Development and testing of a random forest-based machine learning model for predicting events among breast cancer patients with a poor response to neoadjuvant chemotherapy

BACKGROUND

Breast cancer (BC) is the most common malignant tumor around the world. Timely detection of the tumor progression after treatment could improve the survival outcome of patients. This study aimed to develop machine learning models to predict events (defined as either (1) the first tumor relapse locally, regionally, or distantly; (2) a diagnosis of secondary malignant tumor; or (3) death because of any reason.) in BC patients post-treatment.

METHODS

The patients with the response of stable disease (SD) and progressive disease (PD) after neoadjuvant chemotherapy (NAC) were selected. The clinicopathological features and the survival data were recorded in 1 year and 5 years, respectively. Patients were randomly divided into the training set and test set in the ratio of 8:2. A random forest (RF) and a logistic regression were established in both of 1-year cohort and the 5-year cohort. The performance was compared between the two models. The models were validated using data from the Surveillance, Epidemiology, and End Results (SEER) database.

RESULTS

A total of 315 patients were included. In the 1-year cohort, 197 patients were divided into a training set while 87 were into a test set. The specificity, sensitivity, and AUC were 0.800, 0.833, and 0.810 in the RF model. And 0.520, 0.833, and 0.653 of the logistic regression. In the 5-year cohort, 132 patients were divided into the training set while 33 were into the test set. The specificity, sensitivity, and AUC were 0.882, 0.750, and 0.829 in the RF model. And 0.882, 0.688, and 0.752 of the logistic regression. In the external validation set, of the RF model, the specificity, sensitivity, and AUC were 0.765, 0.812, and 0.779. Of the logistics regression model, the specificity, sensitivity, and AUC were 0.833, 0.376, and 0.619.

CONCLUSION

The RF model has a good performance in predicting events among BC patients with SD and PD post-NAC. It may be beneficial to BC patients, assisting in detecting tumor recurrence.

The impact of postmastectomy radiotherapy on cT1-2N1 breast cancer patients with ypN0 after neoadjuvant chemotherapy: a retrospective study based on real-world data

BACKGROUND

The role of postmastectomy radiation therapy (PMRT) in clinical T1-2N1 breast cancer patients who achieve axillary pathological complete response (ypN0) after neoadjuvant chemotherapy (NAC) is controversial.

METHODS

Data from cT1-2N1 breast cancer patients who converted to ypN0 after NAC and subsequent surgery were retrospectively analyzed. Disease-free survival (DFS) and overall survival (OS) were estimated using the Kaplan‒Meier method. Univariate and multivariate Cox regression models were applied to investigate the correlations between clinical or pathological parameters and survival.

RESULTS

From 2012-2019, we identified 116 cases for analysis, including 31 (26.7%) who received PMRT and 85 (73.3%) who did not. At a median follow-up time of 56.4 months, the 5-year DFS and OS rates were 90.2% and 96.7% with PMRT and 93.7% and 97.3% without PMRT, respectively. PMRT did not affect either DFS (p = 0.234) or OS (p = 0.878). On multivariate analyses, no differences in DFS or OS between the two groups were detected, taking into consideration the following factors: age, molecular subtype, Ki67 index, cT stage, and in-breast pathologic complete response (DFS: HR 2.260; 95% CI 0.465-10.982; p = 0.312. OS: HR 1.400; 95% CI 0.138-14.202; p = 0.776). This nonsignificant difference was also consistent in subgroup analyses (all p > 0.05).

CONCLUSIONS

PMRT has limited ability to confer DFS or OS benefits for cT1-2N1 breast cancer patients who achieved axillary pathological complete response after NAC and total mastectomy. It is imperative to conduct prospective studies to investigate the safety and feasibility of omitting PMRT.

TRIAL REGISTRATION

This research was approved by the Ethics Committee of The First Affiliated Hospital of Chongqing Medical University (ID: No. 2021-442).

Comparison of the pCR Rate and DFS Among Breast Cancer Patients with Different Hormone Receptor and HER2 Statuses

Background

Recent studies have investigated the features of breast cancer (BC) with low human epidermal growth factor receptor 2 (HER2) expression or HER2-0 expression. However, the results were inconsistent. In this study, we investigated the differences in the pathological complete response (pCR) rate and disease-free survival (DFS) between HER2-low and HER2-0 BC patients and between subgroups.

Methods

HER2-negative BC patients who received neoadjuvant chemotherapy between January 2013 and December 2019 in our hospital were retrospectively reviewed. First, the pCR rate and DFS were compared between HER2-low and HER2-0 patients and among different hormone receptor (HR) and HER2 statuses. Subsequently, DFS was compared between different HER2 status populations with or without pCR. Finally, a Cox regression model was used to identify the prognostic factors.

Results

Overall, 693 patients were selected: 561 were HER2-low, and 132 were HER2-0. Between the two groups, there were significant differences in N stage (P = 0.008) and HR status (P = 0.007). No significant difference in the pCR rate (12.12% vs 14.39%, P = 0.468) or DFS was observed, independent of HR status. HR+/HER2-low patients had a significantly worse pCR rate (P < 0.001) and longer DFS (P < 0.001) than HR-/HER2-low or HER2-0 patients. In addition, a longer DFS was found in HER2-low patients versus HER2-0 patients among those who did not achieve pCR. Cox regression showed that N stage and HR status were prognostic factors in the overall and HER2-low populations, while no prognostic factor was found in the HER2-0 group.

Conclusion

This study suggested that HER2 status is not associated with the pCR rate or DFS. Longer DFS was found only among patients who did not achieve pCR in the HER2-low versus HER2-0 population. We speculated that the interaction of HR and HER2 might have played a crucial role in this process.