Celera Presents Data Replicating Its Immunodiagnostic Assay To Detect Lung Cancer From Blood Serum

Celera Corporation (NASDAQ:CRA) announced yesterday the presentation of data replicating its novel mass spectrometry-based approach to identify and validate circulating protein biomarkers that detect non-small cell lung cancer (NSCLC) in an independent cohort of individuals with lung cancer. This study was performed in collaboration with scientists at NYU Langone Medical Center, NY. A key outcome of the study was the validation of a 9 biomarker immunoassay on a cohort of samples that is enriched for stage I disease, important for screening of early stage disease. The assay detected lung cancer with 92% sensitivity at 93% specificity. The panel also accurately distinguished malignant cases from benign lung disease. The data is being presented at the 13th World Conference on Lung Cancer as part of the International Association for the Study of Lung Cancer (IASLC) in San Francisco, CA.

This study confirms the performance of a 9 member biomarker panel previously presented at the 100th Annual American Association for Cancer Research (AACR) Conference in April 2009. In addition, the authors report the high performance of a 6 marker panel derived from the 9 marker panel which distinguishes malignant cases with 92% sensitivity at 88% specificity. Celera expects to continue the evaluation of the clinical utility of both the 6 and 9 marker panels in an effort to determine which panel would offer optimal performance in a broad clinical setting.

“This is one of the most promising assays I’ve seen as a predictor of non-small cell lung cancer,” said Harvey I. Pass, M.D., Professor of Cardiothoracic Surgery and Director of Thoracic Surgery and Oncology at the New York University Langone Medical Center and its NYU Cancer Institute. “Such a test with this accuracy, if validated prospectively in other larger cohorts, may impact the management of individuals with presumed or diagnosed non-small cell lung cancer in the future.”

“The Early Disease Research Network initiative of National Cancer Institute is committed to identifying a biomarker test for early detection of lung cancer that demonstrates robust performance,” said William N. Rom M.D., M.P.H., the Sol and Judith Bergstein Professor of Medicine and Environmental Medicine at the New York University Langone Medical Center. “The work from this collaboration has resulted in a very promising set of biomarkers that warrant continued study. Such a test could potentially provide clinically useful information for physicians and their patients.”

“We’re pleased to see the performance confirmed in an independent cohort since we believe this assay presents a robust test to detect lung cancer using a simple blood test,” said Steve Ruben, Ph.D., Vice President of Proteomics at Celera. “We believe the ability to identify a collection of biomarkers, which we have subsequently shown to be elevated in the blood of non-small cell lung cancer patients relative to appropriate controls, presents an opportunity to detect the disease earlier, enabling more effective intervention and monitoring. This test in combination with helical CT imaging may increase the predictive value of these tests over CT alone. We’re exploring the commercial opportunities for this program as we continue our efforts to validate these findings in a clinical setting.”

A panel of 9 candidate biomarkers capable of distinguishing lung cancer serum samples from normal samples was previously identified. In the current study, these biomarkers were measured in an independent collection of samples that included: 39 subjects with NSCLC; 33 of which are stage I or II; 38 non-cancer smoking controls; 20 COPD (Chronic Obstructive Pulmonary Disease, emphysema and/or chronic bronchitis) subjects, which is one group of several benign lung diseases often seen in smokers; and 20 subjects with Ground Glass Opacity who continue to be followed to determine which of these patients may progress to malignant disease.

The 9 marker panel demonstrated a significantly high level of performance in distinguishing malignant lesions from a control population that included smokers and patients with COPD, showing 92% sensitivity at 93% specificity.

Study details

The study included 13 candidate biomarkers from a larger set of markers that had been selected from approximately 500 proteins previously identified by a proteomic discovery effort. These biomarkers were used to test serum specimens using ELISA (Enzyme Linked Immunosorbent Assay) methods in a preliminary validation set of sera comprising 103 patients with NSCLC and 104 healthy smoker controls who were matched for age, gender and smoking history. This set of markers includes examples novel to the mass spectrometry discovery effort that have not previously been reported to be elevated in the blood of lung cancer patients. Markers that showed elevated disease expression were then validated in sera from 39 individuals with NSCLC, 38 non-smoking controls and 20 subjects COPD.

The test panels that were configured detected lung cancer with favorable sensitivity and specificity to those previously reported (J Clin Onc. 2007, 25: 5578-5582; J Thorac Oncol. 2006, 1: 513-519; Clin Cancer Res. 2008, 14: 1355-1362). This panel and others described in the study are expected to provide the flexibility to design tests with performance specifications suitable for a variety of diagnostic applications, such as screening individuals at risk for lung cancer and for monitoring of disease following diagnosis and treatment.

Biomarkers and Lung Cancer

Lung cancer is the leading cause of cancer mortality in the U.S. accounting for about 29% of all cancer deaths that occurred in 2008 (Cancer J Clin, 2008; 58:71-96). Currently, there are no guidelines for screening of individuals at risk for lung cancer, and current biomarkers are used only for monitoring of disease following diagnosis and treatment. Novel biomarker panels with improved performance, such as the one described in the current study, may have value at various stages in the detection and diagnosis of lung cancer, which includes independent screening for high risk individuals, and/or for subsequent CT analysis used in combination with imaging technologies to improve resolution of malignant nodules. Such a panel could also be used to improve the monitoring of lung cancer progression by enhancing the management of pulmonary nodules that are identified through CT scanning.

Genes Identified May Help Breast Cancer Diagnosis

The research team, which also included colleagues from Nottingham and Cambridge universities and King’s College London, are identifying and studying genes which control whether a cell lives or dies.

They found that the survival rate for patients with a low expression of a gene known as Fau, a tumour suppressor, is twice as bad as for people with normal levels, while a high expression of cancer-causing gene MELK has a similar effect.

Professor Gwyn Williams, who has been working on the study for 20 years, said he was excited by the discovery, published in the journal Breast Cancer Research, as it had clear real world relevance.

“Our ongoing research is about finding the genes which may go wrong in people with cancer,” he said. “Genetic changes give hints to where to target therapy and can also help diagnose cancer.”

“When you know enough about genes like these you could carry out a general screening of people who might be at risk. This is not an overnight solution but in the next 20 years I would like to see our knowledge of Fau and MELK being used in more practical areas.”

Research teams will now study the two genes in greater depth to identify their uses in diagnosis. The findings may also prove significant in ovarian and prostate cancer research.