Thai Bulletin of Pharmaceutical Sciences

IMPACT OF OBESITY ON CANCER: FROM PATHOGENESIS TO PHARMACOTHERAPEUTIC CHALLENGES AND MANAGEMENT OF TREATMENT-INDUCED WEIGHT GAIN

Nutthada Areepium — Mon, 27 Apr 2026 00:00:00 +0700

Obesity is a critical risk factor influencing both cancer incidence and prognosis. This narrative review aims to provide readers with a comprehensive understanding of the obesity–cancer relationship across three key dimensions: (1) the pathogenic mechanisms by which obesity promotes tumor development and progression, (2) the impact of obesity on anticancer drug pharmacokinetics and the associated challenges in dose optimization, and (3) proactive pharmaceutical care strategies for obese cancer patients, including the management of treatment-induced weight gain. Pathogenetically, obesity promotes tumor progression through three principal mechanisms: chronic low-grade inflammation, hyperinsulinemia-driven dysregulation of the Insulin/IGF-1 signaling axis leading to aberrant cancer cell proliferation, and alterations in sex hormone metabolism that favor the growth of hormone-sensitive cancers. Regarding pharmacotherapy, obesity significantly alters the pharmacokinetics of anticancer agents, particularly by increasing the volume of distribution for lipophilic drugs and impairing drug clearance, thereby heightening the risk of severe adverse reactions including cardiotoxicity and peripheral neuropathy. Clinical evidence supports dosing based on actual body weight to prevent subtherapeutic exposure. Furthermore, this review highlights sarcopenic obesity, frequently exacerbated by corticosteroids and hormonal therapies, as a clinically underrecognized concern. Pharmacists play a pivotal role in proactive pharmaceutical care by optimizing dose individualization, monitoring drug safety, and managing metabolic complications to enhance therapeutic efficacy and patient quality of life.

PREDICTION OF NEUTROPENIA IN CANCER PATIENTS RECEIVING CHEMOTHERAPY USING MACHINE LEARNING TECHNIQUE

Nutcha Sutthisanseth, Nattee Phornprapa, Lawan Sratthaphut — Mon, 27 Apr 2026 00:00:00 +0700

Neutropenia is a common complication in patients with cancer receiving chemotherapy. Without appropriate prevention or management, this condition can lead to severe complications and may be life-threatening. Previous studies revealed that machine learning techniques can be effectively predicted chemotherapy-related complications. The objective of this study was to develop and compare the optimal machine learning models for predicting neutropenia in cancer patients receiving chemotherapy. A retrospective dataset of 511 patients was collected from the hospital information system of Ratchaphiphat Hospital between January 1, 2022, and March 31, 2024. Data were derived from medical records, physical examination, and laboratory information, and were further screened to identify relevant variables prior to model development. The dataset was split into two groups; a training set (80%) and a testing set (20%). The training dataset was balanced using the SMOTE-NC method. Five machine learning techniques: Logistic Regression (LR), Random Forest (RF), Adaptive Boosting (AdaBoost), Support Vector Machine (SVM), and Artificial Neural Network (ANN) were developed using optimal hyperparameters. Model selection was based on accuracy, precision, recall, and F1-measure. Model performance was evaluated using the testing dataset, based on its sensitivity, specificity, and AUC. The results demonstrated that the best-performing models were those developed using RF and AdaBoost, with sensitivity, specificity, and AUC values of 100%, 98.21%, and 1.000, respectively. Therefore, RF and AdaBoost, which are ensemble learning techniques, were the most suitable models for predicting neutropenia.