Abstract
Objective: Esophageal cancer (EC) ranks eighth among cancers in cancer-related deaths globally, and ~44% of new cases occur in China. We sought to describe the clinical characteristics and treatment landscape of EC in China before the approval of immunotherapy in 2020.
Methods: CHANNEL was a large, retrospective study using patient-level data from 14 hospitals/cancer centers across China, including adults initiating therapy for newly diagnosed EC (January to December 2018). Demographics, clinicopathologic characteristics, and treatment patterns over 6 months were descriptively summarized.
Results: Of 3,493 patients, 75.7% were men, the mean age was 64.1 years, and 75.0% had no family history of cancer. Most (92.8%) had squamous cell carcinoma, with a primary lesion in the middle esophagus (56.4%). Among patients with resectable EC, 92.9% received initial surgery, and 7.1% received neoadjuvant therapy, primarily chemotherapy (85.5% platinum-taxane). Among patients with unresectable early or locally advanced EC, 50.8% and 49.2% received palliative and radical therapy, respectively, as the initial treatment, primarily chemotherapy (66.5% platinum-taxane) and chemoradiotherapy (50.8% platinum-taxane), respectively. Adjuvant therapy was administered to 22.9% of patients undergoing initial surgery, and 2.4% receiving neoadjuvant therapy and surgery. Among patients with advanced EC, 84.6% received systemic therapy as an initial treatment, primarily chemotherapy (61.5% platinum-taxane).
Conclusions: Before the approval of immunotherapy in China, most patients with resectable early or locally advanced EC underwent radical surgery without preoperative treatment, whereas most patients with advanced EC received platinum-taxane chemotherapy. These findings highlight the need for novel EC treatments before immunotherapy was introduced, and provide a baseline for evaluating the benefits of immunotherapy, now that this treatment is widely used in this setting.
keywords
Introduction
Esophageal cancer (EC) is the eighth most common cause of cancer-related death globally1. China has the highest incidence of new EC cases of any country, and currently accounts for ~44% of total new EC cases worldwide1. Squamous cell carcinoma is the predominant histological EC subtype in China and Asia (> 90% of cases in China), whereas in Western countries, adenocarcinoma is more common2–5. EC is observed in a higher proportion of female patients in Asian countries than Western countries6, and differences in the risk factors for developing EC, disease stage at diagnosis, molecular epidemiology, and sensitivity to treatment have been observed7. EC of squamous cell and adenocarcinoma histology is characterized by differing etiologies, risk factors, and clinical characteristics, and also requires different treatment approaches8. Additional differences in the characteristics of EC exist among Asian countries, for example China and Japan9.
The clinical management of EC typically includes surgery, radiotherapy, or chemoradiotherapy for early-stage and potentially resectable disease, or systemic therapy for unresectable advanced or metastatic cancer5,10,11. Before the introduction of immunotherapy for EC in 2020, the standard neoadjuvant treatment for patients with resectable EC was chemoradiotherapy, combined with targeted therapies in patients with specific molecular aberrations5. However, chemotherapy alone is associated with suboptimal survival response rates and safety concerns, and not all patients with EC bear other targetable mutations12–14. Therefore, immunotherapy has provided an important additional treatment option for a large proportion of patients with EC, particularly in Asia, where squamous cell carcinoma is highly prevalent6.
Two immunotherapy agents, pembrolizumab and nivolumab, are approved globally for the treatment of EC, both of which have been shown to improve survival outcomes in patients with advanced EC, as a first-line treatment combined with chemotherapy (approved in 2021 and 2022, respectively) and a second-line treatment as monotherapy (approved in 2019 and 2020, respectively)10,15–17. In China, pembrolizumab was the first agent for immunotherapy approved in the first-line (2021) and second-line (2020) for treatment of advanced EC18. In addition, nivolumab and several locally produced immunotherapy drugs are approved for the treatment of EC19–23. Perioperative immunotherapy has also been shown to benefit select patients10.
Understanding EC treatment patterns in China before the introduction of immunotherapy is crucial for evaluating the resultant changes in EC management. In particular, the types of chemotherapy used in combination with immunotherapy differ in Asia and China vs. Western countries, with wider use of paclitaxel in East Asia, especially China24,25. However, data on the clinical features and treatment patterns for EC before the approval of immunotherapy in China are scarce, and prior studies have generally included relatively small numbers of patients from single centers26–29, thus limiting the generalizability of the findings to the wider Chinese population. A large-scale study is therefore required to evaluate the treatment landscape for EC before the approval of immunotherapy, to provide a baseline for future analyses. Such a study would also enhance understanding of the clinicopathologic features of EC in patients in China and identify data gaps and unmet medical needs in this population.
