DOI: https://doi.org/10.21203/rs.3.rs-2443266/v1
Background: The aim of this study was to evaluate the impact of adjuvant chemotherapy in patients with radically resected esophageal squamous cell carcinoma (ESCC).
Methods: Patients with esophageal cancer who underwent esophagectomy at our hospital from 2010 to 2019 were retrospectively analyzed. Only patients with radically resected ESCC who did not receive neoadjuvant therapy or adjuvant radiotherapy were enrolled in this study. Propensity score matching (1:1) was used to balance the baseline.
Results: A total of 1249patients met the inclusion criteria and were enrolled in the study, and 263 patients received adjuvant chemotherapy. After matching, 260 pairs were analyzed. The 1-, 3-, and 5-year overall survival (OS) rates were 93.4%, 66.1% and 59.6%, respectively, for patients with adjuvant chemotherapy compared with 83.8%, 58.4% and 48.8%, respectively, for patients with surgery alone (P=0.003). The 1-, 3-, and 5-year disease-free survival (DFS) rates were 82.3%, 58.8% and 51.3%, respectively, for patients with adjuvant chemotherapy compared with 68.0%, 48.3% and 40.8%, respectively, for patients with surgery alone (P = 0.002). In multivariateanalyses, adjuvant chemotherapy was found to be an independent prognostic factor. In subgroup analyses, only the patients in certain subgroups were found to benefit from adjuvant chemotherapy, such as patients who underwent right thoracotomy, pT3 diseases, pN1-pN3 diseases, or pTNM stage III and IVA diseases.
Conclusions: Postoperative adjuvant chemotherapy can improve the OS and DFS of ESCC patients after radical resection but may only work for patients in certain subgroups.
Esophageal cancer is one of the most common malignancies in China,[1] and esophageal squamous cell carcinoma (ESCC) is the predominant histological type. [2] Surgical resection is still a standard therapeutic approach for patients with resectable ESCC, but the prognosis is still disappointing. [2] Although neoadjuvant chemoradiotherapy plus surgery is currently recommended for patients with locally advanced ESCC, it is still an infrequently used procedure in China. In a previous study that analyzed the national database of China, neoadjuvant therapy was only given to 18.5% of the patients with esophageal cancer in 2015 (6.2% of patients received neoadjuvant chemoradiotherapy, 9.0% of patients received neoadjuvant chemotherapy, and 3.3% of patients received neoadjuvant radiotherapy), while 21.2% of the patients received adjuvant chemotherapy. [3]
The survival benefit of postoperative adjuvant chemotherapy has been demonstrated in many malignancies, including non-small cell lung carcinoma, gastric cancer, breast cancer, and colon cancer. [4–7] However, the efficacy of adjuvant chemotherapy on ESCC is still controversial. Few randomized, controlled trials have been conducted to explore the efficacy of adjuvant chemotherapy in ESCC patients after radical surgery due to disappointing results. [8, 9] Currently, no optimal postoperative adjuvant therapy is recommended in the National Comprehensive Cancer Network (NCCN) guidelines. However, an increasing number of retrospective studies have found that adjuvant chemotherapy could significantly improve survival in ESCC patients after radical resection. [10–13]
Therefore, we think that the role of adjuvant chemotherapy in patients with ESCC should be further elucidated. In this study, we retrospectively assessed the efficacy of adjuvant chemotherapy in ESCC patients after radical resection compared with those who underwent surgery alone. Propensity score matching (PSM) was also used in this study to minimize baseline differences between groups.
A total of 2324 patients with esophageal carcinoma underwent esophagectomy at Shantou University Medical College Cancer Hospital between May 2010 and July 2019. The inclusion criteria for this study were as follows: (1) thoracic esophageal squamous cell carcinoma; (2) no neoadjuvant therapy before surgery; (3) underwent radical resection; and (4) no adjuvant radiotherapy. Patients who met the following criteria were excluded: (1) cervical ESCC; (2) death or tumor relapse within 3 months after esophagectomy; and (3) concurrent or previous history of other malignancies. Approval was obtained from the institutional review board, and informed consent was acquired from all participants.
