Cancer metastasis is a heterogeneous disease that is influenced not only by the acquisition of specific genetic and epigenetic aberrations, but also by the pre-malignant origin of tumor cells and their interaction with the microenvironment. Despite early detection and treatment, lung cancers spread to multiple tissues, sometimes within months of diagnosis, and account for more deaths than any other type of cancer. The mechanisms by which lung cancer cells invade and colonize distant organs remain largely unknown. Our goals are to identify molecular determinants of lung cancer metastasis, study their biological functions, and assess their clinical relevance.
To achieve this, we integrate experimental modeling of both normal and aberrant biological phenomena with bioinformatic analysis, on various genomic/epigenomic platforms (Figure 1). This multi-faceted approach is being applied to the following topics:
1. Molecular heterogeneity and deregulated cell fate.
Part of the challenge in diagnosis lung cancer patients, is the existence of various lung cancer subtypes, each with unique molecular and cellular properties. For instance non-small cell lung cancers (NSCLCs) such as adenocarcinomas, squamous cell carcinomas, and large cell carcinomas, form in different regions of the airways and vary in their response to therapy. We are interested in determining the molecular etiology of distinct lung cancer subtype. For example, we recently uncovered a tumor gene expression program that correlates with poor outcome in patients with lung adenocarcinoma, an aggressive sub-class of NSCLC. This program is driven by a transcriptional network, which normally controls epithelial differentiation in the airways. We are now studying how perturbations in this particular cell differentiation pathway affect metastatic competence in various mouse models (Figure 2). We are also using genomic insights to understand the function and fate of distinct malignant epithelial cell types at various stages of lung adenocarcinoma metastasis.
2. Cues from the microenvironment.
Disseminating cancer cells also encounter paracrine signals from their microenvironment. Depending on the context, tumor cells may interface with the vasculature, immune cells, and other tissue specific stromal cell types. Many of these interactions are akin to physiological responses triggered during normal tissue injury and repair. We are employing in vivo imaging techniques (Figure 2) and ex vivo- assays to understand how invading lung cancer cells can alter their surroundings (Figure 3) to take advantage of physiological signals and colonize distant organs.
3. Linking metastasis with resistance to therapy.
The resistance of some cancers to therapy often correlates with metastatic relapse. Yet it is unclear how these two phenomena are mechanistically linked. A particularly intriguing example is the secondary metastasis observed in lung adenocarcinoma patients after they have undergone systemic therapy. Is this acquired resistance due to tumor cell intrinsic alterations, factors from the new metastatic niche, or inefficient drug delivery to the affected organ site? In collaboration with other investigators of the Yale Cancer Center, we are attempting to answer these questions and identify new therapeutic possibilities.