Radiotherapy is one of the most important treatments for NSCLC, and radiation dose levels significantly affect treatment outcomes. Fletcher  believed that a radiotherapy dose of 80–100 Gy was required to cure lung cancer. This assume has been confirmed through a clinical trial of stereotactic body radiotherapy, which indicated that significantly higher survival rates were observed among patients who were treated with a BED of at least 100 Gy than among patients who were treated with a BED of less than 100 Gy . Relative to traditional radiotherapy approaches, the techniques of three-dimensional conformal radiotherapy and intensity-modulated radiotherapy can significantly reduce the radiation doses that are received by normal tissues and organs; in the context of radiotherapy alone (without accompanying chemotherapy), these techniques allow tolerable toxicity to be achieved even at radiation doses as high as 83.3 Gy or 103 Gy [26, 27]. NSCLC cells demonstrate accelerated repopulation during radiotherapy, and increases in treatment duration can therefore reduce local control, negatively impacting survival. Thus, to improve local control it is critical to achieve higher BEDs at constant or reduced overall treatment durations [9, 28]. However, hyperfractionated radiotherapy not only produces more severe acute reactions than conventional radiotherapy but also requires three treatments per day, producing inconvenience and much more treatment costs for patients; thus, it is not easy to clinically utilise hyperfractionated radiotherapies [12, 13].
Mehta et al.  suggested that for cancers with very short cell doubling times, such as NSCLC, increasing the dose per fraction constitutes a promising method of achieving dose escalation. In cases of stage I/II NSCLC, hypofractionated stereotactic radiotherapy has achieved effects comparable to the results of surgery; in these cases, the radiotherapy regimen includes extremely high doses of up to 18–22 Gy per fraction [29, 30]. However, in cases of locally advanced NSCLC, because of concerns regarding toxicities (particularly late toxicities) to the oesophagus, trachea, ribs, nerves, and other vital organs, it has not been possible to achieve these high doses per fraction . In fact, the appropriate dose for hypofractionated radiotherapy for locally advanced NSCLC remains unclear .
Thirion et al.  conducted a study of accelerated hypofractionation using 3 Gy per fraction in which a 72 Gy radiation dose was delivered in 24 fractions over five weeks. The 25 NSCLC cases that Thirion et al. examined included 16 cases of stage III NSCLC. In the investigation conducted by Thirion et al., one case experienced grade 3 radiation pneumonitis, two experienced grade 3 radiation oesophagitis, and no grade 4 or higher adverse events occurred among the entire group. Xie et al.  performed a dose escalation trial of accelerated hypofractionated radiotherapy with 3 Gy increments using cases from the Chinese population and found that patients could tolerate the radiotherapy treatment at doses of up to 75 Gy at V20 levels of no greater than 20% or at doses of up to 69 Gy at V20 levels of between 20% and 30%. The Fudan University Cancer Centre reported the results of a Phase II trial of accelerated hypofractionated radiotherapy with sequential chemotherapy for NSCLC . In this trial, the initial radiation dose was 50 Gy with 2.5 Gy fractions; the dose was subsequently increased at 3 Gy/fraction to a total radiation dose of 65 to 68 Gy, and elective nodal irradiation was not performed. All patients received 2 cycles of induction chemotherapy (with full-dose NVB and cisplatin), and good treatment efficacy was observed. The median PFS, median OS, and three-year OS were 10 months, 19.0 months, and 32.1%, respectively, with acceptable treatment toxicities.
Conventional fractionated radiotherapy with concurrent chemotherapy achieves superior efficacy to sequential chemoradiotherapy; moreover, experimental studies have demonstrated that hypofractionated radiotherapy combined with chemotherapy can achieve significantly increased biological effects . Therefore, compared with these treatments, accelerated hypofractionated radiotherapy combined with concurrent chemotherapy should theoretically be capable of producing additional increases in therapeutic efficacy. However, due to concerns that hypofractionated radiotherapy with concurrent chemotherapy could have aggressive toxicity, little research has been conducted in this area. In addition, investigations of this topic have utilised very different fractionation regimens and concurrent chemotherapy drugs. In particular, the studied doses per fraction have ranged from 2.4 Gy to 2.75 Gy, and the concurrent chemotherapy drugs have included low daily doses of cisplatin, weekly paclitaxel/CBP, and full-dose NVB/cisplatin [19, 34–36].
