Radiotherapy for small cell carcinoma of the esophagus: outcomes and prognostic factors from a retrospective study

Small-cell carcinoma is one kind of highly malignant neoplasm that usually originates from the lung. The first case of primary small cell carcinoma of the esophagus (SCCE) was reported by Mckeown in 1952 [1]. SCCE is characterized by its aggressive feature with poor prognosis and is distinct from the squamous cell carcinoma or adenocarcinoma of the esophagus, but similar to small cell carcinoma arising from lungs or other organs. A consensus on the standard therapeutic modality for SCCE has not been established, but based on retrospective studies, a multimodal therapeutic approach combining surgery, radiotherapy (RT), and chemotherapy is recommended [2, 3]. When combined with chemotherapy, RT is considered a major local therapy approach for treating patients with localized or locoregional disease. Though RT is common in clinical practice in western countries, esophagectomy is more widely used compared to RT as the treatment for localized SCCE in China [2, 4]. A previous retrospective study of all the SCCE cases in our center from September 1990 to June 2011 showed that of 64 SCCE patients, only 15 (26.7%) patients received RT [5]. Due to increasing evidence showing its non-inferiority to esophagectomy, RT technology was developed over the years in China for treating SCCE. In this retrospective study, we focused on SCCE cases who had received RT and explored their outcomes and prognostic factors.

Several studies have shown that SCCE only accounts for 0.4–2.8% of primary esophageal carcinoma [6, 7]. Reports of increasing incidence of SCCE have come in from East Asian countries, including China and Japan, reaching global trends [8, 9]. Indeed, a comparison with our previous findings in 2012, the incidence rate of SCCE in our center rose slightly from 0.94 to 1.15%, perhaps due to the improvement in diagnostic pathology [5]. One study described discrepancies in tumor characteristics and therapeutic modalities in different ethnic groups [2]. For instance, the most common primary location of SCCE was the lower thoracic esophagus in the U. S population, while the Chinese population tends to have middle thoracic SCCE. Localized treatment modalities also differ between these two countries. Esophagectomy was the most popularized local therapy in China, which was performed in ~ 52 to 85% of patients [10, 11], whereas approximately half of the U.S. patients were reported to receive RT as the principal local therapy, regardless of chemotherapy [2]. Nevertheless, more Asian patients were diagnosed with earlier localized disease and the lack of feasibility of RT during the last few decades might explain this difference. We previously reported that 26.7% of patients in our center received RT as one part of the treatment, which increased to 40.9% in the current study, suggesting that RT combined with or without chemotherapy gained more popularity among treating physicians for SCCE recently [5].

The role of RT on the local control of SCCE is not well-illustrated because of the lacking of the information about the response rate in previous reports. In our study, the overall response (CR + PR) rate of 60% was obtained, suggested that RT alone or combined with chemotherapy was an effective option for locoregional disease control. Also, most published studies focus only on OS, but not PFS or DFS, leading to difficulty in evaluating the efficiency of RT in long-term local control. Our result indicated that the high response rate achieved by RT contributes to the survival benefit through long-duration remission. The median OS was 12.9 months in this study which was similar to previous reports (8–16 months) [2, 4]. The reported OS rates at 1-, 3- and 5-years in SCCE patients who received multiple therapeutics varied from 30 to 74.8%, 13.2–38.8%, and 7.8–18% respectively, and the survival data obtained in our study are almost equivalent to the best outcome [10‐13).

