Early2 factor (E2F) deregulation is a prognostic and predictive biomarker in lung adenocarcinoma.

Clinicians routinely prescribe adjuvant chemotherapy (ACT) for resected non-small cell lung cancer patients. However, ACT only improves five-year disease-free survival in stage I-III non-small cell lung cancer by 5-15%, with most patients deriving no benefit. Herein, deregulation of the E2F pathway was explored as a biomarker in lung adenocarcinoma patients. An E2F pathway scoring system, based on 74 E2F-regulated genes, was trained for RNA from two platforms: fresh-frozen (FF) or formalin-fixed paraffin-embedded (FFPE) tissues. The E2F score was tested as a prognostic biomarker in five FF-based cohorts and two FFPE-based cohorts. The E2F score was tested as a predictive biomarker in two randomized clinical trials; JBR10 and the NATCH (Neo-Adjuvant Taxol-Carboplatin Hope) trial. The E2F score was prognostic in untreated patients in all seven datasets examined (p < 0.05). Stage-specific analysis of combined cohorts demonstrated that the E2F score was prognostic in stage I patients (p = 0.0495 to <0.001; hazard ratio, HR, =2.04- 2.22) with a similar trend in other stages. The E2F score was strongly predictive in stage II patients from the two combined randomized clinical trials with a significant differential treatment effect (p = 0.015). Specifically, ACT improved survival in stage II patients with high E2F (p = 0.01; HR= 0.21). The 5-year survival increased from 18% to 81%. In contrast, in patients with low E2F, 5-year survival was 57% in untreated patients and 41% in ACT-treated patients with a HR of 1.55 (p = 0.47). In summary, the E2F score provides valuable prognostic information for Stage I and predictive information for Stage II lung adenocarcinoma patients and should be further explored as a decision support tool for their treatment.

A Sensitive NanoString-Based Assay to Score STK11 (LKB1) Pathway Disruption in Lung Adenocarcinoma.

INTRODUCTION: Serine/threonine kinase 11 gene (STK11), better known as liver kinase ß1, is a tumor suppressor that is commonly mutated in lung adenocarcinoma (LUAD). Previous work has shown that mutational inactivation of the STK11 pathway may serve as a predictive biomarker for cancer treatments, including phenformin and cyclooxygenase-2 inhibition. Although immunohistochemical (IHC) staining and diagnostic sequencing are used to measure STK11 pathway disruption, there are serious limitations to these methods, thus emphasizing the importance of validating a clinically useful assay. METHODS: An initial STK11 mutation mRNA signature was generated using cell line data and refined using three large, independent patient databases. The signature was validated as a classifier using The Cancer Genome Atlas (TCGA) LUAD cohort as well as a 442-patient LUAD cohort developed at Moffitt. Finally, the signature was adapted to a NanoString-based format and validated using RNA samples isolated from formalin-fixed, paraffin-embedded tissue blocks corresponding to a cohort of 150 patients with LUAD. For comparison, STK11 IHC staining was also performed. RESULTS: The STK11 signature was found to correlate with null mutations identified by exon sequencing in multiple cohorts using both microarray and NanoString formats. Although there was a statistically significant correlation between reduced STK11 protein expression by IHC staining and mutation status, the NanoString-based assay showed superior overall performance, with a -0.1588 improvement in area under the curve in receiver-operator characteristic curve analysis (p < 0.012). CONCLUSION: The described NanoString-based STK11 assay is a sensitive biomarker to study emerging therapeutic modalities in clinical trials.

The Retinoblastoma Tumor Suppressor Transcriptionally Represses Pak1 in Osteoblasts.

We previously characterized the retinoblastoma tumor suppressor protein (Rb) as a regulator of adherens junction assembly and cell-to-cell adhesion in osteoblasts. This is a novel function since Rb is predominantly known as a cell cycle repressor. Herein, we characterized the molecular mechanisms by which Rb performs this function, hypothesizing that Rb controls the activity of known regulators of adherens junction assembly. We found that Rb represses the expression of the p21-activated protein kinase (Pak1), an effector of the small Rho GTPase Rac1. Rac1 is a well-known regulator of adherens junction assembly whose increased activity in cancer is linked to perturbations of intercellular adhesion. Using nuclear run-on and luciferase reporter transcription assays, we found that Pak1 repression by Rb is transcriptional, without affecting Pak1 mRNA and protein stability. Pak1 promoter bioinformatics showed multiple E2F1 binding sites within 155 base pairs of the transcriptional start site, and a Pak1-promoter region containing these E2F sites is susceptible to transcriptional inhibition by Rb. Chromatin immunoprecipitations showed that an Rb-E2F complex binds to the region of the Pak1 promoter containing the E2F1 binding sites, suggesting that Pak1 is an E2F target and that the repressive effect of Rb on Pak1 involves blocking the trans-activating capacity of E2F. A bioinformatics analysis showed elevated Pak1 expression in several solid tumors relative to adjacent normal tissue, with both Pak1 and E2F increased relative to normal tissue in breast cancer, supporting a cancer etiology for Pak1 up-regulation. Therefore, we propose that by repressing Pak1 expression, Rb prevents Rac1 hyperactivity usually associated with cancer and related to cytoskeletal derangements that disrupt cell adhesion, consequently enhancing cancer cell migratory capacity. This de-regulation of cell adhesion due to Rb loss could be part of the molecular events associated with cancer progression and metastasis.

Role of LKB1-CRTC1 on glycosylated COX-2 and response to COX-2 inhibition in lung cancer.

BACKGROUND: Cyclooxygenase-2 (COX-2) directs the synthesis of prostaglandins including PGE-2 linking inflammation with mitogenic signaling. COX-2 is also an anticancer target, however, treatment strategies have been limited by unreliable expression assays and by inconsistent tumor responses to COX-2 inhibition. METHODS: We analyzed the TCGA and Director's Challenge lung cancer datasets (n = 188) and also generated an LKB1-null lung cancer gene signature (n = 53) to search the Broad Institute/Connectivity-MAP (C-MAP) dataset. We performed ChIP analyses, real-time polymerase chain reaction, immunoblotting, and drug testing of tumor cell lines (n = 8) and primary lung adenocarcinoma surgical resections (n = 13). RESULTS: We show that COX-2 is a target of the cAMP/CREB coactivator CRTC1 signaling pathway. In addition, we detected a correlation between LKB1 status, CRTC1 activation, and presence of glycosylated, but not inactive hypoglycosylated COX-2 in primary lung adenocarcinoma. A search of the C-MAP drug database discovered that all high-ranking drugs positively associated with the LKB1-null signature are known CRTC1 activators, including forskolin and six different PGE-2 analogues. Somatic LKB1 mutations are present in 20.0% of lung adenocarcinomas, and we observed growth inhibition with COX-2 inhibitors in LKB1-null lung cancer cells with activated CRTC1 as compared with LKB1-wildtype cells (NS-398, P = .002 and Niflumic acid, P = .006; two-tailed t test). CONCLUSION: CRTC1 activation is a key event that drives the LKB1-null mRNA signature in lung cancer. We also identified a positive feedback LKB1/CRTC1 signaling loop for COX-2/PGE2 regulation. These data suggest a role for LKB1 status and glycosylated COX-2 as specific biomarkers that provide a framework for selecting patients for COX-2 inhibition studies.