Immune responses against major histocompatibility complex and interleukin-2 gene transfected K1735 murine melanoma: Potential vaccines for the immunotherapy of cancer

Tumor specific immunity is mediated by cytotoxic T lymphocytes (CTL) that recognize peptide antigen (Ag) in the context of major histocompatibility complex (MHC) class I molecules and by helper T (Th) lymphocytes that recognize peptide Ag in the context of MHC class II molecules. The purpose of this study is (1) to induce or augment the immunogenicity of nonimmunogenic or weakly immunogenic tumors by genetic modification of tumor cells, and (2) to use these genetically altered cells in cancer immunotherapy. To study this, I transfected a highly tumorigenic murine melanoma cell line (K1735) that did not express constitutively either MHC class I or II molecules with syngeneic cloned MHC class I and/or class II genes, and then determined the tumorigenicity of transfected cells in normal C3H mice. K1735 transfectants expressing either $\rm K\sp{k}$ or $\rm A\sp{k}$ molecules alone produced tumors in normal C3H mice, whereas most transfectants that expressed both molecules were rejected in normal C3H mice but produced tumors in nude mice. The rejection of K1735 transfectants expressing $\rm K\sp{k}$ and $\rm A\sp{k}$ Ag in normal C3H mice required both $\rm CD4\sp+$ and $\rm CD8\sp+$ T cells. Interestingly, the $\rm A\sp{k}$ requirement can be substituted by IL-2 because transfection of $\rm K\sp{k}$-positive/A$\sp{\rm k}$-negative K1735 cells with the IL-2 gene also resulted in abrogation of tumorigenicity in normal C3H mice but not in nude mice. In addition, 1735 $(\rm I\sp+II\sp+)$ transfected cells can function as antigen presenting cells (APC) since they could process and present native hen egg lysozyme (HEL) to HEL specific T cell hybridomas. Furthermore, the transplantation immunity induced by K1735 transfectants expressing both $\rm K\sp{k}$ and $\rm A\sp{k}$ molecules completely cross-protected mice against challenge with $\rm K\sp{k}$-positive transfectants but weakly protected them against challenge with parental K1735 cells or $\rm A\sp{k}$-positive transfectants. Finally, I demonstrated that MHC $(\rm I\sp+II\sp+)$ or $\rm K\sp{k}$-positive/IL-2-positive cells can function as anti-cancer vaccines since they can abrogate the growth of established tumors and metastasis.^ In summary, my results indicate that expression of either MHC class I or II molecule alone is insufficient to cause the rejection of K1735 melanoma in syngeneic hosts and that both molecules are necessary. In addition, my data suggest that the failure of $\rm K\sp{k}$-positive K1735 cells to induce a primary tumor-rejection response in normal C3H mice may be due to their inability to induce the helper arm of the anti-tumor immune response. Finally, the ability of MHC $(\rm I\sp+II\sp+)$ or $\rm K\sp{k}$-positive/IL-2-positive cells to prevent growth of established tumors or metastasis suggests that these cell lines can serve as potential vaccines for the immunotherapy of cancer. (Abstract shortened by UMI.) ^

p75 receptor-mediated neurotrophism: A new mechanism of melanoma cell survival and invasion

Trophism as a "clonal dominance" support mechanism for tumor cells is an unexplored area of tumor progression. This report presents evidence that the human melanoma low-affinity neurotrophin receptor (p75) can signal independently of its high-affinity tyrosine kinase counterparts, the TRK family of kinases. Signaling may be accomplished by a p75-associated purine-analog-sensitive kinase and results in enhanced invasion into a reconstituted basement membrane with a corresponding stimulation of matrix metalloproteinase-2 expression. Additionally, a "stress culture" survival assay was developed to mimic the growth limiting conditions encountered by melanoma cells in a rapidly growing primary tumor or metastatic deposit prior to neoangiogenesis. Under these conditions, p75, promotes the survival of high p75 expressing brain-colonizing melanoma cells. Extensive 70W melanoma cell-cell contact, which downregulates p75, immediately precedes the induction of cell death associated with diminished production of two key cell survival factors, bcl-2 and the p85 subunit of phosphoinositol-3-kinase, and an elevation in apoptosis promoting intracellular reactive oxygen species (ROSs). Since one function of bcl-2 may be to control the generation of ROSs via the antioxidant pathway, these cells may receive a apoptosis-prompting "double hit". 70W melanoma cell death occurred by an apoptotic mechanism displaying classical morphological changes including plasma membrane blebbing, loss of microvilli and redistribution of ribosomes. 70W apoptosis could be pharmacologically triggered following anti-p75 monoclonal antibody-mediated clustering of p75 receptors. 70W cells fluorescently sorted for high-p75 expression (p75$\sp{\rm H}$ cells) exhibited an augmented survival potential and a predilection to sort with the S + G2/M growth phase, relative to their low p75 expressing, p75$\sp{\rm L}$ counterparts. Apoptosis is significantly delayed by p75$\sp{\rm H}$ cells, whereas p75$\sp{\rm L}$ cells are exquisitely prone to initiate apoptosis. Importantly, the p75$\sp{\rm L}$ cells that survive apoptosis, highly re-expressed p75 and were remarkably responsive to exogenous NGF.^ These are the first data to implicate p75-mediated neurotrophism as an invasion and survival support mechanism employed by brain-metastatic cells. In particular, these results may have implications in little understood phenomena of tumor progression, such as the emergence of "clonal dominance" and tumor dormancy. ^

