eduzhai > Physical Sciences > Physics Sciences >

Causal Measures for Prognostic and Predictive Biomarkers

  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: Researchers conducting randomized clinical trialswith two treatment groups sometimes wish to determine whether biomarkers arepredictive and or prognostic. They can use regression models with interactionterms to assess the role of the biomarker of interest. However, although theinteraction term is undoubtedly a suitable measure for prediction, the optimalway to measure prognosis is less clear. In this article, we define causalmeasures that can be used for prognosis and prediction based on biomarkers. Thecausal measure for prognosis is defined as the average of two differences instatus between biomarker-positive and -negative subjects under treatment andcontrol conditions. The causal measure forprediction is defined as the difference between the causal effect of thetreatment for biomarker-positive and biomarker-negative subjects. We also explain the relationship between the proposed measures and theregression parameters. The causal measure for prognosis corresponds to the termsfor the biomarker in a regression model, where the values of the dummy variablesrepresenting the explanatory variables are -1 2 or 1 2. The causal measure for prediction is simply the causal effectof the interaction term in a regression model. In addition, for a binaryoutcome, we express the causal measures interms of four response types: always-responder, complier, non-complier,and never-responder. The causal measure for prognosis can be expressed as a functionof always- and never-responders, and the causal measure for prediction as afunction of compliers and non-compliers. This enables us to demonstrate thatthe proposed measures are plausible in the case of a binary outcome. Our causal measures should be used to assess whethera biomarker is prognostic and or predictive.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...