CA2521812A1 - Optimal polyvalent vaccine for cancer - Google Patents

Abstract

This invention also provides a method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of: a) selection of an appropriate cancer cell line; and b) detection of the expression of antigens on the surface of said cell line of the cancer, wherein the antigens expressed will be used in the polyvalent vaccine. This invention also provides a method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of: a) selection of an appropriate cancer cell line and b) detection of the immunogenicity will be used in the polyvalent vaccine. This invention provides various uses of the identified polyvalent vaccine.

Description

OPTIMAL POLYVALENT VACCINE FOR CANCER
This application claims benefit of U.S. Serial No.
60/461,622, filed April 09, 2003, the content of which is incorporated by reference here into this application.
This application was supported in part by NIH Grant No.
P01CA33049. Accordingly, the United States Government may have certain rights in this invention.
Throughout this invention, various references are cited.
Disclosures of these publications in their entireties are hereby incorporated by reference into this application to more fully describe the state of the art to which this invention pertains.
BACKGROUND OF THE INVENTION
Small cell lung cancer (SCLC) biopsy specimens previously have been screened with monoclonal antibodies (mAb) against thirty potential target antigens to identify those that are most widely expressed, ie. on >50% of cancer cells in >60 0 of biopsy specimens (1-3). The glycolipids GM2, fucosyl GM1, sLea and globo H, and polysialic acid (polySA) on embryonal NCAM filled these criteria. Two additional glycolipids, GD2 and GD3, have been described by others to also be prevalent on SCLC (4,5) and a multicenter randomized Phase 3 trial with an anti-idiotype vaccine targeting GD3 is currently in progress. These are all cell surface antigens that were demonstrated to be consistently immunogenic in patients when conjugated to Keyhole Limpet Hemocyanin (KLH) and mixed with immunological adjuvant QS
21 (6,7,8,9,10) (excepting sialyl Lewisa (sLea) which has not been tested). They are all excellent candidates for inclusion in a polyvalent, antibody-inducing vaccine against SCLC.

GM2, Fucosyl GM1, Globo H and polySA were the most widespread of the SCLC cell surface antigens in our initial screen using immunohistochemistry with biopsy specimens.
These four antigens were the first choices for incorporation into a polyvalent vaccine against SCLC cell surface. Prior to preparing this tetravalent conjugate vaccine, there was a wish to confirm that mixtures of antibodies against these antigens result in stronger cell surface reactivity than any individual antibodies and to l0 determine whether inclusion of additional antigens would yield higher cell-surface reactivity against SCLC. There were two relevant concerns. First that the SCLC cell lines would prove resistant to complement activation and complement dependent cytotoxicity (CDC), suggesting SCLC in patients would be resistant to complement targeting and cytotoxicity. Second, that antibodies against polySA which we know to be poor target for CDC as a consequence of the great distance it extends from the cell surface (17), would block CDC mediated by mAbs against other antigens. Toward this end 10 SCLC cell lines were tested by flow cytometry and complement dependent cytoxicity (CDC), with monoclonal antibodies against these seven target antigens individually or pooled in different combinations. Slight suboptimal mAb concentrations which demonstrated levels of reactivity by fluorescence activated cell sorter assay (FRCS) comparable to those seen with sera from patients vaccinated with these same antigens were used.

SZJN.~2ARY OF THE INVENTION
The invention disclosed herein provides a general methodology to determine the optimal combination of a single polyvalent vaccine against different cancers. This invention provides a system which would identify the optimal combination.
This invention also provides a method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of: a) selection of a cancer cell line; and b) detection of the expression of antigens on the surface of said cell line of the cancer, wherein the antigens expressed will be used in the polyvalent vaccine.
This invention further provides a method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of: a) selection of an appropriate cancer cell line and b) detection of the immunogenicity of antigens on the surface of said cell line, wherein the antigens showing said immunogenicity will be used in the polyvalent vaccine.
Finally, this invention provides an optimal combination of a polyvalent vaccine against cancer. In an embodiment this invention provides a tetravalent vaccine for small cell lung cancer comprising GM2, Fucosyl GM1,~ Globo H and polysialic acid.

DETAILED DESCRIPTION OF THE FIGURES
Figure 1. Glycolipid and glycoprotein antigens expressed at tr SCLC cell surface.
Figure 2. IgM FACS results against 10 SCLC cell lines after mixture with the 4 mAb pool containing PGNX (GM2), F12 (fucosyl GM1) , VK9 (globo H) and 5A5 (polysialic acid) .
Figure 3. .Anti-CD59 mAb greatly increases Pool 2 mediated CDC
of H345.

