Annual Meeting Proceedings Part 1 - American Society of Clinical ...

120s Central Nervous System Tumors
2019 Poster Discussion Session (Board #7), Fri, 1:00 PM-5:00 PM and
4:30 PM-5:30 PM
IDH1 status and survival benefit from surgical resection of enhancing and
nonenhancing tumor in malignant astrocytomas. Presenting Author: Daniel
P. Cahill, Massachusetts General Hospital, Boston, MA
Background: The value of maximal safe resection for malignant astrocytic
gliomas (AA, WHO Grade III anaplastic astrocytoma and GBM, WHO Grade
IV glioblastoma) has sometimes been controversial, because of confounding
between measures of surgical resection and other prognostic factors.
IDH1 gene mutations are associated with improved survival in glioma
patients, and are thought to identify tumors with a distinct molecular
evolutionary origin. We sought to determine the prognostic impact of
surgical resection on survival after controlling for IDH1 status in malignant
astrocytomas. Methods: Clinical parameters including preoperative and
postoperative MRI-based tumor volume were recorded prospectively on
407 malignant astrocytoma patients – AA (n�157) and GBM (n�250).
IDH1 status was assessed by sequencing and R132H-specific immunohistochemistry.
Results: The measures of surgical resection associated with
longer survival differed between IDH1 wild-type and mutant tumors. In
multivariate analyses of IDH1 wild-type tumors (controlling for age,
Karnofsky performance score, tumor location, and tumor grade), residual
postoperative enhancement was associated with a median survival of 9.9
mo vs. 17.4 mo with no enhancement (HR�1.73, 95% CI, 1.19-2.52,
p�.004). Residual non-enhancing disease, however, was not associated
with survival (scored as continuous volumetric cc, 95% CI 0.99-1.01,
p�.608). These results are consistent with prior studies of GBM, which are
largely IDH1 wild-type lesions (Lacroix et al., J Neurosurg 95:190-8,
2001). In contrast, in IDH1 mutant tumors, both residual enhancing
(HR�7.93, 95%CI 1.14-55.22, p�.037) and non-enhancing (HR�1.03,
95% CI 1.01-1.05, p�.005) postoperative tumor burden were associated
with worse survival. Conclusions: These data suggest surgical resection in
malignant astrocytic gliomas may be individualized based on IDH1 genotype.
IDH1 mutant tumors have a better baseline overall prognosis,
therefore more aggressive surgery and tolerance of temporary peri-operative
neurologic deficits can be weighed in an attempt to gain the additional
survival benefit that appears to be associated with reducing non-enhancing
tumor burden.
2021 Poster Discussion Session (Board #9), Fri, 1:00 PM-5:00 PM and
4:30 PM-5:30 PM
MET and ALK in glioblastoma multiforme (GBM): Comparison of IHC and
FISH. Presenting Author: Kimary Kulig, National Comprehensive Cancer
Network, Fort Washington, PA
Background: GBM, the most common and lethal malignant primary brain
tumor in adults, has an overall one-year life expectancy of 33.7%.
Personalized medicine on the basis of predictive biomarkers remains
elusive in GBM; however a recent case report showed clinical improvement
after treatment with a mesenchymal-epithelial transition (MET) inhibitor.
The authors of that report ascribe clinical response to the MET inhibitor,
but the drug is also an inhibitor of anaplastic lymphoma kinase (ALK). We
explored the co-expression, gain, or amplification of both MET and ALK by
IHC and FISH in GBM. Methods: A convenience sample of 56 available
tumors from gross-resected GBM cases within the Duke Brain Tumor tissue
bank was stained for MET using clone 8F11 (Leica) and for ALK using clone
5A4 (Novocastra). FISH employed the LSI ALK (2p23) breakpointspanning
dual-color DNA probe and the LSI D7S486 (MET)/CEP 7 dual
color DNA probe (Vysis). Results: Of 56 GBM cases, MET was expressed in
69.6% and ALK in 17.9% by IHC. By FISH, gain or amplification was found
in 100% of cases for MET and in 48.2% for ALK. Co-expression of MET and
ALK by IHC was 14.3% and 48.2% by FISH. Co-expression by either IHC or
FISH was also 48.2%. OS estimation for this cohort is premature, with
77% of patients still alive. Conclusions: MET and ALK are (co-)expressed in
a significant proportion of GBM. IHC detection of MET and ALK did not
correlate well with FISH results, suggesting that gene gain or amplification
does not necessarily lead to abnormal MET and ALK protein expression or
that IHC methods or antibodies used can be optimized. Prior studies
reported MET amplification in GBM as low as 4%. Our study indicates that
MET gain/amplification and protein expression may be much higher, having
implications for MET-targeted therapy. Additionally, co-expression of ALK
and MET has implications for dual inhibitors of these signaling pathways as
well as resistance mechanisms. Our continued work will explore significance
of MET and ALK as predictors of therapeutic response, with careful
control for clinically relevant patient and disease characteristics.
