Abstract
Background Chemotherapy-induced cytopenia is the most frequent side effect of chemoradiotherapy in glioblastoma patients which may lead to reduced delivery of treatment. This study aims to develop a predictive model that is able to forecast the cytopenia induced by temozolomide (TMZ) during concomitant chemoradiotherapy. Methods Medical records of 128 patients with newly diagnosed glioblastoma were evaluated to extract the baseline complete blood test before and during the six weeks of chemoradiotherapy to create a dataset for the development of ML models. Using the constructed dataset, different ML algorithms were trained and tested. Results Our proposed algorithm achieved accuracies of 85.6%, 88.7%, and 89.3% in predicting thrombocytopenia, lymphopenia, and neutropenia, respectively. Conclusions The algorithm designed and developed in this study, called PrACTiC, showed promising results in the accurate prediction of thrombocytopenia, neutropenia, and lymphopenia induced by TMZ in glioblastoma patients. PrACTiC can provide valuable insight for physicians and help them to make the necessary treatment modifications and prevent the toxicities.
Highlights
Glioblastoma is the most common primary brain tumor in adults [1, 2]. e standard treatment for newly diagnosed glioblastoma is the maximal safe resection followed by concomitant chemoradiotherapy with temozolomide (TMZ) followed by adjuvant TMZ [3, 4]
Patients receiving TMZ are at risk of hematologic toxicity during therapy
Our results show that the accuracy of PrACTiC achieves 85.6%, 88.7%, and 89.3% with true positives of 92%, 78%, and 89% in predicting thrombocytopenia, lymphopenia, and neutropenia, respectively
Summary
Glioblastoma is the most common primary brain tumor in adults [1, 2]. e standard treatment for newly diagnosed glioblastoma is the maximal safe resection followed by concomitant chemoradiotherapy with temozolomide (TMZ) followed by adjuvant TMZ [3, 4]. Patients receiving TMZ are at risk of hematologic toxicity (thrombocytopenia, lymphopenia, and neutropenia) during therapy. Chemotherapy-induced cytopenia is the most frequent side effect of chemoradiotherapy in glioblastoma patients which may lead to reduced delivery of treatment. Is study aims to develop a predictive model that is able to forecast the cytopenia induced by temozolomide (TMZ) during concomitant chemoradiotherapy. Medical records of 128 patients with newly diagnosed glioblastoma were evaluated to extract the baseline complete blood test before and during the six weeks of chemoradiotherapy to create a dataset for the development of ML models. E algorithm designed and developed in this study, called PrACTiC, showed promising results in the accurate prediction of thrombocytopenia, neutropenia, and lymphopenia induced by TMZ in glioblastoma patients. PrACTiC can provide valuable insight for physicians and help them to make the necessary treatment modifications and prevent the toxicities
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have