ICANCER RESEARCH56. 4749—4757. October IS, 19961
Immunization against Epitopes in the Human Melanoma Antigen gplOO following Patient Immunization Michael
L. Salgaller,'
with Synthetic Peptides
Francesco
M. Marincola,
Janice
N. Cormier,
and Steven
A. Rosenberg
Surgery Branch, National Cancer Institute. NIH, Bethesda, Maryland 20892
ABSTRACT
Freund's Adjuvant. PBMCs were analyzed for the ability to be sen sitized
gplOO is a melanocytic lineage-specific antigen recognized by tumor infiltrating lymphocytes, the adoptive transfer of which is associated with tumor regression In melanoma patients. In this study, peripheral blood
in vitro to the gplOO peptides
G9-209,
G9-280,
and G9-154,
and to the influenza Ml peptide. These gplOO peptides were selected based on recognition by bulk TTh cultures, the adoptive transfer of which was associated with tumor regression (1) as well as the previ ous demonstration that these epitopes were sufficiently immunogenic to elicit HLA- and antigen-restricted CTLs after in vitro stimulation with the native epitopes (6). Moreover, Parkhurst et al.3 have shown that selected peptides modified at HLA@A*020l@binding anchor
mononuclear cells (PBMCs) were harvested from HLA-A2' melanoma patients before and after immunization with G9-209 (ITDQVPFSY), G9280 (YLEPGPVTA), or G9-154 (KTWGQYWQV) peptides in Incomplete Freund's Adjuvant and were tested for the ability to be sensitized in vitro using PBMCs pulsed with the native peptides. In addition, PBMCs from patients receiving the G9-209 or G9-280 peptide were stimulated in vitro with peptides modified at anchor residues to enhance binding to HLA-A2:
Two of these, G9-209/2M
G9-209/2M
(IMDQVPFSY), and G9-280/9V containing a position 9 substitution
(IMDQVPFSY)
or G9-280-9V
(YLEPGPVTV).
In patients
immunized with G9-209, a single in vitro restimulation with G9-209/2M resulted In the generation of specific antipeptide lymphocytes from seven of seven postimmune PBMCs and only three of seven preimmune PBMCS. In patients immunized with G9-280, a single in vitro restimuladon with G9-280/9V resulted in the generation of specific antipeptide lymphocytes from five of six postimmune
PBMCS and four of six preimmune
PBMCs.
In almost all cases, CTLs raised against modified epitopes were capab'e of recognizing targets displaying the native nonamers. Several anti-G9-209 and antl-G9-209/2M CTLs also demonstrated specific lysis of, and specific IFN-@ release in response to, gplOO'-established cell lines. Thus, using peptides modified to enhance immunogenicity for in vitro stimulation improved the sensitivity of immune monitoring of patients immunized with synthetic peptides. These results demonstrate
that immunization
with
a peptide derived from a tumor-associated protein such as gplOOcan provoke a measurable antitumor immune response in cancer patients.
INTRODUCTION The gpl00 and MART-l proteins were identified as the target antigens recognized by TILs,2 the adoptive transfer of which was associated with the regression of metastases in patients with mela noma (1). Several peptides in these molecules have been identified that associate with the MHC class I heavy chain and f32-microglobulin to form the trimolecular MHC complex recognized by these antigen specific TILs. These findings raised the possibility that the specific peptide epitopes in gplOO and MART-l could be used as immunogens in active immunotherapy approaches to human cancer. However, these melanoma-associated epitopes are “self'peptides, expressed more abundantly, but not exclusively, on tumor compared to autolo gous normal cells. Therefore, an important concern is the degree to which such self antigens are immunogenic in vivo in humans, and whether the corresponding reactive precursor T cells have been meg atively selected, anergized, or rendered tolerant (2—5). To assess the immunogemcity of a gplOO-based peptide vaccine in vivo, we obtained PBMCs from melanoma patients before and after a series of injections with gplOO-derived nonamers in Incomplete
positions
had
increased
18U.S.C.Section1734solelyto indicatethisfact. I To
whom
all
correspondence
and
requests
for
reprints
should
be
addressed,
I MHC
containing
binding
a position
affinity.
2 substitution
(YLEPGPVTV), reduced the number of restimulations required to elicit antigen-restricted CTLs and confirmed previous findings relat ing enhanced
MHC binding
to greater
immunogenicity
(7—9).
