Persistent Lymph Node Metastases After Neoadjuvant Chemoradiotherapy for Rectal Cancer

Importance Patients with locally advanced rectal cancer and persistent lymph node metastases (PLNM) after neoadjuvant treatment are at high risk of developing locoregional and distant metastasis, yet optimal postsurgical treatment of these patients is limited. Objective To analyze the association of PLNM with pretreatment clinical parameters, intensity of neoadjuvant treatment, and long-term oncological outcomes. Design, Setting, and Participants This cohort study is a post-hoc analysis of 3 randomized clinical trials (Surgical Oncology Working Group of Germany [CAO], Radiological Oncology Working Group of Germany [ARO], and Working Group for Internal Oncology in the German Cancer Society [AIO]) conducted in Germany in 1994, 2004, and 2012 that included 1948 patients with locally advanced rectal cancer recruited between February 1995 and January 2018. Statistical analysis was conducted between September 2023 and February 2024. Exposures Receiving preoperative fluorouracil-based chemoradiotherapy (CRT, comprising the preoperative group of CAO/ARO/AIO-94 and the control group of CAO/ARO/AIO-04), fluorouracil-based CRT plus oxaliplatin (experimental group of CAO/ARO/AIO-04), or total neoadjuvant treatment (TNT) with fluorouracil-based CRT plus oxaliplatin with induction or consolidation leucovorin calcium (folinic acid), fluorouracil, and oxaliplatin chemotherapy within the CAO/ARO/AIO-12 trial. Main Outcome and Measures The associations of PLNM with clinical parameters, intensity of neoadjuvant treatment, and cumulative incidences of LR, DM, and overall survival were assessed. Results A total of 1888 patients (1333 male participants [70.6%]; median [range] age, 62 [19-84] years) with locally advanced rectal adenocarcinoma (clinical tumor stage 3 to 4 and/or clinically node-positive) treated within 3 consecutive clinical trials were analyzed. A total of 522 (29%) experienced PLNM; 378 had lymph node stage (ypN) 1 (20%) after neoadjuvant treatment (ypN) 1 (20%), and 174 had ypN2 (9%). Age, clinical T-stage, N-stage, grading, carcinoembryonic antigen levels, and time interval from completion of CRT to surgery were significantly associated with PLNM, whereas sex and tumor location were not. The percentage of patients with ypN2 stage was almost halved after TNT (18 of 293 patients [6%]) compared with patients treated with fluorouracil-based CRT (114 of 1009 patients [11.3%]; chi(2)(6) = 16.693; P = .01). After a median (IQR) follow-up of 54 (37-62) months, 5-year overall survival was 86.1% (95% CI, 83.9%-88.4%) for ypN0, 74.0% (95% CI, 83.9%-88.4%) for ypN1, and 43% for ypN2 (95% CI, 35.4%-52.2%) (P < .001). The 5-year cumulative incidences of locoregional and distant metastases were, respectively, 3% (95% CI, 2.1%-4.2%) and 20% (95% CI, 18%-23%) for ypN0, 6% (95% CI, 3.4%-8.8%) and 40% (95% CI, 34%-46%) for ypN1, and 19% (95% CI, 13%-26%) and 72% (95% CI, 63%-79%) for ypN2 (both P < .001). Conclusions and Relevance In this cohort study, PLNM unmasked an unfavorable phenotype of rectal cancer at high risk for treatment failure. More aggressive adjuvant treatment might be considered; however, risk-adapted surveillance strategies and early recurrence-directed surgery, if feasible, are important strategies in this group of patients with CRT- and/or chemotherapy-resistant disease.