| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Surgery, University of California and San Francisco Veterans Affairs Medical Center, San Francisco, CA 94121
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Due to the ability of IL-12 to target, among other cells of the immune
system, NK, NKT, and T cells, the antitumor activities of IL-12 have
been demonstrated in both established and nonestablished tumor models.
IL-12-induced antitumor activity in nonestablished tumor models is
mediated mainly by NK and NKT cells, and conventional T cells are not
required (3, 4, 5). Because both immunogenic (FBL-3,
leukemia) and nonimmunogenic (Lewis lung carcinoma (LLC) and B16
melanoma) tumors were found to respond to IL-12 treatment equally in
these nonestablished tumor models, tumor immunogenicity does not seem
to play a significant role in the first type of IL-12-induced antitumor
response. In contrast, unlike NK and NKT models, in which the antitumor
effect of IL-12 affects the development of newly inoculated tumor
cells, the second type of IL-12-mediated antitumor effect is found in
established s.c. tumor models (2, 6, 7, 8, 9, 10). In this second
type of IL-12-induced tumor rejection, T cells (6, 11) and
IFN-
(2, 6, 12) are essential for tumor rejection.
Clearly different from the NKT-mediated antitumor response, rejection
of established large nonimmunogenic tumors has not been reported.
In our previous report, we have shown that a single dose (100–125 mg/kg) of Cy followed by three injections of IL-12 results in complete eradication of long-term (3–4 wk) established, large (15–20 mm in diameter) s.c. tumors that are resistant to treatment with either Cy or IL-12 alone (2). Although the antitumor effect of Cy+IL-12 is much stronger than that of IL-12 alone, it is still based on a mechanism similar to that of IL-12, and the role of Cy is likely to be a potentiating agent through an as yet unknown mechanism (2). Because most previous antitumor studies using the MCA207 tumor model use experimental lung metastases to measure the antitumor efficacy (13, 14, 15, 16), it remains a possibility that the Cy+IL-12-mediated tumor regression is limited to certain anatomic sites but is not effective against pulmonary metastases. In addition, even if superior antitumor efficacy by Cy+IL-12 can be demonstrated in various anatomic sites of the MCA207 tumor model, it is critical to know whether similar effects of Cy+IL-12 can be obtained in other tumor models. In the first part of this study, we tested the antitumor activity of Cy+IL-12 in a number of tumor models and under different tumor locations. The results indicate that Cy+IL-12 is highly effective in eradicating established tumors in all anatomic sites in immunogenic tumor models. In contrast, Cy+IL-12 fails to induce regression in established palpable nonimmunogenic tumors. In the second part of the paper, we describe experiments leading to the identification of tumor rejection conditions that explain this differential response to IL-12-based therapy by immunogenic and nonimmunogenic tumors. Findings from this study are critical to our understanding of the dramatic antitumor effects by Cy+IL-12 and bear clinical relevance in identifying potential responders to IL-12-based immunotherapy.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Murine MCA203, MCA205, and MCA207 sarcomas, each of which is a methylcholanthrene-induced transplantable tumor in the C57BL/6 strain of mice (17), were obtained from Dr. J. Yang (Surgery Branch, National Cancer Institute, Bethesda, MD). Sa1 is a chemically induced sarcoma from an A/J mouse. B16 is a spontaneously developed melanoma from a C57BL/6 mouse. LLC is a spontaneously developed tumor from a C57BL/6 mouse. Pan02 is a chemically induced pancreatic carcinoma from a C57BL/6 mouse. These tumors were obtained from Division of Cancer Treatment Tumor Depository at Biological Testing Branch, Division of Cancer Treatment, National Cancer Institute, National Institutes of Health (Frederick, MD). All tumor cells were maintained by cell culture from in vivo harvested tumor implants in RPMI 1640 medium supplemented with 10% heat-inactivated FCS, 2 mM glutamine, 100 µg/ml streptomycin, 100 IU/ml penicillin, and 5 x 10-5 M 2-ME. When inoculated in the s.c. position, tumor development and growth were assessed by measurement of perpendicular tumor diameters. Cure is defined as complete tumor regression following treatment and the absence of recurrent tumor during follow-up period (3–6 mo). All normal C57BL/6 and A/J x C57BL/6 (AB6F1) mice were obtained from Biological Testing Branch, National Cancer Institute, National Institutes of Health. All mutant and gene knockout mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Experiments were performed with 8- to 12-wk-old female mice.