An EORTC study reported dose escalation results from hypofractionation radiotherapy with 2.75 Gy fractions for NSCLC and suggested that radiotherapy with a total dose of 66 Gy in 24 fractions in combination with concurrent low-dose cisplatin could be well tolerated . Based on these results, a phase III randomised study was conducted to compare concurrent chemoradiotherapy with sequential chemoradiotherapy . Because this study administered elective nodal irradiation, high rates of severe acute oesophagitis were observed among the concurrent group; in particular, grade 3 and grade 4 acute oesophagitis was observed in 14% and 3%, respectively, of the patients in concurrent group, whereas significantly lower rates of acute oesophagitis were observed in the sequential group. Interestingly, rates of late oesophageal injury were not significantly higher in the concurrent group (5%) than in the sequential group (4%). The OS did not differ between the two groups; in particular, the concurrent group and the sequential group exhibited 2-year OS of 39% and 34%, respectively, and three-year OS of 34% and 22%, respectively. This lack of improvement in OS with the concurrent therapy compared to the sequential therapy may relate to the low doses of chemotherapy that were administered in this study. A 2011 report of the American Society of Clinical Oncology (ASCO) discussed the preliminary results of a British study of radiotherapy with concurrent two-drug chemotherapy for NSCLC, using a total radiation dose of 55 Gy in 20 fractions with 2.75 Gy/fraction; in particular, this treatment produced a median survival time (MST) of 27.4 months, a 2-year OS of 54%, and a 3-year OS of 38% . In this study, full-dose of chemotherapy drugs were administered in a regimen that included 15 mg/m2 NVB on d1, d6, d15, and d20 as well as 20 mg/m2 cisplatin on d1-4 and d16-19 . A phase II trial by the Korean Radiation Oncology Group also achieved good survival results, including an MST of 28.1 months and 2- and 3-year OS of 56.4% and 43.8%, respectively . This study used a radiotherapy regimen consisting of a total radiation dose of 60 Gy in 25 fractions of 2.4 Gy/fraction, which was combined with concurrent paclitaxel and CBP chemotherapy. Elective nodal irradiation was not performed, and tolerable toxicity was observed.
The aforementioned three regimens of hypofractionated radiotherapy with concurrent chemotherapy involved significant differences in radiotherapy fractionation, total radiation dose, and concurrent chemotherapy protocols. In addition, distinct inclusion criteria were used in these studies, and different restrictions on the radiation dose received by normal tissues and organs were employed. Thus, it is difficult to horizontally compare the advantages and disadvantages of these prior investigations to identify a regimen that could receive widespread acceptance. However, it is evident that the efficacy of hypofractionated radiotherapy with concurrent chemotherapy exhibits good comparability with the efficacy of conventional fractionation with concurrent chemotherapy [3–5].
In our trial, we chose 3 Gy as the dose per fraction because the use of hypofractionated radiotherapy at 3 Gy per fraction was a relatively mature approach for radiation treatment alone; in addition, the chosen radiotherapy regimen was efficient and was expected to be completed in a total treatment time of five weeks. Thus, this regimen could achieve a high rate of tumour control . This phase I trial revealed that hypofractionated radiotherapy at 3 Gy/fraction with concurrent full-dose chemotherapy exhibited high levels of safety. No treatment-related deaths occurred during the entire study. The main toxicities that were observed included radiation oesophagitis and radiation pneumonitis. The rate of radiation oesophagitis reached 84.6% (11/13) in whole group. Because we utilised the involved-field irradiation technique and refrained from administering elective nodal irradiation, the observed cases of radiation oesophagitis were mainly of grade 1 or grade 2 (69.2%, 9/13). The adverse events to the radiotherapy were tolerable after general symptomatic treatment and did not affect the implementation of the radiotherapy regimen. The 2 observed cases of grade 3 radiation oesophagitis occurred at the 72 Gy dose level. One case of grade 3 radiation pneumonitis occurred in the 72 Gy group, whereas the other observed cases of radiation pneumonitis in this study were all grade 1 or grade 2. Other non-haematologic toxicities were easily managed through clinical treatment and did not affect the implementation of concurrent chemoradiotherapy. Although all patients suffered from neutropenia, most of the cases of these adverse events were mild and did not affect the implementation of concurrent chemoradiotherapy. Only one case of grade 4 neutropenia occurred in this study, and this case arose in the 72 Gy group. In this case, after the administration of treatments to prevent infection and GSF to enhance white blood cell counts, agranulocytosis was resolved within one week, and no serious consequences were observed. Mild degrees of anaemia and thrombocytopenia were observed in the whole group. The short-term tumour control rate of 84.6% indicated that our regimen of hypofractionated radiotherapy with concurrent chemotherapy was able to achieve good local tumour control. Because of the short follow-up time and the small number of examined cases, mature OS data for this study are not yet available; however, the 12-month PFS of this investigation is highly comparable to the 12-month PFS results from other studies .