A multi-center, clinical analysis of IDH-mutant gliomas, WHO Grade 4: implications for prognosis and clinical trial design.

PURPOSE: Mutations in the Isocitrate Dehydrogenase (IDH) genes, IDH1 or IDH2, define a group of adult diffuse gliomas associated with a younger age at diagnosis and better prognosis than IDH wild-type glioblastoma. Within IDH mutant gliomas, a small fraction of astrocytic tumors present with grade 4 histologic features and poor prognosis. In molecular studies, homozygous deletion of CDKN2A/B is independently predictive of poor prognosis and short survival. As a consequence, 2021 WHO classification now also recognizes this molecular feature, CDKN2A/B deletion, as sufficient for classifying an astrocytoma as IDH-mutant, WHO Grade 4, regardless of histological grading. Here, we investigate outcomes of patients with WHO Grade 4 IDH-mutant astrocytoma both with and without CDKN2A/B deletion, to compare these groups and evaluate clinical and radiographic factors that contribute to survival. METHODS: We retrospectively identified 79 patients with IDH-mutant astrocytoma with CDKN2A/B deletion detected at initial diagnosis across five international institutions as well as a comparison group of 51 patients with IDH-mutant, astrocytoma, histologically Grade 4 without detectable CDKN2A/B deletion. We assembled clinical and radiographic features for all patients. RESULTS: We find that CDKN2A/B deletion was associated with significantly worse overall survival (OS; p = 0.0004) and progression-free survival (PFS; p = 0.0026), with median OS of 5.0 years and PFS of 3.0 years, compared to 10.1 and 5.0 years for tumors with a grade 4 designation based only on histologic criteria. Multivariate analysis confirmed CDKN2A/B deletion as a strong negative prognosticator for both OS (HR = 3.51, p < 0.0001) and PFS (HR = 2.35, p = 0.00095). In addition, in tumors with CDKN2A/B deletion, preoperative contrast enhancement is a significant predictor of worse OS (HR 2.19, 95% CI 1.22-3.93, p = 0.0090) and PFS (HR = 1.74, 95% CI = 1.02-2.97, p = 0.0420). CONCLUSIONS: These findings underscore the severe prognostic impact of CDKN2A/B deletion in IDH-mutant astrocytomas and highlight the need for further refinement of tumor prognostic categorization. Our results provide a key benchmark of baseline patient outcomes for therapeutic trials, underscoring the importance of CDKN2A/B status assessment, in addition to histologic grading, in clinical trial design and therapeutic decision-making for IDH-mutant astrocytoma patients.

CDKN2A/B Homozygous Deletion Sensitizes IDH-Mutant Glioma to CDK4/6 Inhibition.

PURPOSE: Treatment paradigms for isocitrate dehydrogenase (IDH)-mutant gliomas are rapidly evolving. Although typically indolent and responsive to initial treatment, these tumors invariably recur at a higher grade and require salvage treatment. Homozygous deletion of the tumor suppressor gene CDKN2A/B frequently emerges at recurrence in these tumors, driving poor patient outcomes. We investigated the effect of CDK-Rb pathway blockade on IDH-mutant glioma growth in vitro and in vivo using CDK4/6 inhibitors (CDKi). EXPERIMENTAL DESIGN: Cell viability, proliferation assays, and flow cytometry were used to examine the pharmacologic effect of two distinct CDKi, palbociclib and abemaciclib, in multiple patient-derived IDH-mutant glioma lines. Isogenic models were used to directly investigate the influence of CDKN2A/B status on CDKi sensitivity. Orthotopic xenograft tumor models were used to examine the efficacy and tolerability of CDKi in vivo. RESULTS: CDKi treatment leads to decreased cell viability and proliferative capacity in patient-derived IDH-mutant glioma lines, coupled with enrichment of cells in the G1 phase. CDKN2A inactivation sensitizes IDH-mutant glioma to CDKi in both endogenous and isogenic models with engineered CDKN2A deletion. CDK4/6 inhibitor administration improves survival in orthotopically implanted IDH-mutant glioma models. CONCLUSIONS: IDH-mutant gliomas with deletion of CDKN2A/B are sensitized to CDK4/6 inhibitors. These results support the investigation of the use of these agents in a clinical setting.

A single-institution retrospective analysis of pathologically determined malignant transformation in IDH mutant glioma patients.

Background: Lower-grade IDH mutant glioma patients frequently undergo malignant transformation (MT), with apparent worse prognosis. Many studies examine MT in mixed IDH status cohorts and define MT using imaging, not histopathology. Our study examines the timing, predictors, and prognostic implications of pathologically determined MT in a large, exclusively IDH mutant cohort. Methods: We identified 193 IDH mutant lower-grade glioma patients at UCLA who received multiple surgeries. We examined the outcomes of pathologically determined MT patients. Results: Time to MT is longer in grade 2 oligodendroglioma (G2 Oligo) than in grade 2 astrocytoma (G2 Astro) (HR = 0.46, P = .0007). The grade 3 astrocytoma (G3 Astro) to grade 4 astrocytoma (G4 Astro) interval is shorter in stepwise MT (G2 to G3 to G4 Astro) patients than in initial G3 Astro patients (P = .03). Novel contrast enhancement had 65% positive predictivity, 67% negative predictivity, 75% sensitivity, and 55% specificity in indicating pathologically defined MT. In G2 Astro, initial gross total resection delayed MT (HR = 0.50, P = .02) and predicted better overall survival (OS) (HR = 0.34, P = .009). In G2 Oligo, spontaneous MT occurred earlier than treated MT (HR = 11.43, P = .0002), but treatment did not predict improved OS (P = .8). MT patients (n = 126) exhibited worse OS than non-MT patients (n = 67) in All (HR = 2.54, P = .0009) and G2 Astro (HR = 4.26, P = .02). Conclusion: Our study expands the understanding of MT to improve IDH mutant lower-grade glioma management.

Retrospective examination of pseudoprogression in IDH mutant gliomas.

Background: Tumor surveillance of isocitrate dehydrogenase (IDH) mutant gliomas is accomplished via serial contrast MRI. When new contrast enhancement (CEnew) is detected during postsurgical surveillance, clinicians must assess whether CEnew indicates pseudoprogression (PsP) or tumor progression (TP). PsP has been better studied in IDH wild-type glioblastoma but has not been well characterized in IDH mutant gliomas. We conducted a retrospective study evaluating the incidence, predictors, natural history, and survival of PsP patients in a large cohort of IDH mutant glioma patients treated at a single institution. Methods: We identified 587 IDH mutant glioma patients treated at UCLA. We directly inspected MRI images and radiology reports to identify CEnew and categorized CEnew into TP or PsP using MRI or histopathology. Results: Fifty-six percent of patients developed CEnew (326/587); of these, 92/326 patients (28% of CEnew; 16% of all) developed PsP and 179/326 (55%) developed TP. All PsP patients had prior radiation, chemotherapy, or chemoradiotherapy. PsP was associated with longer overall survival (OS) versus TP patients and similar OS versus no CEnew. PsP differs from TP based on earlier time of onset (median 5.8 vs 17.4 months from treatment, P < .0001) and MRI features that include punctate enhancement and enhancement location. Conclusion: PsP patients represented 28% of CEnew patients and 16% of all patients; PsP patients demonstrated superior outcomes to TP patients, and equivalent survival to patients without CEnew. PsP persists for <1 year, occurs after treatment, and differs from TP based on time of onset and radiographic features. Poor outcomes after CEnew are driven by TP.