Nicholas J. Dyson - US grants

The p107 protein was originally identified through its interaction with the adenovirus E1A oncoprotein. A number of observations have suggested that p107 plays an important role in growth regulation. These observations include: (1) the regions of E1A 's growth regulating abilities; (2) cloning and sequencing of p107 showed that it was structurally related to the retinoblastoma gene product, perhaps the best characterized tumor suppressor gene product; (3) in cells without E1A, p107 binds to the E2F transcription factor and inhibits its ability to act as a transcriptional transactivator. E2F participates in the temporal regulation of many genes that are involved in proliferation including c-myc, B-myb, DHFR, DNA po lymerase-alpha, and cdc2; (4) expression of p107 in sensitive cells causes growth arrest in the G1 phase of the cell cycle. In all of these cases, it appears that p107 acts as a negative regulator of proliferation. In mammalian systems, analyzing the biological function of proteins that act as negative regulators of proliferation has been particularly difficult. However, a useful method to study the function of these proteins in vivo has been particularly difficult. However, a useful method to study the function of these proteins in vivo has been to prepare mouse strains with germline mutations that block the synthesis of the particular protein. Breeding these mice allows the examination of the phenotype of mice that lack the negative regulatory functions of these proteins and helps determine the physiological roles of these proteins. This strategy has proven to be particularly powerful in the analysis of other negative regulators including p53, pRB, WT-1, and APC. Recently, we have succeeded in establishing germ line transmission of a p107-null allele (p107-). The construction of a p107(-) allele provides the first opportunity to examine the biological consequences of inactivating p107 mutations. The overall objective of this work are to examine the physiological effects of loosing the function of p107. Our immediate goals center on three objectives. First, we wish to determine the phenotype of mice that carry inactivating p107 mutations. The p107(+/-) heterozygous mice will be examined during their normal life span for any pathology. We will try to make p107 null mice (p107-/-) by breeding the heterozygous mice. Second, we will breed the heterozygote p107(+/-) mice and, if available, the homozygous p107(-/-) mice with Rb(+/-) mice. The Rb(+/-) mice characteristically develop pituitary tumors beginning about 6 to 0 months of age. This pathology provides a recognizable benchmark to test for synergism between these related proteins. Third, we will prepare and analyze cell lines from mice with heterozygous and homozygous p107.

protein structure function; transcription factor; genetic transcription; cell cycle; protein protein interaction; cell proliferation; retinoblastoma protein; cyclins; protein degradation; gene expression; cell differentiation; cell growth regulation; phenotype; apoptosis; gene targeting; Drosophilidae; genetically modified animals; immunoprecipitation; immunocytochemistry; in situ hybridization; gel mobility shift assay;

pRB, p107 and p130 comprise a family of structurally and functionally related proteins. Insight into the normal functions of these proteins has been provided by the analysis of knockout mice. Mice lacking pRB die between days 13.5 and 15.5 of embryogenesis and contain many additional 5-phase cells. Although mice lacking either plO7 or p13O have no obvious abnormalities, mice deficient for both proteins die within a few hours of birth and display defects in cell cycle. Thus, these proteins normally act to restrict cell proliferation and, either directly or indirectly, promote differentiation. How do pRB, p 1O7 and p13O regulate cell proliferation? It is generally thought that their control of the E2F transcription factor is central to their biological functions. All three proteins associate with E2F in vivo in regulatory complexes that exist transiently through the cell cycle. E2F is a critical component of the cell cycle machinery. It is a highly complex activity that co-ordinates the increase in expression of a large number of genes during cell cycle progression from G1 to S phase. The elevation of E2F activity is sufficient to drive quiescent cells into S phase and E2F is required for S-phase entry pRB, p1O7 and p13O are thought to provide different aspects of E2F regulation. However, many genes have been proposed to be targets of E2F and it has been unclear how the functions of pRB, p107 and p13O are integrated to regulate the expression of any of these targets. We have begun to solve this problem by investigating how the expression of E2F target genes is changed in the absence of pRB, p107 or P130. Our initial results show that completely different sets of E2F-target genes are disregulated in primary cells lacking p1O7 and pl30 compared with cells lacking pRB. These data lead us to the hypothesis that pRB and p107 and p130 provide distinct elements of E2F regulation and that the loss of these functions disregulate cell cycle progression in different ways. In this proposal we will (l) distinguish the roles of pRB, p107 and p13O in E2F regulation by completing the analysis of E2F-target genes in primary cells prepared from single and double knockout mice. (2) investigate the functional changes in cell cycle control that occur in the absence of pRB-family proteins by comparing the ease with which these cells can be driven into the cell cycle and asking whether any of these cells are unable to respond to defined mechanisms of G I arrest. (3) identify and characterize a novel regulator E2F that is specifically upregulated in cells lacking both p1O7 and p130.