Biomarker IHC �
FISH �
(gain or amplification) IHC- FISH-
MET 39 56 17 0
ALK 10 27 46 29
MET � ALK 8 27 15 0
2020 Poster Discussion Session (Board #8), Fri, 1:00 PM-5:00 PM and
4:30 PM-5:30 PM
Genomic characterization of meningiomas. Presenting Author: Priscilla
Kaliopi Brastianos, Dana-Farber Cancer Institute, Boston, MA
Background: Understanding the genetic alterations in cancer has lead to
groundbreaking discoveries in targeted therapies. Meningiomas are among
the most common primary brain tumors, with approximately 18,000 new
cases diagnosed annually. Though certain genes have been associated with
the development of meningiomas, most notably the tumor suppressor gene
neurofibromatosis 2 (NF2), the genetic changes that drive meningiomas
remain poorly understood. Our objective was to comprehensively characterize
the somatic genetic alterations of meningiomas to gain insight into the
molecular pathways that drive this disease. Methods: Fresh frozen specimens
and paired blood were collected from 16 consented patients. DNA
was extracted from regions of high tumor purity determined by evaluation of
H&E slides. Whole-genome sequencing from 10 tumor-normal pairs and
whole-exome sequencing from 6 tumor-normal pairs was carried out. We
performed an unbiased screen for point mutations, insertions-deletions,
rearrangements and copy-number changes across the exomes and genomes.
Recurrent (potential driver) events were then analyzed with
additional algorithms for statistical significance. Results: Alterations in the
NF2 gene were present in 9 of 16 patients. Multiple novel rearrangements
and recurrent non-NF2 mutations were also identified in the cohort.
Massive genomic rearrangement termed chromothripsis was observed in
chromosome 1 in one sample, which has never previously been described in
meningiomas, and represents a potentially new mechanism of malignant
transformation in this tumor type. Conclusions: While NF2 mutations
appear to drive a majority of these tumors, our analysis has uncovered
additional potential driver genes in meningiomas, particularly in those
tumors negative for NF2 alterations. To our knowledge, this is the first study
to comprehensively characterize the totality of somatic genetic alterations
in meningiomas, and brings us closer to the development of new therapeutic
targets for this disease.
2022 Poster Discussion Session (Board #10), Fri, 1:00 PM-5:00 PM and
4:30 PM-5:30 PM
Pilocytic astrocytomas in adults: A retrospective study of 125 patients.
Presenting Author: Brett James Theeler, University of Texas M. D. Anderson
Cancer Center, Houston, TX
Background: Pilocytic astrocytomas (PAs) are rare primary brain tumors in
adults. The natural history is poorly defined with the largest published
series including only 30-40 patients. To better define clinical characteristics
of adult PAs, we present an analysis of adult PAs at MD Anderson
Cancer Center. Methods: We conducted an IRB-approved, retrospective
chart review of patients �18 years diagnosed with PAs from 1990-2011 at
our institution. Results: We identified 125 patients with a median age of 29
(range 18-72y), female to male ratio of 1.7:1, and median follow-up of 62
months. The most common anatomic locations were brainstem 22%,
cerebellum 14%, intraventricular 14%, and cortical/lobar 14% (36% other
locations). 21% of cases had an initial pathologic diagnosis discordant with
neuropathologic review at our institution. Comorbidities included neurofibromatosis
type 1 (5 pts) and a history of asthma or autoimmune disease
(14%). 38 patients had a gross total (GTR), 55 a sub-total resection (STR)
and the remainder a biopsy. 62 patients were treated with radiotherapy
(RT): 45 as adjuvant (adj), 17 at recurrence; 2 received RT in both adj and
recurrent settings. 73 (58%) patients were stable after initial treatment
(surgery �/- adj RT). Median progression free survivals (mPFS) based on
upfront treatment as follows: GTR only (n�30) - not been reached, STR
only (n�32) 226.6 months, GTR plus adj RT (n�8) 13.2 months, STR plus
adj RT (n�23) is 52.4 months (p�0.026, logrank test for trend). At last
follow-up, 77% of all patients were stable and 13% were deceased.
Conclusions: This is the largest series of adult PAs reported to date.
Extracerebellar location, slight female predominance, and history of
allergic/autoimmune disease were associations found in this cohort.
Neuropathologic review is essential to avoid misdiagnosis. A significant
subset of adult PAs behave more aggressively than expected for a grade I
tumor with over 40% of patients experiencing recurrence after initial
treatment. Adj RT following GTR or STR did not improve PFS, and a trend
towards worse PFS with adj RT was observed although these analyzed
subgroups were small. These results may help provide a framework for
prospective studies in this tumor type.
Visit abstract.asco.org and search by abstract for the full list of abstract authors and their disclosure information.