In the current study, both native peptides and those containing single amino acid substitutions were used to sensitize human PBMCs harvested before and after in vivo immunization. In seven of seven patients following a single in vitro restimulation with G9-209/2M and in five of six patients following a single in vitro restimulation with G9-28019V, CTLs from postimmune PBMCs displayed epitope-re stricted IFN-'y secretion and cytolytic activity. The majority of CTLs generated against modified peptides recognized the nonmutated epitope as well. Therefore, in vitro stimulations with peptides modi fled to increase immunogenicity could be effectively used for the immune monitoring of patients receiving synthetic peptide-based im munization. These studies demonstrated that immunization with tu mor-associated gplOO peptides in WA can elicit an immune response in melanoma patients.
MATERIALS AND METHODS Cultured
Cell Lines. Established
tumor cell lines were initiated in the
Surgery Branch, National Cancer Institute, as described previously (10) by overnight incubation of minced, fresh tumor with a mixture of 0. 1% collagen
ase type IV, 0.002% DNase type I, and 0.01% hyaluronidase type V (Sigma Chemical Corp., St. Louis, MO) at 25°C. Those melanoma cultures established in our Branch have been given the suffix “mel.― They were maintained monolayers
in RPMI 1640 (Biofluids,
Inc., Rockville,
MD) supplemented
as
with
10% fetal bovine serum, 50 mist HEPES (pH 7.0), 10 flMglutamine, 250 units/ml penicillin-streptomycin,
and 0.25 mg/mI gentamicin.
We used the Daudi lymphoma line as a LAK cell target in chromium-release assays. LAK cells were prepared from normal buffy coats as described previ ously (11). They were maintained
in RPMI 1640 plus 2% human AB serum,
1000 units/ml rIL-2, in addition to the supplements listed above. Assay
of PBMCs
for Reactivity
to gplOO
Peptides.
PBMCs
were col
lected from buffy coats or lymphocytopheresis specimens of malignant mela noma patients treated in the Surgery Branch, National Cancer Institute. Cells were separated
Received 5/23/96; accepted 8/14/96. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with
peptide-class
by laying over lymphocyte
separation
medium (Organon
Teknika Corp., Durham, NC) washed with Ca2@-and Mg2@-freeHBSS, then used as fresh or cryopreserved
samples. HLA typing of peripheral
blood
lymphocytes from melanoma patients was carried out at the NIH HLA Typing Laboratory using the modified Amos microcytotoxicity assay (12).
at
Northwest Biotherapeutics, Inc., 120 Northgate Plaza. Suite 236, Seattle, WA 98125. 2 ‘@ abbreviations
used
are:
TIL,
tumor
infiltrating
lymphocyte;
PBMC,
3 M. Parkhurst,
peripheral
blood mononuclear cell; WA, Incomplete Freund's Adjuvant; LAK, lymphokine-ectivated killer, IL-2, interleukin-2; IVS, in vitro stimulation; pCl'L, precursor C1'L; VIL, vacci nation infiltrating lymphocyte.
M. L. Salgaller,
S. Southwood,
P. F. Robbins, A. Sette, S. A.
Rosenberg, Y. Kawakami, Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gpl00 modified at HLA-A50201 of binding residues,
J. Immunol.,in press,1996. 4749
IMMUNIZATION AGAINSTgplOoAF1@ER PATIENTIMMUNIZATION
On day 0, C'rLcultureswereestablishedin 24-well platesfromthe PBMCs of melanoma patients by plating 1.5—1.75 X 106cells/mI in 2 ml of Iscove's medium (Biofluids, Inc.) supplemented with 10% heat-inactivated human AB serum (Sigma Chemical Co.), as well as 50 misi HEPES, 1 mM sodium
Peptides. 09-154 (KTWGQYWQV),G9-209 (ITDQVPFSV),G9-209/2M (IMDQVPFSV), 09-280 (YLEPGPVTA), or 09-280/9V (YLEPOPVTV) pep tides were generated with a peptide synthesizer (model AMS 422; Gilson Co., Inc., Worthington, OH) (>90% purity) using a solid phase technique. The
pyruvate, 1% Non-Essential Amino Acids, 10
[email protected], 250 units/rnl penicillin-streptomycin, and 50 @g/mlgentamicin. All cultures were estab lished in the presence of 1 mg/ml final concentration of peptide for the in vitro monitoring of vaccine recipients. Forty-eight to 72 h later, 300 IU/ml IL-2 (kind gift of Chiron Corp., Emeryville, CA) were added to the cultures. Lymphocytes were harvested at day 11 for restimulation, washed once in Iscove's medium, then replated at a concentration of 4—5X l0@ cells/tad in 24-well plates. All responders were restimulated with autologous preimmune
influenza matrix peptide MIss_as (GILGFVFTL), derived from the influenza matrix protein Ml, was synthesized by Multiple Peptide Systems (San Diego,
CA). All are presentedin the context of HLA-A2 (1, 15, 16). Clinical
PBMCsthathadbeen thawed,washedtwice, andpulsedwith 1 @g/ml peptide for at least 3 h at 37°C in 15-miconicaltubes(5 ml volume;3 X lO@maximum concentration).
