| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Departments of Cell Biology and Anatomy [S. S. J.], Internal Medicine [S. R. T., A. K.], and Pathology and Microbiology [C. A. K.], University of Nebraska Medical Center, Omaha, Nebraska 68198-6395
ABSTRACT
In this study, in vitro and in vivo antitumor effects of mononuclear cells from human umbilical cord blood cells (UCBCs) and peripheral blood stem cells (PBCs) harvest obtained by leukapheresis were compared. Interleukin 2 (IL-2)-activated mononuclear cells from UCBCs showed increased cytotoxicity against K562 and Raji hematopoietic malignant cells compared with PBCs (P < 0.05). After IL-2 activation, both UCBCs and PBCs showed significant cytotoxicity against MDA-231 human breast cancer cells. The UCBC population involved in this antitumor activity appeared to be CD56+ natural killer precursors. The cytotoxicity of UCBCs was inhibited in the absence of Ca2+(P < 0.05), supporting a perforin/granzyme-mediated target of cell lysis. In addition, antibodies to Fas ligand blocked cytotoxic activity, suggesting that some of the antitumor cytotoxicity was Fas ligand mediated. In vivo antitumor effects of UCBCs and PBCs were studied using a human leukemic cell-bearing severe combined immunodeficient mouse model. There was a significant increase in the survival of K562 leukemia-bearing mice that also received 5 million in vitro IL-2-activated UCBCs or PBCs i.v. on days 3 and day 5 after tumor transplantation compared with untreated mice (P < 0.01). Similar antitumor cytotoxicity of UCBCs and PBCs was also observed against MDA-231 human breast cancer grown in severe combined immunodeficient mice (P < 0.01). These studies suggest that IL-2-activated UCBCs may be a useful source of cellular therapy for patients with hematological malignancies and breast cancer.
INTRODUCTION
Human umbilical cord blood is a rich source of hematopoietic stem cells for hematopoietic reconstitution in cancer patients after high-dose therapy (1, 2, 3, 4, 5, 6) . Umbilical cord blood contains significantly more numbers of hematopoietic stem/progenitor cells compared with normal human peripheral blood, as determined by the number of CD34+ cells and colony forming unit-granulocyte/macrophage progenitors (7) . Transplantation of UCBCs3 has several advantages over other sources of stem/progenitor cells for hematopoietic rescue. These include: widespread availability; absence of donor risk; absence of donor attrition; low risk of transmissible infectious diseases; decreased graft-versus-host disease without an increased incidence of relapse, even in mismatched situations; and increased precursors of immune effector cells (8, 9, 10) . Endogenous hematopoietic reconstitution induced by UCBCs in immunocompromised mice has been reported, indicating another beneficial property of UCBCs for transplantation (11) . In vitro antitumor cytotoxicity of UCBCs against several hematological malignancies has been reported previously (12 , 13) . The natural killer cell-mediated cytotoxicity of UCBCs against standard tumor targets is significantly low compared with mononuclear cells from blood stem cell harvests (14) . However, upon in vitro activation with IL-2, even at a low-dose, short-term incubation, the in vitro antitumor cytotoxic ability of UCBCs increases significantly (14) . These reports indicate the potential clinical usefulness of UCBCs against hematological cancers and solid tumors such as breast cancer, particularly after in vitro activation with cytokines such as IL-2. Additional preclinical studies comparing mononuclear cell cord blood versus peripheral blood using human tumor models of hematological malignancy and solid tumors to confirm the therapeutic ability of UCBCs are essential.
Therefore, to evaluate the potential clinical use of UCBCs as a cellular immunotherapeutic modality, studies were designed to compare the antitumor effects of IL-2-activated UCBCs and PBCs for in vitro and in vivo antitumor cytotoxicity against human hematological malignant cells and human breast cancer cells. In this report, we have described the results of these studies, including the mechanism of action of cytolysis of tumor cells by activated UCBCs in vitro.
MATERIALS AND METHODS
Cell Collections.
All human cell specimens were obtained using protocols approved by the
Institutional Review Board at the University of Nebraska Medical
Center.
Cord Blood Cells.
