In the present study, we retrospectively analyzed the results of radiotherapy in 15 patients with intact SCC of the EAC. The 3 year LC rate for all patients was 51%, the DFS rate was 44%, and the OS rate was 73%. In the univariate analysis, T-category was found to be a significant prognostic factor for LC and DFS. The 3 year LC, DFS, and OS rates in T1−3 cases were 74%, 62%, and 89%, respectively. However, in T4 cases, the 3 year LC, DFS, and OS rates were 17%, 17%, and 50%, suggesting that the efficacy of chemoradiotherapy for T4 cases was inadequate. Direct comparison with previous publications is difficult because of the inclusion of inner ear cancer or differences in stage classification. However, most studies demonstrated that T4 cases showed a poor prognosis after treatment (Supplementary Data 2).
Takenaka et al. performed a meta-analysis of data from 274 patients who were treated with chemoradiotherapy (and surgery) for EAC cancer [13]. They concluded that the treatment outcomes of chemoradiotherapy without surgery in T3−4 cases were comparable with those of surgery with postoperative radiotherapy, and that T4 cases had a significantly poorer prognosis than T3 cases. These findings are consistent with the results of the present study, in which the T4 cases had a poorer prognosis than the T1−3 cases (Fig. 3, Supplementary Data 2). In the present study, the LC rate of T4 cases was low (17%), and all five recurrences in T4 cases occurred in the high-dose area within the irradiation field. This suggests that T4 cases are resistant to X-rays, and that the chemoradiotherapy for T4 cases showed insufficient local efficacy. To address this issue, dose-escalation or more intense modalities are needed. For example, carbon-ion radiotherapy is a promising modality with a strong antitumor effect on X-ray-resistant tumors [14]. Koto et al. reported the treatment outcomes of carbon-ion radiotherapy for T3−4 squamous cell carcinoma of the EAC and middle ear [15]. Of the 13 patients in their study, three had T3 tumors and ten had T4 tumors. The LC rates at 1 and 3 years were 72% and 54%, respectively, and the OS rates were 70% and 40%. Although most of the patients had T4 tumors, the LC rate of carbon-ion radiotherapy appeared to be favorable compared with other studies using X-ray-based treatment, including the present study (Supplementary Data 2).
Several studies on treatment strategies for EAC cancer concluded that a combination treatment of surgery and postoperative radiation was the most curative for T2−3 disease [3, 5, 6, 16, 17]. However, adverse events from surgery are not uncommon [3, 18]. In cases with middle ear cavity invasion, subtotal temporal resection is often performed, and the facial nerve and inner ear are resected together. Generally, facial paralysis or hearing impairment caused by surgery does not recover [19]. In the present study, patients with T2−3 disease were treated with chemoradiotherapy without surgery. Only one patient with T2 disease experienced local recurrence, and no patients developed a grade 3 or worse later adverse event. Although the number of T2−3 cases in our study was too small to compare the two treatment strategies, considering the balance between efficacy and toxicity, chemoradiotherapy as a primary treatment may be a reasonable choice for T2−3 cases.
The optimal regimen for chemotherapy remains unknown. In the present study, patients with T2−4 tumors were treated by radiotherapy combined with concurrent CDDP, CDDP and DTX, or cetuximab. In the group who underwent radiotherapy with CDDP, there were three cases of T2 disease, two cases of T3 disease, and three cases of T4 disease. The CDDP was administered weekly, and the total dose was at least 200 mg/m2. In the T2–3 cases, only one out of five patients experienced local recurrence. A previous study recommended a total CDDP dose of more than 200 mg/m2 for nasopharyngeal cancer [20]. The results of the present study suggest that the total CDDP dose should also be more than 200 mg/m2 for EAC cancer. Of the two patients who underwent radiotherapy with CDDP and DTX, one had T2 disease and one T4 disease, and both achieved CR. It should be noted that among the T4 cases, only the combination of radiotherapy with CDDP and DTX resulted in CR. However, it is important to interpret this result carefully because only two patients underwent this treatment. Radiotherapy with cetuximab was also performed in T3−4 cases. One patient with T3 disease achieved CR, and their follow-up was terminated after 5 years. Two patients with T4 disease experienced local recurrence at a relatively early period (4.4 and 7.7 months), and the treatment effect of radiotherapy with cetuximab appears to be insufficient for T4 disease.
In the present study, acute adverse events improved quickly, while late adverse events did not improve. One patient with T3 disease (patient number 7 in Supplementary Data 1) experienced G2 hearing impairment as a late adverse event (Table 3). Marks et al. showed that the mean dose to the cochlea should be less than 45 Gy to avoid hearing impairment [21]. In our case, the mean dose to the cochlea was 64 Gy because the tumor was very close to the inner ear, and high-dose irradiation to the cochlea was unavoidable, despite using an IMRT technique. Hearing impairment should be avoided as it is directly linked to worse quality of life, but tumor localization can make it difficult to avoid hearing impairment in some cases, especially in T3–4 cases with inner ear invasion. When treating such patients, clinicians need to fully explain the possibility of hearing impairment at the time of informed consent. Brain necrosis, bone necrosis, and soft tissue necrosis, which have all been reported in previous studies, did not occur in the present study [8, 17].
One limitation of the present study is its single-institution retrospective design with a small number of patients. Another limitation is the variation in treatment methods in terms of beam delivery technique, radiation dose, and combined systemic therapy. Further prospective studies using unified radiotherapy and concurrent chemotherapy are warranted.