Does time to retreatment matter? An NTCP model to predict radionecrosis after repeat SRS for recurrent brain metastases incorporating time-dependent discounted dose

Purpose: To develop and compare normal tissue complication probability (NTCP) models for recurrent brain metastases (BMs) treated with repeat single-fraction stereotactic radiosurgery (SRS), considering time-dependent discounted prior dose. Methods: We developed three NTCP models of BMs treated with GammaKnife-based SRS. The maximum dose to 0.2cc (D0.2cc) of each lesion-specific brain and one-year radionecrosis was fitted using a logistic model with equivalent-dose conversions in 2 Gy (EQD2). The M0 and M1-retreat modeled radionecrosis risk following SRS to 1029 non-recurrent lesions (patients=262) and 2nd SRS to 149 recurrent lesions (patients=87). The M1-combo model accounted for 2nd SRS and time-dependent discounted 1st SRS dose for recurrent lesions estimated by a modified Gompertzian function. Results: All three models fitted the data well (Chi-2 = 0.039-0.089 and p = 0.999-1.000). The fitted EQD250 was ~103 Gy for M0, ~88 Gy for M1-retreat, and ~165 Gy for M1-combo. The fitted EQD2_50 exhibited a progressively flatter dose-response curve across the three models, with values of 1.2 Gy for M0, 0.6 Gy for M1-retreat, and 0.4 Gy for M1-combo. For the brain D0.2cc of 29Gy and 19Gy, the steepest to shallowest dose-response or largest change in NTCP, i.e., NTCP29Gy - NTCP19Gy was observed in M1-retreat (0.16), M0 (0.14) and M1-combo (0.06). Conclusions: The model-fitted parameters predict that recurrent BMs have a lower threshold dose tolerance and a more gradual dose response for the 2nd SRS than non-recurrent BMs. This gradual dose-response becomes even more apparent when considering the time-dependent discounted 1st SRS as a cumulative 2nd SRS. Tailoring SRS retreatment protocols based on NTCP modeling can potentially enhance therapeutic efficacy.