
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Division of Surgery, Department of Thoracic and Cardiovascular Surgery [M. N., J. A. R., L. J.], and Departments of Experimental Radiation Oncology [R. E. M.] and Biomathematics [L. B. L., E. N. A., R. A. W.], The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Docetaxel, a derivative of taxane, is a highly active agent in lung cancer. The taxanes mediate cytotoxicity by stabilizing the microtubules of the cellular mitotic apparatus and preventing their depolymerization into free tublin (3 , 4) . Docetaxel in particular, has been shown to inhibit cell replication and be cytotoxic in vitro to various tumor cell lines, to exhibit antitumor activity in different animal tumor systems, and to be effective in the treatment of common human cancers (5, 6, 7, 8, 9, 10) . Chemotherapeutic agents used as induction therapy may be effective in treating systemic micrometastasis but may also have radiosensitizing effects that increase local tumor control. Improved local tumor control can substantially increase long-term survival in lung cancer (11) . Because taxanes have been shown to arrest cells in both G2 and M (the phases in the cell cycle most sensitive to ionizing radiation) their radiosensitizing potential has been explored both in vitro and in vivo, and in clinical trials (12, 13, 14, 15) .
The tumor suppressor gene p53, which is frequently mutated somatically or deleted in various human cancers, plays an important role in tumorigenesis. The tumor suppressor activity of the gene is mainly through its ability to induce cell growth arrest or apoptosis in response to a variety of stress signals, whereas a lack of functional p53 usually leads to increased genomic instability, deregulated cell proliferation, accelerated tumor progression, and elevated cellular resistance to anticancer therapy (16, 17, 18) . Gene therapy by adenovirus-mediated introduction of the wild-type human p53 gene into tumors that are deficient in functional p53 has shown a significant tumor-suppressing efficacy both in animal systems and in human clinical trials (19, 20, 21, 22, 23, 24, 25) . Furthermore, the tumor-suppressing activities of such therapy have been reported to be enhanced by combination with many chemotherapeutic drugs or ionizing radiation in various human tumors (26, 27, 28) .
Gene replacement with Ad-p53 could potentially restore chemotherapy and radiation therapy sensitivity to primary lung cancers that lack p53 function. However, clinical trials of combination therapy are frequently conducted empirically in the absence of both in vitro and in vivo supporting data (29) . A model to investigate complex three-agent interactions may be useful in designing clinical trials. No information is available on the combined effects of the three therapeutic agents Ad-p53, docetaxel, and radiation on tumor growth in vitro and in vivo.
Therefore, in the study described here, we developed mathematical modeling and statistical analysis methods for evaluating (in terms of synergism or antagonism) the therapeutic efficacy of the combination treatment with these three therapeutic agents in vitro and in vivo. Our methodology will provide a valuable tool for presenting critical supporting data for human clinical trials. We also demonstrated for the first time that combination treatment with docetaxel, Ad-p53, and radiation synergistically inhibited the growth of human NSCLC cells in vitro and in vivo (in animal models), which suggests that this three-modality combination treatment may constitute an effective approach to human cancer therapy.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Agents.
Docetaxel (Taxotere) was obtained from Rhône-Poulenc Aventis Pharma (Frankfurt, Germany). Three recombinant adenoviral vectors were used: Ad-p53, which contains a wild-type p53 gene, was used as a gene therapy agent; Ad-Luc, which contains a luciferase gene, was used as a nonspecific negative control; Ad-EV, an empty vector with no transgene insert, was used as a negative control. All three were made replication-deficient by deleting the E1 and E3 regions from the viral genome (30)
. All three recombinant adenoviral vectors were constructed as described previously (31
, 32)
. Viral stocks were prepared by the Vector Core Facility at M. D. Anderson Cancer Center (Houston, TX). Viral titers were determined by absorbance measurement (vp/ml) and plaque assay (plaque-forming units/ml). Potential contamination of the viral preparations by wild-type virus was monitored by PCR analysis.
Growth Inhibition and XTT Assay.
