[Cancer Biology & Therapy 7:3, 1-8; March 2008]; ©2008 Landes Bioscience
1Cancer Biology & Therapy2008; Vol. 7 Issue 3
This manuscript has been published online, prior to printing.Once the issue is complete and page numbers have been assigned, the citation will change accordingly.
Both gemcitabine and synthetic double-stranded RNA (dsRNA)
are known to be pro-apoptotic and immuno-stimulatory (-modula-
tory). We sought to evaluate the extent to which a combination
therapy using gemcitabine and a synthetic dsRNA, polyinosine-
cytosine (poly(I:C)), would improve the resultant anti-tumor
activity. Using model lung and breast cancers in mice, we demon-
strated that combination treatment of tumor-bearing mice with
the poly(I:C) and gemcitabine synergistically delayed the tumor
growth and prolonged the survival of the mice. The combination
treatment also synergistically inhibited tumor cell growth in vitro
and promoted more tumor cells to undergo apoptosis in vivo.
Finally, the combination therapy generated a strong and durable
specific anti-tumor immune response, although the immune
response alone was unable to control the tumor growth after the
termination of the therapy. This approach represents a promising
therapy to improve the clinical outcomes for tumors sensitive to
both dsRNA and gemcitabine.
Cancer is one of the leading causes of death in the U.S.
Chemotherapy remains an important cancer treatment modality.
Traditionally, small cytotoxic molecules that activate only a single
killing mechanism are used. Combination therapy involves treating
patients with a number of different drugs simultaneously. The drugs
differ in their killing mechanisms. Gemcitabine is a nucleoside
analog used as chemotherapy in various carcinomas: non-small cell
lung cancers, pancreatic cancers, breast cancers and bladder cancers.
However, the clinical outcome of the current gemcitabine formula-
tion (Gemzar®) is rather limited. For example, although being the
drug of choice for pancreatic cancer, gemcitabine shows only slightly
increased response rate and median survival of patients. This weak
systemic activity was partially attributed to the very short half-life
of the gemcitabine in the plasma (8–17 min)1 because recent data
from animal studies indicated that the encapsulation of gemcitabine
into a simple liposome formulation significantly improved its in vivo
anti-tumor activity.2,3 Another approach to improve the outcome of
gemcitabine therapy is to utilize the commonly practiced combina-
A recent development is the use of synthetic dsRNA (e.g.,
polyinosine-cytosine, poly(I:C)) as a tumor chemotherapy agent.
Certain dsRNA molecules are potent inducer of type I interferons
(IFNs), which are anti-proliferative, pro-apoptotic and anti-angio-
genic.4,5 The IFN-inducing activity of dsRNA has been exploited
in numerous pre-clinical and clinical tumor therapy trials. More
importantly, dsRNA can also directly promote apoptosis through
mechanisms such as the intracellular dsRNA-dependent protein
kinase (PKR) and the 2,5-oligo A synthetase pathways,4 or by
directly interacting with the Toll-like receptor-3 (TLR3) on certain
tumor cell surface.6 Recently, it becomes evident that besides the
indirect anti-tumor activity of the dsRNA-induced IFNs, the dsRNA
itself can also be utilized to directly kill tumor cells.7,8 Thus, we
hypothesized that a combination therapy using gemcitabine and a
synthetic dsRNA, especially when delivered inside tumor cells, will
synergistically improve the resultant anti-tumor activity, as compared
to using either of them alone.
Moreover, dsRNA is also known to activate both innate and adap-
tive immune responses by interacting with TLR3 and the retinoic
acid-inducible gene-1 protein (RIG1).9,10 The type I IFNs induced
by dsRNA are immunostimulatory too.4,11,12 In addition, the effects
of the gemcitabine on the immune system are well-documented.13-16
The apoptotic bodies generated by gemcitabine are a good source of
tumor antigens.17 It was shown that gemcitabine selectively elimi-
nated splenic myeloid suppressor cells in tumor-bearing animals, but
increased the activity of tumor-specific CD8+ T cells and natural killer
(NK) cells.15 Also, gemcitabine selectively depleted B cells and was
>2-fold more potent in inhibiting B-lymphocyte proliferation than T-
lymphocyte proliferation.16 Data generated in patients with pancreatic
cancers also suggested that gemcitabine may promote the activation
Tumor chemo-immunotherapy using gemcitabine and a synthetic dsRNA
Uyen M. Le,1 Nijaporn Yanasarn,1 Christiane V. Löhr,2 Kay A. Fischer2 and Zhengrong Cui1,*
1Department of Pharmaceutical Sciences; College of Pharmacy; 2Department of Biomedical Sciences; College of Veterinary Medicine; Oregon State University; Corvallis,
Abbreviations: Poly(I:C) or pI:C, polyinosine-cytosine; dsRNA, double-stranded RNA; IFN, interferon; PKR, RNA-dependent
protein kinase; TLR, toll-like receptor; RIG1, retinoic acid-inducible gene 1; NK, natural killer; CTL, cytotoxic T lymphocytes; MTT,
3-(4,5-dimethylthiazol-z-yl)-2,5-diphenyltetrazolium bromide; HPV, human papillomavirus; pI:C/LP, poly(I:C)-lipoplexes; DOTAP, 1,2-
dioleoyl-3-trimethylamonium-propane; Fa, fraction of affected/killed cells; Fu, fraction of unaffected/live cells; D, dose; Dm, dose required
to produce the median effect; CI, combination index; DC, dendritic cell; Gem, gemcitabine
Key words: combination therapy, dsRNA, synergistic, apoptosis, immune response
*Correspondence to: Zhengrong Cui; Department of Pharmaceutical Sciences;
College of Pharmacy; Oregon State University; Corvallis, Oregon 97331 USA; Tel:
541.737.3255; Fax: 541.737.3999; Email: Zhengrong.email@example.com
Submitted: 11/08/07; Revised: 12/07/07; Accepted: 12/13/07
Previously published online as a Cancer Biology & Therapy E-publication:
www.landesbioscience.com Cancer Biology & Therapy2
of naïve T cells.13 A very recent article reported that
gemcitabine has a significant immuno-modulatory
activity in murine tumor models, independent of its
cytotoxic effects.14 Thus, we further hypothesized
that a combination therapy using gemcitabine and a
synthetic dsRNA will not only synergistically inhibit
the tumor growth, but also induce innate and specific
tumor-killing CTL immune responses.
