Journal Thoracic Oncology

Abstracts Journal of Thoracic Oncology • Volume 12 Issue S1 January 2017
POSTER SESSION 1 - P1.02: BIOLOGY/PATHOLOGY
DRIVER GENES IN NSCLC, RESISTANCE, AND OTHER –
MONDAY, DECEMBER 5, 2016
P1.02-060 EGFR MEDIATES ACTIVATION OF RET IN LUNG
ADENOCARCINOMA WITH NEUROENDOCRINE DIFFERENTIATION
CHARACTERIZED BY ASCL1 EXPRESSION
Farhad Kosari, Kaustubh Bhinge, Lin Yang, Simone Terra, Aqsa Nasir, Prasuna
Muppa, Marie Christine Aubry, Joanne Yi, Nafiseh Janaki, Irina Kovtun,
Stephen Murphy, Hamed Rahi, Aaron Mansfield, Mariza De Andrade, Ping
Yang, George Vasmatzis, Tobias Peikert
Mayo Clinic, Rochester/United States of America
Background: Achaete-scute homolog 1 (ASCL1) is a neuroendocrine
transcription factor expressed in 10-20 % of lung adenocarcinomas (AD) with
neuroendocrine (NE) differentiation. Previously, we demonstrated that
ASCL1 functions as an upstream regulator of the RET oncogene in AD with
high ASCL1 expression (A + AD). In this study, we examined the potential role
of wild type RET in influencing the oncogenic properties of A + AD. We also
screened for drugs that could selectively target RET signaling and examined
the role of the two RET isoform separately. Methods: The association of
the mRNA expression for the long (RET51) and short (RET9) RET isoforms
with overall survival (OS) were assessed in a case-control study of stage-1
A + AD patients surgically resected at the Mayo Clinic (1994-2007). Cases and
controls were defined as patients who survived < 3.5 years after surgery
(n= 29) and > 5 years after surgery (n=38), respectively. mRNA was isolated
from FFPE tissue and analyzed by a nanostring assay. Associations of each
isoform mRNA with the OS was determined by the area under the receiver
operative characteristics (AUC). For drug screening, HCC1833 lung AD cells
with endogenously high expression of ASCL1 were stably transfected with
either empty vector or an ASCL1-shRNA. Differential sensitivities of tyrosine
kinase inhibitors (TKIs) in the pair of syngeneic cell lines were measured by
Cell-Titer Glo (Promega). Interactions between EGFR and RET was examined
by co-immunoprecipitation. Results: Expression of RET51 mRNA was
associated with poor OS (p=0.005, AUC 0.71). We detected modestly increased
sensitivity to sunitinib and vandetanib in A + AD compared with A - AD cells.
However, the EGFR inhibitors gefitinib and the dual EGFR and HER2 inhibitor
lapatinib resulted in ≥ 10 fold higher cytotoxicity in A + AD cells than in A - AD
cells. Subsequent experiments demonstrated that EGF stimulation of EGFR
mediates the phosphorylation of RET in multiple A + AD cells. RET and EGFR
were found to interact only in presence of EGF and predominantly through
the long RET isoform (RET51). Conclusion: Herein we demonstrate that wild
type EGFR predominantly interacts with the long isoform of RET (RET51) in
A + AD cells. In the presence of EGF this results in activation of RET. High RET51
is associated with worse OS. Furthermore, compared to A - AD cells, A + AD cells
appear to be more sensitivity to EGFR inhibitors. In summary, our results
suggest that A + AD patients may benefit from treatment with EGFR inhibitors
even in the absence of an EGFR mutation.
Keywords: ASCL1, neuroendocrine, RET, EGFR
POSTER SESSION 1 - P1.02: BIOLOGY/PATHOLOGY
DRIVER GENES IN NSCLC, RESISTANCE, AND OTHER –
MONDAY, DECEMBER 5, 2016
P1.02-061 KINASE FUSIONS IN NON-SMALL CELL LUNG CARCINOMA
IDENTIFIED BY HYBRID CAPTURE BASED CTDNA ASSAY
Lauren Young 1 , Siraj Ali 1 , Alexa Schrock 1 , Mark Kennedy 1 , Laurie Gay 1 , Justin
Allen 1 , James Suh 1 , Victoria Wang 2 , Eric Bernicker 3 , Sai-Hong Ignatius Ou 4 ,
Davood Vafai 5 , Jeffrey Ross 1 , Phil Stevens 1 , Vincent Miller 1
1 Foundation Medicine, Cambridge/United States of America, 2 Hematology-
Oncology, University of California San Francisco Medical Center, San Francisco/CA/
United States of America, 3 Houston Methodist Hospital, Houston/TX/United States
of America, 4 University of California, Irvine School of Medicine, Orange County/
CA/United States of America, 5 Eisenhower Medical Associates, Rancho Mirage/CA/
United States of America
Background: For the detection of genomic driver alterations in NSCLC,
comprehensive genomic profiling(CGP) or focused molecular testing of
biopsied tissue is a well-accepted approach for matching targeted therapies
in first line treatment. For NSCLC patients where invasive biopsy represents
a serious risk, assessment of circulating tumor DNA(ctDNA) is an emerging
alternative. Methods: In patients with clinically advanced NSCLC, two 10 mL
aliquots of peripheral, whole blood were collected and plasma was isolated.
ctDNA was extracted to create adapted sequencing libraries prior to hybrid
capture and sample-multiplexed sequencing on an Illumina HSQ2500 to
>5000x unique coverage. Results were analyzed with a proprietary pipeline to
call substitutions, indels, rearrangements and copy number amplifications.