DETAILED DESCRIPTION OF THE INVENTION
The invention disclosed herein provides a general methodology to determine the optimal combination of antigens for polyvalent vaccines against different cancers.
In the literature, many antigens have been described as being expressed on the surface of cancerous cells. In designing which antigens should be used for vaccine, this invention provides a system which would identify the optimal combination.
This invention also provides a method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of : a) selection of an appropriate cancer cell line; and b) detection of the expression of antigens on the surface of said cell line of the cancer, wherein the antigens expressed will be used in the polyvalent vaccine.
International Patent Application No. PCT/US02/21348 (International Publication No. WO 03/003985 A2, January 16, 2003) discloses a polyvalent vaccine comprising at least two conjugated antigens selected from a group containing glycolipid antigen, polysaccharide antigen, mucin antigen, glycosylated mucin antigen and an appropriate adjuvant.
PCT/US02/21348 also provides a multivalent vaccine comprising at least two of the following: glycosylated MUC-1-32mer, Globo H, GM2, Ley, Tn(c), sTN(c), and TF(c).
The current invention provides an in vitro system which predicts and optimizes the combination of said vaccine.
In an embodiment, more than one cancerous cell line is used for said identification of the optimal confirmation of a polyvalent vaccine. In another embodiment, the expression of the antigens is detected by specific antibody. In a further embodiment, the antibody is a monoclonal antibody.
In a separate embodiment, the expression is detected by Fluorescence Activated Cell Sorter (FACS).
This invention further provides a method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of: a) selection of an appropriate cancer cell line and b) detection of the immunogenicity of antigens on the surface of said cell line, wherein the antigens showing said immunogenicity will be used in the polyvalent vaccine.
As used herein, immunogenicity describes the quality of a substance which is able to provoke an immune response against the substance, a measure of how able the substance is at provoking an immune response against it. This response includes cell-mediated and humoral responses.
In an embodiment, the immunogenicity of antigens is determined by the Complement Dependent Cytoxicity assay. In another embodiment, the cancer is a small cell lung cancer.

This invention further provides the optimal combination identification by the above methods.
This invention also provides a polyvalent vaccine for small cell lung cancer comprising GM2, Fucosyl GM1, Globo H and polysialic acid and GD2 or GD3. In an embodiment, the vaccine further comprises sialyl Lewisa.
In an embodiment, the antigens are conjugated. In a further embodiment, the antigens are conjugated to Keyhole Limpet Hemocyanin.
In yet another embodiment, the above vaccine includes an appropriate adjuvant. The appropriate adjuvant should be able to booster the immunogenicity of the vaccine. In a further embodiment, the adjuvant is saponin-based adjuvant.
The saponin-based adjuvants include but are not limited to QS21 and GPI-0100.
The invention will be better understood by reference to the Experimental Details which follow, but those skilled in the art will readily appreciate that the specific experiments detailed are only illustrative, and are not meant to limit the invention as described herein, which is defined by the claims which follow thereafter.
Experimental Details Small cell lung cancer (SCLC) biopsy specimens previously have been screened with monoclonal antibodies against thirty potential target antigens to identify those that are most widely expressed, ie. on >50% of cancer cells in >60%
of biopsy specimens. The glycolipids GM2, fucosyl GM1, sLea, globo H, and polysialic acid on embryonal NCAM filled these criteria. Prior studies have been performed with GM2 (4, 12) , fucosyl GMl (4, 11, 12) and polySA (13, 14) . Two additional glycolipids, GD2 and GD3, have been described by others to also be prevalent on SCLC (4, 5), and a multicenter randomized Phase 3 trial with an anti-idiotype vaccine targeting GD3 (16) is currently in progress. These are all cell surface antigens that have been demonstrated to be consistently immunogenic in patients when conjugated to KLH and mixed with immunological adjuvant QS-21. They are all excellent candidates for inclusion in a polyvalent, antibody-inducing vaccine against SCLC.
GM2, Fucosyl GM1, Globo H and polySA were the most widespread of the SCLC cell surface antigens in our initial screen using immunohistochemistry with biopsy specimens.
These four antigens were the first choices for incorporation into a polyvalent vaccine against SCLC.
Prior to combining these four conjugates into a single polyvalent vaccine, there was a wish to confirm that mixtures of antibodies against these antigens results in stronger cell surface reactivity than any individual antibodies and whether inclusion of additional antigens would yield higher cell-surface reactivity against SCLC and to identify the optimal combination. There were two relevant concerns. First, that the SCLC cell lines would l0 prove resistant to complement activation and complement dependent cytotoxicity (CDC), suggesting SCLC in patients would be resistant to complement targeting and cytotoxicity. Second, that antibodies against polySA which are known to be a poor target for CDC as a consequence of the great distance it extends from the cell surface thereby blocking CDC mediated by mAbs against other antigens.
Toward this end ten SCLC cell lines were tested by flow cytometry with monoclonal antibodies against these seven target antigens individually and pooled in different combinations. Slightly suboptimal mAb concentrations which demonstrated levels of reactivity by FACS comparable to those seen with sera from patients vaccinated with these same antigens were used.
Cell lines: All cell lines were purchased from the American Type Culture Collection (ATCC) (Manassas, VA). The cell lines are listed in Tables 1 and 2. The origin of each is listed by the ATCC as SCLC, obtained from biopsy of lung nodules except for H82, H187 and H196 which originated from pleural effusions and H211 and H345 which originated from bone marrow biopsies. SHP77 is listed as large cell variant SCLC.
Monoclonal antibodies (mAbs): The target antigens for the seven mAbs, the source of the mAbs and the concentration used in the FRCS studies are described below.

