Tooth Failure Post-Radiotherapy in Head and Neck Cancer: Primary Report of the Clinical Registry of Dental Outcomes in Head and Neck Cancer Patients (OraRad) Study

Abstract

Purpose: To elucidate long-term sequelae of radiation therapy (RT) in head and neck cancer (HNC) patients, a multicenter, prospective study, Clinical Registry of Dental Outcomes in Head and Neck Cancer Patients (OraRad), was established with tooth failure as its primary outcome. We report tooth failure and associated risk factors.

Methods and materials: Demographics and cancer and dental disease characteristics were documented in 572 HNC patients at baseline and 6, 12, 18, and 24 months after RT. Eligible patients were aged 18 or older, diagnosed with HNC, and receiving RT to treat HNC. Tooth failure during follow-up was defined as losing a tooth or having a tooth deemed hopeless. Analyses of time to first tooth-failure event and number of teeth that failed used Kaplan-Meier estimators, Cox regression, and generalized linear models.

Results: At 2 years, the estimated fraction of tooth failure was 17.8% (95% confidence interval, 14.3%-21.3%). The number of teeth that failed was higher for those with fewer teeth at baseline (P < .0001), greater reduction in salivary flow rate (P = .013), and noncompliance with daily oral hygiene (P = .03). Patients with dental caries at baseline had a higher risk of tooth failure with decreased salivary flow. Patients who were oral-hygiene noncompliant at baseline but compliant at all follow-up visits had the fewest teeth that failed; greatest tooth failure occurred in participants who were noncompliant at baseline and follow-up.

Conclusions: Despite pre-RT dental management, substantial tooth failure occurs within 2 years after RT for HNC. Identified factors may help to predict or reduce risk of post-RT tooth failure.

Fig. 1.

Recruitment, enrollment, and follow-up for tooth failure.

Fig. 2.

Relative hazard of first tooth loss event, by participant characteristics, in univariate analyses. Abbreviations: BL = baseline; C/C = compliant at baseline, compliant during follow-up; C/NC = compliant at baseline, noncompliant during follow-up; FU = follow-up; HS = high school; “Nadir as % of BL” = the decrease from baseline to the smaller of the 6- and 18month salivary flow measures, as a percent of baseline, so a 50% decrease is –50%; NC/C = NC/NC analogous; PC1 = first principal component of 4 radiation dose measures (maximum and mean dose to right and left parotids); PD = probing depth; RT = radiation therapy; BW Brigham and Women’s Hospital; CMC Atrium Health’s Carolinas Medical Center; NYU New York University; UConn University of Connecticut; UNC University of North Carolina; UPenn University of Pennsylvania; “18 mo as % of BL” = the 18-month salivary flow measure as a percent of the baseline measure, with the 6-month flow measure substituted if no 18-month measure was available. *Hazard ratio is for increasing the measure by one half of the interquartile range, which is PC1 0.73; larger maximum dose 519; larger mean dose 934; 18 mo as % of BL 25; Nadir as % of BL 21; mini- mum 6- and 18-month flow 0.21.

Fig. 3.

Relative risk of a tooth failing, by participant characteristics, in univariate analyses. Abbreviations: BL = baseline; C/ C = compliant at baseline, compliant during follow-up; C/NC = compliant at baseline, noncompliant during follow-up; FU = follow-up; HS = high school; “Nadir as % of BL” = the decrease from baseline to the smaller of the 6- and 18-month salivary flow measures, as a percent of baseline, so a 50% decrease is –50%; NC/C = NC/NC analogous; PC1 = first principal component of 4 radiation dose measures (maximum and mean dose to right and left parotids); PD = probing depth; RT = radiation therapy; BW Brigham and Women’s Hospital; CMC Atrium Health’s Carolinas Medical Center; NYU New York University; UConn University of Connecticut; UNC University of North Carolina; UPenn University of Pennsylvania; “18 mo as % of BL” = the 18-month salivary flow measure as a percent of the baseline measure, with the 6-month flow measure substituted if no 18-month measure was available. *Hazard ratio is for increasing the measure by one half of the interquartile range, which is PC1 0.73; larger maximum dose 519; larger mean dose 934; 18 mo as % of BL 25; Nadir as % of BL 21; minimum 6- and 18- month flow 0.21.