Esophagectomy was performed through a right thoracotomy or left thoracotomy, and esophagogastric anastomosis was performed in a neck incision for most of the patients. For lymphadenectomy, only the middle and lower mediastinal and the upper abdominal lymph nodes (LNs) could be dissected through a left thoracotomy. The LNs around the bilateral recurrent laryngeal nerves were also dissected through a right thoracotomy. Pathological stage was defined based on the eighth edition TNM classification.
Chemotherapy was first administered to patients at 4–8 weeks after the surgery. The most commonly used chemotherapy included the 5-fluorouracil plus cisplatin regimen, docetaxel plus cisplatin regimen, docetaxel plus nedaplatin regimen, and S-1 single-agent (tegafur, gimeracil, and oteracil potassium capsules). Combination chemotherapy was administered every 3–4 weeks for 1–6 cycles (median 4 cycles). S-1 (80–120 mg/day, d1-14, q3w) single-agent chemotherapy was administered every 3 weeks for 1 year or until tumor recurrence.
Pearson's χ2 test or Fisher’s exact test was used to compare categorical variables. The Kaplan–Meier method was used to compare overall survival (OS) and disease-free survival (DFS) between groups, and the log-rank test was used to test the survival differences. Variables with P < 0.2 in univariate analysis were included in multivariate Cox regression analysis to investigate independent prognostic factors. PSM was performed with the 1:1 nearest neighbor matching method and included the following covariates: sex, age, tumor location, tumor length, histologic grade, thoracotomy, pT category, pN category, and pTNM stage. P < 0.05 was considered statistically significant. All statistical analyses were conducted using SPSS 26.0 software (IBM, Armonk, New York, USA).
A total of 1249 patients were enrolled in this study (Fig. 1). The clinicopathological features of the study group are shown in Table 1. The median age for the whole group was 61 years (range, 37 to 84 years). The mean number of LNs removed was 26.2 ± 11.2, and the median number was 25 (range, 3–85). LN metastases were found in 529 patients (42.4%).
Variable | Original cohort | P value | Matched cohort | P value | ||
---|---|---|---|---|---|---|
S group (n = 986) | S + C group (n = 263) | S group (n = 260) | S + C group (n = 260) | |||
Sex | 0.002 | 0.912 | ||||
Male | 697 (70.7%) | 211 (80.2%) | 208(80.0%) | 209(80.4%) | ||
Female | 289 (29.3%) | 52 (19.8%) | 52(20.0%) | 51(19.6%) | ||
Age (yr) | 0.094 | 0.332 | ||||
≤ 60 | 467(47.4%) | 150 (53.0%) | 137(52.7%) | 148(56.9%) | ||
> 60 | 519(52.6%) | 133 (47.0%) | 123(47.3%) | 112(43.1%) | ||
Tumor location | 0.050 | 0.506 | ||||
Upper third | 135(13.7%) | 26 (9.9%) | 25(9.6%) | 25(9.6%) | ||
Middle third | 665(67.4%) | 172(65.4%) | 181(69.6%) | 170(65.4%) | ||
Lower third | 186 (18.9%) | 65(24.7%) | 54(20.8%) | 65(25.0%) | ||
Tumor length | < 0.001 | 0.643 | ||||
≤ 4cm | 530 (53.8%) | 86(32.7%) | 90(34.6%) | 85(32.7%) | ||
> 4cm | 456(46.2%) | 177 (67.3%) | 170(65.4%) | 175(67.3%) | ||
Thoracotomy | 0.006 | 0.617 | ||||
Left thoracotomy | 352 (35.7%) | 70(26.6%) | 65(25.0%) | 70(26.9%) | ||
Right thoracotomy | 634(64.3%) | 193(73.4%) | 195(75.0%) | 190(73.1%) | ||
Histologic grade | 0.009 | 0.757 | ||||
Well | 352(35.7%) | 73(27.8%) | 70(26.9%) | 73(28.1%) | ||
