Stacy J. Park

CancerLocator: non-invasive cancer diagnosis and tissue-of-origin prediction using methylation profiles of cell-free DNA

We propose a probabilistic method, CancerLocator, which exploits the diagnostic potential of cell-free DNA by determining not only the presence but also the location of tumors. CancerLocator simultaneously infers the proportions and the tissue-of-origin of tumor-derived cell-free DNA in a blood sample using genome-wide DNA methylation data. CancerLocator outperforms two established multi-class classification methods on simulations and real data, even with the low proportion of tumor-derived DNA in the cell-free DNA scenarios. CancerLocator also achieves promising results on patient plasma samples with low DNA methylation sequencing coverage.

Phase I Trial of Intratumoral Injection of CCL21 Gene–Modified Dendritic Cells in Lung Cancer Elicits Tumor-Specific Immune Responses and CD8+ T-cell Infiltration

Purpose: A phase I study was conducted to determine safety, clinical efficacy, and antitumor immune responses in patients with advanced non–small cell lung carcinoma (NSCLC) following intratumoral administration of autologous dendritic cells (DC) transduced with an adenoviral (Ad) vector expressing the CCL21 gene (Ad-CCL21-DC). We evaluated safety and tumor antigen–specific immune responses following in situ vaccination (ClinicalTrials.gov: NCT01574222). Experimental Design: Sixteen stage IIIB/IV NSCLC subjects received two vaccinations (1 × 106, 5 × 106, 1 × 107, or 3 × 107 DCs/injection) by CT- or bronchoscopic-guided intratumoral injections (days 0 and 7). Immune responses were assessed by tumor antigen–specific peripheral blood lymphocyte induction of IFNγ in ELISPOT assays. Tumor biopsies were evaluated for CD8+ T cells by IHC and for PD-L1 expression by IHC and real-time PCR (RT-PCR). Results: Twenty-five percent (4/16) of patients had stable disease at day 56. Median survival was 3.9 months. ELISPOT assays revealed 6 of 16 patients had systemic responses against tumor-associated antigens (TAA). Tumor CD8+ T-cell infiltration was induced in 54% of subjects (7/13; 3.4-fold average increase in the number of CD8+ T cells per mm2). Patients with increased CD8+ T cells following vaccination showed significantly increased PD-L1 mRNA expression. Conclusions: Intratumoral vaccination with Ad-CCL21-DC resulted in (i) induction of systemic tumor antigen–specific immune responses; (ii) enhanced tumor CD8+ T-cell infiltration; and (iii) increased tumor PD-L1 expression. Future studies will evaluate the role of combination therapies with PD-1/PD-L1 checkpoint inhibition combined with DC-CCL21 in situ vaccination. Clin Cancer Res; 23(16); 4556–68. ©2017 AACR.

Inflammation and Lung Cancer: The Role of Epithelial–Mesenchymal Transition

Epithelial–mesenchymal transition (EMT) is a type of cellular plasticity by which epithelial cells acquire the form and function of mesenchymal cells. Physiologic EMT is an essential part of normal embryonic development and an adult organism’s ability to overcome acute injury. However, chronic injury and inflammation can yield dysregulated or pathologic EMT that drives organ fibrosis and cancer development. In this chapter, we review seminal work and recent findings regarding the molecular, cellular, microenvironmental, and environmental factors that drive inflammation-induced EMT-dependent lung carcinogenesis. We also discuss potential approaches for treating or perhaps preventing lung cancer by targeting the inflammation-EMT-cancer axis.

