Esophageal carcinoma (EC) is a highly prevalent and dangerous malignancy of the gastrointestinal (GI) tract (1, 2), accounting for the seventh most common cancer worldwide as well as ranking sixth for cancer-related deaths (3). EC claims approximately 450,000 lives annually (4), with a 5-year survival rate of only 18% (5). Notably, the incidence of EC is increasing rapidly in several countries, including China, in recent times (6, 7). In China, EC causes about 150,000 annual deaths, which equals 21.8% of all cancer deaths, ranking fourth in the list of deadliest carcinomas. The etiology of EC is complex, many factors such as drinking, smoking, diet, genetics, microorganisms and gender have been reported to be associated with elevated risk of EC by previous studies (5–7). Therefore, it is technically challenging and difficult to reduce the occurrence of EC. EC patients primarily undergo endoscopic diagnosis for GI symptoms. Esophagectomy is the major surgical intervention in EC patients, while other treatment modalities involve chemo alone or a combination of chemo and radiotherapy, depending on the clinical characteristic of the cancerous lesion (8, 9). Most often, patients are diagnosed with an advanced stage EC, leaving no option for radical surgery. Importantly, postoperative recurrence and metastasis are frequently detected in surgical EC patients, hence, chemotherapy is considered a potentially effective second line of treatment (10). Interestingly, there is a wide variation in EC tumor classification in different geographical regions, such that >90% of EC cases in Western countries are adenocarcinomas, while esophageal squamous cell carcinoma (ESCC) is the most prevalent (>90% of all cases) among the Chinese populations (11). Therefore, this wide range of variabilities hinders the universal application of the National Comprehensive Cancer Network (NCCN) guidelines to Chinese ESCC patients, directly affecting their treatment outcomes and tumor recurrence. Moreover, establishing an in vitro screening strategy for determining patient-specific sensitivity of chemotherapeutics may resolve the discrepancies in the generalized application of NCCN guidelines.

High-throughput drug sensitivity (HDS) screening strategy is a new method with the advantages of high efficiency, wide coverage, and potential adaptation to personalized therapy, which are promising for determining the antitumor drug sensitivity in liver cancer, head and neck cancer, and lymphomas (12–14). However, the application of this technique in antitumor drug screening of ESCC is rarely reported. Therefore, this study aimed to generate ESCC patient-derived cell models for HDS analyses and develop a patient-derived xenograft (PDX) model of human ESCC in rodents for in vivo investigation. This study provides optimized assay platform able to offer drug screening outcomes in ESCC samples with varying clinical features and guides the precision therapy for ESCC patients (see Figure 1).

Since the majority of patients with progressive ESCC are diagnosed at the advanced stage, most clinicians adopt palliative radiotherapy, chemotherapy, biological therapy, and traditional Chinese medicine (TCM) therapies (19–22). Notably, due to extensive heterogeneity in tumor characteristics across ESCC patients, individualized chemotherapeutic strategies are urgently needed to achieve the best therapeutic outcome and minimum drug toxicity (23, 24). Most clinical researchers exploit the patient-derived tumor cell-derived PDX rodents to assess the cell growth inhibitory effect of different regimens of anticancer drugs to overcome the problem of adverse drug toxicity in critically ill ESCC patients (25, 26). However, these in vitro testing models have their limitations, such as high cost and limited capability in accommodating large panels of therapeutic drugs. Some studies have demonstrated the broader benefits of organoid models using patient-derived cancer stem cells mimicking in vivo tumor microenvironment (TME) for drug screening (27, 28). Again, the efficacy of this model is limited by challenges in accurately modeling cancer organoids, high maintenance costs, and difficulty in simultaneously generating numerous organoids for high-throughput screening strategies.

Recently, high-throughput gene sequencing (HGS) technology has been used for the targeted therapy of ESCC. Some groups have reported the genetic landscape of human ESCCs using the whole genome and exome sequencing approaches (29–31), which has led to promising results in early studies of personalized and targeted therapies (32). However, due to an increase in the oncogenic mutational burden, differences in cellular signaling pathways, and complex interactions among tumor genes, the same tumor gene target may respond differentially to different drugs affecting the overall outcome of individualized therapy (33).

In contrast, the HDS technology used by our group extended the scope of screening with a large panel of FDA-approved drugs for determining the best possible drug regimen exerting maximum cell growth inhibitory effects. With this technically upgraded screening platform, we can rapidly identify highly sensitive and targeted chemotherapeutics for different ESCC patients, thereby solving most of the previously mentioned technical hurdles.

