Intensity-modulated radiotherapy with more than 60 Gy improved the survival of inoperable patients with locally advanced esophageal squamous cell carcinoma: a population-based real-world study

Intensity-modulated radiotherapy (IMRT) is widely applied during the treatment of esophageal squamous cell carcinoma (ESCC), but the optimal radiation dose still lacks a consensus. The aim of this study was to explore the optimal radiation dose for inoperable locally advanced ESCC patients treated with IMRT in a real-world clinical setting. A total of 90 inoperable ESCC patients with locally advanced stages of (cid:0) -IVA treated with IMRT in our institute between February 1, 2014 and June 30, 2019 were included in this retrospective study. Sixty patients had received > 60 Gy (high dose group) and 30 patients had received ≤ 60 Gy (low dose group). The median radiation dose was 66 Gy (range: 61–70 Gy) and 50.2 Gy (range: 40–60 Gy), respectively. Concurrent chemotherapies were platinum-based regimens. (> 60 Gy) resulted in better survival outcomes for inoperable patients with locally advanced ESCC treated with IMRT.


Introduction
Esophageal cancer (EC) is the sixth most common cause of cancer death in the world [1,2]. In Eastern Europe and Asia, the main types of pathology in EC patients are esophageal squamous cell carcinoma (ESCC) [3]. Most ESCC patients are at advanced stages when diagnosed, resulting in poor life quality. National Comprehensive Cancer Network (NCCN) recommend chemoradiotherapy (CRT) as the standard treatment for locally advanced EC patients with the radiation dose of 50-50.4 Gy [4]. This recommendation is based on the RTOG9405 prospective randomized clinical trial results. In this prospective phase III randomized controlled trial [5], 109 patients were included in the high ( uorouracil + cisplatin + 64.8 Gy) and low dose groups ( uorouracil + cisplatin + 50.4 Gy). There was no signi cant difference in the median overall survival (OS) (13.0 vs 18.1 months), 2-year OS rate (31% vs 40%) or local control rate (LCR) (56% vs 52%) between the 2 groups. Therefore, lower radiation dose at 50-50.4 Gy was recommended.
Although the RTOG9405 study suggested no signi cant advantage in the high dose group, it was based on 2-dimensional radiotherapy (2DRT). Since the LCR remained low in EC, there are increasing debate on the optimal radiation dose. For instance, Zhang et al. [6] reported that > 51 Gy (high dose) had signi cantly better LCR than ≤ 51 Gy (lower dose) in EC patients treated with 2DRT or 3-dimensional conformal radiotherapy (3DCRT) (P = 0.01). However, whether higher dose would be more favorable was not clear. Hurmuzlu et al. [7] retrospectively analysed 46 EC patients at stages IIA-III, and Loading [MathJax]/jax/output/CommonHTML/fonts/TeX/fontdata.js found that high dose (66 Gy)-related toxicities were signi cantly increased and the survival rates were not improved compared with the low dose.
As IMRT delivers higher dose within the tumor and protect the critical organs around the tumor better, it is widely used to treat EC and improves e cacy. Lin et al. [8] conducted a study with 676 nonrandomized EC patients to estimate the survival effects of 3DCRT and IMRT. The results suggested signi cantly lower risk of dying, lower risk of cardiac death, higher rates of OS, higher rates of locoregional control (LRC) after IMRT than 3DCRT.Whether higher dose delivered by IMRT could improve clinical outcomes reemerge as an important question in esophageal cancer treatment. In such setting, Chang et al. [9] rstly retrospectively compared radiation dose ≥ 60Gy versus < 60 Gy in 2061 thoracic esophageal squamous cell carcinoma (TESCC) treated with IMRT. The 2-year OS rate of the high dose group (≥ 60 Gy) was signi cantly higher than that in the low dose group (< 60Gy) (35.47% vs 26.74%,P < 0.0001). However, Chang et al. [9] only included patients at relatively earlier stages of A-C. The cervical EC was not investigated in this study. Moreover, the database used for this study failed to provide detailed information of the patients, such as tumor length, clinical N stage and clinical T stage, as well as progression free survival (PFS), LRC and distant metastasis free survival (DMFS).
In our institution, de nitive radiotherapy with or without chemotherapy has long been the preferred approach for the cervical EC patients. In this retrospective study, we initially explored the optimal radiation dose for ESCC patients at locally advanced stages of -IVA with IMRT, and provided detailed information of clinicopathological, OS, PFS, LRC and DMFS.
This nding prompted us that higher radiation dose > 60 Gy would be necessary for inoperable locally advanced ESCC patients treated with IMRT in a real-world clinical setting.

