Probabilistic assessment method of the non-monotonic dose-responses-Part I: Methodological approach

Area of Science:

• Pharmacology and Toxicology
• Biostatistics
• Data Science

Background:

• Characterizing non-monotonic dose-response curves (NMDRCs) is crucial but challenging due to limitations in existing study data.
• Common data limitations include few tested doses, low sample sizes per dose, and the absence of raw data, hindering accurate assessment.
• Existing methods struggle to reliably evaluate the statistical plausibility of NMDRCs under these constraints.

Purpose of the Study:

• To propose and present a new methodological approach for the probabilistic characterization of non-monotonic dose-response curves (NMDRCs).
• To enable the statistical assessment of NMDRCs from previously conducted dose-response studies, even with limited summary data.
• To demonstrate the general principle and application of this probabilistic method for characterizing various dose-response curve types.

Main Methods:

• A three-step probabilistic methodology is introduced for characterizing NMDRCs.
• Step 1: Sampling from summary data to generate a new raw database, encompassing potential response variations.
• Step 2 & 3: Statistical analysis of sampled curves to determine slopes and their sign variations, enabling comprehensive curve characterization for both continuous and discrete data.

Main Results:

• The proposed method successfully generates a probabilistic characterization of NMDRCs.
• It addresses the limitations of few doses, low sample sizes, and missing raw data in existing studies.
• The methodology demonstrates applicability to various dose-response curve types and data types (continuous and discrete).

Conclusions:

• A novel probabilistic method provides a robust framework for assessing non-monotonic dose-response curves (NMDRCs) from summary data.
• This approach overcomes significant data limitations inherent in many existing dose-response studies.
• The method offers a versatile tool for characterizing diverse dose-response relationships, enhancing toxicological and pharmacological research.