Does Accumulated Dose to Parotid Glands Improve Prediction of Patient Reported Outcomes Related to Salivary Function in Nasopharyngeal Cancer Patients

To investigate the dose delivered to parotid glands in nasopharyngeal patients at our institution, using deformable image registration (DIR) and dose accumulation, compared to planned dose and investigate the relationship to patient reported outcomes (PRO) related to salivary function. Retrospective dose accumulation (using daily cone-beam CT (CBCT) images) was performed on 50 nasopharynx patients treated with chemoradiotherapy [70 Gy (high-dose) and 56 Gy (low-dose) in 35 fractions] from 2013 to 2015. An updated method of DIR is being explored for this cohort. Of this cohort, 21 patients had baseline (pre-radiation therapy (RT)) and 6 months post-RT MDASI (MD Anderson Symptom Inventory)-HN scores reported (Scale 1 to 10). For each patient, the change in score per item (dry mouth and mucus) from baseline to 6 months post-RT was used for analysis. All patients had right and left parotid gland contours except one patient that did not have a left parotid contoured. A total of 99 parotid glands were analyzed. Daily delivered dose was calculated on the CBCTs and deformed back to and accumulated onto the planning CT to estimate the delivered dose. Dosimetric parameters were extracted for the parotids for both planned and delivered dose. Thresholds for classifying the parotids as spared or not spared were 26 Gy and 30 Gy for mean dose and D50% respectively as per our institutional protocol. The MDASI score changes were then compared to the parotid gland doses, using both the planned and delivered dose to classify the parotids as spared or not spared. The average planned mean dose to the parotid gland was 41.2 Gy ± 9.9 Gy. Three out of ninety-nine parotids were planned to < 26 Gy mean dose and two of those received a delivered mean dose greater than 26 Gy. The average planned D50% to the parotid gland was 38.4 Gy ± 13.3 Gy. The planning goal for D50% was achieved in 31.3% of the parotids, with 19.3% of these receiving a delivered dose > 30 Gy. There was an increase in delivered mean dose in 87.9% of the parotids, and an increase in delivered D50 dose in 90.9% of the parotids. Of the 21 patients with PRO, the mean MDASI score increase for mucus was 3.15(R: -2 to 8) for the cases that were planned to spare and 3.0(R: -3 to 6) in patients that were not planned to spare. The mean MDASI score increase for mucus was 2.6(R: -2 to 8) for the parotids that were actually spared and 3.5(R: 0 to7) for patients that were not spared, evaluated using delivered dose. Those with planned parotid sparing had a increase of 2.23(R: 0 to 6) and those that were not planned to spare had a increase of only 0.75(R: -3 to 4) in MDASI dry mouth scores. Patients that were actually spared had a score increase of 1.8(R: 0 to 4) while the patients that were not actually spared increased 1.5(R: 0 to 7). Dose accumulation is a useful tool to estimate delivered dose to the parotids which is higher than planned dose in this cohort. Using accumulated dose is a better predictor of MDASI mucus outcomes and potentially a better predictor of MDASI dry mouth outcomes. Use of this information could prove useful in management of patients and in designing appropriate adaptive strategies.