Identification of a Human Airway Epithelial Cell Subpopulation with Altered Biophysical, Molecular, and Metastatic Properties

Lung cancers are documented to have remarkable intratumoral genetic heterogeneity. However, little is known about the heterogeneity of biophysical properties, such as cell motility, and its relationship to early disease pathogenesis and micrometastatic dissemination. In this study, we identified and selected a subpopulation of highly migratory premalignant airway epithelial cells that were observed to migrate through microscale constrictions at up to 100-fold the rate of the unselected immortalized epithelial cell lines. This enhanced migratory capacity was found to be Rac1-dependent and heritable, as evidenced by maintenance of the phenotype through multiple cell divisions continuing more than 8 weeks after selection. The morphology of this lung epithelial subpopulation was characterized by increased cell protrusion intensity. In a murine model of micrometastatic seeding and pulmonary colonization, the motility-selected premalignant cells exhibit both enhanced survival in short-term assays and enhanced outgrowth of premalignant lesions in longer-term assays, thus overcoming important aspects of “metastatic inefficiency.” Overall, our findings indicate that among immortalized premalignant airway epithelial cell lines, subpopulations with heritable motility-related biophysical properties exist, and these may explain micrometastatic seeding occurring early in the pathogenesis of lung cancer. Understanding, targeting, and preventing these critical biophysical traits and their underlying molecular mechanisms may provide a new approach to prevent metastatic behavior. Cancer Prev Res; 10(9); 514–24. ©2017 AACR. See related editorial by Hynds and Janes, p. 491

Abstract B22: The role of e-cigarette exposure on pulmonary epithelial cell transformation

Despite a strong correlation between cigarette smoking and the onset of lung cancer, the prevalence of smoking still remains high. The electronic cigarette (ECIG) is designed to deliver nicotine without combusting tobacco. Since nicotine is widely considered the addictive component in tobacco with limited ability to initiate cancer, ECIGs have been advertised to be a safer alternative to tobacco cigarettes. However, the potential health risks and carcinogenicity of ECIGs have not previously been evaluated. In this study, we assess the impact of ECIG exposure on the carcinogenic potential of immortalized human bronchial epithelial cells on a background of silenced p53 and activated KRAS (H3mut-P53/KRAS). This model is utilized because p53 and KRAS mutations are often observed in the airway of current and former smokers at risk for lung cancer. In anchorage independent growth assays, the in vitro correlate of malignant transformation, we found enhanced colony growth in the HBEC-P53/KRAS cells following a 12-day treatment with high concentrations of ECIG-conditioned media compared to the untreated and low concentration treatment groups. We next assessed the effect of ECIG and exposure on cell invasion using three-dimensional air-liquid interface (ALI) models. HBEC-P53/KRAS cells exhibited invasion-associated morphological changes following a 12-day treatment with the high conditioned media, including increased proliferation, diminished cell-cell cohesion and the appearance of cells percolating out of and breaching the modified basement membrane. Finally, to identify the biological impact of in vitro ECIG exposure in HBECs, we profiled the gene expression of P53/KRAS cells following a 96-hour exposure to ECIG- or tobacco cigarette (TCIG)-conditioned media. We found that epithelial cells exposed to clinically relevant concentrations of ECIG vapor-conditioned media have a gene expression pattern similar to those exposed to TCIG smoke-conditioned media and whole cigarette smoke. Rank-rank hyper-geometric overlap (RRHO) analysis indicated that differential expression - based ranked genes in TCIG and ECIG exposed groups were consistently overlapped at significant levels. There were 263 differentially expressed genes in the cells treated with high ECIG media versus untreated control. Annotations of the identified genes by the Molecular Signature database revealed several enriched biological pathways involved in malignant transformation and epithelial-mesenchymal transition (EMT). We have compared the resulting list of genes to publicly available microarray datasets and identified several transformation-related gene candidates. We are in the process of evaluating their contribution to ECIG-induced dissemination and carcinogenesis in vitro and in vivo. These studies will determine the impact of ECIG exposure on lung carcinogenicity and provide needed scientific guidance to the FDA regarding the physiologic effects of ECIGs. These studies were supported by funding from the following: NIH/NCI #U01CA152751 (SMD, TCW), NCI #U01CA152751-S1 (SMD, TCW, SJP), NCI #U01CA152751-AS (SMD, KK), NCI #T32-CA009120-36 (SMD, SJP, PCP), NIH/NHLBI #T32HL072752 (SMD, EL), University of California Tobacco-Related Disease Research Program (TRDRP) #18FT-0060 (TCW), TRDRP #20KT-0055 (TCW), Prevent Cancer Foundation (SJP), Lung Cancer SPORE (P50CA70907, JDM, JEL) Citation Format: Stacy J. Park, Tonya C. Walser, Linh M. Tran, Catalina Perdomo, Teresa Wang, Long-Sheng Hong, Paul C. Pagano, Elvira L. Liclican, Jill E. Larsen, Kostyantyn Krysan, Michael C. Fishbein, John D. Minna, Marc E. Lenburg, Spira Avrum, Steven Dubinett. The role of e-cigarette exposure on pulmonary epithelial cell transformation. [abstract]. In: Proceedings of the Thirteenth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2014 Sep 27-Oct 1; New Orleans, LA. Philadelphia (PA): AACR; Can Prev Res 2015;8(10 Suppl): Abstract nr B22.

