Additional Surgery After Endoscopic Resection for T1 Colorectal Cancer in Older Adults: From Oncologic Benefit to Net Clinical Benefit

Abstract

Following the endoscopic resection of pT1 colorectal cancer, additional surgery with lymph node dissection is recommended for patients with pathological features associated with an increased risk of lymph node metastasis (LNM) [1]. This strategy is supported by substantial evidence showing that established pathological risk factors identify patients at increased oncological risk and that additional surgery is associated with improved long-term outcomes, including disease-specific survival (DSS) and overall survival (OS), in appropriately selected patients [2]. A recent meta-analysis by Chen et al. demonstrated that surgical resection was associated with improved DSS compared to local resection, although this benefit became more apparent only after prolonged follow-up [3]. The 5-year DSS rates were similar between the two groups (96.7% vs. 98.3%), whereas a more substantial difference emerged after 10 years (86.9% vs. 97.1%). In contrast, the corresponding differences in OS remained relatively modest (86.3% vs. 94.5% at 5 years and 72.9% vs. 84.4% at 10 years), reflecting the influence of competing mortality. These findings suggest that the oncological benefits of surgery accumulate gradually over time and may not be fully realized in older adults with limited life expectancy and a high risk of competing mortality. In a real-world setting, among patients aged ≥ 80 years with high-risk pT1 colorectal cancer, there was no significant difference in OS between patients managed with endoscopic resection alone and those who underwent additional surgery when treatment was selected based on the patients' overall condition and surgical tolerability [4]. This raises an important clinical question: How should additional surgery be considered in older adults after endoscopic resection of pT1 colorectal cancer?

Ichimasa et al. made two important contributions to this discussion [5]. First, they reframed the clinical questions. Rather than asking whether additional surgery is indicated according to pathological criteria alone, they asked whether additional surgery is likely to provide meaningful benefits for an individual older patient. Second, they systematically synthesized the available evidence regarding the potential benefits and harms of additional surgery. Although experienced clinicians have long recognized the importance of considering the patient's overall condition when making treatment decisions, evidence supporting these decisions remains scattered across multiple domains. Ichimasa et al. have provided clinicians with a practical framework for weighing the potential benefits and harms of additional surgery in older adults and identified key areas where further evidence is needed.

In practice, decision-making requires clinicians to balance multiple, often competing factors, including oncological risk, physiological reserve, comorbidities, life expectancy, and patient preferences, when determining whether the expected benefit of additional surgery outweighs its potential harm to older patients. However, translating these considerations into individualized treatment decisions is often challenging in clinical practice. One practical way to approach this complexity is to consider the decision-making process across three broad domains: tumor-related, patient-related, and surgery-related factors.

Tumor-related factors represent the first domain. Current guidelines recommend additional surgery based on the presence or absence of pathological risk factors for LNM [1]. However, pathological risk extends beyond this binary classification and varies substantially with the type and number of pathological risk factors. In a nationwide multicenter study [6], the LNM rate increased from 6.1% in tumors with a submucosal invasion depth of 1000–1999 μm to 11.0% in those with an invasion depth of ≥ 2000 μm. Similarly, substantial differences were observed among individual pathological risk factors, with LNM rates of 17.2% for high-grade tumor budding (BD2/3), 13.7% for lymphovascular invasion, and 21.7% for grade 3 histology. Furthermore, the risk of LNM varies according to the number and combination of pathological risk factors [7]. Although risk estimation based on multiple individual pathological factors and their combinations can be challenging, available nomograms may facilitate quantitative estimation of LNM risk as an absolute probability [6], thereby supporting individualized surgical decision-making.

Patient-related factors represent the second domain. In older adults, patient assessments should consider both short- and long-term outcomes. In the short term, older adults are at an increased risk of postoperative complications, which may result in prolonged functional decline and loss of independence, even after successful surgery. In the long term, competing mortality may prevent patients from surviving long enough to realize the oncological benefits of additional surgery. As older adults are highly heterogeneous, chronological age alone is insufficient for treatment decision-making. Instead, biological age, including frailty, comorbidities, functional status, and life expectancy, should be considered. Comprehensive geriatric assessment offers a practical framework for identifying patients most likely to tolerate surgery and derive meaningful benefits [8].

Surgery-related factors constitute the third domain. Surgical burden is not uniform and depends on the extent and location of the planned resection. While minimally invasive colectomy is generally associated with favorable postoperative recovery, rectal resection carries unique risks beyond perioperative morbidity, including bowel, urinary, and sexual dysfunction as well as temporary or permanent stoma formation. These functional consequences may substantially impair the quality of life and independence, particularly in older adults.

Considering treatment decisions through the framework of tumor-, patient-, and surgery-related factors may facilitate a more structured approach to individualized care. However, the relative weight assigned to each domain ultimately depends on individual patient values and preferences. While some patients may prioritize long-term cancer control, others may place greater value on preserving independence, avoiding functional impairment, or maintaining quality of life. Therefore, shared decision-making is essential to ensure that treatment decisions align not only with the best available evidence but also with what matters most to an individual patient.

Beyond optimizing decision-making after endoscopic resection, improving initial endoscopic treatment may further reduce the need for additional surgery. Achieving a secure negative vertical margin may be particularly important in older adults. First, a positive vertical margin constitutes an absolute indication for additional surgery under the Japanese guidelines and should therefore be avoided whenever possible [1]. Second, secure deep resection allows the confident identification of patients in whom deep submucosal invasion is the only pathological risk factor. Because these patients appear to have a relatively low risk of LNM [9], achieving an adequate deep resection margin may help avoid unnecessary additional surgery. In this context, further refinement of endoscopic techniques capable of secure deep resection, including endoscopic muscularis dissection and full-thickness resection, may expand organ-preserving treatment strategies for carefully selected older adults [10].

In conclusion, by systematically synthesizing the available evidence, Ichimasa et al. provided a valuable framework for individualized decision-making in older adults with pT1 colorectal cancer after endoscopic resection [5]. The central question is not whether additional surgery is indicated but whether it is truly worthwhile for the individual patient.

H.T. conceived, drafted, and revised the manuscript. S.O. critically revised the manuscript and approved the final version.

The authors have nothing to report.

The authors declare no conflicts of interest.

This article is linked to Ichimasa et al. To view the linked article, visit https://doi.org/10.1111/den.70222.

The authors have nothing to report.