CA2849640A1 - Serum bactericidal assay for n. meningitidis specific antisera - Google Patents

Abstract

The present invention relates to the field of Serum Bactericidal Activity (SBA) assays for Gram negative bacteria, in particular N. meningitidis. The SBA assay is the most important method for measuring functional activity of serum antibodies against meningococcus. In order to determine whether a subject or a population is seropositive against invasive meningococcus the SBA test should ideally be both sensitive and specific. The inventors have found the standard N. meningitidis serotype A and W SBA can be significantly improved in this regard.

Description

SERUM BACTERICIDAL ASSAY FOR N. MENINGITIDIS SPECIFIC ANTISERA
FIELD OF THE INVENTION
The present invention relates to the field of neisserial vaccine compositions, and methods to assess the quality of the immune response elicited after their administration, for instance laboratory correlates of protection, and in particular serum bactericidal activity (SBA).
BACKGROUND OF THE INVENTION
Neisseria meningitidis (meningococcus) is a Gram negative bacterium frequently isolated from the human upper respiratory tract. It is a cause of serious invasive bacterial diseases (on an endemic and epidemic scale) such as bacteremia and meningitis.
The incidence of meningococcal disease shows geographical, seasonal and annual differences (Schwartz, B., Moore, P.S., Broome, C.V.; Clin. Microbiol. Rev. 2 (Supplement), S18-S24, 1989). The bacterium primarily affects children from 6 months to 2 years, but also affects teenagers. The bacterium is commonly classified according to the serogroup of its capsular polysaccharide.
Most disease in temperate countries is due to strains of serogroup B and varies in incidence from 1-10/100,000/year total population - sometimes reaching higher values (Kaczmarski, E.B. (1997), Commun. Dis. Rep. Rev. 7: R55-9, 1995; Scholten, R.J.P.M., Bijlmer, HA., Poolman, J.T. et al. din. Infect. Dis. 16: 237-246, 1993; Cruz, C., Pavez, G., Aguilar, E., et al. Epidemiol. Infect. 105: 119-126, 1990).
Epidemics dominated by serogroup A meningococci, mostly in central Africa, sometimes reach incidence levels of up to 1000/100,000/year (Schwartz, B., Moore, P.S., Broome, C.V. Clin. Microbiol. Rev. 2 (Supplement), S18-S24, 1989). Nearly all cases as a whole of meningococcal disease are caused by serogroup A, B, C, W-135 and Y
meningococci, and a tetravalent A, C, W-135, Y capsular polysaccharide vaccine is available (Armand, J., Anninjon, F., Mynard, M.C., Lafaix, C., J. Biol. Stand.
10: 335-339, 1982). The available polysaccharide vaccines are currently being improved by way of chemically conjugating them to carrier proteins (Lieberman, J.M., Chiu, S.S., Wong, VIC., et al. JANIA 275 : 1499-1503, 1996). The serogroup B capsular polysaccharide has been found to be noninununogenic - most likely because it shares structural similarity with host components (Wyle, F.A., Artenstein, M.S., Brandt, M.L. et al. J. Infect. Dis.
126: 514-522, 1972; Finne, J.M., Leinonen, M., Makela, P.M. Lancet ii.: 355-357, 1983).
The frequency of Neisseria meningitidis infections has risen in the past few decades in many European countries. This has been attributed to increased transmission due to an increase in social activities under crowded conditions (for instance discos, swimming pools, theatres, etc.). It is no longer uncommon to isolate Neisseria meningitidis strains that are less sensitive or resistant to some of the standard antibiotics. This phenomenon has created an unmet medical need and demand for new anti-microbial agents, vaccines, drug screening methods, and diagnostic tests for this organism.
The development of effective vaccines requires reliable tests for establishing whether an effective immune response has been elicited in vaccinated individuals. For N. meningitidis serogroup A, B and C vaccines Serum Bactericidal Activity (SBA) assays are seen as being the gold standard in the field (with a four-fold increase in SBA
being accepted as surrogates for protection for these serogroups) [Borrow et al., 2001, Infect Immun 69:1568-1573; Vermont et al., 2002, FEMS Immun and Med Microbiol 34:89-96]. The assay can give information on: a) whether any particular serum sample has a level (titre) of bactericidal antibodies sufficient to reach a protective threshould, b) a % seropositivity (%SP) of subjects surpassing this level, c) a %
seroconversion (%SC) of subjects either i) with a ratio of SBA titres (post- to pre-immunisation sera titres) which increase beyond a particular multiple (4-fold), or ii) as for %SP if the level of SBA titres pre-immunisation in the subjects (i.e. the very young) is below the protective SBA threshould. This assay is therefore used as a primary read-out (of protection and vaccine efficacy). In addition to SBA assays IgG ELISA assays are often used as a secondary readout. These ELISA assays measure the concentration (in jig/m1) of IgG
antibodies in a serum sample which bind to a meningococcal capsular polysaccharide (A, C, W135 or Y).
In A, C, W-135 (or W), and Y SBA assays the effect of respectively specific anti-serogroup A, C, W and Y Neisseria meningitidis antibodies are evaluated in the presence of complement (typically from baby rabbit [a rSBA assay], or alternatively from human sera [a hSBA assay]). A mixture of bacteria and complement is added to the sera. The meningococcal specific antibodies bind to the target cell-surface via meningococcal-specific protein or carbohydrate moieties. The Cl q subunit of Cl binds to the Fc portion of the surface-bound Ig. The binding of Cl q to Ig activates the classical pathway of the complement which ultimately results in death of the target cell. The bactericidal titre for each serum is expressed as the reciprocal serum dilution corresponding to 50% killing.
SBA assay methods are well-known in the art. In 1976 WHO Expert Committee on Biological Standardization recommended an SBA assay to be used to satisfy the requirements for production and release of meningococcal polysaccharide vaccines (WHO Tech. Rep. Ser. 1976, 594:72-73). Since this time Centers for Disease Control and Prevention (CDC), Atlanta, Georgia, USA has researched into the various to variables in an SBA assay (No. of CFU per well, assay buffer, growth of the target strain, assay incubation time, source of complement, complement concentration, and starting dilution of serum) and has recommended standard SBA assay protocols well-known and used in the field (Maslanka et al. Clin. Diag. Laboratory Immun.
1997, 4:156-167). Typically the CDC also recommends and provides meningococcal strains to be used in SBA assays in the meningococcal vaccine field. For instance for N
meningitidis serogroup C (MenC) SBA assay (WHO guidelines) a consensus has been reached with regard to the recommended assay strain to use.

SUMMARY OF THE INVENTION
As stated above the SBA assay is the most important method for measuring functional activity of serum antibodies against meningococcus. In order to determine whether a subject or a population is seropositive against invasive meningococcus (or has seroconverted post vaccination) the SBA test should ideally be both sensitive and specific. It should be sensitive to protective antibodies induced post immunization with effective vaccines (such as MenAJW/C/Y capsular saccharide conjugate vaccines) but not merely to complement not in the presence of antibodies. It should also be specific so that it registers the presence of functional antibodies against invasive strains, but minimizes the effect of the presence of any low avidity antibodies that might be effective against carriage strains but not against invasive strains.
The present inventors have found that the SBA methods currently employed to assess protective antibodies against meningococcal serogroups A and W135 can be significantly improved in one or more of the above respects.
For N. meningitidis serogroup A (MenA) a particular CDC strain is recommended for SBA assays (F8238) ¨ see Maslanka et al., supra. The inventors have found in a clinical trial, however, that although SBA assays employing this strain may determine the protective immune response caused by vaccination (with a MenA
capsular saccharide conjugate vaccine) around 5 months of age, SBA assays using the strain did not efficiently differentiate the responses pre- and post-immunization at a later age (10-15 months). The inventors have found that natural immunity to MenA
strains can start to occur between 1 and 2 years of age by the recognition of epitopes =
on carrier isolates (primarily of the L11 LOS immunotype). This is due in large part to an immune response against non-capsular polysaccharide antigens, and in particular due to natural lipooligosaccharide (LOS) immunization, and possibly low-avidity antibodies against capsular polysaccharide antigens. Surprisingly, it has also been determined that the CDC-approved MenA SBA strain (F8238) predominantly expresses L11 epitopes.
There is therefore a significant problem in the use of conventional CDC-approved MenA assays for evaluating MenA capsular saccharide vaccine compositions. The present inventors have developed advantageous SBA strategies using novel MenA reference strains. The present inventors have also developed SBA
strategies that may be implemented in general for all Gram negative bacteria that have LOS as a component which may lead to natural immunity such that efficacy of a vaccine composition may be differentiated from natural immunity against LOS
epitopes from bacteria isolated from asymptomatic carriers.
In addition the inventors have surprisingly found that the strain provided/suggested by the CDC for N. meningitidis serogroup W-135 (MenW) SBA
assays (strain 54383) is not optimal for this purpose in that it is too sensitive to being killed by the complement source itself without having serum present. This poses significant problems in the use of this strain for SBA assays. Again advantageous SBA strategies using novel MenW reference strains have been developed.
Accordingly there is provided in one embodiment a N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay (e.g. a hSBA or rSBA
assay) comprising the step of determining the SBA titre of a serum sample against, respectively, a N. meningitidis serogroup A or serotype W reference strain having the following characteristics:
a) it has a % complement toxicity of less than 40%, 35, 30, 20, 10, 5%, or is around 0%, and/or that its % complement toxicity is less than that of the MenW

S4383 standard strain available from the Centers for Disease Control and Prevention, Atlanta, Georgia, USA or from the BCCM;
b) the % seropositivity against the strain in a group of at least 25 naive 1-2 year old children where seropositivity is measured by an rSBA assay with titres of 8 or above (to achieve 50% cell killing) is less than 40, 35, 30, 25, 20 or 15%, and/or the % seropositivity against the strain in a group of at least 25 naïve 1:2 year old children where seropositivity is measured by an rSBA assay with titres of 8 or above (to achieve 50% cell killing) is less than that of the % seropositivity against the F8238 standard strain available from the Centers for Disease Control and Prevention, Atlanta, Georgia, USA or from the BCCM; and c) the % seroconversion against the strain in a group of at least 25 3-5 year old children immunized with a unconjugated N. meningitidis ACWY capsular polysaccharide vaccine is more than 50, 60, 70, 80, 90 or 95%, where seroconversion is measured by assessing the % of subjects with at least a 4 fold increase in rSBA titre (to achieve 50% cell killing) when sera is compared just before immunization (when they are naïve) and one month after immunization with the vaccine.