GM2, mAb PGNX. Progenics Pharmaceuticals Inc. (Tarrytown, NY), ascites 0.5j~1/ml. Fucosyl GM1, mAb F12, Dr. Thomas Brezicka (Goteborg, Sweden), 0.1~,g/ml. Globo H, mAb VK9, Kenneth Lloyd (MSKCC), 20~.g/ml. Polysialic acid, mAb 5A5, Urs Rutishauser (MSKCC), ascites 0.1~.g/ml. GD2, mAb 3F8, Dr. Nai-Kong Cheung (MSKCC), 0.4~,g/ml. GD3, mAb R24, Dr.
Paul Chapman (MSKCC), 0.4~g/ml. sLea, mAb 19.9, purchased from Signet (Dedham, MA), supernatant 0.05~,1/ml. These mAbs, concentrations and mAb subclasses are listed in Table 1. The antigens recognized by these mAbs are shown in Figure 1.
Fluorescence Activated Cell Sorter (FACS) Assay: The ten SCLC cell lines served as targets. Single cell suspensions of 2 x 105 cells/tube were washed with 3 o fetal calf serum in PBS and incubated with 20 ~,l of 1:20 diluted mAb for 30 min on ice. After washing the cells twice with 3% FCS in PBS, 20 ~,l of 1:15 rabbit anti-human IgG or IgM-labeled with FITC was added. The suspension was mixed, incubated for 30 min and washed. The percent positive population and mean fluorescence intensity of stained cells were analyzed using a FACS Scan (Becton-Dickinson, CA) (8, 9) with percent positive cells for second antibody alone set at 1%.
Complement Dependent Cytotoxicity (CDC) and Antibody Dependent Cellular Cytotoxicity (ADCC): Complement dependent cytotoxicity was assayed on the ten cell lines using a 2-hour 51 chromium release assay as previously described (10) with human complement with MoAb at 10~,g/ml.
Approximately 10' cells were labeled with 100 ~,Ci of Nay 51Cr04 (New England Nuclear, Boston, MA) in 3 o HSA for 2h at 37°C, shaking every 15 min. The cells were washed four times and brought to a concentration of 106 live cells/ml.
Fifty ~,1 of labeled cells were mixed with 50 ~.1 of undiluted pre- or post-vaccination serum or with medium alone in 96-well, round-bottomed plates (Corning, New York, NY) and incubated at 4°C on a shaker for 45 min. Human complement (Sigma Diagnostics, St. Louis, MO) diluted 1:5 with 3% HSA was added, at 100 ~,1/well, and incubated at 37°C for 2h. The plates were spun at 1008 for 3 min, and an aliquot of 100 ~,1 of supernatant from each well was read by a gamma counter to determine the amount of 5lCr released.
All samples were carried out in triplicate and included control wells for maximum release and for spontaneous release in the absence of complement.
Spontaneous release (the amount released by target cells incubated with complement alone) was subtracted from both experimental and maximal release values. Maximum release was the amount of radioactivity released by target cells after a 2-hour incubation with 1% Triton X-100. Percent specific release was calculated as corrected experimental/corrected maximal release. Concentrations of anti-CD55 and anti-CD59 between 25 and 150 ~.g/ml were added to CDC assay wells with the mAbs or mAb pools to counteract inhibition mediated by CD55 and CD59. Mab clone BRIC 216 against CD55 and mAb MEM-43 against CD59 were purchased from Serotec Inc. (Raleigh, N.C.).
Cell surface reactivity demonstrated by FACS
Cell surface reactivity for the 7 monoclonal antibodies utilized at the concentrations summarized in Table 1 ranged from 1% to more than 90% in the 10 SCLC cell lines. Two of the mAbs (PGNX recognizing GM2 and 5A5 recognizing polySA) resulted in 50% or more positive cells in 6 of the 10 SCLC
cell lines. The other mAbs demonstrated comparable reactivity with 5 or fewer cell lines. On the other hand, when the antibodies were pooled in different combinations using the same concentration of antibody, 9 of 10 cell lines demonstrated 500 or greater positive cells.
Combinations containing mAbs against fucosyl GM1, GM2, globoH and polysialic acid (the four mAb pool) were optimal, the addition of antibodies against GD2, GD3 and sialyl LewisA had little additional impact. While some cell lines such as DMS79 and H187 were strongly positive with 6 of the 7 mAbs, others such as SHP77, H211 and H82 or H196 were positive with only 0, 1 or 2 of the mAbs.
However, when the antibodies were pooled in different combinations only SHP77 continued to express less than 50%
positive cells. Cell surface reactivity by FACS for the 10 cell lines with the 4 mAb pool is demonstrated in greater detail in Figure 2. With the exception of cell line SHP77, strong cell surface reactivity was demonstrated against all cell lines.