Results: In 288 NSCLC patients evaluated, 20 (6.9%) harbored kinase fusions.
17/20 (85%) were adenocarcinomas, all stage IV. Median patient age was
61 years (range 41-81), and 59% were female. 13 (4.5%) cases harbored ALK
fusions with partners as follows: nine EML4 (one each of novel partners
PPFIBP1 and CACNB4), and two with unidentified partners. All but one case
had breakpoints in ALK intron 19, the remaining harboring a novel intron
17 breakpoint. Three cases (1%) harbored KIF5B-RET (canonical breakpoint
intron 12), three (1%) had CD74-ROS1 (breakpoints: ROS1 intron 33(2) and
intron 32(1)), and one had FGFR3-TACC3. ALK, RET, and ROS1 fusions were
observed by tissue testing of NSCLC in the FoundationCore database with
similar frequencies. Five patients had a biopsy with insufficient tissue for
CGP; three had both sufficient tissue and ctDNA available. The remainder
had no tissue available. For one patient, EML4-ALK fusion was detected in
both ctDNA and tissue, collected six days apart. For another, CGP identified
EGFR L858R + EGFR L709K and the patient had a durable response to afatinib/
cetuximab. After progression, ctDNA assay identified FGFR-TACC3 as well as
EGFR L858R. For a pre-menopausal, therapy naïve never smoker, female of
east Asian heritage, both assays detected a CD74-ROS1 fusion, whereas ROS1
rearrangement was not identified by the prior use of another commercially
available ctDNA test. The patient had a major radiographic response by
the second cycle of crizotinib treatment. Conclusion: Hybrid capture based
ctDNA assay can identify kinase fusions in NSCLC when CGP of biopsied tissue
cannot be performed and can direct rational use of first line TKIs. This series
identified a novel mechanism of acquired resistance to EGFR inhibitors, novel
fusion partners and intronic breakpoints for ALK, and a case of false negative
testing by another ctDNA assay.
Keywords: fusions, ctDNA, NSCLC, kinase
POSTER SESSION 1 - P1.02: BIOLOGY/PATHOLOGY
Other Mutations in Thoracic Malignancies –
MONDAY, DECEMBER 5, 2016
P1.02-062 CONSENSUS OF GENE EXPRESSION PHENOTYPES
AND PROGNOSTIC RISK PREDICTORS IN PRIMARY LUNG
ADENOCARCINOMA
Markus Ringner 1 , Johan Staaf 2
1 Lund University, Lund/Sweden, 2 Department of Oncology and Pathology, Lund
University, Lund/Sweden
Background: Transcriptional profiling of lung adenocarcinomas has
identified numerous gene expression phenotype (GEP) and risk prediction
(RP) signatures associated with patient outcome. However, classification
agreement between signatures, underlying transcriptional programs,
and independent signature validation are less studied. Methods:
Published transcriptional profiles from 17 cohorts comprising 2395 lung
adenocarcinomas with available patient outcome data were collected from
authors’ websites or public repositories as described in the original studies.
Tumors were classified according to 18 different GEPs or RPs derived from
microarray analysis of lung adenocarcinoma or NSCLC cohorts, using reported
original classification schemes or consensus clustering of gene signatures
on a per cohort basis. To independently assess the connection between
classifications by the lung cancer derived signatures and classification by
expression of proliferation-related genes only, a 155-gene breast cancer
proliferation signature was used to classify all tumors as low-proliferative
(average expression of the 155-genes < median) or high-proliferative. Tumors
were also scored according to five reported expression metagenes in lung
cancer representing different biological processes; proliferation, immune
response, basal / squamous, stroma / extra cellular matrix (ECM), and
expression of Napsin A / surfactants on a per cohort basis to contrast subtype
classifications with biological functions. Results: 16 out of 18 signatures
were associated with patient survival in the total cohort and in multiple
individual cohorts. For significant signatures, total cohort hazard ratios were
~2 in univariate analyses (mean=1.95, range=1.4-2.6). Signatures derived in
adenocarcinoma generally displayed better classification agreement than
signatures derived in mixed NSCLC cohorts. Strong classification agreement
between signatures was observed, especially for predicted low-risk patients
by adenocarcinoma-derived signatures, despite a generally low gene overlap.
Expression of proliferation-related genes correlated strongly with GEP
subtype classifications and RP scores, driving the gene signature association
with prognosis. A three-group consensus definition of samples across 10
GEP classifiers demonstrated aggregation of samples with specific smoking
patterns, gender, and EGFR/KRAS mutations, while survival differences were
only significant when patients were divided into low- or high-risk resembling a
terminal respiratory (TRU) like and non-TRU division, respectively. Conclusion:
By providing this consensus of current GEPs and RPs in lung adenocarcinoma
we have connected molecular phenotypes, risk predictions, patient outcome
and underlying transcriptional programs of the different classifier types. Our
results provide a general insight into the nature and agreement of GEP and RP
signatures in the disease, and their prognostic value.
S270 Journal of Thoracic Oncology • Volume 12 Issue S1 January 2017