Estimating sample size for a randomized clinical trial of lung cancer screening

Area of Science:

• Medical Screening
• Clinical Trial Design
• Biostatistics

Background:

• Randomized clinical trials (RCTs) are crucial for demonstrating the effectiveness of lung cancer screening but require substantial sample sizes and long follow-up periods.
• Key study design elements influencing sample size include accrual length, screening frequency, follow-up duration, and test interpretation methods.

Purpose of the Study:

• To present a tiered approach for sample size estimation in screening trials.
• To enable investigators to assess the impact of various study design features on sample size requirements for lung cancer screening RCTs.

Main Methods:

• A four-tiered approach to sample size estimation is detailed.
• Tier 1: Determines person-years at risk.
• Tiers 2-4: Incorporate study population characteristics, screening test performance, treatment benefits, and disease-specific mortality rates to estimate preclinical and clinical disease incidence.

Main Results:

• The tiered approach was applied to a lung cancer screening RCT using chest X-rays and computer-aided detection.
• Sample size calculations were performed considering diverse study design features and assumptions regarding cure rates, treatment complications, and endpoint misclassification.

Conclusions:

• The proposed tiered methodology provides a framework for sample size determination in screening trials.
• This approach empowers researchers to systematically evaluate how modifications in study design influence the required sample size for lung cancer screening RCTs.