Radiotherapy (RT), either alone or in combination with chemotherapy, is the standard treatment for localized NPC. RT targets the gross tumor volume (GTV) of the primary tumor, metastatic lymph nodes (LN), and risk areas, considering the tumor spread patterns. Primary tumors of the nasopharynx tend to invade the surrounding soft tissues and bones, and spread along several foramina of the skull base. Cervical LN metastasis is widespread, with 60%–90% of patients present with LN metastasis at diagnosis [3]. The pattern of cervical LN metastasis in NPC is predictable and ordered. Skip metastasis is rare, with a risk of 0.5% to 2.7% [4,5]. Level II and lateral retropharyngeal LNs are the most commonly involved areas, followed by levels III, VA, and IV. As the nasopharynx is a midline structure, the efferents of lymphatics draining the central location often reach lymph nodes on both sides, resulting in bilateral lymph node metastases in the neck. This is particularly common in NPC, affecting up to 50% of patients.

Traditionally, radiation targets routinely included the primary tumor, retropharyngeal area, and whole neck bilaterally. The Radiation Therapy Oncology Group 0225 protocol [6] and an institution in Hong Kong [7] routinely include bilateral level I to V LNs. This was based on the pattern of LN metastases, the radiation field of the conventional two-dimensional RT technique, and the use of less accurate imaging. In contrast, with more advanced imaging methods available, such as magnetic resonance imaging (MRI) and positron emission tomography/computed tomography, LN metastases more easily and accurately detected [8]. Intensity-modulated RT (IMRT) is now the standard technique. IMRT delivers a more precise and conformed radiation dose, allowing irradiation of the selected target volume. In the era of precision medicine, the routine use of traditional RT for the treatment of NPC is currently being challenged due to advancements in diagnostic and therapeutic techniques.

As target volume delineation has become more sophisticated, evidence-based consensus guidelines for target volumes in NPC have been suggested [9]. In addition, the accumulated tumor control and failure pattern data, after selected target volume irradiation, have led to significant advances in personalized treatment for RT targeting in NPC [10-17]. As a result of these efforts, a recent randomized phase III trial involving that elective upper neck irradiation of the uninvolved neck has shown similar regional control with less toxicity compared with whole-neck irradiation in patients with N0-N1 NPC [18].

Kim et al. [19] provided further evidence for this trend. They retrospectively analyzed 236 patients with NPC who were treated with IMRT. Of these, 212 received level IB-sparing RT, and 24 received non-IB-sparing RT. There was no level IB recurrence despite the use of level IB-sparing RT. After propensity score matching analysis, they reported that level IB LN-sparing RT is sufficiently safe and feasible for treating patients with NPC without level IB LN metastasis. The inclusion of level IB LNs in the target volume during NPC treatment is controversial. Level IB involves lymphatic drainage from the hard and soft palates, oral cavity, anterior nasal cavity, gums, and cheeks. The incidence of level IB LN metastasis is very low, approximately 2%–4% in patients with NPC. Moreover, isolated metastasis to level IB LN is rare [5]. As such, level IB irradiation is not routinely recommended. However, it is used in selected cases, such as in patients with level II LN involvement. The rationale for using level IB irradiation for these patients is that nearly all patients with level IB LN metastasis simultaneously have level II LN. It is assumed that the involvement of level II LN can block the lymphatic flow and cause the aberrant flow to level IB. Some authors have suggested that level II LN with a large or extracapsular extension (ECE) is a high-risk factor for level IB LN metastasis. Hence, patients with these features may need to be irradiated in the level IB areas. Zhang et al. [15] suggested that the level IB sparing technique is safe and feasible for low-risk patients with NPC and in the absence of level IIA LNs ≥2 cm or level IIA LNs with ECE, bilateral neck involvement, or oropharynx involvement. No level IB recurrence (0/904) was observed in among those who were low-risk. Ou et al. [17] also confirmed that the percentage of level IB recurrence in those treated with level IB-sparing IMRT was only 0.46% (1/216). Interestingly, data from Kim et al. [19] showed that about 40% of patients treated with level IB LN-sparing RT have high-risk features including ≥2 cm level IIA LNs and/or with ECE. Bilateral LN involvement was observed in approximately 35% of patients. Despite these high-risk features, no recurrence of IB LN was observed after level IB LN-sparing RT. Our institution has similar experiences. Level IB was not electively irradiated in our practice, regardless of the level II LN status. There was only one (0.3%) recurrence among 293 patients with NPC [20]. A recent study by Wang et al. [11] also supported tthe use of level IB-sparing RT and suggested that it is safe and feasible in patients with NPC with high-risk features. After propensity matching, the level IB failures in the level IB-sparing and level IB irradiation groups were 3/169 (1.8 %) vs. 2/169 (1.2 %) in patients with ECE or level II LN ≥2 cm.

Level IB LN-sparing RT aims to decrease the submandibular gland (SMG) dose and volume, which may reduce xerostomia. However, previous results have not shown clinical improvement in patient-reported xerostomia [15]. They explained that RT planning was performed without a dose constraint on the irradiated SMGs. Despite level IB LN-sparing RT, the mean dose exceeded the threshold dose of 39 Gy. Contrary to these results, Kim et al. [19] showed a 23.4% absolute reduction in the incidence of grade 2 xerostomia in the bilateral IB-sparing RT group compared with the non-IB-sparing RT group in their study, even though the mean SMG doses of level IB LN-sparing RT on the ipsilateral and contralateral sides were 51.2 Gy and 45.2 Gy, respectively [19]. Wang et al. [11] also showed that grade ≥1 dry mouth incidence after 5 years was lower in the level IB-sparing group (27.5% vs. 16.5%, p = 0.029). It is now feasible to reduces the mean dose to SMGs by 39 Gy using sophisticated RT, which may increase the possibility of a clinically significant improvement in symptoms related to xerostomia [21]. Furthermore, some authors have reported the feasibility of reducing the elective RT dose to approximately 36–40 Gy [22-24]. The lower elective dose can further decrease the RT dose to the SMGs, which may reduce RT-induced xerostomia, and improve patient outcomes.

In summary, RT for NPC is becoming increasingly sophisticated as the evidence for the use of advanced techniques accumulates. Level IB-sparing RT may be feasible and safe for patients with NPC and negative IB LN. Using this technique, quality of life may be improved by reducing the RT dose to SMGs in patients with NPC.