In a study reported in the Journal of Clinical Oncology, Maas et al developed a novel integrated molecular-morphologic risk classification score that improved prediction of disease recurrence in patients with meningioma.
Study Details
The study involved DNA methylation, copy-number, and mutation data from meningioma samples for analysis of predictive power for disease recurrence according to DNA methylation subgroups, copy number variations, and World Health Organization (WHO) 2016 grading. An integrated risk score—defining low, intermediate, and high risk—was developed using multivariate analysis including WHO grading, methylation family, and copy number variations risk stratification in a retrospective cohort of 514 patients; it was validated in a retrospective cohort of 184 patients and a prospective cohort of 287 patients.
Key Findings
In the discovery cohort, both copy number variation–based risk stratification (c-index = 0.706) and methylation family-based stratification (c-index = 0.721) improved accuracy of prediction of recurrence risk vs WHO grading (c-index = 0.699). Combination of these risk stratification approaches into the integrated molecular-morphologic score further improved predictive accuracy (c-index = 0.744; P values = .004 vs WHO, .008 vs copy number variations, and .06 vs methylation).
The integrated score consistently provided superior predictive accuracy in all three cohorts, with significantly improved accuracy vs WHO grading in the retrospective and prospective validation cohorts, in addition to the discovery cohort (c-index difference P = .005).
In the retrospective validation cohort, the integrated score exhibited significant risk stratification (overall P < .0001). In addition to significantly improving prediction vs WHO grading, the model significantly improved prediction vs methylation family-based risk (c-index difference P = .011).
The integrated score exhibited significant risk stratification in the prospective cohort (overall P = .0249). The c-index for the integrated score was 0.665, compared with 0.596 to 0.652 with the other three stratification approaches.
In addition to overall superior risk stratification, the integrated score more precisely distinguished risk of progression at the interface of WHO grade 1 and grade 2 tumors, with hazard ratios of 4.34 (95% confidence interval [CI] = 2.48–7.57) in the retrospective and 3.34 (95% CI = 1.28–8.72) in the prospective validation cohorts.
The investigators concluded, “Merging these layers of histologic and molecular data into an integrated, three-tiered score significantly improves the precision in meningioma stratification. Implementation into diagnostic routine informs clinical decision-making for patients with meningioma on the basis of robust outcome prediction.”
Felix Sahm, MD, PhD, of the University Hospital Heidelberg and the German Cancer Research Center, Heidelberg, is the corresponding author for the Journal of Clinical Oncology article.
Disclosure: The study was supported by the Else Kröner Fresenius Foundation, German Cancer Aid, and Hertie Foundation. For full disclosures of the study authors, visit ascopubs.org.
