| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Departments of Gynecologic Oncology [B. W. A., D. M. G., C. G. I.], Surgery [G. E. P.], Bioimmunotherapy, [J. L. M., M. E. G.], and Immunology [C. G. I.], The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
ABSTRACT
The
presence of tumor-reactive CTLs in tumor infiltrates and in the
peripheral blood of cancer patients demonstrates an immune response
against tumors that apparently cannot control disease spread. This
raises concerns as to whether amplification of this response may be
useful during disease progression. Induction of tumor-reactive CTLs in
healthy donors at risk, as well as in patients free of disease, may be
therapeutically important, based on the hypothesis that CTLs that
recognize tumors early may be more effective in containing their
progression than CTLs that expand only when the disease progresses. To
address the feasibility of priming cytolytic activity in healthy
donors, we used the HER-2 peptide E75 (369–377) as an immunogen and
autologous peripheral blood mononuclear cell-derived dendritic cells as
antigen-presenting cells. We found that of 10 healthy donors tested,
two responded at priming with E75 presented on autologous dendritic
cells by induction of E75-specific CTL activity. Three other responders
were identified after two additional restimulations. Of these five
responders, three recognized E75 presented on the ovarian tumor line
SKOV3.A2, as demonstrated by cold-target inhibition experiments.
Induction of cytolytic activity at priming was enhanced in responders
by tumor necrosis factor-
and interleukin 12 but not in the
nonresponders. B7.1 monoclonal antibody added at priming enhanced
induction of lytic activity in only one of the four nonresponding
donors tested, suggesting that in the majority of donors,
E75precursor CTLs were not tolerized. Because of the possibility
that disease may develop in nonresponders, strategies to improve the
immunogenicity of tumor antigens for healthy donors may be required for
development of cancer vaccines.
INTRODUCTION
Identification of human epithelial tumor Ags,4 such as the ones expressed on ovarian and breast cancers, allows antitumor vaccination strategies to be developed. Among the most interesting are those that focus on HER-2 because this proto-oncogene is overexpressed in 20–40% of patients with highly aggressive breast, ovarian, pancreas, colon, and prostate cancers and with consequent poor prognosis. Two clinical trials have targeted HER-2 (1 , 2) using peptides and various adjuvants. The immunogen of choice in these trials was of the HER-2 peptide E75 (369–377, KIFGSLAFL), which maps an epitope frequently recognized by CTLs from tumor-infiltrating/associated lymphocytes of breast and ovarian cancer patients (3 , 4) .
Although peptide immunization is an appealing approach to tumor immunotherapy because it removes concerns of toxicity and safety while focusing the effectors, the methodology for vaccination and immunological evaluation it is not yet defined (5) . Important questions need to be addressed before this approach can be developed to its therapeutic potential. The first question is whether CTLs generated by primary in vitro and in vivo immunizations will lyse targets endogenously expressing the Ag. It has been shown with model Ag that the majority of peptide-induced CTLs at priming recognized the peptide used as immunogen, but only a small fraction recognized the endogenously presented Ag (6) . In some instances, CTLs recognizing endogenously presented Ag could be induced only with a variant peptide (7) . Although peptide-specific CD8+ cells may not always be expected to directly lyse tumors in vitro and in vivo, such cells can recognize peptides derived from extracellular degradation of Ag from dying tumor cells and tumor debris. This may lead to secretion of various cytokine patterns in the tumor environment and conditioning of APCs, resulting in an indirect but significant impact on antitumor responses.
The second concern is how frequently tumor peptide vaccinations induce Ag-specific CTLs in the human population. This concern is attributable to the reported low precursor frequency of tumor-reactive CTLs in healthy individuals, however, this concept has been recently challenged (8 , 9) . There is also concern over the weak ability of tumor Ag to induce massive Ag-specific CTL expansions, as reported with viral Ags (10) . The third concern is whether self-reactive (tumor-reactive) CTLs in healthy donors are silenced by active tolerance or anergy, and stimulation with Ag in peptide form cannot reactivate memory effectors because of B7-CTLA4-mediated peripheral tolerance. Recent studies have shown that induction of melanoma tumor Ag and tumor-reactive CTLs in healthy donors is much less effective than in cancer patients (10) . However, induction of tumor-reactive CTLs in healthy donors (as well as in breast and ovarian cancer patients in long-term remission and without evidence of disease) is important based on the hypothesis that CTLs that recognize tumors early may be more effective in containing their progression than CTLs that expand only when the tumor Ag is overexpressed. (11) .