DESCRIPTION: (Adapted from the investigator's abstract) RB-1 was the first tumor suppressor gene to be identified and it has served as a prototype for this class of cancer-related genes. Genetic studies show that pRB is a multifunctional protein and over eighty different cellular pRN-related proteins have been reported. However, a detailed analysis of this protein has been difficult because most of its properties require a common domain, the pocket domain, and this domain has been refractory to structure/function studies. The recently published crystal structure of the pRN pocket has provided the information needed to specifically mutate conserved residues on the surface of pRB, without disrupting the overall structure of the pocket domain. These studies are needed because of the multiple activities, and the large number of pRB-binding proteins, that depend on this domain. Currently it is unclear which interactions are needed for specific functions of pRB, and it is uncertain which functions of pRB are most relevant for tumor suppression. In the last cycle of this grant they have used the crystal structure to begin to dissect the functions of the pRB pocket domain. A panel of mutants has been prepared that disrupt the LXCXE-binding cleft of pRB. This structure is used by multiple viral oncoproteins to bind to pRB, and sequences similar to the viral LXCXE motif have been found in at least 21 of the cellular pRB-binding proteins reported to date. The LXCXE-binding cleft is maintained in all pRB-homologs that have been identified, suggesting that it is critical for a conserved function of pRB.

[unreadable] DESCRIPTION (provided by applicant): The E2F transcription factor is an important regulator of the G1 to S transition of the cell cycle. E2F activity is rate-limiting for S-phase entry, and the deregulation of E2F is believed to occur in most tumor cells as a result of mutations in the pRB pathway. The vast majority of E2F research has been carried out using mammalian cells. Such studies show that there are many different forms of E2F, that these have a wide variety of transcriptional activities, and that E2F activity can be regulated in many different ways. The immense complexity of mammalian E2F has made it difficult to understand how different components of E2F regulation are integrated together, and it is often difficult to place studies of individual proteins in the larger picture of E2F function. To complement the study of mammalian E2F proteins, we initiated an investigation of E2F in Drosophila. The first two cycles of this grant have provided the basic groundwork needed to use this system. We identified and characterized two E2F homologs, a DP homolog, and two RB family proteins and the genome sequencing efforts revealed that these are the only family members present in flies, dE2F, dDP and RBF1 regulate S-phase entry in vivo in ways that are strikingly analogous to the action of their mammalian counterparts. In the latest cycle of this grant we explored the functional relationships between dE2F1 and dE2F2, and RBF1 and RBF2, and uncovered distinct roles for each of these proteins in E2F-regulated transcription. In the next cycle of this grant we propose to use this system to explore, in detail, three fundamental aspects of E2F/RBF function. Each of these Aims is based on findings made in the previous funding period. Aim 1 will investigate a new class of dE2F2/RBF1 and dE2F2/RBF2-repressed promoters that do not appear to be expressed in cycling cells, but are expressed in developmentally restricted patterns. Aim 2 will investigate the functional overlap between RB family proteins in E2F regulation, and the way that this changes during development. In the previous period we discovered that cells homozygous mutant for both de2fl and de2f2 can proliferate. Aim 3 will use E2F- or DP-deficient cells to determine which aspects of cell cycle regulation require E2F activity. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): RB-1 was the first tumor suppressor gene to be identified and it has served as a prototype for this class of cancer-related genes. pRB is believed to prevent appropriate cell proliferation, in large part, by repressing E2F-dependent transcription. pRB antagonizes E2F-dependent activation in two general ways: pRB binds to E2F proteins and inhibits their ability to activate transcription;in addition, when bound to E2F, pRB recruits complexes that modify chromatin structure and actively repress transcription. These two activities of pRB are mediated via distinct binding sites. Currently it is unclear how many genes are controlled by pRB-mediated recruitment of repressor complexes, as opposed to the more direct inhibition of activator E2Fs. It is also uncertain which of the many cellular functions ascribed to pRB depend on its ability to recruit repressor complexes to DNA. Indeed there is very little information about which repressor complexes are required at E2F-controlled genes, and precisely which of these repressors are recruited by pRB. Using homologous recombination we have generated a knock-in allele of Rb that specifically lacks the LXCXE-binding cleft (Rb delta/LXCXE}. The mutant protein retains the ability to interact with E2F and to block E2F-mediated activation but it lacks the ability to interact with several proteins, including several repressor complexes. Our preliminary experiments with RB/delta/LXCXE/deltaLXCXE MEFs show that these cells are defective for certain types of pRB-dependent cell cycle arrest, but not for others. Since the mutant protein is expressed from the natural Rb promoter and is present at normal levels we can be certain that the phenotypes of these cells reflect physiological functions of pRB. We plan to take advantage of these cells to discover which genes are controlled by this specific element of pRB action, and to identify the proteins that pRB recruits to exert its effects. In doing so we will discover whether pRB recruits the same types of complexes to all its target promoters, or whether it is needed for the recruitment of different complexes at different promoters. By examining cells as they respond to different types of cell cycle arrest signals we will also discover whether pRB recruits different types of repressor complexes in different contexts.