Moderate | 491(49.8%) | 135(51.3%) | 130(50.0%) | 134(51.5%) | ||
Poor | 143(14.5%) | 55 (20.9%) | 60(23.1%) | 53(20.4%) | ||
pT category | < 0.001 | 0.426 | ||||
pT1 | 200(20.3%) | 14 (5.3%) | 9(3.5%) | 14 (5.4%) | ||
pT2 | 198(20.1%) | 30(11.4%) | 40(15.4%) | 30(11.5%) | ||
pT3 | 572(58.0%) | 210(79.8%) | 204(78.5%) | 207(79.6%) | ||
pT4 | 16(1.6%) | 9(3.4%) | 7(2.7%) | 9(3.5%) | ||
pN category | < 0.001 | 0.544 | ||||
pN0 | 679(68.9%) | 41 (15.6%) | 53 (20.4%) | 41 (15.8%) | ||
pN1 | 184(18.7%) | 115(43.7%) | 106(40.8%) | 115(44.2%) | ||
pN2 | 91(9.2%) | 77(29.3%) | 70(26.9%) | 75(28.8%) | ||
pN3 | 32 (3.2%) | 30 (11.4%) | 31 (11.9%) | 29(11.2%) | ||
pTNM stage | < 0.001 | 0.936 | ||||
I | 255(25.9%) | 12(4.6%) | 11(4.2%) | 12(4.6%) | ||
II | 440 (44.6%) | 41 (15.6%) | 46(17.7%) | 41 (15.8%) | ||
III | 257 (26.4%) | 177(67.3%) | 170(65.4%) | 175(67.3%) | ||
IVA | 34 (3.4%) | 33 (12.5%) | 33(12.7%) | 32(12.3%) | ||
C, chemotherapy; S, surgery. |
Two hundred and sixty-three patients received adjuvant chemotherapy (S + C group), including 52 patients with 5-fluorouracil plus cisplatin chemotherapy, 95 patients with docetaxel plus cisplatin chemotherapy, 67 patients with docetaxel plus nedaplatin chemotherapy, and 49 patients with S-1 single-agent chemotherapy. Compared with patients who underwent surgery alone (S group), there were more males (P = 0.002) with longer tumor lengths (P < 0.001) in the S + C group. Moreover, patients in the S + C group underwent more right thoracotomy (P = 0.006) and had higher histological grade (p = 0.009) and advanced-stage tumors (P < 0.001). After 1:1 PSM, 260 well-balanced pairs were enrolled for further analysis (Table 1).
The follow-up data were updated to June 2022, and the mean follow-up time was 62.1 months (range, 4-145 months). In the whole study group of 1249 patients, 540 patients had recurrent diseases, 468 patients had died, and 33 patients were lost to follow-up (2.6%). The 1-, 3- and 5-year OS rates for the entire study group were 92.1%, 72.6% and 64.7%, respectively, and the 1-, 3- and 5-year DFS rates were 82.1%, 64.9% and 58.2%, respectively.
Before PSM, the 1-, 3- and 5-year OS rates for patients in the S group were 91.8%, 74.3% and 66.1%, respectively, which were better than the rates of 93.5%, 65.7% and 59.2%, respectively, among patients in the S + C group (Fig. 2A), although the P value of 0.052 indicated that the difference was nonsignificant. The 1-, 3- and 5-year DFS rates for patients in the S group were 82.2%, 66.6% and 60.0%, respectively, which were higher than those of 81.7%, 58.5% and 51.0%, respectively, for patients in the S + C group (P = 0.025, Fig. 2B).
In the PSM cohort, the 1-, 3- and 5-year OS rates for patients in the S group were 83.8%, 58.4% and 48.8%, respectively, compared with the rates of 93.4%, 66.1% and 59.6%, respectively, for patients in the S + C group (Fig. 3A), and the difference was significant (P = 0.003). The 1-, 3- and 5-year DFS rates for patients in the S group of 68.0%, 48.3% and 40.8%, respectively, were also significantly worse than those of 82.3%, 58.8% and 51.3%, respectively, for patients in the S + C group (P = 0.002, Fig. 3B). Other factors that were significantly correlated with survival included tumor length, pN category, and pTNM stage (Table 2).