r.
o~, ~ ~ ~ o ' ~ o o s o ~. '~M N OOa '~~ s 6~1 \ o~o o Nb \ o 0~10~50~1o +r '~~ C"~ oo ~ V~Io'~oo 0~1 o O~O~N-' OmO~iO
rte.\ o o o \ o o N ~'. '~4~~tV'~ oo 'r'~ aoo_ ~
i~ o ,~-n 0~0 ~.I O M ~.C..~ r~,'~I'oOt O
~ .'-~ N 4101 4't ......' ...................."..._..to('1_.._........ _.....
o \ \ \ o SI' ~ o a' <J' o .......'~'~N N N ~ N O M
._.,._. M 4l ~' ~u~ ....__ '5y-o .,...... _._,..
o\ _......
~ __.._ ~ ooo M ~
N

~ o 0 0 0 0 \ 0 0 0 0 v 0 0 ~

,.~.1 ~ '~,-N-nN CR_ '~O~45 4h (3V
\ D~ OS
~l' .+ 0 0 0 \ 0 o ~ 0 0' ~r~ ~ ~ 0 ' 0 N ~ r31 ch CI'4't~ 00 Ch o ,--~M M '-'O m-nN OS 01 Ot ~ 01 ,-- DS

40 0 0 0 0 0 \ a o~ o o + 0 a r1 t0tDO O O 4hCh 01 l,T
~Ot0M SROv ~ cT~ 6v 6S ~ ~
Ot OS

......,._......_..._........_._.._......_........._..,.,._......,....._........
........_..__.._..,................_.,_._.._.._ ......,_..,. ............. _..
._._._.
..,._._....

0 0 0 0 0 0 0 0 0 ~
.'~...o o a o 0 ~ o 0 o o o~ 0 0 ~ 0 .-~00O M 00 ~ SI'O N Qt rx N ~ot~-- O o~.tf"~ c~-\ wn m vo M

r' 0 0 0 0 0 0 0 0 0 0 ~ o..\ ~ 0 \ \ 0 ~ o o ~OO N ~'41 r"01OS ~s t31 a1N r~V1lp m10101 L~
OL t31 01 Oa ~

A

4.1 .~ N
~,~ ~ ~ ~
ifs ~

~ ~ O 41 a7 ~ O ON O a _ O

G t.~' y H
~ ~
~

C~ .
-~., a w N

~
~ ~

~ O
N O ~ O
Y ~ O .fl .fl O

. Q7 ~ _Op _O
-n, N ~
O

l~~ ~

~1 N
V

+~ v (s.iM Rj~ ~ ~ N N

Pi w w w w w l II II

V N M N y~ DAp p O ~
.f7 ~ O O
O

w ~ ~ ~ w ~ w w w ~ w Cell surface reactivity demonstrated by CDC
Complement dependent cytotoxicity (CDC) assays using human complement demonstrated 300 or greater lysis in 5 of the 10 cell lines with PGNX against GM2, in 3-4 of the 10 cell lines with mAbs against fucosylated GM1, GD2 and GD3, and none of the cell lines with mAb against polysialic acid, globoH and sialyl LeA (Table 2). The 4 antibody pool including fucosylated GM1, GM2, globoH and polysialic acid resulted in greater than 30% cytotoxicity for 9 of the 10 cell lines. This was increased slightly by the addition of antibodies against GD2 and GD3 but still one cell line, H345, had less than 30% cytotoxicity despite the fact that 99% of the H345 cells had strong reactivity by FACS with the same pools. Aside from H345, FRCS and CDC correlated fairly closely, with some such as HSP77 and H211 demonstrating stronger than expected CDC.

'.oc o c oc '~
~

Sl'M O ,NM O'M N MM M M

o o o o o'~ o o o o ' ' ., o' ' , , ~
.

r5yN ~.,~r,r N O . 01Y1 M N ,1V1~ .-~ V1f1 00QS
4~V1 r V1 a , H n~ N o'N,,ac c c \ ~~;oc c c o o ~ -.o ~ ~
' ~
M

a ~ ~ ~ V N Vt.. ~ . O 0 ~0 O
O

~ .
' a ~ C~ ~~\~'1?~' ~ ~. ~.o o .c o c H ~ o _.. ..~... ~..............~.....:

c ' o o c 0 0 ~ ~ N 0 o .
0 ' rte. O ~ ,p~' 00N .
w-, ~ ~,-0 u-, ~ b ~ ~
~r :

o o ': oo 0 0 o .

