Journal of Surgical Oncology最新文献

Background

We assessed causes of death in patients with early-onset colorectal cancer (EOCRC) and factors associated with non-CRC-related deaths.

Methods

SEER database was screened between 2000 and 2020 for patients with EOCRC. Causes of death were classified into CRC-related and non-CRC-related, and stratified by demographics, disease location, and stage. The main study outcome was the cause of death in EOCRC.

Results

A total of 67 353 patients (53.9% male) had EOCRC. In total, 13.2% of 25 441 deaths were unrelated to CRC. The most common cause of non-CRC-related deaths was medical conditions (36.1%), mainly heart disease (16.6%). CRC was more often the cause of death in patients aged < 30 years, female, and stages III and IV disease; whereras medical conditions accounted for more deaths in patients aged 40–50 years, males, and Black. Other primary cancers were more often the cause of death in patients aged 45–50 years, female patients, and Asian patients. Death due to causes other than CRC was significantly more likely when surgery for CRC was performed (OR: 2.35; p = 0.028) and when CRC was one of multiple primary cancers (OR: 3.7; p < 0.001).

Conclusions

Most common causes of non-CRC-related deaths were medical conditions and other primary cancers accounting for more deaths in patients aged 40–50 years, males, Black and Asian patients, and with early-stage CRC.

Background

Total mesorectal excision (TME) is the standard surgical treatment for rectal cancer. Transanal local excision (TLE) after neoadjuvant chemoradiotherapy (nCRT) is an organ-preserving option, avoiding morbidity of TME. This study compared TLE versus TME following nCRT.

Methods

We searched Medline, PubMed, Embase, Web of Science, Scopus, Cochrane databases, Google Scholar, and CINHAL to 30 April 2025. Eligible studies were adults with nonmetastatic mid or low rectal cancer treated with nCRT followed by TLE or TME. Outcomes included local recurrence, disease free survival (DFS), overall survival (OS), and postoperative complications. Risk of bias was assessed using Cochrane RoB 2 for randomized trials and Newcastle-Ottawa Scale (NOS) for cohorts.

Results

Nineteen studies were included. TLE was associated with higher local recurrence in cohorts (RR = 1.823; 95% CI = 1.222–2.720; p = 0.003), but no difference in RCTs (RR = 1.248; 95% CI = 0.618–2.518; p = 0.537). DFS (HR = 1.121; p = 0.174) and OS (HR = 1.032; p = 0.830) did not differ. Postoperative morbidity was lower after TLE (RR = 0.429; p = 0.005).

Conclusion

Strengths include robust search, study quality and number of patients, while heterogeneity in nCRT protocol, follow up, and complication reporting are limitations. Higher recurrence in TLE in cohorts but not in RCTs suggests safety of TLE when strict selection criteria are applied.

Registration: PROSPERO CRD420251076513.

Objective

Endoscopic submucosal dissection (ESD) is increasingly preferred as an alternative to surgery and as an organ-preserving method for the treatment of rectal lesions. However, its impact on bowel function—particularly the development of low anterior resection syndrome (LARS)—remains poorly understood. This study aimed to evaluate the incidence of LARS following rectal ESD and to identify associated clinical and pathological factors.

Methods

Data from 118 patients who underwent rectal ESD between January 2018 and December 2024 were retrospectively analyzed. Demographic characteristics, lesion location and size, histopathological findings, and LARS scores were recorded. All procedures were performed by the same experienced colorectal surgeon. For functional assessment, patients were contacted by telephone, and the LARS score was administered.

Results

Of the patients included in the study, 60.9% were male, with a mean age of 68.2 ± 11.0 years. In total, 93.2% of lesions were located in the rectum, while 6.8% were located in the anal canal or anorectal junction. The mean lesion size was 5.04 ± 3.19 cm. Lymphovascular invasion was detected in four patients. Minor LARS was observed in 3.4% (n = 4) of cases, and no cases of major LARS were identified. All patients with LARS had adenomas located in the lower rectum containing high-grade dysplasia.