Abstract B15: Cell motility and deformability in the pathogenesis of lung cancer.

The most effective therapy for lung cancer is surgical resection, but it recurs in approximately 50% of patients with early stage disease, with recurrence most commonly presenting as metastatic disease. This suggests that micrometastatic disease is often already present at the time of surgery, but below the level of detection of our current imaging studies. This is consistent with the reports of circulating tumor cells in patients with stage-I non-small cell lung cancer (NSCLC). Although metastatic behavior is often considered a late stage event, these clinical findings suggest that the metastatic process is also operative early in the pathogenesis of the disease. These clinical observations are also consistent with recent laboratory-based investigations indicating that dissemination may occur during early tumor development, particularly in the context of the genetic program associated with epithelial-mesenchymal transition (EMT). Our preliminary studies indicate that a key regulator of the EMT program, Snail, is responsible for both transformation and enhanced motility of human bronchial epithelial cells (HBECs). Cells that over-express Snail (HBEC-Snail) show structural features indicative of enhanced motility, including filopodia, lamellipodia, membrane ruffling, and front-rear polarity. HBEC-Snail cells move with up to 50% greater velocity than vector control cells. We find that HBECs expressing genes associated with EMT or bearing common NSCLC driver mutations (p53/KRAS), show marked heterogeneity in their capacity for cell motility. Greater than an order of magnitude difference between the fastest and slowest moving cells was consistently observed in these cells. We hypothesize that enhanced epithelial cell motility is operative during premalignancy and is a driver of early metastatic dissemination. The overarching objective of this research is to elucidate the fundamental mechanisms involved in epithelial cell motility and their potential impact on disease onset and progression by coupling novel in vitro and in vivo models of human lung carcinogenesis and an innovative motility-based cell isolation technique. As a result of isolating these fast moving cells, we have shown that the selected cells move through transwell membranes at a 45- to 120-fold increased frequency compared to their unselected counterparts. We have also found that cellular deformability relates to a motile phenotype. By using atomic force microscopy (AFM) and deformability cytometry (DC), we have determined that highly migratory HBEC-Snail cells are more deformable than unselected cells. In the future, we intend to implant selected cells into mice to determine if there is a physiologically-relevant transformation- or cancer-associated phenotype, such as decreased time to metastasis or increased final metastatic burden. These are the first investigations to tie the physical traits of cell motility and deformability to lung carcinogenesis and early metastatic behavior. We anticipate that further investigation, including transcriptome analysis of selected cells by RNAseq, will yield a more complete understanding of the molecular determinants of lung cancer pathogenesis, which will stimulate development of more effective chemoprevention and early detection, benefiting those at risk for lung cancer and those with early stage disease. These studies were supported by funding from the following: NIH/NCI #T32-CA009120-36 (SMD, PCP, SJP), NCI #U01CA152751 (SMD, TCW), NCI #U01CA152751-S1 (SMD, TCW, SJP), NCI #U01CA152751-AS (SMD, KK), NIH/NHLBI #T32HL072752 (SMD, EL), Department of Veteran Affairs #5I01BX000359 (SMD), University of California Tobacco-Related Disease Research Program (TRDRP) #22DT-0005 (PCP), TRDRP #18FT-0060 (TCW), and TRDRP #20KT-0055 (TCW), Lung Cancer SPORE #P50CA70907 (JDM, JEL), Packard Foundation Fellowship (DDC). Citation Format: Paul C. Pagano, Shivani Sharma, Sean O9Byrne, Henry T. Tse, Stacy J. Park, Elvira L. Liclican, Kostyantyn Krysan, Tonya C. Walser, Jill E. Larsen, John D. Minna, James K. Gimzewski, Dino Di Carlo, Steven M. Dubinett. Cell motility and deformability in the pathogenesis of lung cancer. [abstract]. In: Proceedings of the AACR-IASLC Joint Conference on Molecular Origins of Lung Cancer; 2014 Jan 6-9; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2014;20(2Suppl):Abstract nr B15.