The MenA reference strain in one embodiment is an invasive strain of L10 inununotype. Strain 3125 is well suited for the MenA SBA assays of the invention.
Strain 3193 (W135:2a:P1.5,2:L3) is well suited for the MenW SBA assays of the invention.
In a further aspect of the invention there is provided a Serum Bactericidal Activity (SBA) assay for a Gram negative bacterium comprising the step of determining the SBA titre of a serum sample against said Gram negative bacterium, wherein before use the serum sample is depleted of antibodies against lipooligosaccharide (LOS) epitopes present in said Gram negative bacterium.
to This is achieved in one embodiment by preincubating the serum sample with one or more lipooligosaccharide (LOS) epitopes (and preferably the entire LOS
molecule) present in the Gram negative bacterium to deplete the serum sample of antibodies against the LOS epitopes. In particular this approach is useful for MenA, where serum samples can be advantageously removed of anti-L10 and/or L3,7,9 and/or L1,8 antibodies and/or anti-Lll antibodies, by incubation with each respective LOS epitope/molecule.
Furthermore, kits for carrying out the SBA assays of the invention are provided, as is the use of the above strains of the invention in an SBA assay.

DESCRIPTION OF THE INVENTION
The subject matter of and information disclosed within the publications and patents or patent applications mentioned in this specification are incorporated by reference herein.
Reference to "lipooligosaccharide" (or "LOS") may also be referred to as "lipopolysaccharide" or "LPS". Reference to an LOS epitope of a particular immunotype is also reference to a complete LOS molecule of the particular inununotype.
The terms "comprising", "comprise" and "comprises" herein is intended by the inventors to be optionally substitutable with the terms "consisting of', "consist of', and "consists of', respectively, in every instance.
The SBA assays of the invention as mentioned herein can either be rSBA
(using baby rabbit complement) or hSBA (using human complement) assays except where it is clear from the test that a particular type is specified.
As stated above, the SBA assay is the most important method for measuring functional activity of serum antibodies against meningococcus. To date SBA has proved to be the best surrogate of protection against meningococcus (it correlates highly with immunity to meningococcal disease). Typically 'pre' and 'post' immunization sera are collected from a naive subject (for instance 'pre' collected from the subject just before vaccine administration [never having been immunized with a vaccine comprising the antigen in the vaccine previously], and 'post' 1 month post immunization) and bactericidal dilution titres to meningococcal strains are determined by an hSBA assay (e.g. see Borrow et al., 2005 Vaccine 23:2222-2227) or an rSBA
assay according to the CDC protocol [e.g. the MSG96 protocol] (Maslanka et al., supra). Bactericidal titres for pre- and post- immunization sera are expressed as the reciprocal of the dilution resulting in 50% killing of the bacteria (either the 2-fold serum dilution corresponding to at least 50%, or determined exactly by a mathematical fit from a series of 2-fold serum dilutions) as compared to the number of target cells present before incubation with serum and complement (for instance if the bacteria are in the presence of active complement but without serum). The assay can give information on: a) whether any particular 'pre' or `post' serum sample has a level (titre) of bactericidal antibodies sufficient to satisfy the correlate of protection (usually this is a titre of >= 4 in an hSBA assay, or >= 8 for a rSBA assay [Borrow et al., 2005, supra], b) a % seropositivity (%SP) of subjects in a study (typically at least 25, e.g. at least 50) pre- and/or post- vaccination with >= the relevant serum correlate level, c) a Geometric Mean Titre (GMT) of all (typically at least 25, e.g. at least 50) the 'post' (and 'pre' if relevant) SBA titres in a vaccine study d) a % seroconversion (%SC) of subjects in a study (typically at least 25, e.g. at least 50) with either: i) a ratio of SBA titres (post-to pre-immunisation sera SBA titres) which increase beyond a particular level (usually >= to a 4 fold increase in SBA titres post immunisation) [typically used for adults where natural immunity may have raised 'pre' SBA titres beyond the correlate of protection level]; or ii) % subjects (typically at least 25, e.g. at least 50) post-vaccination with >=
the relevant serum correlate level (typically used for very young children).
Serum Bactericidal Activity (SBA) assays for Gram negative bacteria are provided in this invention. Suitable methods of carrying out SBA assays are well known to a skilled person (e.g. see Maslanka et al., supra and Examples 1 and herein for optimized Nmen assays). Such assays involve testing a serum sample (pre-and post-immunisation for instance) for the titre required to cause a 50% cell killing against the Gram negative bacterium.
In one embodiment of the invention (as described above) a N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay comprising the step of determining the SBA titre of a serum sample against, respectively, a N.
meningitidis serogroup A or serotype W reference strain having the following beneficial characteristics over strains used in the prior art for such purposes (for instance strains recommended by the CDC):
a) i) it has a % complement toxicity of less than 40%, 35, 30, 20, 10, 5%, or is around 0%, and/or ii) that its % complement toxicity is less than that of the MenW
S4383 standard strain;
b) i) the % seropositivity against the strain in a group of at least 25 naïve year old children where seropositivity is measured by an rSBA assay with titres of 8 or above (to achieve 50% cell killing) is less than 40, 35, 30, 25, 20 or 15%, and/or ii) the % seropositivity against the strain in a group of at least 25 naive 1-2 year old children where seropositivity is measured by an rSBA assay with titres of 8 or above (to achieve 50% cell killing) is less than that of the % seropositivity against the F8238 standard strain; and c) the % seroconversion against the strain in a group of at least 25 3-5 year old children immunized with a unconjugated N. meningitidis ACWY capsular polysaccharide vaccine is more than 50, 60, 70, 80, 90 or 95%, where seroconversion is measured by assessing the % of subjects with at least a 4 fold increase in rSBA titre (to achieve 50% cell killing) when sera is compared just before immunization (when they are naive) and one month after immunization with the vaccine.
Any permutation of the above characteristics is envisaged by the inventors. In particular for any embodiment of the invention with a combination of charactistics a) and b) the following combinations of charactistics are specifically envisioned:
a)i) <40% a)i) <35% a)i) <30% a)i) <20% a)i) <10% a)i) <5% a)i) 0% a)ii) b)i) <40% b)i) <40% b)i) <40% b)i) <40% b)i) <40% b)i) <40% b)i) <40% b)i) <40%
a)i) <40% a)i) <35% a)i) <30% a)i) <20% a)i) <10% a)i) <5% a)i) 0% a)ii) b)i) <35% b)i) <35% b)i) <35% b)i) <35% b)i) <35% b)i) <35% b)i) <35% b)i) <35%
a)i) <40% a)i) <35% a)i) <30% a)i) <20% a)i) <10% a)i) <5% a)i) 0% a)ii) b)i) <30% b)i) <30% b)i) <30% b)i) <30% b)i) <30% b)i) <30% b)i) <30% b)i) <30%
a)i) <40% a)i) <35% a)i) <30% a)i) <20% a)i) <10% - a)i) <5% a)i) 0%
a)ii) b)i) <25% b)i) <25% b)i) <25% b)i) <25% b)i) <25% b)i) <25% b)i) <25% b)i) <25%
a)i) <40% a)i) <35% a)i) <30% a)i) <20% a)i) <10% a)i) <5% a)i) 0% a)ii) b)i) <20% b)i) <20% b)i) <20% b)i) <20% b)i) <20% b)i) <20% b)i) <20% b)i) <20%
a)i) <40% a)i) <35% a)i) <30% a)i) <20% ¨ a)i) <10% a)i) <5% a)i) 0%
a)ii) b)i) <15% b)i) <15% b)i) <15% b)i) <15% b)i) <15% b)i) <15% b)i) <15% b)i) <15%
a)i) <40% a)i) <35% a)i) <30% - a)i) <20% a)i) <10% a)i) <5% a)i) 0%
a)ii) b)ii) b)ii) b)ii) b)ii) b)ii) b)ii) = b)ii) b)ii) Each of these may be separately combined with more than 50, 60, 70, 80, 90 or 95% from characteristic c). Characteristic c) should be assessed using any unconjugated N. meningitidis ACWY capsular polysaccharide vaccine e.g.
Mencevaxn4 from Glax.oSmithKline Biologicals s.a. or the like. This should be assessed in subjects over the age of 3 (e.g. between the ages of 3 and 5, or between the ages of 18 and 25).
With respect to characteristic a)ii) in the embodiments of the invention.
"less than" that of the standard strain can mean at least 10, 20, 30, 40 or 50 %
complement toxicity units fewer than the % complement toxicity for the standard strain.

