Elective lymph node irradiation late course accelerated hyper-fractionated radiotherapy plus concurrent cisplatin-based chemotherapy for esophageal squamous cell carcinoma: a phase II study

Based on the results of the Radiation Therapy Oncology Group (RTOG) phase III intergroup trial 85–01 and 95–04, the standard therapy for patients with localized esophagus carcinoma selected for nonsurgical treatment is radiation therapy (RT) plus concurrent chemotherapy [1, 2]. In RTOG 85–01, combined therapy significantly increased 5-year overall survival (OS) to 26% (95% confidence interval [CI], 15%–37%) compared with RT alone. However, the incidence of local/regional failure and local/regional persistence of disease was high up to 45.9% [1]. In an attempt to improve these results, RTOG 95–04 increased radiation dose from 50.0 Gy to 64.8 Gy, however, intensification of the radiation dose did not improve local/regional control or survival [2]. Although the reason for the lack of benefit using higher dose is unclear, a significant prolongation of treatment time may have contributed, in part, to this result [2].

Several animal experiments and clinical investigations have shown that accelerated proliferation of surviving tumor clonogen during a standard schedule of RT is one of the major reasons for treatment failure [3, 4]. It is postulated theoretically that a shortened irradiation course, still keeping the total radiation dose, or increased dose of radiation delivered in the late course of the treatment would improve local control for esophageal carcinoma by overcoming the deleterious effects of accelerated repopulation. Clinical data from China [5] has shown late course accelerated hyperfractionated radiotherapy (LCAHRT) improved the 5-year OS (odds ratio [OR] = 2.93, 95% CI: 2.15–4.00, p < 0.00001) and 5-year local control (OR = 3.96, 95% CI: 2.91–5.38, p < 0.00001) than standard fractionated RT in the localized esophageal carcinoma. Clinical investigation in nasopharyngeal [6, 7] and lung carcinoma [8] also displayed promising treatment outcome.

For esophageal squamous cell carcinoma (ESCC), lymph nodal failure remains a major reason for poor prognosis. Since the early 1980s, Japanese surgeons have practiced 3-field regional lymph node dissection for esophageal cancer and led to an improved survival [9, 10]. It is thought that prophylactic 3-field lymph node dissection improves the survival rate by eliminating micrometastases and reducing the regional lymph node recurrence rate. In accordance with the concept of 3-field lymph node dissection in curative surgery, elective nodal irradiation (ENI) has been adopted for definitive chemoradiotherapy in Japan. Although, the benefit of ENI for ESCC remains controversial, recently studies from Japan [11, 12] have confirmed ENI was effective for preventing regional and distant nodal failure in patients with esophageal carcinoma undergoing concurrent chemoradiation.

In randomized clinical trials, no consistent benefit was seen for any specific chemotherapy regimens for locally advanced ESCC. Cisplatin is one of the most active agents, with a single agent response rate consistently in the range of 20% or greater. Older agent 5-fluorouracil and newer agents such as capecitabine, pemetrexed, were reported effectiveness for locally advanced ESCC [13].

Based on above thinking, our institution had engaged in clinical trials of ENI LCAHRT concurrently with cisplatin-based chemotherapy (CHT) for ESCC since 2004. The results of our phase I studies [14, 15] demonstrated that this treatment scheme was feasible and tolerated, and the treatment outcomes were encouraging. The results with longer follow-up are now reported in the present study.

ESCC remains one of the most lethal carcinomas in China. The poor survival is mainly related to the advanced stages at diagnosis. The optimal management for locally advanced disease remains controversial. Surgery or RT alone is associated with a poor survival, usually less than 10% at 5 years [16]. Randomized studies [1, 2] have demonstrated the survival advantage of concurrent chemoradiotherapy compared with RT alone for patients with squamous cell or adenocarcinoma of the esophagus. Although concurrently administered chemotherapy acts as a promoter of the locoregional effects of RT, the locoregional control remains unsatisfactory in RTOG 85–01 trail [1]. Dose escalation trial by RTOG 94–05 failed to improve the locoregional control as well [2]. It may thus be concluded that by simply increasing the local radiation dose cannot easily convert to survival benefits for esophageal carcinoma patients. Therefore, many efforts have been made to explore the more reasonable treatment schedules based on this thinking and understanding to radiobiology.

From last two decades, a further increase in local-regional tumor control may be expected by augmenting radiation effects by altering fractionation schedule, because rapid tumor clonogen repopulation during treatment seems to be a cause of poor prognosis for cancer patients. One phase II study from Japan showed a promising result for ESCC patients treated with accelerated hyperfractionated RT plus 5-fluorouracil/cisplatin chemotherapy [17]. In China, Shi and their teams [18] initiated the study of LCAHRT for ESCC treatment, and yielded very encouraging results. Comparing with conventional fractionation (CF), the 5-year survival (34% versus 15%) and local control (55% versus 21%) were markedly improved with the LCAHRT regime. Henceforth, unremitting results from randomized and retrospective trials came out [19‐23]. Recently, three independent meta-analysis from China strengthened the evidence of therapeutically beneficial of LCAHRT compared with CF for esophageal carcinoma [5, 24, 25].

