Utilizing tumor deposit count as a stratification criterion in revising TNM staging system for patients with colorectal cancer: a nomogram review study.

Background: Tumor deposit (TD) is an independent risk factor associated with recurrence or metastasis for patients with colorectal cancer (CRC). The scenario in which both TD and lymph node metastasis (LNM) are positive is not clearly illustrated by the current TNM staging system. Simply treating one TD as one or two LNMs by a weighting factor is inappropriate. The aim of this study was to evaluate the prognostic impact of TD counts and revise TNM staging by utilizing TD count as a stratification criterion in patients with CRC. Methods: All the cases diagnosed with CRC between 2010 and 2019 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Patients who met all inclusion criteria were grouped as TD=0, TD=1, TD=2, or TD≥3 based on their TD counts. To compare overall survival (OS) between groups, survival curves were plotted using Kaplan-Meier methods with log-rank tests. Utilizing TD count as a stratification criterion, the NM classification was regrouped. Using the Cox proportional hazards model, univariate and multivariate analyses were performed to identify significant factors in the revised TNM staging system. A nomogram was then created. The C-index, calibration plots, and receiver operating characteristic (ROC) curve were used to verify the model's accuracy. Results: A total of 60,145 patients who satisfied all inclusion requirements were ultimately included in the datasets for analysis. Among them, 6,092 (10.1%) had TDs, with only 1,384 (22.7%) staged as N1c. The 3-year OS rates were 74.6%, 50.4%, 39.8%, and 29.1% in the TD=0, TD=1, TD=2, and TD≥3 groups, respectively (p < 0.001), indicating that a greater TD count was linked to a worse prognosis. We introduced a novel stage, N2c, for patients with 10 or more LNMs. We found a striking overlap between the survival curves of this new subgroup and those of the M1a group, with p = 0.39. We observed that the survival trajectories of CRC patients with more than three TDs were similar to those of patients in the N2c group and the M1a group, with p = 0.15 and p = 0.42, respectively, which means that CRC patients with more than 10 LNMs or three TDs are equivalent to the presence of distant metastases. Utilizing TD count as a stratification criterion, we regrouped the NM classification. We finally assigned CRC patients with three or more TDs to the M1a. Finally, the revised TNM staging's prognostic significance was demonstrated by the nomogram and dynamic nomogram. The C-indices for OS prediction in the training cohort and validation cohort were 0.751 (95% CI: 0.748-0.754) and 0.752 (95% CI: 0.747-0.756), respectively. The ROC curve study revealed that both the training cohort and the validation group had areas under the curve (AUCs) of approximately 0.8 at 1, 3, and 5 years. The calibration curves demonstrated good agreement between actual observation and the nomogram-predicted Cancer-Specific Death (CSD) probability. The clinical application study demonstrated that the model outperformed the TNM staging approach in terms of net benefit increases and fewer needless procedures. Conclusions: TDs have significant predictive significance in CRC. The revised TNM staging using TD counts as a stratification criterion predicts survival more accurately than the current staging. CRC patients with more than 10 LNMs or three TDs have a level of malignancy comparable to the existence of distant metastases. The dynamic nomogram could assist medical practitioners in diagnosing, providing prognoses, and optimizing treatment strategies more quickly and effectively.