LINKAGE RELATIONSHIPS AMONG PROTEIN-CODING LOCI IN FISHES OF THE GENUS XIPHOPHORUS

Inbred strains of three species of fishes of the genus Xiphophorus (platyfish and swordtails) were crossed to produce intra- and interspecific F(,1) hybrids, which were then backcrossed to one or both parental stocks. Backcross hybrids were used for the analysis of segregation and linkage of 33 protein-coding loci (whose products were visualized by starch gel electrophoresis) and a sex-linked pigment pattern gene. Segregation was Mendelian for all loci with the exception of one instance of segregation distortion. Six linkage groups of enzyme-coding loci were established: LG I, ADA --6%-- G(,6)PD --24%-- 6PGD; LG II, Est-2 --27%-- Est-3 --0%-- Est-5 --23%-- LDH-1 --16%-- MPI; LG III, AcPh --38%-- G(,3)PD-1 (GUK-2 --14%-- G(,3)PD-1 is also in LG III, but the position of GUK-2 with respect to AcPh has not yet been determined); LG IV, GPI-1 --41%-- IDH-1; LG V, Est-1 --38%-- MDH-2; and LG VI, P1P --7%-- UMPK-1 (P1P is a plasma protein, very probably transferrin).^ Sex-specific recombination appeared absent in LG II and LG IV locus pairs; significantly higher male recombination was demonstrated in LG I but significantly higher female recombination was detected in LG V. Only one significant population-specific difference in recombination was detected, in the G(,6)PD - 6PGD region of LG I; the notable absence of such effects implies close correspondence of the genomes of the species used in the study. Two cases of possible evolutionary conservation of linkage groups in fishes and mammals were described, involving the G(,6)PD - 6PGD linkage in LG I and the cluster of esterase loci in LG II. One clear case of divergence was observed, that of the linkage of ADA in LG I. It was estimated that a minimum of (TURN)50% of the Xiphophorus genome was marked by the loci studied. Therefore, the prior probability that a new locus will assort independently from the markers already established is estimated to be less than 0.5. A maximum of 21 of the 24 pairs of chromosomes could be marked with at least one locus.^ Only the two LG V loci showed a significant association with a postulated gene controlling the severity of a genetically controlled melanoma caused by abnormal proliferation of macromelanophore pigment pattern cells. The independence of melanotic severity from all other informative markers implies that one or at most a few major genes are involved in control of melanotic severity in this system. ^

Role of the transcription factor activator protein-2alpha in prostate carcinogenesis

Activator protein 2α (AP-2) is a transcription factor known to play a crucial role in the progression of malignant melanoma, colorectal carcinoma, and breast cancer. Several AP-2 target genes are known to be deregulated in prostate cancer, therefore, we hypothesize that loss AP-2 expression plays a causal role in prostate carcinogenesis. Immunofluorescent staining for AP-2 of 30 radical prostatectomy specimens demonstrated that while AP-2 was highly expressed in normal prostate epithelium, its expression was lost in most cases of high grade prostatic intraepithelial neoplasia (PIN), and all cases of prostate cancer studied. Additional analyses demonstrated that AP-2 was associated with normal luminal differentiation and it was not expressed in the basal cell layer. In cell lines, AP-2 was strongly expressed in immortalized normal prostate epithelial cells, whereas low expression was observed in the LNCaP, LNCaP-LN3, and PC3M-LN4 prostate cancer cell lines. Transfection of the highly tumorigenic and metastatic cell line PC3M-LN4 with the AP-2 gene significantly decreased tumor growth in the prostate of nude mice (p = 0.032) and inhibited metastases to the lymph nodes. Moreover, transfection of the low tumorigenic, low metastatic cell line LNCaP-LN3 with full length AP-2; resulted in complete inhibition of tumor incidence in the AP-2 transfectants (0/19) vs. neo control (10/16). A potential mechanism for this loss of tumorigenicity was the modulation of gene expression in prostate cancer cells that mimicked the normal phenotype. Analysis of differential expression between neo control- and AP-2-transfected cells in vitro and in tumors demonstrated low VEGF expression in AP-2 transfectants. We further demonstrated that AP-2 acted as a transcriptional repressor of the VEGF promoter by binding to a GC-rich region located between −88 and −66. This region contains an AP-2 consensus element overlapping two Sp1 consensus elements. We found that Sp3 and AP-2 bound to this region in a mutually exclusive manner to promote activation or repression. Increased VEGF expression has been observed in high grade PIN and in prostate cancer. Here we provide evidence that this early molecular change could be a result of loss of AP-2 expression in the prostatic epithelium. ^