In vivo treatment models
In s.c. tumor models, 5 x 105 tumor cells in 0.2 ml of saline were injected s.c. on the flank of syngeneic or semisyngeneic F1 mice. Tumor size was assessed with calipers. Recombinant murine IL-12 (Genetics Institute, Cambridge, MA) was administered i.p. at a dose of 200 ng in 0.5 ml of 1% mouse serum in saline given once every other day for three doses. Cy+IL-12 treatment was composed of a single i.p. injection of 3 mg (120 mg/kg) of Cy (Sigma-Aldrich, St. Louis, MO) in 0.5 ml of saline followed 3–4 days later by a course of IL-12 as described above. In some cases, additional single-dose IL-12 injection was given weekly following the initial IL-12 treatment during tumor regression.
In the experimental lung metastases model, 5 x 105 MCA207 tumor cells in 0.5 ml of saline were injected i.v. via tail vein. Cy+IL-12 treatment, same as described for the s.c. model above, was initiated on the days described in the text. Survival of animals was followed by daily inspection. Visualization of pretreatment tumor burden in the lung was conducted by injection of india ink into the lungs of euthanized animals followed by bleaching as described before (13).
In the peritoneal sarcomatosis model, 5 x 105 MCA207 tumor cells in 0.2 ml of saline were injected i.p. Cy+IL-12 treatment, as described for the s.c. model above, was initiated on the day described in the text. The survival of the animals was followed by daily inspection. Pretreatment tumor burden was determined by inspection and photography of the peritoneal cavity of euthanized animals.
Concomitant and prophylactic immunity test
Concomitant immunity tests were conducted as described previously (18, 19). Briefly, s.c. tumors were first established in naive mice in one flank with 5 x 105 tumor cells. At the indicated time points after first tumor inoculation, another inoculation of 5 x 105 tumor cells was given to naive or tumor-bearing mice in the contralateral flank. The development of tumors from the second inoculation in both naive and tumor-bearing mice was assessed. Protective immunity was tested by immunizing naive mice with 1 x 106 irradiated (5000 rad) tumor cells s.c. once a week for two times followed by challenge with 5 x 105 live tumor cells on the opposite flank 1 wk after the second immunization. Naive mice were used as control at the time of live tumor challenge.
Adoptive cell transfer model
TCR
gene knockout mice were used as recipients of
adoptively transferred spleen and T cells from various donor sources.
TCR knockout mice were first inoculated with 5 x
105 MCA207 tumor cells s.c. Fourteen days after
tumor establishment when most tumors were 7–12 mm in diameter,
tumor-bearing TCR mice received indicated numbers of spleen or
purified T cells from indicated donor mice via i.v. tail vein
injection. Two to 7 days after adoptive cell transfer, recipients were
treated with IL-12 (200 ng i.p. three times). The donor T cells were
isolated from single cell suspension of spleen cells by
anti-Thy1.2-conjugated magnetic beads (Miltenyi Biotec, Auburn,
CA). Five to ten million (5–10 x 106)
purified T cells (>93% CD3 positive) were used for each adoptive
transfer. Spleen cells were prepared by removing RBCs from single cell
suspension by osmolysis followed with extensive washing with saline.
Tumor-immune donor mice were generated by immunization of naive mice
twice with 0.5–1 x 106 irradiated (5000
rad) tumor cells s.c. The immunized mice were challenged with 5 x
105 live tumor cells 1 wk after the second
immunization. Only mice that resisted tumor challenge were used as
donors in adoptive transfer experiments 1–3 mo after rejection of
tumor rechallenge.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have previously demonstrated that large (established 3–4 wk)
s.c. MCA207 tumors can be completely eradicated by a single low dose of
Cy (125 mg/kg) followed by three injections of recombinant murine IL-12
(Cy+IL-12) (2). In subsequent experiments reported in this
work we found that Cy+IL-12 treatment was also highly effective against
MCA207 tumors established by the i.p. and i.v. routes. The antitumor
effects in these treatment models are truly clinically significant and
surpass other previously developed immunotherapies in this tumor model,
in that premorbid tumor-bearing mice can be cured by Cy+IL-12
treatment. Fig. 1
shows the pretreatment
tumor burden in mice following s.c., i.p., and i.v. tumor inoculation.
In the case of experimental lung metastases induced by i.v. tumor
inoculation, there has been no previously demonstrated therapy that is
curative in this tumor model when administered 14 days after tumor
inoculation when lung metastases are grossly visible, as shown in Fig. 1C. It is under these advanced tumor-bearing conditions that
we are able to demonstrate the curative effects of Cy+IL-12 treatment
(Table I). First, as we have reported in
our previous study (2), a single dose of Cy plus a short
course of IL-12 can cure long-term established large (17–25 mm) MCA207
s.c. tumors such as the ones established for 32 days shown in Fig. 1A. Treatment with either Cy or IL-12 alone inhibited tumor
growth only temporarily and no mice were cured (data not shown).