With respect to characteristic b)ii) in the embodiments of the invention "less than" that of the standard strain can mean at least 10, 20, 30, 40 or 50 %
seropositivity units fewer than the % seropositivity for the standard strain.
Characteristic b)ii) may optionally be replaced herein by the characteristic that the ratio of % seropositivity (carried out as described by b)ii)) of the 3125 strain to the unknown reference strain is 0.1-2.5, 0.5-2.0, 0.7-1.5, or around 1(0.9-1.1).
The reference strain characteristic a), b) or c) [or a) and b), a) and c), b) and c), or a), b) and c)} can be suitably assessed by carrying out the % complement toxicity and/or the % seropositivity and/or the % seroconversion measures, respectively, as described in Example 2h, lb and lc, respectively.
In a further aspect of the invention there is provided a N. meningitidis serogroup W Serum Bactericidal Activity (SBA) assay comprising the step of determining the SBA titre of a serum sample against a N. meningitidis serogroup W
reference strain, wherein the N meningitidis serogroup W reference strain is not sensitive to complement dependent killing in the absence of serum, or at least less sensitive to complement dependent killing in the absence of serum than the standard reference strain available from the Centers for Disease Control and Prevention, Atlanta, Georgia, USA or from the BCCM. For instance the %
Complement Toxicity for the N. meningitidis serogroup W reference strain may be less than 40%, 35, 30, 20, 10, 5%, or is around 0%. The term "less than" that of the standard strain can mean at least 10, 20, 30, 40 or 50 % complement toxicity units fewer than the % complement toxicity for the standard strain. Optionally the reference strain also has characteristic c) as described above.
The sensitivity of the N. meningitidis serogroup W reference strain to complement dependent killing can be ascertained by measurement of % complement toxicity by the method of Example 2b.
The sensitivity of the N. meningitidis serogroup W reference strain used in the assay to complement dependent killing is ascertainable by using a control in the absence of sera in any suitable SBA assay known in the art (e.g. see Maslanka et al., supra). For instance the SBA assay of Example 2b may be used (where %
complement toxicity is measured). The colony forming units (CPUs) obtained for this control are thus the result of the growth of bacteria without addition of human serum sample, the average of the obtained CFIJs for reference strains should therefore ideally correspond to 0% killing.
In particular % Complement Toxicity is a useful assessment of the complement sensitivity of a MenW strain, or any SBA reference strains discussed herein. This can be calculated if a further SBA control takes place with only diluent, bacteria and inactivated complement [e.g. after heat treatment ¨ such as 40 minutes at 56 C before use] (instead of the active complement in the first control) is added to the plates. Ideally 8 or more (e.g. 48) pairs of active and inactive samples are measured to obtain an average measurement. Each active/inactive pair of measurements can be with the same lot of complement source or with different lots, though within each pair the same lot of complement should be used. Any complement source can be used which is suitable for SBA assays (such as baby rabbit complement). The bacterial killing obtained in the samples using active complement is compared with the bacterial killing obtained in the samples using inactivated complement. The results are expressed as % Complement Toxicity by calculating:
% Complement Toxicity Average # CFU Inactivated Complement ¨ Average # CFU
Active Complement * 100%
Average # CFU Inactivated Complement The % Complement Toxicity for the MenW strain obtained from the CDC for SBA assays ¨ S4383 ¨ for instance was around 55% in one experiment.
In typical SBA assays plates are rejected where the % Coinplement Toxicity control on the plate exceeds 40% (otherwise very high SBA titres would be obtained and a high variability of the results). The present invention can overcome this problem and may lead to a reduction in loss of such wasted plates. =
In a still further aspect of the invention there is provided a N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay comprising the step of determining the SBA titre of a serum sample against a N. meningitidis serogroup A
reference strain having the following characteristic:
b)i) the % seropositivity against the strain in a group of at least 25 naïve 1-year old children where seropositivity is measured by an rSBA assay with titres of 8 or above (to achieve 50% cell killing) is less than 40, 35, 30, 25, 20 or 15%, or b)ii) the % seropositivity against the strain in a group of at least 25 naïve 1-2 year old children where seropositivity is measured by an rSBA assay with titres of 8 or above (to achieve 50% cell killing) is less than that of the % seropositivity against the F8238 standard strain. The % seropositivity measure may optionally be carried out as described in Example lb. Optionally the reference strain also has characteristic c) as described above.
With respect to characteristic b)ii) in the embodiments of the invention "less than" that of the standard strain can mean at least 10, 20, 30, 40 or 50 %
seropositivity units fewer than the % seropositivity for the standard strain.
Characteristic b)ii) may optionally be replaced herein by the characteristic that the ratio of %seropositivity (carried out as described by b)ii)) of the 3125 strain to the unknown reference strain is 0.1-2.5, 0.5-2.0, 0.7-1.5, or around 1(0.9-1.1).
In general for the N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assays of the invention for characteristic b) the at least 25 naive 1-2 year old children may be native to the Philippines. For the N
meningitidis serogroup A Serum Bactericidal Activity (SBA) assays of the invention, the N.
meningitidis serogroup A reference strain can be an invasive strain, and can optionally be of immunotype L10 (or predominantly this immunotype), for instance strain 3125, or can optionally be of immunotype L3,7,9 (or predominantly this immunotype), for instance strain 3048. Optionally the group A strain Z1092 (4,21:P1.10) from Marc Achtman (Max-Planck Institut fur Infektionsbiologie, Berlin Gerniany) [Granoff et al.
2005 Vaccine 23:4307-4314] is not a MenA reference strain of the invention.
For the N.
meningitidis serogroup W Serum Bactericidal Activity (SBA) assays of the invention, the N. meningitidis serogroup W reference strain can be an invasive strain, and can optionally be of immunotype L3,7,9 (or predominantly this immunotype), for instance strain 3193. Optionally the group W-135 strain M9262 (2a:P1.5,2) from US
Centers for Disease Control and Prevention (CDC) [Granoff et al. 2005 Vaccine 23:4307-4314] is not a MenW reference strain of the invention. Clearly the F8238 and S4383 strains are not SBA standard strains of the above aspects of the invention.
In the N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assays of the invention the serum sample may be taken from a host immunized with a vaccine composition comprising a capsular saccharide (oligosaccharide or polysaccharide) antigen (preferably conjugated to a carrier protein) from N.
meningitidis serogroup A and/or W.
In the N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assays of the invention the serum sample may have been taken from a human aged 0-55 years, 1 month ¨ 20 years, 2 months ¨ 16 years, 6 months ¨ 10 years, months ¨ 7 years, 10 months ¨4 years, 0-2 years, 6 months - 2 years, or 1-2 years.
A kit is also provided for carrying out the SBA assays of the invention. In such a kit the reference strain(s) of the invention may be provided. This could be provided along with necessary instructions of use in an SBA assay, and optionally other SBA
assay reagents.
The use of the reference strains of the invention (and in particular N.
meningitidis W135:2a:P1.5,2:L3 isolate 3193 from the Netherlands Reference Laboratory for Bacterial Meningitis, Amsterdam [of from ECACC] - the %
Complement Toxicity for this strain was found to be very low [around -9% in one experiment] - or N. meningitidis A:NT:P1.6:L3,6 isolate 3048 from the Netherlands Reference Laboratory for Bacterial Meningitis, Amsterdam [or from ECACC], or an invasive LION. meningitidis serogroup A strain [such as isolate 3125], or strains with similar attributes as described herein) in a SBA assay is a further aspect of this invention.
The inventors have further found that it is at least in part an antibody response against LOS from carriage strains of Gram negative bacteria (e.g.
Meningococcus e.g.
of serogroup A) that can lead to high apparent SBA titres in sera from naive children aged 1-2 years old.
Therefore in a further aspect of the invention there is provided a Serum Bactericidal Activity (SBA) assay for a Gram negative bacterium comprising the step of determining the SBA titre of a serum sample against said Gram negative bacterium, wherein before use the serum sample is depleted of antibodies against lipooligosaccharide (LOS) epitopes present in said Gram negative bacterium. By "depleted" it is meant that antibodies against at least one immunotype of LOS
is reduced in concentration by 20, 30,40, 50, 60, 70, 80, 90, 95 or 100%.
The serum sample may have been taken from naïve host (not immunized with a vaccine composition comprising antigens from said Gram negative bacterium) or from a host that has been so immunized (e.g. one month before serum collection) with a vaccine composition not comprising LOS antigens from the Gram negative bacterium, and wherein the SBA titre is measured with reference to the Gram negative bacterium which comprises an antigen present in the vaccine composition. For instance, the vaccine composition may comprise a capsular saccharide (oligosaccharide or polysaccharide) antigen (preferably conjugated to a carrier protein such as tetanus toxoid, diphtheria toxoid, CRM197, protein D from H
influenzae, etc.) from the Gram negative bacterium.
In one embodiment the Serum Bactericidal Activity (SBA) assay of the invention is performed such that before use the serum sample has been preincubated with one or more lipooligosaccharide (LOS) epitopes or one or more immunotypes of LOS molecule of said Gram negative bacterium to deplete said serum sample of antibodies against said LOS epitopes/immunotypes. By "LOS epitopes" it is meant that oligosaccharide structures present on the LOS responsible for the LOS
iz-nmunotype are used, or mimotopes thereof, or even the entire LOS molecule is used.
In one embodiment the one or more LOS epitopes/immunotypes used to deplete the serum sample comprise one or more LOS epitopes/immunotypes which predominate in carriage strains of the Gram negative bacterium. By "predominate" it is meant that over 10, 20, 30, 40, 50 or 60% of the carriage strains of the bacterium comprise the epitope.
It is envisioned that the SBA assays of the invention will be applicable to any Gram negative bacterium SBA assay where the Gram negative bacterium has LOS.
In one embodiment, it will be of use in SBA assays for neisserial strains, for instance Neisseria meningitidis serogroup A, B, C, W or Y, in particular serogroup A
(MenA).
In such a neisserial SBA assay before use the serum sample can be preincubated with LOS epitopes derived from one or more LOS immunotypes, or with one or more LOS immunotypes, selected from the group consisting of: Li, L2, L3, IA, L5, L5, L6, L7, L8, L9, L10, L11, and L12. For instance a preincubation step may be done with L11 (the most common immunotype of MenA carriage strains) and/or L3,7,9 and/or Ll 0 and/or Li and/or L8 and/or L2.
Pre-incubation can be readily achieved by, for instance, incubating sera samples for 1 hour at 37 C with a 100 g/m1 solution of LOS (a VN mix can be made thus giving a final concentration of 501.tg/mL LOS).