Conclusion

The incidence of LARS after rectal ESD is low and limited to minor symptoms. Lesion location, size, and dysplasia grade may influence LARS development. These findings support ESD as a functionally safe treatment approach. Further validation in larger, prospective studies is warranted.

Trial Registration

This study was a retrospective observational analysis.

Background and Objectives

The indication for sublobar resection is determined based on radiologic findings, but some cases exhibit radiology-pathology discordance. This study aimed to examine the impact of histologic subtypes on radiology-pathology discordance and their preoperative predictability.

Methods

We reviewed 585 patients with clinical stage IA adenocarcinoma and examined the relationship between radiology-pathology discordance and histologic characteristics, focusing on high-grade components: solid (SOL) or micropapillary (MIP). The predictive ability of radiologic or cytopathologic examinations for those subtypes was evaluated.

Results

Radiology-pathology discordance was found in 148 (25.2%) patients and was significantly associated with the presence of histologic high-grade components, with 71.9% and 70.4% of patients with upstaged lymph node and pleural invasion statuses having high-grade components. The preoperative prediction of high-grade components varied between subtypes, and radiographically pure-solid appearance and high maximum standardized uptake value were independent predictors of the SOL subtype, but not MIP. Among pre- or intraoperative cytopathologic examinations, intraoperative touch imprint cytology exhibited superior detection ability for MIP component.

Conclusions

Histologic high-grade components are highly associated with radiology-pathology discordance in early-stage lung adenocarcinoma. Radiologic assessment would be beneficial for predicting the SOL subtype, but not MIP. Alternatively, intraoperative cytologic assessment would complement the detection of MIP subtype.

Background

We assessed predictors of overall (OS) and cancer-specific survival (CSS) in stage I colorectal cancer (CRC).

Methods

Retrospective analysis of patients with stage I colon or rectal adenocarcinomas from the SEER database (2010–2020) Survival was assessed using Kaplan-Meier statistics and multivariable Cox regression analyses. The primary outcomes were 5-year OS and CSS.

Results

40,001 patients (51.3% male; mean age: 65.1 ± 12.6 years) were included. Colon and rectal cancers accounted for 75.8% and 24.2%, respectively. Five-year OS and CSS were 83.1% (95% CI: 82.6–83.5%) and 93.2% (95% CI: 92.9–93.5%), respectively. Factors independently associated with worse OS were age (HR: 1.07; p < 0.001), male sex (HR:1.48; p < 0.001), Black race (HR: 1.25; p < 0.001), single, divorced, or widowed status (HR: 1.49, 1.46, and 1.43; p < 0.001), tumor size (HR: 1.001; p = 0.008), poorly differentiated carcinomas (HR: 1.32; p < 0.001), undifferentiated carcinomas (HR:1.44; p = 0.026), perineural invasion (HR: 1.84; p < 0.001), elevated CEA levels (HR: 1.68; p < 0.001), and systemic therapy (neoadjuvant: HR: 1.3; p = 0.032, adjuvant: HR: 2.2; p < 0.001, both: HR: 1.97; p < 0.001). Factors independently associated with worse CSS were age (HR: 1.05; p < 0.001), male sex (HR: 1.32; p < 0.001), Black race (HR: 1.43; p < 0.001), marital status (HR: 1.44, 1.28, and 1.68; p < 0.001), tumor size (HR: 1.003; p < 0.001), poorly differentiated carcinomas (HR: 1.77; p < 0.001), perineural invasion (HR: 2.29; p < 0.001), elevated CEA levels (HR: 2.24; p < 0.001), and systemic therapy (neoadjuvant: HR: 2.53; p = 0.032, adjuvant: HR: 4.22; p < 0.001, both: HR: 3.83; p < 0.001).

Conclusions

Although patients with stage I CRC had excellent survival, single, older, Black, male patients with large, high-grade tumors associated with perineural invasion and elevated CEA levels had a higher mortality risk.