SIR 2024
Interventional Oncology
Mohammad Mirza-Aghazadeh-Attari, MD, MPH
Postdoctoral Research Fellow
Johns Hopkins University
Financial relationships: Full list of relationships is listed on the CME information page.
Tara Srinivas, MS
Medical Student
Johns Hopkins School of Medicine
Disclosure information not submitted.
Arun Kamireddy, MD, MBBS
Research fellow
Johns Hopkins University School of Medicine
Financial relationships: Full list of relationships is listed on the CME information page.
Alan J. Kim, BS (he/him/his)
Medical Student
Johns Hopkins University School of Medicine
Financial relationships: Full list of relationships is listed on the CME information page.
Clifford R. Weiss, MD, FSIR
Professor of Radiology and Biomedical Engineering
The Johns Hopkins Hospital
Financial relationships: Full list of relationships is listed on the CME information page.
The aim of this study is to offer meta-analytic evidence regarding the effectiveness of radiomics features extracted from preprocedural or early post-procedural imaging in assessing the objective response to trans-arterial radioembolization in individuals with hepatic malignancies. Additionally, the study seeks to evaluate the methodological quality of radiomics features research in this specific context, employing the Radiomics Quality Score (RQS) and the CheckList for EvaluAtion of Radiomics research (CLEAR) tools.
Materials and Methods:
A systematic review was performed on PubMed, Embase, Google Scholar, and Scopus. Relevant original studies with a similar patient profile, index test, comparator and outcome were included. RQS and CLEAR were used to determine the methodologic quality of the studies. A Random effects model was used to determine the pooled accuracy and area under the ROC curve of the studies.
Results:
The systematic review encompassed a total of 13 studies, out of which eleven studies were integrated into the meta-analytic analysis. Among these, six studies involved patients with hepatocellular carcinoma (HCC), two studies centered around cholangiocarcinoma, and five studies enrolled patients with a range of primary and secondary liver lesions. The cumulative participant count for the analysis was 447 patients, with a recorded objective response prevalence of 59%. The pooled sensitivity, specificity, and diagnostic odds ratio were 87.5% (95% CI: 81 to 91), 81.6% (95% CI: 72 to 88), and 31 (95% CI: 17.8 to 53.9), respectively.
Meta-regression analysis revealed that neither the imaging modality nor the type of lesion significantly influenced the prognostic outcomes derived from radiomics features (P = 0.9 and 0.8, respectively). The mean Radiomics Quality Score (RQS) was determined to be 10.3 ± 3.3 (28% ± 9%), while the average count of positive items on the CLEAR checklist was 32 ± 6.4 (55% ± 11%), underscoring the limited adherence to quality assessment measures across the studies included.
Conclusion:
Radiomics features demonstrate promising potential for predicting the response to Trans-Arterial Radioembolization (TARE) in patients with hepatic lesions, regardless of lesion type and imaging modality. This offers a pathway to personalized therapy through TARE. Nonetheless, the quality of the existing body of evidence poses a limitation to widespread generalizability. Future investigations should prioritize conducting multicenter studies with external validation, while strictly adhering to the reporting standards established by RQS and CLEAR.