Inhibition of tumor cell growth by combination treatment with docetaxel, Ad-p53, and radiation were analyzed by quantitatively determining cell viability using an improved XTT assay (Roche Molecular Biochemicals, Indianapolis, IN; Ref. 33
). The XTT assay is based on the cleavage of the yellow tetrazolium salt XTT {sodium 3'-[1-(phenylamino-carbonyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzene sulfonic acid hydrate} by mitochondrial dehydrogenases of viable cells, which results in the formation of an orange formazan dye. Briefly, cells were plated in 96-well microtiter plates at 1 x 103 cells/well in 100 µl of medium. One day after the cells were plated, 25-µl aliquots of media containing varying concentrations of docetaxel were added into each well. After 24 h of exposure to docetaxel, the medium was removed from each well and replaced with a 100-µl aliquot of medium containing Ad-p53 or Ad-Luc vector at various MOI in units of vp/cell. After 4 h of incubation with adenoviral vectors, cells in 96-well plates were irradiated with various doses of
-radiation in a 137Cs unit at room temperature. Cells were then incubated at 37°C in a humidified atmosphere under 5% CO2. Five days after incubation, cell growth and viability were quantified by XTT assay. Briefly, the culture medium was removed, and 50 µl of XTT reaction mixtures were added into each well with fresh medium at a final concentration of 0.3 mg/ml/well. Cells were then incubated for 2 h at 37°C. The absorbance was measured at a wavelength of 450 nm against a reference wavelength of 630 nm in a microplate reader (Model MRX; Dynatech Laboratories, Chantilly, Virginia). Percentage cell viability was calculated in terms of the absorbency in treated cells relative to the absorbency in the untreated control cells. Experiments were repeated at least three times with quadruplicate samples for each treatment in each individual experiment.
The treatment schedule and experimental design were chosen and optimized based on: (a) experimental observations that docetaxel could reduce the number of clonogenic cells in tumors undergoing radiotherapy by its own cytotoxic action and its radiosensitizing ability and that docetaxel given within 2448 h before irradiation acted as a potent enhancer of tumor radioresponse and increased the therapeutic gain of irradiation (14 , 34) ; (b) the clinically relevant treatment schedule in which a single bolus of docetaxel (3040 mg/kg i.v.) was usually given 24 h before 35 daily fractions of radiation (14 , 34) ; and (c) the facts that p53-deficent cells were usually shown to be resistant to radiation therapy and wild-type p53 expression could sensitize the response of tumor cells to irradiation and enhance p-53-induced apoptosis through the mechanism of irradiation-induced DNA-damaging repair (28) . This schedule takes advantage of reoxygenation of hypoxic tumor cells during the interval between drug treatment, exogenous wild-type p53 gene expression, and radiation delivery, and is in the closest relevance to clinical practice.
Isobolograms Analysis of Two- and Three-Agent Combinations.
The effects of two-agent combinations and three-agent combinations on tumor cell growth of four human NSCLC cells (H1299, H460, H322, and A549) in vitro were respectively analyzed by two-dimensional and three-dimensional isobolograms originally described by Steel and Peckham (35)
and improved by us. The schematic representations of the two-dimensional and three-dimensional isobologram methods are illustrated in Fig. 1
.
|
To determine the significance of synergism in a combination, Wilcoxons signed-rank tests were performed to compare the mean values of the observed data with the predicted minimum or maximum values for the additivity. If the mean value of the observed data was equal to, or smaller than, the predicted maximum value but equal to, or larger than, the predicted minimum value, the effect of the combination was considered additive. If the mean value of the observed data was smaller than the predicted minimum value, the effect of the combination was considered synergistic; if larger than the predicted maximum value, antagonistic, P
0.05 was considered significant. All of the statistical analyses were performed using Statistica software (StatSoft Inc., Tulsa, Oklahoma).
Efficacy of Combination Treatments in Animal Models.