Human UCBC samples were obtained by the Labor and Delivery nursing
staff. Briefly, after the delivery of an infant, the umbilical cord was
double clamped and ligated, and venipuncture of the umbilical cord vein
was performed using a heparinized syringe prior to expulsion of the
placenta. MNCs in the cord blood samples were isolated using lymphocyte
separation medium. The cells were further processed in a manner
identical to the PBCs described below.
MNC Harvest from Peripheral Blood Apheresis Product.
Normal volunteers received no mobilizing cytokine prior to apheresis.
PBCs were collected with the Cobe Spectra apheresis machine, as
described previously (15)
. Briefly, an aliquot of the PBC
collection was layered and centrifuged over lymphocyte separation
medium to obtain MNCs. The MNCs were washed once with RPMI 1640 and
resuspended in RF10 medium, which consisted of RPMI 1640 supplemented
with 10% fetal bovine serum, 2 mM L-glutamine,
100 units/ml penicillin, and 100 µg/ml streptomycin
(15)
. The MNCs were activated in vitro with
IL-2 and then used for cytotoxicity assays and/or for the in
vivo therapeutic studies described below.
Cell Line Cultures.
The human chronic myeloid leukemic cell line, K562 (NK
sensitive), and human B-cell lymphoma cell line, Raji (NK resistant/LAK
sensitive), were grown in vitro in
25-cm2 culture flasks in RF10 medium. As a
representative of solid tumors, MDA-231 human breast cancer cells were
grown in DF10 medium (DMEM medium supplemented with 10% fetal bovine
serum, L-glutamine, and antibiotics, as above) as
monolayer cultures and used in both in vitro and in
vivo studies. Raji, K562, and MDA-231 cells were labeled with
chromium-51
(Na251CrO4)
and used as target cells for cytotoxicity assays.
Activation of Effector Cells with IL-2.
MNCs obtained from UCBCs and PBCs were activated in vitro
with IL-2 and assayed for cytotoxicity. Five million cells were
cultured in T-25 flasks containing 5 ml of RF10, 50
µM ß-mercaptoethanol, and 1000 units/ml IL-2
for 72 h at 37°C in 5% CO2 and 95% air.
Control flasks contained PBCs or UCBCs in 5 ml of the above media
without IL-2. After the incubation period, the cells were harvested
using a diSPo cell scraper, washed twice with RF10 or DF10 medium,
resuspended, counted, and assayed for their cytotoxicity against
51Cr-labeled K562, Raji, or MDA-231 target cells.
Activated cells were also injected into tumor-bearing mice to evaluate
their in vivo antitumor activity.
Cytotoxicity Assay.
MNCs obtained from UCBC and PBC harvests were tested for cytotoxic
activity against 51Cr-labeled K562, Raji, or
MDA-231 tumor target cells before and after in vitro
activation, as described previously (15)
. Briefly, the
effector cells and 51Cr-labeled target cells were
mixed to obtain the E:T ratios of 12.5:1, 25:1, 50:1, and 100:1 in 200
µl of RF10 or DF10 medium/well in 96-well plates. The plates were
incubated at 37°C for 4 h, and the supernatants were then
harvested. The radioactivity in the supernatant was measured using a
Beckman 5500 gamma counter. The percentage of target cells lysed was
calculated with the following formula:
 |
where the experimental cpm is effector + target cells, spontaneous cpm is target cells + medium, and maximum cpm (total release) is target cells + 1% Triton X-100.
Immunophenotyping Analysis.
Cell surface immunophenotyping of MNCs from UCBCs and normal PBCs was
performed by flow cytometric analysis using a Becton Dickinson FACStar
PLUS flow cytometer as described previously (16)
. Three
hundred thousand cells were suspended in 200 µl of PBS containing
0.1% BSA and 0.05% NaN3 (fluorescence buffer)
in 12 x 75-mm culture tubes. Cells were incubated on ice for 30
min with 10 µl of FITC or 5 µl of phycoerythrin-labeled antibody
specific for the following cell surface markers: CD3, CD4, CD8, CD49e,
CD49d, CD56, CD80, and CD83 (Becton Dickinson Corp., San Jose, CA).
After incubation, the cells were washed with 1 ml of fluorescence
buffer, centrifuged, and resuspended in 800 µl of PBS for analysis.
Purification of CD3- and CD56-positive Cells.