In the present study, we first demonstrated that a
synthetic dsRNA poly(I:C) in the form of poly(I:C)-
cationic liposome lipoplexes was cytotoxic in a model
mouse lung tumor cell line and significantly inhibited
the growth of model lung and pancreatic tumors in
vivo. The lipoplexes were used to improve the delivery
of the poly(I:C) into tumor cells. We then showed that
the combined anti-tumor activity of the poly(I:C)-
lipoplexes and the gemcitabine was synergistic both in
vitro and in vivo in the model mouse lung tumor cells.
The combination therapy also generated a strong and
long-lasting tumor-specific CTL response, although
the CTL response alone did not control the tumor
growth after the combination treatment was termi-
nated. Finally, we showed that the same combination
therapy also synergistically inhibited the growth of a
model breast tumor in mice.
Poly(I:C) inhibited tumor growth in vitro and in
vivo. First, we confirmed the anti-tumor activity of
the poly(I:C) using the TC-1 tumor cells in culture.
Although ‘naked’ poly(I:C) did not inhibit the TC-1
cell growth at a concentration as high as 300 nM (data
not shown), the poly(I:C) in the form of the poly(I:C)-
lipoplexes was apparently cytotoxic, with an IC50 value
of 1.42 nM (Fig. 1A). The supernatant of the TC-1 cells cultured in
the presence of the poly(I:C)-lipoplexes also significantly inhibited
the growth of fresh TC-1 cells (Fig. 1B).
We then evaluated the poly(I:C)’s ability to inhibit the growth of
the TC-1 tumors in mice. Data in Figure 1C showed that tumors in
mice injected with the liposomes alone grew uncontrolled, and the
mice died ~20 days after the tumor implantation. The poly(I:C)-lipo-
plexes significantly delayed the tumor growth (Fig. 1C). However,
the tumor growth was delayed for only 7 days, thereby indicating that
the poly(I:C)-lipoplexes alone were not sufficiently effective. Finally,
when a similar experiment was carried out using a model pancreatic
tumor cells (Panc-02) in mice, the poly(I:C)-lipoplexes also signifi-
cantly extended the survival of the Panc-02 tumor-bearing mice
(Fig. 1D), and 2 of the 7 mice became tumor-free in the end. The
following combination therapy experiments were mainly carried out
using the TC-1 lung cancer model because the TC-1 cells grew much
faster in mice than the Panc-02 cells. Also, the poly(I:C)-lipoplexes
alone tended to be less effective in the TC-1 tumor model.
Combination therapy using the poly(I:C)-liposome lipoplexes
and gemcitabine synergistically further inhibited tumor growth in
vitro and in vivo. Shown in Figure 2A are the dose-effect plots when
the TC-1 cells in culture were treated with the poly(I:C)-lipoplexes,
gemcitabine, or their mixture. The IC50 values for the gemcitabine,
the pI:C/LP, and their mixture were 0.04 nM, 1.42 nM and 0.60
nM, respectively. Because the dose-effect plots in Figure 2A did not
indicate whether the activities of the gemcitabine and the poly(I:C)-
lipoplexes were mutually exclusive or non-exclusive, the combination
index (CI) was calculated on the basis of both mutually exclusive and
non-exclusive assumptions. When assumed mutually exclusive, at
high fraction affected (Fa) values, there was a marked synergism for
their combined effect (CI < 1) (Fig. 2B). When assumed mutually
non-exclusive, the combined effect was also synergistic, except that
an antagonism was likely at very high Fa values (Fig. 2B). Thus, the
combined tumor-inhibitory effect of the gemcitabine and the poly(I:
C)-lipoplexes was generally synergistic in cell culture.
In vivo, combination treatment with gemcitabine and poly(I:
C)-lipoplexes, dosing schedule shown in Figure 2C, significantly
further delayed the tumor growth (Fig. 2D) and prolonged the
survival of the mice (Fig. 2E), as compared to treatment with the
poly(I:C)-lipoplexes alone or the gemcitabine alone. In fact, the
time it took for the tumors in mice treated with PBS, poly(I:C)-
lipoplexes, gemcitabine, or the combination of poly(I:C)-lipoplexes
and gemcitabine to reach 800 mm3 (or 11.5–12.5 mm in diam-
eter) was 15.7 ± 0.7, 18.6 ± 1.7, 35.4 ± 0.9 and 43.2 ± 4.3 days,
respectively, and the combination treatment significantly delayed
the tumor growth as compared to the gemcitabine single treatment
Figure 1. Poly(I:C)-in-lipoplexes inhibited tumor growth in vitro and in vivo. (A) Dose-effect
plot of the % TC-1 cells alive as a function of the concentration of the pI:C in pI:C/LP (n =
3). (B) Indirect killing effect of the pI:C. The supernatant of TC-1 cells incubated with the pI:
C/LP inhibited the growth of fresh TC-1 cells (n = 3). The (*) indicates that the value of the
(+) pI:C/LP differs from that of the other two (ANOVA of 3, p = 0.001; fresh medium vs. (+)
pI:C/LP, p = 0.003; (+) pI:C/LP vs. (-) pI:C/LP, p = 0.003). (C) pI:C/LP delayed the growth
of TC-1 tumors in mice (n = 7). For example, on day 19, the tumors of the control was larger
than that of the pI:C/LP (p = 0.0002, t-test). (D) pI:C/LP prolonged the survival of Panc-02
tumor-bearing mice (p = 0.02, n = 7). Two of the 7 mice became tumor-free.
3 Cancer Biology & Therapy2008; Vol. 7 Issue 3
(p = 0.003). Interestingly, the tumors in mice received the combi-
nation treatment were suppressed for more than 30 days, and for
unknown reasons started to grow about 30 days after the tumor cell
injection (or 16 days after the last dosing). Also, the combined anti-
tumor activity was synergistic. For example, on days 7 and 19 after
the tumor cell injection, the tumor volume synergistic indexes were
1.63 and 2.06, respectively. Shown in Figure 2F are typical photo-
graphs of the tumors 17 days after the TC-1 cell injection.
