br At registration patients were randomly allocated to one
At registration, patients were randomly allocated (1:1:1) to one of the three arms through a sealed envelope system and were una-ware of the treatment given. In arm A, patients received EGCG solution orally from the start of radiotherapy. Patients in arm B or C were given oral EGCG or mLDG solution respectively, and began medications as soon as grade I esophagitis occurred during radiation. The grading definition of ARIE in the Radiation Therapy Oncology Group (RTOG) scoring system was also listed in Supple-mentary Table S1 . Three groups of patients all discontinued oral liquid preparations until two weeks after the completion of radiotherapy. Fig. 1 showed an overview of the study design.
RT was not interrupted unless persistent or worsening dyspha-gia was present after therapy. Other treatments were not per-formed until grade 4 esophagitis occurrence or radiotherapy suspension. In case of unresponsive to therapy and RT interruption, patients were supported with methylprednisolone, analgesics, antifungal therapy, or intravenous infusion as appropriate until recovery. The nasogastric or nasoduodenal tube was an option for non-responders with grade 4 toxicity lasting for at least 3 days.
From the beginning of the FLAG tag Peptide treatment to two weeks after the end of radiotherapy, esophageal toxicity was measured every week according to the RTOG scoring system. Symptoms
Fig. 1. An overview of the study design.
188 EGCG in lung cancer
related to esophagitis were also measured using the numerical rat-ing scale (NRS) weekly, including pain and dysphagia.
Esophagitis index (EI) was referred to assess therapeutic effi-cacy [14,15]. Esophagitis grades for each observation time point were plotted on a graph against time. Due to the different duration of individual treatment, we took a Leading strand conversion method of one-percentage-mark system. The area under the curve was calculated for each patient’s graph using the trapezoidal method and was defined as an adjust esophagitis index (AEI) for an example in Fig. 2. The AEI was designed to account for not only the maximum grade, but also the duration of esophagitis. The same statistical method was also used to calculate the adjusted pain index (API) and the dysphagia index (ADI) based on the NRS scores for pain and dysphagia.
A medical team consisting of two experienced physicians, una-ware of the patient’s clinical history and treatment allocation, con-ducted the routine esophagitis assessment to avoid subjective bias.
The primary endpoint was to evaluate the relative efficacy of EGCG to improve esophagitis grade over time during lung cancer radiotherapy compared to routine regimens. Secondary endpoints included maximum esophagitis score, API and ADI.
Based on our previous experience and other documentation, the expected number of patients with AEI was estimated [15–17]. The study compared the treatment groups (arm A or B) with the control arm (arm C). Assuming a mean AEI of 5 (standard deviation = 4) in treatment groups and 10 (standard deviation = 10.5) in control arm , the required sample size would be 25 patients in each arm with 80% statistical power and a two-sided 5% type 1 error rate. Assuming a 10% attrition rate, a total of 83 patients would be required.
Measurement data of the different groups were expressed as mean ± standard deviation and analyzed by t-test. Pearson’s exact tests were used for categorical variables. A value of P < 0.05 was considered statistically significant. All statistical analyses were performed with SPSS software (version 17.0; SPSS Inc., Chicago, IL).