Variable | 5-yr OS (%) | P value | 5-yr DFS(%) | P value |
---|---|---|---|---|
Sex | 0.602 | 0.836 | ||
Male | 54.5 | 46.0 | ||
Female | 52.8 | 46.1 | ||
Age (yr) | 0.583 | 0.423 | ||
≤ 60 | 55.1 | 45.1 | ||
> 60 | 52.4 | 46.6 | ||
Tumor location | 0.095 | 0.500 | ||
Upper third | 66.7 | 54.4 | ||
Middle third | 52.4 | 44.6 | ||
Lower third | 53.5 | 46.9 | ||
Tumor length | 0.008 | 0.048 | ||
≤ 4cm | 61.5 | 50.7 | ||
> 4cm | 50.4 | 43.6 | ||
Thoracotomy | 0.173 | 0.178 | ||
Left thoracotomy | 50.0 | 41.1 | ||
Right thoracotomy | 55.8 | 48.0 | ||
Histologic grade | 0.696 | 0.574 | ||
Well | 56.3 | 46.7 | ||
Moderately | 53.9 | 47.2 | ||
Poorly | 52.8 | 41.7 | ||
pT category | 0.051 | 0.066 | ||
pT1 | 76.1 | 64.5 | ||
pT2 | 56.4 | 52.1 | ||
pT3 | 53.1 | 44.3 | ||
pT4 | 37.0 | 36.5 | ||
pN category | < 0.001 | < 0.001 | ||
pN0 | 81.7 | 72.7 | ||
pN1 | 60.2 | 53.6 | ||
pN2 | 37.3 | 31.3 | ||
pN3 | 28.4 | 12.4 | ||
pTNM stage | < 0.001 | < 0.001 | ||
I | 76.1 | 59.3 | ||
II | 80.1 | 73.2 | ||
III | 50.8 | 44.4 | ||
IVA | 28.1 | 13.0 | ||
Adjuvant chemotherapy | 0.003 | 0.002 | ||
No | 48.8 | 40.8 | ||
Yes | 59.6 | 51.3 | ||
DFS, disease-free survival; OS, overall survival. |
In multivariate analysis, pN category and adjuvant chemotherapy were independently correlated with OS and DFS, while thoracotomy was only independently correlated with DFS. None of the other factors were independent risk factors in this matched cohort (Table 3).
Prognostic factor | Hazard Ratio | 95%CI | P value |
---|---|---|---|
Overall survival | |||
Tumor location | 1.152 | 0.907–1.464 | 0.245 |
Tumor length | 1.173 | 0.873–1.575 | 0.290 |
Thoracotomy | 0.775 | 0.585–1.028 | 0.077 |
pT category | 1.016 | 0.733–1.407 | 0.924 |
pN category | 1.660 | 1.286–2.142 | < 0.001 |
pTNM stage | 1.164 | 0.778–1.742 | 0.461 |
Adjuvant chemotherapy | 0.579 | 0.448–0.748 | < 0.001 |
Disease-free survival | |||
Tumor length | 1.021 | 0.781–1.336 | 0.877 |
Thoracotomy | 0.769 | 0.593–0.997 | 0.048 |
pT category | 1.101 | 0.816–1.486 | 0.528 |
pN category | 1.856 | 1.469–2.346 | < 0.001 |
pTNM stage | 1.093 | 0.756–1.579 | 0.636 |
Adjuvant chemotherapy | 0.576 | 0.454–0.731 | < 0.001 |
CI, confidence interval. |
The impact of adjuvant chemotherapy on survival in subgroup analyses is shown in Table 4. Adjuvant chemotherapy was found to improve survival in patients who underwent right thoracotomy (P = 0.006) but not in patients who underwent left thoracotomy (P = 0.178). Patients with pT3 diseases, pN1-pN3 diseases, or pTNM stage III and IVA diseases were more likely to benefit from adjuvant chemotherapy than patients with pT1-2 diseases, pT4 diseases, pN0 diseases, or pTNM stage I - II diseases. Moreover, adjuvant chemotherapy was also found to improve survival in the subgroups of male patients, age > 60 years, tumor > 4 cm, tumor located in middle or lower third of the thorax, and moderately or poorly differentiated tumors.