The combination therapy enhanced the proportion of the tumor
cells undergoing apoptosis. Very few apoptotic cells were detected
in the tumors in mice left untreated (Fig. 3A, part a). Apoptosis was
apparent in the tumors in mice received the poly(I:C)-lipoplexes
(Fig. 3A, part b) or the gemcitabine alone (Fig. 3A, part c). The
combination treatment generated significantly more positive staining
than each of the single treatment alone (Fig. 3A, part d). In fact, the
apoptotic index of the combination therapy was more than 3-fold
higher than any of the single treatments alone (Fig. 3B).
Histology. Significant areas of necrosis (20–60%) were detected
in the large tumors (central or disseminated) in mice injected with
the sterile PBS. Infiltration of the tumors by mononuclear cells and
polymorphonuclear cells was rare and peritumoral (Fig. 4A and B).
Necrosis was also extensive (40–60%) in the tumors in mice received
the poly(I:C)-lipoplexes alone but mainly in the center of the tumors.
There was severe infiltration of mononuclear and polymorphonuclear
cells in the periphery of the tumors. Moreover, severe edema was also
apparent in these tumors (Fig. 4C and D). Necrosis was rare in the
tumors in mice received the gemcitabine alone. Cellular infiltration
and edema were rare to moderate in these tumors (Fig. 4E and F).
Finally, necrosis was random but extensive (60–80%) in tumors in
mice received the combination therapy. Cellular infiltration of the
tumors was rare, but located inside the tumors (Fig. 4G and H).
The combination therapy generated a specific tumor lytic
immune response, which alone was unable to control the tumor
growth. Data in Figure 5A showed that a TC-1 specific CTL
response was induced in tumor-bearing mice received the combi-
nation treatment. At the effector to target ratio of 50:1, a weak
target cell lysis was also detectable in mice injected with the poly(I:
C)-lipoplexes alone (data not shown). However, there was not any
detectable CTL activity in mice injected with the gemcitabine alone,
nor in mice injected with sterile PBS. In addition, when the 24JK
cells were used as the target, no significant CTL activity was detected
in any of the mice (data not shown), indicating that the tumor cell
lytic activity was specific against the TC-1 tumor cells.
To understand why the specific CTL activity in tumor-bearing
mice received the combination therapy failed to control the tumor
growth after the termination of the treatment (Fig. 2D), we evaluated
the durability of the CTL activity. As shown in Figure 5B, although a
TC-1 specific CTL activity was not detectable in mice 10 days after
the beginning of the combination therapy, a CTL response became
apparent 20 days after the beginning of the therapy, and was still
strong on days 30 and 40 (Fig. 5B). Thus, the combination therapy
had induced a relatively strong and long-last specific tumor lytic
immune response, but the response alone failed to effectively control
the tumor growth after the combination therapy was terminated.
Figure 2. The combination of poly(I:C)-in-lipoplexes and gemcitabine synergistically inhibited tumor growth in vitro and in vivo. (A) Dose-effect plots showing
the killing of TC-1 cells by gemcitabine, pI:C/LP, or their mixture (3290:1, pI:C/Gem). Fa is the fraction of cells affected/killed. Fu is the fraction of cells
un-effected. D is the dose in nM. (B) Combination index (CI) with respect to the Fa. (C) In vivo dosing schedule. (D) TC-1 tumor growth kinetics. The values
of the Gem and pI:C/LP+Gem differed starting from day 5. Data reported are mean ± S.E.M. (n = 6–7). (E) Survival of mice with TC-1 tumors (p = 0.004,
Gem vs. pI:C/LP+Gem). (F) Photographs of typical tumors in mice 17 days after tumor cell injection.
www.landesbioscience.com Cancer Biology & Therapy4
The combination therapy synergistically inhibited the growth
of breast cancer cells in a mouse model. The 410.1 breast cancer
cells were shown to be sensitive to both poly(I:C)-lipoplexes and
the gemcitabine, and the combination of them significantly further
inhibited the growth of the 410.4 breast cancer cells in culture (data
not shown). In vivo, a combination therapy using the poly(I:C)-lipo-
plexes and the gemcitabine also synergistically delayed the growth of
the 410.4 tumors (Fig. 6). On days 10, 19 and 35 after the tumor
cell injection, the synergistic indexes for the tumor volume were
1.68, 3.87 and 4.63, respectively. In fact, it took an average of 7.1
± 1.5, 11.4 ± 2.1, 12.0 ± 4.2 and 30.1 ± 3.9 days for the tumors in
mice treated with PBS, pI:C/LP, gemcitabine, or the combination of
pI:C/LP and gemcitabine to reach 25 mm3, respectively. Again, the
combination therapy delayed the tumor growth significantly (p = 2.3
x 10-8 vs. pI:C/LP, p = 2.3 x 10-6 vs. Gem).
In the present study, we showed that the treatment of model solid
tumors in mice with the poly(I:C)-lipoplexes and gemcitabine syner-
gistically inhibited the tumor growth and prolonged the survival
of the mice. The combination therapy also induced a strong and
durable tumor-specific CTL response, although the CTL response
alone was unable to effectively control the tumor growth after the
termination of the treatment. When fully optimized, this approach
may represent a promising modality to improve the clinical outcomes
of the therapy of cancers sensitive to both gemcitabine and synthetic
dsRNA. For example, a dsRNA therapy may be integrated into the
standard gemcitabine therapy to improve the resultant therapeutic
The anti-tumor activity of the poly(I:C) was known for decades. It
is known that dsRNA activates a number of pro-apoptotic processes,
including the PKR and the 2,5-oligo A pathways, both of which turn
Figure 3. The combination therapy promoted more tumor cells to undergo
apoptosis in vivo. (A) Micrographs of tumors stained against activated
Caspase-3 into red. (a) PBS, (b) pI:C/LP, (c) Gem, (d) pI:C/LP+Gem. (B)
Apoptotic index. Data reported are mean ± S.D. (n = 3). (*) indicates that
the value of the pI:C/LP+Gem differs from that of the others (ANOVA, p =
0.0004; ANOVA without pI:C/LP+Gem, p = 0.18). The values of the Gem
and the pI:C/LP alone were comparable (t-test, p = 0.84).