~,_a . ~;.M'.~, a ~

~ o o o~'o c a- o~~ N'c ', ~.
~
, Cd ,~oN ,~ .~-co'o,-. o vn.sy-~_. ~_ ~...:
a ~ ~ gypM ~ M Q N~O ~tb'= 4'1V1 M t 41 a .

-..,~~
~ :~1 o 'o.o o 00 0 0 N r~
~

" a mR ovco m ~- o0 N .

~o~b'~

;~ a i d ~ ~ o U-,u-, U

'~..~ ~~CV, CV.

'~, m' i o O
a s~ ~,. . - pi.
a M

~ ~ R
~

~ ~

N

C
~~J

.

~ N ~ O O p _O
' .~

-nN, .'~

,~ TN-'WI~N1~~ ~.01~~'~~
M ~ ~ : 'N
-~ ~ C7 ~ C7 G 41 r-n ~

~ w ~ w w w a a n d ~ ~ ~a r1N 't1' M

N ~ .-m-.m--n~
H ~ ~
~

~ w A ~'c w ~" w ww w 7 ' . , v Cell surface expression of CD55 and CD59 CD55 was strongly expressed on 3 of the 10 cell lines (SHP77, H524 and H196) and CD59 was strongly expressed on all cell lines except H211 and H82. There was no clear correlation between expression of these 2 complement resistance factors and the level of complement dependent cytotoxicity (Table 3). H345 was one of the many strongly CD59 positive cell lines but was only moderately positive for CD55. H345 may have been negative by CDC because the predominate antigen recognized by these mAbs at the cell surface is polysialic acid. Nevertheless, to explore the role of CD55 and CD59 in complement lysis against this apparently complement resistant cell line the CDC assay in the presence of anti-CD55 or anti-CD59 mAbs was performed (see Table 3). Neither anti-CD55 nor anti-CD59, (nor the two in combination), were able to mediate detectable complement cytotoxicity on their own against H345.
However, CDC mediated by the four mAb pool increased from 5% to 15o in the presence of 100~,g per ml of anti-CD55, and 70% in the presence of 100~,g per ml of anti-CD59 (Figure 3) .

