Nomogram-based strategy to predict relapse-free survival in patients with gastrointestinal stromal tumor using inflammatory indicators

Abstract

Zhao et al’s investigation on the assessment of inflammatory markers prognostic value for relapse-free survival in patients with gastrointestinal stromal tumor (GIST) using a nomogram-based approach is a scientific approach. This study explored the potential of an inflammatory marker-based nomograph model, highlighting the relapse-free survival-associated risk factors prognostic potential in patients with GIST. The author assessed 124 samples from patients with GIST to find an association between inflammatory markers and tumor size in a retrospective study using multivariate regression analysis. Further, a nomogram model was developed to identify the independent risk factors for the prognosis. GIST clinical treatment can use preoperative monocyte/lymphocyte ratio and platelet/lymphocyte ratio for relapse-free survival prognosis as independent factors.

Key Words: Recurrence-free survival; Gastrointestinal stromal tumor; Nomogram model; Inflammatory markers; Independent risk factor; Cancer

Core Tip: This retrospective study evaluated inflammatory indicators' predictive significance for relapse-free survival in patients with gastrointestinal stromal tumor (GIST) utilizing a nomogram-based technique. The author tried to associate test preoperative inflammatory indicators with recurrence-free survival. The study created a GIST prognostic nomogram using inflammatory markers, but the limitation(s) and its possible effect on the outcome of GIST prognosis were not discussed. Inclusion and exclusion criteria should be broad enough to minimize the false-positive effects of immune marker levels in test samples. Normalization of inflammatory markers in response to different chemotherapeutic drug(s) is important for obtaining a stable prognostic outcome in GIST.

TO THE EDITOR

I was delighted to read the high-quality article by Zhao et al[1], published in the World Journal of Gastrointestinal Oncology. The main focus of this article was to predict the prognosis of patients with gastrointestinal stromal tumor (GIST) after surgery by examining the predictive value of inflammatory indicators such as the systemic immune inflammation index (SII), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and monocyte-to-lymphocyte ratio (MLR). The study included 124 patients from 2014 to 2017 and identified independent risk factors influencing prognosis. The results showed a correlation between MLR and PLR and tumor size, and preoperative SII, MLR, NLR, and PLR were significantly associated with recurrence-free survival. The study suggests that nomograms based on these factors can be used for clinical treatment.

In the present study, the author used the predictive value of inflammatory markers along with the creation of related nomograms to extend this idea to predict the post-surgery relapse in patients with GIST. While the study is interesting, I need to express some concerns regarding the report. Although the author mentioned some of the limitations, such as lack of external data verification, post-surgery chemotherapy interventions, and small sample size, a discussion on the limitations of the proposed nomogram was lacking. Some critical parameters, such as decision curve analysis (DCA) and calibration plots, were missing in the study, which raises the question of the authenticity of the proposed nomogram for postoperative recurrence in patients with GIST.

In the 19^th century, the German pathologist Virchow[2] proposed a close correlation between inflammation and tumors. This connection has been confirmed by epidemiological and molecular biology research. Tumor-related inflammatory factors include cytokines, extracellular matrix proteins (secreted by tumor cells during injury or infection), and tumor-related inflammatory cells can cause tissue atrophy, destruction, and malignant progression of tumors. The persistent inflammatory microenvironment induces tumorigenesis and aggravates the inflammatory response, and a systemic inflammatory response is strongly linked to tumor prognosis. Tumour cell invasion and metastasis are linked to the progression of many tumours by peripheral blood immune and inflammatory cells such as neutrophils, monocytes, platelets, and lymphocytes. Neutrophils, monocytes, and platelet ratios to lymphocyte count are effective systemic inflammatory scoring indicators[3].

Nomograms are increasingly used in cancer studies to predict clinical results, providing accurate predictions through easy-to-understand shapes and sizes. However, increasing nomogram suggestions can lead to misunderstandings and uncertainty in clinical situations[4]. It remains unclear whether the nomogram(s) is beneficial for patient results. One reason that stands out is that most nomograms are made by looking backwards. Prospective evaluation is not always done because it is hard to do, so there would be a lot of doubt when nomograms are used in clinical settings. Because of this limitation, scientists have come up with DCA to help figure out how useful prediction models are in the real world. This method takes into account the results of the analysis, and while the analysis is ongoing, the clinical value of a prediction could be checked to see if the model works in this case[5,6]. The nomogram's main idea is that its effectiveness will remain consistent over time, even if the clinical setting undergoes significant changes, which is actually not the case in reality. The disease progression or relapse is a dynamic process that occurs at different rates. Thus it is important to note that nomograms are believed to be effective only at one time point but may be incorrect in mid-term changes (during the disease progression/relapse process) due to their accuracy[7-9]. The nomogram's accuracy is crucial for demonstrating high sensitivity, but calibration is necessary for quantifying it. However, calibration plots are often omitted, making it difficult to estimate accuracy at different probabilities. Oncologists should be aware of published nomograms without calibration curves to ensure consistency in predictions[10]. Comparing the area under the curve of nomograms with different databases is not accurate, as differences may not indicate the importance of each nomogram in real-world scenarios, even if variables remain constant[11].

CONCLUSION

The present retrospective clinical study-based relationships between preoperative inflammatory markers and postoperative relapse in patients with GIST is in need of improvement. It is unable to answer some of the valuable limitations of nomograms, and also lacks some assessments. A prospective study should be designed to avoid the bias in the collection of data in the past (retrospective data collection). Inclusion of molecular and genetic prognostic indicators is suggested to overcome the limitations of clinical and pathologic prognostic markers. Including these recommendations will provide better accuracy in the development of nomograms for the prognosis of postoperative relapse in patients with GIST.