Wei Wei1, Fan Li Hua1, Nan Yu1, and Yong Yu1
1Affiliated Hospital of Shaanxi University of Chinese Medicine, XianYang, China
Synopsis
Genes
play a crucial
role in the development, progression and therapeutic outcome of tumors. Several studies have linked codeletion of chromosomal
arms 1p/19q in LGG with positive response to chemotherapy and radiotherapy and
longer progression-free survival, so 1p/19q status of LGG can impact optimal
therapy options. Unfortunately, determining 1p/19q status requires surgical
biopsy to identify chromosomal deletion. Radiomics, which extracts
high-throughput features from medical images, showing great advantages in tumor
phenotype classification, treatment options and prognosis analysis.
Objective
To evluate the
value of radiomics in predicting the 1p/19q status of Low-Grade Gliomas (LGG)
based on T2-weighted MR images.Materials and Methods
All study patients (n=159) from
the Cancer Genome Atla (TCGA) data portal who had pre-operative MRI images and
biopsy proven 1p/19q status consisting either no deletion (n=85) or co-deletion
(n=74) were included. Patients were divided into training (n=111) and validation
cohorts (n=48) in a ratio of 7:3. T2-weighted images were imported into the ITK-SNAP
to manually delineate volume of interest (VOI) of the entire-tumor. Each VOI
produced 396 radiomics features including Histogram, GLCM, GLSZM, RLM, Form
Factor and Haralick. LASSO regression was used for feature screening. A formula
was generated using a linear combination of selected features that were
weighted by their respective LASSO coefficients. A radiomics score was
calculated for each patient by the formula to reflect the 1p/19q status. The
predictive accuracy of radiomics was quantified by the area under curve (AUC)
of a ROC curve in both cohorts. The calibration degree (CD) of the radiomics
was evaluated by Hosmer-Lemeshow test. The clinical usefulness of the radiomics
signatures was assessed by decision curve analysis (DCA). Results
Seven radiomics features with non-zero
coefficients were chosen to build a radiomics label that that significantly
correlated with the 1p/19q status with an AUC, sensitivity, specificity and CD
of 0.754, 79%, 74% and 0.647 in the training cohort; and 0.757, 80%, 71% and
0.731 in the validation cohort. DCA for the radiomics label showed that if the
threshold probability was between 0.14 and 0.90, using the radiomics label to
predict 1p/19q status added more benefit than treating either all or no
patients.Conclusion
The radiomics
label can be used as a noninvasive method to predict the 1p/19q status of LGG
for clinical decision optimal therapy options.Acknowledgements
No acknowledgement found.References
1.Clark
K , Vendt B ,
Smith K , et al. The Cancer Imaging Archive (TCIA): Maintaining and Operating a
Public Information Repository[J]. Journal of Digital Imaging, 2013, 26(6):1045-1057.
2.Erickson,
Bradley; Akkus, Zeynettin; Sedlar,
Jiri; Korfiatis,
Panagiotis. (2017). Data From LGG-1p19qDeletion. The Cancer Imaging Archive.
3.Yushkevich P A
, Piven J , Hazlett H C
, et al. User-guided 3D active contour segmentation of anatomical structures:
Significantly improved efficiency and reliability[J]. NeuroImage,
2006, 31(3):1116-1128.
4.Wang
Q, et al. Radiomics
nomogram building from multiparametric MRI
to predict grade in patients with glioma: a cohort study. J Magn Reson
Imaging. 2019;49(3):825–833.