| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Meeting Reports |
1 Department of Veterinary Biosciences, The Ohio State University, Columbus, Ohio; 2 SLK Kliniken Heilbronn, Medizinische Klinik III, Heilbronn, Germany; and 3 Research Fellowship Program at the 4 Division of Hematology and Oncology, Department of Medicine, University of Freiburg Medical Center, Freiburg, Germany
Requests for reprints: Michael Lübbert, Division of Hematology and Oncology, University of Freiburg Medical Center, Hugstetter Street 55, 79106 Freiburg, Germany. Phone: 49-761-270-3279; Fax: 49-761-270-3697; E-mail: michael.luebbert{at}uniklinik-freiburg.de.
| Introduction |
|---|
|
|
|---|
The 23rd biannual Symposium of the International Association for Comparative Research on Leukemia and Related Diseases (IACRLRD), the oldest international hematologic society, was held September 7 to 11, 2007 in Freiburg, Germany during the 550th anniversary of the Freiburg University. The mission of the IACRLRD Symposia was to provide, at a time of increasing specialization, an active, interdisciplinary discourse between the areas of clinical hematology and oncology, molecular biology, virology, and drug development. More than 300 participants from 19 countries attended the Symposium, which allowed for comprehensive discussion of present approaches in therapy, targeting genetic and epigenetic lesions in leukemia, preleukemia, and solid tumors as well as immunotherapy, transplantation approaches, and novel trends in virology. Results from 14 oral sessions, and five poster sessions are summarized.
| Molecular Targets of Leukemia, Preleukemia, and Cancer Therapy |
|---|
|
|
|---|
The basis for variability in response to imatinib is in part caused by differences in dose and duration of therapy. Furthermore, the high genetic instability caused by BCR-ABL is critical for imatinib resistance, due to decreased DNA repair, and potentially leading to spontaneous acquisition of new mutations (Giuseppe Saglio, John Goldman). Thus, in case of suboptimal response to imatinib, BCR-ABL mutational analyses are indicated. In highly resistant BCR-ABL clones harboring a T315I mutation, imatinib could even provide a growth and survival advantage of these cells by inducing the activation of mitogen-activated protein kinase-1/2 in the absence of interleukin 3 (Thomas Klag).
Second-generation TK inhibitors (TKI), such as dasatinib and nilotinib, show activity against imatinib-resistant CML with BCR-ABL kinase domain mutations (Table 1 ), except for the T315I mutation. In imatinib-resistant CML blast crisis, dasatinib results in major cytogenetic responses in 31% of these patients, and in a 95% complete cytogenetic remission rate after 6 months of therapy in patients with previously untreated CML (1). With nilotinib, CML blast crisis patients reached a major cytogenetic response in 48% of cases, whereas in patients with untreated CML, 100% achieved a complete cytogenetic remission (2). Third-generation TKIs include Bosutinib, a potent inhibitor of Src kinase activity, or Aurora kinase inhibitors, such as MK-0457, active in highly refractory CML, notably also with the T315I mutation.
|
, PDGFRβ, and c-KIT and induces high response rates in patients with PDGFR mutations (Andreas Reiter, Oliver Ottmann). This includes chronic myelomonocytic leukemia, MDS/MPD-U and idiopathic hypereosinophilic syndrome. Down-regulation of the eukaryotic translation initiation factor 5A occurs in imatinib-treated BCR-ABL–positive CD34+ cells (Tim Brümmendorf). Eukaryotic translation initiation factor 5A represents the only known eukaryotic protein activated by posttranslational hypusination (transfer of an amino-butyl residue to lysine) and is essential for cellular proliferation. Additionally, overexpression of cyclin G2 with imatinib treatment in normal CD34+ stem cells was detected, which might be another explanation for the suppression of normal hematopoiesis by blocking cell cycle entry, and thus, differentiation. Activation of Jak2 by point mutations (Jak2 V617F) is mainly associated with polycythemia vera (PV; >90%), essential thrombocythemia (ET; 50%), and idiopathic osteomyelofibrosis (IMF; 58%). In the Frank Rauscher Memorial Lecture, Radek Skoda challenged the question of how a single mutation can cause three different diseases, and whether activating Jak2 mutations are really the primary cause for those diseases. The phenotypic switch is correlated with the expression level of Jak2-V617F, and the highest expression levels were detected for IMF followed by PV and ET. However, the causal clonal event might occur even before the Jak2 point mutation. MPS patients with secondary AMLs often do not carry the activating Jak2 mutation. Therefore, the treatment of PV patients with Jak2 inhibitors may reduce the symptoms of the primary disease, but not the risk of evolving AML. First results from phase I clinical trials with the Jak2 inhibitor LS104 showed responses in PV patients with in vivo reduction of phosphorylated STAT5 (Thomas Fischer).