The role of the sonic hedgehog signaling pathway in epidermal homeostasis

Non-melanoma skin cancer is the most frequently diagnosed malignancy in the United States of which basal cell carcinoma (BCC) accounts for 65%. It has recently been determined that deregulation of the sonic hedgehog (shh) pathway leads to the development of BCC. Shh, gli-1, gli-2 gli-3, ptc and smo are overexpressed in BCC and overexpression of these genes in the epidermis results in formation of BCC-like tumors. Despite these observations, the mechanisms by which the pathway controls epidermal homeostasis and the development of the malignant phentotype are unknown. This study assessed the role of the shh pathway in epidermal homeostasis through regulation of apoptosis and differentiation. ^ The anti-apoptotic protein, bcl-2 is overexpressed in BCC, however transcriptional regulators of bcl-2 in the epidermis are unknown. Transient transfection of primary keratinocytes with gli-1 resulted in an increase of bcl-2 expression. Database analysis revealed seven candidate gli binding sites on the bcl-2 promoter. Cotransfection of increasing amounts of gli-1 in keratinoycytes resulted in a corresponding dose-dependent increase in bcl-2 promoter luciferase activity. An N-terminal mutant of gli-3 inhibited gli-1 transactivation of the bcl-2 promoter. The region −428 to −420 was found to be important for gli-1 regulation through gel shift, luciferase assays and site-directed mutagenesis. ^ In order to assess the ability of the shh pathway to regulate keratinocyte differentiation, HaCaT keratinocytes overexpressing sonic hedgehog, were grown in organotypic raft culture. Overexpression of shh induced a basal cell phenotype compared to vector control, as evidenced by transmural staining of cytokeratin 14 and altered Ki67 staining. Shh also induced keratinocyte invasion into the underlying collagen. This was associated with increased phosphorylation of EGFR, jnk and raf and increased expression of c-jun, mmp-9 and Ki67. Interestingly, shh overexpression in HaCaTs did not induce the typical downstream effects of shh signaling, suggesting a gli-independent mechanism. Sonic hedgehog's ability to induce an invasive phenotype was found to be dependent on activation of the EGF pathway as inhibition of EGFR activity with AG1478 and c-225 was able to reduce the invasiveness of HaCaT shh keratinocytes, whereas treatment with EGF augmented the invasiveness of the HaCaT shh clones. ^ These studies reveal the importance of the sonic hedgehog pathway in epidermal homeostasis by regulation of apoptosis through bcl-2, and control of keratinocyte differentiation and invasion through activation of the EGF pathway. They further suggest potential mechanisms by which deregulation of the shh pathway may lead to the development of the malignant phenotype. ^

Mechanisms involved in suppression of the elicitation of immune responses by ultraviolet light

The ultraviolet radiation (UVR) present in sunlight is the primary cause of nonmelanoma skin cancer and has been implicated in the development of cutaneous malignant melanoma. Ultraviolet radiation also suppresses the immune response. In the majority of studies investigating the mechanisms regulating UV-induced immune suppression, UV is used to suppress the induction of immune responses. Equally important, is the ability of UVR to suppress established immune responses, such as the recall reaction in humans, which protects against microbial infections. We established a murine model to help elucidate the immunological mechanisms governing UV-induced suppression of the elicitation of immune responses. 80 kJ/m2 of UVR nine days after sensitization consistently suppressed the elicitation of delayed type hypersensitivity reaction to C. albicans . We found ultraviolet A (320±400 nm) radiation was as effective as solar-simulated ultraviolet A + B (290±400 nm) in suppressing the elicitation of an established immune response. The mechanisms involved in UV-induced suppression of the induction & elicitation of the immune response are similar. For example, mice irradiated with UV after immunization generated antigen-specific T suppressor cells. Injection of monoclonal antibodies to IL-10 or recombinant IL-12 immediately after exposure to UVR blocked immune suppression. Liposomes containing bacteriophage T4N5 to the skin of mice also prevented immune suppression, demonstrating an essential role for ultraviolet-induced DNA damage in the suppression of established immune reactions. ^ In addition to damaging DNA, UV initiates immune suppression through the isomerization of urocanic acid in the epidermis. Here we provide evidence that cis-UCA induces systemic immunosuppression via the serotonin (5-hydroxyyryptamine; 5-HT) receptor. Biochemical and immunological analysis indicate that cis-UCA binds to, and activates, the serotonin receptor. Moreover, serotonin specific antibodies block UV- and/or cis-UCA-induced immune suppression. Our findings identify cis-UCA as novel serotonin receptor ligand and indicate that serotonin receptor engagement can activate immune suppression. Cumulatively, our data suggest that similar immune regulatory mechanisms are activated regardless of whether we expose mice to solar-simulated UV (UVA + UVB) radiation or UVA only, and that ultraviolet radiation activates similar immunologic pathways to suppress the induction or the elicitation of the immune response. ^