Figure 4. H & E micrographs. (A and B) PBS, (C and D) pI:C/LP, (E and F)
Gem, (G and H) pI:C/LP+Gem. (A, C, E and G) 4 x magnification; (B, D, F
and H) 40 x magnification.
5 Cancer Biology & Therapy2008; Vol. 7 Issue 3
off protein synthesis and ultimately lead to cell death.25
More recent data also showed that the direct binding of
the TLR3 on the surface of certain human breast cancer
cells with poly(I:C) also triggered apoptosis.6 In addi-
tion, dsRNA has indirect anti-tumor activities. Poly(I:
C) is a very potent inducer of type I IFNα/β,26 which
are known to have multiple anti-tumor mechanisms
(see review4). Firstly, they are anti-proliferative and
can affect all phases of the mitotic cell cycling, most
commonly with a blockade of the G1 phase.5 Secondly,
type I IFNs are pro-apoptotic and were shown to be
cytotoxic to malignant cells.27 In general IFN-induced
cell death occurs > 48 hours after treatment and can
be prevented by inhibitors of caspase-8 or caspase-3.27
Finally, type I IFNs also inhibit tumor angiogenesis.
Following treatment with IFNs, tumor endothelial
cells exhibit microvascular injury and necrosis.28
Starting from the end of 1960, there had been
numerous studies using synthetic dsRNA to inhibit
tumors in experimental animal models and in clinical trials. The
poly(I:C) was generally dosed systemically, and the anti-tumor
activity of the poly(I:C) was mainly attributed to its ability to induce
the production of type I IFNs, which then either kill tumor cells
or enhance the host immune response. However, the effects of the
systemic poly(I:C) on tumors are not consistent, and higher doses of
systemic poly(I:C) may generate unwanted effects.29-34 In contrast,
the local administration of poly(I:C) has been shown to be rather
more effective in suppressing tumor growth.7,31,35 For example, it
was shown that the growth of transplanted rat tumors was retarded
and in some cases completely suppressed when the tumor cells were
injected subcutaneously (s.c.) in admixture with poly(I:C). Systemic
treatment of the same tumor with the poly(I:C) failed to prevent
the progressive growth of a range of rat tumors.31 The total tumor
regression observed by Shir et al. after direct intratumoral injection
of tumor cell-targeting poly(I:C) is another example.7 Thus, the
direct cytotoxic effects of the poly(I:C) seemed to be more effective
in suppressing tumor growth than the indirect effects from the IFNs
induced by the poly(I:C), which prompted us to inject the poly(I:
C) locally in the present study. The cationic liposomes were used
as a carrier for the poly(I:C) to improve the uptake of the poly(I:C)
by the tumor cells. Only when delivered inside tumor cells, can the
poly(I:C) cause cell death directly by interacting with the intracel-
lular proteins such as the PKR and the 2,5-oligo A synthetase.25 It
needs to be emphasized that the peritumoral route was chosen in
the present study simply to prove the feasibility of this combina-
tion therapy. This local injection is feasible for a few tumors such as
melanoma and certain head and neck cancers, but may be clinically
difficult to practice for other tumors. Therefore, we are currently
focused on developing a delivery system to target dsRNA into tumor
cells after intravenous injection.
It was interesting that a relatively strong and durable tumor-
specific CTL immune response was induced by the combination
therapy (Fig. 5). The immunostimulatory activity of the poly(I:C)
was originally established in the 1960–1970’s.36 However, it was
not until recently that the TLR3 was identified as a receptor of
dsRNA,9 and interest in studying its immunostimulatory activity
was revived again. To summarize, dsRNA was shown to induce
the maturation of human DCs,37 to augment NK cell-mediated
cytotoxicity,38,39 to promote the survival of T cells.23,40,41 Moreover,
the cellular protein RIG-1 also senses intracellular dsRNA and
stimulate dsRNA-induced innate immune responses.10 Recently, it
was shown that mouse CD8a+ DCs were activated by virus-infected
cells, but not by uninfected cells.42 Similarly, the CD8α+ DCs were
activated by irradiated epithelial Vero cells pre-loaded with poly(I:
C), but not by irradiated Vero cells in the absence of poly(I:C).
The activation of DCs required the phagocytosis of infected cells
(or the irradiated cells) by the DCs, followed by TLR3 signaling.42
Immunization of mice with dead tumor cells with poly(I:C) trans-
fected inside also led to a stronger anti-tumor immune responses than
with the same dead tumor cells admixed with poly(I:C).43 Moreover,
the type I IFNs induced by dsRNA are also immunostimulatory.
Type I IFNs were shown to augment cytotoxic T cells, NK cells and
DCs and promote the cross-priming of CD8+ T cells by stimulating
the maturation of DCs.12 All these direct and indirect immuno-stim-
ulatory activities of the dsRNA, as well as the immuno-modulatory
effects of the gemcitabine mentioned early,14-16 may explain the
induction of the tumor-specific CTL response after the combination
therapy. We speculate that the poly(I:C) and the type I IFNs induced
by the poly(I:C) have likely promoted the dead or dying tumor cells
generated by the gemcitabine and the poly(I:C) to cross-prime tumor
cell-specific CD8+ T cells. The tumor cell apoptotic bodies generated
by gemcitabine were shown to be a good source of tumor antigens,17
and intratumoral or peritumoral (p.t.) injection of poly(I:C) was
also shown to generate some weak specific CD8+ T cell responses.44
However, a more interesting finding in this study is that the specific
tumor-killing immune response, being strong and durable when
assayed in vitro (Fig. 5), evidently failed to control the tumor cell
growth after the combination treatment was terminated (Fig. 2D).
