Tumor Volume Growth Rates and Doubling Times during Active Surveillance of IDH-mutant Low-Grade Glioma

Authors

Ankush Bhatia

Raquel Moreno

Anne S Reiner

Subhiksha Nandakumar

Henry S Walch

Teena M Thomas

Philip J Nicklin

Ye Choi

Anna Skakodub

Rachna Malani

Vivek Prabhakaran

Pallavi Tiwari

Maria Diaz

Katherine S Panageas

Ingo K Mellinghoff

Tejus A Bale

Robert J Young

Doi

10.1158/1078-0432.CCR-23-1180

PMID: 37910594 · DOI: 10.1158/1078-0432.CCR-23-1180 · Journal: Clinical Cancer Research (2024)

TL;DR

Bhatia et al. retrospectively analyzed 128 adult patients at MSKCC with WHO 2016 Grade 2 IDH-mutant gliomas (astrocytoma and oligodendroglioma) who underwent active surveillance after resection, integrating 3D volumetric MRI segmentation with clinical and MSK-IMPACT molecular profiling. Using log-linear mixed-effects modeling, IDH-mt low-grade gliomas grew at ~10.46% per 6 months with a doubling time of 3.5 years. IDH-mt gliomas with homozygous CDKN2A/B deletion (molecular grade-high) grew roughly twice as fast (19.17% per 6 months), and joint modeling showed that each natural log tumor-volume increase conferred a >3-fold increase in risk of death (HR=3.83). The authors propose tumor volume growth rate (TVGR) as a surrogate endpoint for trials of novel IDH-targeted therapies during the watch-and-wait period (PMID:37910594).

Cohort & data

  • 128 consecutive adult patients with WHO 2016 Grade 2 IDH-mutant astrocytoma or oligodendroglioma evaluated at Memorial Sloan Kettering Cancer Center (MSKCC) between 1997 and 2019, all observed without upfront chemoradiation after diagnostic biopsy or resection (PMID:37910594).
  • 69 (53.9%) 1p19q codeleted oligodendrogliomas and 59 (46.1%) 1p19q intact astrocytomas; median follow-up 6.3 years (range 0.5–23.1); median 8 MRI scans per patient (range 2–28).
  • Cancer types: DIFG, ASTR, ODG.
  • Dataset: difg_msk_2023 (data released on Synapse syn52658621 and cBioPortal study difg_msk_2023).
  • Assay: MSK-IMPACT, a 468-gene targeted hybrid-capture NGS panel, applied to 73/128 patients with tissue available.
  • Imaging: manual 3D volumetric segmentation of T2/FLAIR abnormality (TeraRecon iNtuition 4.4.13) with VASARI feature annotation.

Key findings

  • Entire cohort (n=128) continuous TVGR = 10.46% per 6 months (95% CI 9.11–11.83), doubling time 3.5 years (95% CI 3.10–3.98) (PMID:37910594).
  • Log-linear mixed-effects model fit (adjusted r²=0.9864) was marginally superior to linear (r²=0.9556).
  • No significant TVGR difference by 1p19q status: intact 11.73% vs codeleted 9.62% per 6 months (P=0.16); doubling times 3.1 vs 3.8 years.
  • No significant TVGR differences by initial postoperative tumor size (<25 cm³ vs ≥25 cm³, P=0.68) or by extent of resection (near total vs subtotal vs partial, all P≥0.20).
  • Molecular grade-high (CDKN2A/B homozygous deletion, n=4): TVGR 19.17% per 6 months (95% CI 15.57–22.89), significantly faster than molecular grade-intermediate (9.09%) and molecular grade-low (9.54%); pooled low+intermediate 9.37% vs high 19.17% remained significant.
  • Joint longitudinal-survival modeling: each 1-unit increase in ln(tumor volume) → HR=1.61 (95% CI 1.33–1.95, P<0.0001) for next-intervention-free survival (NIFS), and HR=3.83 (95% CI 2.32–6.30, P<0.0001) for overall survival. Each 10% increase in tumor volume corresponded to a 5% increase in NIFS risk and a 14% increase in risk of death.
  • Mutation landscape in MSK-IMPACT subcohort (n=73): 1p19q codeleted tumors — TERT promoter mutations 100%, CIC 42%, FUBP1 24%; 1p19q intact tumors — TP53 94%, ATRX 77%.
  • T2-FLAIR mismatch sign was present in 38.9% of 1p19q intact tumors and absent in 96.9% of codeleted tumors (P<0.0001), and was present in 100% of molecular grade-high tumors vs 13.5% low / 14.3% intermediate (P=0.008) — a novel association of the T2-FLAIR mismatch sign with higher molecular grade.
  • Tumor mutational burden and fraction genome altered did not differ between slow, intermediate, and fast TVGR strata.