The CD3-positive T cells and CD56-positive NK cell populations were
purified from UCBC and PBC MNCs using flow sorting techniques as
described by Pirruccello et al. (16)
. Briefly,
the cells were stained with CD3 antibodies conjugated to phycoerythrin
and CD56 antibodies conjugated to FITC. CD3-positive and CD56-positive
cells were sorted and separated to at least 95% or greater purity
using a Becton Dickinson FACStar PLUS flow sorter. The purified cells
were then used to determine cytotoxicity activity after in
vitro activation with IL-2 as described above.
Role of Perforin and Granzyme in Cytolysis of Tumor Cells by UCBCs.
To determine the mode of cytolysis of target cells by activated UCBCs,
calcium-free RF10 medium was used in the cytotoxicity assay. Calcium is
an essential component for cytolysis mediated by perforin and granzyme
present in the effector cells (17)
. The UCBC effector
cells were activated as above and used in an in vitro
cytotoxicity assay. The effector and 51Cr-labeled
target cell mixtures were suspended in either calcium-free RF10 medium
or RF10 medium with calcium. The rest of the procedure for the
cytotoxicity assay was similar to that described above.
Role of Fas Ligand in Cytolysis of Tumor Cells by UCBC.
Tumor cell lysis by the effector cells may be mediated by FasL. Some
effector cells express FasL on their surface, which recognizes the Fas
molecule expressed on target cells. The FasL-Fas binding leads to the
activation of apoptosis pathways in the target cells, resulting in
death (18)
. Therefore, to determine the role of FasL in
the cytolysis of tumor target cells by the UCBCs, anti-FasL antibodies
were used to block FasL-Fas binding. Purified, activated cells from
UCBCs were mixed with 51Cr-labeled target cells,
and different concentrations of anti-FasL antibody, NOK1 (PharMingen,
San Diego, CA), were added. The control wells contained
isotype-matched, nonspecific antibodies. The rest of the cytotoxicity
assay was similar to that described above.
RT-PCR Analysis for the Expression of Perforin and FasL Genes in
Activated Cells.
A RT-PCR technique was used to determine the expression of FasL and
perforin genes in activated UCBC cells as we have described previously
(19)
. Briefly, RNA was isolated from activated UCBCs using
TRIzol reagent obtained from Life Technologies, Inc. (Gaithersburg,
MD). One µg of RNA was reverse transcribed using 150 ng of random
primers following standard procedures, with the exception of the
addition of DMSO (1% final concentration), and incubated at 44°C for
1 h. PCR was then performed on 10 µl of cDNA in 50-µl reaction
medium including 20 pmol of each primer (for perforin, 5'-GTT GCA TCT
CAC CCT CAT GGG ACC AGA CTT-3' and 5'-TAA GCC CAC CAG CAA TGT GCA TGT
GTC TGT-3'; and for FasL, 5'-AGG GGC AGG TTG TTG CCA GA-3' and 5'-CAC
CCC AGT CCA CCC CCT GA-3'), 2.5 mM
MgCl2, 200 µM deoxynucleotide
triphosphates and 5% DMSO. The temperature conditions were as follows:
5 min at 95°C, followed by 2 min at 62°C, followed by 25 cycles of
20 s at 72°C, 45 s at 94°C, and 20 s at 62°C. PCR
products were visualized by electrophoresis in ethidium
bromide-stained, 2% agarose gels. As a control,
ß-tubulin gene expression was measured
using the same procedure and appropriate primers,
5'-AAGAAATCCAAGCTGGAGTTC-3' and 5'-GTTGGTCTGGAATTCTGTGAG-3'. All primer
sets were designed to span introns to eliminate any potential signal
from contaminating DNA in the RNA preparation (19)
.
In Vivo Experiment with Human Leukemic Target Cells.
To evaluate the in vivo effects of activated UCBCs and PBCs
against malignant hematological cells, K562 cells were grown in SCID
mice, 6–8 weeks of age, with CB17 background (Charles River,
Wilmington, MA). Each group consisted of 10 mice. The animals were
transplanted with 5 x 106 K562 cells i.v.
