Journal Thoracic Oncology

Abstracts Journal of Thoracic Oncology • Volume 12 Issue S1 January 2017
of 37 days and accumulation in tumors with permeable vasculature. This
phase II trial evaluated the CNS activity of etirinotecan pegol in patients with
lung cancer and refractory brain metastases. Methods: Patients with lung
cancer and brain metastases were eligible who had received prior systemic
therapy and prior brain-directed neurosurgery, radiation/radiosurgery, or
refused whole brain radiotherapy (WBRT). Measurable brain metastases
were defined as one >= 10 mm; or one 5-9 mm with others >= 3 mm, totaling >=
10 mm. Etirinotecan pegol was administered at 145 mg/m2 IV every 3 weeks.
Response (modified RECIST 1.1) was assessed with brain MRI and systemic
CT every 6 weeks. The primary endpoint was a 25% or greater 12-week CNS
disease control rate (CNS-DCR; defined as unconfirmed response or stable
disease with systemic non-progression) in a non-small cell lung cancer (NSCLC)
cohort. Another exploratory cohort enrolled small cell lung cancer (SCLC)
patients. Results: In the NSCLC cohort, twelve patients were enrolled, all
with adenocarcinoma. Genomic alterations included six (50%) with EGFR
mutation, and one each with HER2, KRAS, ROS1, and NRAS. Patients received
a median of 2.5 prior systemic treatments. Two (17%) patients had prior
neurosurgery, ten patients (83%) had irradiation - 3 WBRT, 9 radiosurgery.
Common related toxicities were nausea and diarrhea (each in 50%), vomiting
(22%) and blurred vision (22%). One patient died after developing diarrhea
and dehydration. CNS responses lasting 24, 8, and 6 weeks were observed in
25% (3/12) - all EGFR mutation positive. The 6-week CNS-DCR was 50% (6/12),
but 12-week CNS-DCR was 17% (2/12). Median progression-free survival was
11.4 weeks (95% CI 5.3-11.7) and median overall survival from study entry was
29.6 weeks (95% CI 22.3-38.2).
In the SCLC cohort, two patients were enrolled. One patient with prior
PCI had CNS response at 5 weeks but died of neutropenic infection; one
who refused prior WBRT had CNS progression at 6 weeks. Conclusion:
Radiographic responses of brain metastases were observed in patients
following administration of etirinotecan pegol, but the study did not meet
the primary endpoint because the 12-week CNS-DCR was 17%. Further study
of etirinotecan pegol is ongoing in patients with breast cancer and brain
metastases.
Keywords: Brain metastasis, chemotherapy, non small cell lung cancer
POSTER SESSION 2 – P2.03B: ADVANCED NSCLC & CHEMOTHERAPY/TARGETED THERAPY/
IMMUNOTHERAPY
BRAIN META –
TUESDAY, DECEMBER 6, 2016
P2.03B-013 OUTCOME OF PATIENTS WITH ALK+ NSCLC AND BRAIN
METASTASES IN RELATION TO DISEASE BURDEN AND CLINICAL
MANAGEMENT
Luigi De Petris 1 , Signe Friesland 1 , Daniel Brodin 2 , Hanna Carstens 1 , Johan
Falkenius 1 , Oscar Grundberg 2 , Magnus Löfdahl 2 , Clara Lenneby Helleday 1 ,
Reza Karimi 2 , Anna Öjdahl-Boden 2 , Georgios Tsakonas 1 , Simon Ekman 1 , Karl-
Gustav Kölbeck 2
1 Oncology, Karolinska University Hospital, Stockholm/Sweden, 2 Respiratory
Medicine and Allergology, Karolinska University Hospital, Stockholm/Sweden
Background: The management of ALK+ NSCLC patients with CNS metastases
represents a clinical challenge. We conducted this single institution
retrospective analysis in order to evaluate the frequency of CNS metastases
in this patient group and explore clinical features associated with survival
Methods: Between 2011 and 2016, 70 patients with advanced ALK+
adenocarcinoma were treated at our institution. Data on CNS imaging
modality, treatment strategy and outcome was collected by chart review
Results: CNS imaging was performed with either MRI(36%) or CECT(64%) in
59 cases, and CNS metastases were diagnosed in 56% of examined subjects.