Although it is unknown whether the tumor cell-specific T cells have
reached the tumor tissues or not, the H & E micrographs shown
in Figure 4 indicated that there were infiltration of the immune
cells into the center of the tumor tissues. Thus, it is possible that
these tumor cells that escaped the action of the poly(I:C) and the
gemcitabine were also resistant or became insensitive to the killing by
the specific immune cells. Detailed mechanisms to explain the lack of
Figure 5. Combination therapy using pI:C/LP and Gem induce tumor-specific CTL responses
(A) and the CTL response was still strong 40 days after the tumor cell implantation. (B) Data
in A and B were from two separate experiments. In B, the (x) and (O) were the values of
the negative and positive controls, respectively. The numbers in the parentheses (100:1 or
40:1) are the E to T ratios.
www.landesbioscience.comCancer Biology & Therapy6
effectiveness of the specific immune responses induced by the combi-
nation treatment will be further elucidated in future experiments.
However, the data in Figure 2D and Figure 5 did suggest that more
efforts may need to be directed towards “persuading” more tumor
cells to die from the direct cytotoxic effects of the poly(I:C) and the
gemcitabine. This may be achieved by actively targeting the chemi-
cals into tumor cells. On the other hand, the tumor-specific immune
responses induced by the combination of the poly(I:C)-lipoplexes
and the gemcitabine may have partially contributed to the observed
synergism of the in vivo anti-tumor activity. Another possible expla-
nation to the observed synergism, both in vivo and in vitro (Figs. 2
and 6), could be that the gemcitabine rendered the tumor cells more
sensitive to the killing by the poly(I:C) or vice versa. We will further
elucidate the mechanisms responsible for the observed synergism in
We speculate that the anti-tumor activity of the combination
therapy using the poly(I:C) and the gemcitabine may be further
improved by optimizing the dosing schedule. In the present study, we
dosed the poly(I:C) and the gemcitabine concurrently to the tumor-
bearing mice, reasoning that this schedule will expose the tumor
cells to both poly(I:C) and gemcitabine simultaneously. Since the
cytotoxic mechanisms of the poly(I:C) and gemcitabine differ from
each other, simultaneous exposure of the same cell to both chemicals
may result in a faster and more effective killing of the tumor cell. In
addition, the concurrent dosing of the gemcitabine and the poly(I:
C) may also allow the poly(I:C) to more effectively promote the
dead or dying tumor cells generated by the gemcitabine to prime
the induction of specific immune responses. However, this concur-
rent dosing schedule may not be optimal. In previous studies where
the therapy involved the gemcitabine and other immunostimulatory
molecules, the gemcitabine was dosed first, and the immunostimula-
tory molecules (e.g., CpG oligos or the anti-CD40 antibody FGK45)
were dosed after the gemcitabine application was completed.45,46
It was shown that this sequential dosing schedule was much more
effective than when they were dosed concurrently.46 It is possible
that if the tumor-bearing mice in our study were dosed with the
gemcitabine first followed by the poly(I:C)-lipoplexes, the resultant
anti-tumor activity would be stronger. More experiments will have to
be carried out to identify the more effective dosing schedule. Finally,
we completed only two cycles of combination dosing in the present
study (1 cycle per week). We expect that the anti-tumor activity of
the combination therapy can be further improved by increasing the
number of dosing cycles and/or by allowing more appropriate time
between each dosing cycle to allow the small fraction of unaffected
tumor cells to enter into an active growth stage and be more effec-
tively killed in the following dosing cycle.
Finally, the combination therapy also synergistically inhibited
the growth of a model breast tumor in mice, suggesting that this
approach is applicable for many tumors. Also, we would like to
point out that compared to the TC-1 cells, the 410.4 breast cancer
cells grew much more slowly in mice and tended to metastasize.
Moreover, the 410.4 cells are weakly immunogenic, while the TC-1
cells are strongly immunogenic. Thus, more experiments need to
be carried out to evaluate the extent to which the adaptive immune
response induced by the combination therapy was responsible to the
tumor growth delay observed in Figures 2D and 6.
In conclusion, we reported a novel combination tumor chemo-
therapy approach that synergistically improved the resultant
anti-tumor activity. The combination therapy increased the percent
of tumor cells undergoing apoptosis and promoted the induction of
a strong and durable tumor-specific immune response. After further
optimization, this approach may improve the clinical outcomes
for the therapy of the cancers sensitive to both gemcitabine and
dsRNA. A similar strategy may also be adopted to combine
synthetic dsRNA and other chemotherapy agents to more effectively
fight other cancers.
Materials and Methods
Materials and cell lines. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide), concanavalin A and mouse IL-2 were
from Sigma-Aldrich (St. Louis, MO). Poly(I:C) (or pI:C) was from GE
Healthcare (Piscataway, NJ). It was a duplex polymer composed of a
poly(I) strand (152–539 b) annealed to a poly(C) strand (319–1,305
b). GEMZAR® (gemcitabine hydrochloride) was purchased from
Eli Lilly (Indianapolis, IN). The E749-57 peptide (RAHYNIVTF)
was synthesized by the GenScript Corp. (Piscataway, NJ). TC-1
cells (ATCC # CRL-2785TM) were engineered by Dr. T.C. Wu’s
group in the Johns Hopkins University by transforming primary
lung cells from C57BL/6 mice with human papillomavirus (HPV)
type 16 E6 and E7 oncogenes and an activated H-ras.18 The TC-1
cells are also used by many researchers as model cervical cancer cells.
The Panc-02 cell line and the 410.4 cell line were kindly provided
by Dr. Joyce Solheim from the University of Nebraska and Dr. Fred
Miller in the Karmanos Cancer Institute (Detroit, MI), respectively.
The TC-1 and the Panc-02 cells were grown in RPMI1640 medium
(Invitrogen, Carlsbad, CA). The 410.4 cells were grown in a folic
acid-deficient RPMI1640 medium. All media were supplemented
with 10% fetal bovine serum (FBS, Sigma), 100 U/mL of penicillin
(Sigma) and 100 μg/mL of streptomycin (Sigma).
Preparation of poly(I:C)-liposome lipoplexes (poly(I:C)-lipo-
plexes or pI:C/LP). Cationic liposomes comprised of cholesterol, egg
phosphatidylcholine and 1,2-dioleoyl-3-trimethylamonium-propane
Figure 6. The combination therapy also synergistically inhibited the growth of
a model breast cancer (410.4) in vivo. Data reported are mean ± S.E.M. (n
= 7–8). The values of the Gem and the pI:C/LP+Gem differ from each other
starting from day 5. The 410.4 cells in the injection site grew very slowly
and tended to metastasize spontaneously.