Three days after tumor transplantation, 1 x
107 MNCs from UCBCs or PBCs activated in
vitro with IL-2 were infused i.v. into each mouse. After
activation, cells were washed three times with sterile PBS and then
used for infusion. The absolute numbers of CD56-positive NK cells
infused were 3.4 x 106 cells for UCBCs and
2.2 x 106 cells for PBCs per mouse. The
control mice were untreated. The mice were observed for signs of
morbidity/death. After necropsy, the spleen, liver, kidney, heart, and
lung were fixed in 10% buffered formalin and processed for
histological evidence of tumor. Tumors that developed in the control
mice were also removed for pathological confirmation of the presence of
a tumor.
In Vivo Experiment with Human Breast Cancer Target
Cells.
A human breast cancer cell line, MDA-231, was used as a solid tumor
model. For in vivo therapeutic analysis, SCID mice, 6–8
weeks of age, with CB17 background (Charles River, Wilmington, MA) were
used. Animals were transplanted with 3 x
106 MDA-231 cells s.c. in the flank along with
Matrigel (Collaborative Biomedical Product, Bedford, MA) to enhance the
tumor uptake and growth. The control group of six mice did not receive
any treatment, whereas each treatment group consisted of seven mice.
Twelve days after tumor transplantation, 5 x
106 in vitro IL-2-activated UCBCs or
PBCs were infused i.v. After IL-2 activation, cells were washed three
times with sterile PBS and then used for infusion. This cell therapy
treatment was repeated three additional times over a period of 2 weeks.
In total, the absolute numbers of CD56-positive NK cells infused into
these mice were 3.9 x 106 cells/mouse for
UCBCs and 3.25 x 106 cells/mouse for PBCs
as a result of three infusions. Tumor growth was determined by
measuring the tumor volume using a caliper twice a week and plotted on
a linear scale. Ten weeks after tumor transplantation, all remaining
mice were sacrificed and processed as described above.
Statistical Analysis.
Data from control and experimental groups were analyzed using the
Tukey-Kramer Multiple Comparison Test, and differences were considered
significant when P < 0.05.
RESULTS
In Vitro Cytotoxicity of UCBCs versus
PBCs against Hematological Tumor Targets.
The cytotoxicity levels of normal UCBCs and PBCs against NK-sensitive
K562 and NK-resistant but LAK-sensitive Raji tumor target cells were
determined after in vitro activation with IL-2 (Fig. 1)
. The cytotoxicity of IL-2-activated
UCBCs was significantly (P < 0.05) higher than that
generated by IL-2-activated PBCs against K562 tumor target cells (Fig. 1
A). Similarly, the cytotoxicity of IL-2-activated UCBCs
against Raji tumor cells was significantly greater (P < 0.001) than the levels in IL-2-activated PBCs at all E:T ratios
(Fig. 1
B). These results demonstrate the greater antitumor
cytotoxicity of UCBCs compared with PBCs against hematological
malignant cells.
|
show the
cytotoxicity of both UCBCs and PBCs after in vitro
activation with IL-2. In this case, there was no significant difference
between the in vitro cytotoxicity of activated UCBCs and
PBCs against MDA-231 breast cancer.
|
indicated an increased cytotoxicity of
activated MNCs from UCBCs against different tumor targets compared with
activated MNCs from PBCs. In an attempt to dissect out the effector
cell types involved in the cytotoxicity, CD3-positive and CD56-positive
cells were purified by flow cytometric cell sorting (>95% purity)
from UCBCs and tested for cytotoxicity using the in vitro
cytotoxicity assay (Fig. 3)
. The purified
CD56 cells showed significant levels of cytotoxicity against both K562
and Raji target cells (P < .001), whereas the
CD3-positive cells did not, suggesting that the CD56-positive NK cells
may be the predominant antigen-nonspecific effector cells in the UCBCs,
as is the case of PBCs (data not shown).
|
.