Patient selection and pre-treatment evaluation
The study ow diagram is shown in Fig. 1. A total of 90 patients with pathologically con rmed inoperable ESCC and without known metastases were recruited from the Department of Radiation and Medical Oncology, Zhongnan Hospital of Wuhan University from February 1, 2014 to June 30, 2019. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study protocol was approved by the Ethics Committee of Zhongnan hospital of Wuhan University (2020105-1), and the requirement for informed consent was waived because of the retrospective nature of the research. Cancer stages were determined based on the AJCC/UICC 8th edition [10]. The inclusion criteria were as follows: (a) patients with pathologically con rmed ESCC; (b) inoperable patients treated with radiotherapy or CRT using IMRT technique; (c) patients with complete data of clinicopathological information, radiation dose, and serum hemoglobin levels; (d) patient without clinical evidence of distant or retroperitoneal lymph node metastasis; (e) patient without coexisting malignancies.
Pre-treatment evaluation included a medical history collection and physical examination, focusing on performance status and a history of dysphagia. Laboratory studies included a complete blood cell count and blood chemistries. Barium swallow, chest computed tomography (CT) and transesophageal endoscopic ultrasonography were performed to assess the clinical T and N stages. Positron emission tomography (PET)-CT, bone and abdomen CT, and brain magnetic resonance imaging (MRI) were performed to evaluate distant and retroperitoneal lymph node metastasis prior to treatment.

Treatment approaches
Patients were treated 5 days per week at 1.8-2.0 Gy/fraction, one fraction/day. The total radiation dose ranged from 40 to 70 Gy (median: 64 Gy). The gross target volumes (GTV) were delineated based on CT results, including gross tumor volumes (GTVt) and gross nodal tumor volumes (GTVn). The clinical target volumes (CTV) consisted of the CTVn and CTVt. The CTVt was de ned by a 0.5-to 1-cm radial margin expansion and a 3-to 4-cm proximal and distal margin expansion around the GTVt. The CTVn was de ned by a 0.5-to 1-cm expansion around the GTVn. The CTV should not Loading [MathJax]/jax/output/CommonHTML/fonts/TeX/fontdata.js cover normal tissues and organs at risk, such as the spinal cord and vertebral body, and minimize the dose to the heart and lungs. The planning target volume (PTV) was the CTV plus a uniform 0.5-cm expansion margin. For both the low (≤ 60 Gy) and high dose (> 60 Gy) groups, prescribed dose was given to the PTV. Concurrent chemotherapies were platinum-based regimens.

Follow-up and evaluation
Follow-up was conducted by outpatient review, inpatient review and telephone contact. The last follow-up time was December 17, 2019, and the median follow-up time was 14.1 months (range: 2.2-64.4 months). Follow-up examinations were performed every 3 months in the rst 2 years, every 6 months in years 3-5, and annually thereafter. Tumor response and nodal disease were evaluated with repeated CT,barium swallow, and endoscopy. MRI or PET-CT was also performed if clinically necessary. Treatment-related toxicities were graded using National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE, version 4.0). LRC was de ned as the period from the date of diagnosis to the date of the rst evidence of locoregional disease progression or recurrence. DMFS was de ned as the period from the date of diagnosis to the date of the rst evidence of distant metastasis. PFS was de ned as the period from the date of diagnosis to the date of any treatment failure (including distant metastasis, locoregional disease progression or recurrence) or death from any cause. OS was de ned as time from diagnosis to death from any cause.

Statistical analysis
The Pearson's chi-square test was used to compare categorical variables. Rank-Sum test was used to compare continuous variables without Gaussian distribution. T test was used to compare continuous variables with Gaussian distribution. The Kaplan-Meier method with log-rank test was used to analyse survival outcomes between groups. The optimal cut-off was de ned as the hemoglobin value with the smallest P-value of log-rank tests. Univariable and multivariable Cox regression analyses were performed to explore prognostic factors. All statistical tests were two-sided with a P < 0.05 considered statistically signi cant. The analyses were conducted using IBM SPSS statistics software version 25.0 and GraphPad Prism 6.

Patient characteristics
A total of 90 ESCC patients at Zhongnan Hospital of Wuhan University between February 1, 2014 and June 30, 2019 were included in this retrospective study. Among the patients, 30 received ≤ 60 Gy radiation doses (low dose), and the other 60 received > 60 Gy radiation doses (high dose). The median radiation dose was 66 Gy (range: 61-70 Gy), and the median fraction size was 2 Gy (range: 1.8-2.0 Gy) in the high dose group. The median radiation dose was 50.2 Gy (range: 40-60 Gy), and the median fraction size was 2 Gy (range: 1.8-2.0 Gy) in the low dose group. Patients' clinicopathological characteristics, disease information and treatment pro les were shown in Table 1. No statistically signi cant difference was found between the 2 groups in age, gender, tumor location, tumor length, clinical N stage, clinical TNM stage, hemoglobin or treatment regimens (P > 0.05). A larger proportion of patients in the high dose group had cT3 (P = 0.004), but the clinical TNM stage had no statistically signi cant difference between the two groups (P = 0.139).  (Fig. 2).
We performed Log-rank comparisons between groups to investigate the impacts of radiation doses on OS, PFS, DMFS and LRC (Fig. 3)