The transcription factor NF-E2 is up-regulated in Jak2 V617F–positive erythroid progenitors in 93% of cases of PV (Heike Pahl). Induction of NF-E2 can promote erythroid maturation in the absence of Epo and can reprogram precursor cells towards erythroid maturation. Thus, NF-E2 might be a major mediator of the transforming activity of Jak2 V617F.
Molecular targets in solid tumors: modulating angiogenesis. In a Keynote Lecture, Axel Ullrich reviewed the development of the antibody Trastuzumab directed against HER2/neu and the small molecule SU11248 (Sunitinib), a receptor TKI, which was approved in 2006 for the treatment of imatinib-resistant gastrointestinal stroma tumors and renal cell carcinoma.
An angiogenic switch is induced by the disruption of E-cadherin and RAF, providing a rational target for the prevention of metastasis using sorafenib (Ulf Rapp). Activation of coagulation in tumor vessels is mediated in vitro by fusion proteins consisting of the extracellular domain of tissue factor and different peptides (RGD, NGR) targeting
V-integrins or aminopeptidase N (CD13), surface receptors preferentially expressed on tumor endothelial cells (Wolfgang Berdel). In vivo studies revealed that administration of these fusion proteins induced partial or complete thrombotic occlusion of tumor vessels and significant tumor growth retardation of the xenografts, without apparent side effects.
The efficacy of conventional chemotherapy is increased by antiangiogenic drugs (Yuval Shaked). Within hours after administration of chemotherapy, bone marrow–derived circulating endothelial progenitor cells are mobilized, in some cases, stimulating angiogenesis. Cotreatment with an antiangiogenic drug such as anti–vascular endothelial growth factor receptor-2 antibodies can block this response and amplify the antitumor effects of the cytotoxic drug. In a phase I trial of SU14813 (a sunitinib-related receptor TKI), of 49 patients with advanced solid tumors, one patient with renal cell carcinoma achieved a complete remission (CR), and several patients achieved a partial remission (PR), and 13 were treated for more than 12 months (Walter Fiedler).
To bring conventional drugs specifically to the sites of tumor formation, adipose tissue mesenchymal stem cells were used as cellular delivery vehicles in targeted transgene-delivery for prodrugs being converted in the tumor (Cestmir Altaner). The transgenic adipose tissue mesenchymal stem cells in combination with 5-fluorocytosine exerted a strong bystander cytotoxic effect in an in vitro model of colon cancer.
| Epigenetic Alterations in Neoplasia: Diagnostic and Therapeutic Implications |
|---|
|
|
|---|
Clara Nervi introduced "epigenetic priming" with the VPA and the differentiating agent retinoic acid (ATRA) in acute promyelocytic leukemia (APL, M3) with the fusion protein PML/RAR
, and also in non–M3 AML. Chromatin remodeling drugs may increase the AML blasts' sensitivity to conventional chemotherapy, which is supported by studies of non–M3 AML treated with a combination of VPA and ATRA (5). Her group also showed the recruitment of leukemic blasts from the resting G0-G1 cell cycle phases into the S-G2-M phases, resulting in terminal differentiation of these blasts, and AML1/ETO silencing of the retinoic acid receptor pathway via methylation of RARβ2 (6).
Preclinical studies also show a potential benefit of HDACis such as the hydroxamic acid SAHA (Vorinostat) in CML or Ph-negative acute lymphoblastic leukemia (ALL) in combination with the TK inhibitor dasatinib (Mark Kirschbaum). In two Ph-negative cell line models, this combination treatment led to a significant increase in apoptosis and decreased cell growth accompanied by up-regulation of genes reported as related to CML progression such as GADD45G.