[unreadable] DESCRIPTION (provided by applicant): A very precise control of cell proliferation is necessary for the normal development of multicellular organisms. Deregulated, or inappropriate, cell proliferation is the root cause of many human diseases, particularly Cancer. The E2F transcription factor is a key element in the control of cell proliferation. In a role that is conserved from flies to humans, E2F coordinates the cell cycle-dependent expression of hundreds of genes that are necessary for cells to divide and proliferate. The analysis of E2F and RB homologs in Drosophila has provided a very valuable experimental system that complements the more traditional studies of pRB and E2F family members in mammalian cells. These studies give an opportunity to study E2F function in vivo, in the larger context of animal development. Progress in the previous funding period has provided a detailed picture of the functions of the components of this network, and the ways that the activities of the individual components are integrated. The clear perspective on E2F function provided by this system leads us to the conclusion that, for the control cell proliferation, the key component of this network is the transcriptional activator, dE2F1. To identify rate-limiting steps in dE2F1 function we have taken a genetic approach and have discovered a remarkable number of mutant alleles that strongly modify dE2F1-dependent phenotypes in two different tissues. These provide the first glimpse of the spectrum of cellular activities that have a significant impact on E2F-dependent control of cell proliferation in vivo. These interactors provide a unique opportunity to identify new aspects of E2F regulation and function. We propose to extend our analysis of the dE2F/RBF network by studying the biochemical processes that underlie a newly discovered set of functional interactions between dE2F1 and the components of a repressive submodule of the Mediator complex. Starting from the list of genes isolated in the screen we plan to identify and characterize proteins that limit E2F-dependent proliferation, and proteins that are needed for the activation of dE2F1-dependent transcription. [unreadable] [unreadable] [unreadable]

abstracting; Alleles; Animal Model; Animals; anticancer research; Apoptosis; Apoptotic; base; Biological; Biological Models; cancer cell; Cancer Etiology; cancer type; Cell Cycle Regulation; Cell Death; Cell division; cell killing; Cell Proliferation; cell type; Cells; Collection; Colon Carcinoma; Colorectal Cancer; Development; Drosophila genus; E2F transcription factors; E2F1 gene; EGFR inhibition; Environment; Epidermal Growth Factor Receptor; Evolution; Eye; Eye Development; fly; Funding; Genes; Genetic; Genetic Screening; Genetic Transcription; Goals; Human; imaginal disc; in vivo; Individual; inhibitor/antagonist; innovation; interest; Intestinal Cancer; Killings; Lead; Lesion; Malignant Neoplasms; Molecular; Mortality Vital Statistics; mouse model; Mus; mutant; Mutate; Mutation; neoplastic cell; novel; Nuclear; Orthologous Gene; Outcome; Pathway interactions; Phenotype; prevent; Proliferating; Property; Proteins; public health relevance; Ras/Raf; Regulation; Research; Retinoblastoma; Retinoblastoma Protein; retinoblastoma tumor suppressor; Screening procedure; Signal Transduction; Staging; Susceptibility Gene; System; Testing; Therapeutic; Time; tissue culture; tissue/cell culture; tool; tumor; Tumor Cell Line; Tumor-Derived; United States; Work