Abstract A08: The role of miR-125a in the pathogenesis of lung cancer

Lung cancer is the leading cause of cancer-related mortality in the U.S., and despite focused research pertaining to conventional therapies, the five-year survival rate remains only 17%. The lack of a systematic approach for chemoprevention agent selection and the disregard for individual differences when randomizing to intervention groups have led to clinical prevention trials that have failed or, in some cases, increased the risk for lung cancer. In this regard, understanding the molecular pathogenesis of lung cancer development in a given individual can facilitate identification of targets and/or biomarkers specific to their unique airway molecular profile and the application of targeted chemoprevention. This is expected to yield more effective lung cancer prevention and control. microRNAs (miRNA) have the potential to be robust biomarkers for lung cancer. Our work focuses on preclinical studies to identify mechanisms by which loss of miRNA-125a-3p (miR-125a) expression contributes to the malignant potential of pulmonary epithelial cells harboring an activating point mutation of the K-ras proto-oncogene, one of the most clinically challenging genetic changes commonly found in current and former smokers. Normal human bronchial epithelial cells (HBECs) isolated from large airways of patients were immortalized with hTERT and Cdk4 and subsequently manipulated to express oncogenic K-ras (HBEC K-ras). Compared to their respective vector controls (HBEC Vector), the basal expression of miR-125a is significantly reduced by 3- to 4-fold in premalignant HBEC K-ras lines. Subsequent studies demonstrated that the loss of miR-125a expression in K-ras-mutated HBECs is regulated by the PEA3 transcription factor. Compared with Vector, PEA3 expression is upregulated in HBEC K-ras cells at both the mRNA and protein level, and genetic knockdown of PEA3 with siRNA in HBEC K-ras cells restores miR-125a expression. To assess the impact of miR-125a loss on tumor-promoting factors in K-ras-mutated HBECs, we evaluated a panel of inflammatory and angiogenic proteins and found that the basal protein expression of CXCL1 and VEGF was significantly elevated in HBEC3 K-ras cells compared with Vector. Importantly, overexpression of miR-125a significantly reduced the levels of these tumorigenic factors in HBEC K-ras cells, while no changes were observed in Vector. miR-125a overexpression also reduced the increased proliferation rate of HBEC K-ras cells to the level of Vector and suppressed anchorage-independent growth of K-ras/P53-mutated HBECs by 50%, an inhibition found to be dependent on CXCL1. Importantly, our data also show that pioglitazone, a model chemopreventive agent, increases levels of miR-125a in HBEC K-ras cells through downregulation of PEA3. Moreover, pioglitazone exhibits similar anti-tumor activity as miR-125a overexpression. Treatment of HBEC K-ras cells with pioglitazone downregulates expression of VEGF, suppresses proliferation and induces morphological changes indicative of mesenchymal-to-epithelial transition. Taken together, our novel findings identify the loss of miR-125a expression as a putative biomarker for lung carcinogenesis and possible chemoprevention target. Moreover, our studies illustrate how augmentation of miR-125a by a model chemoprevention agent, in the setting of the K-ras-mutated pulmonary epithelium, can abrogate some of the deleterious downstream events associated with this mutation. These studies were supported by NHLBI #T32HL072752 (EL) and Lung Cancer SPORE #P50CA70907 (JDM, JEL). Citation Format: Elvira L. Liclican, Saswati Hazra, Jill E. Larsen, Stacy J. Park, Paul C. Pagano, Tonya C. Walser, Kostyantyn Krysan, John D. Minna, Steven M. Dubinett. The role of miR-125a in the pathogenesis of lung cancer. [abstract]. In: Proceedings of the Twelfth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2013 Oct 27-30; National Harbor, MD. Philadelphia (PA): AACR; Can Prev Res 2013;6(11 Suppl): Abstract nr A08.