The combination of decitabine and 3-deazauridine (3DU) increases the effectiveness of the demethylating drug (Richard Momparler). 3DU inhibits the CTP synthetase, resulting in decreased intracellular levels of CTP and dCTP. As high levels of dCTP inhibit the deoxycytidine kinase, an enzyme that is crucial for the conversion of decitabine into its active metabolite via phosphorylation, inhibition of the CTP synthetase seems to be an ideal additive to increase decitabine's antineoplastic activity in MDS and leukemia treatment.
Down-regulation of death-associated protein kinase 1 (DAPK1) in chronic lymphocytic leukemia (CLL) can result from germ line predisposition as well as from epigenetic or somatic aberrations (Christoph Plass; ref. 7). Silencing of DAKP1 in sporadic cases of CLL was achieved via promoter methylation, whereas in a case of familial CLL, besides promoter methylation, a single nucleotide change proximal of the transcription start site increased the binding affinity of HOXB7 and resulted in the down-regulation of DAPK1.
In a combination assay of Chip-CHIP with gene expression arrays, Carsten Müller-Tidow's group identified 375 direct genomic targets of PML/RAR
, including genes which were important for transcriptional programming, cell cycle control (p21/WAF/CIP1), and apoptosis (BCL2; ref. 8). Mechanistically, mRNA repression of PML/RAR
target genes was accomplished via HDAC1 recruitment, loss of histone H3 acetylation and increased trimethylation of histone H3K4 and H3K9. Interestingly, global histone H3 acetylation and H3K9 trimethylation patterns were capable of distinguishing ALL from AML samples. Within the AML patient sample set, global histone modification patterns revealed several distinct subgroups independent of their karyotype.
Aberrant expression of microRNAs is frequent in tumorigenesis. A study of lung cancer revealed miR-29 family members to be significantly down-regulated as compared with normal lung tissue (Carlo Croce; ref. 9). Driven by the striking complementarity to the 3'-untranslated region of the de novo DNA methyltransferases DNMT3a and DNMT3b, the potential interaction of miR-29 and the DNMTs was studied. In vitro, miR-29 targets DNMT3a and DNMT3b that restoration of miR-29 results in down-regulation of the two DNMTs, which are frequently up-regulated in lung cancer and are associated with poor prognosis. Furthermore, down-regulation of DNMT3a and DNMT3b in vitro also restored normal DNA methylation patterns of epigenetically silenced tumor suppressors such as FHIT, suggesting that miR-based strategies might become a potential treatment option for patients with lung cancer.
A crosstalk is active between epigenetic alterations: microRNAs themselves can be epigenetically regulated by DNA methylation (Supplemental Fig. S1; Björn Hackanson). In leukemia cell lines, miR-124a-3 is silenced via DNA methylation, becomes up-regulated upon demethylating treatment and down-regulates one of its targets, the transcription factor CAAT/enhancer-binding protein-
(C/EBP
; ref. 10).
| Update on Acute Myeloid Leukemia, Myeloproliferative and Myelodysplastic Syndromes |
|---|
|
|
|---|
Whereas AML1/ETO is classically considered as dominant-negative, gains-of-function are also mediated by AML1/ETO (Supplemental Fig. S2; Stephen Nimer). Using retroviral models of AML1/ETO expression in human CD34+ cells, some genes are down-regulated (C/EBP
, MEF/ELF4), others up-regulated (p21/WAF1/Cip1, ID1, TRKa), thereby altering hematopoietic stem/progenitor cell potential (13). Overexpression of p21/WAF1/Cip1 and down-regulation of MEF leads to stem cell quiescence, ID1 restrains cells from differentiating and protects from stem cell exhaustion. Together, these gains-of-function may play an important role for the malignant phenotype. AML1 and AML1/ETO activity is regulated by protein-protein interactions and posttranslational modifications such as acetylation, methylation, phosphorylation, and ubiquitination. For example, the binding of AML1 to SIN3A is modulated by arginine methylation catalyzed by PRMT1.
