ESCC often metastasizes to LN at very early stages, with metastase sites ranging from the neck to the abdomen (8). Postoperative pathological results show a high rate of LN metastases, even in cases treated with NCRT, which is considered the most important prognostic factor for patients (1). Although there have been studies on the location of metastatic LN after neoadjuvant therapy in both squamous cell carcinoma and adenocarcinoma, treatment protocols vary among authors, resulting in inconsistent outcomes (9, 10) and there also no focus on the group of patients like our study. Hagens et al. published an article about LN metastases after NCRT but in esophageal adenocarcinoma rather than ESCC, as in our study (9). Compared with Kim (11) and Miyata (12), about haft of patients in their study underwent three-field lymphadenectomy rather than two-field lymphadenectomy in our study.
In this study, the median number of dissected LN was 24 (range 8-28), comparable to several other studies (3, 10). NCRT reduces both the number of dissected LN and the number of metastatic LN. A recent study showed that the LN metastase rate was 74% for patients undergoing esophagectomy alone, compared to only 31% for those receiving neoadjuvant therapy (3). Our study observed a low LN metastase rate of 4.11% with a median of 2 metastatic LN (range, 1-20). Compared to Goubing Pan's study, the LN metastase rates for middle-third and lower-third ESCC were 30.7% and 15.7%, respectively (13). This demonstrates that NCRT significantly reduces the rate of LN metastases.
We cannot identify specific pattern of lymphatic metastasis, although some LN stations had high rate of metastatis after NCRT, such as in perigastric region (8.97%). When comparing the thoracic and abdominal regions, the abdominal LN had a higher median number of positive LN (2 vs. 1.5) and a higher rate of LN metastases (5.94% vs. 2.84%). This demonstrates the role of thoracic radiation in reducing the rate of chest LN metastases. Meanwhile, abdominal radiation is only considered when abdominal LN metastases are suspected – that is, when a CT scan detects nodes with a short-axis diameter >1 cm, they are deemed metastatic and are then irradiated (6). However, abdominal LN often measure less than 1 cm but may still be metastatic. This indicates a gap in the treatment approach, suggesting that more attention should be paid to the radiation field in the abdominal area without considering lymph node size, given the high risk of metastasis in this region.
Although NCRT reduces the number of metastatic LN, it cannot replace systematic lymphadenectomy. One study reported that 7% of LN metastases within the radiation field, despite no suspected metastases on chest CT scans (11). Another study evaluated the correlation between LN metastases and radiation field in patients with a complete response to NCRT. The results showed that 11 patients (21%) had metastatic LN, with 8 of these having metastases outside the radiation field. The authors identified high-risk areas, including LN around the esophagus, along the aorta, and near the gastroesophageal junction (14). To evaluate the impact of the radiation field on LN metastases, we assessed abdominal LN metastases, as this area includes nodes outside and inside of the radiation field. Our results showed that LN metastases outside the radiation field were significantly higher than within the radiation field. This finding is consistent with Kim et al.'s study, where the rate of LN metastases within the radiation field was 3% compared to 11% outside the radiation field (11). Therefore, abdominal LND is crucial, as the radiation field does not cover all these nodes, leading to a high risk of metastases outside the radiation field. In addition, in our study, among 30 patients with a complete response at the ypT0-stage, seven patients had LN metastases (23.3%). Therefore, systematic lymphadenectomy remains crucial and cannot be replaced by NCRT.
Although LN metastasis has a poor prognosis in esophageal cancer (1), extensive LN dissection is a complex surgery with risks and complications (15-17). Therefore, some studies have aimed to identify risk factors for LN metastasis to minimizing lymphadenectomy for patients without risk factors (11, 18). However, the results remain inconclusive. In this study, based on multivariate regression analysis, we found that LN metastasis is closely associated with ycN-stage tage (HR=2.03, 95% CI 1.02-4.04, p=0.044). However, similar to previous studies, this factor cannot be definitively identified clinically. Therefore, we cannot recommend minimizing lymphadenectomy based on this finding.
Regarding long-term results, the LN metastases were not significantly associated with OS or DFS (p=0.891 and p=0.653, respectively). However, we divided patients into two groups based on the number of LN removal and found that it was significantly associated with both OS and DFS (p=0.040 and p=0.049, respectively). Our findings align with the current literature (5, 9, 19, 20). These results further reinforce the role of systematic lymphadenectomy in managing ESCC. Surgeons must ensure they dissect the necessary number of LN, as the number of LN is related to OS and DFS.
Our study, while insightful, is not without its limitations, which must be acknowledged to contextualize our findings accurately. Firstly, the relatively small cohort of patients limits the statistical power of our analysis and restricts our ability to detect smaller yet potentially significant differences in outcomes. A larger sample size would have enabled more robust conclusions and potentially unveiled subtler patterns and correlations that remain obscured in this study. Secondly, all patients included in our study were diagnosed with ESCC. This singular focus on ESCC limits the generalizability of our results across the broader spectrum of esophageal cancer types. Moreover, the study was conducted within a single institution, which may introduce institutional bias and limit the external validity of our findings. The surgical techniques, perioperative care protocols, and postoperative management strategies can vary widely between institutions. As a result, the outcomes observed in our study may reflect the specific practices and expertise of our institution rather than a universal standard. Multi-center studies would be beneficial in providing a more comprehensive picture and enhancing the generalizability of the results.