DESCRIPTION (provided by applicant): The retinoblastoma susceptibility gene (RB1) was one of the first tumor suppressor genes to be discovered and it has long served as prototype for this group of cancer-associated genes. Mutation of RB1 is the rate-limiting event in the development of the childhood cancer, retinoblastoma, and the RB1 gene product (pRb) is absent or misregulated in over 90% of human cancers. Mammalian cells contain three Rb-family members and homologous genes have been conserved during the evolution of both plants and animals. pRB binds to transcription factors and recruits chromatin modifying enzymes to promoters. pRB family members are believed to co-ordinate global changes in gene expression programs during cell cycle exit, allowing cells to switch off genes that are needed for cell proliferation and switch on genes that are needed for differentiation. How pRB family members orchestrate these changes is unclear. In the previous cycle of this grant we discovered a new activity of pRB-family proteins. We found that they physically interact with CAP-D3 a component of Condensin II complexes. Condensin complexes are conserved from yeast to humans and are required for normal chromosome structure during mitosis. Experiments using the Drosophila model system show that dCAP-D3 is extremely important for RBF1 function. We hypothesize that dCAP-D3 and Condensin II complexes have functions in addition to their well-studied roles during mitosis. We propose that Condensin II complexes are recruited to specific locations in the genome by pRB family proteins and that, at those sites, they act in conjunction with pRB family proteins to regulate levels of gene expression. To test this, we will identify genes regulated by dCAP-D3/Condensin II and RBF1. We will generate interaction-defective alleles to test the hypothesis that RBF1 and dCAP-D3 provide an element of transcriptional regulation that is distinct from its interactions with E2F. Using biochemical approaches we will purify RBF1/dCAP-D3 complexes and will examine the mechanisms underlying the functional co-operation between dCAP-D3 and RBF1.

DESCRIPTION (provided by applicant): Many human tumor cells are chromosomally unstable, showing an elevated rate of gains and losses of whole chromosomes that is 10-100 times higher than that seen in diploid primary cells. Chromosome instability (CIN) enhances the evolution of tumor cells causing genomic changes that can promote metastasis and chemotherapeutic resistance. CIN can have a causal role in tumorigenesis and it correlates with poor patient prognosis. Identifying the mutations that cause CIN in human cancer, and understanding why they reduce the fidelity of mitosis, are important objectives in cancer research. The inactivation of the pRB pathway promotes cell proliferation and is a common event in tumor cells. Interestingly, the functional inactivation of pRB causes aneuploidy. Recent experiments show that the specific loss of pRB increases rates of chromosome mis-segregation to levels that are remarkably similar to CIN tumor cells. This strongly suggests that a significant fraction of the chromosome instability seen in tumor cells is a byproduct of the inactivation of pRB. This grant investigates the exciting link between pRB inactivation and chromosome mis-segregation. Newly obtained results show that the loss of pRB causes defects in centromere function and chromatid cohesion. We will test the hypothesis that defects in cohesion and condensation allow chromosome mis-segregation when pRB-deficient cells are delayed in mitosis. Cohesin is the primary determinant of chromosome cohesion. In Aim 1 we will determine how the association of cohesin complexes with chromatin is altered in the absence pRB, and will test the functional significance of physical interactions between pRB and cohesin. The experiments in Aim 2 will test whether the mitotic defects in pRB-deficient cells can be either suppressed or enhanced. Such experiments may lead to new therapeutic opportunities. In Aim 3 we will extend these studies to cancer cells and will measure and compare the changes in centromere function, chromosome cohesion, and the rates of chromosome mis-segregation seen when functional pRB is re-introduced into tumor cells that lack it, or is specifically removed from tumor cells with functional pRB. We will determine whether similar changes occur when pRB is inactivated by deregulated cdk activity and when pRB-related proteins are targeted.