Meningococcal lipooligosaccharides (LOS) are outer membrane bound glycolipids which differ from the lipopolysaccharides (LPS) of the Enterobacteriaceae by lacking the 0 side chains, and thus resemble the rough form of LPS
(Griffiss et al.
Rev Infect Dis 1988; 10: S287-295). Heterogeneity within the oligosaccharide moiety of the LOS generates structural and antigenic diversity among different meningococeal strains (Griffiss et al. Inf. hnmun. 1987; 55: 1792-1800). This has been used to subdivide the strains into 12 inununotypes described above (Scholtan et al. J
Med Microbiol 1994, 41:236-243). Immunotypes L3, L7, & L9 are immunologically identical and are structurally similar (or even the same) and have therefore been designated L3,7,9 (or, for the purposes of this specification, generically as "L3").
Meningococcal LOS L3,7,9 (L3), L2 and L5 can be modified by sialylation. Antibodies to LOS
have been shown to contribute to the bactericidal activity in children infected with N.
meningitidis (Griffiss et al I Infect Dis 1984; 150: 71-79).
In a further embodiment of the invention the LOS preincubation step is added to any or all the SBA assays described herein (e.g. using the meningococcal serogroup A or W reference strains of the invention).
A kit is also provided for carrying out this SBA assay of the invention comprising means for depleting a sertun sample of antibodies against LOS (for instance those means described above) and instructions for its use in an SBA
assay.
The LOS epitopes (preferably entire LOS) for preincubating sera and/or reference strains of the invention may be provided in the kit. Necessary instructions of use, and optionally other SBA assay reagents may also be included. Methods of isolating LOS
from bacteria of appropriate immunotype are well known in the art (see for instance the hot water-phenol procedure of Wesphal & Jann [Meth. Carbo. Chem. 1965, 5:83-91]). See also Galanos et al. 1969, Eur J Biochem 9:245-249, and Wu et al.
1987, Anal Bio Chem 160:281-289. Such methods can be used to make the LOS included in the kits of the invention.
Alternatively, the effect of carriage strains on anti-LOS antibodies in neve hosts may be removed as an SBA assay issue if the SBA assay strain did not cross-react with anti-LOS antibodies. Thus in a still further aspect of the invention a Serum Bactericidal Activity (SBA) assay for a Gram negative bacterium is provided comprising the step of determining the SBA titre of a serum sample against a Gram negative bacterium reference strain which is a LOS mutant either unable to synthesise LOS or which produces a truncated LOS oligosaccharide structure (such that the strain is not sensitive to anti-LOS antibodies in sera samples). For meningococcus, for instance, an lpx% mutant or other known mutants may be used to remove LOS.
Alternatively, highly truncated LOS structures can be produced in lgtA(-), lgtE(-) or galE(-) strains which may be used as reference strains in any or all of the SBA assays of the invention. A kit is also provided for carrying out this SBA assay of the invention comprising such a mutant reference strain, and instructions for its use in an SBA assay.
The examples below are carried out using standard techniques, which are well known and routine to those of skill in the art, except where otherwise described in detail.
The examples are illustrative, but do not limit the invention.

EXAMPLES
Example la: A Typical MenA SBA assay procedure (e.g. using the 3125 or F8238 Men A strain as the reference strain) Day Change Day 1 1. Pre-culture of the Worldn% Seed Working Seed Men A 3125 or F82381 (-70 C).
Spread 50 tit, of bacteria in isolation way across the following plates:
> 3 brain heart infusion + 1% horse serum (BHI) (4 C).
> 1 thayer Martin (TM) -3 selective media for Nrnen with Antibiotic (4 C).
[control plate which showing Nmeningitidis cells are present]
)=. 1 nutrient agar (NA) -3 detection of the contamination (4 C). [control plate showing there is no contaminants growing ¨ Nmen should not grow]
Incubate overnight or for approximately 40-42 hours at 37 C, 5% CO2 with humidity for colonies to grow.
Day 2 2. Culture of the WS on solid media Control the TM and NA media: verify before the test.
Take about 20 colonies (well isolated) of 1 BM petri dish.
Spread these colonies across 1 MIL repeat this operation 3 times.
Incubate during 4 hours at 37 C, 5% CO2 with humidity...
3. Samples and controls preparation (serum samples and controls are inactivated for 40 min at 56 C before use) Dilutions in tube with assay diluent PBS+glucose 0.1%+MgC12 0.5mM+CaC120.9mM pH 7.4 (4 C).
CTRL1: (-20 C) dil: 1/32 in tube (1/128 in microplate).
CTRL2: (-20 C) dil: 1/128 in tube (1/512 in microplate.
(CTRL1/2 are optional controls of serum samples with known SBA titres) Samples 1/8 in microplate: dil: Y2 in tube.
Samples 1/256 in microplate dil 1/64 in tube.
4. Sample and control addition 4.1. Assay diluent addition (in sterile microplate Nunc 96 wells fiat bottom, with microplates lids) Diluent: PBS+glucose 0.1%+MgCh 0.5mM+CaC120.9mM pH7.4.
Distribute 254 of diluent in each well of the microplate.
4.2. Sample and control addition Distribute 25 p.L of sample and control into the first row, column 1 to 5 and to 11.
Mix the solutions in the first row.
Transfer 25 iL from A to B. Repeat the transfers from B to C, ..., row G to H.
After mixing the serum in row H, remove 25 pL and discard.
Distribute 25 1.1.1, of sample into the column 12, sample 1 into the well A, ..., 8 into the well H.
Mix.
After mixing, remove 25 sL and discard.
5. Bacterial suspension preparation 5.1. Harvest colonies from 1 BHI with 5 mL of assay diluent.
Dilute this suspension with assay diluent to obtain an OD of +/- 0.400 (0.380 ¨ 0.420) at 600 nm.
=
5.2. Bacterial suspension Prepare a 1:175000 dilution of bacterial suspension [approx. 114 CFU/25 u1]:
1:10: 1 mL 0.400 absorbance +9 mL assay diluent dil (1:10).
1:100: 1 mL of 1:10 solution + 9 mL assay diluent dil (1:10).
1:1000: 1 mL of 1:100 solution 9 mL assay diluent dil (1:10).
1:175000: 2 mL of 1:1000 solution + 33 mL assay diluent dil (1:17.5).
6. Bacteria / complement mixture addition Rabbit complement Pel Freez (-70 C) [e.g. complement from 3-4 week old baby rabbits]
Thawing of the inactive and active complement is done at room temperature.

Prepare the mixtures of bacteria with inactive and active complement corresponding to 50% of complement and 50% bacterial suspension 1:175000.
Add 25 pL of the inactive mixture into the column 12.
Add 25 L of the active mixture into the column Ito 11.
Cover the microplates with plastic film.
Incubate during 1 hour 30 at 37 C in the orbital shaker (210 rpm).
Day 3 7. Addition of the agar medium Liquefy the agar medium in a microwave and equilibrate its temperature in a water-bath (48 C).
Remove the plastics from the plates and change the microplates lids.
Coat 1: Add 50 pL of the TSB [Tryptic Soy Broth] + 0.9% agar (4 C).
Leave the microplates under the flow-hood at room temperature during 15' to solidify the agar.
Coat 2: Add 50 L of the TSB + 0.9% agar (4 C).
Leave the microplates under the flow-hood at room temperature during 30' (or 15 minutes if sufficient time) to solidify the agar.
Incubate 16 ¨ 20 (or 18-20) hours at 33 C with humidity (without CO2).
Thus rows A to H correspond to 8 2 fold dilutions of sera sample (e.g. from 8 to 1024; or starting at 64 if the serum is known to have high titres), except -for lane 6 which is the blank (0% killing: bacteria + active complement but no serum) and lane 12 (bacteria with the eight serum samples at lowest dilution + Inactive complement [40 min at 56 C before use]). Lanes 7 and 8 are optional control serum samples of known bactericidal titre. Lanes 1-5, 9-11 are the 8 test serum samples on the plate.
8. Reading of the plates and data transfer Remove the microplates lids.
CFUs in each well are counted by any method know in the art, for instance the plates can be read with the KS400 imaging system (Zeiss): MA!
protocol.
Transfer of the raw data and calculation of titres with SoftMax pro.

The bactericidal titer for each serum is expressed as the reciprocal serum dilution corresponding to 50% killing (relative to the 0% killing control [blank] of bacteria + active complement only¨ i.e. 50% killing = 50% of the average CFU for all such controls on a plate).
Example lb: Typical method to calculate % seropositivity Carry out the method of Example la or 2 with the chosen reference strain with sera from at least 25 (for instance at least 50) different naïve children between the ages of 1 and 2 (when protective immunity is low in the majority of such children).
To look at the specificity of the SBA assay, sera from naive children [i.e.
who have not been vaccinated with a MenA and/or MenW vaccine (for instance a capsular saccharide conjugate)], the sera can be assessed for SBA titres achieving 50%
cell =
killing. The % of these subjects with sera achieving a reciprocal titer equal to or more than 8 in the SBA test is defined as % seropositivity.
% seropositivity post immunization with MenA and/or MenW vaccine can also be done in this way.
Example lc: Typical method to calculate % seroconversion Carry out the method of Example la or 2 with the chosen reference strain with sera from at least 25 (for instance at least 50) different children (naive [just prior to immunization] and one month post immunization with a unconjugated N.
meningitidis ACWY capsular polysaccharide vaccine e.g. MencevaxTM from GlaxoSmithKline Biologicals s.a.) over the age of 3 (e.g. between the ages of 3 and 5, or between the ages of 18 and 25).
To assess the sensitivity of the SBA assay, the % seroconversion against the strain is measured by assessing the % of subjects with at least a 4 fold increase in rSBA titre (to achieve 50% cell killing).

