Individualized Re-Irradiation Dose for Recurrent Nasopharyngeal Carcinoma Based on Outcome Modeling

Abstract

The normal tissue toxicities such as temporal lobe necrosis, trismus and massive hemorrhage are a serious concern for re-irradiation of recurrent nasopharyngeal carcinoma (rNPC). This study explores a method to determine the optimal and patient-specific prescription dose for rNPC based on a quantitative consideration of normal tissue toxicity probability (NTCP) and tumor control probability (TCP). It has been reported that the total tolerable biologically effective dose (BED) calculated with α/β of 3 Gy (BEDmax3Gy) after two radiation treatments for rNPC can be higher than the tolerable BED of the initial radiation treatment, indicating normal tissue repair from radiation damage of the initial treatment. For example, BEDmax3Gy corresponding to 20% NTCP could be 129% the tolerable BED of the initial radiation treatment. Such observation can be described by the following expression: BED23Gy=BEDmax3Gy(1-BED13Gy/BEDmax3Gy)1/(r+1), where r=1.5+e0.5(τ-1) and τ is the time interval in year (gap) between two radiation treatments. BED13Gy and BED23Gy are BEDs of the initial and second radiation treatment. On the other hand, the multivariate analysis of TCP data from 295 rNPC patients revealed that the gap was a significant factor affecting TCP. The longer the gap, the better the tumor controls. A TCP model, called tumor regrowth-gap model, assuming that tumor regrowth after the initial treatment leads recurrence, was developed to investigate the TCP dependence on the time interval between the two treatment courses. The model parameters were determined by the TCP data of the 295 recurrent NPC patients and validated by another independent data set of 47 rNPC patients. Model fitting showed the prescribed re-irradiation dose and the speed of tumor cell regrowth significantly affected TCP of rNPC patients. The faster the tumor regrowth the earlier the recurrence and the poorer the clinic outcome of re-irradiation. The tumor BED of the second treatment (BED210Gy) that leads to 20% NTCP was found to be between 48 Gy to 58 Gy depending on the recurrence time, which corresponds to physical doses of 57 to 68 in 2 Gy per fraction. With the extracted model parameters, TCP at 5 years for irradiation of the rNPC patients were predicted based on allowable combination of BED210Gy and gap for a NTCP<20%. Assume that the prescription dose of the initial treatment is 70 Gy in 2 Gy per fraction.Abstract 1109; Tableτ (yr)0.51.01.52.02.53.03.54.04.55.05.56.0BED2 10 Gy (Gy)5887899193949596568083858890929395547073778184878991925259626670757882858789504850535762677276808386483940404348535964697478814629303134394450566166717544202222252935404652586368 Open table in a new tab The dependence of TCP/NTCP on BED and gap was established for re-irradiation of the rNPC. The model along with the extracted tumor-specific radiobiological parameters may be used to optimize re-irradiation dose and/or fractionation scheme for individual patients with rNPC.