d v W \ m o o O~ ~ 01 O1 M M
O

N

O \ \ 0 0 ~ ~ M 00 ~ ~

U

H d N

U ~ ~ ~ s s o ~ o ~ ~ 0 0 A

U

~ a~

N
U ~ ~ ~ ~ o o w ~

_ N o o \ p \ \ \ o C/~\ v0 O O ue' a o0 .-r M .d O

a a ~ o U ~ o ''H

~
Z \ \

V o ~ o o o m o , ~

M ~ 01 O1 01 ,-.a M

z ~ ~ ~ o y o 0 0 ~
~

z o ~ ~ o , M ~ 01 01 d' H l~

M
N

ue d ' A

U

x ~ s A

U
~

w ~ o O rig~

z ' ~ ~
A ~
\

O \ ~ o~ ov a m v ~ ~

H a\ a1 a1 a\

d a U ~

o o O ''_' ''' V o o p, M a .yn U U U U
,- ,-H ~~ ~~ wa w qw ~

CONCLUSION
Biopsies of SCLC demonstrate a rich array of cell carbohydrate surface antigens. Fucosyl GM1, GM2, polysialic acid, globo H, sialyl Lea, GD2 and GD3 are the most widely expressed of these. These are each excellent targets for active or passive antibody mediated immunotherapy of SCLC, but no one of these antigens has been shown to be expressed on more than 70 or 800 of SCLC biopsy specimens. This is the basis for the focus on constructing a polyvalent vaccine against several of these antigens, but which ones?
Are all required? Will antibodies induced against some of these antigens detract from the impact of antibodies induced against other antigens? Would a pool of antibodies be able to target all SCLCs. These questions were addressed by testing the reactivity of mAbs against these 7 different cell surface antigens on a panel of 10 SCLC cell lines using flow cytometry. The concentrations of the mAbs used was selected to give ELISA and FRCS titers of reactivity comparable to those achieved in patients receiving KLH conjugate vaccines against these antigens (6-10, 17, 26). The four antigens recognized most widely by these mAbs on biopsy specimens, and now these ten cell lines, were fucosylated GM1, GM2, globoH and polysialic acid. The number of cell lines demonstrating 50% or more positive cells by FACE increased to 9 of the 10 when this pool of antibodies was utilized and the remaining cell line was positive as well, demonstrating 30% positive cells.
The addition of antibodies against GD2, GD3 and sialyl Lea had little additional impact. Based on these results we selected fucosylated GM1, GM2, globoH and polysialic acid for inclusion in the polyvalent vaccine that we are in the process of constructing against SCLC. However in the previous clinical trials with polySA-KLH conjugate vaccines, antibodies against polysialic acid were unable to mediate CDC). Consequently it was our wish to confirm the selection of these 4 target antigens using a complement dependent cytotoxicity assay to be sure that antibody against polySA would not interfere with CDC mediated by antibodies against the other 3 antigens.
Several mechanisms for cancer cells to evade complement dependent cytotoxicity have been described (20 - 22).
CD55 which interferes at the level of C3 convertase and CD59 which interferes with assembly of the membrane attack complex are the most widely studied of the complement activation resistance factors. It has been reported that tumor cells can avoid CDC in the face of potent FRCS
reactivity at the cell surface when the antigens are on elongated molecules such as mucins. This was initially detected with monoclonal antibodies and vaccine induced antibodies against MUC1 (23) but more recently also against polysialic acid (17). CDC resistance was assumed to be a consequence of the great distance from the cell surface that complement activation occurs, similar to the resistance to CDC described against Salmonella minnesota, Salmonella monte video, sudamina aeruginosa and other "smooth" bacterial strains with long lipopolysaccharide chains (18, 19). Complement activation initiates a cascade of enzyme activities resulting in binding of C3b and eventually insertion of the C5b-9 protein complement membrane attack complex (MAC) into cell membranes to form pores. Dimensions of the MAC are 100 by 150 angstroms (25). The molecular weight of the NCAM C-terminal extracellular subunit and flanking sequence are in excess of 100KD (24), making it likely that the polysialic acid portion begins 100 angstroms or more from the cell membrane. If complement activation occurs at sites more distant than 100 angstroms from the cell membrane, the membrane attack complex would not form or if formed would not reach the cell membrane and a number of serum proteins would quickly inactivate the forming membrane attack complex (25). Even in this case, C3 mediated inflammation and opsonization would remain in place.
In the study presented here, mAb5A5 again proved to be a highly reactive IgM antibody, resulting in potent cell surface reactivity by FACS against 6 of the 10 SCLC cell lines, but it was unable to mediate complement cytotoxicity against any cell line. This is consistent with our previous finding with sera from SCLC patients after vaccination (17). When mAb5A5 was added to pools of other monoclonal antibodies, however, no diminution in CDC was detected, demonstrating that there was no steric or other hindrance to CDC mediated by antibodies binding to antigens that are more intimately associated with the cell surface lipid bilayer. Overall, the CDC assay gave results which were quite similar to those obtained with FACS. The number of cell lines demonstrating more than 30% cytotoxicity with any one mAb increased from 0-5 cell lines with single mAbs to 9 of the 10 cell lines with the pools of antibodies.
Most cancers of the colon and stomach are known to express CD55 but CD55 has not been found on either of the two SCLC
biopsies described to date (27, 28) and was seen in 0/4 (29) or 29% (30) of SCLC cell lines, consistent with our findings of strong CD55 expression in 3 of 10 SCLC cell lines. CD55 was only minimally expressed in the one SCLC
that was resistant to CDC. It seems unlikely, therefore, that CD55 mediated CDC resistance will be a major problem in the SCLC patients that we plan to immunize, but further studies with primary and metastatic SCLC biopsy specimens are indicated to confirm this. CD59 was strongly expressed on 8 of the 10 cell lines, but there was no clear correlation between this expression and CDC. While cell line H345 (which expressed CD55 weakly and CD59 strongly) had 12% peak CDC, it demonstrated 99o positive cells by FACS. In the presence of inhibiting levels of mAbs against CD59, however, CDC increased from 12% to 72%. This demonstrates complement activation by the 4 mAb pool which was being inhibited only at the membrane attack complex level by CD59. This strongly suggests that with the four antibody pool, all 10 SCLC cell lines tested, including even H345, should be good targets for antibody mediated effector mechanisms such as those mediated by complement activation (inflammation and opsonisation). These results demonstrate that vaccines containing the four cell surface antigens fucosylated GM1, GM2, globo H and polysialic acid should be sufficient for inducing antibodies against the great majority of SCLCs and that these antibodies should be able to mediate tumor cell destruction.

