Prognostic factors in spinal chordoma: An update of current systematic review and meta‐analysis

Zou, Ming-Xiang; Lv, Guohua; Wang, Xiaobin

doi:10.1002/jso.24542

Cited by 7 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bootstrap analysis was used to test the robustness of the IRS as a prognostic predictor 45,46 . With combined data from the training and validation cohorts, the nomogram was constructed based on the IRS and previously reported significant clinical factors, 2,47,48 as well as the results from our multivariate analysis. The nomogram's predictive accuracy was determined by a concordance index (C‐index) and receiver operating characteristic (ROC) curve, which was also applied to assess the sensitivity and specificity of the variables for LRFS and OS prediction.…”

Section: Methodsmentioning

confidence: 99%

A four‐factor immune risk score signature predicts the clinical outcome of patients with spinal chordoma

Zou

Pan

Huang

et al. 2020

Clinical & Translational Med

Self Cite

View full text Add to dashboard Cite

Background Currently, the measurement of immune cells in previous studies is usually subjective, and no immune‐based prognostic model has been established for chordoma. In this study, we sought to simultaneously measure tumor‐infiltrating lymphocyte (TIL) subtypes in chordoma samples using an objective method and develop an immune risk score (IRS) model for survival prediction. Methods Multiplexed quantitative immunofluorescence staining was used to determine the TIL levels in the tumoral and stromal subareas of 114 spinal chordoma specimens (54 in the training and 60 in the validation cohort) for programmed death‐1 (PD‐1), CD3, CD8, CD20 (where CD is cluster of differentiation), and FOXP3. Flow cytometry was performed to validate the immunofluorescence assay for lymphocyte measurement on an additional five fresh chordoma specimens. Subsequently, the IRS model was built using the least absolute shrinkage and selection operator (LASSO) Cox regression method. Results Flow cytometry and quantitative immunofluorescence showed similar lymphocytic percentages and TIL subpopulation proportions in the fresh tumor specimens. With the training data, the LASSO model identified four immune features for IRS construction: tumoralFOXP3, tumoralPD‐1, stromalFOXP3, and stromalCD8. In both cohorts, a high IRS was significantly associated with tumoral programmed cell death‐1 ligand 1 expression, Enneking inappropriate tumor resection, and surrounding muscle invasion by tumor. Multivariate Cox regression and stratified analysis in the two cohorts revealed that the IRS was an independent predictor and could effectively separate patients with similar Enneking staging into different risk subgroups, with significantly different survival rates. Further receiver operating characteristic analysis found that the IRS classifier had a better prognostic value than the traditional clinicopathological factors and compensated for the deficiency of Enneking staging for outcome prediction. More importantly, a nomogram based on the IRS and clinical predictors showed adequate performance in estimating disease recurrence and survival of patients. Conclusions These data support the use of the IRS signature as a reliable prognostic tool in spinal chordoma and may facilitate individualized therapy decision making for patients.

show abstract

Section: Methodsmentioning

confidence: 99%