A maximum Likelihood Approach to Analyzing Incomplete Longitudinal Data in Mammary Tumor Development Experiments with...

Area of Science:

• Biomedical research
• Cancer research
• Statistical modeling

Background:

• Longitudinal studies of mammary tumor development in mice often encounter missing data due to natural death or euthanasia.
• Censored data, arising from instrumental detection limits, also presents challenges in analyzing tumor progression.
• Accurate statistical methods are crucial for interpreting results from preclinical cancer studies.

Purpose of the Study:

• To develop and evaluate a statistical test for K-group comparisons in longitudinal mammary tumor studies.
• To address the challenges of missing and censored data in mouse cancer models.
• To provide a robust method for analyzing tumor development data in preclinical research.

Main Methods:

• A maximum likelihood methodology was employed to develop the statistical test.
• A likelihood ratio test was derived based on general distributions.
• The properties of the test were investigated theoretically and evaluated through a simulation study.

Main Results:

• The developed likelihood ratio test effectively handles missing and censored data in longitudinal studies.
• The simulation study demonstrated the performance of the proposed test.
• The test was successfully applied to real-world data from a mouse breast cancer study.

Conclusions:

• The new statistical test provides a reliable approach for analyzing mammary tumor development in mice.
• This method improves the analysis of preclinical cancer data by accommodating common data limitations.
• The findings contribute to more accurate interpretations of longitudinal cancer studies in experimental models.