As the two main local therapeutic modalities, surgery and definitive RT, especially for LD SCCE, show vast discrepancies when conducted in individuals with different demographics and clinical characteristics. While Sun et al., retrospectively reported 1-, 3- and 5-year OS rates of 50.7, 13.7 and 8.2% in 73 SCCE patients treated by surgery, these rates are lower than findings we reported previously and here [14]. However, in an analysis of the National Cancer Data Base, esophagectomy was associated with the best OS for patients with localized or local-advanced SCCE comparing to chemoradiation or chemotherapy alone. The 1-, 3- and 5-year OS rates reported in another study were higher than those obtained in our study, but the median overall OS of LD-SCCE patients who received surgery was only 18.0 months, which was consistent with other report but shorter than current analysis [15, 16]. Interestingly, another study comparing the RT and chemotherapy (RT + CT) with surgery and chemotherapy (S + CT) in the management of LD SCCE indirectly favored RT, in which a significantly longer OS with RT + CT compared to S + CT (33.0 vs. 17.5 months, p = 0.02) was observed [13]. To sum up, there is a lack of consensus on whether surgery or RT should be the localized treatment modality for SCCE, especially for LD-SCCE, which must be further investigated in prospective controlled studies.

Here we also have shown that RT is also beneficial for the survival of patients with ED SCCE. Similar to the result of the current study, one accumulative analysis of the SEER data demonstrated that the addition of RT leads to a reduction of 30% risks of death in distant stage ESCC. Despite these promising results, the survival benefit of SCCE from RT was much poorer than that from squamous carcinoma or adenocarcinoma of the esophagus, suggesting the need for novel treatment modalities strategy for SCCE [17, 18]. Chemotherapy is one of the most important modalities for SCCE, it was well-acknowledged as the fundamental part of the multimodalities therapy and improved the survival both in LD and ED patients [13, 19]. In our study, the survival of RT + CT group prone to be longer compared to RT alone but without statistically significant difference (p = 0.088) of the whole cohort. In patients with LD, RT + CT is correlated with better survival (p = 0.046). By summary, the combination of RT and chemotherapy will further improve the survival of SCCE patients. Distant metastasis is the major failure pattern of SCCE despite stages. In accordance with this, over half (65%) of patients were failed in the distant control of our study, which further suggested that systemic chemotherapy is essential for the long-term control of SCCE.

The principles of chemotherapy for SCCE is poorly illustrated. According to a study focusing on the SCCE in locoregional disease treated with RT ± chemotherapy, more cycles of chemotherapy improved the survival, while the regiment and sequence of RT and chemotherapy had no impact on the disease control [4]. Unfortunately, due to a relatively small size of patients received chemotherapy in our study, we could not yield a reliable analysis to answer these questions. There is no consensus on the practice of RT neither, especially the technology and dose in SCCE, which are based on routine RT for SCLC. The dose used in this study is similar to the common RT doses ranging from 40-70Gy based on previous case reports and retrospective studies [20‐22]. We found that a higher dose of RT (≥56Gy) is correlated with better survival. However, Jeene et al. reported that no effect of radiotherapy dose on overall survival in the multivariable analysis, while as it was mentioned, the results should have to be interpreted with caution, as only 5 patients received a higher dose of RT in that study. Despite knowing that IMRT is better at target positioning and avoiding errors for esophageal cancer compared to conventional RT [23, 24]. There are no comparative studies of these two RT technologies for SCCE treatment. We did not find significant differences between the OS of patients treated with 3DCRT or IMRT, which might be related to the small sample size of this study.

The main shortcomings of the current study are the small sample size and its retrospective nature. Nevertheless, compared to other studies, we only enrolled the patients who received RT as one part of the treatments and addressed on the local-control role of RT in SCCE by using consistent detailed individual patient data from a single institution. However, the confounding effect caused by enrolling patients with other treatment strategies cannot be overlooked.

Due to its low incidence, prospective and randomized studies on exploring the novel treatments for SCCE is limited. A phase II study (NCT03811379) in our center was initiated in 2018 and is currently recruiting patients to explore the efficacy of toripalimab (PD-1 antibody) as a monotherapy for patients with SCCE, which provides a hint for the novel exploration of a combination of immunotherapy and radiation therapy for SCCE.

In summary, RT was one of the effective and safe treatment options for locoregional control of SCCE. Lower ECOG PS score, shorter lesion length, treated with chemotherapy, and a higher dose of RT were identified as favorable independent prognostic factors.