Clinical Implementation and Surveillance Strategies

Dr. Melody outlines her current practice approach, following NCCN guidelines recommending physical examination and blood work every 3 months post-treatment with imaging every six months for the first year. She includes ctDNA with quarterly blood work but typically stops after 1 year, acknowledging evolving practice as technology emerges. For Maria's case, with negative PET-CT (relapse risk <20%) and negative ctDNA (single-digit relapse risk), the 18-month positive result creates anxiety-provoking scenarios.

Single positive values require trend monitoring rather than immediate action, leading to increased testing, imaging, and follow-up. If subsequent ctDNA draws return negative, questions arise about resetting monitoring periods for additional years. Dr. Melody emphasizes that ctDNA currently serves as prognostic indicators rather than treatment-guiding algorithms, similar to traditional markers like International Prognostic Index (IPI) or comorbidities.

Dr. Roschewski discusses three Journal of Clinical Oncology papers showing largely concordant results, providing reassuring critical mass evidence pointing in similar directions. These data demonstrate 75% agreement between blood tests and PET scans, introducing new paradigms in the remaining 25% of cases. Although medicine dislikes uncertainty, existing single-test approaches already involve substantial uncertainty: positive PET scans at therapy completion predict progression in only 50% of patients, whereas negative PET scans still result in 15% to 20% progression rates.

Blood tests reduce but do not eliminate uncertainty. In discordant cases where PET scans are positive but blood tests undetectable, clinicians should carefully consider salvage therapy initiation. The challenging scenario involves negative PET scans with detectable ctDNA, occurring approximately 15% of the time (1 in 6 patients), creating decision-making dilemmas requiring additional data.

Dr. Roschewski contrasts 2 surveillance approaches: single time-point screening (like colonoscopies) versus serial testing for trend monitoring. Most screening tests predict future outcomes from one assessment, whereas serial monitoring provides dynamic information over time.