Cancer Risk Assessment and the Biostatistical Revolution of the 1970s-A Reflection.

Area of Science:

• Toxicology
• Environmental Health
• Risk Assessment

Background:

• Early toxic substance risk assessment relied on the NOAEL-safety factor approach, assuming a risk-free exposure threshold.
• The understanding that single-cell mutations can cause cancer challenged the threshold concept, suggesting linear dose-response relationships.

Purpose of the Study:

• To review the historical evolution of cancer risk assessment methodologies.
• To analyze the adoption and persistence of specific risk assessment models by regulatory agencies like the EPA.

Main Methods:

• Historical review of toxicological risk assessment paradigms.
• Analysis of the development and application of statistical models in cancer risk assessment.
• Examination of EPA's cancer risk assessment guidelines and practices over time.

Main Results:

• The shift from threshold-based risk assessment to model-based extrapolation, notably the Linearized Multistage Model, occurred in the 1970s.
• The EPA's adoption of the Linearized Multistage Model emphasized statistical modeling for low-dose extrapolation.
• Despite updated guidelines allowing varied approaches, current EPA cancer risk assessments largely mirror methods established over four decades ago.

Conclusions:

• Cancer risk assessment has undergone significant methodological shifts driven by scientific understanding of carcinogenesis.
• The persistence of older assessment methods suggests inertia in regulatory practices, despite advancements in the field.