Although a number of studies focused on improving the immunogenicity of tumor peptides in selected responding patients and donors using DCs as APCs and inflammatory cytokines, there is little information on the frequency of induction of these responses in unselected healthy donors. However, this question is important because cancer vaccines are expected to be given to distinct individuals, of which some may be at risk to develop disease, whereas others may be free of disease and otherwise considered healthy individuals. Thus, the frequency of CTL responses to a tumor Ag in the population becomes an important issue. We rationalized that if the frequency of responses to E75 priming is similar to or lower than the frequency of responses to MART-1 (10) , initial screening of a large panel of healthy donors may identify at least one responder. Cells of this responder can be then used as positive controls to address the questions of costimulation and of cytokine help in elicitation of cytolytic function in nonresponders.
We developed a model for priming T cells of PBMCs from healthy donors
with the HER-2 peptide E75. We used as APCs autologous DCs, always
freshly generated in the presence of GM-CSF + IL-4 from the same PBMC
sample. To determine the role of costimulation in this system, B7.1
antibodies were added at priming. To establish whether IL-12 and
TNF- are essential for CTL priming, stimulations were performed in
the presence or absence of these cytokines. We found that 2 of 10
healthy donors responded by inducing E75-specific cytolysis at peptide
priming and 5 of 10 at restimulation. Although IL-12 and TFN-
potentiated CTL induction in the responsive donors, they did not help
induce CTLs in nonresponders, suggesting that additional factors to the
nature of APCs and inflammatory cytokine conditioning regulate the
induction of CTLs specific for HER-2 by synthetic peptides.
MATERIALS AND METHODS
Cells, Antibodies and Cytokines.
HLA-A2+ PBMCs were obtained from healthy
volunteers from the Blood Bank of M. D. Anderson Cancer Center. The
HLA phenotypes of the donors used in this study were: donor 1 (A2, B7,
44); donor 2 (A2, 33, B40, 44); donor 3 (A2, 33, B41, 81); donor 4 (A1,
2, B27, 44); donor 5 (A1, 2 B44, 57, Cw5, 6); and donor 6 (A2, 31, B35,
44, Cw4, w5). For the other four donors only, the HLA-A2 expression was
determined. T2 cells, ovarian SKOV3, SKOV3.A2 cells, and indicator
tumors from ovarian ascites were described (3
, 4)
. mAb to
CD3, CD4, CD8 (Ortho), CD13 and CD14 (Caltag Laboratories, San
Francisco, CA), B7.1 and B7.2 (CD80 and CD86, Calbiochem),
intercellular adhesion molecule-1 (ICAM-1 CD54; Calbiochem), CD40L
(Ancell, Bayport, MN), HLA-A2 (clone BB7.2; American Type Culture
Collection), and MHC-II (L243; Dako Corp., Carpinteria, CA) were used
as unconjugated, FITC, or phycoerythrin conjugated. The following
cytokines were used: GM-CSF (Immunex Corp., Washington, DC; specific
activity, 1.25 x 107 colony-forming
units/250 mg); TNF-a (Cetus Corp., Emeryville, CA; specific
activity, 2.25 x 107 units/mg); IL-4
(Biosource International; specific activity, 2 x
106 units/mg); and IL-2 (Cetus Corp.; specific
activity, 18 x 106 IU/mg). IL-12 at 5 x 106 units/mg was a kind gift from Dr. Stanley
Wolf (Department of Immunology, Genetics Institute, Cambridge, MA).
Synthetic Peptides.
The HER-2 peptides used were: E75 (369–377) and the unnatural modified
Muc-1 peptides D125: (GVTSAKDTRV) and D132
(SLADPAHGV). The corresponding natural peptides
do not bind HLA-A2. Introduction of an HLA-A2 anchor and sequence
modification in Muc1 in residues contacting TCR lead to an unnatural
sequence (12)
. All peptides were prepared by the Synthetic
Antigen Laboratory of M. D. Anderson Cancer Center and purified
by high-performance liquid chromatography. Peptides were 95–97% pure
by amino acid analysis. Peptides were dissolved in PBS and stored
frozen at -20°C in aliquots of 2 mg/ml Polyglycol bead-containing
E75 were a kind gift of Dr. Kenneth Grabstein (Corixa Corp., Seattle,
WA).
Immunofluorescence.
Antigen expression by DCs and T cells was determined by
fluorescence-activated cell sorter using a flow cytometer (EPICS
Profile Analyzer; Coulter Co., Hialeah, FL). DCs were defined by
the presence of CD13 and absence of CD14 marker after culture in GM-CSF
and IL-4. For phenotype analysis, DCs were incubated with
phycoerythrin-conjugated anti-CD13 mAb and FITC-conjugated mAb specific
for a surface Ag.
Generation of PBMC-derived DCs.