All of the animals were maintained, and animal experiments were performed, under NIH and institutional guidelines established for the Animal Core Facility at M. D. Anderson Cancer Center. The animals used in this study; female Nu/nu mice (68 weeks of age), were purchased from Charles River Laboratories (Wilmington, MA). Prior to tumor cell inoculation, mice were subjected to 3.5 Gy of total body irradiation from a 137Cs radiation source. Both H1299 and A549 cells were used to establish s.c. tumors in mice. Briefly, 1 x 106 cells were injected into the right flank of each mouse. When the average size of tumors reached 50100 mm3, mice were randomly divided into treatment groups containing 510 mice each. Mice were treated according to the following schedule: on day 0, mice were i.v. injected with docetaxel diluted in 0.2 ml of PBS at a dose of 30 mg/kg/mouse (PBS alone was used as a mock treatment); on day 1, each mouse was intratumorally injected with adenoviral vector Ad-p53, Ad-EV, or Ad-luc at a dose of 5 x 1010 vp/tumor in a volume of 0.2 ml; on day 2, tumors were locally irradiated with a vertical 60Co-
-radiation beam from a Theratron 780C Cobalt unit (Theratronics International, Ltd., Kanata, ON, Canada) at a dose of 5 Gy/tumor. Each mouse was restrained in a custom-designed jig without anesthesia and positioned in the radiation field such that only the tumor xenograft implanted on the hind flank was exposed to the radiation beam, and the rest of the body was shielded by a lead block. Mice were ear-tagged so that data obtained from individual animals could be traced. Tumor dimensions were measured three times per week using a digital caliper. Tumor volume was calculated using the equation V (mm3) = a x b2/2, where a is the largest diameter and b is the smallest diameter. Differences in tumor volumes between treatment groups were analyzed using an ANOVA test with the Statistica software. A difference was considered to be statistically significant when P
0.05.
A mathematical model and a statistical method were developed to analyze the effects of combination treatments in our animal model. The log of tumor volume was fitted using a quadratic model in time. To avoid numerical difficulties, 1.0 was added to each volume prior to analysis. Treatment and time were entered as fixed effects, mouse as a random effect. Interaction terms were entered between the treatment term and both the linear and quadratic terms in time. A number of covariance structures were examined for each model. The final structure was selected using the Akaikes Information Criterion (37) . For data sets from A549 tumor models, an unstructured covariance matrix was chosen; for H1299 models, a Toeplitz heterogeneous structure was chosen. Interaction between two treatments (A and B) was defined in the following way. Given time t, the mean tumor volume in the control group is V0(t); in treatment group A, VA(t); in treatment group B, VB(t); and in the combined treatment group A+B, VAB(t). Let pA(t) = VA(t)/V0(t), where pA(t) is the proportion in which the treatment A reduces tumor volume at time t and thus VA(t) = pA(t) V0(t). Similarly, define pB(t) and pAB(t). Then, if there is no synergy between treatments A and B, pAB(t) = pA(t) pB(t). For example, if the tumor volume in treatment group A falls to 80% and the tumor volume in treatment group B falls to 70% of that in the control group at time t, then, in the absence of synergy, the combined treatment should reduce the tumor volume to 56% of that in the control group. Under this definition, synergy may be present at one time point (t) and not at others. Our hypothesis is that the equation pAB(t) = pA(t) pB(t) may be reformulated as a linear hypothesis suitable for testing by substituting for the definitions of PA, PB, and PAB. Taking the logs of both sides of the equations yields lnVAB = lnVA + lnVB - lnV0. This may be tested using an appropriate contrast. All of the analyses were performed using SAS procedure MIXED in SAS version 6.12 (SAS Institute Inc., Cary, NC).
| RESULTS |
|---|
|
|
|---|
-radiation, were analyzed by determining the relative viability of the treated cells. The cytotoxicity of each agent was assayed in vitro against four NSCLC cell lines (A549, H460, H322, and H1299). Ad-Luc was used as a nonspecific control agent for gene transfer. The dose effect of each agent on tumor cell growth was subjected to the median-effect plot (29
, 38)
. The concentrations of agents that caused 50 and 80% inhibition of growth (ID50 and ID80) were determined (Table 1)
|
|
|
Isobolograms Analysis of Three-Agent Combinations.