PrOtOCOl. All patients had metastatic
melanoma
histologically
confirmed by pathologists at the National Cancer Institute. The treatment protocol was approved by the Clinical Research Committee of the National Cancer Institute, and all patients signed an informed consent before treatment. To be eligible for treatment,patientswere requiredto have evaluable meta static melanoma. Each course of therapy consisted of the s.c. injection of 1, 3, or 10 mg of peptide emulsified in WA, administered s.c. every 3 weeks for a planned total of four doses.
During peptide pulsing, the cells were gently resuspended
several times to minimize clumping, and viability did not decrease. After this, stimulators
were irradiated
with 3000 rads using a ‘37Cssource,
washed
once
RESULTS
with Iscove's, and added to wells containing the appropriate effectors at stimulator:responder
ratios ranging between 5: 1 and 10: 1 (1 ml volume each).
Summary of Treatment Groups for Patients Undergoing gplOO + WA Immunization Protocol. Twenty-eight HLA@A*020l melanoma patients were accrued in a Phase I, dose-escalating inocu lation protocol involving 2—7s.c. immunizations of the 09-154, 09-209, or 09-280 peptides emulsified in WA (Table 1). Eight
The next day, 300 IU/ml IL-2 were added to the cultures. Some cultures were restimulated
a second time on day 18 using an identical procedure.
C1'L
activity was assessed before restimulation (day 11), as well as I week after the first or second rounds of restimulation
(days 18 and 25, respectively).
Phenotypic Analysis. Flow cytometry of CFLs was performed using a FACScan (Becton Dickinson, MountainView, CA), as describedelsewhere (13). Cells were stained with murine antihuman mAbs against CD3 (Leu-4), CD4 (Leu-3), CD8 (Leu-2), and TCR a/j3 (all from Becton Dickinson). Isotype-matched murine antibodies (IgG1 and IgG2a) served as a negative control. Cytotoxicity
Assays.
Target
cells (1 X 106) were labeled
patients had rapid disease progression and were unable to endure collection of PBMCs after at least three 09-209 or 09-280 doses of peptide, or at least two G9-154 vaccinations. Thus, 20 patients formed the basis of this analysis. Of the 20 patients given gplO0-based immunotherapy, 3 had a final pheresis after only two immunizations, and an additional 4 had a final pheresis after three immunizations. The 20 patients analyzed in this study received the treatments summarized in Table 1. Only one patient receiving the 09-209 peptide, patient 4, experienced an objective tumor regression after immunization. In Vitro Induction of Specific WN-y Secretion in gplOO-reactive CTLS Using gplOO Peptides. In preliminary experiments, it ap peared that better in vitro sensitization to gplOO peptides and superior cell expansion were obtained using 300 compared to 30 lU/mi IL-2. Similar results were obtained using 1 or 5 @&M concentrations of peptide (data not shown). Thus, all assays were conducted using 300 lU/mi IL-2 and 1 @tM peptide. PBMCs were sensitized on day 0 with peptide and on day 11 with peptide-pulsed PBMCs and assayed at day 18, as described in “Materialsand Methods.― Also, we sometimes restimulated at day 18 and retested at day 25. PBMCs were collected before the first immunization, then at triweekly intervals immediately preceding subsequent immunizations. PBMCs from seven melanoma patients receiving vaccinations with 09-209 peptide in IFA were analyzed for the ability to secrete IFN-'y following stimulation in vitro with the immunizing peptide, the mod ified G9-209/2M, or an irrelevant gplO0 peptide, 09-154. We arbi trarily defined PBMCs to be specifically sensitized when IFN-y release in response to the specific peptide was at least 50 pg/mi, and at least three times that in response to either the 09-154 or Ml irrelevant peptide-pulsed targets. An example of the induction of specific cytokine release in response to peptide-pulsed T2 cells using
with 200 uCi of
5tCr(New EnglandNuclear,Boston, MA) in 0.5 ml of RPM!with 2%human AB serum for 2 h at 37°C,then washed three times before addition of cultured CTL or LAK effectors. Targets displaying 70% viability were used. Targetand effector cells were incubated together for 4 h at 37°C.Supernatants were collected and counted on
a gammacounter.The percentageof specific lysis was calculatedas: % specific lysis
Sample counts —spontaneous counts Maximum counts —spontaneous counts
x 100
Spontaneous release was calculated from targets incubated with RPMI, max imum release from targets incubated with 2% SDS. All assays were executed in triplicate. Spontaneous counts were always 1011 ND2
the background
without
68
exogenous
peptide
ranged
260
ND
of CTLs cultured
NDND
for 24 h at 37C
from
0 pg/mI
ND
in the absence
for all unpulsed
T2
ND ND ND ND
ND ND ND NDND
of stimuletors.