7Cancer Biology & Therapy2008; Vol. 7 Issue 3
(DOTAP) (all from Avanti Polar Lipids, Alabaster, AL) at a molar
ratio of 4.6:10.8:12.9 were prepared by the thin film hydration
method.19 The final concentration of the DOTAP in the liposomes
was 10 mg/mL. The poly(I:C)-liposome lipoplexes were prepared by
mixing equal volumes of a poly(I:C) (50 μg) solution and a liposome
suspension containing 8 μg of DOTAP, followed by gentle pipet-
ting. The resultant lipoplexes were net negatively charged. The final
poly(I:C) concentration in the lipoplexes was 250 μg/mL. Dextrose
was used to adjust the tonicity of the lipoplexes.
In vitro cytotoxicity assays. To evaluate the combined effect of
the gemcitabine and the pI:C/LP, TC-1 cells (3,000/well) were incu-
bated in the presence of various amounts of gemcitabine, pI:C/LP,
or their mixture at a molar ratio of 3290:1 (pI:C vs. gemcitabine)
for 48 h at 37°C, 5% CO2. Fresh serum-free RPMI 1640 medium
alone was added in the control samples. The number of cells alive was
quantified using an MTT assay. The fraction of affected/killed cells
(Fa) and the fraction of unaffected/live cells (Fu) at every dose (D)
were calculated and plotted following the equation of Log(Fa/Fu) =
m*Log(D)-m*Log(Dm) . Dm is the dose required to produce
the median effect (analogous to the IC50 value), and the m is a
coefficient.20 The combination index (CI) was calculated using the
m and Dm values of the gemcitabine, pI:C/LP, or their mixture as
previously described.20 A CI value of 1, >1 and <1 indicates additive,
antagonism and synergism, respectively.
The ‘by-stander’ cytotoxic effect of the poly(I:C) was evaluated
using TC-1 cells as previously described.7 Briefly, cells (5 x 105) were
cultured in complete RPMI medium in the presence or absence of
the pI:C/LP (1 μg/mL of pI:C) for 24 h, and the culture superna-
tant was collected. Fresh TC-1 cells (5,000 cells/well) were cultured
for 48 h in the presence of the supernatant collected above (ratio of
supernatant to fresh medium, 2:1, v/v). As a control, the TC-1 cells
were also cultured in fresh RPMI medium.
Animal studies. NIH guidelines for animal use and care were
followed. Animal protocol was approved by our institutional
IACUC. Female C57BL/6 and BALB/c mice (6–8 weeks) were
from the Simonsen Labs (Gilroy, CA, USA). TC-1 tumors were
established in the flank of mice by subcutaneous (s.c.) injection of 5
x 105 cells. Starting on day 3 after the tumor cell implantation, the
tumors became visible (~3 mm), and the mice were injected peritu-
morally (p.t.) with the liposomes alone or the pI:C/LP (25 μg of pI:
C per day, 80–100 μL) for 5 consecutive days. The tumor size was
measured and calculated based on the following equation: Tumor
volume (mm3) = ½ [length x (width)2].
To evaluate the anti-tumor activity when mice were treated with
the pI:C/LP and gemcitabine, starting on day 3 after the tumor cell
injection, mice were p.t. injected with the pI:C/LP (50 μg of pI:C
per day) for 5 consecutive days and intraperitoneally injected with
the gemcitabine in PBS (10 mM, pH 7.4) (100 mg per kg of body
weight, twice a week). After a two-day break, the injections were
repeated once. On day 17, 3 mice in each group were euthanized
to collect tumor samples and spleens. The tumor samples were used
for histological evaluations, and the spleens were used to prepare
splenocytes to carry out in vitro CTL assays. The CTL assay was
completed as previously described, and the target cells used were the
TC-1 cells.21 The 24JK cells, another C57BL/6 mouse lung cell line
that does not express the HPV 16 E6 and E7 oncogene proteins,22
were used as the non-target control cells. To evaluate the kinetics of
the CTL response, a group of TC-1 tumor-bearing mice were treated
as above. On days 10, 20, 30 and 40 after the tumor injection, mice
(n = 4) were sacrificed to collect their splenocytes for CTL assay.
Negative control mice were tumor-free and left untreated. Positive
control mice were tumor-free, but immunized with the complexes of
poly(I:C) with an MHC class I-restricted epitope derived from the
HPV 16 E7 protein (pI:C/E749-57) on days 0 and 14 , and their
spleens were collected on day 40.
BALB/c mice with the breast cancer (410.4 cells, 1 x 106)
were also treated similarly. The 410.4 cells were originally isolated
from a single spontaneous mammary tumor. The cells are highly
metastatic and weakly immunogenic. They grew very slowly in the
Histology. To detect apoptosis, the immunohistochemistry was
performed using the cleaved caspase-3 (Asp175) (5A1) rabbit mAb
(Cell Signaling Technology, Danvers, MA) following a protocol
provided by the manufacturer. Assessment of the apoptosis was
performed at 400 x magnification using a counting grid eyepiece
graticule. A minimum of 1,000 cells per section within ten random
fields were scored for cleaved caspase-3 positivity. The apoptotic
index was defined as the % of the cleaved caspase-3+ cells among the
total counted cells. For haematoxylin and eosin (H & E) staining,
tumor tissues were fixed in 10% neutral buffered formalin, processed
on a Tissue-Tek VIP5 Tissue Processor (Sakura), and then embedded
in Paraffin Type 9 (Richard-Allen Scientific, Kalamazoo, MI). The
tissues were sectioned at 4–5 microns and then stained with Gill-
3 Haematoxylin (Fisher Scientific) followed by Eosin Y Alcoholic
(Fisher). The slides were observed under a light microscope.
Calculation of combination index in in vivo studies. The combi-
nation effect was evaluated as previously described.24 An index of
greater than 1 indicates the presence of a synergistic effect, whereas
an index of less than 1 indicates a less than additive effect.