Example 2: A Typical MenW SBA assay procedure (using 3193 strain as the reference strain for example) Day Change Day 1 1. Samples and controls preparation (samples and controls are inactivated during 40 min at 56 C before use) Dilutions in tube with assay diluent PBS+glucose 0.1%+MgC12 0.5mM+CaC120.9mM p117.4 (4 C).
to CTRL1: (-20 C) dil: 1/128 in tube (1/512 in microplate).
CTRL2: (-20 C) dil: 1/128 in tube (1/512 in microplate).
(CTRL1/2 are optional controls of serum samples with known SBA titres) Samples 1/8 in microplate: dil: IA in tube.
Samples 1/256 in microplate dil 1/64 in tube.
2. Sample and control addition 2.1. Assay diluent addition (in sterile microplate Nunc 96 wells flat bottom, = with microplates lids) Diluent: PBS+glucose 0.1%+MgC12 0.5mM+CaC12 0.9mM pH7.4.
Distribute 25AL of diluent in each well of the microplate.
=
2.2. Sample and control addition =
Distribute 25 AL of sample and control into the first row, column 1 to 5 and 7 to 11.
Mix the solutions in the first row.
Transfer 25 !IL from A to B. Repeat the transfers from B to C, ..., row G to H.
After mixing the serum in row FL, remove 25 1.11.. and discard.
Distribute 25 1, of sample into the column 12, sample 1 into the well A,.
8 into the well H.
Mix.
After mixing, remove 25 1..rL and discard.

3. Bacterial suspension preparation Working Seed Men W135 3193 (-70 C).
3.1. Subculture of the working seed.
25 L is spread across 1 TM plate and 1 NA plate in massive way (control media).
Incubate overnight at 37 C, 5% CO2 with humidity.
3.2. Bacterial suspension Prepare a 1:18000 dilution of bacterial suspension (no sub-culture needed:
the bacterial suspension is made directly by diluting the working seed (WS) to up to 1:18000 [0D600 of the WS between 0.400 - 0.500]; the diluted bacterial suspension contains approx. 1111 CFU /25 I):
1:10: 200tiL WS + 1800 L assay diluent (1:10).
1:100: 1 mL of 1:10 solution + 9 mL assay diluent (1:10).
1:1000:1 rnL of 1:100 solution + 9 mL assay diluent (1:10).
1:18000:2 mL of 1:1000 solution + 34 niL assay diluent (1:18).
4. Bacteria / complement mixture addition Rabbit complement Pel Freez (-70 C) (e.g. complement from 3-4 week old baby rabbits]
Thawing of the inactive and active complement is done at room temperature.
Prepare the mixtures of bacteria with inactive and active complement corresponding to 60% of complement and 40% bacterial suspension 1:18000.
Add 25 L of the inactive mixture into the column 12.
Add 25 L of the active mixture into the column 1 to 11.
Cover the microplates with plastic film.
Incubate during 1 hour at 37 C in the orbital shaker (210 rpm).
5. Addition of the agar medium Liquefy the agar medium in a microwave and equilibrate it's temperature in a water-bath (48 C).
Remove the plastics from the plates and change the microplates lids.
at 1: Add 1004 of the TSB + 0.9% agar (4 C).

Leave the microplates under the flow-hood at room temperature to solidify the agar (approx. 30 minutes).
Coat 2: Add 254 of the TSB + 0.9% agar (4 C).
Leave the microplates under the flow-hood at room temperature to solidify the agar (15 min).
Incubate 18 ¨20 hours at 35 C with humidity (without CO2).
Day 2 6. Reading of the plates and data transfer Control the TM and NA media.
CFUs in each well are counted by any method know in the art, for instance the plates can be read with the ICS400 imaging system (Zeiss): MAI
protocol.
Transfer of the raw data and calculation of titres with SoftMax pro.
The bactericidal titer for each serum is expressed as the reciprocal serum dilution corresponding to 50% killing (relative to the 0% killing control [blank] of bacteria + active complement only ¨ i.e. 50% killing = 50% of the average CPU for all such controls on a plate).
Example 2b - Calculation of complement toxicity The bactericidal titre for each unknown serum is expressed as the reciprocal serum dilution corresponding to exactly 50% killing of the bacteria. the 0% killing (i.e.
maximum CFUs) is obtained through the blank. The blank in the GSK rSBA assay contains only diluent, bacteria and active complement.
In order to test the complement toxicity of a strain, a simplified SBA assay (compared to [but based on] Examples 1 a and 2) can be set up where only blanks comprising diluent, bacteria and active complement are compared with controls comprising diluent, bacteria and inactivated complement on the plate. The CFUs obtained in wells filled with diluent, bacteria and active complement are then compared to the CFUs obtained in wells filled with diluent, bacteria and inactivated complement (40 min at 56 C before use). In a 96 well plate 8 or more of each should be tested, for instance 48 of each. The CFUs of all active samples can be averaged and compared with the average CFUs or inactive complement samples using the following formula:
% Complement toxicity =
1 Average # CFU Inactivated Complement 1 ¨ Average # CFU Active Complement 1 * 100 Average # CFU Inactivated Complement 1 We have found that Neisseria meningitidis strains with a % complement toxicity comparable to that of MenW 3193 strain (i.e. around 0%) have beneficial %
complement toxicity properties for use in SBA assays (see Example 4).
Example 3: MenA SBA assay results It is known that natural irrununity may play an important role for the protection against Neisseria men ingitidis A.
In a recent clinical study performed in the Philippines in which infants were primed with a menA conjugate vaccine a non-primed group was added as a control.
When the children were aged 10 months it was observed that more antibodies were detected with the MenA SBA test than with the MenA IgG ELISA test (More than 70% of the infants had a MenA rSBA response, while less than 10% had a MenA
IgG
ELISA response in the non-primed groups). This effect may be due to several reasons:
Presence of anti-PSA antibodies which are not measured by the GSK menA IgG
ELISA, or presence of non-PSA antibodies induced by carriage of non-pathogenic neisseria or some other bacteria expressing cross-reactive surface antigens shared with meningococci.
Unimmunised subjects around 5 months old had low SBA titres (<10% with titres more than or equal to 8 using the CDC-approved F8238 strain). However pre-boost at 10 months of age unimmunised SBA went up to 77% with titres more than or equal to 8). Post-boost at 12-13 months, of the 23% unimmunised control subjects that did not have titres more than or equal to 8, 80% of these now did have such a titre.
Distinguishing SBA due to immunization vs. natural immunity was thus difficult.

Presence of other than IgG anti-PSA antibodies were investigated by an anti-total Ig PSA ELISA.
Regarding the MenA strains we hypothesised that LPS was involved in natural immunity. Twelve LPS immunotypes have been described in Neisseria meningitidis corresponding to a diverse range of carbohydrate-structures- The L9-11 are predominantly found among serogroup A meningococci.
We endeavoured to find a way to differentiate between the immune response caused by vaccination and by natural immunity.
= MenA IRTOT ELISA:
First, the presence of antibodies IgG other than anti-PSA antibodies was investigated. In the standard assay the anti-human IgG conjugate was replaced by anti-IgA + M + G conjugate. As can be seen below the concordance between the MenA IgG ELISA and the MenA SBA test is somewhat improved, but doesn't explain the observed natural immunity, consequently a further search related to assay strains and LPS inhibition was performed.
Comparison N IRTOTIELISA Concordance IgG Concordance IQTot, ELISA/SBA ELISA/SBA
DTPw-HepB/HibMenAC-TT-002* (10 Mths of age) Pre Challenge 102 1.31 52% 63%
-Post Challenge 97 1.57 _ 85% 967.
MenCY-P1)-001 (20-24 Mths of age) Pre 16 69% 81%
Post 1 (Mencevax group) 24 1.85 96% _ 100%
* Since the booster phase of the study is still ongoing, it's not known whether subjects were primed or not.
= Use of a different MenA assay strain.
Two actions with the complete strains:
= Tricine gels were run (Figure 1) in order to determine the irnmunotype profile of the current strain ie the F8328 (DC F8238) strain and potential new strains isolated in the Netherlands (3227, 3048, 760570). Although the exact profile of both strains cannot be determined, the gel does indicate that both groups of strains have a different LOS profile.
= In addition an ELISA and rSBA test were run to differentiate strains. In both, the ELISA
and rSBA the respective strains were used as test strain and the the 9-2-L379 mouse monoclonal was used for screening. The assays indicated that F8238 seemed not to be a L3, L7 or L9 strain (actually immunotyped to be a Ll 1 profile - see Example 5 - in contrast to the LI 0 profile suggested by Maslanka et al, supra), while the the Dutch strains were found to react. Based on complement toxicity data the 3048 strain was selected for further evaluation (Data not shown).
LPS inhibition:
= LPS purified from MenB 1144/76 was obtained and sera were inhibited with this L3 LPS.
This was done by incubation for I hour at 37 C with a 1001g/m1 solution (a VN
Mix was made thus giving a final concentration of L3 LPS of 50 g/mL). As seen in the attached tables a full neutralization could be obtained for the discordant serum samples mentioned above (ELISA vs rSBA), while only a partial inhibition sera could be obtained for concordant samples (ELISA vs rSBA).Tbe discordant samples related to pre-immunization (pre-booster) samples, are expected to be highest Sample Time-Point = MenA ELISA (pg/mL) MenA rSBA MenA rSBA
. 196 IgTOT
After neutralization with 50 pg LOS
207 Pre <0.30 0.51 494 .64 , P1M1 5.02 7.08 1466 474 218 Pre 40.30 0.37 265 464 PIM1 121 1.37 245 464 331 Pre <0.30 0.39 195 <64 . _ PIM1 17.05 , 25.70 684 250 337 Pre <0.30 <0.30 371 <64 PIM! 28.50 38.28 2334 1665 339 Pre <0.30 , <0.30 410 <64 PIM] 0.60 1.77 1514 1112 - , 246 Pre 0.77, 1.76 4195 2960 PIM1 9.85 21.07 5108 2144 238 Pre 0.94 1.19 502 <=64 PIM I 38_16 70.25 3333 2291 36 Pre 10_84 17.24 449 171 PIMI 27.75 4122 1876 964 151 Pre 35.91 45.75 1245 386 PIM1 35.66 69.00 1844- 1337 I
µ
Sample Time-Point MenA ELISA (pg/mL) _MenA rSBA MenA rSBA
/96 IgTOT After neutralization with 100 pg 1.05 322 Pre 40.30 40.30 304 464 PlAA1 0.31 0.63 406 464 323 Pre <0.30 1 <0.30 289 <64 PI MI 4.16 I 5.73 1681 315 _-344 Pre Ø30 Ø30 791 <64 PIM1 I <0.30 0.53 - 2571 158 3 Pre 1.91 222 767 464 PIM1 18.34 , 19.28 1673 825 The crucial importance of anti-LPS in the natural immunity against menA has not been described before, and the critical use of this new information to develop assays to evaluate A PS induced response is new.