Values
ND ND ND ND
. 50.
performed
a@40.
were at least 89% CD3@, nearly all of those being TCR TCt/I3@cells (data not shown). The day 25 phenotypic analysis is representative of CTLs used to produce the data in this report, because all the data presented in this study used PBMCs harvested within 1 week of this time point.
@1
/
30-
30
2010.
20. 10
on day 25 of culture.
For every patient and condition,
CTLs
0. 0
4
0
100
100 .
90
90.
80
80.
70
70.
C/) 60
@
d
4
DISCUSSION
patient5
In the present report, we have described a method for immunolog ical monitoring of patients receiving peptide-based vaccines by meas uring the induction in vitro of epitope-restricted CTLs from the PBMCs of melanoma patients before and after immunization. In vitro stimulations were carried out using common epitopes from the mela nocytic lineage protein gplOO (16, 18, 19). After a standardization of procedure, we studied the in vitro immunogenic properties of three
C/) 60. C/)
C/) @;50
50.
a@40 30
20
a
80
b
10 70
0
60 NO. OF IMMUNIZATIONS
@
NO. OF IMMUNIZATIONS
50
C,,
Fig. 1. Peptide recognition by C'FL.sinduced with G9-209/2M peptide. CTLs from patients 4-7 were generated by weekly stimulation with autologous PBMCs pulsed with G9-209/2M peptide. After one (day 18) restimulation, recognition of G9-209/2M, 09280/9V, or Ml was assessed using peptide-pulsed T2 cells as targets in 5tCr release
Patient 6
C,,
40 30
cytotoxicity assays (40:1 E:T). The data show target lysis at two time points at which
specific melanoma activity was examined. Date are presented for those patients in which melanoma-reactive PBMCs could and could not be induced. In graphs (a—tI), T2 + 09209/2M (U) represents lysis of T2 cells pulsed with the peptide used to in vitro-sensitize those cells; T2 + 09-280/9V (•)represents lysis of T2 cells pulsed with a negative control gploo peptide; and T2 + Ml (A) represents lysis of a HLA—A2-restricted nonmelanocytic lineage peptide from influenza A virus.
20
10
0
1 2 3 4 No.of immuruzations
5
0
1 2 3 4 No. of immurizations
In two of four patient PBMCs vaccinated with 09-209 and specif ically lyric toward 09-209/2M-pulsed T2 cells by day 25, PBMCs harvested after two immunizations exhibited HLA-A2- and gplOO restricted
@
cytotoxicity
against
the established
cell line 624.38mel
at
day 25 of culture (patients 6 and 7; Fig. 2). It is interesting that cultures in vitro stimulated with modified G9-209/2M, as well as the native G9-209, generated CTLs specifically lytic against 624.38me1.
U,
>.