Statistical analysis. Except that the survival curves were compared
using the Kaplan-Meier method (GraphPad Prism 5), other statis-
tical analyses were completed using ANOVA followed by pair-wise
comparisons using the Fisher’s protected least significant different
procedure. A p-value of <0.05 was considered to be statistically
This work was supported in part by a grant from the Elsa U.
Pardee Foundation (to ZC). UML was supported in part by an
Oregon Sports Lottery Scholarship from OSU. NY was supported by
a fellowship from the Payap University in Thailand. UML would like
to thank Tuan Tran for photographing the tumors in mice.
1. Abbruzzese JL, Grunewald R, Weeks EA, Gravel D, Adams T, Nowak B, Mineishi S,
Tarassoff P, Satterlee W, Raber MN. A phase I clinical, plasma, and cellular pharmacology
study of gemcitabine. J Clin Oncol 1991; 9:491-8.
2. Moog R, Burger AM, Brandl M, Schuler J, Schubert R, Unger C, Fiebig HH, Massing
U. Change in pharmacokinetic and pharmacodynamic behavior of gemcitabine in human
tumor xenografts upon entrapment in vesicular phospholipid gels. Cancer Chemother
Pharmacol 2002; 49:356-66.
3. Bornmann C, Graeser R, Esser N, Ziroli V, Jantscheff P, Keck T, Unger C, Hopt UT,
Adam U, Schaechtele C, Massing U, von Dobschuetz E. A new liposomal formulation
of Gemcitabine is active in an orthotopic mouse model of pancreatic cancer accessible to
bioluminescence imaging. Cancer Chemother Pharmacol 2007; Epub.
4. Chawla-Sarkar M, Lindner DJ, Liu YF, Williams BR, Sen GC, Silverman RH, Borden EC.
Apoptosis and interferons: role of interferon-stimulated genes as mediators of apoptosis.
Apoptosis 2003; 8:237-49.
5. Balkwill F, Taylor-Papadimitriou J. Interferon affects both G1 and S+G2 in cells stimulated
from quiescence to growth. Nature 1978; 274:798-800.
Combination chemo-immunotherapy Download full-text
www.landesbioscience.comCancer Biology & Therapy8
6. Salaun B, Coste I, Rissoan MC, Lebecque SJ, Renno T. TLR3 can directly trigger apoptosis
in human cancer cells. J Immunol 2006; 176:4894-901.
7. Shir A, Ogris M, Wagner E, Levitzki A. EGF receptor-targeted synthetic double-stranded
RNA eliminates glioblastoma, breast cancer and adenocarcinoma tumors in mice. PLoS
Med 2006; 3:e6.
8. Hirabayashi K, Yano J, Inoue T, Yamaguchi T, Tanigawara K, Smyth GE, Ishiyama K,
Ohgi T, Kimura K, Irimura T. Inhibition of cancer cell growth by polyinosinic-poly-
cytidylic acid/cationic liposome complex: a new biological activity. Cancer Res 1999;
9. Alexopoulou L, Holt AC, Medzhitov R, Flavell RA. Recognition of double-stranded RNA
and activation of NF-kappaB by Toll-like receptor 3. Nature 2001; 413:732-8.
10. Yoneyama M, Kikuchi M, Natsukawa T, Shinobu N, Imaizumi T, Miyagishi M, Taira K,
Arika S, Fujta T. The RNA helicase RIG-I has an essential function in double-stranded
RNA-induced innate antiviral responses. Nat Immunol 2004; 5:730-7.
11. Liu S, Yu Y, Zhang M, Wang W, Cao X. The involvement of TNF-alpha-related apoptosis-
inducing ligand in the enhanced cytotoxicity of IFN-beta-stimulated human dendritic cells
to tumor cells. J Immunol 2001; 166:5407-15.
12. Le Bon A, Etchart N, Rossmann C, Ashton M, Hou S, Gewert D, Borrow P, Tough DF.
Cross-priming of CD8+ T cells stimulated by virus-induced type I interferon. Nat Immunol
13. Plate JM, Plate AE, Shott S, Bograd S, Harris JE. Effect of gemcitabine on immune cells
in subjects with adenocarcinoma of the pancreas. Cancer Immunol Immunother 2005;
14. Suzuki E, Sun J, Kapoor V, Jassar AS, Albelda SM. Gemcitabine Has Significant
Immunomodulatory Activity in Murine Tumor Models Independent of its Cytotoxic
Effects. Cancer Biol Ther 2007; Epub.
15. Suzuki E, Kapoor V, Jassar AS, Kaiser LR, Albelda SM. Gemcitabine selectively eliminates
splenic Gr-1+/CD11b+ myeloid suppressor cells in tumor-bearing animals and enhances
antitumor immune activity. Clin Cancer Res 2005; 11:6713-21.
16. Nowak AK, Robinson BW, Lake RA. Gemcitabine exerts a selective effect on the humoral
immune response: implications for combination chemo-immunotherapy. Cancer Res 2002;
17. Nowak AK, Lake RA, Marzo AL, Scott B, Heath WR, Collins EJ, Frelinger JA, Robinson
BW. Induction of tumor cell apoptosis in vivo increases tumor antigen cross-presentation,
cross-priming rather than cross-tolerizing host tumor-specific CD8 T cells. J Immunol
18. Lin KY, Guarnieri FG, Staveley-O’Carroll KF, Levitsky HI, August JT, Pardoll DM, Wu
T-C. Treatment of established tumors with a novel vaccine that enhances major histocom-
patibility class II presentation of tumor antigen. Cancer Res 1996; 56:21-6.
19. Le UM, Cui Z. Long-circulating gadolinium-encapsulated liposomes for potential applica-
tion in tumor neutron capture therapy. Int J Pharm 2006; 312:105-12.
20. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects
of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984; 22:27-55.
21. Ren J, Zheng L, Chen Q, Li H, Zhang L, Zhu H. Co-administration of a DNA vaccine
encoding the prostate specific membrane antigen and CpG oligodeoxynucleotides sup-
presses tumor growth. J Transl Med 2004; 2:29.
