Value of size and malignant features of lateral lymph nodes in risk stratification at lateral local recurrence of rectal cancer: a national cohort study

Abstract

Over the past few decades, treatment strategies for patients with advanced rectal cancer have developed rapidly, and now the combination of neoadjuvant (chemo)radiotherapy ([C]RT) and total mesorectal excision (TME) has become the cornerstone of Western curative therapy.1 This has resulted in a decrease in local recurrence (LR) rates but a proportional increase in lateral local recurrences (LLRs), now comprising half of all LRs.2 These LLRs are most likely caused by spread to lateral lymph nodes (LLNs) located beyond the TME plane.3 An LLN dissection (LLND) has been proven effective in reducing LLRs and is widely used in Asian countries.4,5 However, it is a technically difficult procedure and is associated with increased morbidity. Hence, it is important to accurately predict which LLNs would lead to increased LLR rates in order to carefully select patients who might benefit from LLND, while sparing others who are at less risk for LLR.

The current American Society of Colon and Rectal Surgeons guidelines recommend an LLND for patients with suspicious LLNs,6 although the definition of “suspicious” remains controversial, with short-axis sizes of either >5.0 mm or ≥7.0 mm suggested as the cutoff.7,8 Conversely, for mesorectal lymph nodes, a clear definition of “suspicious” exists in the European Society of Gastrointestinal and Abdominal Radiology guideline based on a combination of size (<5 mm, 5–8 mm, and ≥9 mm) and malignant features (internal heterogeneity, irregular margins, and round shape), and the guideline recommends the same criteria be used for LLNs.9 However, multiple studies recently reported short-axis size (≥7.0 mm) to be the main predictor for LLR, which is associated with 5-year LLR rates up to 20% after (C)RT and TME only.9–11 This short-axis cutoff (≥7.0 mm) is implemented in the current Dutch guidelines, but other radiologic parameters, such as malignant features and the presence of multiple enlarged LLNs, have not yet been incorporated.12–14

Only a few studies have previously investigated the presence of malignant features in LLNs. Ogura et al15 found no association between malignant features and LLR rates, whereas Kroon et al16 found an increased distant metastasis (DM) rate in the presence of malignant features. It is also unclear which malignant features are important; the study by Kroon et al16 used only irregular margins and internal heterogeneity, whereas other studies have also used loss of fatty hilum and round shape to describe potential malignant lymph nodes.12,17 Moreover, the focus of LLN research has thus far been on the largest LLN because it is thought to have the most malignant potential, but the potential impact of multiple enlarged LLNs has not yet been assessed. We previously reported that the presence of any malignant feature was a predictor for LLR.11 The objective of this study was to perform an in-depth analysis of malignant features on primary or restaging MRI and the number of LLNs to determine their prognostic impact in addition to short-axis size.