Evaluating minimal important differences for the FACT-Melanoma quality of life questionnaire

Objectives. Minimal Important Differences (MIDs) establish benchmarks for interpreting mean differences in clinical trials involving quality of life outcomes and inform discussions of clinically meaningful change in patient status. As such, the purpose of this study was to assess MIDs for the Functional Assessment of Cancer Therapy–Melanoma (FACT-M). ^ Methods. A prospective validation study of the FACT-M was performed with 273 patients with stage I to IV melanoma. FACT-M, Karnofsky Performance Status (KPS), and Eastern Cooperative Oncology Group Performance Status (ECOG-PS) scores were obtained at baseline and 3 months following enrollment. Anchor- and distribution-based methods were used to assess MIDs, and the correspondence between MID ranges derived from each method was evaluated. ^ Results. This study indicates that an approximate range for MIDs of the FACT-M subscales is between 5 to 8 points for the Trial Outcome Index, 4 to 5 points for the Melanoma Combined Subscale, 2 to 4 points for the Melanoma Subscale, and 1 to 2 points for the Melanoma Surgery Subscale. Each method produced similar but not identical ranges of MIDs. ^ Conclusions. The properties of the anchor instrument employed to derive MIDs directly affect resulting MID ranges and point values. When MIDs are offered as supportive evidence of a clinically meaningful change, the anchor instrument used to derive thresholds should be clearly stated along with evidence supporting the choice of anchor instrument as the most appropriate for the domain of interest. In this analysis, the KPS was a more appropriate measure than the ECOG-PS for assessing MIDs. ^

Psychosocial measures reported by parents in studies of skin cancer prevention

Background. Because it is important to minimize children's sun exposure to reduce skin cancer risk, much of the extensive skin cancer prevention literature consists of studies of children's sun protection, sun avoidance and ultraviolet radiation (UVR) exposure. Little attention has been focused on the measurement of psychosocial constructs in these studies. Identification of the psychosocial correlates or determinants of children's skin cancer risk or risk-reduction behavior is critical to more fully understand and predict behavior. Furthermore, psychosocial variables may be influenced by interventions to reduce risk. Thus, it is important to examine the psychosocial measures used in studies of children's skin cancer prevention. Information on the validity and reliability of psychosocial measures may increase confidence in study findings based on these measures. In particular, self-efficacy and barriers are key constructs in several major theoretical frameworks and parental measures have been associated with children's sun protection. However, there is conceptual overlap of self-efficacy and barriers measures and little is known about the psychometric properties of these measures.^ Study Aims and Methods. The overall goal of this dissertation was to examine the measurement of psychosocial constructs relevant to children's skin cancer prevention. Because children depend primarily on their parents for skin cancer prevention, measures of parents' psychosocial constructs are the focus. Study 1 was a systematic review of parental psychosocial measures used in studies of children's sun protection, sun avoidance and UVR exposure. The specific aims of Study 1 were to (1) describe psychosocial measures reported by parents, including available information on the psychometric properties of these measures and their use in analyses and (2) provide recommendations for the development, refinement and standardized reporting of measures. ^ Study 2 examined the psychometric properties of measures of parental self-efficacy and barriers regarding children's sun protection. Melanoma patients (N=205) who were parents of children ≤ 12 years of age completed a telephone interview that included self-efficacy and barriers measures specific to sunscreen, clothing, shade and limiting time outdoors. The specific aims of Study 2 were to (1) use a confirmatory factor analytic approach to examine the factorial validity of parental self-efficacy and barriers measures, (2) examine the convergent and discriminant validity of behavior-specific measures of self-efficacy and barriers and (3) assess the reliability of item and scale measures.^ Results. In Study 1, a search of standard databases yielded 48 eligible studies. Most studies assessed only one or two psychosocial constructs. Knowledge was measured most frequently. There was little discussion of measure source, development, theoretical background or psychometric properties, besides internal consistency reliability. There was conceptual overlap of some measures. In Study 2, confirmatory factor analytic findings supported the factorial validity of the self-efficacy and barriers measures. When all eight self-efficacy and barriers measures were included in the same model, a modified eight-factor model adequately fit the data, providing preliminary evidence that the measures are distinct. Measure associations supported the convergent validity of all measures and the discriminant validity of most measures. The self-efficacy and barriers measures were reliable.^ Conclusions. Recommendations based on the literature review include developing and refining psychosocial measures based on theory. Describing a measure's theoretical basis and psychometric properties would facilitate critical evaluation. Standardized reporting of source, development, theory, construct, items and analytic role would facilitate comparison of findings, continual refinement and future applications of measures. In the validation study, self-efficacy and barriers measures were examined in a sample of parents with a personal history of melanoma. Findings suggested that these measures are valid and reliable for use in studies of children's sun protection. There was preliminary evidence that these measures are distinct but additional study is needed. ^