Second, mice bearing i.p. sarcomas died of tumor burden between 20 and
27 days, but treatment with Cy+IL-12 started on day 18 at a time of
massive i.p. tumor burden (Fig. 1B) cured all treated mice.
Even very late treatment started on day 20 cured three of five mice.
Finally, in the experimental lung metastases model, untreated mice died
of tumor burden between 21 and 31 days, corresponding to the varying
pulmonary tumor burden on day 14, as seen in Fig. 1C.
Nevertheless, all mice treated with Cy+IL-12 on days 7 and 14 were
cured.
|
|
Cy+IL-12 treatment was tested in a panel of other murine tumors,
each of which, except the Sa1 tumor, is derived in the C57BL/6
background like the MCA207 tumor. Although Cy+IL-12 was effective in
eradicating established s.c. tumors of Sa1, MCA105, MCA203, and MCA205,
it was ineffective against s.c. tumors of B16 melanoma, LLC, and Pan02
pancreatic carcinoma (Table II). For most
of these other tumors that respond to Cy+IL-12, the size of the tumor
that completely responds to therapy is smaller than that seen in the
MCA207 tumor model. These results show that the curative response to
Cy+IL-12 treatment is not limited to MCA207 tumor model alone, but
varies among different tumor models.
|
shows, there were already a few
CD4 and CD8 T cells present in day-21 MCA207 tumor of 13–17 mm before
treatment with Cy+IL-12. The infiltration by T cells and Mac-1-positive
macrophages increased dramatically following Cy+IL-12 treatment. In
contrast, s.c. day-14 LLC tumors of 10–14 mm had almost no sign of T
cells and macrophages before treatment with Cy+IL-12. However,
following Cy+IL-12 treatment there were a few CD4-positive cells and no
CD8 T cells and macrophages in the tumor. This vivid contrast of immune
infiltration between a responding and a nonresponding tumor has been
observed in several other responding and nonresponding tumor
models.
|
Why do some tumors respond to Cy+IL-12 so well and others not at
all? One suggestion from the list of responders and nonresponders to
Cy+IL-12 (Table II
) is that all responders are tumors that have been
characterized as immunogenic, whereas the nonresponders are tumors that
are usually described as poorly immunogenic or nonimmunogenic. An
immunogenic tumor is able to induce an antitumor immune response upon
establishment in normal, but not T cell-deficient, hosts. This T
cell-mediated antitumor immunity is usually not strong enough to
eliminate the primary tumor but is detectable by the rejection of a
secondary challenge with the homologous tumor cells in a classic
concomitant immunity test (18, 20). Using this criterion,
we could see that all of the responders to Cy+IL-12 treatment are
indeed immunogenic tumors. As Table III
shows, at 7–8 days following tumor establishment, mice bearing tumors
that respond to Cy+IL-12 treatment (MCA203, MCA205, MCA207, and Sal)
resisted a rechallenge with the same tumor cells, whereas the same
tumor inoculation resulted in 100% tumor take in control naive mice.
The resistance of rechallenge in three of the four tumor models (all
except for the Sal model) was found to be dependent on T cells, as it
was not observed in 
T cell-deficient mice. T cell-dependent
concomitant immunity in the Sa1 tumor model has already been described
in previous studies (19). In contrast, none of the three
tested nonresponders (LLC, B16, and Pan02) showed resistance to
rechallenge in a T cell-dependent manner following primary tumor
establishment. There was an inhibition of second tumor development by a
primary Lewis lung tumor. However, this inhibition was also seen in T
cell-deficient mice, indicating that the inhibition is not due to a T
cell-mediated mechanism, but may be caused by antiangiogenesis factor
secreted by the tumor itself (21). To test the
immunogenicity of various tumors without the interference by a primary
tumor, we conducted another test of passive immunization and
rechallenge. In this prophylactic immunity test, Lewis lung, together
with B16 and Pan02, behaved like nonimmunogenic tumors, in that
repeated immunization of naive mice with irradiated tumor cells failed
to protect the immunized mice from subsequent challenge with live tumor
cells. In contrast, immunization of naive mice with the three
immunogenic tumors MCA207, MCA205, and Sa1, showed 100, 80, and 100%
protection, respectively, against a subsequent live tumor
challenge.