Abstract C58: The gene expression and functional impact of E-cigarette exposure on pulmonary epithelial transformation

Lung cancer is the leading cause of cancer deaths in the United States. Despite a strong correlation between cigarette smoking and the onset of lung cancer, the prevalence of smoking still remains high. Significant reduction or elimination of the use of combustible tobacco products would substantially lower tobacco-caused morbidity and mortality. The electronic cigarette (ECIG) is a battery-powered electronic nicotine delivery system (ENDS) designed to deliver nicotine without combusting tobacco. ECIGs are widely believed to be a safer alternative to tobacco cigarettes (TCIG), however, the toxicity and potential carcinogenicity of ECIGs have not previously been evaluated. In this study, we assess the impact of ECIG exposure on the carcinogenic potential of immortalized human bronchial epithelial cells on a background of silenced p53 and activated KRAS (H3mut-P53/KRAS). This model is utilized because p53 and KRAS mutations are often observed in the airway of current and former smokers at risk for lung cancer. The epithelial cells were exposed to a low and high nicotine concentration of ECIG vapor- or TCIG smoke-conditioned media. The lower nicotine concentration was selected to mimic plasma nicotine levels in ENDS users and did not demonstrate toxic or anti-proliferative effects on the cells. The higher concentration was chosen to represent the anticipated nicotine levels to which the epithelial cells of smokers are actually exposed. In anchorage independent growth assays, the in vitro correlate of malignant transformation, we found enhanced colony growth in the H3mut-P53/KRAS cells following a 10-day treatment with the high nicotine concentration of ECIG- and TCIG-conditioned media compared to the untreated and low nicotine treatment groups. We next assessed the effect of ECIG and TCIG exposure on cell invasion using three-dimensional air-liquid interface (ALI) models. While treatment with the low nicotine concentration of ECIG-conditioned media did not induce invasion-associated morphological changes, cells treated with the low nicotine concentration of TCIG-conditioned media exhibited slightly invasive behavior by breaching the modified basement membrane and invading through the field of fibroblasts/collagen. At baseline, H3mut-P53/KRAS cells exhibit invasive behavior in the ALI model, due to the downstream effects of P53 silencing and KRAS activation. Treatment of H3mut-P53/KRAS cells with the low nicotine concentration of ECIG- and TCIG-conditioned media did not further enhance the degree of invasion observed in the untreated group. We will next examine the effect of the high nicotine concentration on cell invasion. In addition, we are currently evaluating gene expression profiles by microarray of H3mut-P53/KRAS cells exposed to high and low nicotine concentrations of ECIG-conditioned media. This will directly evaluate the impact of ECIGs on carcinogenicity-related gene expression signatures established in previous and ongoing clinical investigations. These studies will determine the impact of ECIG exposure on lung carcinogenicity, which will provide needed scientific guidance to the FDA regarding the physiologic effects of ECIGs. These studies were supported by the NIH/NCI #U01CA152751-S1 (SMD, AS, MEL, TCW) and NIH/NHLBI #T32HL072752 (SJP). Citation Format: Stacy J. Park, Tonya C. Walser, Catalina Perdomo, Paul P. Pagano, Daniel Brass, Elvira L. Liclican, Kostyantyn Krysan, Marc E. Lenburg, Avrum Spira, Steven M. Dubinett. The gene expression and functional impact of E-cigarette exposure on pulmonary epithelial transformation. [abstract]. In: Proceedings of the Twelfth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2013 Oct 27-30; National Harbor, MD. Philadelphia (PA): AACR; Can Prev Res 2013;6(11 Suppl): Abstract nr C58.