LAK cells as a nonspecific effector in the cytolytic analysis of pulsed targets and established cell lines. LAK cell lysis of G9-154-pulsed and Mi-pulsed T2 cells was 59 and 53%, respectively, at an E:T ratio of 40:1. LAK cell lysis of the HLA-A2, gp100@ melanoma line 624.28mel was 29% at an E:T ratio of 40: 1. LAK cell lysis of the
Patient 7
40
In these experiments, stimulation in vitro with either G9-209 or 09-209/2M induced specific PBMCs of nearly equal potency. At an E:T ratio of 40:1, those PBMCs in vitro stimulated with the G9-209 peptide were only specifically lytic using post-ist and post-2nd PB MCs, after which they were only specifically cytotoxic by our defi nition at the lower E:T ratio due to high background lysis versus 624.28mel and A375 (data not shown). Due to the greater number of weekly restimulations required to observe cytotoxicity testing against established melanoma cell lines compared with pulsed targets (6), we did not perform this assay with any day 18 cultures. The lysability of nonspecific targets was established by including
50
U,
30 20
2
3
4
5
0
1
2
3
4
Fig.2. Kineticsof the HLA—A2-restricted melanomarecognitionby patients6 and7. The figure shows PBMCs induced with 09-209 or G9-209/2M peptide. PBMCs were
harvested before and 3 weeks after each round of immunization with 09-209. PBMC stimulation in vitro was performed twice (days I I and 18), then assessed for recognition of 624.38mel (}4J@@j@ gplOO'), 624.28mel (HLA—A2@,gplOO'), and A375 (HLA A2@, gplOO) as established melanoma cell line targets in “Crrelease cytotoxicity assays
(40:1 and 10:1E:T). The data show target lysis at time points for which specific melanoma activity was examined. Data are presented only for those patients in which melanoma reactive CFLs were induced. U, lysis of the positive control, 624.38mel. •,lysis of the
negative control for HLA—A2 expression, 624.28mel. A, lysis of the negative control for gplOOexpression, A375.
4755
IMMUNIZATION AGAINST gploo AFFER PATIENT IMMUNIZATION
peptides
recognized
by bulk TIL cell lines that were associated
with
clinical response following adoptive immunotherapy (I): G9-l54, G9-209, and G9-280. Because of the recognition of these gplO0 epitopes
by several
distinct
bulk
TIL
lines
associated
with
tumor
regression in vivo, it appeared that these may represent tumor rejection antigens
(1). In addition,
we studied
whether
G9-209/2M
and G9-
280/9V synthetic peptides, modified at HLA-A2 major binding resi dues to enhance stability of the trimolecular complex, could evoke specific CTLs more rapidly than that obtained with its normal cognate. In addition, we had successfully induced melanoma-reactive CTLs with gpl0O native (6) and modified peptides3 using 1—5p.g/ml per stimulation. Peptide titration analysis had shown that maximum lysis of T2-pulsed target cells could be achieved with —I p.Mnative peptide (6) and 1—100flMmodified peptide.3 In initial cocultures, we observed no consistent benefit in using S ,.LMinstead of 1 p@ peptide as measured
by the kinetics
or frequency
of epitope-specific
CTL induc
tion; thus, the lower concentration was subsequently used. IL-2 stim ulates the proliferation of CTL, ThI , and Th2 cells and enhances TNF-a and IFN-y production (20). The higher IL-2 concentration in culture wells promoted the expansion (data not shown) and specificity of CTLs stimulated with both gplO0 and Ml peptides. Importantly, such enhancement was observed without a concurrent increase in nonspecific
activity.
Therefore,
it appears
such
cocultures
can be
established in 300 lU/mi IL-2 without diminishing the effectiveness of this monitoring technique. Although patients were immunized solely with native epitopes, we used both native and modified synthetic peptides for in vitro sensiti zation. Whereas G9-154-, G9-209-, and G9-280-restricted responses were observed in CTLs from nonimmunized melanoma patients after at least four rounds of weekly restimulation (6), stimulation with peptides modified to enhance HLA-A2 induced reactivity more rap idly and in a greater percentage of patients.3 This report confirms previous work with viral epitopes that peptides with amino acid substitutions that result in enhanced binding to HLA also renders them more immunogenic in vitro (7, 21). Because high-affinity HLA binding peptides are able to induce antigen-restricted CTLs most frequently (22), it was important to investigate amino acid substitu tions of G9-209 and G9-280, both of which are intermediate binders
(I). The ability of CTLs generatedagainstmodifiedepitopesto recognize the native cognate also confirms prior studies.3 One important finding of this study is the more rapid kinetics of PBMC response to stimulation in vitro with either native or modified gplOO epitopes using PBMCs from subjects after a series of immu nizations with gplOO peptides in IFA. None of the preimmunization CTLs from seven patients subsequently vaccinated with 09-209 or any of six patients subsequently vaccinated with G9-280 demon strated specificity to any T2-pulsed target when stimulated in vitro with the native peptide, except for the recall antigen Ml. Although some preimmune reactivity could be seen when stimulating in vitro with either G9-209/2M or G9-280/9V, the percentage of specific induction and the strength of response were greater using autologous postimmunization
PBMCs.
After only one in vitro restimulation
G9-209/2M, antigen-restricted harvested
2 weeks
after a single
with the native
release.