Assessment of cancer incidence in Titus County, Texas, 1977--1984

The purpose of this study was to determine the incidence of cancer in Titus County, Texas, through the identification of all cases of cancer that occurred in residents of the county during the period from 1977 to 1984. Data gathered from Texas Cancer Registry, hospital records, and death certificates were analyzed with regard to anatomic site, race, sex, age, city of residence, and place of birth. Adjustment of incidence rates by sex and race allowed comparisons with U.S. rates provided by the Surveillance, Epidemiology, and End Results Program (SEER).^ Seven hundred sixty-six (766) cancer cases were identified for the eight year period during 171,536 person-years of observation. In whites, statistically significant standardized incidence ratios (SIR) were found for leukemia (males SIR = 2.70 and females SIR = 2.26), melanoma (males SIR = 1.90 and females SIR = 2.25), lung (males SIR = 1.45) and for multiple myeloma (both sexes combined SIR = 1.86). In blacks, significant excess numbers of cases were found for Hodgkin's disease (males SIR = 8.33 and females SIR = 13.3) and for esophagus and bone considering both sexes together (SIR = 2.68 and 12.54, respectively). Rates for blacks were based on a small population and therefore unstable. A statistically significant excess number of cases for all sites combined was found in Mount Pleasant residents (age-adjusted incidence rate = 563.6 per 100,000 per year).^ A review of possible environmental risk factors in the area: hazardous waste disposal site, lignite deposits, and petrochemical and poultry industries are presented. A need for further epidemiological and environmental studies to identify etiological factors that could be responsible for the excess number of leukemia cases are recommended. For melanoma, a public health educational program to teach the population methods of protection from sun exposure is also suggested. ^

THE CYTOPLASMIC TAIL OF MHC CLASS I MOLECULES PLAYS A CRITICAL ROLE IN DENDRITIC CELL-INDUCED T CELL IMMUNITY

The presentation of MHC class I (MHC-I)/peptide complexes by dendritic cells (DCs) is critical for the maintenance of central tolerance to self and for the regulation of cytotoxic T lymphocytes (CTL)-mediated adaptive immune responses against pathogens and cancer cells. Interestingly, several findings have suggested that the cytoplasmic tail of MHC class I plays a functional role in the regulation of CTL immune responses. For example, our previous studies demonstrated that exon 7-deleted MHC-I molecules not only showed extended DC cell surface half-lives but also induced significantly increased CTL responses to viral challange invivo. Although exon 7-deleted variant of MHC-I does not occur naturally in humans, the animal studies prompted us to examine whether exon 7-deleted MHC-I molecules could generate augmented CTL responses in a therapeutic DC-based vaccine setting. To examine the stimulatory capacity of exon 7-deleted MHC-I molecules, we generated a lentivirus-mediated gene transfer system to induce the expression of different MHC-I cytoplasmic tail isoforms in both mouse and human DCs. These DCs were then used as vaccines in a melanoma mouse tumor model and in a human invitro co-culture system. In this thesis, we show that DCs expressing exon 7-deleted MHC-I molecules, stimulated remarkably higher levels of T-cell cytokine production and significantly increased the proliferation of meanoma-specific (Pmel-1) T cells compared with DCs expressing wild type MHC-I. We also demonstrate that, in combination with adoptive transfer of Pmel-1 T-cell, DCs expressing exon 7-deleted Db molecules induced greater anti-tumor responses against established B16 melanoma tumors, significantly extending mouse survival as compared to DCs expressing wild-type Db molecules. Moreover, we also observed that human DCs expressing exon 7-deleted HLA-A2 molecules showed similarly augmented CTL stimulatory ability. Mechanistic studies suggest that exon 7-deleted MHC-I molecules showed impaired lateral membrane movement and extended cell surface half-lives within the DC/T-cell interface, leading to increased spatial availability of MHC-I/peptide complexes for recognition by CD8+ T cells. Collectively, these results suggesr that targeting exon 7 within the cytoplasmic tail of MHC-I molecules in DC vaccines has the potential to enhance CD8+ T cell stimulatory capacity and improve clinical outcomes in patients with cancer or viral infections.