|
Why are only immunogenic tumors able to respond to Cy+IL-12? The
major difference between an immunogenic and a nonimmunogenic tumor is
that the former induces a T cell-mediated antitumor response upon tumor
establishment whereas the latter does not. T cells from mice bearing
immunogenic tumors are thus tumor-sensitized. To study the role of T
cells, especially tumor-sensitized T cells, in tumor rejection induced
by IL-12 and Cy+IL-12, we tested tumor rejection by Cy+IL-12 in the
MCA207 tumor model in various T cell-deficient mice. Using athymic nude
mice that are defective in thymus-derived T cells, but not
extra-thymus-derived NKT cells (22), we found that tumor
growth was only transiently inhibited by Cy+IL-12 treatment, and the
curative effect of Cy+IL-12 against large s.c. MCA207 tumors was
abolished (Table IV). This is clearly in
contrast to the NKT-mediated antitumor response that is equally
functional in both normal and nude mice (3), and suggests
that thymus-derived T cells are essential for tumor rejection in this
model. Further analysis using specific TCR gene knockout mice showed
that the curative effects of Cy+IL-12 found in normal mice was
completely abolished in TCR knockout mice lacking the classic
T and NKT cells, but not in TCR
knockout mice lacking T cells
(Table IV). As expected, rejection of established MCA207 tumors was
also abolished in TCR double knockout mice lacking all T cells.
Among subsets of T cells, either CD4 or CD8 T cells alone were found to
be able to mediate complete tumor rejection (Table IV). This
observation is consistent with previous studies by others who also
reported that depletion of either CD4 or CD8 T cells using Abs did not
abolish the antitumor response by IL-12 in the MCA207 small tumor
model, but depletion of both T cell subsets eliminated the antitumor
effects (6). Finally, complete tumor rejection was
obtained in mice lacking
2-microglobulin (Table IV), which
affects the development of MHC I-dependent CD8 T cells and
CD1-dependent NKT cells (23, 24, 25). These results indicate
that rejection of established large tumors by Cy+IL-12 requires
conventional T cells of either the CD4 or CD8 subset.
|
knockout mice. For creating a tumor-bearing host
containing naive, but not tumor-sensitized, T cells, we transferred
naive spleen cells or Thy1.2+ T cells purified
from naive spleen into TCR knockout mice bearing established s.c.
MCA207 tumors. In parallel, we transferred spleen cells or purified
spleen Thy1.2+ T cells from tumor-immune donors
to tumor-bearing TCR knockout mice to create the situation of a
tumor-bearing host containing tumor-sensitized T cells. Because
splenocytes show very low NKT-mediated antitumor effects due to the
lack of this cell population (4) and the
Thy1.2+ T cells we have purified are
NK1.1- by staining (data not shown), the
reconstituted recipients with purified T cells or spleen cells remained
NKT deficient. Following adoptive cell transfer, the recipients were
treated with either saline or IL-12 or Cy+IL-12, and the sizes of
tumors were assessed. As Table V shows, T
cell-deficient MCA207 tumor-bearing mice reconstituted with naive
spleen cells or naive T cells did not respond to either IL-12 or
Cy+IL-12 treatment, in that none of the tumors in these recipients were
eradicated by IL-12 and Cy+IL-12 treatments. In contrast,
reconstitution of tumor-bearing T cell-deficient mice with spleen cells
or T cells from tumor-immune donors restored the response of
tumor-bearing mice to both IL-12 and Cy+IL-12 treatments. It should be
noted that complete tumor rejection was achieved in T cell-deficient
recipients of tumor-sensitized T cells treated with IL-12 alone (Table V). Because Cy is not necessary for tumor rejection in the adoptive
transfer model, we conducted future adoptive transfer experiments using
IL-12 treatment alone instead of Cy+IL-12.
|
shows. Thus all MCA207 tumors
progressed in T cell-deficient mice treated with IL-12 before receiving
T cells from tumor-immune donors. In contrast, in recipient mice that
received tumor-sensitized T cells before IL-12 treatment, all tumors
were completely eradicated. Furthermore, tumor rejection by IL-12 not
only requires tumor-sensitized T cell from tumor-immune donors, but the
T cells must be tumor-specific. In an experiment we transferred
T cells from MCA205 and MCA207 tumor-immune donors to either MCA205- or
MCA207-bearing recipients. All four of four MCA207 tumors in mice
receiving T cells from MCA207-immune donors and zero of four tumors in
mice receiving T cells from MCA205-immune donors were eradicated by
IL-12 following adoptive cell transfer. In the same experiment, three
of five MCA205 tumors in mice receiving T cells from MCA205-immune
donors and zero of five tumors in mice receiving T cells from
MCA207-immune donors were eradicated by IL-12 following adoptive cell
transfer. The combined results from these adoptive transfer experiments
show that pre-existing tumor-sensitized Ag-specific T cells are
essential for rejection of established large tumors by IL-12/Cy+IL-12
treatment.
|
, in which spleen T cells from normal
mice bearing early (day-12) or late (day-28) MCA207 tumors were
isolated and transferred into T cell-deficient mice bearing established
MCA207 tumors. It is shown in Table VI that T cells from mice bearing
early (day-12) MCA207 are able to prime T cell-deficient recipients for
tumor rejection following IL-12 treatment. Furthermore, when spleen
cells from mice bearing long-term established large tumors were tested
as donor cells in the adoptive transfer experiment, complete tumor
eradication in the recipients required Cy+IL-12 instead of IL-12 alone.