Because
only a single patient
receiving
the G9-209
peptide experienced tumor regression, we could not correlate the strength of the in vitro response to its clinical effectiveness. Overall, reactivity was slower to develop and weaker when established mela noma cell lines were used as targets, but this is not surprising because, unlike pulsed targets, tumor cells display a much greater array of surface antigens. However, all four patient (7—10)CTLs tested against melanoma cell lines at day 25 of culture were positive for specific IFN-'y secretion. Also, our findings suggest the absence of tolerance induction by gplOO antigens in vivo, and that in vivo immune reac tivity
might
be enhanced
by vaccination
even
with an immunogen
such as synthetic peptide in adjuvant. Estimation of pCTL frequency is another approach to assessing im mune status, although we have not been successful in using this method
to detect anti-gplOOreactivity in our patients. Recent reports used limit ing dilution analysis to estimate pCFL frequencies toward viral antigens such as influenza (24). The powerful anti-influenza Ml response that was noted in our studies using primary CTL cultures without restimulation is not an uncommon observation (15, 25).@Indeed, pCTL levels can be estimated ifhighly immunogenic antigens such as hepatitis C (26) or HIV (23, 27) serve as stimulators, precluding the need for restimulation. Most investigations reporting nonprimary pCFL levels used a single stimulator
concentration to restimulate several dilutions of responders (23, 24, 26). The result would be higher stimulator.responder ratios in microwells containing fewer cells and lower ratios in those wells containing more effectors. Because greater stimulator.responder ratios can enhance spe cific CFL induction,5 this approach could affect the accuracy of pCFL frequency determinations. Human
CU
analysis
following
peptide-based
vaccination
has re
cently been reported. Mukherji et a!. (28) used the MAGE-l nonapep tide (EADPTGHSY) pulsed onto adherent mononuclear cells as an autologous melanoma vaccine. Er vivo expanded infiltrating T cells from the vaccination site (VIL) and from metastatic tumor deposits
(TIL) were analyzed in two patients after four monthly immunizations and at least three restimulations in vitro. Postimmune VIL and TIL displayed enhanced cytolytic activity against relevant targets, yet random TNF-(3 synthesis by pre- versus postimmune VIL and ilL was noted (28). Tsang et a!. (29) observed the induction of CTLs specific for carcinoembryonic antigen in 3 of 26 colon carcinoma patients after three monthly immunizations with a recombinant vac cinia virus expressing carcinoembryonic antigen. In this report, we were able to demonstrate the induction of specific CTLs from >50% of G9-209- and 09-280-vaccinated subjects, and to generate specific CTLs
from PBMCs
after only one or two immunizations.
We are currently developing pox- and adenovirus-based vaccines expressing the gplOO protein for use in patient immunization. The techniques described here will be of value in the assessment of the patients receiving these vaccines. ACKNOWLEDGMENTS We thankArnoldMixon for fluorescence-activatedcell sortinganalysis.
with
CT!.. activity was detected in PBMCs vaccination
chromium
peptide;
REFERENCES
I. Kawakami,Y.,Eliyahu,S., Delgado,C. H.,Robbins,P. F., Sakaguchi,K.,Appella, this occurred less frequently when G9-209 was the in vitro stimulating E., Yannelli,J. R., Adema,0. J., Mild,T., and Rosenberg,S. A. Identificationof a peptide. Because gplO0 is a self-antigen, the possibility existed that human melanoma antigen recognized by tumor-infiltrating lymphocytes associated high cell-surface expression would elicit tolerance resulting from the with in vivo tumor rejection. Proc. NatI. Acad. Sd. USA, 91: 6458, 1994. 2. Schild, H., Rdtzschke, 0., Kalbacher, H., and Rammensee, H-G. Limit of T cell eradication or deactivation of reactive T cells (23). However, these intermediate-affinity and potentially subdominant epitopes were able 4 F. Marmncola, L. Rivoltini, M. L Salgaller, M. Player, and S. A. Rosenberg, to generate specific CTLs in five of eight 09-209-immunized and four Differential anti-MART-l/melan A C1'L activity in comparison to healthy donors: cvi of nine G9-280-immunized subject PBMCs. IFN-'y secretion appears dence for in vivo priming by tumor cells, J. Immunother., in press, 1996. to be a more sensitive readout of patient response compared to 5 M. Salgaller, F. Marmncola, and L. Rivoltini, unpublished observations. 4756
IMMUNIZATION AGAINST gpIOO AFTER PATIENT IMMUNIZATION
tolerance to self proteins by peptide presentation. Science (Washington DC), 247:
17. Kawakami, Y., Eliyahu, S., Sakaguchi, K., Robbins, P. F., Rivoltini, L., Yannelli,
1587, 1990. 3. Vidard, L., Colarusso, L. J., and Benacerraf, B. Specific T-celI tolerance may be
J. R., Appella,E., and Rosenberg,S. A. Identificationof the immunodominant
preceded by a primary response. Proc. Nail. Acad. Sci. USA, 91: 5627, 1994. 4. Ferber, I., Schönrich,0., Schenkel, J., Mellor, A. L., Hammerling, 0. J., and Arnold, B. Levels of peripheral T cell tolerance induced by different doses of tolerogen.