PROJECT SUMMARY (See instructions): The Cancer Genetics Program's overall mission is to expand the understanding of the genetic underpinnings of cancer development, and to use this information to improve the care of cancer patients. To advance this mission, the Program has assembled a large and vibrant membership, including investigators with a broad range of scientific interests in all major aspects of cancer genetics, such as: cancer gene discovery in both human cancer samples and model organisms, technology development and efficient application of high throughput DNA sequencing, detailed cancer genome analysis, high throughput approaches to cancer gene analysis and annotation, identification and analysis of both cancer initiating cells and induced pluripotent stem cells, clinical cancer genetics and risk counseling, and development and use of CLIA-certified testing for clinically relevant cancer diagnostics. The specific aims of the Program are to: 1. Support the discovery of new genes and cellular pathways implicated in cancer. 2. Enhance identification and understanding of the germline genetic variations that influence cancer risk and response to therapy. 3. Increase the understanding of the full spectrum of somatic mutation that occurs in cancer and how it contributes to the genesis and progression of cancer. 4. Support the translation of these research findings to both clinical research in oncology and routine cancer patient care. The Program has been funded by the CCSG since 2000 when DF/HCC was established, and received an Outstanding merit score at the last renewal in 2005. The Program's membership includes 100 investigators, representing all seven institutions in the consortium, 14 departments of HMS, and one department of HSPH. In 2009, the Program received $52.5 million in cancer-relevant funding (total costs), which includes $18.5 million from NCI and $25 million from other peer-reviewed sponsors. Program members have published 1,630 publications over the project period (2006 to 2010), of which 9% were intra-programmatic, 44% were inter-programmatic and 32% were inter-institutional.

DESCRIPTION (provided by applicant): pRB is one of the best-known tumor suppressors. It plays a central role in the processes that enable cells to stop dividing and to permanently withdraw from the cell cycle. Consistent with its role as a key negative regulator of cell proliferation, pRB's ability to function is thought to be compromised in most, if not all, human cancer cells. The majority of human tumors do not mutate the RB1 gene but acquire changes, such as mutations in the locus encoding the p16 cdk inhibitor that reduce pRB's activity. Despite extensive investigation, the molecular details of pRB's mechanism of action are vague. More than 150 different proteins have been reported to physically interact with pRB and have been proposed to be relevant for pRB function. However, most studies of have investigated just pRB-associated protein, or a small group of pRB-interactors, at any one time. Different groups have championed the importance of individual interactions but because the literature on pRB-associated proteins has accumulated in a piecemeal fashion, there is no information about the relative significance of these interactions, and no consensus over which pRB- associated proteins are genuinely important. We propose to take advantage of the rapid developments in shRNA technologies to systematically test the effect of knocking-down each of the reported pRB-associated protein on a panel of pRB-induced phenotypes. This will enable us to assess the relative importance of each pRB-binding protein for various activities of pRB. To identify pathways that impinge on these protein/protein interactions we will complement this analysis by screening the same set of functional assays with a collection of shRNAs that target the kinome. Preliminary data shows that these function-based screens not only identify proteins that are needed for pRB to act, but also identify pathways that can be targeted to enhance specific outcomes, including the ability of pRB to induce senescence in human tumor cells. Using pancreatic cancer cells as an example of cancers that retain an intact RB gene, we will identify the interacting proteins that are essential for pRB-mediated suppression of cell proliferation, and we will test whether targeting co-operative pathways can enhance the activity of endogenous pRB and improve its ability to induce a permanent cell cycle arrest. Together, these experiments will provide a framework for future analysis of pRB function and will uncover ways to enhance the activity of endogenous pRB in human cancer cells.

? DESCRIPTION (provided by applicant): The functional inactivation of the Rb tumor suppressor is an important event in the genesis of many cancers. pRb is a negative regulator of cell proliferation and it is thought that tumors acquire changes that allow them to escape pRb's ability to suppress cell cycle progression. A key goal at the heart of all Rb research is to understand how cells are changed by the loss of pRb. This information might suggest ways to suppress these changes, or might reveal weaknesses that can be targeted therapeutically. pRb associates with chromatin and regulates transcription. Numerous studies have profiled the transcriptional changes that occur when Rb is lost, and it has generally been assumed that these transcriptional profiles give a meaningful picture of the altered state of Rb-mutant cells. As an alternative approach, in this application we propose to use proteomic profiling to identify proteins whose abundance changes significantly when Rb is removed. Preliminary data from analysis of Rb mutant tissues shows that loss of this tumor suppressor leads to extensive proteomic changes that are strikingly different from the transcriptional signatures that have been studied previously. In Aim 1 we propose to generate a detailed timecouse of these changes, to identify the types of proteins that are altered and to determine whether the proteomic changes occur before, or after, the changes in transcription. We propose to study one of the largest groups of proteomic changes and to determine which changes promote the survival and proliferation of pRb-deficient cells. One of the surprising features of the proteomic data is that the transcriptional upregulation of classic E2F targets leads to only minor changes in overall protein levels in Rb-mutant issues. We hypothesize that one of the reasons for this is that Rb loss increases the levels of RNA-binding proteins that bind to the 3'UTR sequences of multiple E2F-induced mRNAs and suppress translation. Aim 2 of the proposal will test this model and will identify transcripts repressed by Nanos and Pumilio in pRb-deficient cells. Collectively these results will provide new insights into the cellular consequences of pRb inactivation, and the ways that cells adapt to cope with deregulated E2F activity.