REFERENCES
1. Zhang, S., Cordon-Cardo, Zhang, H.S., Reuter, V.E.
Adluri, S., Hamilton, W.B., Lloyd, K.W., and Livingston, P.O. Selection of carbohydrate tumor antigens as targets for immune attack using immunohistochemistry. I. Focus on Gangliosides. Int. J. Cancer 73:42-49, 1997.
2. Zhang, S., Zhang, H.S., Cordon-Cardo, C., Reuter, V.E., Singhal, A.K., Lloyd, K.O., Livingston, P.O. Selection of tumor antigens as targets for immune attack using l0 immunohistochemistry. II. Blood group-related antigens.
Int. J. Cancer 73: 50-56, 1997.
3.. Zhang, S., Zhang, H.S., Cordon-Cardo, C., Ragupathi, G.
and Livingston, P.O. Selection of tumor antigens aw targets for immune attack using immunohistochemistry: III
protein antigens. Clin. Cancer Res. 4: 2669-2676, 1998.
4. Brezicka, T., Bergman, B., Olling, S., Fredman, P.
Reacitivity of monoclonal antiboides with ganglioside antigens in human small cell lung cancer tissues. Lung Can 28: 29-36, 2000.
5. Cheresh, D., Rosenberg, J., Mujoo, K., Hirschowitz, L., Reisfeld, R.A. Disialoganglioside Gd3 on human melanoma serves as a relevant target antigen on small cell lung carcinoma for monoclonal antibody-medicated cytolysis. Can Res 46: 5112-5118, 1986.
6, helling, F., Zhang, A., Shang, A., Adluri, s., Calves, M., Koganty, r., Longenecker, b.M., Oettgen, H.F. and Livingston P.O. GM2-KLH conjugate vaccine: Increased immunogenicity in melanoma patients after administration with immunological adjuvant QS-21. Cancer Res. 55:2783 2788, 1995.
7. Ragupathi, G., Slovin, S., Adluri, S., Sames, D., Kim, I-J., Kim, H.M., Spassova, M., bornmann, W.G., Lloyd, K., Scher, H.I., Livingston, P.O. and Danishefsky, S.J. A

fully synthetic globo H carbohydrate vaccine induces a focused humoral response in prostate cancer patients: A
proof of Principle. Angewandte. Chemie 38: 563-566, 1999.
8. Dickler, M.N., Ragupathi, G., Liu, N.X, Musselli, C., Martino, D.J., Miller, V.A., Kris M.G., Brezicka, F.T., Livingston, P.O., and Grant S.C. Immunogenicity of the fucosyl-GM1-keyhole limpet hemocyanin (KLH) conjugate vaccine in patients with small cell lung cancer. Can Res 5: 2773-2779, 1999.
9. Ragupathi, G., Meyers, M., Adluri, S., Howard, L., Yu, R.K., Ritter, G. and Livingston, P.O. Phase I trial with GD3-lactone-KLH conjugate and immunological adjuvant QS-21 vaccine with malignant melanoma. Int. J. Can 85: 659-666, 2000.
lO.Ragupathi, G., Livingston, P.O., Hood, C., Gathuru, J., Krown, S.E., Chapman, P.B., Wolchok, J.D., Williams, L.J., Oldfield, R.C., Hwu, W-J. Consistent antibodies against ganglioside GD2 induced in patients with Melanoma by a GD2 Lactone (GD2L)-KLH conjugate vaccine plus immunological adjuvant QS-21. Clin Can Res. In Press ll.Brezicka, F. T., S. Olling, et al. (1989).
"Immunohistological detection of fucosyl-GM1 ganglioside in human lung cancer and normal tissues with monoclonal antibodies." Cancer Res 49(5): 1300-5.
l2.Nilsson, O., Mansson, J.E., Brezicka, T. et al. (1984) e<
fucosyl-GM1-A ganglioside associated with small cell lung carcinomas. Glycoconjugate J 1: 43-49. 13.
l3.Komminoth, P., J. Roth, et al. (1991). "Polysialic acid of the neural cell'adhesion molecule distinguishes small cell lung carcinoma from carcinoids." Am J Pathol 139(2):
297-304. 14.
l4.Scheidegger, E. P., P. M. Lackie, et al. (1994). "In vitro and in vivo growth of clonal sublines of human small cell lung carcinoma is modulated by polysialic acid of the neural cell adhesion molecule." Lab Invest 70(1): 95-106.
15.
l5.Souhami, R. L., P. C. Beverley, et al. (1991). "Antigens of lung cancer: results of the second international workshop on lung cancer antigens." J Natl Cancer Inst 83 (9) : 609-12.
l6.Chapman, P.B., Livingston, P.O., Morrison, M.E., Williams, L., and Houghton, A.N. Immunization of melanoma patients with anti-idiotypic monoclonal antibody BEC2 (which mimics GD3 ganglioside): Pilot trials using no immunological adjuvant. Vaccine Res. 3:59, 1994.
l7.Krug,.L.M., Ragupathi, G., Ng, K.K., Hood, C., Jennings, H.J., Miller, V.A., Kris, M.G., Pizzo., B., Tyson, L., Baez, V., Livingston, P.O. Vaccination of small cell lung cancer pateints with polysialic acid or N-propionylated r polysialic acid conjugated to KLH. Clin Ca Res, in press.
l8.Joiner, K. A. (1988). "Complement evasion by bacteria and parasites." Annu Rev Microbiol 42: 201-30.
l9.Moffitt, M. C. and M. M. Frank (1994). "Complement resistance in microbes." Springer Semin Immunopathol 15(4):
327-44.
20.Jarvis, G.A., Li, J., Hakulinen, J., Brady, K.A., Nordling, S., Dahiya, R. and Meri, S. Expression and function of the complement membrane attack complex inhibitor protectin (CD59) in human prostate cancer. Int.
J. Cancer 71: 1049-1055, 1997.
2l.Cheung, N. K., E. I. Walter, W. H. Smith-Mensah, W. D.
Ratnoff, M. L. Tykocinski and M. E. Medof. Decay-accelerating factor protects human tumor cells from complement-mediated cytotoxicity in vitro. J Clin Invest 81 (4) : 1122-8, 1988.