In a mouse model of CALM/AF10–positive leukemia, a stem cell population with lymphoid characteristics emerges (Christian Buske). Also, the human nucleoporin 98 (NUP98), a component of the nuclear pore complex, contributes to leukemogenesis (Malcolm Moore). NUP98 is fused to several HOX genes in AML, and NUP98/HOXA9 enhances stem cell proliferation and inhibits myeloerythroid differentiation in human CD34+ cells. NUP98/HOXA9 is protected from ubiquitin degradation, resulting in the persistent expression of HOXA9. Allelic haploinsufficiency (loss of heterozygosity) of NUP98 is also implicated in patients with MDS (8%) and AML (29%). In AML, loss of heterozygosity of NUP98 is associated with a poor prognosis.
Clinical trials in younger AML patients. In a Keynote Lecture, Clara Bloomfield reviewed the great advances that have been made in the classification of AML. In 1968, only 18 adult patients with acute leukemias could be identified that survived more than 5 years. In 1976, the French-American-British classification first distinguished ALL from AML and separated different subgroups of AML based on cytology and immunocytochemistry. The importance of genetic aberrations was first recognized in the early 1980. In 2001, the WHO classification for the first time subclassified AML not just on the basis of morphology, but also on the basis of cytogenetics. By molecular methods, it was possible to identify mutations in c-kit in core-binding factor leukemias as adverse prognostic marker, and in AML with a normal karyotype, both adverse (MLL-PTD, FLT3-ITD, overexpression of ERG, or BAALC) and favorable (mutations of CEBP
, NPM1) prognostic factors (14, 15). These can be used not only to guide treatment decisions, but also to design targeted therapies.
In the AML-10 trial of the European Organization for Research and Treatment of Cancer, idarubicin and mitoxantrone were superior to daunorubicin with respect to disease-free survival and overall survival (Roel Willemze). Cytogenetically defined high-risk patients with a compatible donor had a better disease-free survival and overall survival than those without a donor. High-dose cytarabine during induction was compared with standard-dose cytarabine with comparable toxicities during induction in the AML-12 trial. Patients with high-dose cytarabine treatment during induction had a slower platelet recovery and a lower rate of stem cell mobilization. However, patients with a slower platelet recovery had a lower risk for relapse.
In APL, ATRA and simultaneous anthracycline-based chemotherapy cures a high percentage of patients (Miguel Sanz). In low-risk patients, idarubicin combined with ATRA ("AIDA") is sufficient, but there is a clear trend towards better outcome with cytarabine in patients at high risk. Resistant leukemia is virtually absent with these treatment approaches, but infection, bleeding, and differentiation ("ATRA") syndromes remain complicating issues. Relapsed patients can be salvaged with the use of arsenic trioxide (ATO) and stem cell transplantation. The use of ATO in the initial treatment of APL could further improve the results, and a large randomized Europe-wide intergroup trial, coordinated by the Italian GIMEMA group, comparing a chemotherapy-free induction-consolidation regimen (ATRA/ATO) to the AIDA standard is under way.
Older and/or medically non–fit AML patients: an important therapeutic challenge. In AML patients ineligible for intensive chemotherapy, a highly significant advantage has been shown for the use of low-dose cytarabine when compared with hydroxyurea (MRC-14B), thus defining a new standard for the treatment of this patient subgroup (Alan Burnett; ref. 16). However, low-dose cytarabine has only limited activity in high-risk cytogenetics, and new treatments are under development including clofarabine, demethylating agents, cloretazine, ATO, mylotarg, and tipifarnib. With a CR rate of 52% in one trial, clofarabine is among the most promising agents for this difficult-to-treat subgroup of AML patients (Alan Burnett). A new approach ("pick-a-winner") was adopted by the Medical Research Council to rapidly identify active agents for the treatment of elderly patients with AML. Tipifarnib induces remissions in 14% of poor-risk, newly diagnosed patients with AML above the age of 65 years (Judith Karp). In combination with oral etoposide, a 25% CR rate could be achieved, and expression profiling can aid in the selection of tipifarnib-responsive AML patients.
In view of numerous upcoming new treatments for AML, new methods for performing clinical trials are clearly warranted (Elihu Estey). It may be preferable to perform small trials to be able to more rapidly identify more effective and more tolerable treatments. Randomized trials could also use adaptive randomization based on Bayesian statistical inference to allocate more patients to the more successful treatments.
Detection of chromosomal loss provides a major therapeutic option in MDS. Phillip Koeffler presented data on the molecular pathogenesis of MDS generated by the use of single nucleotide polymorphism arrays. Using these arrays, it is possible to narrow down commonly deleted regions of chromosomes. Regions of uniparental disomy are found in
30% of all patients with MDS. By uniparental disomy, loss of heterozygosity can occur without changes in chromosomal copy numbers. Single nucleotide polymorphism arrays even enable the identification of novel fusion genes through small imbalances occurring during translocation.
Lenalidomide, besides acting as an angiogenesis inhibitor and modulating immune responses, has specific, remarkable effects in MDS with the deletion of 5q (Aristoteles Giagounidis). It can induce transfusion independence in 70% of patients with the 5q syndrome, with a high rate of cytogenetic responses (17). Significant activity was also seen in 26% of patients without the 5q deletion. Thus, the marked activity in 5q patients may be associated with the ability of lenalidomide to compensate (e.g., by enhanced gene expression on the remaining allele) for the haploinsufficiency generated by this interstitial deletion (18).
| Lymphoid Malignancies |
|---|
|
|
|---|
Targeting of TdT-positive leukemias by 3'-deoxyadenosine (Cordycepin) has now entered a clinical phase I trial (Richard McCaffrey; ref. 20). The DNA polymerase TdT is expressed in immature B- and T-lymphoid cells and blast cells in several hematopoietic malignancies. It plays a role in N-region insertion in the rearrangement of immunoglobulin genes and creates 3'-OH sites in genomic DNA. Cordycepin additions to these sites favor chain termination and DNA fragmentation, resulting in cytotoxic activity in TdT-positive leukemia cells as well as in a xenograft mouse model.
For diffuse large B-cell lymphoma in patients <65 years, the challenge consists of finding an optimized risk stratification of patients (Michael Pfreundschuh). Ongoing studies will show the feasibility and efficacy of fewer cycles and shorter intervals compared with the standard of six cycles of R-CHOP in patients with different risk profiles. Moreover, the significance of high-dose regimens for unfavorable subgroups of younger patients is being evaluated. In elderly patients with diffuse large B-cell lymphoma, the RICOVER-60 trial studying six versus eight cycles of CHOP-14 with or without Rituximab noted the best results ever reported with six cycles of R-CHOP-14 (78% overall survival at 36 months of follow-up).
Bone marrow aspiration is not required to diagnose chronic CLL according to the guidelines of the IWCLL-NCI working group (Clemens Wendtner). However, the detection of genetic aberrations by interphase fluorescence in situ hybridization is currently considered an essential diagnostic tool to predict the individual course of the disease, and should therefore routinely be conducted when CLL is first diagnosed. Mutation analysis of the immunoglobulin heavy chain should be restricted to clinical trials.
| Immunotherapy of Cancer: Vaccination and Allogeneic Blood Stem Cell Transplantation Strategies |
|---|
|
|
|---|
Adoptive T-cell transfer did result in clinical responses and tumor-associated antigen–specific cellular immunity in 3 of 11 patients with melanoma following adoptive T-cell transfer (Andreas Mackensen; ref. 22). To improve these results, the genetic engineering of T cells is pursued to render them resistant to cell-extrinsic forms of immunosuppression. Paradigmatic for molecules relevant for immune evasion, the coinhibitory factor programmed death receptor-ligand 1 (PD-L1) was discussed. The interaction of PD-L1 on the tumor cell and PD-1 on T cells was critical for immune escape mechanisms of malignant tumors, as evidenced in mouse models, and decreased survival of patients with PD-L1–expressing tumors.
Host-derived IL-18 has striking effects upon conventional CD4 T-cell expansion following allogeneic hematopoietic cell transplantation (aHCT; Robert Zeiser). In a mouse model using IL-18 gene-deficient donor versus recipient animals, the major cytokine production during the early phase of acute graft-versus-host disease induction was recipient-derived, whereas donor hematopoietic cells contributed significantly less (23). This indicates that host-derived IL-18 is a major factor for IFN-
production that may have a protective effect on CD4+-mediated acute graft-versus-host disease, but is nonessential for Treg expansion in an allogeneic environment.
In order to make aHCT similarly accessible to patients with higher age or disease status, different nonmyeloablative regimens have been investigated. Clinical data on the use of reduced-toxicity conditioning according to the FBM protocol were reported by Reinhard Marks (24). In a prospective trial, 133 patients (mostly AML and MDS), of which 106 had advanced disease, were studied at two centers. Interestingly, no significant differences were observed when patients above the age of 55 versus patients below the age of 55 were treated, although a female donor was a predictor for a low relapse rate and chronic graft-versus-host disease. These data are very encouraging and indicate that advanced disease and age are not factors that per se should preclude aHCT.
| Viruses in Cancer: Antiviral Therapy and Viral Vectors for Tumor Targeting |
|---|
|
|
|---|
The discovery of HTLV-1 ATL and the details of carcinogenesis of this virally induced leukemia exemplify the thioredoxin system as a key to both carcinogenesis and therapeutic intervention (Junji Yodoi). Thioredoxin has important biological roles in the regulation of various intracellular molecules via thiol redox control. Thioredoxin expression is elevated in various types of human cancer and the overexpression of thioredoxin introduces resistance to anticancer drugs.
Recently, a novel mitochondrial localizing protein of HTLV-1, p13 was discovered (Vincenzo Ciminale). The full-length HTLV-1 genomic RNA encodes for structural and enzymatic proteins critical for virus replication, whereas a singly spliced mRNA encodes the viral envelope precursor. In HTLV-1, a unique doubly spliced mRNA from ORF II encodes the HTLV-1 p30 protein and a singly spliced mRNA within the same ORF II encodes a protein of 13 kDa (p13). HTLV-1 p13 alters the oxidative balance and ion channel responses of mitochondria, likely important in lymphocyte survival and viral persistence. Michael Lairmore gave a summary on HTLV-1 p30 functions. His work has shown the essential role of HTLV-1 pX ORF II in virus replication in vivo. The group has focused new attention on p30 as a nucleolar localizing protein with unique transcriptional and posttranscriptional properties. Recent data suggests that this unique protein may alter the balance of DNA damage signaling in favor of viral persistence and lymphocyte survival (25).
In a Keynote Lecture, Anton Berns provided insights in bioanalytically supported "cross-species oncogenomics," a novel and powerful tool for cancer gene identification. Slow transforming retroviruses (M-MuLV) induce tumors in tumor suppressor "knockout" mice. The underlying mechanism leading to cell transformation is the activation of proto-oncogenes or inactivation of tumor suppressor genes as a consequence of proviral insertions into the host genome. The insertion sites of approximately 1,000 tumors have been sequenced. Bioinformatics yielded approximately 600 significant common insertion sites that mark oncogenes and tumor suppressor genes. Two gene families were further analyzed. Whereas Pim acts in strong collaboration with Myc, Frat seems to act in the noncanonical Wnt signaling pathway (26).
The therapeutic potential of oligodeoxynucleotides in preventing viral replication in cell-free particles and newly infected cells were reported by Karin Moelling. Mice infected with the spleen focus-forming virus were treated with a specific oligodeoxynucleotide M, which targets the polypurine tract (27). This resulted in a significant delay of disease progression, increased survival rates, and prevented viral infection.
An important highlight of the meeting was the presentation of the Howard Temin Memorial Lecture. A touching summary of the courageous career of Howard Temin was presented by Cestmir Altaner. Jan Svoboda provided the lecture by summarizing his outstanding career accomplishments in the face of adversity throughout his career, inspiring all in attendance.
| Conclusions |
|---|
|
|
|---|
In summary, the Symposium which was perceived as very successful by the participants will be followed by the next Symposium in the fall of 2009 at The Ohio State University, Columbus, OH.
| Disclosure of Potential Conflicts of Interest |
|---|
|
|
|---|
| Acknowledgments |
|---|
| Footnotes |
|---|
Received 1/18/08. Revised 4/17/08. Accepted 5/19/08.
| References |
|---|
|
|
|---|
repress critical targets in leukemogenesis as analyzed by ChIP-Chip. Blood 2008;111:2887–95.
expression in acute myeloid leukemia. Cancer Res 2008;68:3142–51.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |