Given the high malignancy of glioma, it is crucial for clinicians to make precise and individualized prognostic assessments based on a comprehensive set of clinicopathological features. Traditional staging systems, such as the WHO grade, tend to incorporate only a limited number of pathological attributes, but overlook other critical clinical and treatment-related data, which in turn restricts their predictive power. Therefore, there exists an imperative need to develop a prediction system that can more accurately forecast the prognosis of glioma patients by integrating these diverse factors. A nomogram serves as a graphical statistical tool that facilitates the calculation of probable outcomes through the incorporation of multiple variables into a single model. It has gained widespread acceptance due to its ability to holistically consider a range of elements including both clinical pathology and patient demographics characteristics14. This is the driving force behind our focus on constructing a nomogram specifically designed for predicting OS in glioma patients.
By leveraging the advantages of the CGGA database, we developed a prediction model based on patients with shorter survival time (less than 360 days) and longer survival time (greater than 1800 days). The LASSO-COX dimension reduction analysis was utilized to select an optimum prognostic signature comprising the most representative gene markers for identifying the 8-gene signature in short-term survival glioma patients. Subsequently, a risk score based nomogram was devised to forecast the probability of early mortality in individuals diagnosed with glioma. The nomogram, incorporating the risk score, PRS status, WHO grade, age, chemotherapy status and 1p/19q codeletion status, effectively predicts patients at a high risk of early death.
According to the nomogram, the risk score is the most significant factor affecting patient prognosis. The risk score incorporates 8 genes, including EN1, HOXD13, HOXD9, IGF2BP3, IGFBP2, NNMT, PRLHR, and RP1-293L6.1, which have been proven to impact the prognosis of glioma patients15–22. The second important factor is age. Multiple studies have emphasized that age serves as an independent and critical prognostic factor in glioma; Even when the histopathological diagnosis is identical, differing patient ages can significantly impact glioma prognosis. Thus, it is proposed that individualized treatment modalities be based on meticulous age considerations23.
The WHO grade for gliomas is a critically important classification, where grade IV (such as glioblastoma) have the poorest prognosis, and grade II tumors generally fare better than grade III24,25. The WHO grade classification is currently one of the most significant means for predicting the prognosis in glioma patients. The likelihood of recurrence of malignant tumors will have a significant impact on the prognosis of patients. Due to the infiltrative nature of glioma, recurrence is typically inevitable for the tumor, particularly with glioblastoma, where the rate of recurrence approximates nearly 100%26–28. This invasiveness plays a critical role in tumor progression16. Codeletion of 1p/19q has also been demonstrated to be a prognostic marker, patients harboring 1p/19q codeletion exhibit notably longer survival compared to those without the alteration, furthermore, the demonstration of the 1p/19q status as the main factor associated with sensitivity to PCV (procarbazine, lomustine, and vincristine) chemotherapy has gained increased clinical significance29,30. Chemotherapy plays a significant role in glioma prognosis, aligning with prior research findings31.
This study has some limitations. Firstly, both the training group and the validation group included only Chinese patients. It is necessary to validate using data from other countries or regions. Secondly, the included data did not have IDH mutant status, while IDH mutation has been proven to be an important factor affecting the prognosis of glioma patients32. In addition, some latest treatment methods were not included. With the continuous progress in glioma research, these new methods and technologies may have impact on the prognosis of glioma patients. Lastly, the calculation of risk score will increase the economic burden on patients, so it needs to be replaced with more cost-effective methods.
In conclusion, this study provides clinicians with a practical tool specifically designed for assessing survival rates in glioma patients. The tool is designed to be concise and easy to use, suitable for various clinical scenarios. By integrating multiple influencing factors, the tool can generate highly accurate predictions of survival probabilities, thereby enhancing overall management and treatment outcomes for glioma patients.