Issues possibly associated with misinterpreting survival data: A method study

Area of Science:

• Oncology
• Biostatistics
• Clinical Trials

Background:

• Kaplan-Meier analyses are crucial for interpreting survival data in clinical cancer studies.
• Misinterpretations of survival data can lead to erroneous conclusions in research.

Purpose of the Study:

• To raise awareness of potential pitfalls in interpreting survival data from clinical cancer studies.
• To prevent errors in future studies by highlighting issues in survival analyses.

Main Methods:

• Evaluation of a randomized controlled trial involving high-risk neuroblastoma patients.
• Review of survival analyses, statistical approaches, baseline characteristics, and primary endpoints.
• Reenactment of survival functions from pictured data to estimate hazard ratios.

Main Results:

• No significant difference in overall survival between treatment groups was identified, contrary to trial reporting.
• No effective crossing of survival curves was observed.
• No significant difference in event-free survival between comparable treatment groups was found, also contrary to trial reporting.

Conclusions:

• Statistical issues, including assumed curve crossing and changes in statistical approach, can complicate survival data interpretation.
• Differences in pretreatment characteristics and the use of event-free survival as a primary outcome can also lead to misinterpretation.
• Careful statistical handling and interpretation are essential to avoid potential misinterpretations in future clinical studies.