CD13+ DCs were generated from freshly isolated
PBMCs by following the established CD14 methods (13
, 14)
.
Complete RPMI 1640 (containing 10% FCS) supplemented with 1000
units/ml GM-CSF, and 500 units/ml IL-4 were added to each well
containing plastic-adherent cells, changed every 48 h, and
maintained for 7 days. In separate studies, performed in parallel, we
attempted to grow DCs in medium containing either HS or in AIM-V
medium. The growth and expression of surface markers of DC cultured in
complete RPMI 1640 was significantly better than of DCs cultured in
other conditions; thus, in this study only DCs cultured in complete
RPMI 1640 were used. CD8+ cells were isolated by
removing first the CD4+ and then the
CD16+ and CD56+ cells from
the nonadherent population using Dynabeads (Dynal, Oslo, Norway). After
depletion, the resulting cells were 97% CD8+, as
determined by flow cytometry.
T-Cell Stimulation by Peptide-pulsed DCs.
DCs were washed three times in serum-free medium, plated at 1.2 x
105 cell/well in 24-well culture plates, and
pulsed with peptides at 25–50 µg/ml in serum-free medium for 4 h before the addition of responders. TNF- (50 units/ml) was added to
DCs for the last hour to stimulate Ag uptake and presentation
(13)
. Autologous PBMCs or isolated
CD8+ cells in RPMI 1640 containing 10% HS were
added to DCs at 1.5 x 106/ml, followed 60
min later by IL-12 at 3 IU/ml, IL-2 was added 16 h later to each
well at 60 IU/ml and every 48 h thereafter. For inhibition
studies, mAbs specific for B7.1, B7.2, and HLA-A2 were added to DCs
1 h before responders in amounts reported to be inhibitory by the
manufacturers.
CTL and Cytokine Assays.
Recognition of peptides used as immunogens by CTLs was performed as
described (3
, 15)
. Equal numbers of viable effectors from
each well were used in all assays. To minimize cross-reactive
recognition of human peptides, all stimulations were preformed in
medium containing HS, whereas CTL assays were performed using medium
containing FCS (15)
. This minimized background cytolysis
attributable to activation of other autoreactive cells to Ags present
in HS. Specific LU were determined as described (10
, 16)
and are expressed as LU 30/107 cells.
RESULTS
Priming of Healthy Donors PBMCs with E75-pulsed Autologous
CD14-derived DCs.
To address the question of whether priming with E75 presented on DCs
induce E75specific CTL activity, plastic nonadherent PBMCs from
healthy donors were stimulated with autologous DCs generated by culture
in GM-CSF + IL-4. DCs were pulsed with E75 at 25–50 µg/ml. Seven
days later, cytolytic activity was determined against E75-pulsed T2
using as control T2 cells that were not pulsed with peptide. Cumulative
results are presented in Table 1
. The
results show that of 10 healthy donors tested, only 2 showed stable and
consistent recognition of E75 after priming with DC-E75. These
cumulative results show that E75 can induce specific cytolytic activity
at priming only in a small fraction of healthy donors (2 of 10; 20%).
These results were confirmed because each CTL induction experiment was
repeated at least three times at different time points and always with
freshly isolated PBMCs (except donor 10). In 2 donors (donors 2 and 9),
specific CTL activity was occasionally detected in one of three
independent experiments, but this activity was unstable and could not
be further expanded. Induction of this activity in donor 9 required
B7.1 at priming (shown in Fig. 5
). In contrast, in the two
responding donors, specific CTL activity was detected at priming in 4
of 6 (donor 1) and 3 of 4 (donor 5) independently performed induction
experiments over periods of 1 year and 6 months, respectively.
|
|
and IL-12 at priming and IL-2 every 48 h for the
following 7 days. A representative experiment of the specificity of
recognition of CTLs from donor 1 primed by DC-E75 is shown in Fig. 1
. The results show that the specificity
of recognition of CTLs induced by E75 at three concentrations of
peptide (25, 50, and 100 µg/ml) was similar, with 50 µg/ml E75
inducing the highest differences in lytic activity between targets
pulsed with E75 or targets pulsed with no peptide. CTLs could not be
induced even by high concentrations of E75 in any of the nonresponders.
To address whether E75-stimulatory potency was increased by delivery of
Ag encapsulated in polyglycol beads, PBMCs from two nonresponders
(numbers 2 and 9) were stimulated with autologous DCs pulsed with
polyglycol beads in numbers to generate equivalent E75 concentrations
with peptides. The results are shown for donor 9 (Fig. 1
B).