Novel Imaging-Based Techniques Reveal a Role for PD-1/PD-L1 in Tumor Immune Surveillance in the Lung

The binding of immune inhibitory receptor Programmed Death 1 (PD-1) on T cells to its ligand PD-L1 has been implicated as a major contributor to tumor induced immune suppression. Clinical trials of PD-L1 blockade have proven effective in unleashing therapeutic anti-tumor immune responses in a subset of patients with advanced melanoma, yet current response rates are low for reasons that remain unclear. Hypothesizing that the PD-1/PD-L1 pathway regulates T cell surveillance within the tumor microenvironment, we employed intravital microscopy to investigate the in vivo impact of PD-L1 blocking antibody upon tumor-associated immune cell migration. However, current analytical methods of intravital dynamic microscopy data lack the ability to identify cellular targets of T cell interactions in vivo, a crucial means for discovering which interactions are modulated by therapeutic intervention. By developing novel imaging techniques that allowed us to better analyze tumor progression and T cell dynamics in the microenvironment; we were able to explore the impact of PD-L1 blockade upon the migratory properties of tumor-associated immune cells, including T cells and antigen presenting cells, in lung tumor progression. Our results demonstrate that early changes in tumor morphology may be indicative of responsiveness to anti-PD-L1 therapy. We show that immune cells in the tumor microenvironment as well as tumors themselves express PD-L1, but immune phenotype alone is not a predictive marker of effective anti-tumor responses. Through a novel method in which we quantify T cell interactions, we show that T cells are largely engaged in interactions with dendritic cells in the tumor microenvironment. Additionally, we show that during PD-L1 blockade, non-activated T cells are recruited in greater numbers into the tumor microenvironment and engage more preferentially with dendritic cells. We further show that during PD-L1 blockade, activated T cells engage in more confined, immune synapse-like interactions with dendritic cells, as opposed to more dynamic, kinapse-like interactions with dendritic cells when PD-L1 is free to bind its receptor. By advancing the contextual analysis of anti-tumor immune surveillance in vivo, this study implicates the interaction between T cells and tumor-associated dendritic cells as a possible modulator in targeting PD-L1 for anti-tumor immunotherapy.

BAYESIAN STATISTICAL METHODS IN GENE-ENVIRONMENT AND GENE-GENE INTERACTION STUDIES

Complex diseases such as cancer result from multiple genetic changes and environmental exposures. Due to the rapid development of genotyping and sequencing technologies, we are now able to more accurately assess causal effects of many genetic and environmental factors. Genome-wide association studies have been able to localize many causal genetic variants predisposing to certain diseases. However, these studies only explain a small portion of variations in the heritability of diseases. More advanced statistical models are urgently needed to identify and characterize some additional genetic and environmental factors and their interactions, which will enable us to better understand the causes of complex diseases. In the past decade, thanks to the increasing computational capabilities and novel statistical developments, Bayesian methods have been widely applied in the genetics/genomics researches and demonstrating superiority over some regular approaches in certain research areas. Gene-environment and gene-gene interaction studies are among the areas where Bayesian methods may fully exert its functionalities and advantages. This dissertation focuses on developing new Bayesian statistical methods for data analysis with complex gene-environment and gene-gene interactions, as well as extending some existing methods for gene-environment interactions to other related areas. It includes three sections: (1) Deriving the Bayesian variable selection framework for the hierarchical gene-environment and gene-gene interactions; (2) Developing the Bayesian Natural and Orthogonal Interaction (NOIA) models for gene-environment interactions; and (3) extending the applications of two Bayesian statistical methods which were developed for gene-environment interaction studies, to other related types of studies such as adaptive borrowing historical data. We propose a Bayesian hierarchical mixture model framework that allows us to investigate the genetic and environmental effects, gene by gene interactions (epistasis) and gene by environment interactions in the same model. It is well known that, in many practical situations, there exists a natural hierarchical structure between the main effects and interactions in the linear model. Here we propose a model that incorporates this hierarchical structure into the Bayesian mixture model, such that the irrelevant interaction effects can be removed more efficiently, resulting in more robust, parsimonious and powerful models. We evaluate both of the 'strong hierarchical' and 'weak hierarchical' models, which specify that both or one of the main effects between interacting factors must be present for the interactions to be included in the model. The extensive simulation results show that the proposed strong and weak hierarchical mixture models control the proportion of false positive discoveries and yield a powerful approach to identify the predisposing main effects and interactions in the studies with complex gene-environment and gene-gene interactions. We also compare these two models with the 'independent' model that does not impose this hierarchical constraint and observe their superior performances in most of the considered situations. The proposed models are implemented in the real data analysis of gene and environment interactions in the cases of lung cancer and cutaneous melanoma case-control studies. The Bayesian statistical models enjoy the properties of being allowed to incorporate useful prior information in the modeling process. Moreover, the Bayesian mixture model outperforms the multivariate logistic model in terms of the performances on the parameter estimation and variable selection in most cases. Our proposed models hold the hierarchical constraints, that further improve the Bayesian mixture model by reducing the proportion of false positive findings among the identified interactions and successfully identifying the reported associations. This is practically appealing for the study of investigating the causal factors from a moderate number of candidate genetic and environmental factors along with a relatively large number of interactions. The natural and orthogonal interaction (NOIA) models of genetic effects have previously been developed to provide an analysis framework, by which the estimates of effects for a quantitative trait are statistically orthogonal regardless of the existence of Hardy-Weinberg Equilibrium (HWE) within loci. Ma et al. (2012) recently developed a NOIA model for the gene-environment interaction studies and have shown the advantages of using the model for detecting the true main effects and interactions, compared with the usual functional model. In this project, we propose a novel Bayesian statistical model that combines the Bayesian hierarchical mixture model with the NOIA statistical model and the usual functional model. The proposed Bayesian NOIA model demonstrates more power at detecting the non-null effects with higher marginal posterior probabilities. Also, we review two Bayesian statistical models (Bayesian empirical shrinkage-type estimator and Bayesian model averaging), which were developed for the gene-environment interaction studies. Inspired by these Bayesian models, we develop two novel statistical methods that are able to handle the related problems such as borrowing data from historical studies. The proposed methods are analogous to the methods for the gene-environment interactions on behalf of the success on balancing the statistical efficiency and bias in a unified model. By extensive simulation studies, we compare the operating characteristics of the proposed models with the existing models including the hierarchical meta-analysis model. The results show that the proposed approaches adaptively borrow the historical data in a data-driven way. These novel models may have a broad range of statistical applications in both of genetic/genomic and clinical studies.