This is similar to the situation in normal mice bearing long-term
established large tumors in which treatment with Cy+IL-12 is better
than IL-12 alone.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is produced by
IL-12-activated NKT cells (5), it is not clear that it is
essential for the antitumor effects. In contrast, the second type of
IL-12-induced antitumor immune response, including the one described in
the current study, is demonstrated in established palpable tumor models
in which s.c. tumors are treated with IL-12 or Cy+IL-12 at 7–35 days
after tumor inoculation when tumor sizes are 4–25 mm in diameter
(2, 6, 7, 8, 9, 10). Unlike in the nonestablished tumor models,
complete tumor rejection (cure) is often achieved in responders only
when IL-12 treatment is initiated at least 7 days, but not 1–4 days,
after tumor establishment (8, 10, 27, 28). The antitumor
response in the established tumor model is dependent on T cells (Refs.
6 and 11 and Table IV) and IFN- (2, 6, 9). The effector mechanism in these models is not yet
elucidated, but is unlikely to be dependent on perforin, because rapid
and complete tumor rejection was found to take place in both normal and
perforin gene knockout mice (our unpublished results).
As the current study shows, the antitumor activity of Cy+IL-12, when
observed under the right conditions, is dramatic and clinically
relevant (Fig. 1 and Table I). In terms of efficacy, there is no other
previously described immunotherapy that is equal to the antitumor
effects of Cy+IL-12 in any of the responder tumor models. For example,
previous studies involving the use of MCA207 tumor model have been
mainly limited to adoptive T cell transfer (13, 14, 15) and
active immunization (16) with dendritic cells in
experimental lung metastasis model. In these experiments, antitumor
activity was detected only when the tumor burden was minimal at 3, but
not 7 or 10, days after tumor inoculation. Successful immunotherapy
treatments against established s.c. tumors are rare and are limited to
small (days 3–5) tumors (16). Similarly, a number of
previous studies using the Sa1 tumor model have demonstrated limited
therapeutic effects against established small s.c. tumors
(29). In contrast, treatment with Cy+IL-12 in both these
tumor models leads to complete eradication of advanced tumor burdens
established at various anatomic sites. The degree of antitumor efficacy
shown in the current study even surpasses those observed in most
previous reported studies with IL-12 alone (6, 27). This
is due to the inclusion of Cy before IL-12 treatment as an
immunopotentiating agent, as we have described in a recent study
(2). Some of the tumors used in the current study, such as
MCA203 and MCA205, responded only to Cy+IL-12, but not to IL-12 or Cy
alone even when treated at a stage of small (3–6 mm) tumor (our
unpublished observation). Although the mechanism by which Cy enhances
IL-12-induced antitumor response is still under investigation, our
previous study has shown that tumor rejection induced by both IL-12
alone and Cy+IL-12 is similar and is mediated by a Th1 response
(2). The difference between the responses activated by
IL-12 alone and Cy+IL-12 is likely at the level of response, in that
Cy+IL-12 activates a stronger Th1 response than IL-12 alone
does.
Despite the ability of Cy+IL-12 to activate a strong antitumor response
in responding tumors, the same treatment barely affects other tumors,
including the two nonimmunogenic tumors, LLC and B16 melanoma, that are
used in most studies of IL-12-induced, NKT-mediated antitumor response
in nonestablished tumor models (3, 4). What are the
conditions that render some established tumors responsive to Cy+IL-12
and others not at all? One important finding from the current study is
the demonstration that the presence of tumor-sensitized T cells at the
time of IL-12 treatment is essential for tumor rejection in these
models. This conclusion is supported by our adoptive transfer
experiments in which we created two situations differing in the
presence or absence of tumor-sensitized, but not total, T cells. If
IL-12 is able to activate an antitumor response from naive T cells that
is able to reject the established tumors, we would expect to see this
in T cell-deficient mice receiving naive T cells. Our data show that
this is not the case, in that tumors in recipients of naive T cells did
not regress in response to subsequent IL-12 treatment (Table V).
