Functional dose to the parotid gland: a new regional based dose metric in NTCP models for xerostomia
Maria I. van Rijn - Dekker,
The Netherlands
OC-0109
Abstract
Functional dose to the parotid gland: a new regional based dose metric in NTCP models for xerostomia
Authors: Maria I. van Rijn - Dekker1, Peter van Luijk1, Ewoud Schuit2, Arjen van der Schaaf1, Johannes A. Langendijk1, Roel J.H.M. Steenbakkers1
1University Medical Center Groningen, Department of Radiation Oncology, Groningen, The Netherlands; 2University Medical Center Utrecht, Julius Center for Health Sciences and Primary Care, Utrecht, The Netherlands
Show Affiliations
Hide Affiliations
Purpose or Objective
Despite improvements, HNC patients still experience xerostomia due to RT-induced salivary gland damage. The parotid gland’s (PG) stem cells, concentrated in the gland’s main ducts (stem cell rich (SCR) region), play a critical role in the PG’s response to radiation. However, treatment optimization requires dose metrics accounting for the relative contributions of dose to this SCR region and dose to the remainder of the gland (non-SCR region) to the risk of xerostomia. This study aimed to develop such dose metric: the functional dose. Subsequently, NTCP models including additional contributions of other risk factors were constructed using multivariable logistic regression analyses.
Material and Methods
Treatment and toxicity data of 1,013 HNC patients treated with definitive RT was obtained from our prospective data registration program. Xerostomia was measured with the EORTC QLQ-H&N35, the Groningen Radiotherapy-Induced Xerostomia questionnaire and the CTCAE version 4.
For xerostomia endpoints associated with PG dose, functional dose (Dfunc) to the PG was defined as Dfunc,PG = Dmean,SCR + r * Dmean,non-SCR. In our cohort for weighting factor r = 3.6, Dfunc,PG corresponds to Dmean,PG. This reflects the difference in volume between the SCR and the non-SCR region. A value of r < 3.6 indicates an enhanced contribution of Dmean,SCR to the risk of xerostomia. The value of r for which Dfunc,PG was the best predictor for xerostomia endpoints, was estimated in a subset of 102 patients included in a previously-published randomized controlled trial testing stem cell sparing RT(1).
Next, for each endpoint Dfunc,PG, dose to other organs and clinical factors were used to develop multivariable NTCP models in 663 patients. Finally, these models were validated in the remaining 350 patients.
Results
Dose to contralateral PG (sub)structures was significantly associated with patient-rated daytime, eating-related and physician-rated grade ≥2 xerostomia. No associations with patient-rated general, activity-related and nighttime xerostomia were found.The estimations of r were smaller than 3.6 for most xerostomia endpoints (Table 1), demonstrating that Dmean,PG underestimates the contribution of dose to the SCR region to the risk of xerostomia.
The most-frequently additional risk factors identified in multivariable modeling were pretreatment xerostomia and Dmean oral cavity (Table 1).
The models showed moderate-to-good discrimination and calibration after internal and external validation (Table 2).
Conclusion
We developed a new dose metric to account for the larger contribution of dose to the stem cell rich region to the risk of xerostomia. This new dose metric was subsequently used to develop novel multivariable models that can be used for the routine implementation of stem cell sparing RT.
(1) Steenbakkers, RJHM et al. Parotid Gland Stem Cell Sparing Radiation Therapy for Patients with Head and Neck cancer: A Double-Blind Randomized Controlled Trial. Int. J. Radiat. Oncol. 112, 306-316 (2022).