Ultrasound-based radiomics combined with immune status to predict sentinel lymph node metastasis in primary breast cancer

In the past few years, the axillary lymph node dissection technique has been steadily replaced by sentinel lymph node biopsy for treating and diagnosing breast cancer, thereby minimizing the complications and sequelae of the patients. Nevertheless, sentinel lymph node biopsy still presents limitations, such as high operation requirements, prolonged surgical duration, and adverse reactions to tracer agents. This study developed a novel non-invasive method to predict sentinel lymph node metastasis in breast cancer by analyzing the ultrasound imaging characteristics of the primary tumor, combined with the analysis of peripheral blood T-cell subsets that reflect the immune status of the body. The radiomic features analyzed in this study were extracted from preoperative ultrasound images of 199 solitary breast cancer patients, who were undergoing surgery and were pathologically diagnosed at the Yancheng First People's Hospital. All cases were randomly categorized in a 4:1 ratio to the training (n = 159) and validation (n = 40) cohorts. The extracted radiomics features were subjected to dimensional reduction with the help of the least absolute shrinkage and selection operator technique, resulting in the inclusion of 19 radiomics features. Four classifiers, including naïve Bayesian, logistic regression, classification decision tree, and support vector machine, were utilized to model the radiomics features, conventional ultrasound features, and peripheral blood T cell subsets in the training dataset, and validated using the validation dataset. The best-performing model was chosen for constructing the combined model. The radiomics model constructed using the logistic regression showed the best performance, with the training and validation cohorts showing areas under the curve (AUCs) of 0.77 and 0.68, respectively. The conventional ultrasound and peripheral blood T cell models constructed using the classification decision tree showed the best performance, wherein the training cohort presented AUCs of 0.71 and 0.81, respectively, while the validation cohort presented AUCs of 0.68 and 0.69, respectively. The combined model constructed by logistic regression showed AUCs of 0.91 and 0.79 in the training and validation datasets, respectively. The resulting combined model can be considered a simple, non-invasive method with strong reproducibility and clinical significance. Thus, it can be utilized to predict sentinel lymph node metastasis in breast cancer. Furthermore, the combined model can be effectively used to guide clinical decisions related to the selection of surgical procedures in breast surgery.

Relationship of body mass index with prognosis in breast cancer patients treated with adjuvant radiotherapy and chemotherapy

BACKGROUND

The aim of this study was to investigate the relationship of body mass index with overall and progression-free survival as well as other prognostic factors of breast cancer in patients with non-metastatic breast cancer.

MATERIALS AND METHODS

We retrospectively reviewed 456 patients diagnosed with breast cancer in the Radiation Oncology department of Kayseri Teaching Hospital between 2005 and 2013. We investigated relationship of body mass index with prognosis and other prognostic factors.

RESULTS

The study included 456 patients (447 women and 9 men). Mean age at presentation was 55.6 years. Of the cases, 96.9% underwent modified radical mastectomy and 95.0% received chemotherapy, while 82.4% received radiotherapy and 60.0% were given hormone therapy. Body mass index was >25 mg/kg(2) in 343 cases. Five- and 10-years overall survival rates were 77% and 58% whereas progression-free survival rates were 65% and 49%, respectively. In univariate analyses, factors including stage (p=0.046), tumor diameter (p=0.001), lymph node metastasis (p=0.006) and body mass index (p=0.030) were found to be significantly associated with overall survival, while perinodal involvement was found to be significantly associated with progression-free survival (p=0.018). In multivariate analysis, stage (p=0.032; OR: 3.8; 95% CI: 1.1-13), tumor diameter (p<0.000; OR: 0.0; 95% CI: 0.0-0.3), lymph node metastasis (p=0.005; OR: 0.0; 95% CI: 0.0-0.5) and BMI (p=0.027; OR: 0.02; 95% CI: 0.0-0.8) remained as significantly associated with OS.

CONCLUSIONS

In our study, it was seen that overall survival time was shorter in underweight and obese patients when compared to normal weight patients.

A systemic analysis on pemetrexed in treating patients with breast cancer

BACKGROUND

This systemic analysis was conducted to evaluate the efficacy and safety of pemetrexed based chemotherapy in treating patients with metastatic breast cancer as first or second line chemotherapy.

METHODS

Clinical studies evaluating the efficacy and safety of pemetrexed based regimens on response and safety for patients with breast cancer were identified using a predefined search strategy. Pooled response rate (RR) of treatment were calculated.

RESULTS

In first line pemetrexed based regimens, 10 clinical studies which including 513 patients with advanced breast cancer were considered eligible for inclusion. For second line pemetrexed based chemotherapy, 5 clinical studies which including 281 patients with advanced breast cancer were considered eligible. Systemic analysis suggested that, in all patients, pooled RR was 32.6% (167/513) in pemetrexed based first line regimens, and 13.9 % (39/281) in pemetrexed based second line regimens. Major adverse effects were neutropenia, leukopenia, fatigue, and anemia in pemetrexed based first line treatment; and lymphopenia, neutropenia, leukopenia, as well as anemia in second line chemotherapy. One treatment related death occurred with pemetrexed based second line treatment.

CONCLUSION

This systemic analysis suggests that pemetrexed based first line regimens are associated with a reasonable response rate and acceptable toxicity, however with low response rate for treating patients with metastatic breast cancer when is used in the second line.