Comparison of clinical and economic outcomes between high-intensity care and hospice care for the end-of-life melanoma patients

The intensity of care for patients at the end-of-life is increasing in recent years. Publications have focused on intensity of care for many cancers, but none on melanoma patients. Substantial gaps exist in knowledge about intensive care and its alternative, hospice care, among the advanced melanoma patients at the end of life. End-of-life care may be used in quite different patterns and induce both intended and unintended clinical and economic consequences. We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked databases to identify patients aged 65 years or older with metastatic melanoma who died between 2000 and 2007. We evaluated trends and associations between sociodemographic and health services characteristics and the use of hospice care, chemotherapy, surgery, and radiation therapy and costs. Survival, end-of-life costs, and incremental cost-effectiveness ratio were evaluated using propensity score methods. Costs were analyzed from the perspective of Medicare in 2009 dollars. In the first journal Article we found increasing use of surgery for patients with metastatic melanoma from 13% in 2000 to 30% in 2007 (P=0.03 for trend), no significant fluctuation in use of chemotherapy (P=0.43) or radiation therapy (P=0.46). Older patients were less likely to receive radiation therapy or chemotherapy. The use of hospice care increased from 61% in 2000 to 79% in 2007 (P =0.07 for trend). Enrollment in short-term (1-3 days) hospice care use increased, while long-term hospice care (≥ 4 days) remained stable. Patients living in the SEER Northeast and South regions were less likely to undergo surgery. Patients enrolled in long-term hospice care used significantly less chemotherapy, surgery and radiation therapy. In the second journal article, of 611 patients identified for this study, 358 (59%) received no hospice care after their diagnosis, 168 (27%) received 1 to 3 days of hospice care, and 85 (14%) received 4 or more days of hospice care. The median survival time was 181 days for patients with no hospice care, 196 days for patients enrolled in hospice for 1 to 3 days, and 300 days for patients enrolled for 4 or more days (log-rank test, P < 0.001). The estimated hazard ratios (HR) between 4 or more days hospice use and survival were similar within the original cohort Cox proportional hazard model (HR, 0.62; 95% CI, 0.49-0.78, P < 0.0001) and the propensity score-matched model (HR, 0.61; 95% CI, 0.47-0.78, P = 0.0001). Patients with ≥ 4 days of hospice care incurred lower end-of-life costs than the other two groups ($14,298 versus $19,380 for the 1- to 3-days hospice care, and $24,351 for patients with no hospice care; p < 0.0001). In conclusion, Surgery and hospice care use increased over the years of this study while the use of chemotherapy and radiation therapy remained consistent for patients diagnosed with metastatic melanoma. Patients diagnosed with advanced melanoma who enrolled in ≥ 4 days of hospice care experienced longer survival than those who had 1-3 days of hospice or no hospice care, and this longer overall survival was accompanied by lower end-of-life costs.^

Dissection of antitumor activity by lymphokine -activated killer cells: Role of FcR+ precursors and obligatory role of tumor necrosis factor-$\alpha$ in antibody -dependent cellular cytotoxicity

