Owing to the reports of the insufficient outcomes of RT alone with conventional fractionation, more-intensive treatment strategies have been used for T3 N0 glottic carcinoma without vocal cord fixation. Okumura et al. reported the treatment results of 74 patients with T3 N0 glottic carcinoma without vocal cord fixation . They divided the patients into three groups according to treatment regimen as follows: accelerated fractionated RT (AFRT) with 63–65.25 Gy in 27–29 fractions (2.25–2.4 Gy per fraction, once daily) without ENI (AFRT group, n = 41), CCRT using cisplatin or cetuximab with ENI of 40–46 Gy in 20–23 fractions (2 Gy per fraction, once daily) followed by boost irradiation with 24–30 Gy in 12–15 fractions (2 Gy per fraction, once daily; CCRT group, n = 10), hyperfractionated RT (HFRT) with ENI of 40–46 Gy in 20–23 fractions (2 Gy per fraction, once daily) followed by boost irradiation with 24–30 Gy in 16–20 fractions (1.5 Gy per fraction, twice daily; HFRT group, n = 23). The 3-year local failure and OS rates were 10% and 77%, 20% and 100%, and 26% and 87% for the AFRT, CCRT, and HFRT groups, respectively. Murakami et al. prospectively evaluated CCRT for T2 N0 glottic carcinoma staged according to UICC TNM staging system, fifth edition . The RT dose of 64 Gy was delivered with a once-daily fraction of 2 Gy using low-dose chemotherapy, including cisplatin and UFT. Of 51 cases, 24 were restaged as T3 N0 without vocal cord fixation according to UICC TNM staging system, sixth edition. Their 5-year LC and LP rates were 62% and 83%, respectively. Lin et al. evaluated the efficacy of CCRT for 17 patients with T3 glottic carcinoma . Of the 17 patients, 16 (94%) had N0 and one (6%) had N1 disease. Nine patients (53%) had vocal cord fixation. A median RT dose of 70 Gy (range, 62–72.5 Gy) was delivered with a once-daily fraction of 2–2.07 Gy using cisplatin-based chemotherapy. The 5-year LC, laryngectomy-free survival, and OS rate were 87%, 81%, and 89%, respectively. The LC, LP, and OS rates in our study were considered as equivalent to or greater than those in the previous reports.
The major acute toxicities of TPF-CRT are hematologic toxicities and mucositis. Katori et al. performed TPF-CRT for 34 patients with advanced SCC of the head and neck . The chemotherapy regimens consisted of two cycles of docetaxel 60 mg/m2 (day 1), 5-FU 750 mg/m2/day (days 1–5), and cisplatin 70 mg/m2 (day 4) for induction chemotherapy (n = 15, induction CRT group) and two cycles of docetaxel 50 mg/m2 (day 1), 5-FU 600 mg/m2/day (days 1–5), and cisplatin 60 mg/m2 (day 4) for CCRT (n = 19, CCRT group), which were similar to or the same as ours. The total RT doses were 64–70 Gy (mean, 66.9 Gy) for the induction CRT group and 63–74 Gy (mean, 67.8 Gy) for the CCRT group. Grade 3–4 leukocytopenia, neutropenia, and mucositis were observed in 6 (40%), 9 (60%), and 6 patients (40%) in the induction CRT group and 10 (53%), 8 (42%), and 15 patients (79%) in the CCRT group. Komatsu et al. performed TPF-CCRT for 24 patients with nasopharyngeal carcinoma . Chemotherapy included two cycles of docetaxel 50 mg/m2 (day 1), 5-FU 600 mg/m2/day (days 1–5), and cisplatin 60 mg/m2 (day 4), which were the same as ours. RT included ENI with 40–45 Gy followed by boost irradiation, for a total of 70 Gy. Grade 3–4 neutropenia and mucositis were observed in 10 (42%) and 16 patients (67%), respectively. Our results on hematologic toxicities were similar with theirs. On the other hand, our results indicated obviously milder mucositis as compared with their results, probably because we omitted ENI. Overall, the toxicities we encountered were acceptable.
Currently, ENI for levels of II, III, and IVa is recommended and widely performed for patients with T3 N0 glottic carcinoma [16, 17]. However, T3 N0 glottic carcinoma without vocal cord fixation was previously categorized into early T1–T2 N0 disease and usually treated with RT without ENI. Regional failure after treatment completion was uncommon. We previously reported the recurrence pattern of staged or restaged T3 N0 disease without vocal cord fixation by reviewing 64 cases treated with RT with or without chemotherapy . Of the 64 patients, 22 (34%) underwent RT alone and 42 (66%) underwent CCRT with low-dose or TPF-chemotherapy. Ten patients (16%) underwent total laryngectomy or LP surgery after the delivery of 40 Gy because the tumor showed no regression. The remaining 54 patients received a total RT dose of 60–72 Gy (median, 66 Gy). Eighteen patients (28%) developed recurrence, and all of them were recorded as local failure alone. None of the patients developed regional and distant failure. Similarly, as mentioned earlier, Okumura et al. performed AFRT without ENI for 41 patients with T3 N0 glottic carcinoma without vocal cord fixation . Five patients (12%) developed disease progression. The initial recurrence site was local in three patients (7%) and regional in only two patients (5%). It is essential to maintain a fine balance between treatment intensity for tumor control and avoiding toxicities especially for mucositis in head and neck cancer. From this point of view, intensive local treatment with CCRT while omitting ENI is a reasonable approach for T3 N0 glottic carcinoma without vocal cord fixation.
Previously, surgery was one of the main treatment modalities for T3 glottic carcinoma, and radical RT with or without chemotherapy was performed only for selected patients . Therefore, in the early part of our series, five patients underwent surgery after the delivery of 40 Gy, but more than half of the patients were confirmed to have complete pathological response to the treatment with RT of 40 Gy and one cycle of TPF. Considering the excellent LC in our study, complete cure can be obtained with radical TPF-CCRT even if patients have no tumor regression at the time of interim assessment.
Our study has some limitations. First, this was a retrospective study involving a relatively small number of patients. Second, five patients underwent surgery after the delivery of 40 Gy and one cycle of chemotherapy. The potential bias of surgical intervention might influence our conclusions.