Stepwise model-based treatment plan optimization to prevent dypshagia.in head and neck cancer.
OC-0106
Abstract
Stepwise model-based treatment plan optimization to prevent dypshagia.in head and neck cancer.
Authors: Langendijk|, Hans(1)*[j.a.langendijk@umcg.nl];van den Bosch|, Lisa(1);van den Hoek|, Anne(1);Oldehinkel|, Edwin(1);Meijer|, Tineke(1);Kierkels|, Roel(1);Scandurra|, Dan(1);Wolter|, Atje-Marleen(1);Mulder|, Dennis(1);van der Laan|, Hans-Paul(1);Both|, Stefan(1);van der Schaaf|, Arjen(1);Steenbakkers|, Roel(1);
(1)UMCG University Medical Center Groningen, Department of Radiation Oncology, Groningen, The Netherlands;
Show Affiliations
Hide Affiliations
Purpose or Objective
In 2007, prospective data collection started for all head and neck (HNC) patients treated with radiotherapy (RT). Physician- and patient-rated outcome measures were systemically collected at fixed time points. These data were used to develop and validate NTCP-models for dysphagia. The DVH-parameters in the NTCP-models were subsequently used to reduce dysphagia by using them as optimization parameter in treatment planning (NTCP-guided optimization).
This approach resulted in a shift from standard IMRT (ST-IMRT) with sparing of the parotid glands only, towards swallowing-sparing RT (SW-RT with either IMRT or VMAT) with additional sparing of the superior pharyngeal constrictor muscle (PCMsup) and supraglottic area. Based on a new NTCP-model, further optimization was obtained with oral cavity sparing RT (OCS-RT) with additional sparing of the oral cavity, using either VMAT or IMPT. The aim of this study was to test the hypothesis that dysphagia was significantly reduced by these two NTCP-guided treatment plan optimizations. The primary endpoint for these NTCP-models was grade ≥ 2 dysphagia at 6 months after RT.
Material and Methods
In total 1118 patients were included of which 201 were treated with ST-IMRT (2007-2011), 704 with SW-RT (2012-2017) and 213 with OCS-RT (2018-2019). Assessments were made at baseline, weekly during RT and at fixed time points after RT (Table 1). Physician-rated toxicity was scored according to the CTCAEv4.0 and patient-rated dysphagia with the EORTC QLQ-HN35. Univariable and multivariable logistic and linear regression analysis was performed whenever appropriate.
Results
The oral cavity Dmean decreased from 48.5 Gy (±15.7) to 40.9 Gy (±17.9) and 27.7 Gy (±16.3) with ST-IMRT, SW-RT and OCS-RT, resp. (P<0.001). The PCMsup Dmean decreased significantly from 57.6 Gy (±13.8) to 47.9 Gy (±18.8) and 41.5 Gy (±18.9) with ST-IMRT, SW-RT and OCS-RT, resp. (P<0.001). The prevalence of grade ≥ 2 acute dysphagia significantly decreased in week 3-12 with the introduction of SW-RT compared to ST-IMRT (Table 1) and in week 4-12 with oral cavity sparing versus SW-RT.
Both SW-RT and OCS-RT subsequently resulted in significant reductions of grade≥2 and grade≥3 dysphagia after radiotherapy up to 12 months post-RT. Patient-rated swallowing problems reduced with the introduction of SW-RT and OCS-RT, with significantly lower dysphagia scores at most time points during RT and post-RT (Figure 1).
Conclusion
The model-based approach, together with technological improvements resulted in two optimization steps that resulted in lower dose to relevant OARs resulting in significant reductions of different aspects of dysphagia. Although the NTCP-models were developed for grade ≥ 2 dysphagia at 6 months after RT, dysphagia decreased at multiple time points. Adding oral cavity sparing to sparing of the PCMsup and supraglottic area further prevented long-term swallowing problems.