Copenhagen, Denmark
Onsite/Online

ESTRO 2022

Session Item

Monday
May 09
09:00 - 10:00
Mini-Oral Theatre 2
18: CNS
Barbara Diletto, Italy;
Ghaiet El Fida Noubbigh, Tunisia
Mini-Oral
Clinical
Clinicogenomic predictors for the pattern of failure in high-grade glioma
Byung-Hee Kang, Korea Republic of
MO-0721

Abstract

Clinicogenomic predictors for the pattern of failure in high-grade glioma
Authors:

Byung-Hee Kang1, Joo Ho Lee2, Tae Min Kim3, Jae-Kyung Won4, Hongseok Yun5, Seung Hong Choi6, Soon-Tae Lee7, Sung-Hye Park4, Chul-Kee Park8

1Seoul national university hospital, Radiation oncology, seoul, Korea Republic of; 2Seoul national university hospital, Radiation onocology, Seoul, Korea Republic of; 3Seoul National University Hospital, Hematology and Medical Oncology, , Seoul , Korea Republic of; 4Seoul National University Hospital, Pathology, Seoul, Korea Republic of; 5Seoul National University Hospital, Biomedical Research Institute, Seoul, Korea Republic of; 6Seoul National University Hospital, Radiology, Seoul , Korea Republic of; 7Seoul National University Hospital, Neurology, Seoul , Korea Republic of; 8Seoul National University Hospital, Neurosurgery, Seoul, Korea Republic of

Show Affiliations
Purpose or Objective

High-grade gliomas, the most common primary brain tumors, are highly lethal and treatment options remain limited. Despite advances in genomic technologies, there are few molecular biomarkers to guide precision medicine for high-grade glioma. Here, we aimed to identify the clinicogenomic features associated with its prognosis and recurrence patterns.


Material and Methods

Our single-institution retrospective analysis included 182 patients diagnosed with high-grade gliomas who underwent next-generation sequencing targeting 82 brain tumor-relevant genes. We analyzed clinicopathological status, treatment characteristics, survival, and genomic profiles.

Results

 At a median follow-up of 23 months (range, 2-229 months), 151 patients (83%) had progression or recurrences. Local and distant recurrences were observed in 132 (72.5%) and 101 (54.9%) patients, respectively. The most common genomic variants in high-grade gliomas were TP53 (42.9%), IDH1/2 (23.1%), TERT promoter (38.5%), ATRX (13.2%), H3F3A (7.1%), and SETD2 (6.0%) mutation. Regarding copy number variants, amplification of EGFR (20.9%), PDGFRA (9.9%), MYCN (2.2%) and loss of CDKN2A/2B (49.5%), PTEN (37.9%), RB1 (17.6%), and 1p19q codeletion(9.3%) were the most common copy number aberrations. On multivariate cox regression anlalysis, MYCN amplification (HR 6.08 95% CI 1.91-19.35, p = 0.002), and SETD2 mutation (HR 0.19 95% CI 0.06-0.62, p =0.06) were independent predictors of overall survival, in relation to previously established prognostic factors including age, Eastern Cooperative Oncology Group Performance Status (ECOG PS) scale, extent of resection, MGMT promoter methylation, and IDH1/2 mutation. Interestingly, MYCN amplification (HR 5.24 95% CI 1.69-16.27, p = 0.004), SETD2 mutation (HR 0.35 95% CI 0.12-0.99, p =0.048) were independent predictors of failure from distant recurrences. Homozygous deletion of CDKN2A/2B was associated with an increased risk of local failure.

Conclusion

The joint analysis of genomic characteristics and pattern of failure for high-grade glioma aids in identifying patients who may benefit from dose or target volume optimization of postoperative radiotherapy.