HLA-A2-restricted tumor infiltrating lymphocytes. J. Exp. Med., /80: 347, 1994.
Science (Washington DC), 263: 674, 1994.
5. Aichele,P., Brduscha-Riem, K.,Zinkernagel,R. M.,Hengartner,H.,andPircher,H. T cell priming versus T cell tolerance induced by synthetic peptides. J. Exp. Med., 182: 261, 1995.
6. Salgaller, M. L., Afshar, A., Marmncola,F. M., Rivoltini, L., Kawakami, Y., and Rosenberg, S. A. Recognition of multiple epitopes in the human melanoma antigen gpIOO by peripheral blood lymphocytes stimulated in vitro with synthetic peptides.
Cancer Res., 55: 4972, 1995. 7. Seue, A., Vitiello, A., Reherman, B., Fowler, P., Nayersina. R., Kast, W. M., Melief,
C.J. M.,Oseroff,C.,Yuan,L.,Ruppert,J., Sidney,J.,delGuercio,M-F.,Southwood, S., Kubo, R. T., Chesnut, R. W., Grey, H. M., and Chisari, F. V. The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epitopes. J. Immunol.. 153: 5586, 1994. 8. Chen, W., Khulko,S., Fecondo, J., Margulies, D. H., and McCluskey, J. Determinant selection of major histocompatibility complex class I-restricted antigenic peptides is explained by class 1-peptide affinity and is strongly influenced by nondominant anchor residues. J. Exp. Med., 180: 1471, 1994. 9. Sette, A., Sidney. J., del Guercio, M., Southwood, S., Ruppert, J., Dahlberg, C., Grey, H., and Kubo, R. Peptide binding to the most frequent HLA-A class I alleles measured
by quantitativemolecularbindingassays.MoI.Immunol.,31: 813, 1994. 10. Topalian,S. L., Solomon,D., and Rosenberg,S. A. Tumor-specificcytolysisby lymphocytes infiltrating human melanomas. J. Immunol., 142: 3714, 1989. 11. Rosenberg, S. A., Lotze, M. T., Muul, L M., clang, A. E., Avis, F. P., Leitman, S., Linehan. W. M., Robertson, C. N., Lee, R. E., Rubin, J. T., Seipp, C. A., Simpson, C. G., and White, D. E. A progress report on the treatment of 157 patients with
advanced cancer using lymphokine-activated killer cells and interleukin-2 or high dose interleukin-2 alone. N. Engl. J. Med., 316: 889, 1987. 12. Hopkins, K. A., van Leeuwen, A., Tardiff, G. N., and LeFor, W. M. Lymphotoxicity testing. In: A. A. Zachery and 0. A. Teseri (eds.), ASHI Laboratory Manual. American Society for Histocompatibility and Immunogenetics, Lenexa, KS. 195, 1990. 13. Topalian, S. L., Muul, L. M., Solomon, D., and Rosenberg, S. A. Expansion of human tumor infiltrating lymphocytes for use in immunotherapy trials. J. Immunol. Methods, 102: 127, 1987.
14. Horn, S. S., Schwartzentruber, D. J., Rosenberg, S. A., and Topalian, S. L. Specific release of cytokines by lymphocytes infiltrating human melanomas in response to shared melanoma antigens. J. Immunol. Methods, 13: 18, 1993.
