Nasopharyngeal necrosis following intensity-modulated radiation therapy of primary nasopharyngeal carcinoma-incidence rate and predictors of risk

Abstract

Objectives: This study aimed to investigate the incidence of post radiation nasopharyngeal necrosis (PRNN) in primary NPC after intensity modulated radiation therapy (IMRT) and identify the predictors of risk.

Methods: Data of 5798 NPC patients who received IMRT-based treatment between April 2009 and December 2015 were retrospectively reviewed. PRNN was diagnosed by MRI or nasopharyngoscopy. Dosimetric factors were selected by the least absolute shrinkage and selection operator logistic regression and applied to Cox proportional hazards modeling with clinical predictors.

Results: Among the 5798 patients, 53 developed PRNN-an incidence rate of 0.89%. Age > 55 years, diabetes, LDH > 170 U/L, and tumor volume of nasopharynx > 60.5 cm³,were independently associated with risk of PRNN(all p < 0.05. Dosimetric analysis showed that D_0.5cc^EQD2 of 80.20 Gy might be the dose constraint for nasopharynx (sensitivity = 62.3%, 33 out of 53; specificity = 84.2%, 4897 out of 5925). Besides, the RTOG dose constraints of V_110% (V_77.0) should be less than 0.2% in case of increasing risk of PRNN(HR = 2.28, 95% CI: 1.26-4.41, p = 0. 01).

Conclusion: Nasopharyngeal necrosis is rare after primary IMRT. The independent risk factors for this rare complication include age > 55 years, diabetes mellitus, LDH > 170 U/L, tumor volume of nasopharynx > 60.5 cm³, D_0.5cc^EQD2 > 80.20 Gy, and V_77.0 < 0.2% to the planning treatment volume of nasopharynx.

Keypoints: High radiation dose may lead to devastating nasopharyngeal necrosis after primary IMRT. Real world analysis will provide valuable information for prevention.

Findings: The aged, diabetes mellitus, large tumor volume, D_0.5cc^EQD2 > 80.20 Gy and V_77.0 < 0.2% to planning treatment volume increased the risk of nasopharyngeal necrosis.

Clinical relevance: This real-world study provided valuable information for prevention of PRNN. Compared with RTOG protocol, D_0.5cc^EQD2 > 80.20 Gy is a reliable evidence-based new complement to dose constraint, especially for T3-4 disease, who received high prescribe dose in China.

Keywords: Dose constraint; Intensity modulated radiation therapy; Least absolute shrinkage and selection operator logistic regression; Nasopharyngeal carcinoma; Nasopharyngeal necrosis.

Fig. 1

Study flow chart

Fig. 2

MRI examination of PRNN and Overall survival in PRNN and non-PRNN groups. A: Transverse T1-weighted image; B: Transverse contrast-enhanced T1-weighted image; C: Transverse T2-weighted image; D: Coronal T1-weighted image; E: Coronal contrast-enhanced T1-weighted image; F: Kaplan–Meier curves of overall survival in PRNN and non-PRNN groups

Fig. 3

AUC of important dosimetric factor and cut-off value. * Converted to equivalent 2 Gy fractionation (EQD2) using α/β = 10 Gy. **Translated according to the recommend total dose of 70 Gy in protocols

Fig. 4

Univariate analysis for covariates to estimate the risk of PRNN. Abbreviations: EBV = EpsteineBarr virus; N = node; T = tumor; HGB = hemoglobin, ALB = albumin, CRP = C-reactive protein; LDH = lactate dehydrogenase. * Converted to equivalent 2 Gy fractionation (EQD2) using α/β = 10 Gy

Fig. 5

Multivariate analysis for covariates to estimate the risk of PRNN. A: Multivariate analysis with clinical covariates and D_0.5cc; B: Multivariate analysis with clinical covariates and V_77.0(V_110%). *: Converted to equivalent 2 Gy fractionation (EQD2) using α/β = 10 Gy