Variable | No of patients | 5-yr OS (%) | P value | 5-yr DFS (%) | P value | ||
---|---|---|---|---|---|---|---|
S | S + C | S | S + C | ||||
Sex | |||||||
Male | 417 | 49.2 | 60.0 | 0.009 | 40.1 | 52.0 | 0.003 |
Female | 103 | 47.8 | 57.9 | 0.137 | 44.1 | 48.1 | 0.285 |
Age (yr) | |||||||
≤ 60 | 285 | 53.9 | 56.2 | 0.445 | 42.4 | 47.5 | 0.167 |
> 60 | 235 | 42.7 | 64.5 | < 0.001 | 38.5 | 56.1 | 0.001 |
Tumor location | |||||||
Upper third | 50 | 61.5 | 70.4 | 0.373 | 43.6 | 66.0 | 0.053 |
Middle third | 351 | 48.4 | 56.9 | 0.036 | 41.2 | 48.4 | 0.054 |
Lower third | 119 | 43.3 | 62.0 | 0.036 | 38.9 | 53.2 | 0.034 |
Tumor length | |||||||
≤ 4cm | 175 | 59.0 | 64.1 | 0.212 | 46.9 | 55.2 | 0.228 |
> 4cm | 345 | 43.4 | 57.4 | 0.004 | 37.4 | 49.4 | 0.002 |
Thoracotomy | |||||||
Left thoracotomy | 135 | 46.0 | 53.8 | 0.178 | 39.7 | 42.3 | 0.446 |
Right thoracotomy | 385 | 49.7 | 62.7 | 0.006 | 41.1 | 55.3 | 0.001 |
Histologic grade | |||||||
Well | 143 | 57.8 | 55.1 | 0.987 | 46.6 | 46,3 | 0.749 |
Moderately | 264 | 47.0 | 61.0 | 0.007 | 40.5 | 53.8 | 0.009 |
Poorly | 113 | 43.5 | 63.6 | 0.038 | 33.4 | 51.0 | 0.017 |
pT category | |||||||
pT1 | 23 | 88.9 | 64.3 | 0.547 | 88.9 | 47.6 | 0.158 |
pT2 | 70 | 54.4 | 59.4 | 0.416 | 49.1 | 56.3 | 0.434 |
pT3 | 411 | 46.7 | 59.7 | 0.004 | 37.8 | 50.8 | 0.001 |
pT4 | 16 | 17.1 | 51.9 | 0.276 | 14.3 | 53.3 | 0.080 |
pN category | |||||||
pN0 | 94 | 88.2 | 72.7 | 0.473 | 77.9 | 65.5 | 0.310 |
pN1 | 221 | 52.8 | 67.7 | 0.035 | 47.7 | 59.3 | 0.031 |
pN2 | 145 | 24.4 | 49.5 | 0.002 | 19.6 | 42.9 | 0.001 |
pN3 | 60 | 20.2 | 37.9 | 0.011 | 3.2 | 21.9 | < 0.001 |
pTNM stage | |||||||
I | 23 | 90.9 | 59.5 | 0.489 | 79.5 | 41.7 | 0.104 |
II | 87 | 86.4 | 72.1 | 0.458 | 77.5 | 68.1 | 0.497 |
III | 345 | 41.3 | 60.6 | 0.001 | 35.8 | 53.1 | < 0.001 |
IVA | 65 | 18.4 | 38.3 | 0.005 | 3.0 | 23.1 | < 0.001 |
C, chemotherapy; DFS, disease-free survival; OS, overall survival; S, surgery. |
The CROSS study published in 2012 established neoadjuvant chemoradiotherapy plus surgery as the standard treatment for locally advanced esophageal cancer. [14] However, the long-term survival is still very disappointing, with a 10-year OS of only 38% for patients treated with the CROSS strategy. [15] Recently, the CheckMate 577 trial showed that adjuvant nivolumab therapy could improve DFS for patients with residual disease after neoadjuvant chemoradiotherapy plus surgery. [16] However, because of the different tumor prevalence and surgery preferences, the optimal treatment strategies for locally advanced ESCC remain unclear. [17]
Over 80% of the patients with ESCC chose surgical resection as their primary therapy in China. [3] However, no optimal postoperative adjuvant therapy for these patients was recommended in the NCCN guidelines. Due to the high recurrence rate and poor prognosis for patients with ESCC who underwent surgery alone, postoperative adjuvant therapy has been used to improve the likelihood of survival. However, the efficacy of adjuvant chemotherapy on ESCC is still controversial. Few randomized trials have evaluated the effect of adjuvant chemotherapy on patients with ESCC after radical surgery. The JCOG8806 study revealed that no survival benefit was obtained from adjuvant chemotherapy using a combination of cisplatin and vindesine. [8] In the JCOG9204 study, the 5-year DFS was significantly