Genes & alterations

  • IDH1 / IDH2 — mutation required for cohort inclusion; verified by IHC (93.8%) and/or NGS (57%).
  • CDKN2A / CDKN2B — homozygous deletion defined “molecular grade-high” (WHO 2021 Grade 4 IDH-mt astrocytoma) and drove ~2× faster TVGR (19.17% vs 9.54% per 6 months).
  • TERT — promoter mutations in 100% of 1p19q codeleted oligodendrogliomas sequenced.
  • CIC (42%) and FUBP1 (24%) — canonical co-alterations in 1p19q codeleted oligodendrogliomas.
  • TP53 (94%) and ATRX (77%) — canonical co-alterations in 1p19q intact astrocytomas.
  • MYCN, CDK4, PDGFRA — focal amplifications used to define “molecular grade-intermediate” in 1p19q intact tumors.
  • PIK3R1, PIK3CA — mutations used (with TP53 and broad CNV load) to define “molecular grade-intermediate” in 1p19q codeleted tumors.

Clinical implications

  • TVGR on volumetric MRI is proposed as an earlier surrogate endpoint for active-surveillance trials of IDH-targeted therapies in IDH-mt LGG, aligning with WHO 2021 molecular grading and correlating with NIFS and OS (PMID:37910594).
  • CDKN2A/B homozygous deletion identifies Grade 2 IDH-mt gliomas that behave aggressively on imaging even during the watch-and-wait period, supporting the WHO 2021 upgrade of these tumors to Grade 4.
  • The T2-FLAIR mismatch sign may flag higher-molecular-grade IDH-mt gliomas, a novel potentially predictive imaging biomarker warranting prospective evaluation.
  • Absence of a growth-rate difference between 1p19q codeleted vs intact tumors in the untreated phase suggests the known survival advantage of codeleted tumors may be driven by chemosensitivity rather than intrinsically slower natural history.

Limitations & open questions

  • Retrospective, single-center (MSKCC) design limits generalizability.
  • Heterogeneous scan parameters across ~22 years (pre-dating consensus brain tumor imaging protocols); slice thickness, interslice gap, and scanner type may bias volume estimates.
  • Manual segmentation, although blinded and expert-reviewed, may include edema or post-surgical changes indistinguishable from tumor on MRI.
  • Molecular grade-high subgroup is very small (n=4), limiting precision of the CDKN2A/B deletion effect size.
  • The authors chose to restrict molecular grade-high to CDKN2A/B deletions; whether PDGFRA/CDK4 amplifications and PI3K pathway mutations should also upgrade tumors (as some prior literature suggests) is unresolved.
  • Prospective validation, automated/AI-assisted segmentation, and integration with advanced MRI/PET are called out as next steps.

Citations from this paper used in the wiki

  • “the entire cohort (n=128) had a continuous %TVGR per 6 months of 10.46% (95% CI: [9.11%, 11.83%]) and a doubling time of 3.5 years (95% CI: [3.10-3.98])” — Results/Abstract.
  • “High molecular grade IDH-mt gliomas, defined by the presence of homozygous deletion of CDKN2A/B, had %TVGR per 6 months of 19.17% (95% CI: [15.57%, 22.89%])” vs low “9.54% (95% CI: [7.32%, 11.80%]) (P < 0.0001)” — Abstract.
  • “each one natural logarithm tumor volume increase resulted in more than a 3-fold increase in risk of death (HR=3.83, 95% CI: [2.32-6.30], P < 0.0001)” — Abstract / Results.
  • “The most common genomic alterations in the 1p19q codeleted group were TERT promoter mutations (100%), CIC (42%) and FUBP1 (24%)… in the 1p19q intact group the most frequently altered genes were TP53 (94%) and ATRX (77%)” — Results.
  • “The T2-FLAIR mismatch sign was present in 21 (38.9%) 1p19q intact patients and absent in 63 (96.9%) of 1p19q codeleted patients (P < 0.0001)” — Results.
  • “data generated in this study are available… for visualization in cBioPortal (https://cbioportal.org/study/summary?id=difg_msk_2023)” — Methods/Data Availability.

This page was processed by crosslinker on 2026-05-04.