Development of Gene Expression-Based Random Forest Model for Predicting Neoadjuvant Chemotherapy Response in Triple-Negative Breast Cancer

Using machine learning models to predict synchronous genitourinary cancers among gastrointestinal stromal tumor patients

Objectives

Gastrointestinal stromal tumors (GISTs) can occur synchronously with other neoplasms, including the genitourinary (GU) system. Machine learning (ML) may be a valuable tool in predicting synchronous GU tumors in GIST patients, and thus improving prognosis. This study aims to evaluate the use of ML algorithms to predict synchronous GU tumors among GIST patients in a specialist research center in Saudi Arabia.

Materials and Methods

We analyzed data from all patients with histopathologically confirmed GIST at our facility from 2003 to 2020. Patient files were reviewed for the presence of renal cell carcinoma, adrenal tumors, or other GU cancers. Three supervised ML algorithms were used: logistic regression, XGBoost Regressor, and random forests (RFs). A set of variables, including independent attributes, was entered into the models.

Results

A total of 170 patients were included in the study, with 58.8% (n = 100) being male. The median age was 57 (range: 9-91) years. The majority of GISTs were gastric (60%, n = 102) with a spindle cell histology. The most common stage at diagnosis was T2 (27.6%, n = 47) and N0 (20%, n = 34). Six patients (3.5%) had synchronous GU tumors. The RF model achieved the highest accuracy with 97.1%.

Conclusion

Our study suggests that the RF model is an effective tool for predicting synchronous GU tumors in GIST patients. Larger multicenter studies, utilizing more powerful algorithms such as deep learning and other artificial intelligence subsets, are necessary to further refine and improve these predictions.

Understanding the mechanistic pathways and clinical aspects associated with protein and gene based biomarkers in breast cancer

Cancer is one of the most widespread and severe diseases with a huge mortality rate. In recent years, the second-leading mortality rate of any cancer globally has been breast cancer, which is one of the most common and deadly cancers found in women. Detecting breast cancer in its initial stages simplifies treatment, decreases death risk, and recovers survival rates for patients. The death rate for breast cancer has risen to 0.024 % in some regions. Sensitive and accurate technologies are required for the preclinical detection of BC at an initial stage. Biomarkers play a very crucial role in the early identification as well as diagnosis of women with breast cancer. Currently, a wide variety of cancer biomarkers have been discovered for the diagnosis of cancer. For the identification of these biomarkers from serum or other body fluids at physiological amounts, many detection methods have been developed. In the case of breast cancer, biomarkers are especially helpful in discovering those who are more likely to develop the disease, determining prognosis at the time of initial diagnosis and choosing the best systemic therapy. In this study we have compiled various clinical aspects and signaling pathways associated with protein-based biomarkers and gene-based biomarkers.

Triple-Negative Breast Cancer and Predictive Markers of Response to Neoadjuvant Chemotherapy: A Systematic Review

Around 40-50% of all triple-negative breast cancer (TNBC) patients achieve a pathological complete response (pCR) after treatment with neoadjuvant chemotherapy (NAC). The identification of biomarkers predicting the response to NAC could be helpful for personalized treatment. This systematic review provides an overview of putative biomarkers at baseline that are predictive for a pCR following NAC. Embase, Medline and Web of Science were searched for articles published between January 2010 and August 2022. The articles had to meet the following criteria: patients with primary invasive TNBC without distant metastases and patients must have received NAC. In total, 2045 articles were screened by two reviewers resulting in the inclusion of 92 articles. Overall, the most frequently reported biomarkers associated with a pCR were a high expression of Ki-67, an expression of PD-L1 and the abundance of tumor-infiltrating lymphocytes, particularly CD8+ T cells, and corresponding immune gene signatures. In addition, our review reveals proteomic, genomic and transcriptomic markers that relate to cancer cells, the tumor microenvironment and the peripheral blood, which also affect chemo-sensitivity. We conclude that a prediction model based on a combination of tumor and immune markers is likely to better stratify TNBC patients with respect to NAC response.