Furthermore, our study demonstrates that the T cells must be not only
tumor-sensitized, but also tumor-specific. It should be noted
that although eradication of 2-wk established MCA207 tumors of >10 mm
in diameter in normal mice requires combined treatment of Cy+IL-12
(2), IL-12 alone is sufficient to induce complete
eradication of 14-day established MCA207 tumors in the adoptive cell
transfer model. Some of the tumors in the adoptive transfer experiments
had progressed to >15 mm before exhibiting complete regression
following IL-12 treatment. This is consistent with our previous
hypothesis that the direct cytotoxicity of Cy to tumor cells is not
critical for the dramatic antitumor effect of Cy+IL-12, because large
tumors in the adoptive transfer experiments can be eradicated by IL-12
treatment alone. A possible explanation is that Cy is required in
normal mice to counteract a down-regulation of antitumor immunity that
cannot be overcome by IL-12 treatment alone. In the adoptive transfer
model, tumor-sensitized T cells were from mice immunized with
irradiated tumor cells. It is possible that the down-regulation of
antitumor immunity found in normal long-term tumor-bearing mice does
not develop in these mice. Furthermore, it does not seem to develop in
the T cell-deficient recipients. Therefore, there is no need for Cy
treatment following adoptive T cell transfer in the T cell-deficient
recipients. Indeed, when spleen cells from long-term tumor-bearing mice
were transferred into T cell-deficient recipients, tumor rejection
required the addition of Cy, because IL-12 alone was found ineffective
(Table VI). The presence of a down-regulation of antitumor immunity in
long-term tumor-bearing hosts is indicated by the loss of concomitant
immunity after 3 wk of tumor establishment. It seems that a suppression
of antitumor immunity is cotransferred with spleen cells.
One important implication coming from the requirement of
tumor-sensitized T cells for tumor rejection is that, without further
immune intervention, only immunogenic tumors will likely meet this
requirement. This is because only these tumors are able to induce a
host T cell response to the incipient tumor upon tumor establishment
(Table III) (19). It should be noted that although
sensitive in vitro assays have been used to detect host response to
tumor (30), we chose the in vivo assays of prophylactic
and concomitant immunity for two reasons. First, these assays have been
used to assign the immunogenicity of a tumor historically. Thus B16
melanoma has been referred to as poorly immunogenic or nonimmunogenic
by all investigators not because it does not induce any kind of
detectable host response to the tumor, but because it does not induce
the kind of response that is able to reject a subsequent tumor
challenge. Secondly, these assays are rather stringent in terms of
measuring antitumor response, and they seem to provide more meaningful
correlation between tumor immunogenicity and response to immunotherapy
in our study. The antitumor immunity induced by immunogenic tumors
rarely affects the progression of the incipient tumor but is detectable
by the ability of the immunity to reject a second tumor challenge in a
concomitant immunity test. The onset of this tumor-induced immunity
takes >3 days after tumor inoculation, because second tumor challenge
given 3 days after first tumor establishment is not rejected in the
concomitant immunity test in both MCA207 and Sa1 tumor models (our
unpublished results). Without further manipulation, this spontaneously
generated antitumor immunity dissipates after a period of 3 wk. Thus
second tumor challenge given to mice bearing 3-wk MCA207 tumors is no
longer rejected (our unpublished result). North et al.
(31) have attributed the dissipation of this spontaneous
antitumor primary response to negative regulation by suppressor cells.
In immunogenic tumor models, the effectiveness of treatment with IL-12
seems to correlate to this window of primary response. For example, it
has been repeatedly seen by several investigators that IL-12 treatment
initiated too early (day 1–3) in tumor establishment before the
establishment of the primary antitumor response is less effective than
that initiated later (7–10 days), at the peak of the primary response
(8, 10, 27, 28). This phenomenon may be explained by the
requirement of the pre-existing tumor-sensitized T cells for tumor
rejection as identified in the current study. Although treatment
initiated early after tumor establishment faces a lighter tumor burden,
the absence of an adequate number of tumor-sensitized T cells is such a
limiting factor at this early time that it does not allow the
activation of a tumor-specific T cell response by IL-12. In contrast,
IL-12 treatment initiated after the onset of tumor-sensitization of T
cells faces a larger tumor burden, but the presence of tumor-sensitized
T cells make it possible for a strong T cell response, resulting in
eradication of established tumors. When IL-12 treatment is further
delayed until the dissipation of the primary response after 3 wk of
tumor growth, the response to IL-12 alone by immunogenic tumors such as
MCA207 is again lost (2). Different from the early tumor
establishment situation, the presence of tumor-sensitized T cells in
late tumor-bearing hosts is preserved, most likely in the form of
resting memory T cells. Treatment of late tumor-bearing mice with
Cy+IL-12, but not IL-12 alone, seems to reactivate these memory cells
and result in a strong antitumor response.