CD56+ NK cells were significantly increased in
UCBCs compared with PBCs (P < 0.05). The frequencies
of other markers, such as CD4+-positive T helper
cells, or the expression of adhesion molecules CD49e and CD49d were
comparable in both UCBCs and PBCs. Expression of CD54, a marker for
intercellular adhesion molecule expression, was significantly lower
(P < 0.05) in UCBCs compared with PBCs.
|
depicts the results of
these studies, showing significantly decreased (P <
0.05) cytotoxicity of IL-2-activated UCBCs against either K562 or Raji
target cells compared with cytotoxicity levels of similar UCBCs in the
presence of Ca2+. These results support the
hypothesis that tumor cell lysis by activated UCBCs is mediated through
the perforin/granzyme pathway.
|
. A
nonspecific, isotype-matched control antibody did not reduce the
toxicity of activated UCBC-derived MNCs, indicating the FasL-specific
inhibition of cytotoxicity by the NOK1 antibody.
|
and Table 1
indicate that both perforin- and
FasL-mediated pathways are involved in cytolysis of tumor cells by UCBC
MNCs. The expression of perforin and FasL in activated UCBCs was
confirmed by RT-PCR analysis using specific primers for each gene.
Expression of the ß-tubulin gene was determined
as a control. Fig. 6
demonstrates
expression of perforin and FasL gene in UCBCs after
activation.
|
shows the in vivo
therapeutic effects of activated UCBCs and PBCs against K562 cells as
measured by the survival of tumor-bearing treated versus
tumor-bearing control animals. All 10 mice in the untreated control
group died by 110 days or less after tumor cell infusion, whereas 50%
of the PBC-treated animals and 10% of the UCBC-treated animals were
dead in that time frame. One hundred sixty days after tumor
inoculation, 90% of the UCBC-treated mice and 50% of the PBC-treated
mice survived. These results suggest a significant antitumor activity
of both UCBCs and PBCs (P < 0.01). Activated UCBCs
appear to have a greater antileukemic effect compared with PBCs.
|
demonstrates the in vivo
antitumor therapeutic effects of IL-2-activated MNCs from UCBCs and
PBCs. The tumor size values represent the mean value derived from at
least six animals from each group. There was an exponential growth of
tumor in control untreated mice and a significantly (P < 0.01) reduced growth in mice treated with IL-2-activated PBCs or
UCBCs. UCBCs appeared more effective than PBCs. After 80 days, all mice
were sacrificed, and no metastatic tumor was identified. The
histological analysis of the primary tumors revealed more necrosis in
tumors from the treated group than the untreated control group.
However, the tumor nodules appeared to contain some areas with live
tumor cells based on cytomorphology.
|
In this study, the in vitro and in vivo
antitumor properties of the IL-2-activated MNCs from UCBCs in
comparison with IL-2-activated MNCs from PBCs were explored. The NK
cytotoxicity levels of UCBCs were very low compared with PBCs prior to
in vitro activation (data not shown). After IL-2 activation,
however, there was significant antitumor cytotoxicity against
NK-sensitive/LAK-resistant and LAK-sensitive tumor target cells (Fig. 1)
. Similar results have been observed by other investigators
(20, 21, 22, 23, 24, 25, 26)
. The activated UCBCs were not only cytotoxic to
hematological malignant cells but also considerably cytotoxic to human
breast cancer cells (Fig. 2)
. The level of cytotoxicity of activated
cord blood cells against breast cancer was not different from that of
similarly activated MNCs from peripheral blood apheresis product. This
result is different from what we observed with K562 and Raji tumor
targets. The precise reason for this difference is not clear at this
time.
The in vitro results reported here suggest that the CD56+ CD3- MNC population provided the antitumor activity of cord blood MNCs. The purified CD3-positive cells did not demonstrate significant cytotoxicity in vitro after IL-2 activation in the studies reported here. LAK precursor cells were identified based on the expression of CD56 cell surface antigens. Although the majority of the LAK precursors are CD56-positive cells, a very small population of T cells that express both CD3 and CD56 is also capable of generating LAK cells. Only a few LAK precursors expressed the T-cell surface markers CD3 and CD5. Relatively immature NK precursor cell populations have been isolated from umbilical cord blood (23) , which was phenotypically different from that observed in adult peripheral blood MNCs. One population was shown not to express NK markers (CD7+, NK-) but developed into phenotypically and functionally active NK cells after incubation with IL-2. Another population of precursor NK cells did express NK markers CD16 and CD56 (CD7+, NK+). Both cell populations were deficient in their ability to generate NK-like activity but readily acquired this activity after activation with IL-2. The CD3+ lymphocytes from UCBCs displayed strong LAK activity after in vitro activation with IL-2 (24) . The majority of these CD3+ lymphocytes (>80%) expressed the high-affinity IL-2 receptor consisting of both p55 and p75 polypeptide chains, indicating that UCBC T lymphocytes, in the presence of IL-2, can differentiate into LAK cells. The results presented in this report, as well as reports from other investigators described above, suggest that in the presence of IL-2, human umbilical cord blood T cells are endowed with the ability to function as a first-line defense against aberrant cell growth. Because in the neonatal period both T and B lymphocyte-mediated immune responses are still on their way to full development, the T cells in the cord blood might function as nonspecific functional effector cells. These studies also confirm that the MNCs from human cord blood have the potential to be used in the clinical setting as a source for cellular therapy.