The CHANNEL study was a large, nationwide, multicenter, retrospective study aimed at describing the clinical features and treatment patterns of EC in China before immunotherapy approval. These data provide a foundation for exploring the correlation between immunotherapy and EC and its effects on patient outcomes.
Materials and methods
Data source and study design
This retrospective analysis was conducted on patient-level data collected from electronic medical records (EMRs) from 14 large hospitals and cancer centers across various geographic regions in China (Table 1). Patients who initiated first-line therapy for newly diagnosed EC between January 1, 2018, and December 31, 2018, were identified. The start date of this initial treatment was designated as the index date. EMRs were consecutively screened, beginning from the earliest index date and progressing in reverse to the most recent index date. Data on demographics, clinicopathologic characteristics, and anti-tumor treatment patterns for as many as 6 months after the index date were obtained and descriptively summarized.
Study centers and patient enrollment
The CHANNEL study protocol was approved by an institutional review board/independent ethics committee (IRB/IEC) at each study site before study initiation. The study was registered on EU PAS (register number: EUPAS48396).
Study population
Eligible patients were ≥ 18 years of age at EC diagnosis and had histologically or cytologically confirmed EC regardless of stage (including advanced or metastatic, early-stage, or locally advanced), had not received prior treatment for EC, and received the first anti-tumor treatment for EC during the eligibility period at the corresponding study site. All patients provided a signed informed consent form to participate in the study, or had conditions accepted by the IRB/IEC to waive this requirement. Patients were excluded if they had multiple primary cancers indicated in the EMR, or had participated in any investigational program or clinical trial involving interventions outside routine clinical practice for anti-tumor purposes.
Enrolled patients were classified into early, locally advanced, or advanced stage EC groups. Early-stage EC was defined as T1N0M0; locally advanced EC was defined as M0 and T2 or N+; and advanced EC with metastatic disease was defined as M1.
Outcomes and measurements
The primary outcomes of this analysis were the clinicopathological features and initial treatment patterns of the patients with EC. Subsequent treatment patterns (second line and beyond) were evaluated as a secondary outcome.
Demographics
The collected demographic data included age at EC diagnosis, gender, residential area, smoking and alcohol consumption history, and family history of cancer.
Clinicopathological features
The recorded clinicopathological variables included the location of the primary tumor site; histology and tumor grade; tumor stage; presence and location of distant metastases; and information on any biomarkers tested as part of the patient’s clinical management, including programmed cell death ligand 1 (PD-L1), epidermal growth factor receptor (EGFR), and human epidermal growth factor receptor 2 (HER2).
Treatment patterns
Treatments were classified into initial or subsequent lines of therapy. Initial treatment was defined as the first treatment approach received after EC diagnosis, and included multiple cycles or doses of the same anti-cancer therapy. Subsequent treatment referred to any new treatment (i.e., received after the initial treatment had been completed or discontinued) within 6 months after the start of the initial treatment.
Initial treatment patterns were described by disease status (early-stage, locally advanced, or advanced). Categories of treatment comprised surgery, neoadjuvant or adjuvant therapy, curative therapy, and palliative therapy; and treatments included radiotherapy, chemoradiotherapy, chemotherapy, and targeted therapies.
When disease recurrence occurred in patients with early-stage or locally advanced EC, similar variables to those recorded for the initial treatment were recorded for the subsequent treatment, on the basis of patients’ current disease status, including the date and type of recurrence (local, regional, or distant). For patients with advanced EC receiving systemic therapy, subsequent treatments after disease progression were documented.
Statistical analysis
All variables were summarized descriptively. Continuous variables were summarized with mean, median, range, and standard deviation, and categorical variables were analyzed with frequency distributions.
The initial treatment modality (surgery, radiotherapy, or systemic therapy) was recorded. Surgery type, dose of radiation, and number of chemoradiotherapy cycles were assessed. For each initial systemic treatment, the regimen (reported by class only), median duration of therapy, and median number of cycles administered were summarized. Similarly, the proportion of patients receiving subsequent treatments was recorded, together with the modality, treatment regimen, and number of cycles administered. Second- and third-line systemic therapies were analyzed separately. The frequency of recurrence after curative treatment and progression after systemic treatment were also recorded.
No tests of statistical significance were included in this study; 95% confidence intervals were calculated with the Clopper–Pearson method.
