Conditional Cancer-Specific Survival for Inflammatory Breast Cancer: Analysis of SEER, 2010 to 2016

Dynamic Prediction of Rectal Cancer Relapse and Mortality Using a Landmarking-Based Machine Learning Model: A Multicenter Retrospective Study from the Italian Society of Surgical Oncology-Colorectal Cancer Network Collaborative Group

Background: Almost 30% of patients with rectal cancer (RC) who submit to comprehensive treatment experience relapse. Surveillance plays a leading role in early detection. The landmark approach provides a more flexible and dynamic framework for survival prediction. Objective: This large retrospective study aims to develop a machine learning algorithm to profile the patient prognosis, especially the risk and the onset of RC relapse after curative resection. Methods: A cohort of 2450 RC patients were analyzed using landmark analysis. Model A applied a classical cause-specific Cox approach with a landmarking approach, while Model B implemented a landmarking-based RSF (random survival forest) competing risk algorithm. The two models were compared in terms of predictive and interpretative ability. A bootstrapped validation strategy was employed to validate the model's performance and prevent overfitting. The best-performing hyperparameters were selected systematically, ensuring the model's robustness within the landmark approach. The study assessed these factors' importance and interactions using RSF and compared the predictive accuracy to that of the classical Cox model. Results: Model B outperformed Model A (mean C-index 0.95 vs. 0.78), capturing complex interactions and providing dynamic, individualized relapse predictions. Clinical factors influencing survival outcomes were identified across time with the landmark approach allowing for more accurate and timely predictions. Conclusions: The landmark approach offers an improvement over traditional methods in survival analysis. By accommodating time-dependent variables and the evolving nature of patient data, this approach provides a precise tool for profiling RC survival, thereby supporting more informed and dynamic clinical decision-making.

Conditional survival analysis and dynamic prediction of long-term survival in Merkel cell carcinoma patients

Background

Merkel cell carcinoma (MCC) is a rare type of invasive neuroendocrine skin malignancy with high mortality. However, with years of follow-up, what is the actual survival rate and how can we continually assess an individual's prognosis? The purpose of this study was to estimate conditional survival (CS) for MCC patients and establish a novel CS-based nomogram model.

Methods

This study collected MCC patients from the Surveillance, Epidemiology, and End Results (SEER) database and divided these patients into training and validation groups at the ratio of 7:3. CS refers to the probability of survival for a specific timeframe (y years), based on the patient's survival after the initial diagnosis (x years). Then, we attempted to describe the CS pattern of MCCs. The Least absolute shrinkage and selection operator (LASSO) regression was employed to screen predictive factors. The Multivariate Cox regression analysis was applied to demonstrate these predictors' effect on overall survival and establish a novel CS-based nomogram.

Results

A total of 3,843 MCC patients were extracted from the SEER database. Analysis of the CS revealed that the 7-year survival rate of MCC patients progressively increased with each subsequent year of survival. The rates progressed from an initial 41-50%, 61, 70, 78, 85%, and finally to 93%. And the improvement of survival rate was nonlinear. The LASSO regression identified five predictors including patient age, sex, AJCC stage, surgery and radiotherapy as predictors for CS-nomogram development. And this novel survival prediction model was successfully validated with good predictive performance.

Conclusion

CS of MCC patients was dynamic and increased with time since the initial diagnosis. Our newly established CS-based nomogram can provide a dynamic estimate of survival, which has implications for follow-up guidelines and survivorship planning, enabling clinicians to guide treatment for these patients better.