The effects of combinations of the three therapeutic agents under study (Ad-p53, docetaxel, and radiation) on tumor cell growth of four human NSCLC cells (H1299, H460, H322, and A549) were analyzed quantitatively and statistically in vitro (Fig. 3)
. The observed and predicated values for the combined effects are summarized in Table 2
. At both the ID50 and ID80 levels, the combination of docetaxel, Ad-p53, and radiation showed synergistic cancer cell cytotoxicity (P < 0.01 at ID50 and P < 0.001 at ID80) in all four cell lines tested. In contrast, the combination of docetaxel, Ad-Luc, and radiation produced no synergistic effects in any of the cell lines tested at either the ID50 or ID80 levels (Fig. 3, A and B
, D+L+R). These observations were consistent with those plotted on the two-dimensional isobolograms with Ad-Luc as a control and suggested that Ad-p53 made an important contribution to the observed synergistic inhibition of NSCLC cell growth by the three-agent combination of docetaxel, Ad-p53, and radiation.
|
-radiation. For each experiment, mice were divided into 10 treatment groups: four control groups (PBS, docetaxel alone, Ad-p53 alone, and radiation alone); three groups treated with two-agent combinations (docetaxel + Ad-p53, docetaxel + radiation, and Ad-p53 + radiation); and three groups treated with three-agent combinations of docetaxel + Ad-p53 + radiation, docetaxel + Ad-Luc + radiation, and docetaxel + Ad-EV + radiation). Each treatment group contained five mice, and all experiments were repeated twice.
|
-radiation significantly inhibited tumor growth in both the H1299 (Fig. 4A)| DISCUSSION |
|---|
|
|
|---|
A major impediment to successful therapy, however, is the high rate of local failure (>40%; Ref. 2 ). One possible solution to this problem is to increase the radiation dose to the tumor by using radiosensitizing agents and adding an additional local treatment such as gene therapy. The potential radiation sensitizing effects of certain genes that are inactivated during cancer progression make this an attractive approach. Promising therapeutic responses have been demonstrated in advanced NSCLC treated with wild- type p53 delivered locally with recombinant adenoviral vectors alone or in combination with chemotherapy such as the DNA-damaging agent cisplatin (20 , 22 , 24 , 25) .
An aggressive treatment approach combining new chemotherapeutic agents such as docetaxel, gene therapy with Ad-p53, and radiotherapy may improve the outcome for patients with advanced NSCLC. However, what is currently lacking is critical information on the cytotoxicity and target response to such a complicated combined-modality therapy. This study was intended to help fill this gap. In brief, we have developed methods for two- and three-dimensional isobologram modeling and for statistical analysis to evaluate triple combined-modality therapy in human NSCLC cell lines in vitro and in animal models in vivo. We used the improved isobologram method of Steel and Peckham for evaluating the interaction of combined treatments in vitro (35) . This method, generally, is much more stringent for analyzing synergism or antagonism than other methods because it coordinates a wider area of uncertainty (envelope of additivity) instead of an additive line. Thus, the possibility of predicting false-positive synergistic or antagonistic interactions, a problem inherent in other methods, was eliminated. The original methods of constructing isobolograms were very complicated and usually involved plotting by hand. We improved the method for making isobolograms for this study by using computer-based dose-response curves and fitting them with the median-effect equation (43) . This reduced the inaccuracy and inconvenience of the conventional hand-plotting method. We also developed mathematical and statistical analytical methods for producing sophisticated three-dimensional isobolograms that could be used to evaluate the interaction of the three-agent combinations.