Stimulation with encapsulated-E75 failed to induce specific CTL
activity at priming, as concluded from 5-h CTL assays. However, in
longer CTL assays (16 h), lysis of T2-E75 pulsed targets increased
significantly compared with that of control targets, and the levels of
recognition were similar. This suggested that either the numbers of
E75-specific CTLs were low, or CTLs in these two donors had weak
cytolytic activity. It should be mentioned that in both experiments,
IL-2 was added at 48 and 96 h in higher concentrations (150–200
IU/ml) to facilitate T-cell expansion. Although this increased the
background lysis, it did not change the patterns of recognition.
|
and IL-12 in Induction of CTL Activity at
Priming. and IL-12 have been described in different systems to
augment cytotoxicity of CD8+ cells (17
, 18)
. To determine whether the E75-induced CTL activity could be
enhanced by TNF- and IL-12, CTL priming experiments were performed
in the presence and absence of these cytokines (Fig. 2)
. The results from donor 5 show that
the addition of TNF- at the time of pulsing with peptide-enhanced,
E75-specific activity compared with cultures that received only IL-2.
When IL-12 was added, the increase in T2-E75 killing was paralleled by
an increase in nonspecific killing. The addition of IL-12 at higher
concentrations during priming did not increase the specific but rather
the nonspecific CTL activity. This was equally true when isolated
CD8+ cells were used as effectors (data not
shown). To address whether these cytokines enhanced cytolytic activity
at priming in nonresponders, the experiment was repeated with donor 1
(responder) and donor 4 (nonresponder). The results in Fig. 3
show that TNF- increased specific
CTL activity in donor 1 but not in donor 4. Although T2-E75 lysis
increased in TNF--treated cultures, it was still not significantly
different from the control. Similar to donor 5 (in both donors 1 and
4), IL-12 increased both the nonspecific and specific lysis. These
results were confirmed with all donors tested (Table 1)
. IL-12 alone,
or together with IL-2 and TNF-, failed to induce specific CTL
activity in nonresponders.
|
|
and IL-12 at priming was attributable to changes in the levels
of CD8+ cells in these cultures, we determined
the percentages of CD8+ and
CD4+ in E75-primed cultures from donor 5 used in
the experiment shown in Fig. 2
. The results in Table 2
show only small differences between the
percentage of CD8+ cells in cultures containing
IL-2 only compared with IL-2 + TNF- or IL-2 + TNF- + IL-12. A
caveat of this analysis is that E75 tetramers are not yet available;
thus, we could not determine whether TNF- and IL-12 increased the
numbers of E75-specific cells in these cultures.
|
B7.1 and B7.2 mAbs were added at priming.
The results in Fig. 4
show that B7.1
significantly inhibited induction of specific CTL activity in donor 5
(by >80%), whereas B7.2 had a much smaller effect.
|
B7.2 mAb was insufficient for
blocking. The strong inhibition of E75-specific CTL induction by
B7.1 suggested that the responder CTLs in this donor are more likely
naïve T cells.
To determine whether the nonresponders were activated but tolerized T
cells, which cannot expand because of B7-CTLA4 interaction, the
experiment was repeated with four nonresponders (nos. 3- 6) using the
same amounts of B7.1 as in donor 5. The results are shown in Fig. 5
. In donor 9, the addition of B7.1 at
priming led to induction of specific CTL activity (Fig. 5
A).
It is tempting to speculate that in this donor, activated but tolerized
E75-specific CTLs were present, and they cannot expand because of the
B7-CTLA4 interaction. Additional studies are needed to address this
point. In donors 4 and 5, the addition of B7.1 at priming failed to
induce significant specific cytolytic activity. These results were
confirmed with donor 3 (not shown). Thus, of five donors tested,
B7.1 inhibited E75-specific CTL priming in one (no. 5), enhanced CTL
priming in another one (no. 9), but failed to enhance induction of
specific CTL activity in three (nos. 3–5). These experiments were
repeated, and the results were confirmed. Thus, the requirements for
B7-CD28 costimulation appeared to be dependent on the donor.
Induction of CTL Activity at Restimulation.
To address whether E75 restimulation enhanced specific cytolytic
activity, E75 primed PBMCs from all donors were restimulated with
DC-E75. Of the eight nonresponders at primary stimulation, only three
(donor nos. 3, 6, and 7) increased their E75-specific lysis at
restimulation. In two of three donors (donors 3 and 6), specific E75
recognition was borderline after restimulation. E75-specific cytolysis
was observed at the fourth stimulation with DC-E75 in these two donors
(data not shown). In the other five donors that were both primed and
restimulated with E75 but failed to show specific CTL activity,
additional restimulations were not attempted, because of the low levels
of recognition of T2-E75 at restimulation compared with nonspecific
lysis. We rationalized that the five nonresponders will require a
minimum of four and even more restimulations to possibly elicit
E75-specific CTLs. Thus, even if cytolytic activity would be detected
after four to five stimulations, this finding would also support the
hypothesis of weak E75 immunogenicity in these individuals.