Example 4 ¨MenW SBA studys = INVENTION
Our proposal is to use a rSBA assay which uses the 3193 (or similar) strain and not the S4383 strain. The S4383 strain was found to be subject to complement toxicity in the GSK rSBA set-up.
= = Killing of N. meningitidis W in SBA tests should occur due to the combination of both anti-MenW
antibodies in the serum sample and complement (in this case an exogeneous source of complement i.e. rabbit complement is used). To check on the intrinsic bactericidal acitivity of rabbit complement two controls are added in each assay. In the first control active complement is added =
to bacteria without antibodies, in the second control inactivated complement is added to the bacteria without antibodies (see Example 2b). Rabbit complement lots are refused in case a bactericidal effect is seen in the controls.
4 The complement toxicity criteria were met with the 3193 strain but not with S4383.
Background = Bactericidal effect should only occur when both antibodies and active complement are added in the presence of complement. To check on this, a bactericidal control with only diluent, bacteria and active complement is added in each assay. The colony forming units (CFUs) obtained for this control are thus the result of the growth of bacteria without addition of human serum sample, the average of the obtained CPUs thus corresponds to 0% killing.
= When a new complement is used a second bactericidal control with only diluent, bacteria but now inactivated complement is added to the plates as well. The bacterial killing obtained in the bacterial control using active complement is then compared with the bacterial killing obtained in the bacterial complement using inactivated complement. This is done as following:
% Complement Toxicity = # CFU Inactivated Complement - # CFU Active Complement * 100%
# CFU Inactivated Complement 4 In case the VG Complement Toxicity exceeds 40% the assay plate is rejected.
Bacteria would thus be killed through addition of complement only. If this criterium were not used, this could lead to very high SBA titres and a high variability of the results.
Data = 108 assay plates were tested using the classical rSBA assay set-up with the S4383 strain. On each plate both bacterial controls were added. Although when all complement toxicities obtained in the different plates are averaged a value of 39% is obtained, the figure below indicates that for many plates the 40% criterium was not met.

= It was GSK's intention to adapt the assay set-up in a later stage in order to harmonize between the SBA assays. Since the complement toxicity problem occurred it was decided to harmonize the assay set-up upfront and then compare the S4383 and the 3193 strain in that assay format Prior to further testing the complement toxicity was now compared by dividing assay plates in two parts:
one part with the bacterial complement with active complement, one part with the bacterial complement with inactive complement. A complement toxicity of 55% was now obtained with the 4383 strain, -9% with the3193 strain.
4 Based on these data it was decided to continue with the 3193 strain.
Example 5: Immunatypes of MenA LOS
Recently, in non-primed toddlers a discrepancy was observed between anti-PSA
IgG ELISA
data and MenA bactericidal data obtained with the reference strain F8238. A
significant level of MenA bactericidal antibodies were measured whilst almost no MenA PS IgG
antibodies were found. These observations suggest that these children have developed a natural immunity against Neisseria meningitidis serogroup A. It was also suggested that this discrepancy could be linked to the use of a SBA MenA strain that expressed a LOS close or similar to the LOS of carriage MenA strains (and not similar to the LOS of invasive MenA
strains).
The aim of this example is to review the literature data in order to assess if carriage and invasive serogroup A meningococci strains express or not similar LOS and to determine their major immunotypes.
Methods Only the papers using the most accurate immunotyping methods were selected_ They are the following:
Zollinger W and Mandrel! R (1980) In this paper, the immunotools are rabbit polyclonal antibodies. Based on the data discussed in the article, it can be conclude that:
= the anti-L10 is specific of the L10 immunotype, = the anti-LI I reacts strongly against the L 11 immunotype but shows some cross-reactivity against the LIO immunotype = the anti-L9 does not react against the L10 and LI 1 imrnunotypes but shows some cross-reaction against L3, L4, L6 and L7 strains Kim JJ et al (1988); Salih Metal (1990) and Achtman Metal (1992) These three papers were selected because they combined immunotyping and the relative molecular weights (MrS) of the LOS. The LOS of N. meningitidis strains is composed of one to six components that can be separated by SDS-PAGE. The IlfrS have been estimated to be between 3150 and 7100. These different eleetrophoretic mobilities reflect differences in the chemical composition of their oligosaccharides and the oligosaccharide compositional differences also account for antigenic differences.

Immunotype AS

Poolman Jet al (1982) and Scholten R et al (1994).
In these two articles, microprecipitation with polyclonal antibodies and/or whole-cell ELISA
typing using a set of 14 different monoclonal antibodies were used.
The microprecipitation was demonstrated to be a specific technique, whilst the assignment of immtmotype using an algorithm, based on the reactivity of the 14 MAbs, gives similar results to the microprecipitation technique.
Other papers found in the literature were rejected based on absence of clear demonstration of the specificity of inununotyping methods.
Results Immunotvve of invasive MenA strains The following table summarizes the data from the first four articles.
RwinziiiiiVievoing:P.,.41.f _4- = --'7:4,44.A..1 W. ''''.:' µ..'-ft,,,,.. '''.
µ.- ,'' '' . c 712': % .. .. 1"..d- '1'''' ---r".41 . ,A.?' a China ! 2 7 6 0 0 1 others 4 2 1 1 0 1 b Sudan 1 18 I 2 0 4 0 Sweden 9 5 _ 9 0 2 0 = _ c Africa Subgroup lit 29 122 49 0 9 13 Clone IV-1 125 0 0 0 0 8 _ d Germany 0 9 1 . 0 0 0 Africa 7 5 0 0 0 0 Finland 0 0 4 0 1 1 Brazil 0 6 0 0 0 0 North America 2 24 2 1 0 2 Pe rcent I
'LA
7 ' v.- ARSAaritsVtd*Vi.Wat- .ltit*Y'Mtfs,ragfvgtMEVZOYtMleeN;*VT:(rr-Attt'IE4tiAtViNiAniEWP31:+rtaiCqtXigitttl':"-ZV:t age with clone TV-1 35.9 1 40.0 15.1 0.4 3.4 5_2 ''' C strain isolated from carrier, 1) strain isolated from disease a: Kim J et al (1988); b: Salih M et al (1990); c: Achtman M et al (1992); d:
Zollinger Wand Mandrel!
It (1980) Not taking into account clone IV-1 strains, the results show that most of invasive MenA
strains are L10 (54.4%) followed by Ll 1 immunotype (20.5%) and L9 immunotype (14.8%).
It is of note that most of the MenA strains displaying the iinmtmotype 9 harbor also other immunotypes (most generally L3 and/or L7).

