Using and interpreting surrogate end-points in cancer research

Area of Science:

• Oncology
• Biostatistics
• Epidemiology

Background:

• Molecular, cellular, and histological markers are proposed as surrogate end-points for cancer (SECs).
• The validity of a SEC depends on its necessity in the causal pathway to cancer.
• Alternative causal pathways can weaken the inference from a SEC to cancer incidence.

Purpose of the Study:

• To evaluate the criteria for determining the validity of surrogate end-points for cancer (SECs).
• To identify the necessary conditions for a marker to be considered a valid SEC.
• To guide the selection and interpretation of SECs in cancer research.

Main Methods:

• Review of existing literature on cancer surrogate end-points.
• Analysis of causal pathways linking interventions, SECs, and cancer.
• Emphasis on empirical studies addressing the relationship between SECs and cancer, and interventions and SECs.

Main Results:

• Colorectal adenomatous polyps are a valid SEC as they are obligate precursors to most large bowel cancers.
• The presence of alternative causal pathways or incomplete mediation by the SEC can compromise its validity.
• SEC validity can be intervention-specific, requiring careful consideration of the exposure context.

Conclusions:

• Definitive evidence for cancer etiology and prevention requires SECs that are necessary steps in the causal pathway.
• Studies must rigorously assess the relationship between interventions, SECs, and cancer, including mediation.
• Careful evaluation of SEC measurement error is essential to avoid attenuated results.