In the present study, we evaluated the interaction of three agents in either two-agent or three-agent combinations in four NSCLC cell lines at both ID50 and ID80 levels. We used both ID levels because, although isobologram analysis of combined effects is commonly done at the ID50 level, isobologram analysis at the ID80 level is more relevant to clinical cancer treatment settings. We found that the combination of Ad-p53 + radiation produced significantly synergistic effects in all of the cell lines tested at both the ID50 and ID80 levels, whereas the combinations of docetaxel + Ad-p53 and docetaxel + radiation produced mixed effects ranging between additive and synergistic. The three-agent combination also produced significantly synergistic effects in all of the cell lines tested at both the ID50 and ID80 levels. No antagonistic effects were observed for any two- or three-agent combinations. On the other hand, the combinations containing control vector (docetaxel + Ad-Luc, Ad-Luc + radiation, and docetaxel + Ad-Luc + radiation) showed only additive or additive-to-antagonistic effects in all of the cell lines tested, and no synergistic effects, despite doses of Ad-Luc that exceeded Ad-p53. Together, these findings suggest that the adenoviral vector-mediated wild-type p53 gene therapy specifically induced the observed suppression of NSCLC cell growth. Interestingly, although H460 and A549 cells both contain the wild-type p53 gene and showed resistance to single-agent Ad-p53 treatment in our studies, they also showed a significant response to combination treatments that included Ad-p53 with docetaxel and radiation; moreover, the growth of both cell lines was synergistically inhibited. Together, these results imply that the application of this combined modality therapy might be broadened to include cancers that show resistance to a single therapeutic agent.
Cellular responses to ionizing radiation and to many chemotherapeutics are frequently attributed to (a) the cytotoxicity of such agents to actively proliferating cells and (b) the apoptosis they induce (44 , 45) . Because apoptosis is a genetically programmed cellular response to environmental stresses or stimuli, mutations in apoptotic pathways could theoretically result in cellular resistance to therapeutic drugs and radiation treatment. For example, the status of the p53 tumor suppressor gene in tumor cells has been shown to be a strong determinant of cellular response to treatment with either radiation or chemotherapy; the vulnerability of tumor cells to radiation or chemotherapy is greatly reduced by mutations that abolish p53-dependent apoptosis (17 , 46, 47, 48) . These results suggested that the inactivation of p53 might produce treatment-resistance of tumor cells to docetaxel chemotherapy and radiotherapy. On the other hand, they suggest that restoration of p53 function in p53-deficient cells or overexpression of exogenous p53 in p53-wild type tumor cells might overcome the cellular resistance and enhance the cellular response to either chemotherapy or radiotherapy via a mechanism leading to p53-dependent apoptosis. Together, our results strongly suggest that the combined-modality therapy studied here with docetaxel chemotherapy, Ad-p53 gene therapy, and ionizing radiotherapy might be an effective therapeutic option for patients with advanced NSCLC as well as other types of cancers. In addition, our two- and three-dimensional isobologram models could be useful for predicting the outcomes of complicated therapeutic combinations in vitro and in vivo and for providing critical supporting data for clinical trials.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 Partially supported by grants [a Specialized Program of Research Excellence (SPORE) in Lung Cancer Grant P50-CA70907 and Grant PO1 CA78778] from the National Cancer Institute (NCI) and NIH; a Keck Gene Therapy Career Development Grant (to L. J.); a NCI Cancer Center Support Core Grant CA16672 and a grant from the Tobacco Settlement Funds as appropriated by the Texas State Legislature (Project 8) to The University of Texas (U. T.) M. D. Anderson Cancer Center; and by gifts from Tenneco and Exxon in support of the Core Laboratory Facility at the U. T. Division of Surgery and Anesthesiology. ![]()