Recognition of Tumor Cells by E75-primed CTLs.
To address the question of whether E75-primed CTLs from healthy donors
recognized endogenously presented epitopes, CTLs generated from donors
1, 3, 5, 6, and 7, which showed peptide-specific lytic activity, were
tested for their ability to lyse the tumor line SKOV3.A2 and its
A2- counterpart, SKOV3. Except for HLA-A2, all
other histocompatibility Ags on SKOV3 and SKOV3.A2 are identical. To
verify that the responses are E75 specific, cold-target inhibition
experiments using unlabeled T2-E75 as specific target and T2-NP as
negative control were performed in parallel. T2 express only HLA-A2 and
low levels of HLA-B5. The results are summarized in Table 1
.
Peptide-specific CTLs from donors 6 and 7 did not show specific
recognition of SKOV3.A2 tumor. However, E75-specific CTLs from donors
1, 3, and 5 recognized endogenous E75. These donors do not express
HLA-B5, suggesting that E75 was presented by HLA-A2. It should be
mentioned that E75-specific CTLs were induced in donors 1 and 5 at
priming with DC-E75, whereas in donor 3, expression of this cytolytic
activity required four stimulations with DC-E75. These results
indicated that of 10 healthy donors tested, only three (33%) responded
by inducing CTLs that specifically recognized tumor cells. This
percentage was higher in the group of responders with peptide-specific
CTLs (three of five; 60%). The results with donor 5 are shown in Fig. 6
, A and B. Both
E75-primed cultures from donor 5 lysed SKOV3.A2 better than SKOV3,
suggesting that they recognize an epitope associated with HLA-A2. To
address whether these cultures recognized an endogenous presented
epitope similar to E75, we performed cold-target inhibition
experiments. The results in Fig. 6
C show that T2-E75
significantly inhibited by >50% recognition of SKOV3.A2 by CTLs from
donor 5 compared with control T2-NP, which expressed HLA-A2. This
inhibition was peptide specific because it was not observed with the
control peptide E71 pulsed on T2, suggesting that some E75-primed CTLs
recognized an endogenously presented epitope, but these cells are not
the majority in the effector population. Similar results were obtained
with E75-primed cells from donors 1 (Fig. 6
C) and 3 (not
shown). However, the levels of cold-target inhibition were lower and
ranged between 20 and 25% in two separate experiments. This suggested
that only a subpopulation of E75-induced CTLs recognize endogenous
generated epitopes. Thus, successful induction of E75-specific CTL
activity at priming with E75 using DCs as APCs and inflammatory
cytokine support appears to be dependent of additional factors other
than the nature of APCs and B7 costimulation.
|
In this study, we investigated the ability of the HER-2 peptide
E75 to prime E75-specific cytolytic activity in healthy donors when
presented on autologous DCs. We found that only 2 of the 10
HLA-A2+ healthy donors tested responded by
induction of E75-specific cytolytic activity at priming. This was
confirmed in replicated experiments performed over time, and the use of
various cytokine combinations IL-2+IL-12, IL-2 + TNF- + IL-12, or
preculture in IL-2, preculture in IL-2 +
RANTES.5
These
results indicated that E75-specific or cross-reactive T cells endowed
with cytolytic activity can be elicited at priming in only a fraction
of healthy donors (20%) but induced in an additional 30%. Of
interest, E75-primed CTLs from these two donors recognized E75
presented on tumor cells, because their activity was inhibited in
cold-target inhibition assays.
Two cytokines, TNF- and IL-12, were used to potentiate E75-specific
CTL induction. TNF- has been described to increase Ag uptake and
presentation by DCs (13)
and to potentiate CTL generation
in animal models (17)
. IL-12 has also been described to
potentiate CTL induction and cytolytic activity (18
, 19)
. TNF- and IL-12 increased the levels of cytolytic
activity in responders but had no effect in nonresponders. This
suggests that these cytokines are not essential for priming of
E75-specific CTL activity. Of interest, in the responders, induction of
E75-specific cytolytic activity was inhibited by B7.1, suggesting a
requirement for costimulation in the induction of cytolytic activity as
described (20)
for other Ag raising the possibility of DC-E75 primed
naïve T cells.
Induction of E75-specific cytolytic activity at priming with peptide observed with two healthy donors is of interest in evaluating the potential of this epitope for tumor-specific CTL induction and cancer vaccine development. In human tumor systems (most instances), priming with peptide required several repeated stimulations of healthy donor PBMCs before specific cytolysis was detected. CTLs specific for tyrosinase 369–377 peptide were detected in four of five healthy donors after three restimulations with peptide (21) . Peptide-specific CTLs were induced in healthy donors using DCs and peptides from gp100, tyrosinase, and MART-1/MelanA. Detection of CTL activity required three to four cycles of stimulation (22) . Similarly, presentation of MART-1 by DCs transduced with an adenoviral vector construct carrying the MART-1 gene required three stimulations for induction of specific cytolysis (23) , although in some donors, specific cytolytic effectors were detectable 7 days after priming (24) . In contrast, in another study, MART-1 (27, 28, 29, 30, 31, 32, 33, 34, 35) -specific cytolytic function could be induced in a nonresponder only, using APCs infected with rVVs expressing rVV-B71/2 + peptide, or rVV-B7.1 + MART-1 and restimulated with peptide, but not by peptide stimulation only (25 , 26) . Thus, the potency of E75 to induce cytolytic function in healthy donors appears similar to that of the MART-1 peptide 27–35.
Few studies have investigated the frequency of responses to tumor Ags in healthy donors at priming and restimulation or the consistency of these responses for an individual. This aspect is important because of its implications for protective vaccination in healthy donors or ovarian, breast, and prostate cancer patients without evidence of disease. In one extensive study, Marincola et al. (10) found that after several stimulations with MART-1 (27, 28, 29, 30, 31, 32, 33, 34, 35) , five of nine healthy donors (56%) responded by induction of specific cytolytic effectors. Only one to two donors showed weak activation of cytolysis at priming. In an independent study, 4 of 16 healthy donors (25%) responded to MART-1 (27, 28, 29, 30, 31, 32, 33, 34, 35) after two stimulations (27) . Similarly, anti-p53 (264–272) cytolytic effectors were generated from 2 of 5 healthy donors (40%) after several restimulations with peptide-pulsed DCs (28) , whereas anti-gp100 CTLs were elicited in 1 of 10 healthy donors at priming (29) .
In contrast with melanoma Ag, the immunogenicity of which has been repeatedly investigated in extensive studies, the experience with E75 is limited. Similar to the MART-1 peptide, E75 activated rapid cytokine secretion from cultured ovarian tumor-infiltrating lymphocytes or CTL lines (30 , 31) and activated cytolysis in tumor-associated lymphocytes (32) . Freshly isolated PBMCs from cancer patients that were not vaccinated with peptide rapidly responded to E75 and to another HER-2-epitope, GP2 (33) , in a similar fashion as melanoma patients to MART-1, by induction of specific cytolysis and tumor recognition (34) . The ease by which E75- and GP2-specific cytolytic activities were induced in patients suggested that E75 and GP2 reactivated effector/memory CTLs rather than primed naïve cells (34) . Our results showing 2 of 10 responders at priming (20%) and 5 of 10 responders at restimulation (50%) indicate that E75 is similar to MART-1 (27, 28, 29, 30, 31, 32, 33, 34, 35) , tyrosinase (369–377), and p53 (264–272) in its ability to activate cytolysis in randomly selected healthy donors.
It is possible that E75 cannot elicit a complete response in all donors
during PBMC priming. Our recent studies on cytokine responses by
E75-primed PBMCs in healthy donors show that E75 rapidly activated
specific IFN-
release in five of six healthy donors, an effect that
was enhanced by IL-12 (36)
. Although the same donors were used in these
studies and IL-12 was used in parallel experiments, we could not
observe a similar effect with respect to induction of cytolysis. Thus,
a complete response (cytokines and cytolysis) was observed only in two
donors of the six tested. This suggests that E75 may act as a partial
agonist. In support of this possibility, Zaks and Rosenberg
(1)
reported recently that E75 vaccination in incomplete
freund adjuvant of four cancer patients led to a
peptide-specific response at restimulation in all patients (cytolysis
and IFN-). T cells from two of three E75-vaccinated patients
recognized, occasionally, tumor cells by specific IFN- secretion but
failed to show specific tumor lysis (1)
. Similarly,
tyrosinase 369–377-specific CTLs from two of four responders failed to
recognize tyrosinase-expressing tumors (21)
. Preliminary
results from a vaccine trial in breast and ovarian cancer patients
indicate that PBMCs from only two of six E75-vaccinated patients (33%)
stimulated in vitro elicited specific CTL activity against
peptide and specific tumor recognition, although all responded to E75
by specific IFN- induction (2)
.
The similar response rates for induction of cytolysis in healthy donors to in vitro tumor peptide vaccination raise a number of questions about the application of this approach:
(a) Why, regardless of the Ag used, only 10–20% of healthy donors respond at priming, and only 40–50% at restimulation with peptide? One possibility is that healthy donors have different precursor frequencies for the CTL epitopes, and repeated peptide stimulation (three to four times) is not sufficient to expand the effector population to sufficiently high numbers to detect CTL activity. Thus, one approach is to continue repeated vaccinations and deliver exogenous help by helper peptides and cytokines until such CTL responses are elicited (9 , 35) .
(b) If the 10–20% of donors that respond to peptide priming have higher pCTL frequencies for E75/MART-1 than nonresponders, then what is the reason for this increased frequency? It is tempting to speculate that local inflammatory conditions and cross-reactive priming may activate CTL precursors, such as in donor 9, and these precursors become tolerized.
(c) If pCTL frequency is similarly low in all healthy
donors, then why do some respond better than others? One possibility is
that discrete changes in HLA-A2 attributable to HLA-A2 polymorphism may
lead to a more immunogenic E75 in some donors. In support of this
possibility, Maurer et al. (37)
demonstrated
that mutated HLA-A2 in position 97 can segregate MART-1
(27, 28, 29, 30, 31, 32, 33, 34, 35)
-induced cytolysis from cytokine production. Furthermore, the
pool of E75 precursors may expand or contract over time because of
different environmental factors (38
, 39)
. This may be
supported by the fact that the responders showed E75-specific CTL
activity frequently, whereas some of the nonresponders showed activity
only occasionally in the majority of independently performed
experiments. Additional studies using carboxyfluorescein acetate
to determine cell division, E75-tetramers to determine the frequency of
E75-specific CTLs, and intracellular IFN- staining are required to
distinguish among these possibilities.
Increased HLA-A2 binding affinity by COOH-terminal modification was able to enhance tumor Ag immunogenicity (in some instances), as shown in our previous studies (15) and by other investigators using the melanoma Ag gp100 (40 , 41) . However, increased HLA-A2 binding affinity does not always predict a higher T-cell receptor signaling or a complete T-cell activation (reviewed in Ref. 42 ). In some of the reported cases, CTLs induced by higher HLA-A2 affinity binding variant showed low affinity for the tumor cells (15) , and more recently (40 , 41) , some reports have suggested that they preferentially targeted tumors expressing high numbers of the epitope. Thus, novel immunogens need to be designed with emphasis on modifications in the Ag that would induce high rates of proliferation and select responders of high cytolytic activity, i.e., high catalytic activity as demonstrated by enzymes. Because restimulations may induce apoptosis, it remains to be seen whether tumor-specific CTL expansion would require several agonists, each being capable of activating one effector function at a time. At the present, the results of this study demonstrated that cytolytic effectors to an epitope on HER-2, which is overexpressed on the majority of epithelial tumors, can be elicited in a fraction of healthy individuals at priming, and in nonresponders, the precursors were not tolerized. Because these studies were performed with 10 donors and substantiated in multiple replicate induction experiments using the same stimulation system, this suggests that this stimulation system may be identifying individuals that will respond to vaccine with a tumor Ag. This may have implications for preventative vaccination in high-risk individuals.
ACKNOWLEDGMENTS
We thank the Blood Bank personnel for donating blood repeatedly for this project.
FOOTNOTES
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 18 U.S.C. Section 1734 solely to indicate this fact.
1 Supported in part by Research Grant
DAMD17-97-1-7098. Peptide synthesis was supported in part by Core Grant
CA 16672. 
2 Present address: Department of Surgery,
Uniformed Armed Services, University of the Health Sciences, Bethesda,
MD 20814.
3 To whom requests for reprints should be
addressed, at The University of Texas M. D. Anderson Cancer Center,
Department of Gynecologic Oncology, 1515 Holcombe Boulevard, Box 67,
Houston, TX 77030. Phone: (713) 792-2849; Fax: (713) 792-7586.
4 The abbreviations used are: Ag, antigen; DC,
dendritic cell; HER-2, HER-2/neu proto-oncogene; HS,
human serum; NP, no peptide; APC, antigen-presenting cell; PBMC,
peripheral blood mononuclear cell; GM-CSF,
granulocyte/macrophage-colony stimulating factor; IL, interleukin; mAb,
monoclonal antibody; TNF, tumor necrosis factor; LU, lytic unit(s);
rVV, recombinant vaccinia vector.
5 T. V. Lee and C. G. Ioannides, preliminary
data.
Received 4/17/00; revised 6/30/00; accepted 8/24/00.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  L. C. Benavides, J. D. Gates, M. G. Carmichael, R. Patel, J. P. Holmes, M. T. Hueman, E. A. Mittendorf, D. Craig, A. Stojadinovic, S. Ponniah, et al. The Impact of HER2/neu Expression Level on Response to the E75 Vaccine: From U.S. Military Cancer Institute Clinical Trials Group Study I-01 and I-02 Clin. Cancer Res., April 15, 2009; 15(8): 2895 - 2904. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. OKARVI and I. A. JAMMAZ Design, Synthesis, Radiolabeling and In Vitro and In Vivo Characterization of Tumor-antigen- and Antibody-derived Peptides for the Detection of Breast Cancer Anticancer Res, April 1, 2009; 29(4): 1399 - 1409. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. LI, S. MATSUEDA, C. L. EFFERSON, N. TSUDA, K. KAWANO, H. GAO, G. E. PEOPLES, and C. G. IOANNIDES Distinct Patient Responses to Activation of T-cells by Free HER-2, G89 (777-789) and Protected LRMK-linked HER-2, {AE-39 [p776 (Ava-774-788)], AE-47 [(Ava-776-788)] and AE-37[p776 (774-788)]} Peptides Could Lead to Development of Personalized Cancer Vaccines Anticancer Res, January 1, 2009; 29(1): 41 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Conrad, K. Gebhard, H. Kronig, J. Neudorfer, D. H. Busch, C. Peschel, and H. Bernhard CTLs Directed against HER2 Specifically Cross-React with HER3 and HER4 J. Immunol., June 15, 2008; 180(12): 8135 - 8145. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Mittendorf, C. E. Storrer, C. D. Shriver, S. Ponniah, and G. E. Peoples Investigating the Combination of Trastuzumab and HER2/neu Peptide Vaccines for the Treatment of Breast Cancer Ann. Surg. Oncol., August 1, 2006; 13(8): 1085 - 1098. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Efferson, N. Tsuda, K. Kawano, E. Nistal-Villan, S. Sellappan, D. Yu, J. L. Murray, A. Garcia-Sastre, and C. G. Ioannides Prostate Tumor Cells Infected with a Recombinant Influenza Virus Expressing a Truncated NS1 Protein Activate Cytolytic CD8+ Cells To Recognize Noninfected Tumor Cells J. Virol., January 1, 2006; 80(1): 383 - 394. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Peoples, J. M. Gurney, M. T. Hueman, M. M. Woll, G. B. Ryan, C. E. Storrer, C. Fisher, C. D. Shriver, C. G. Ioannides, and S. Ponniah Clinical Trial Results of a HER2/neu (E75) Vaccine to Prevent Recurrence in High-Risk Breast Cancer Patients J. Clin. Oncol., October 20, 2005; 23(30): 7536 - 7545. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kawano, C. L. Efferson, G. E. Peoples, D. Carter, N. Tsuda, J. L. Murray, and C. G. Ioannides Sensitivity of Undifferentiated, High-TCR Density CD8+ Cells to Methylene Groups Appended to Tumor Antigen Determines Their Differentiation or Death Cancer Res., April 1, 2005; 65(7): 2930 - 2937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Efferson, J. Schickli, B. K. Ko, K. Kawano, S. Mouzi, P. Palese, A. Garcia-Sastre, and C. G. Ioannides Activation of Tumor Antigen-Specific Cytotoxic T Lymphocytes (CTLs) by Human Dendritic Cells Infected with an Attenuated Influenza A Virus Expressing a CTL Epitope Derived from the HER-2/neu Proto-Oncogene J. Virol., July 1, 2003; 77(13): 7411 - 7424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Scheffold, M. Kornacker, Y. C. Scheffold, C. H. Contag, and R. S. Negrin Visualization of Effective Tumor Targeting by CD8+ Natural Killer T Cells Redirected with Bispecific Antibody F(ab')2HER2xCD3 Cancer Res., October 15, 2002; 62(20): 5785 - 5791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Castilleja, D. Carter, C. L. Efferson, N. E. Ward, K. Kawano, B. Fisk, A. P. Kudelka, D. M. Gershenson, J. L. Murray, C. A. O'Brian, et al. Induction of Tumor-Reactive CTL by C-Side Chain Variants of the CTL Epitope HER-2/neu Protooncogene (369-377) Selected by Molecular Modeling of the Peptide: HLA-A2 Complex J. Immunol., October 1, 2002; 169(7): 3545 - 3554. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, 
) of 
). D132 was used to stabilize HLA-A2 at comparable levels
with E75, thus minimizing the natural killer-derived background lysis.
* and **, P < 0.01 and <0.05 respectively,
compared with the lysis of T2-D132.