15. Utz, U., Koenig, S., Coligan, J. E., and Biddison, W. E. Presentation ofthree different viral peptides. HTLV-1TAX, HCMV gB. and influenza virus MI, is determined by common structural features ofthe HLA-A2.l molecule. J. Immunol., 149: 214, 1992. 16. Cox, A. L, Skipper, J., Chen, Y.. Henderson, R. A., Darrow, T. L., Shabanowitz, J., Engelhard, V. H., and Singluff, C. L., Jr. Identification of a peptide recognized by five melanoma-specific
human cytotoxic T cell lines. Science (Washington
peptides of the MART-l
human melanoma antigen recognized by the majority of
18. Kawakami, Y., Eliyahu, S., Jennings, C., Sakaguchi, K., Kang. X.. Southwood, S.,
Robbins, P. F., Sette, A., Appella, E., and Rosenberg, S. A. Recognition of multiple epitopes in the human melanoma antigen gplOo by tumor-infiltrating
T lymphocytes
associated with in vivo tumor regression. J. Immunol., 154: 3961, 1995. 19. Bakker, A. B. H., Schreurs, M. W. J., de Boer, A. J., Kawakarni, Y., Rosenberg, S. A., Adema, 0. J., and Figdor, C. G. Melanocyte lineage-specific
antigen gplO0 is
recognized by melanoma-derived tumor-infiltrating lymphocytes. J. Exp. Med., 179: 1005, 1994. 20. Mass, R. A., Dullens, H. F. i., and Den Otter, W. Interleukin-2 in cancer treatment:
disappointing or (still) promising? Cancer Immunol. Immunother., 36: 141, 1993. 21. Lipford, G. B., Bauer, S., Wagner, H., and Heeg, K. Peptide engineering allows cytotoxic T-cell vaccination against human papilloma Immunology. 84: 298, 1995.
virus tumour antigen, E6.
22. Motioka, N., Kikumoto, Y., Hoon, D. S. B., Morton, D. L., and Inc. R. F. A decapeptide (Gln-Asp-Leu-Thr-Met-Lys-Tyr-Gln-IIe-Phe)from human melanoma is recognized by C-rL in melanoma patients. J. Immunol., 153: 5650, 1994.
23. Livingston, P. 0., Wong, G. Y. C., Adluri, S., Tao, Y., Padavan, M., Parente, R., Hanlon, C., Jones-Calves, M., Helling, F., Rifler, G., Oettgen, H. F., and Old. L. J. Improved survival in stage II melanoma patients with GM2 antibodies: a randomized trial of adjuvant vaccination with GM2 ganglioside. J. Clin. Oncol., 12: 1036, 1994.
24. Lehner, P. J., Wang, E. C. Y., Moss, P. A. H., Williams, S., Platt, K., Friedman, S. M.,
Bell,J. I., and Borysiewicz,L. K. HumanHLA-A020l-restricted cytotoxicT lym phocyte recognition of influenza A is dominated by T cells bearing the v-beta-7 gene segment. J. Exp. Med., 181: 79, 1995.
25. Rivoltini, L., Kawakarni, Y., Sakaguchi, K., Southwood, S., Sette, A., Robbins, P. F., Marincola, F. M., Salgaller, M. L., Yannelli, J. R., Appella, E., and Rosenberg, S. A. Induction of tumor reactive CFL from peripheral blood and tumor infiltrating lym phocytes of melanoma patients by in vitro stimulation with the human melanoma antigen MART-I. J. Immunol., 154: 2257, 1995. 26. Cheever, M. A., Disis, M. L., Bernhard, H., Gralow, J. R., Hand, S. L., Huseby, E. S.,
Qin, H. L., Takahaski, M., and Chen, W. hnmunity to oncogenic proteins. Immunol. Rev., 145: 33, 1995. 27. McFarland, E. J., Harding, P. A., Luckey, D.. Conway, B., Young, R. K., and Kurtizkes, D. R. High frequency of gag- and envelope-specific cytotoxic T lympho cyte precursors in children with vertically acquired human imrnunodeficiency virus type 1 infection. J. Infect. Dis., 170: 766, 1994. 28. Mukheiji, B., Chakraborty, N. 0., Yamasaki, S., Okino, T., Yamase, H., Spom, J. R.,
Kurtzman, S. K., Ergin, M. T., Meehan, J., and Mauri, F. Induction of antigen-specific cytolytic T cells in situ in human melanoma by immunization with synthetic peptide pulsed autologous antigen presenting cells. Proc. NatI. Acad. Sci. USA, 92: 8078, 1995. 29. Tsang. K. Y.. Zaremba, S., Nieroda. C. A., Thu. M. Z., Hamilton, J. M., and Schlom,
J. Generationof humancytotoxicT cells specificfor humanCarcinoembryonic Antigen epitopes from patients immunized with recombinant vaccinia-CEA vaccine.
DC), 264:
716, 1994.
J. Natl. Cancer Inst., 87: 982, 1995.
4757