22. Hwu P, Yang JC, Cowherd R, Treisman J, Shafer GE, Eshhar Z and Rosenberg, SA. In vivo
antitumor activity of T cells redirected with chimeric antibody/T-cell receptor genes. Cancer
Res 1995; 55:3369-73.
23. Cui Z, Qiu F. Synthetic double-stranded RNA poly(I:C) as a potent peptide vaccine adju-
vant: therapeutic activity against human cervical cancer in a rodent model. Cancer Immunol
Immunother 2006; 55:1267-79.
24. Dings RP, Yokoyama Y, Ramakrishnan S, Griffioen AW, Mayo KH. The designed angio-
static peptide anginex synergistically improves chemotherapy and antiangiogenesis therapy
with angiostatin. Cancer Res 2003; 63:382-5.
25. Farrell PJ, Sen GC, Dubois MF, Ratner L, Slattery E, Lengyel P. Interferon Action: Two
Distinct Pathways for Inhibition of Protein Synthesis by Double-Stranded RNA. PNAS
26. Absher M, Stinebring WR. Toxic properties of a synthetic double-stranded RNA.
Endotoxin-like properties of poly I. poly C, an interferon stimulator. Nature 1969;
27. Chawla-Sarkar M, Leaman DW, Borden EC. Preferential induction of apoptosis by inter-
feron (IFN)-beta compared with IFN-alpha2: correlation with TRAIL/Apo2L induction in
melanoma cell lines. Clin Cancer Res 2001; 7:1821-31.
28. Sidky YA, Borden EC. Inhibition of angiogenesis by interferons: effects on tumor- and
lymphocyte-induced vascular responses. Cancer Res 1987; 47:5155-61.
29. Meier H, Myers DD, Huebner RJ. Ineffectiveness of poly rI:rC on transplanted tumors
induced by methylcholanthrene. Naturwissenschaften 1970; 57:248-9.
30. Larson VM, Panteleakis PN, Hilleman MR. Influence of synthetic double-stranded ribonucleic
acid (poly I:C) on SV40 viral oncogenesis and transplant tumor in hamsters. Proc Soc Exp Biol
Med 1970; 133:14-9.
31. Pimm MV, Embleton MJ, Baldwin RW. Treatment of transplanted rat tumours with
double-stranded RNA (BRL 5907). I. Influenced of systemic and local administration. Br J
Cancer 1976; 33:154-65.
32. Anderson KM, Evans A, Kellen JA. Unimpaired development of DMBA-induced mam-
mary tumors in rats treated with poly(I)-poly(C). Res Commun Chem Pathol Pharmacol
33. Weinstein AJ, Gazdar AF, Sims HL, Levy HB. Lack of correlation between interferon induc-
tion and antitumour effect of poly I-poly C. Nat New Biol 1971; 231:53-4.
34. Sakurai M, Iigo M, Sasaki Y, Nakagawa K, Fujiwara Y, Tamura T, Ohe Y, Bungo M, Saijo
N. Lack of correlation between interferon levels induced by polyribonucleotides and their
antimetastatic effect. Oncology 1990; 47:251-6.
35. DuBuy HG. Effect of local or distal therapeutic treatment with polyinosinic-polycytidylic
acid on survival of mice with Ehrlich carcinoma. J Natl Cancer Inst 1972; 48:1525-7.
36. Park JH, Baron S. Herpetic keratoconjunctivitis: therapy with synthetic double-stranded
RNA. Science 1968; 162:811-3.
37. Verdijk RM, Mutis T, Esendam B, Kamp J, Melief CJ, Brand A, Goulmy E. Polyriboinosinic
polyribocytidylic acid (poly(I:C)) induces stable maturation of functionally active human
dendritic cells. J Immunol 1999; 163:57-61.
38. Schmidt KN, Leung B, Kwong M, Zarember KA, Satyal S, Navas TA, Wang F, Godowski
PJ. APC-independent activation of NK cells by the Toll-like receptor 3 agonist double-
stranded RNA. J Immunol 2004; 172:138-43.
39. Sivori S, Falco M, Della Chiesa M, Carlomagno S, Vitale M, Moretta L, Moretta A. CpG
and double-stranded RNA trigger human NK cells by Toll-like receptors: induction of
cytokine release and cytotoxicity against tumors and dendritic cells. Proc Natl Acad Sci USA
40. Gelman AE, Zhang J, Choi Y, Turka LA. Toll-like receptor ligands directly promote acti-
vated CD4+ T cell survival. J Immunol 2004; 172:6065-73.
41. Salem ML, Kadima AN, Cole DJ, Gillanders WE. Defining the antigen-specific T-cell
response to vaccination and poly(I:C)/TLR3 signaling: evidence of enhanced primary and
memory CD8 T-cell responses and antitumor immunity. J Immunother 2005; 28:220-8.
42. Schulz O, Diebold SS, Chen M, Naslund TI, Nolte MA, Alexopoulou L, Azuma Y-T,
Flavell RA, Liljestrom P, Sousa CR. Toll-like receptor 3 promotes cross-priming to virus-
infected cells. Nature 2005; 433:887-92.
43. Cui Z, Le UM, Qiu F, Shaker DS. Learning from Viruses: The Necrotic Bodies of Tumor
Cells with Intracellular Synthetic dsRNA Induced Strong Anti-tumor Immune Responses.
Pharm Res 2007; 24:1645-52.
44. Fujimura T, Nakagawa S, Ohtani T, Ito Y, Aiba S. Inhibitory effect of the polyinosinic-
polycytidylic acid/cationic liposome on the progression of murine B16F10 melanoma. Eur
J Immunol 2006; 36:3371-80.
45. Nowak AK, Robinson BW, Lake RA. Synergy between chemotherapy and immunotherapy
in the treatment of established murine solid tumors. Cancer Res 2003; 63:4490-6.
46. Pratesi G, Petrangolini G, Tortoreto M, Addis A, Belluco S, Rossini A, Selleri S, Rumio C,
Menard S, Balsari A. Therapeutic synergism of gemcitabine and CpG-oligodeoxynucleotides
in an orthotopic human pancreatic carcinoma xenograft. Cancer Res 2005; 65:6388-93.