Commentary: On measurement error, PSA doubling time, and prostate cancer

Area of Science:

• Epidemiology
• Biostatistics
• Oncology

Background:

• Exposure measurement error is a significant challenge in epidemiological research.
• Prostate-specific antigen doubling time (PSADT) is crucial for managing recurrent prostate cancer.
• Existing PSADT calculations often do not account for inherent measurement errors.

Purpose of the Study:

• To highlight a real-world example of exposure measurement error in PSADT calculation.
• To introduce a diagnostic tool that corrects for PSADT measurement error for physicians and patients.
• To provide a more accurate probability for guiding prostate cancer treatment decisions.

Main Methods:

• Derived an expression for the probability of true PSADT being below a critical value, adjusting for measurement error.
• Developed an Excel calculator to compute PSADT and the associated probability from time and ln(PSA) values.
• Analyzed statistical issues related to PSADT estimation using limited PSA data points.

Main Results:

• The study quantifies the probability of a patient's true PSADT being below a concerning threshold, accounting for estimator variability.
• An accessible Excel tool is provided for calculating this adjusted probability.
• Identified limitations in current PSADT estimation methods, often based on few data points.

Conclusions:

• The derived probability offers a more reliable measure than unadjusted PSADT for clinical decision-making.
• The Excel calculator serves as a practical tool for physicians and patients to better understand prostate cancer recurrence risk.
• Future epidemiological studies on PSA doubling time must incorporate corrections for measurement error.