Related Experiment Video
Updated: Aug 17, 2026
High-Content Screening Assay for the Identification of Antibody-Dependent Cellular Cytotoxicity Modifying Compounds
Published on: August 18, 2023
Cytotoxic assays for screening anticancer agents
Lamya A Baharith1, Abeer Al-Khouli, Gillian M Raab
1Department of Statistics, King Abdul Aziz University, Jeddah, Saudi Arabia.
Abstract:
In the process of identifying potential anticancer agents, the ability of a new agent is tested for cytotoxic activity against a panel of standard cancer cell lines. The National Cancer Institute (NCI) present the cytotoxic profile for each agent as a set of estimates of the dose required to inhibit the growth of each cell line. The NCI estimates are obtained from a linear interpolation method applied to the dose-response curves. In this paper non-linear fits are proposed as an alternative to interpolation. This is illustrated with data from two agents recently submitted to NCI for potential anticancer activity. Fitting of individual non-linear curves proved difficult, but a non-linear mixed model applied to the full set of cell lines overcame most of the problems. Two non-linear functional forms were fitted using random effect models by both maximum likelihood and a full Bayesian approach. Model-based toxicity estimates have some advantages over those obtained from interpolation. They provide standard errors for toxicity estimates and other derived quantities, allow model comparisons. Examples of each are illustrated.
Insights
This study introduces non-linear models as a superior alternative to linear interpolation for assessing anticancer agent cytotoxicity. These advanced models offer more precise toxicity estimates and enable robust comparisons for drug discovery.
Area of Science:
- Pharmacology
- Biostatistics
- Computational Biology
Background:
- The National Cancer Institute (NCI) uses linear interpolation to estimate drug doses inhibiting cancer cell growth.
- Current methods provide dose-response profiles but lack detailed statistical insights.
Purpose of the Study:
- To propose and evaluate non-linear fitting models as an alternative to NCI's linear interpolation for anticancer agent evaluation.
- To demonstrate the advantages of model-based toxicity estimates over interpolation methods.
Main Methods:
- Non-linear mixed-effects models were applied to dose-response data from cancer cell lines.
- Two non-linear functional forms were fitted using maximum likelihood and Bayesian approaches.
- The models were tested using data from two agents recently submitted to the NCI.
Main Results:
- Fitting individual non-linear curves was challenging, but a mixed-effects model approach successfully addressed these difficulties.
- Model-based estimates offer standard errors for toxicity and derived quantities, facilitating model comparison.
- The proposed non-linear models provide a more comprehensive analysis of cytotoxic activity.
Conclusions:
- Non-linear mixed models offer significant advantages over linear interpolation for anticancer agent screening.
- These advanced statistical approaches enhance the precision and interpretability of drug toxicity assessments.
- The findings support the adoption of model-based methods for more robust drug discovery pipelines.