better when patients with positive LNs received adjuvant chemotherapy with cisplatin plus fluorouracil (52.0% vs. 38.0%; p = 0.041); however, the difference for OS was not significant. [9] Both of these clinical trials were conducted in an early period. Recently, some retrospective studies and meta-analyses found that adjuvant chemotherapy could improve survival in certain subgroups for patients with ESCC after radical resection, especially for patients with positive LNs,[10, 11, 13,18, 19] indicating that the efficacy of adjuvant chemotherapy should be further determined.
Our current retrospective study enrolled one of the largest patient cohorts to date. We also used PSM to minimize baseline differences between the S group and the S + C group. Our results confirmed that adjuvant chemotherapy not only improved DFS for patients with ESCC who underwent radical resection but also improved OS, and adjuvant chemotherapy was an independent prognostic factor. These results were similar to some of the other retrospective studies. [10, 11, 18] In a recent meta-analysis by Zhao et al.[19] that enrolled 9 studies and a total of 1684 cases, the authors also found that adjuvant chemotherapy could improve OS (HR 0.78, P = 0.002) and DFS (HR 0.72, P < 0.001) for patients with ESCC. There may be three reasons for the positive results in our study. First, all of the surgeries were performed or closely supervised by two senior surgeons (J. S Yang and Y. P Chen). The homogeneity of the surgical treatment will reduce the methodological biases. Second, the mean dissected LN number in this study was 26.2, which was higher than that in most previous studies. The insufficiency of lymphadenectomy would lead to higher locoregional LN recurrence, which might not be very well controlled by adjuvant chemotherapy. Third, most of the patients in our study received sufficient cycles of chemotherapy (88.2% with more than 3 cycles). Patients who undergo esophagectomy always have a slow recovery process, and the lower tolerance to adjuvant chemotherapy would impact the effect of this therapy. [20]
Subgroup analyses in our study showed that not all patients benefited from adjuvant chemotherapy. Our findings that patients with pN1-pN3 diseases and pTNM stage III - IVA diseases, but not pN0 diseases and pTNM stage I - II diseases, were more likely to benefit from adjuvant chemotherapy were consistent with previous studies. [13, 21] Patients with positive LNs and advanced TNM stage are known to have a high risk of tumor recurrence and should be more likely to have systemic disease, so systemic chemotherapy might improve the survival of these patients. [12] Accordingly, patients with moderately and poorly differentiated tumors and tumor lengths >4 cm were also found to benefit from adjuvant chemotherapy, as previous studies showed that these patients might have a higher rate of tumor recurrence and worse survival. [22, 23] However, Pasquer et al.[24] found that adjuvant chemotherapy did not improve the OS and DFS for esophageal cancer patients with positive LNs. Ando et al.[8] also found that adjuvant chemotherapy did not improve 5-year survival for LN-positive ESCC patients (43.7% vs. 35.5%, P = 0.13). The differences in surgical approach, lymphadenectomy, chemotherapy agents, and chemotherapy cycles might contribute to these different results.
