Adaptive IMRT Planning Improves Pharyngeal Sparing for Oropharyngeal Cancer

Abstract

To evaluate adaptive IMRT planning based on intra-treatment CT for normal tissue sparing in oropharyngeal cancer patients. A total of 45 oropharyngeal SCC patients (n = 20 BOT, n = 24 tonsil, n = 1 unknown primary; T stage: n = 1 T0, n = 8 T1, n = 18 T2, n = 9 T3, n = 9 T4) on an institutional review board-approved study were retrospectively re-planned. All patients had FDG-PET and contrast CT before treatment and after 20 Gy. Prescription was 44 or 50 Gy (low risk PTV) and 70 Gy (high risk PTV) in 2 Gy fractions. Accelerated adaptive IMRT planning workflow was developed: Target and organ at risk (OAR) contouring was assisted with rigid and deformable registration from pre-treatment CT and subsequently verified. A physician assessed potential planning changes based on prior OAR doses and volume changes after 20 Gy. Adapted plans were generated using the changed treatment volumes, prioritized dose-volume constraints and original plan parameters. Volumes and median dose (Dmedian) for pre- and intra-treatment targets and OARs were compared for original and adapted plans. Wilcoxon signed rank tests (2-tailed, p < 0.05) were performed to assess these differences. Subgroup analyses for pharynx dose based on site (BOT vs. tonsil) and T stage (T0/T1/T2 vs. T3/T4) were performed using Wilcoxon Rank Sum Test (2-tailed, p < 0.05). Incorporating prioritized dose-volume constraints in adaptive planning reduced planning time compared to a de novo plan. Volumes for GTV (primary and LN), PTVs, pharynx, parotids, larynx, and submandibular glands (SMG) significantly decreased after 20 Gy. Oral cavity volume change was not significant. Dmedian was significantly reduced for pharynx, oral cavity, both parotids, larynx, and right SMG. The adapted plan spared a median of 3.1 Gy to the pharynx, with > 6.7 Gy sparing for a quarter of patients. Pharynx sparing did not differ significantly based on tumor site (p = 0.8) or T stage (p = 0.7). Adaptive IMRT planning based on CT after 20 Gy reduced Dmedian to OARs, while maintaining PTV coverage. Clinically significant pharyngeal dose sparing was observed. Reduced planning time makes this adaptive planning feasible for clinical implementation.Abstract MO_23_2679; Table 1Volume differences (%) between intra- and pre-treatment CT and Dmedian differences (Gy) between adapted and clinical plansStructureΔ Volume intra- and pre-treatment CTΔ Dmedian adapted and clinical plansMedian (Q1, Q3) (%)p-valueMedian (Q1, Q3) (Gy)p-valueGTV primary-38.4 (-57.1 -24.4)<0.001--GTV LN-23.2 (-39.4, -8.4)<0.001 (n=43)--PTV 70 Gy-10.9 (-17.7, -6.1)<0.001--PTV 44/50 Gy-1.4 (-3.7, 1.0)0.003 (n=44)--Pharynx4.3 (-1.7, 8.7)0.018-3.1 (-6.7, -1.3)<0.001Oral Cavity1.8 (-4.1, 7.3)0.247-1.6 (-2.6, -0.3)<0.001R Parotid-11.0 ( -14.9, -2.2)<0.001-0.8 (-2.7, 0.4)0.002L Parotid-13.5 (-18.6, -6.8)<0.001-1.0 (-2.6, 0.9)0.015Larynx3.8 (-5.4, 17.3)0.015-0.9 (-1.7, 0.4)0.007 (n=44)R SMG-11.6 (-17.0, -4.3)<0.001-0.7 (-3.1, 0.3)0.006 (n=26)L SMG-6.7 ( -18.8, -2.4)<0.0010.0 (-1.5, 0.5)0.378 (n=29) Open table in a new tab