ESMO Congress 2024 | OncologyPRO

Abstract 311P

Background

The American Joint Committee on Cancer (AJCC) currently categorizes breast cancer tumor sizes based on empirical, human-defined criteria. This method may not fully capture the complexities of tumor biology or its impact on patient outcomes. Advances in big data technologies now allow for a more nuanced understanding of the relationship between tumor size and breast cancer prognosis. This study aims to leverage big data analytics to refine AJCC's classification of breast cancer tumor sizes, potentially leading to more precise staging criteria and personalized treatment recommendations.

Methods

We analyzed data from the Surveillance, Epidemiology, and End Results (SEER) program for the periods 2004-2015 and 2018-2020, encompassing 88,560 and 35,515 breast cancer patients, respectively. Patients were strictly categorized within T1-T4, N0, and M0 stages. Hierarchical clustering was used to classify tumors into distinct stages based on size and patient survival data, creating both three-category and four-category classifications. The effectiveness of these models was validated through Kaplan-Meier survival analysis.

Results

The newly defined stages are as follows: Stage I includes tumor sizes from 1 to 14 mm (5-year survival rate: 91%, 10-year: 78%, 15-year: 64%), Stage II from 15 to 34 mm (5-year: 84%, 10-year: 69%, 15-year: 56%), Stage III from 35 to 120 mm (5-year: 73%, 10-year: 57%, 15-year: 47%), and Stage IV for tumors exceeding 120 mm (5-year: 63%, 10-year: 44%, 15-year: 35%). Compared to traditional staging, the Log-rank test showed significant differences between survival curves (P < 0.05), indicating that the new staging system more accurately reflects differences in patient prognosis.

Conclusions

This study highlights the potential of big data to refine oncological practices and suggests pathways for further research into optimizing tumor staging criteria.