Discussion
In this retrospective study including 90 inoperable ESCC patients receiving IMRT, we found higher radiation doses brought signi cantly better PFS and OS than lower radiation doses (≤ 60 Gy). In addition, a persistent trend toward better LRC and Loading [MathJax]/jax/output/CommonHTML/fonts/TeX/fontdata.js DMFS in the high dose group was also observed. Meanwhile, no additional grade ≥ 3 treatment-related toxicities were present in the high dose group. These results suggested that IMRT at a radiation dose > 60 Gy would be necessary and safe for inoperable patients with locally advanced ESCC. Our work contributed to explore the optimal dose of IMRT for ESCC patients.
The RTOG8501 study [11] established concurrent CRT as the standard therapeutic strategy for EC patients. Shortly afterwards, the RTOG9405 study [5] identi ed an optimal dose of concurrent CRT at 50.4 Gy for EC patients. However, there is still a lack of consensus on the optimal radiotherapy dose for locally advanced EC. First, although there were more deaths in the high dose group than the low dose group (11 vs 2), 7 of the 11 patients in the higher dose group died before the radiotherapy dose reached 50.4 Gy. Therefore, higher risk of death might not result from the higher radiation doses [5].
Second, more than 60% EC patients in the RTOG9405 trial were at early clinical stages. Third, higher distant metastasis rate might result from the higher proportion of stage III patients in the high dose group. Fourth, both squamous cell carcinoma (85%) and adenocarcinoma (15%) were included in the RTOG9405 study, which might have different optimal radiation dose since ESCC was more sensitive to radiotherapy. Fifth, the lower uorouracil dose in the high dose group of RTOG9405 trial might impact the prognosis. Finally, patients received conventional rather than modern radiotherapy techniques in the RTOG9405 trial. The radiotherapy technology has been improved over the last decades, and the recommended radiation dose should be updated accordingly.
Our study aimed to investigate the e cacy of IMRT at the high dose (> 60 Gy) compared with the low dose (≤ 60 Gy) for inoperable ESCC patients with advanced clinical stages ( -A). The OS and PFS of patients were better in the high dose group than the low dose group in our study (P < 0.05). Previous studies also indicated that increased radiotherapy dose improved the therapeutic effects of CRT on EC patients, as shown in Table 4 [6, 9, 12-16]. The higher radiation doses resulted in signi cantly better OS (P < 0.05) [9,12,15,16]. In our study, we also found that the higher radiation doses increased OS rates compared with the lower doses (1-year OS 57.5% vs 39.5%; 2-year OS 31.4% vs 15.8%, P = 0.007). However, Suh et al. [13] reported that higher doses (≥ 60 Gy) had higher 2-year LCR (69% vs 32%, P < 0.01) and 2-year PFS rate (47% vs 20%, P = 0.01). The median OS of the high and low dose groups were 28 and 18 months, respectively (P = 0.26). Zhang et al. [6] reported that > 51 Gy had signi cantly better LCR than ≤ 51 Gy in EC patients with clinical stages II or III (P = 0.01). Our study also suggested a persistent trend toward better LRC in the high dose group (P = 0.707). In addition to Chang et al., the other 6 studies did not consider the possible effects of the radiation technique on patients ( Table 2). Our data indicated that the higher radiation doses of IMRT might improve the PFS of inoperable patients with locally advanced ESCC. This nding complements previous studies reported by Chang et al. which failed to provide detailed information of the patients, as well as progression free survival (PFS).  [19][20][21][22]. In our study, the optimal cut-off was de ned as the hemoglobin value with the smallest P-value of log-rank tests. Patients with hemoglobin levels < 132.1 g/L should be concerned and the patients' hemoglobin levels should be raised before treatment. Our data identi ed critically prognostic factors in inoperable patient with locally advanced EC with IMRT. Additional studies are still required for validation.
IMRT becomes increasingly popular since it improves target conformality and decreases treatment-related toxicity [23].
Other studies also con rmed that IMRT decreased the radiation doses to protect the normal tissues, such as lungs, heart and thyroid [9,[24][25][26][27]. In our study, no patient died of treatment-related toxicity. No signi cant difference between the high and the low dose groups on treatment-related toxicities of grade ≥ 3, including hematologic toxicity, radiation esophagitis, radiodermatitis and stula (P = 0.402).
It should be noted that this study had several limitations. First, it was a retrospective study in a single institution, which inevitably resulted in a selection bias and treatment heterogeneity. Second, the number of patients included in this study is relatively small. In the future, a large-scale randomized prospective trial is required to further con rm the conclusion.

Conclusion
Higher radiation dose (> 60 Gy) of IMRT performed better survival outcomes for inoperable patients with locally advanced ESCC.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study protocol was approved by the Ethics Committee of Zhongnan hospital of Wuhan University (2020105-1), and the requirement for informed consent was waived because of the retrospective nature of the research.

Consent for publication
Not applicable.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.