Because of unique geographical and pathologic features of esophageal carcinomas in China, furthermore, esophageal carcinomas tend to spread axially to regional lymphatics, producing morbidity and mortality from locoregional effects [1]. Based on the previous studies [9, 10], we supposed that elective HRLNR irradiation might be associated with lower nodal failure in advanced stage ESCC, and this hypothesis was supported by recent studies from Asian [11, 12, 26]. From another point of view, considering of high incidence of nodal failure in combined modality therapy, chemotherapy seems to play a limited role in preventing locoregional or distant nodal failure. Omitting ENI may be not reasonable for locally advanced stage ESCC patients, especially if the primary tumor located in lymphatic-rich regions [27] and PET/CT could not be incorporated in the planning of radiotherapy [28].

Previous study designed by Zhao et al. [22] enrolled 54 cases of local advanced stage ESCC with clinical stage T_1-4N_0-1M₀ treated with LCAHRT (41.4Gy/23 fractions followed by 27Gy/18 fractions with 1.5Gy per fraction, twice a day) and 2 cycles of concurrent chemotherapy with cisplatin (25 mg/m²/day days 1–3) plus 5-FU (600 mg/m²/day days 1–3). And they reported the median survival was 30.8 months (95%CI, 17.6–44.1 months), Survival rates at 1, 3, and 5 years were 67%, 44%, and 40%, respectively. Twenty-six percent (14/54) of patients had locoregional disease presenting as the first failure, and distant metastases as the first failure occurred in 24% (13/54). In our study, 40Gy/20 fractions delivered to primary and metastatic regional tumor(s) and HRLNR, and additional 19.6Gy/14 fractions in 1.4Gy per fraction, twice a day, were given to the boost volume. Cisplatin-based chemotherapy concurrently administered during the RT course. Our regimen improved median survival time from 30.8 to 34.4 months, and the OS at 1-, 3-year improved from 67%, 44% to 75.5%, 46.5%, respectively. However, this trend was not observed at 5-year (40% vs. 22.7%). It should be noticed that approximately 32.3% of the population consisted of stage IV_a cases, and 46% of eligible patients were enrolled between 2010 and 2011 (Figure 1) in our study, which may partially explain the relatively poor 5-year OS. By subgroup analysis, for patients with stage II–III, the 5-year OS reached to 39.9%, which was similar to the results of Zhao et al. [22]. Despite the study cohort consisted of a high proportion of patients with stage IV_a, this regimen was associated with only twenty-nine percent of local or/and regional failure and nineteen percent of distant metastases. Previous studies (Table 5) focusing on LCAHRT regimen reported the patterns of failure from local/locoregional failure and distant metastasis were 12.5–41.8% and 15.4–34.9%, respectively [18‐23] in patients with TNM stage limited to I–III.

Regarding tolerance in our study, high-grade (≥3) acute esophagitis and leucopenia were seen in 26.4% and 32.4% of patients, respectively. Similar findings were reported by Zhao et al. [22]. The higher rates of acute toxicities may be related to this treatment scheme. Our approach allowed us to avoid potential micrometastasis of lymphatics, nevertheless, it was gained at the expense of early toxicities. Therefore, how to minimize the damage to the sensitive normal tissues within irradiated fields without sparing cancer is a big challenge. Recent innovation of RT technology has made it possible to use sophisticated rotational intensity-modulated RT. This can deliver intensified radiation doses to the tumor while minimizing the doses to the normal tissues [29]. Acceleration of radiation treatment involves the delivery of the target dose in less time, and is analogous to the concept of dose-intensity in the delivery of cytotoxic chemotherapy. In order to prevent late toxicity, we reduced the dosage of per fraction (1.4Gy per fraction twice day). The interval between daily fractions (typically ≥ 6 hours) provides normal tissues with time to repair sublethal radiation damage, which may be responsible for the lower rate of late lung toxicity in our study. Notably, treatment-related upper-gastrointestinal hemorrhage (4.4%) and esophagus-tracheal fistula (2.9%) were relatively higher, which should be paid much attention. Although the reason for the higher incidence of such complications was unknown, we suspected that it might be association with advanced T stage in our cohort of patients.

Several strengths and limitations should be noted. This was a single center study with a relatively small sample size, which may limit the generalizability of our findings. This study cohort consisted of a higher proportion of stage IV_a who had worse prognosis. Moreover, a variety of cisplatin-based chemotherapy regimens were enrolled, although OS rate was not associated with combined regimens. The chemotherapy agents are usually 5-fluorouracil and cisplatin. Recently, capecitabine and pemetrexed have been introduced in phase I and II trials and the preliminary results are promising [13]. Therefore, how to best combine chemotherapy and accelerated radiotherapy regimens to maximize local control and survival is an important question. Further trials and observation confirming the efficiency of LCAHRT concurrently with these chemotherapy regimens are still needed. At last, this accelerated radiation scheme is broadly used as a standard treatment for locally advanced ESCC and non-small cell lung cancer in China, however, has not met with widespread clinical acceptance in the United States or Europe probably because of higher acute toxicity, and the use of twice fractions per day.

In conclusion, this phase II study with limited number of patients demonstrated that ENI LCAHRT concurrently with CHT was generally tolerated, and the treatment outcome was satisfactory. Although these results by this interim analysis were not sufficient to confirm the impact on ESCC treatment with this regimen, the results in our cohort patients with a higher proportion of advanced stage were compared favorably with those of previous studies. Further observation with longer time follow up and randomized phase III trial is currently underway.