These data (from invasive MenA strains) correlate well with data obtained by tricroprecipitation (and/or whole cell ELISA with a panel of 14 MAbs) (Poolman et al, 1982 and Scholten et al, 1994).
Ori = in CID* L379 L10 LI1 Other NI) diverse D 5 11 1 9 0 It can be concluded that most of invasive MenA strains (not including the strains belonging to clone IV-1) are L10. Others are either L9(L379) or L11.
fromunotype of carriage MenA strains Ref Ori. in L9 L10 L11 or L10,11 Other ND
d North America 1 2 10 0 2 e Diverse 0 4 1 2 0 OZEcaSAIEURAMAr.iiteWPWAsiEiNsalift*-ftsfstaf*Tter4RiktinOMf "
1.7)**MfakftegrAgM100.
Percentage _ 4.5 _ 27.3 50.0 9.1 9.1 e: Poolman et al (1984) Carriage strains isolated in North America are mainly L10,11 immunotype.
Nevertheless, as the anti-L11 rabbit polyclonal used for this typing cross-reacts with L 10, it is not possible to conclude if these strains display or not the LEO immunotype.
Interestingly, the limited data obtained from carriage MenA strains seems to indicate that Manunotype distribution among these strains is quite different to the distribution observed for invasive cases. Indeed, if most of invasive strains are Lb. the majority of carriage strains express at least the LI1 immunotype.
Conclusion and discussion The classical MenA SBA developed by the CDC used the F8238 strain as target.
We now know this strain expresses a L11 LOS (as immunotyped by Hopman C, AMC lab, Netherlands) and not a L10 as had been previously thought. In other words, the reference MenA strain used for the SBA is more closely related to carriage MenA
strains than to invasive MenA strains. This relationship can explain the lack of concordance observed between anti-PSA ELISA and "classical" MenA SBA in the non-primed toddler population (Example 3).
This observation reinforces the hypothesis that the lack of concordance observed between anti-PSA ELISA and F8238 MenA SBA is linked to a natural immunity acquired by the carriage of non-invasive N meningitidis strains, where immunity is not directed against the polysaccharide (or at least not high avidity anti-PS
antibodies) but probably against at least LOS antigens from carrier isolates.
References = Achtman M, Kusecek B, Morelli G, Eicicmann K, Jianfu W, Crowe B, Wall R, Hassan-King M, Moore P and Zollinger W (1992) A comparison of the variable antigens expressed by clone IV-1 and subgroup III of Neisseria meningitidis serogroup A. J. Infect. Dis. 165:
53-68.
= Kim J., Mandrel! M, Zhen H, Westerink M, Poolman J and McLeod Griffis J
(1988) Electrommphic characterization and description of conserved epitopes of the lipooligosaccharides of groupA Neisseria meningitidis. Infect. Imrnun. 56: 2631-2638.
= Poolman J, Hopman C and Z,anen H (1982) Problems in the definition of meningococcal serotypes.
FEMS Microbiol Letters 13: 339-348 = Salih M, Danielsson D, Baclanan A, Caugant D, Achtman M and Olcen P
(1990) Characterization of epidemic and non-epidemic Neisseria meningitidis serogroup A strains from Sudan and Sweden.
J. Clin. Microbio1.28: 1711-1719.
= Scholten R, Kuipers E, Valkenburg II, Dardcert J, Zollinger W and Poolman J (1994) Lipooligosaccharide inimunotyping of Neisseria meningitidis by a whole-cell ELISA with monoclonal antibodies. J. Med. Microbiol. 41: 236-243.
= Zollinger W and Mandrel! R (1980) Type-specific antigens of group A
Neisseria meningitidis:
lipopolysaccharide and heat-modifiable outer membrane proteins. Infect. Immun.
28: 451-458.
Example 6: Specificity of the CDC F8238 strain vs. strain 3125 in an SBA assay Aim of the experiment:
The impact of using N. meningitidis A strains with different immunotypes (here respectively the F8238 strain with L11 immunotype vs the 3125 strain with L10 immunotype) on the natural immunity observed with the MenA rSBA assay has been checked. It is to be noted that natural immunity was observed to a much lesser extent with the MenA (anti-MenA capsular polysaccharide) IgG ELISA assay. The latter assay is thus capable to differentiate between vaccine induced and naturally induced immunity.
Selected samples:
= A first set of samples was selected among studies which evaluated the immunogenicity of a booster dose of a Hib Vaccine (either conjugated or unconjugated) to children primed with pertussis (either acellular or whole cell) vaccines coadministred with Hib. In a first study performed in Germany (Hib-044) toddlers were aged between 12-18 mths and primed with different formulations of a whole cell pertussis vaccine coadministred with Hib. In a second study performed in Myanmar (Hib-064) toddlers were aged between 15-24 mths and primed with different formulations of a whole cell pertussis vaccine coadministred with Hib.
= A second set of samples was selected among a study (HAV-210) which evaluated the immunogenicity of a Hepatitis A vaccine administered to children aged between 11 to 25 Mths.
Since the purpose of these studies was not to study the effect of a menA
vaccination, it could therefore be anticipated that no MenA vaccine was administered to these children (they were nave with regards to MenA vaccination). A number of samples with sufficient remaining volume to perform both MenA IgG ELISA and the two MenA rSBA assays (With L10 and Ll 1 as assay strain) were selected at random.

Since the purpose of these studies were not to evaluate the menA groups it was not checked to exactly which groups these subjects belonged.
= A third set of samples was selected among studies which evaluated different menA vaccinations. In a first study performed in the Philippines (DTPw-HepB/HibMenAC-TT-002) children aged approx. 10 months and 15-18 months and not previously primed with a DTPw-HepB/HibMenAC vaccine were selected.
In a second study (MenACWY-T1'-003) adults aged 18-25 yrs not previously vaccinated with a MenA vaccine were selected.
Results:
As clearly seen in the table below the % of subjects with a titre >= 0.30 pg,/mL with the menA IgG ELISA is very low, except for adults. The same is seen in terms of subjects >= 8 with the MenA rSBA using the 3125 (L10) strain. The same effect is thus seen with both assays on these placebo samples i.e. they are both able to show a low response when no vaccine was administered. In contrast, performing the menA
rSBA assay using the F8238 (LI I) strain results in a higher % of responders (>=8) and as such does not reflect the fact that the subjects were not vaccinated.
Therefore it is concluded that a MenA rSBA using the 3125 (L10) strain is capable to better differentiate the vaccine induced response.
MenA IgG EUSA MenA 'SBA
% 0.30 pg/mL
Study Country Age Group n Subset F8238 (L11) 3125(110) 115-044 Germany 12-18 Mins No details (Possibly all) 68 39%
2%
Hib-064 Waltman' 15-2419th, No details (POssibly 60 6% 73%
0%
HAV-210 I US A 11-25 Mths I No delaVs (Possibly all) 1 92 I 0% I 30% 1 0%
OTPw-Herdi/HibMenAC-T1-002 PrOPOirles Aorx 10 Mats Trit-Hay r Hib 34 6% 68% 0%
DTPw-1iep8/HibMenACIT-002 Phi5PPInes Ayer 10 Mths Trit-HBV/Hib + Meningttee 39 5% 77% 21%
DTPw-Hepil/HibMenAC-11%002 Philippines 15-18 Mths Trit-HBV + 1-Irb , 29 , 10% 62% 17%
DTPw-Hep13/HibMenACIT-002 Ph-Wines _ 15-18 Mths Tril-HBV/Hib + 1.4eningitee 33 3% 52% 12%
MenACWY-1T-003 Belgium 18-25 Yrs Prlor meneeyas dose 21 62% 100% 76%
MenACWY-TT-003 Belgium 18-25 Yrs Prior MenACWV swinge* dose 22 . = 55% 100% 77%
Example 7: Sensitivity of the CDC F8238 strain vs. strain 3125 in an SBA assay Clinical trial samples were selected among studies which evaluated different menA
vaccinations. In a first study performed in the Philippines naive children aged approx.
14 weeks were primed with one of three different DTPw-HepB/HibMenAC vaccine formulations (where HibMenAC are all conjugated). Seroconversion (%SC) was measured as the % of children 1 month post priming with an rSBA (50%) titre more than or equal to 8.

rSBA GMTs rSBA % SC
rSBA strain F8238 3125 F8238 3125 Vaccine 1 315 160 94 88 Vaccine 2 400 229 100 100 Vaccine 3 284 156 100 100 Clinical trial samples were also selected among studies which evaluated different menACYW vaccinations. In a study performed in the Belgium naive adults (not previously immunized with a menA vaccine) aged 18-25 years were primed with either a unconjugated menACWY capsular polysaccharide vaccine (Mencevaxml ¨
GSICBiologicals s.a.) or a conjugated menACWY vaccine. Seroconversion (%SC) was measured as the % of adults 1 month post immunisation with a 4 fold increase in rSBA (50%) titre compared with just prior to immunisation.
rSBA GMTs rSBA % Sc rSBA strain F8238 3125 F8238 3125 Menceva.xTM 6776 2875 88 100 Conjugate 4626 2121 53 89 The data show that the 3125 strain is sensitive to vaccine induced antibodies from either conjugated or unconjugated menA capsular polysaccharide vaccines. In addition it is possible (for instance in older age groups beyond 3 years of age) that the %SC
data produced with the 3125 strain may be more accurate than with the F8238 strain.
=

Deposited Materials The N nzeningitidir serogroup A strain F8238 is available from the Centers for Disease Control and Prevention (CDC), Atlanta, Georgia, USA. Alternatively it has also been deposited at the Belgian Coordinated Collections of Microorganisms (BCCM), Laboratorium voor Microbiologie-Bacterienverzameling (BCCM/LMG), Universiteit Gent, K.L. Ledeganckstraat 35, 9000 GHENT, Belgium (deposited 20th June 2005 with a statement that it is equivalent to the CDC MenA F8238 strain). It has the Accession Number LMG P-23098.
The N. meningitidis serogroup A strain 3125 is available from the Academic Medical Center (AMC), University of Amsterdam, Netherlands Reference Laboratory for Bacterial Meningitis, Dept. of Medical Microbiology, Meibergdreef 15, 1105 AZ Amsterdam, The IS Netherlands. Alternatively it has also been deposited at the Belgian Coordinated Collections of Microorganisms (BCCM), Laboratorium voor Microbiologie-Bacterienverzanaeling (BCCM/LMG), Universiteit Gent, K.L. Ledeganckstraat 35, GHENT, Belgium (deposited 20th June 2005 with a statement that it is equivalent to the AMC MenA 3125 strain). It has the Accession Number LMG P-23097.
The N meningitidis serogroup A strain 3048 (A:NT:P1.6:L3,6) is available from the Academic Medical Center (AMC), University of Amsterdam, Netherlands Reference Laboratory for Bacterial Meningitis, Dept. of Medical Microbiology, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands. Alternatively it has also been deposited at the European Collection of Cell Cultures (ECACC), Health Protection Agency, Porton Down, Salisbury, SP4 OJG, UK (Accession Number: 04070702). The strain was deposited on 21't June 2004.