22.Sakuma, T., K. Kodama, T. Hara, Y. Eshita, N. Shibata, M. Matsumoto, T. Seya and Y. Mori. Levels of complement regulatory molecules in lung cancer: disappearance of the D17 epitope of CD55 in small-cell carcinoma. Jpn J Cancer Res 84(7): 753-9, 1993.
23.Liu, N.X., Ragupathi, G., Cappello, S., Musselli, C., Lloyd, K. and Livingston, P.O. Antibodies against tumor glycolipids and proteins, but not mucins, mediate complement-dependent cytotoxicity (CDC). Proceedings AACR
41: #5582, 2000.
24.Price, M.R. and Tendler, S.J.B. Polymorphic epithelial mucins (PEM): molecular characteristics and association with breast cancer. The Breast 2: 3-7, 1993.
25.Colten, H.R. and Rosen, F.S. Complement deficiencies.
Ann Rev Immunol. 10:809 834.
26.Livingston, P.~., Zhang, S., Walberg, L., Ragupathi, G., Helling, F. and Fleischer, M. Tumor cell reactivity mediated by IgM antibodies in sera from melanoma patients vaccinated with GM2-KLH is increased by IgG antibodies.
Cancer Immunol. Immunother. 43:324-330, 1997.
27.Nicholson-Welter, A., Burge, J., Fearon, D.T., Welter, P.F., Austen, K.F. Isolation of a human erythrocyte membrane glycoprotein with decay-accelerating activity for C3 convertases of the complement system. J. Immunol 129:
184-189, 1982.
28.Jarvis, G.A., Li, J., Hakulinen, J., Brady, K.A., Nordling, S., Dahiya, R. and Meri, S. Expression and function of the complement membrane attack complex inhibitor protectin (CD59) in human prostate cancer. Int.
J. Cancer 71: 1049-1055, 1997.
29.Cheung, N. K., E. I. Walter, W. H. Smith-Mensah, W. D.
Ratnoff, M. L. Tykocinski and M. E. Medof. ~~Decay-accelerating factor protects human tumor cells from complement-mediated cytotoxicity in vitro." J Clin Invest 81(4): 1122-8, 1988.
30.Sakuma, T., K. Kodama, T. Hara, Y. Eshita, N. Shibata, M. Matsumoto, T. Seya and Y. Mori. "Levels of complement regulatory molecules in lung cancer: disappearance of the D17 epitope of CD55 in small-cell Carcinoma.' Jpn J Cancer Res 84(7): 753-9, 1993.

Claims (15)

1. A method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of:
a) Selection of an appropriate cancer cell line; and b) detection of the expression of antigens on the surface of said cell line, wherein the antigens expressed will be used in the polyvalent vaccine.

2. The method of claim 1, wherein the expression of the antigens is detected by a specific antibody.

3. The method of claim 2, wherein the antibody is a monoclonal antibody.

4. The method of claim 2 or 3, wherein the expression is detected by Fluorescence Activated Cell Sorter.

5. A method for identification of the optimal combination of a polyvalent vaccine against a cancer comprising steps of:
a) selection of an appropriate cancer cell line; and b) detection of the immunogenicity of antigens on the surface of said cell line, wherein the antigens showing said immunogenicty will be used in the polyvalent vaccine.

6. The method of claim 5, wherein the immunogenicity of antigens is determined by the Complement Dependent Cytotoxicity assay.

7. The method of claims 1, 2, 3, 4, 5, or 6, wherein the cancer is a small cell lung cancer.

8. The optimal combination identified by the method of claims 1, 2, 3, 4, 5, 6, or 7.

9. A polyvalent vaccine for small cell lung cancer comprising GM2, Fucosyl GM1, Globo H and polysialic acid, but not GD2 or GD3.

10. The vaccine of claim 9, further comprising sialyl Lewis a.

11. The vaccine of claim 9, wherein the antigens are conjugated.

12. The vaccine of claim 10, wherein the antigens are conjugated.

13. The vaccine of claim 11, wherein the antigens are conjugated to Keyhole Limpet Hemocyanin.

14. The vaccine of claim 12, wherein the antigens are conjugated to Keyhole Limpet Hemocyanin.

15. The vaccine of claim 9, 10, 11, 12, 13, or 14, including an appropriate adjuvant.