Impact of time from diagnosis to endoscopic resection on clinical outcomes in early gastric cancer
This retrospective cohort enrolled 1,646 consecutive patients diagnosed with early gastric cancer (EGC) who underwent curative endoscopic resection (ER) between 2010 and 2023. The report describes patient ascertainment, baseline characteristics, and follow-up used to evaluate the relationship between time from diagnosis to ER and clinical and pathological outcomes. Affiliations and publication details are provided in the source.
Time from diagnosis to ER was the exposure of interest. The median waiting time reported was 15 days with an interquartile range of 11–23 days. Only 2.3% of patients waited more than 90 days for ER. The study modeled waiting time both categorically (<30 days vs ≥30 days) and continuously (per 10-day increment), and also applied restricted cubic spline (RCS) modeling to evaluate nonlinear associations up to a maximum window of 180 days.
Primary endpoints were overall survival (OS) and disease-free survival (DFS). Secondary endpoints included positive resection margins, pathological upstaging, curative resection rate, and requirement for additional gastrectomy after ER.
Statistical approaches included Kaplan–Meier survival curves and multivariable Cox regression for time-to-event outcomes, and logistic regression for binary pathological endpoints. Time was incorporated both as a categorical and continuous variable; RCS modeling assessed potential nonlinear relationships. Sensitivity analyses were prespecified, including exclusion of high-risk deep submucosal (SM2) lesions. Multivariable models adjusted for baseline covariates where there were marginal imbalances (the source notes small differences in age and family history).
Across linear, categorical, and restricted cubic spline analyses, no independent adverse association was observed between preoperative delays (within the 180-day window evaluated) and OS or DFS. Reported p values for associations with primary endpoints were greater than 0.05, indicating no statistically significant relationship between waiting time and survival in this cohort after multivariable adjustment.
The source emphasizes that sensitivity analysis excluding SM2 lesions produced consistent neutral survival findings, supporting the main observation that delay—within the studied range—was not linked to worse survival outcomes.
Secondary pathological endpoints, including positive resection margins, pathological upstaging, curative resection rate, and need for additional gastrectomy, likewise showed no independent adverse associations with delays up to 180 days in both continuous and categorical models. The study reports all corresponding tests as non-significant (all p > 0.05) after adjustment for potential confounders identified in baseline comparisons.
A predefined sensitivity analysis excluded lesions classified as deep submucosal (SM2), which are considered higher risk. Excluding these SM2 lesions did not change the neutral survival results, indicating the findings were robust to this subgroup exclusion. The authors note, however, that only a small proportion of the cohort experienced long waits (for example, >90 days), producing marked sample-size imbalance across waiting-time strata and limiting power to detect effects among long-delay subgroups.
The authors conclude that, in this single-cohort retrospective series of 1,646 EGC patients treated with curative ER, preoperative waiting up to 180 days was not associated with worse oncological or pathological outcomes. They state these are exploratory results that support the possibility of a reasonable up-to-6-month deferral for preoperative assessment when clinically necessary.
Key limitations noted in the source include the retrospective design and the small number of patients with long preoperative waits, which creates imbalance and limits the ability to definitively validate the absence of harm for long delays. The authors recommend large prospective multicenter trials to confirm these findings and to more precisely evaluate outcomes in patients with prolonged preoperative intervals.
Overall, this study adds cohort-level evidence that short-to-moderate delays to ER (median 15 days, with analysis up to 180 days) were not associated with worse clinical outcomes in the reported population, while explicitly acknowledging the need for larger prospective validation.