Will the requirement of tumor-sensitized T cells explain why
nonimmunogenic tumors do not respond to IL-12/Cy+IL-12? On the one
hand, unlike immunogenic tumors such as MCA207, nonimmunogenic tumors
are unable to induce a strong antitumor host immune response upon
establishment of the tumor. This is not necessarily the result of a
lack of antigenicity, because introduction of well-defined viral
surface Ag in the nonimmunogenic tumors of LLC (32) and
B16 melanoma (33) did not make these tumors more
immunogenic. It seems that mice bearing nonimmunogenic tumors tend to
lack the levels and type of tumor-sensitized T cells required for a
response to IL-12/Cy+IL-12. This can be seen from immunohistochemical
analysis of nonresponder tumors (Fig. 2), which indicates that there is
no sign of a strong T cell response at the site of tumors both before
and after Cy+IL-12 treatment. On the other hand, it is not clear that
this lack of pre-existing immunity is the sole reason for the lack of
response to IL-12/Cy+IL-12 therapy by nonimmunogenic tumors. Gao et al.
(34) recently showed that some immunogenic tumors with
demonstrated presence of tumor-sensitized T cells still do not respond
to IL-12. They have attributed the failure to the lack of certain forms
of tumor stroma (35). Alternatively, we have seen evidence
in our preliminary studies to support the hypothesis that enhancing
host recognition of nonimmunogenic would directly contribute to a
better response of these tumors to Cy+IL-12. For example, by immunizing
mice with tumor vaccine made with tumor-derived heat shock proteins
(36), we were able to eradicate large (>10 mm) s.c. LLC
tumors in the immunized, but not naive, mice (our unpublished results).
In another study we found that the spontaneously derived BALB/c breast
carcinoma 4T1 is nonimmunogenic and refractory to Cy+IL-12 therapy in
normal BALB/c mice. However, in Stat6-deficient mice in which host
recognition of 4T1 is enhanced (37), 4T1 tumor behaves
like a typical immunogenic tumor and 14-day established 4T1 tumors were
completely eradicated by Cy+IL-12 therapy (our unpublished results).
Thus, the answer to the above question may not be a simple one, and may
differ among different tumor models.
Why are tumor-sensitized T cells necessary for IL-12-induced tumor
rejection? One likely explanation is that T cells are primary targets
of IL-12 during tumor rejection. Because IL-12R is selectively
expressed in activated, but not naive, T cells (38, 39),
targeting of T cells by IL-12 during tumor rejection requires that
these T cells be in activated state. Sensitization by tumor Ag before
the start of IL-12 treatment would satisfy this requirement. Consistent
with this view, we have observed in the current study that
tumor-sensitized T cells must be present before, but not after, IL-12
treatment in order for tumor rejection to occur (Fig. 3). This view is
further supported by a recent study by Iwasaki et al. (11)
in other IL-12-responding tumor models in which they showed a
correlation between spleen T cell response to IL-12 in vitro and tumor
response to IL-12 in vivo. Finally, if indeed tumor-sensitized T cells
are targets of IL-12 during tumor rejection, we would expect to see a
loss of IL-12-induced tumor rejection by T cell-deficient mice
receiving tumor-sensitized T cells from IL-12R knockout donor in an
adoptive transfer experiment. Results from our preliminary study
support this prediction (our unpublished results). However, the proof
of this hypothesis calls for an adoptive transfer experiment in which
tumor-sensitized T cells from normal mice transferred into T
cell-deficient and IL-12R knockout mice are able to mediate tumor
rejection following IL-12 treatment.
Despite the fact that IL-12 has been shown to possess significant antitumor activities in a large number of animal tumor models, clinical application of IL-12 in cancer patients has not met with great success. Results from our previous and current studies suggest that at least two factors are responsible. First, to eradicate clinically significant large tumor burden, T cell-mediated antitumor response is more effective than those mediated by NK and NKT cells. But the activation of such T cell-mediated antitumor responses by IL-12 requires a pre-existing immunity that is only present in a limited number of tumor-bearing hosts (patients). Second, clinical trials thus far were conducted with IL-12 alone. Even if IL-12 treatment is occasionally effective against established tumors in animal tumor models, its efficacy is still limited to small (<10 mm in diameter) tumors in most cases. Thus, even in potential responders, IL-12 alone may not be effective due to the late tumor-bearing states of these hosts (patients). Our repeated demonstration of better antitumor effects by Cy+IL-12 than by IL-12 alone provides one solution to the low efficacy of IL-12 alone in potential responders. However, the current study also shows that this potential curative treatment is limited only to hosts bearing immunogenic tumors. Thus better clinical response to IL-12/Cy+IL-12-based treatment will likely depend on accurate selection of potential responders and conversion of nonresponders. The latter may be achieved through various forms of immunization with tumor vaccines. Previous immunotherapy trials have indicated that either cytokine treatment or active immunization with tumor vaccine alone is not sufficient to induce complete rejection of long-term established large tumor burdens consistently. It would be interesting to see whether the enhancement of a pre-existing antitumor immunity through immunization followed by treatment with Cy+IL-12 will bring more success in the laboratory as well as in the clinic.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Kangla Tsung, Department of Surgery, San Francisco Veterans Affairs Medical Center, 4150 Clement Street, San Francisco, CA 94121. E-mail address: kangla1{at}itsa.ucsf.edu
3 Abbreviations used in this paper: Cy, cyclophosphamide; Cy+IL-12, the combination of Cy and IL-12; LLC, Lewis lung carcinoma.
Received for publication August 24, 2001. Accepted for publication October 15, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells activated by IL-12 as a major effector in inhibition of experimental tumor metastasis. J. Immunol. 156:3366.[Abstract]
14 NKT cells in IL-12-mediated rejection of tumors. Science 278:1623.14 NKT cells. Int. J. Cancer 91:523.[Medline]
production. J. Immunol. 153:1697.[Abstract]
production by anti-tumor T cells. Int. Immunol. 7:1135.-mediated tumor growth inhibition induced during tumor immunotherapy with rIL-12. Int. Immunol. 8:855. and tumor necrosis factor have a role in tumor regressions mediated by murine CD8+ tumor-infiltrating lymphocytes. J. Exp. Med. 173:647. secretion is associated with in vivo therapeutic efficacy of activated tumor-draining lymph node cells. Cancer Immunol. Immunother. 41:317.[Medline]
T cells in the lungs of euthymic and athymic mice. Immunology 88:82.[Medline]
2-microglobulin-dependent surface expression of functional mouse CD1.1. J. Exp. Med. 182:1913. production in the generalized Shwartzman reaction. J. Immunol. 160:3522.14 NKT cells. Proc. Natl. Acad. Sci. USA 95:5690.-involving host-tumor interaction. Int. J. Oncol. 16:805.[Medline]
2 subunit expression in developing T helper 1 (Th1) and Th2 cells. J. Exp. Med. 185:817.This article has been cited by other articles:
|
|
 |
|
  M. Robinson, B. Li, Y. Ge, D. Ko, S. Yendluri, T. Harding, M. VanRoey, K. R. Spindler, and K. Jooss Novel Immunocompetent Murine Tumor Model for Evaluation of Conditionally Replication-Competent (Oncolytic) Murine Adenoviral Vectors J. Virol., April 15, 2009; 83(8): 3450 - 3462. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. N. Rad, G. Pollara, S. M. A. Sohaib, C. Chiang, B. M. Chain, and D. R. Katz The Differential Influence of Allogeneic Tumor Cell Death via DNA Damage on Dendritic Cell Maturation and Antigen Presentation Cancer Res., August 15, 2003; 63(16): 5143 - 5150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Mandpe, K. Tsung, and J. A. Norton Cure of an Established Nonimmunogenic Tumor, SCC VII, With a Novel Interleukin 12-Based Immunotherapy Regimen in C3H Mice Arch Otolaryngol Head Neck Surg, July 1, 2003; 129(7): 786 - 792. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. Hess, N. K. Egilmez, N. Bailey, T. M. Anderson, E. Mathiowitz, S. H. Bernstein, and R. B. Bankert Human CD4+ T Cells Present Within the Microenvironment of Human Lung Tumors Are Mobilized by the Local and Sustained Release of IL-12 to Kill Tumors In Situ by Indirect Effects of IFN-{gamma} J. Immunol., January 1, 2003; 170(1): 400 - 412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. C. Hill, T. F. Conway Jr., M. S. Sabel, Y. S. Jong, E. Mathiowitz, R. B. Bankert, and N. K. Egilmez Cancer Immunotherapy with Interleukin 12 and Granulocyte-Macrophage Colony-stimulating Factor-encapsulated Microspheres: Coinduction of Innate and Adaptive Antitumor Immunity and Cure of Disseminated Disease Cancer Res., December 15, 2002; 62(24): 7254 - 7263. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Tsung, J. P. Dolan, Y. L. Tsung, and J. A. Norton Macrophages as Effector Cells in Interleukin 12-induced T Cell-dependent Tumor Rejection Cancer Res., September 1, 2002; 62(17): 5069 - 5075. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. G. Segal, N. C. Lee, Y. L. Tsung, J. A. Norton, and K. Tsung The Role of IFN-{gamma} in Rejection of Established Tumors by IL-12 : Source of Production and Target Cancer Res., August 15, 2002; 62(16): 4696 - 4703. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
