785. MANAGEMENT OF ESOPHAGEAL CANCER WITH CONCURRENT CERVICAL NODE METASTASIS: A NATIONWIDE POPULATION-BASED COHORT STUDY

Background In the Netherlands, the standard treatment of locally advanced, resectable esophageal cancer without metastasis is neoadjuvant chemoradiotherapy followed by esophagectomy. There is a small subset of patients that present with concurrent cervical lymph node metastasis (LNM). Historically this was seen as distant metastasis and surgical intervention has usually not been an option for these patients. The contemporary TNM classification now categorizes these lymph node stations as locoregional disease. Our current study aims to describe current treatment paradigms in the Netherlands for patients presenting with esophageal cancer and concurrent cervical LNM. Methods This population-based cohort study utilized data from the Netherlands Cancer Registry (NCR), encompassing patients with locally advanced thoracic esophageal or gastroesophageal junction cancer and concurrent cervical lymph node metastasis. Treatment modalities were categorized into five options: neoadjuvant therapy followed by surgery (Neo + S), definitive chemoradiotherapy (dCRT), chemotherapy with or without radiotherapy < 30 Gray (CT), radiotherapy (RT), and best supportive care (BSC). Overall survival (OS) was assessed using the Kaplan-Meier method and compared via the log-rank test. Hazard rates were computed using Cox proportional hazards regression, with adjustment for confounding achieved through inverse probability of treatment weighting (IPTW). Results Between 2015 and 2021, a cohort of 412 patients was identified from the NCR database. Median survival durations were observed as follows: 24.2 months for Neo + S, 18.0 months for dCRT, 14.5 months for CT, 7.0 months for RT, and 3.2 months for BSC (Figure). A comparison between the Neo + S group and dCRT demonstrated a significant improvement in survival (p=0.02). Further subdivision of the surgical group into neoadjuvant CRT or chemotherapy did not reveal a significant difference in survival (p=0.6). Utilizing IPTW to adjust for confounding factors, Neo + S maintained its survival advantage. Conclusion The retrospective cohort findings suggest that neoadjuvant therapy followed by surgery may represent the optimal approach for managing esophageal cancer patients with cervical LNMs Yet, it's vital to recognize the influence of confounding by indication, which statistical adjustments may not entirely rectify. Furthermore, immortal time bias notably skews results favorably toward surgery. Nevertheless, the results emphasize the importance of considering surgery as a viable option for these patients. These limitations underscore the critical need for a prospective study, prompting the launch of the NODE-II trial.