The N. meningitidis serogroup W strain S4383 is available from the Centers for Disease Control and Prevention (CDC), Atlanta, Georgia, USA. Alternatively it has also been deposited at the Belgian Coordinated Collections of Microorganisms (BCCM), Laboratorium voor Microbiologie-Bacterianverzameling (BCCM/LMG), Universiteit Gent, K.L. Ledeganckstraat 35, 9000 GHENT, Belgium (deposited 20" June 2005 with a statement that it is equivalent to the CDC MenW S4383 strain). It has the Accession Number LMG P-23099 The N meningitidis serogroup W strain 3193 (W135:2A:P1.5,2:L3) is available from the Academic Medical Center (AMC), University of Amsterdam, Netherlands Reference Laboratory for Bacterial Meningitis, Dept. of Medical Microbiology, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands. Alternatively it has also been deposited at the European Collection of Cell Cultures (ECACC), Health Protection Agency, Porton Down, Salisbury, SP4 OJG, UK (Accession Number: 04070701). The strain was deposited on 21st June 2004.
The deposit of the deposited stain has been made under the terms of the Budapest Treaty on the International Recognition of the Deposit of Micro-organisms for Purposes of Patent Procedure. The deposited strain will be irrevocably and without restriction or condition released to the public upon the issuance of a patent. The deposited strain is provided merely as convenience to those of skill in the art and is not an admission that a deposit is required for enablement, such as that required under 35 U.S.C. 112: A license may be required to make, use or sell the deposited strain and no such license is hereby granted.

Claims (43)

1. A N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay comprising the step of determining the SBA titre of a serum sample against, respectively, a N. meningitidis serogroup A or serotype W reference strain having the following characteristics:
a) it has a % complement toxicity of less than 40%, 35, 30, 20, 10, 5%, or is around 0%, and/or that its % complement toxicity is less than that of the MenW S4383 standard strain;
b) the % seropositivity against the strain in a group of at least 25 naïve

2 year old children where seropositivity is measured by an rSBA assay with titres of 8 or above, to achieve 50% cell killing, is less than 40, 35, 30, 25, 20 or 15%, and/or the % seropositivity against the strain in a group of at least naïve 1-2 year old children where seropositivity is measured by an rSBA
assay with titres of 8 or above, to achieve 50% cell killing, is less than that of the % seropositivity against the F8238 standard strain; and c) the % seroconversion against the strain in a group of at least 25 3-5 year old children immunized with a unconjugated N. meningitidis ACWY
capsular polysaccharide vaccine is more than 50, 60, 70, 80, 90 or 95%, where seroconversion is measured by assessing the % of subjects with at least a 4 fold increase in rSBA titre, to achieve 50% cell killing, when sera is compared just before immunization (when they are naïve) and one month after immunization with the vaccine.
2. The N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of claim 1, wherein in characteristic a), b), c), a) and b), a) and c), b) and c), or a), b) and c) the % complement toxicity and/or the % seropositivity and/or the % seroconversion measures, respectively, are carried out as described in Example 2b, 1b and 1c, respectively.

3. The N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of claim I or 2, wherein in characteristic b) the at least 25 naïve 1-2 year old children are native to the Philippines.

4. The N meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1-3, wherein the N. meningitidis serogroup A
reference strain is an invasive strain.

5. The N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1-4, wherein the N. meningitidis serogroup A
reference strain is predominantly of immunotype LIO.

6. The N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1-5, wherein the N meningitidis serogroup A
reference strain is strain 3125.

7. The N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1-4, wherein the N. meningitidis serogroup A
reference strain is predominantly of immunotype L3,7,9.

8. The N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1-4 and 7, wherein the N. meningitidis serogroup A
reference strain is strain 3048.

9. The N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1-3, wherein the N meningitidis serotype W
reference strain is isolate 3193.

10. A N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1-9, wherein the serum sample is from a host immunized with a vaccine composition comprising a capsular saccharide antigen from N. meningitidis serogroup A and/or serotype W.

11. The N meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of claim 10, wherein the capsular saccharide antigen is conjugated to a carrier protein.

12. A N. meningitidis serogroup A or serotype W Serum Bactericidal Activity (SBA) assay of any one of claims 1- 11, wherein the serum sample has been taken from a human aged 0-55 years, 1 month - 20 years, 2 months - 16 years, 6 months -years, 8 months - 7 years, 10 months - 4 years, 0-2 years, 6 months - 2 years, or 1-2 years.

13. A kit for carrying out the SBA assays of any one of claims 1-12 comprising the N. meningitidis serogroup A and/or serotype W reference strain(s) of any one of claims 1-12 and instructions for its/their use in an SBA assay.

14. The use of an invasive L10 N. meningitidis serogroup A strain in a SBA
assay.

15. The use of a N. meningitidis serogroup A isolate 3125 in a SBA assay.

16. A N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay comprising the step of determining the SBA titre of a serum sample against a N.
meningitidis serogroup A reference strain having the following characteristic:
b) the % seropositivity against the strain in a group of at least 25 naïve 1 -2 year old children where seropositivity is measured by an rSBA assay with titres of 8 or above, to achieve 50% cell killing, is less than 40, 35, 30, 25, 20 or 15%, or the % seropositivity against the strain in a group of at least 25 naïve 1-2 year old children where seropositivity is measured by an rSBA
assay with titres of 8 or above, to achieve 50% cell killing, is less than that of the % seropositivity against the F8238 standard strain.

17. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of claim 16, wherein in characteristic b) the % seropositivity measure is carried out as described in Example 1b.

18. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of claim 16 or 17, wherein in characteristic b) the at least 25 naïve 1-2 year old children are native to the Philippines.

19. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of any one of claims 16-18, wherein the N. meningitidis serogroup A reference strain is an invasive strain.

20. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of any one of claims 16-19, wherein the N. meningitidis serogroup A reference strain is predominantly of immunotype L10.

21. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of any one of claims 16-20, wherein the N. meningitidis serogroup A reference strain is strain 3125.

22. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of any one of claims 16-19, wherein the N. meningitidis serogroup A reference strain is predominantly of immunotype L3,7,9.

23. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of any one of claims 16-19 or 22, wherein the N. meningitidis serogroup A
reference strain is strain 3048.

24. A N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of any one of claims 16-23, wherein the serum sample is from a host immunized with a vaccine composition comprising a capsular saccharide antigen from N.
meningitidis serogroup A.

25. The N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of claim 24, wherein the capsular saccharide antigen is conjugated to a carrier protein.

26. A N. meningitidis serogroup A Serum Bactericidal Activity (SBA) assay of any one of claims 16-25, wherein the serum sample has been taken from a human aged 0-55 years. 1 month - 20 years, 2 months - 16 years, 6 months - 10 years, months - 7 years, 10 months - 4 years, 0-2 years, 6 months - 2 years, or 1-2 years.

27. A kit for carrying out the SBA assays of any one of claims 16-26 comprising the N. meningitidis serogroup A reference strain of any one of claims 16-26 and instructions for its use in an SBA assay.

28. A Serum Bactericidal Activity (SBA) assay for a Gram negative bacterium comprising the step of determining the SBA titre of a serum sample against a Gram negative bacterium reference strain which is a LOS mutant either unable to synthesise LOS or which produces a truncated LOS oligosaccharide structure.

29. A N. meningitidis serotype W Serum Bactericidal Activity (SBA) assay comprising the step of determining the SBA titre of a serum sample against a N.
meningitidis serotype W reference strain, wherein the N. meningitidis serotype W
reference strain is not sensitive to complement dependent killing in the absence of serum.

30. A N. meningitidis serotype W Serum Bactericidal Activity (SBA) assay comprising the step of determining the SBA titre of a serum sample against a N.
meningitidis serotype W reference strain, wherein the N. meningitidis serotype W
reference strain is less sensitive to complement dependent killing in the absence of serum than the S4383 standard reference strain.

31. The SBA assay of claim 29 or 30, wherein the sensitivity of the N.
meningitidis serotype W reference strain to complement dependent killing is ascertainable by measurement of % complement toxicity by the method of Example 2b.

32. The SBA assay of any one of claims 29-31, wherein the % Complement Toxicity for the N. meningitidis serotype W reference strain is less than 40%, 35, 30, 20, 10, 5%, or is around 0%.

33. The SBA assay of any one of claims 29-32, wherein the N. meningitidis serotype W reference strain is N. meningitidis isolate 3193.

34. The SBA assay of any one of claims 29-33, wherein the serum sample is from a host immunized with a vaccine composition comprising a capsular saccharide antigen from N. meningitidis serotype W.

35. The SBA assay of claim 34, wherein the capsular saccharide antigen is conjugated to a carrier protein.

36. The SBA assay of claim 28, further comprising the steps of the SBA
assay of any one of claims 1-11, 16-25, and 29-35.

37. The SBA assay of claim 36, wherein the reference strain is a N.
meningitidis LOS mutant either unable to synthesise LOS or which produces a truncated LOS
oligosaccharide structure.

38. A kit for carrying out the SBA assay of any one of claims 28-37 comprising a reference strain which is a LOS mutant either unable to synthesise LOS or which produces a truncated LOS oligosaccharide structure, and instructions for its use in an SBA assay.

39. A kit for carrying out the SBA assay of any one of claims 29-34 comprising the N. meningitidis serotype W reference strain of any one of claims 29-34, and instructions for its use in an SBA assay.

40. The use of N. meningitidis isolate 3193 in a SBA assay.

41. The use of N. meningitidis isolate 3048 in a SBA assay.

42. The SBA assay of any one of claims 1-11, 16-25 and 28-37 which is an hSBA
assay.

43. The SBA assay of any one of claims 1-11, 16-25 and 28-37 which is an rSBA
assay.