Purpose: Durable progression-free survivors (dPFSors) over 2 years have been reported among patients with melanoma or non-small-cell lung cancer (NSCLC) who received PD-(L)1 therapy. However, risk of progression still exists and the optimal imaging surveillance interval is unknown.
Methods: Individual patient data for progression-free survival (PFS) were extracted from PD-1 blockade clinical trials with a follow-up of at least 5 years. Patients with a PFS of at least 2 years were considered as dPFSors. Conditional risks of progression/death (P/D) every 3, 4, 6, and 12 months in each subsequent year were calculated. We prespecified three different levels of risk between scans (10%, 15%, or 20%) to allow clinicians and patients to decide on the scanning interval on the basis of considerations of imaging frequency and risk tolerance. An interval is considered acceptable if the upper bound of the 95% CI of the risk at each scan is lower than a prespecified level.
Results: Of 1,495 and 3,752 patients with melanoma and NSCLC, 474 (31.7%) and 586 (15.6%) were dPFSors, respectively. Among them, the PFS probability for an additional 3 years was 76.4% and 48.1%, respectively. Not more than 8% of patients had P/D in any quarter in the 3 years. With a risk threshold of 10%, melanoma dPFSors can be scanned every 6 months during the third year and then every 12 months in years 4 and 5. The interval for NSCLC would be every 3 months in the third year and every 4 months in years 4 and 5. The higher risk tolerance of 15% and 20% would allow for less frequent scans.
Conclusion: On the basis of their own risk tolerance level, our findings allow clinicians and dPFSors make data-driven decisions regarding the imaging surveillance schedule beyond every 3 months.