Results
Study population and clinicopathologic features
A total of 4,036 patients with EC were screened, among whom 3,493 patients from 14 hospitals met the eligibility criteria and were included (Table 1). The main reasons for exclusion were the presence of multiple primary tumors, missing information on treatment modality, and previous treatment. The overall screening failure rate was 13.5%. Most patients were men (75.7%), with a mean age of 64.1 years at diagnosis, and lived in an urban area (53.4%) (Table 2). A total of 75.0% of patients had no family history of cancer. Among 3,199 patients with available data, 2,142 had resectable disease [early-stage, n = 429 (13.4%); locally advanced, n = 1,713 (53.5%)], 828 (25.9%) had unresectable disease, and 229 (7.2%) had advanced-stage disease.
Patient demographics and baseline clinicopathologic characteristics
At baseline, most patients had squamous cell carcinoma (92.8%), with the primary lesion located in the middle esophagus (56.4%) (Table 2). The proportion of patients with TNM stage I, II, III, and IV disease was 14.7%, 31.2%, 38.6%, and 15.4%, respectively. Among 229 (7.1%) patients with distant metastases at baseline, the most common affected location was the lymph nodes [39 (55.7%) of patients based on non-missing data]. Biomarker data were available for only 67 (1.9%) patients.
Initial treatment patterns
Among the 2,142 patients with resectable EC, 1,989 (92.9%) underwent initial radical surgery (primarily esophagectomy, 99.2%), and 153 (7.1%) received neoadjuvant therapy (Table 3). The mean time from diagnosis to surgery was approximately 2 weeks. Chemotherapy was the most commonly used neoadjuvant therapy (85.6%) and was primarily platinum-based, most frequently platinum + taxane (85.5%).
Initial treatment for patients with resectable EC
For the 828 patients with unresectable early or locally advanced EC, the initial treatment comprised radical therapy in 49.2% of patients, including definitive chemoradiotherapy (31.7%) and radiotherapy (17.5%) (Table 4). The most prevalent radical chemoradiotherapy regimens were platinum + taxane (50.8%) and platinum + pyrimidines (10.3%). Patients treated with chemoradiotherapy received a median total radiation dose of 64.00 (Q1: 60.00, Q3: 116.00) Gy delivered over a median of 2.0 (range 1–9) chemoradiotherapy cycles.
Initial treatment for patients with unresectable EC
Palliative therapy was administered to 50.7% of patients with unresectable early and locally advanced EC (Table 4). Chemotherapy was the most frequently used palliative treatment (27.8%), and the most common regimens were platinum-based doublet therapy: platinum + taxane (66.5%) and platinum + pyrimidines (16.1%). A minority of patients (4.8%) received a platinum-based triplet (platinum + taxane + pyrimidines).
Among the 229 patients with advanced-stage EC, 84.6% received systemic therapy as an initial treatment. The most common form of systemic therapy was chemotherapy (93.4%), primarily platinum-based doublet chemotherapy (Table 5). Patients received first-line systemic chemotherapy for a median of 46.0 (range 1–191) days and a median of 3.0 (range 1–9) cycles.
Initial treatment for patients with advanced-stage EC (palliative treatment)
Subsequent treatment patterns in patients receiving surgery
Among all patients receiving adjuvant therapy after surgery (n = 503; including those also receiving neoadjuvant therapy), the most common treatment was chemotherapy (75.0%), predominantly platinum + taxane (59.6%) or platinum + pyrimidines (27.1%) (Table 6). For patients receiving adjuvant therapy after surgery alone, chemotherapy was the most common (74.8%) postoperative regimen, and was followed by chemoradiotherapy (17.3%), and radiotherapy (7.9%) (Table 7). Platinum + taxane was the most common chemotherapy and chemoradiotherapy regimen (58.5% and 34.2%, respectively). The median number of chemotherapy cycles was 3 (range 1–7), and the mean duration of treatment was 60.6 days. Among patients receiving adjuvant therapy after neoadjuvant therapy and surgery, the most common treatment was chemotherapy (76.6%), which was followed by chemoradiotherapy (14.9%), and radiotherapy (4.3%) (Table 8). Platinum + taxane was the most common chemotherapy and chemoradiotherapy regimen (69.4% and 71.4%, respectively). The median number of chemotherapy cycles was 2.0 (range 1–6), and the mean duration of treatment was 31.7 (26.5) days.
Adjuvant therapy in patients with resectable EC receiving surgery regardless of neoadjuvant therapy
Adjuvant therapy in patients with resectable EC receiving only surgery (no neoadjuvant therapy)
Adjuvant therapy in patients with resectable EC receiving neoadjuvant therapy and surgery
Subsequent treatment patterns in patients receiving systemic therapy
Among patients with advanced-stage disease receiving first-line systemic therapy (n = 197), 18 (9.1%) subsequently received second-line therapy during the study follow-up period, typically further systemic therapy (77.8%), primarily chemotherapy (94.4%). Two patients (11.1%) also started third-line treatment with systemic therapy (chemotherapy, 50.0%), or combined radiotherapy plus systemic therapy (50.0%).
Recurrence
Recurrence occurred in 19/2,549 (0.7%) patients with resectable or unresectable early-stage or locally advanced EC who received radical treatment within the 6 month observation period; most cases involved distant recurrence (17/19 cases) and typically affected the liver (6/17 cases). Progression was recorded in 8/197 (4.1%) patients with advanced-stage disease after first-line systemic therapy.
Discussion
The results of this national, multicenter retrospective study describe the clinicopathologic features and current treatment patterns for patients with EC in China, and is, to our knowledge, the largest dataset of EC characteristics and treatment patterns before the introduction of immunotherapy. These results can therefore serve as baseline data for exploring the effects of immunotherapy on patient outcomes, understanding unmet clinical needs, and providing new ideas for medical practice and research and development.
This study included a lower proportion of patients with stage IV EC (494/3,207, 15.4%) than a prior study describing the characteristics of EC in China (23%)30. This difference might be explained by several factors. First, only patients with EC without prior anti-tumor treatment were included; therefore, patients with recurrent metastases during the eligibility period were excluded from the present study. Patients with recurrent metastases will be included in the subsequent CHANNEL2 study, in which EC treatment patterns in the era of immunotherapy will be explored. Second, the recent promotion of early screening and diagnosis in China might have resulted in EC identification at earlier stages than in the prior study. Finally, because a high proportion of study sites were located in hospitals with strong surgical departments, a high proportion of newly diagnosed patients were seen in surgical departments. Because patients would be seen in such departments only if they had potentially resectable disease, i.e., early and locally advanced disease, the overall proportion of patients with stage IV EC included in this study might have been affected.
Our results indicated that neoadjuvant therapy for tumor downstaging before surgery was uncommon among patients with resectable EC. This finding probably reflects the large proportion of patients with early TNM stages (T0–T2), in whom surgeons were confident that direct R0 resection could be performed without a need for neoadjuvant therapy. In addition, neoadjuvant treatment is not usually considered a standard treatment approach for patients with early-stage resectable EC, because of limited evidence of an associated survival benefit10. When neoadjuvant therapy is warranted, the 2018 Chinese Society for Clinical Oncology (CSCO) guidelines for the diagnosis and treatment of EC recommend neoadjuvant concurrent chemoradiotherapy as a first-level expert recommendation, and neoadjuvant chemotherapy as a second-level expert recommendation for resectable locally advanced EC5. Accordingly, neoadjuvant concurrent chemoradiotherapy and neoadjuvant chemotherapy were the 2 most commonly used modalities among patients receiving neoadjuvant therapy in the present study. Therefore, exploring any changes in the use of preoperative neoadjuvant chemotherapy in patients diagnosed with EC after the introduction of immunotherapy will be of interest.
The proportion of patients receiving adjuvant therapy after surgery (22.9%) in the present study was slightly lower than that reported in prior studies by Luo et al.26 in 2023 (including patients with an EC diagnosis from 2010–2019) and Mao et al.31 in 2020 (including patients with an EC diagnosis from 2009–2014). However, this proportion is in line with findings from a nationwide database analysis in which 21.2% of patients undergoing surgery for EC received adjuvant chemotherapy (including patients with an EC diagnosis in 2015)32. This discrepancy might be explained by a lack of broad consensus regarding the benefits and use of adjuvant therapy when patients in those studies received treatment. In addition, current treatment guidelines recommend adjuvant chemotherapy for only selected patients, for example, those with pathologically confirmed regional lymph node metastasis5,10.
Most patients in the present study received radical treatment. Most patients with resectable EC underwent radical surgery, primarily minimally invasive esophagectomy or open esophagectomy. Approximately half the patients with unresectable disease received initial radical therapy; although radical chemoradiotherapy (31.7%) was the most frequently used regimen, it might have been expected to have been used in a higher proportion of patients, given that definitive chemoradiotherapy-based integrated therapy is recommended as the first choice for patients with unresectable EC in the Chinese national treatment guidelines5. Toxicity concerns with definitive chemoradiotherapy33,34, or discrepancies in access to radiotherapy resources across China35, might potentially have limited the use of this regimen during the study period, before the publication of these guidelines in 2018. However, this result is generally consistent with that from a previous study in Korea (including patients from 2005–2017) reporting the use of definitive chemoradiotherapy in ~40% of patients with unresectable EC36.
In our study, for patients with advanced EC, initial treatments consisted primarily of chemotherapy, most frequently platinum + taxane (61.5%), followed by platinum + pyrimidines (11.8%). Although no consensus exists regarding the optimal chemotherapy regimen for East Asian patients with advanced EC, current data appear to support both platinum + taxane and platinum + fluoropyrimidine regimens24. Taxane-based regimens are widely used in China as concurrent chemotherapy for patients with EC34. In addition, infusion of a taxane can be completed within 1 day, whereas pyrimidines require continuous infusion over 48 h. Because platinum + taxane might provide greater disease control and survival benefits than other regimens37, as well as patient convenience and conservation of medical resources, our finding that platinum + taxane was most frequently used might reflect that this chemotherapy regimen was preferred for advanced EC in China during the study period. Only a small proportion of patients initiated second- or third-line systemic therapy, possibly as a result of the short observation period; this finding might also be largely explained by the paucity of effective second-line therapeutic options for patients with advanced EC38. Indeed, the choice of second- and third-line regimens in the present study was heterogeneous, thereby reflecting a lack of definitive direction regarding which regimen should be used.
Traditional treatment methods have demonstrated limited survival benefits for patients with EC: single and combination chemotherapy regimens have shown a response duration of 4–6 months14 and median overall survival of 5.5–9.5 months39–42. Consequently, an urgent need existed to develop new treatment methods to improve survival benefits with targeted agents and immunotherapy in this patient population. Immunotherapy has since become the standard treatment for advanced EC, while also providing an alternative option for patients with locally advanced disease43. Now that the clinicopathological characteristics and treatment patterns for Chinese patients with EC before the introduction of immunotherapy have been characterized, the ongoing CHANNEL2 study is aimed at describing subsequent treatment characteristics in the era of immunotherapy. That study will further indicate the value of immunotherapy for the real-world treatment of EC in China.
Limitations of this study include those inherent to the retrospective design, such as sample representativeness, primarily because of the inclusion of first-diagnosis inpatients, data completeness, and the implementation of specifications. In addition, the use of rule derivation for some of the missing data might have led to bias. A risk of selection bias also existed, because patients were generally selected from large cancer centers, thus potentially limiting the generalizability of the findings to the wider Chinese population. Finally, the relatively short follow-up period led to a short observation time window and limited the collection of data regarding subsequent treatment patterns. However, examining those patterns was not the main purpose of the present study.
In conclusion, in this large observational study of patients with advanced EC treated before the introduction of immunotherapy in China, most patients with early stage or locally advanced EC underwent initial surgery without preoperative treatment, and patients with advanced EC received initial chemotherapy, with platinum + taxane as the most common regimen. The subsequent CHANNEL2 study will explore the treatment patterns of Chinese patients with EC in the era of immunotherapy, and will enable assessment of any changes or differences with respect to the era of chemotherapy described herein. We believe that these findings will help identify unmet needs in clinical practice.
Grant support
This study was funded by MSD China.
Conflict of interest statement
Danfeng Shi, Jing Qian, Si Shi, and Fengshi Dong are employees of MSD China. Lin Shen received grants to their institution from BeiGene and participated in data safety monitoring boards or advisory boards for MSD China, Boehringer Ingelheim, Servier Pharmaceuticals, AstraZeneca, and Transcenta Holding. The other authors disclose no potential conflicts of interest.
Author contributions
Conceived and designed the analysis: Lin Shen, Zhihao Lu, Yongtao Han, Danfeng Shi, Jing Qian, Si Shi, Fengshi Dong.
Collected the data: Lin Shen, Zhihao Lu, Mingfei Geng, Yongtao Han, Jianzhong Cao, Jun Wang, Tianshu Liu, Xianglin Yuan, Xue Meng, Yanqiao Zhang, Rong Zhao, Lixin Wan, Enxiao Li, Wenran Wang, Zhijie Li.
Contributed to data or analysis tools: Lin Shen, Zhihao Lu, Mingfei Geng, Yongtao Han, Jianzhong Cao, Jun Wang, Tianshu Liu, Xianglin Yuan.
Performed the analysis: Xue Meng, Yanqiao Zhang, Rong Zhao, Lixin Wan, Enxiao Li, Wenran Wang, Zhijie Li.
Wrote the paper: All authors.
Data availability statement
The data generated in this study are available upon request from the corresponding authors.
Acknowledgments
Medical writing support was provided by Jake Burrell, PhD, and Phoebe Kennedy, MSc, at Rude Health Consulting Ltd. This support was funded by MSD China.
- Received August 13, 2024.
- Accepted December 3, 2024.
- Copyright: © 2024 The Authors
This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.