The primary aim of this study was to obtain high-quality primary tumor cells from ESCC patients and subsequently exploit them for in vitro drug sensitivity screening strategies as an alternative to in vivo models. We used fresh tumor tissue masses to isolate primary ESCC cells through trypsin digestion with a success rate as high as 90%, which was considerably higher than that reported in other studies (34). The HDS technique developed by our group could simultaneously measure the sensitivity of more than 300 drugs to patient-derived ESCC tumor cells, along with the screening of optimal drug concentrations to reduce the adverse drug toxicity effects. In this study, primary tumor cells from 40 ESCC patients were tested for drug sensitivity against 9 different chemotherapeutic agents, which fully illustrated the inter-patient heterogeneity of tumor cells (35). Moreover, the results of this study further signify the importance of individualized chemotherapeutic regimens in treating highly heterogeneous carcinomas like ESCC. Through this screening assay, we noticed that most ESCC patients were sensitive to Bortezomib.

Bortezomib is a 20S proteasome inhibitor that exerts its antitumor effects by interfering with cell signaling, causing cell cycle arrest, inducing apoptosis, and inhibiting angiogenesis and is currently used mainly for the treatment of multiple myeloma (36, 37). However, one study (38) reported its role in enhancing the sensitivity of ESCC to radiotherapy. Besides, there are few a studies on the application of Bortezomib in chemotherapy for advanced-stage ESCC (39). Here, we found that Bortezomib was effective in chemotherapy for esophageal squamous carcinoma by in vitro HDS assay, which might provide a new avenue for the treatment of ESCC.

To further verify the results of in vitro experiments, the previously reported PDX model (40) was used to determine the therapeutic effects of screened anticancer drugs in vivo. We exploited patient-derived primary ESCC cells for constructing the PDX mouse model to screen potential anticancer drugs. In addition, we used Paclitaxel (41) as the positive control in this animal study. The results showed that the tumor tissue mass and volume of PDX mice treated with screened high-sensitivity drugs were significantly smaller than those treated with Paclitaxel or the vehicle control (p < 0.05). Finally, we determined the optimal drug concentration to achieve the best possible tumor growth inhibition in this PDX mouse model.

After achieving encouraging results in animal experiments, we conducted individualized treatment for 40 ESCC patients with Bortezomib and compared the anticancer efficacy with 40 ESCC patients who received Paclitaxel combined with cisplatin chemotherapy in the same period of time. We first analyzed the differences in expressions of tumor markers CEA, CYFRA21-1, and SCCA between the two groups. CEA is one of the first tumor markers found to be widely present in tumor cells. CYFRA21-1 is also commonly detected in lung and esophageal tumor cells and exhibits a high sensitivity along with the diagnostic value in the evaluation of therapeutic effect against ESCC (42). SCCA is mainly present in tumor cells of the lung, pharynx, and esophagus and exhibits a high sensitivity for occurrence and progression of ESCC with high specificity (43). Some studies (44–46) have reported that the combination of these three biomarkers could be a valuable prediction tool for the precise diagnosis of EC and for predicting the disease progression rate as well. Here, we found that CEA, CYFRA21-1, and SCCA levels were significantly lower than those in the control group after chemotherapy treatment in the observation group (p < 0.05), suggesting that the HDS-screened anticancer drug was effective in controlling the ESCC tumor growth in the observation group. Furthermore, we analyzed the treatment efficacies in the two groups based on clinical symptoms and found that the 12-month and 18-month OS rates in the observation group were significantly higher than those in the control group (p < 0.05), while rates of LP and DM were significantly lower than those in the control group (p < 0.05), indicating promising efficacy of HDS-screened drugs against ESCC.

In summary, the HDS method is an efficient, reproducible, and easy-to-use drug sensitivity screening tool that can expedite the ESCC diagnosis, especially for terminally ill patients. Moreover, drugs for antitumor sensitivity on the HDS platform consistently exhibited significant inhibitory effects on ESCC tumor growth in disease animal models as well as human ESCC patients.

The original contributions presented in the study are included in the article/Supplementary materials, further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by Ethics Committee of the 901th Hospital of the Joint Support Force. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by Ethics Committee of the 901th Hospital of the Joint Support Force.

The study was conceived and designed by XnH, JT, HY, JH, and WW. XnH, HY, and JH equally contributed to performing all the required experiments, analyzing data, preparing the manuscript, and further revisions. SY and XD contributed to the data acquisition and analysis. JT, HL, JW, JL, YG, XC, and MZ provided human adipose tissue samples. JH, HY, WW, and XnH performed the HDS experiments. QG, XlH, GL, and JH performed animal studies and immunohistochemical analysis. XlH, HY, XD, HL, JW, SG, and JT were responsible for clinical studies. WW and JT provided critical suggestions for this manuscript. All authors contributed to the article and approved the submitted version.

This study was supported by the Anhui Key Research and Development Program (1804 h08020257).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.