Human peripheral blood lymphocytes (PBL) cultured for varying lengths of time in IL-2 are able to mediate antibody independent cellular cytotoxicity (AICC) as well as antibody dependent cellular cytotoxicity (ADCC) against a wide range of tumor targets. The objective of our study is to determine the cytotoxic potential of the subset of LAK cells involved in ADCC, the tumor recognition mechanism in ADCC, the kinetics of ADCC mediated by PBL cultured under various conditions and the role of TNF-$\alpha$ in the development and maturation of ADCC effectors in the LAK population.^ The model system in this study for ADCC used a monoclonal antibody 14G2a (IgG2a), that recognizes the GD2 epitope on human melanoma cell line, SK-Mel-1. The target recognition mechanism operative in AICC (traditionally known as lymphokine activated killing or LAK) is an acquired property of these IL-2 activated cells which confers on them the unique ability to distinguish between tumor and normal cells. This recognition probably involves the presence of a trypsin sensitive N-linked glycoprotein epitope on tumor cells. Proteolytic treatment of the tumor cells with trypsin renders them resistant to AICC by PBL cultured in IL-2. However, ADCC is unaffected. This ADCC, mediated by the relatively small population of cells that are positive for the Fc receptor for IgG (FcR), is an indication that this subset of "LAK" cells does not require the trypsin sensitive epitope on tumor cells to mediate killing. Enriching PBL for FcR+ cells markedly enhanced both AICC and ADCC and also reduced the IL-2 requirement of these cells.^ The stoichiometry of Fc receptor (FcR) expression on the cytotoxic effectors does not correlate with ADCC lytic activity. Although FcRs are necessary to mediate ADCC, other factors, appear to regulate the magnitude of cytolytic activity. In order to investigate these putative factors, the kinetics of ADCC development was studied under various conditions (in IL-2 (10u/ml) and 100u/ml), in IL-2(10u/ml) + TNF$\alpha$ (500u/ml) and in TNF-$\alpha$ (500u/ml) alone). Addition of exogenous TNF-$\alpha$ into the four hour cytotoxicity assay did not increase ADCC, nor did anti-TNF antibodies result in inhibition. On the other hand, addition of anti-TNF antibodies to PBL and IL-2 for 24 hours, resulted in a marked inhibition of the ADCC, suggesting that endogenous TNF-$\alpha$ is obligatory for the maturation and differentiation of ADCC effectors. ^

Regulation of apoptosis during the pathogenesis of non -melanoma skin cancer

Previous studies from our lab have shown distinctive patterns of expression of bcl-2 gene family members in human nonmelanoma skin cancer (NMSC). To further evaluate the significance of these observations and to study the effects of cell death deregulation during skin carcinogenesis, we generated a transgenic mouse model (HK1.bcl-2) using the human keratin 1 promoter to target the expression of a human bcl-2 minigene to the epidermis. Transgenic protein expression was confirmed in all the layers of the epidermis except the stratum corneum using immunohistochemistry. Multifocal epidermal hyperplasia, without associated hyperkeratosis, was observed in newborn HK1.bcl-2 mice. Immunofluorescence staining using monoclonal antibodies specific for a variety of differentiation markers revealed aberrant expression of keratin 6 (K6) in the transgenic epidermis. Epidermal proliferative indexes, assessed by anti-BrdUrd immunofluorescence staining, were similar in control and transgenic newborn mice, but suprabasal proliferating cells were seen within the hyperplastic areas of the transgenic mouse skin. Spontaneous apoptotic indices of the epidermis were similar in both control and HK1.bcl-2 transgenic newborn mice, however, after UV-B irradiation, the number of "sunburn cells" was significantly higher in the control compared to the HK1.bcl-2 transgenic animals.^ Adult HK1.bcl-2 and control littermate mice were used in UV-B and chemical carcinogenesis protocols including DMBA + TPA. UV-B irradiated control and HK1.bcl-2 mice had comparable incidence of tumors than the controls, but the mean latency period was significantly shorter in the HK1.bcl-2 transgenic. Both control and transgenic animals included in chemical carcinogenesis protocols required application of both the initiating (DMBA) and promoting (TPA) agents to develop tumors. The frequency, number, and latency of tumor formation was similar in both groups of animals, however, HK1.bcl-2 mice exhibited a rate of conversion from benign papilloma to carcinoma 2.5 times greater than controls.^ Similar carcinogenesis experiments were performed using newborn mice. HK1.bcl-2 mice treated with UV-B plus TPA have a three fold greater incidence of tumor formation compared to controls littermates. HK1.bcl-2 transgenic animals also exhibited a shorter latency for papilloma formation when treated with DMBA plus TPA.^ HK1.bcl-2/v-Ha-ras double transgenic mice shared phenotypic features of both HK1.v-Ha-ras and HK1.bcl-2 transgenic mice, and exhibited focal areas of augmented hyperplasia. These double transgenic mice were susceptible to tumor formation by treatment with TPA alone.^ Cultures of primary keratinocytes were established from control, HK1.bcl-2, HK1.Ha-ras, and HK1.bcl-2/v-Ha-ras newborn mice. Cell viability was determined after exposure of the cells to UV-B irradiation, DMBA, TPA, or TGF-$\beta$1. Internucleosomal DNA fragmentation ("ladders") and morphological cellular changes compatible with apoptotic cell death were observed after the application of all these agents. HK1.bcl-2 keratinocytes were resistant to cell death induction by all of these agents except TGF-$\beta$1. HK1.Ha-ras cells had a higher spontaneous rate of cell death which could be compensated by co-expression of bcl-2.^ These findings suggest that bcl-2 dependent cell death suppression may be an important component of multistep skin carcinogenesis. ^