2 To whom requests for reprints should be addressed, at Department of Thoracic and Cardiovascular Surgery, Box 109, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. Phone: (713) 792-6935; Fax: (713) 794-4901; E-mail: lji{at}mail.mdanderson.org ![]()
3 The abbreviations used are: NSCLC, non-small cell lung cancer; MOI, multiplicity/multiplicities of infection; vp, viral particle(s). ![]()
Received 3/ 1/01; revised 5/24/01; accepted 5/24/01.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. Jayachandran, J.-i. Sazaki, M. Nishizaki, K. Xu, L. Girard, J. D. Minna, J. A. Roth, and L. Ji Fragile Histidine Triad Mediated Tumor Suppression of Lung Cancer by Targeting Multiple Components of the Ras/Rho GTPase Molecular Switch Cancer Res., November 1, 2007; 67(21): 10379 - 10388. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Ohtani, A. Iwamaru, W. Deng, K. Ueda, G. Wu, G. Jayachandran, S. Kondo, E. N. Atkinson, J. D. Minna, J. A. Roth, et al. Tumor Suppressor 101F6 and Ascorbate Synergistically and Selectively Inhibit Non-Small Cell Lung Cancer Growth by Caspase-Independent Apoptosis and Autophagy Cancer Res., July 1, 2007; 67(13): 6293 - 6303. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Mazumder, D. Plesca, M. Kinter, and A. Almasan Interaction of a Cyclin E Fragment with Ku70 Regulates Bax-Mediated Apoptosis Mol. Cell. Biol., May 1, 2007; 27(9): 3511 - 3520. [Abstract] [Full Text] [PDF] |
||||
![]() |
W.-G. Deng, H. Kawashima, G. Wu, G. Jayachandran, K. Xu, J. D. Minna, J. A. Roth, and L. Ji Synergistic Tumor Suppression by Coexpression of FUS1 and p53 Is Associated with Down-regulation of Murine Double Minute-2 and Activation of the Apoptotic Protease-Activating Factor 1-Dependent Apoptotic Pathway in Human Non-Small Cell Lung Cancer Cells Cancer Res., January 15, 2007; 67(2): 709 - 717. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Ueda, H. Kawashima, S. Ohtani, W.-G. Deng, M. Ravoori, J. Bankson, B. Gao, L. Girard, J. D. Minna, J. A. Roth, et al. The 3p21.3 Tumor Suppressor NPRL2 Plays an Important Role in Cisplatin-Induced Resistance in Human Non-Small-Cell Lung Cancer Cells Cancer Res., October 1, 2006; 66(19): 9682 - 9690. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Zhang, R. M. Cheung, R. Komaki, B. Fang, and J. Y. Chang Radiotherapy Sensitization by Tumor-Specific TRAIL Gene Targeting Improves Survival of Mice Bearing Human Non-Small Cell Lung Cancer Clin. Cancer Res., September 15, 2005; 11(18): 6657 - 6668. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Teraishi, S. Wu, J. Sasaki, L. Zhang, H.-B. Zhu, J. J. Davis, and B. Fang P-Glycoprotein-Independent Apoptosis Induction by a Novel Synthetic Compound, MMPT [5-[(4-Methylphenyl)methylene]-2-(phenylamino)-4(5H)-thiazolone] J. Pharmacol. Exp. Ther., July 1, 2005; 314(1): 355 - 362. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Nishizaki, J.-i. Sasaki, B. Fang, Edward. N. Atkinson, J. D. Minna, J. A. Roth, and L. Ji Synergistic Tumor Suppression by Coexpression of FHIT and p53 Coincides with FHIT-Mediated MDM2 Inactivation and p53 Stabilization in Human Non-Small Cell Lung Cancer Cells Cancer Res., August 15, 2004; 64(16): 5745 - 5752. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. K. Sah, A. Munshi, T. Nishikawa, T. Mukhopadhyay, J. A. Roth, and R. E. Meyn Adenovirus-mediated wild-type p53 radiosensitizes human tumor cells by suppressing DNA repair capacity Mol. Cancer Ther., November 1, 2003; 2(11): 1223 - 1231. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Danesi, F. De Braud, S. Fogli, T. M. De Pas, A. Di Paolo, G. Curigliano, and M. Del Tacca Pharmacogenetics of Anticancer Drug Sensitivity in Non-Small Cell Lung Cancer Pharmacol. Rev., March 1, 2003; 55(1): 57 - 103. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |