fetal thvmus Prior to the full correction of
immune capacity by the stem cell
implanta-tion ( Ammann, et a!., 1970 ). Several
ad-ditional attempts to reconstitute children
having severe combined immunodeficiency i)y implants of stem cell preparations from matched sibling donors have failed.
In one instance, a “marrow stem cell
frac-tioll from the best-matched parent, when
combined with an immunosuppressive
regi-1li(ll based 011 amethopterin and
anti-lym-)llocyte globulin, resulted in encouraging
reconstitution of immune capacity of the
re-cipient. This child died from intercurrent
Sel)ticaemia 7 weeks after the transplant
(I)ekoning, et a!., 1969; Meuwissen,
Tera-saki and Good, 1970).
Marrow transplantation from parent to
child has been carried out in two cases in
‘ilich an attempt was made to suppress
graft-versus-host reaction by antibody
given to the child. One had a single detect-able HL-A antigen that was lacking in the father. Human antiserum to this antigen
was given, without obvious toxic effect, Prior to giving the father’s marrow cells;
tile graft took, and lasted for some weeks
with evidence of reconstitution of humoral
and cellular systems and without detectable
graft-versus-host disease. Nonetheless, the
immune system finally failed and the child
died.
In the second child, marrow
transplanta-tioll fronl parent to child was combined
vitii injecting the child with anti-HL-A
an-tihodies from the mother directed towards
th(’ father’s antigens, 1)0th prior to and
re-peatedly after marrow transplantation from
tile mother. Reconstitution of hunloral
im-lTIillie responses aild, to sonic degree, of
cell-mediated immunity was achieved
( Buckley,et a!., 1970 ). This approach
should be encouraged because many such children do not have a matched sibling.
ESTABLISHMENT OF IMMUNO-COMPETENCE
CAN BE EVALUATED BY:
1. Improvement of clinical status; e.g., rapid resolution of momliasis, and so
forth.
2. Serial measurements of immunoglob-ulin levels; immunoelectrophoresis may be especially useful, since the appearance of
homogeneous immunoglobuhins has been
documented.
3. Appearance of specific humoral
anti-bodies following antigenic stimulation.
4. Appearance of isohaemagglutinins of
donor origin.
5. Appearance of positive cell-mediated
immune reactions.
ESTABLISHMENT OF CHIMAERISM : This is
the fllOst reliable evidence for the take of a
graft. It can he shown by chromosome stud-ies of lymphocytes if the donor was of the
opposite sex to the recipient. Furthermore,
appropriate tests for red cell antigen
mo-saicism and isohemagglutinin changes
silould always be performed.
These tests should be repeated periodi-cally in successful cases, since subsequent
gradual decline has been observed in some
instances.
In summary, combined system
immuno-deficiency has now been repeatedly cor-rected by bone marrow transplantation. Administration of marrow from a matched sibling donor, when it is available, seems the best approach. Thymic grafting is not es-sential to restore immune capacity in most
of these children. Stem cell fractionation
may help to reduce or eliminate
graft-ver-sus-host reaction, but present techniques
may impair the effectiveness of the
trans-plant. \Vhen a nlatciled sibling is not
avail-able, marrow from a parent donor seems
the next best approach. Use of anti-serulll
against HL-A antigens of the host may be a
valuable adjunct for avoiding
Thymic Hypoplasia
Theoretically, the defect in this disorder silould he correctable by an adequate
trans-plant of thymus tissue. Evaluation of the
in-fluence of corrective therapy is difficult
be-cause tile degree of thymic development is
variable and some patients have shown
de-velopment of imnlune capacity in spite of
deficiencies demonstrated early in life.
In two patients, transplantation of
thy-mus from fetuses of 12 and 16 weeks’
ges-tation appeared to reconstitute
thymus-depen dent immune functions that were
previously lacking in these children. As
nugilt be predicted from animal
experi-ments, the grafted thymus was rejected in
olle instance and probably in the other. In
one of these cases, immune function has subsequently declined.
Further efforts to reconstitute immulle
function by thymus transplantation, in
ill)-munodeficient children with complete thymic aplasia or severe thymic hypoplasia, are strongly indicated.
Other Disorders
Long-lasting marrow transplants from
matched sibling donors have now also been
achieved in patients with several other con-ditions, including the Wiskott-Aidrich syn-dronie, chronic mucocutaneous candidiasis, lymphoma, and leukaemia. In these condi-tions, successful transplantation of marrow has been based on administration of whole marrow from matched donors, conlbmed with vigorous immunosuppression. In none of these conditions, however, has the mar-row transplant fully corrected the underly-ing disease. The indications for marrow transplantation from nlatched sibling donor deserve further evaluation in patients with
tile \Viskott-Aldrich syndrome. Marrow’
transplants from ideally matched sibling
donors should be tried in other patients,
also, as in those with hitherto unclassified
immunodeficiency states. Bone marrow’
transplants from nonmatched donors are
probably contraindicated in the
Wiskott-Aldrich syndrome, and marrow’ transplants
seeiii of no ue in X-linked agammaglobuhi-naemia.
VIII. CONCLUSIONS
1. Two registries are to be established in
association with WHO, with the object of
gathering and recording information on
cases of inirnunodeficiency from all over the
world. One registry will comprise cases of
primary inimunodeficiency that have re-ceived transplants of any immunologically competent organ, cell, or cell product, e.g., bone marrow, thymus, or transfer factor. It
will he organized by W. Hitzig and M.
Se-higmann. The other will comprise cases of
malignant tumour found in patients with
primary immunodeficiency; it will i)(’ orga-nized by R. A. Good, who will also arrange for pathological material to be sul)mitted
for study by selected pathologists.
The organizers will collate and
periodli-cally report oii the material in both
regis-tries.
2. Standardization of diagnostic
iroc’-dures and additional studies related to
di-agnostic tests are required.
ci. Tllere is a need for standard
pr(’pa-rations of antigens for assessing antibody
fornlation as described. A sufficiently large
batch of such antigens should be prepared, and ahquots silould be sent, on request, for use by clinicians investigating
immunode-ficiency.
1). ‘s/IethO(ls for measuring seruni
anti-bodies iieed to be standardized, and in citro
tests for the measurement of cell-mediated
immunity, such as tests for cytotoxicitv and
macrophage migration inhibition factors,
need to l)e further developed.
3. Therapeutic proc(’dures should be
standardized and assessed.
a. Aggregates in immunoglobuhin
pr’p-arations for intramuscular or intravenous
injection should be looked for with tests
such as those recommended.
1). National agencies responsii)le for
imlllulloglol)uhiil production should develop
stable aggregate-free intact IgG, without