Our subgroup analyses also showed that patients who underwent right thoracotomy were more likely to benefit from adjuvant chemotherapy than patients who underwent left thoracotomy. Bilateral recurrent laryngeal nerve LNs, with nearly 40% involvement, were the most frequent metastatic nodes in thoracic ESCC. [25, 26] However, these LNs could not be removed through a left thoracotomy. This means that nearly 40% of ESCC patients who undergo left thoracotomy may not receive radical surgery, resulting in a high rate of locoregional recurrence. For these patients, postoperative chemoradiotherapy but not chemotherapy may be a better adjuvant therapy to reduce locoregional recurrence and improve survival. [27, 28]
Surprisingly, we found that older patients (>60 years) might benefit from adjuvant chemotherapy but not younger patients (≤60 years) in our subgroup analyses, which was different from the result by Zhu et al.[21] There may be several reasons for this result. First, the cutoff point of age was different in these two studies (60 years vs. 65 years). Second, the chemotherapy cycles were similar between younger and older patients in our study; however, younger patients received more chemotherapy cycles in the study by Zhu et al.[21] (although the P value was 0.081). Most previous studies have shown that age is a factor that influences treatment choices but not necessarily outcomes. [29, 30] Both younger patients and older patients could benefit comparably from chemotherapy. [29, 30] According to our results, we think that adjuvant chemotherapy should not be withheld based on age alone in ESCC patients after surgery. Although we also found that adjuvant chemotherapy might not benefit female patients or patients with tumors located in the upper third of the thorax, the sample sizes in these subgroups were too small to draw a conclusion. We think that more data should be collected to evaluate these results.
There are some limitations to this study. First, it was a retrospective study, and we could not analyze the toxicity of adjuvant chemotherapy in this study, as most of these data were missing. Second, different chemotherapy agents, such as 5-fluorouracil, cisplatin, and docetaxel, were used, and we could not define the optimal chemotherapy regimens in this study. Third, although PSM was used to balance the baseline differences, some of the other factors that may impact the prognosis, such as performance status, were not included in this study. With the development of new drugs, such as taxanes, increasing data have shown that adjuvant chemotherapy may improve survival in patients with ESCC. We think that a multicenter, randomized clinical trial should be conducted to explore the role of adjuvant chemotherapy in patients with ESCC after radical surgery.
In conclusion, postoperative adjuvant chemotherapy improves the OS and DFS of ESCC patients after radical resection but may only work for patients in certain subgroups. Further multicenter, randomized clinical trials should be conducted to evaluate our findings.
DFS: Disease-free survival; ESCC: Esophageal squamous cell carcinoma; LN: Lymph nodes; NCCN: National Comprehensive Cancer Network; PSM: Propensity score matching; OS: Overall survival.
Ethics approval and consent to participate: All patients involved in this study sign informed consent before operation. This study was approved by an independent ethics committee at Shantou University Medical College Cancer Hospital and conducted in accordance with the principles of the Declaration of Helsinki.
Consent for publication: All patients provided informed consent.
Availability of data and materials: The datasets generated during and/or analyzed during the current study are not publicly available due to hospital regulations.
Competing interests: The authors declare no competing interest.
Funding: Science and Technology Special Fund of Guangdong Province of China(190829105556145) .
Author contributions: Shao-bin Chen and Yu-ping Chen designed the research, analyzed the data and wrote part of the paper. Di-tian Liu analyzed the data and wrote part of the paper.
Acknowledgments: We would like to thank American Journal Experts (www.aje.cn) for English language editing.