Abbreviated Name

Lung Ref Set A App: Goodglick/Diamandis (2009)

Lead Investigator

Chia (Retired), David — University of California Los Angeles

Coordinating Investigator

Feng, Ziding — Fred Hutchinson Cancer Center

Involved Investigators

• Feng, Ziding — Fred Hutchinson Cancer Center
• Srivastava, Sudhir — National Cancer Institute
• Diamandis, Eleftherios P. — Mount Sinai Hospital
• Ujhazy, Peter — National Cancer Institute
• Chia (Retired), David — University of California Los Angeles

Abstact

No abstract availalbe.

Aims

Aim 1. We will coordinate with the Lung Cancer Biomarker Committee to obtain aliquots of serum reference set A (i.e., Phase II validation set: 180 cases and 180 controls, 75 other cancer controls). The characteristics of these sample sets have been described on the EDRN website. Aim 2. Two independent sites (UCLA and University of Toronto) will conduct ELISA tests on blinded serum samples for KLK8, 11, 12, 13, and 14. These assays have been well described previously, so we anticipate no technical hurdles. All samples will be run in duplicate. Aim 3. Data will be analyzed by the DMCC for both single KLK expression as well as for the entire panel (KLK8, 11, 12, 13, and 14) as previously described (1).

Analytic Method

Briefly, the nonparametric Kruskal-Wallis test will be used to examine the relationship between KLK levels, pathology, and clinical information. To evaluate the KLK for disease screening and diagnosis, the classification accuracy will be assessed using receiver operating characteristic (ROC) curves. The ROC curve will plot the true positive fraction (TPF) versus the false positive fraction (FPF) for the set of rules that classify a subject as “test-positive” i.e., if the marker value exceeds a threshold value, where threshold varies over all possible values. It quantifies how well a marker discriminates between diseased and non-diseased individuals. We will perform the ROC analysis first on individual KLK and then in combination. We will use an algorithm that renders a single composite score using the linear predictor, fitted from a binary regression model. A stepwise regression procedure will be used to select markers in the panel with or without additional clinical variables.

Outcome

To study of the utility of KLK8, 11, 12, 13, and 14 profiles in diagnosing lung cancer.

Publications

• No publications available at this time for this protocol.

Biomarkers

• KLK11
• KLK12
• KLK13
• KLK14
• KLK8

Data Collections

• No data collections available at this time for this protocol.

Start Date

Mar 26 2009

Estimated Finish Date

Mar 25 2009

Finish Date

Mar 25 2009

Protocol ID

280

Protocol Type

Reference Set

Field of Research

Proteomics

Collaborative Group

Lung and Upper Aerodigestive Cancers Research Group

Cancer Types

• Malignant neoplasm of bronchus and lung

Phased Status

1