Project summary For simplicity we tend to imagine that most proteins have a single mechanism of action. The retinoblastoma tumor suppressor is an example of a protein that defies this simple classification. RB1 is functionally inactivated in most human cancers and the molecular properties of its protein product (RB) have been studied intensively. Despite this research, RB's mechanism of action has remained an enigma. RB has been reported to physically associate with hundreds of proteins and many different interactions have been proposed to contribute to its tumor suppressive properties. The RB research community is faced with a conundrum: which of these interactions are real, which are not, and how could one protein co-ordinate its effects on so many potential targets? Recent studies from several laboratories have suggested that the answers to these questions lie in a code of RB phosphorylation. The concept is that normal cells do not contain a single form of RB, but that differential phosphorylation generates multiple isoforms of RB that have different binding properties and, presumably, perform different roles. In essence, the action of RB is tailored by phosphorylation. RB is known to have 14 sites of CDK phosphorylation. We have recently developed methods that allow us to use mass spectrometry-based proteomics to profile RB complexes. We have also generated panels of isogenic cell cultures in which we can replace the endogenous RB protein with mutant RB proteins that contain just a single cdk phosphorylation site, or a single phospho-mimicking mutation. In this application we propose to use these tools to decipher this code of RB phosphorylation. In Aim 1 we will use state-of-the-art proteomics to define the binding properties of each of the mono-phosphorylated isoforms of RB. In Aim 2 we will identify the functional consequences of these interactions by identifying the transcriptional programs that they control and by genomic loci that they target. Using this binding information and transcription profiles we will identify the molecular interactions that allow specific mono-phosphorylated isoforms of RB to control distinct programs of transcription. Together these experiments will generate a framework of molecular information that is critical to be able to understand RB's mechanism of action.  

PROJECT SUMMARY Small cell lung cancer (SCLC) afflicts more than 30,000 patients per year and is rapidly fatal in 95% of cases, with median survival is less than one year. Belying this grim prognosis, treatment-naive SCLC is highly sensitive to chemotherapy, with response rates in excess of 70% for etoposide/platinum. However, relapse is nearly inevitable, and relapsed SCLC presents two obstacles that have been insurmountable for at least 30 years: cross-resistance to chemotherapy, and absence of biomarker-driven targeted therapy. Following relapse, resistance often extends beyond etoposide/platinum, and a disease that was once highly chemosensitive becomes inexorably progressive. However, the molecular determinants of cross-resistance in SCLC remain unclear. Although critically important, cross-resistance is difficult to study experimentally, as it requires a model system that faithfully reproduces clinical outcomes. Topotecan is the only approved second-line therapy, but NCCN guidelines list 10 agents of nearly equivalent efficacy. None are particularly effective in unselected patients, and although there is significant molecular heterogeneity in SCLC, this does not guide patient selection. As novel targets and therapeutic regimens emerge, biomarker discovery will require a model system that recapitulates the molecular features of patient tumors, so that molecular heterogeneity can be parsed into clinically meaningful subgroups. We have generated a panel of 44 SCLC patient-derived xenograft models (PDXs) from biopsy specimens and circulating tumor cells (CTCs). Our panel includes successive models from individual patients at time points before and after specific lines of therapy, with detailed information about the corresponding clinical response. For both standard chemotherapy and experimental agents in clinical trial, these models faithfully mirror patient responses. However, unlike the patient experience, multiple strategies can be compared for identical tumors. We propose to use these models to directly compare standard first and second-line chemotherapy with two experimental regimens that have given promising results in the clinic or in preclinical assays: olaparib plus temozolomide, in a phase I/II trial at MGH, and a combined Mcl-1/Bcl-2 inhibitors. Individually, these PDX population trials are designed to reveal biomarkers of sensitivity and mechanisms of resistance for promising experimental therapies. Collectively, they present a novel opportunity to model cross-resistance through comparative analysis with reference to the clinical histories of each model. The successful completion of this work will establish a large collection of PDX models with comprehensive molecular an functional profiles. In addition, these experiments will investigate the molecular determinants of cross-resistance following chemotherapy, a problem that has beleaguered management of SCLC for over three decades.