The Feasibility of Level Ib Sparing Intensity Modulated Radiation Therapy in Nasopharyngeal Carcinoma Patients with High-Risk Factors: Based on International Guideline

Abstract

In spite of the rarity of level Ib recurrence after intensity-modulated radiation therapy, the International Guideline (IG) provides the risk factors for prophylactic coverage. In practice, however, there are significant differences between institutions. The purpose of this study is to examine the feasibility of sparing level Ib IMRT in NPC patients with high-risk factors based on IG. From January 2014 to October 2017, newly-diagnostic, non-metastatic NPC patients in our center were retrospectively reviewed. According to the risk factors of prophylactic level Ib coverage in patients with negative level Ib recommended by IG, the characteristics of pre-treatment MRI were analyzed. Four high-risk factors were identified: a. involvement of the structures that drain to level Ib as first echelon (FES), including anterior half of nasal cavity, oral cavity, b. involvement of submandibular gland (SMG), c. with radiologic extranodal extension (rENE) in level II LNs, or d. maximal axial diameter (MAD)≧2 cm in level II LNs. Patients with risk factors were divided into Cohort A (with risk factors a), Cohort B (with risk factor b, but without a), and Cohort C (only with risk factors c and/or d). Recurrence rates of level Ib and regional relapse-free survival (RRFS) rates were evaluated in different cohorts. A total of 961 patients were finally included. Thirty-six cases (3.7%) presented with radiologically positive level Ib metastasis. For the other patients with negative Ib LNs, there were 18, 65, 421, and 444 cases classified as FES involvement, SMG involvement, level II LNs with rENE, and level II nodal with MAD ≧2 cm. Excluding overlap, a total of 571 patients with risk factors were divided into three groups: Cohort A (n = 18), Cohort B (n = 49) and Cohort C (n = 504). Nine patients (9/961, 0.94%) developed level Ib recurrence. Except for 1 patient with positive Ib LNs at diagnosis, 2 did not meet any of the risk factors, while the other six (6/9, 66.7%) met at least one risk factor. The rate of recurrence at neck level Ib was highest in Cohort A (11.1%, 2/18; Ib-sparing group: 0/10, 0.0% vs Ib-covering group: 2/8, 25.0%; P = 0.183). In Cohort B, no cases were found with level Ib recurrence (0.0%, 0/49). In Cohort C, the rates were rare (0.8%, 4/504) in both groups (0.7%, 2/276 vs 0.9%, 2/228; P > 0.999). Among the three Cohorts, there were no significant differences in 5-year RRFS between two groups, which were 90.0% vs 62.5% (p = 0.248), 90.9% vs 92.0% (p = 0.905), and 92.6% vs 90.1% (p = 0.445), respectively. Among patients with high-risk factors, the incidence of grade 3-4 late dry mouth symptom was higher in the level Ib-covering group (3.1% vs 7.5%, P = 0.033). Level Ib sparing appears safe and feasible for NPC patients with negative level Ib LNs, even if combined with risk factors: SMG involvement, and/or level II with rENE, and/or level II MAD ≧2 cm. Level Ib-sparing irradiation reduces dry mouth symptoms compared with level Ib-irradiation.