Vujanovic et al. (26) have demonstrated that the LGLs activated in culture with IL-2 became adherent to the plastic surface of the flask. These cells, accounting for >95% LGLs based on morphological criteria, demonstrated potent antitumor activity against tumor cells in vitro. The UCBCs contain a relatively immature immune cell population; these cells may not express appropriate adhesion molecules, such as CD11 and CD18, that would allow them to attach to the flask. Studies have already shown that plastic adherence by LGLs can be markedly reduced with monoclonal antibodies to these cell surface molecules (24) . Our results with phenotypic analysis showed that the expression of the adhesion molecule CD54 was significantly less on UCBC lymphocytes compared with PBCs. The precise role of the decreased expression of CD54 on UCBCs is not known. These observations suggest further studies on the role of adhesion molecules in the cytotoxicity of UCBCs in vitro.
More interestingly for this report, UCBCs have demonstrated significantly greater levels of LAK cytotoxicity compared with adult blood or PBCs. Harris et al. (12) found that IL-2-activated UCBCs produced IL-2, IL-4, and granulocyte/macrophage-colony stimulating factor mRNA encoding hematopoietic growth factors, as measured by message and protein levels. IL-2-activated UCBCs may prove useful in the transplantation setting, serving as antitumor effector cells. As in the case of peripheral blood lymphocytes including NK cells (16 , 17) , the results with UCBCs have shown that both perforin- and FasL-mediated cytolysis of tumor cells are methods used by the in vitro IL-2-activated UCBCs (17 , 18) . The most important findings of this study were the in vivo antitumor therapeutic effects of UCBCs against both human leukemic and breast cancer cells. Phenotypic analysis of the activated cells used for in vivo studies showed that there were more CD56+ cells in the UCBCs compared with PBCs. It is possible that the increased antitumor activity could be attributable to the increase in CD56-positive NK cells in UCBC populations. These results provide the foundation for further studies to identify the precise phenotype of the antitumor effector cells in UCBCs. We hope that these studies may also open avenues leading to clinical studies to test the antitumor immunotherapeutic effects of UCBCs against cancer.
ACKNOWLEDGMENTS
We thank Nataliy Babushkina-Patz for technical assistance. We also thank the Labor and Delivery staff at University of Nebraska Medical Center for assistance; Dr. James Lynch for help in statistical analysis; Michelle Desler, Eileen Vu, and Penni Davis for assistance in preparing the manuscript; and Kristi Berger editorial assistance. We are grateful to Prof. David Crouse for critically evaluating the manuscript for its content and interpretation of the results.
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 This work was supported by a research grant from
the Elsa U. Pardee Foundation and a grant from the Susan G. Komen
Breast Cancer Foundation. 
2 To whom requests for reprints should be
addressed, at Cell Biology and Anatomy, University of Nebraska Medical
Center, 986395 Nebraska Medical Center, Omaha, NE 68198-6395. Phone:
(402) 559-4165; Fax: (402) 559-7328; E-mail: ssjoshi{at}unmc.edu
3 The abbreviations used are: UCBC, umbilical cord
blood cell; PBC, peripheral blood stem cell; IL, interleukin; MNC,
mononuclear cell; NK, natural killer; LAK, lymphocyte activated killer;
FasL, Fas ligand; RT-PCR, reverse transcription-PCR; SCID, severe
combined immunodeficient; LGL, large granular lymphocyte.
Received 3/27/00; revised 8/22/00; accepted 8/24/00.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