The characteristics of these 33 patients were as follows: gender male/female
45%/54%; median age 60y (IQR 50-65); # of CNS metastases 1-3/4-10/>10
21%/36%/42% (including 3 subjects with leptomeningeal involvement); timing
for diagnosis of CNS metastases: at primary cancer diagnosis (21%), at PD on
chemotherapy (33%), at PD on crizotinib (39%), at PD on 2 nd generation ALKi
(6%). Radiotherapy was administered as either SRS(42%) or WBRT(58%) in
66% of cases. Overall medical treatment was chemotherapy (n=32); crizotinib
(n=28); 2 nd generation ALKi, either alectinib, ceritinib or brigatinib (n=22). The
first treatment strategy upon diagnosis of CNS metastases was radiotherapy
alone, crizotinib or a 2 nd gen ALKi in 42%, 24% and 33% of subjects,
respectively. In 4 cases, the 2 nd generation ALKi was started directly after
completion of radiotherapy. Median OS from the diagnosis of CNS metastasis
was 18 months (95% CI 9-50). 1- and 2-year survival rates were 59% and 44%,
respectively. Cox proportional hazard analysis showed that neither gender,
age, timing for diagnosis of CNS metastasis nor the use of radiotherapy
were significant prognostic factors for OS in this patient cohort. Survival
analysis stratified by Number of CNS metastasis showed a trend favoring
1-3 met (median OS 59 months) vs 4-10 and >10 lesions (median OS of 25 and
9 months, respectively), with the three survival curves crossing each other
(p=0.1). On the other hand, the first treatment strategy after the diagnosis
of CNS metastases was shown to be indeed a significant prognostic factor
for OS. Median OS for patients treated with crizotinib, radiotherapy alone or
a 2 nd ALKi (with or without RT) was 9 months, 29 months and Not reached,
respectively (p=0.03). Conclusion: This retrospective study confirms the
high incidence of CNS metastases in Caucasian patients with advanced ALK+
NSCLC. The wider implementation of 2 nd generation ALKi in clinical practice
may change the prognosis of these subjects
Keywords: ALK, WBRT, CNS Metastasis, 2nd generation ALKi
POSTER SESSION 2 – P2.03B: ADVANCED NSCLC & CHEMOTHERAPY/TARGETED THERAPY/
IMMUNOTHERAPY
BRAIN META –
TUESDAY, DECEMBER 6, 2016
P2.03B-014 ATEZOLIZUMAB IN ADVANCED NSCLC PATIENTS
WITH BASELINE BRAIN METASTASES: A POOLED COHORT SAFETY
ANALYSIS
Rimas Lukas 1 , Mayank Gandhi 2 , Carol O’Hear 2 , Sylvia Hu 2 , Catherine Lai 2 , Jyoti
Patel 3
1 Department of Neurology, The University of Chicago Medicine, Chicago/IL/United
States of America, 2 Genentech, Inc., San Francisco/CA/United States of America,
3 Section of Hematology/Oncology, The University of Chicago Medicine, Chicago/IL/
United States of America
Background: Brain metastases, occurring in 20% to 40% of patients with
advanced NSCLC, are associated with poor survival. Atezolizumab (anti-PDL1)
inhibits PD-L1/PD-1 signaling and restores tumor-specific T-cell immunity.
Clinical benefits have been observed in patients with NSCLC across PD-L1
expression levels following atezolizumab monotherapy, but the safety
profile in NSCLC patients with brain metastases has not previously been
explored. Methods: Pooled safety analyses were conducted in 843 patients
who received atezolizumab as 2L+ treatment in 4 studies (PCD4989g:
NCT01375842 [N = 76]; BIRCH: NCT02031458 [N = 520]; FIR: NCT01846416
[N = 105]; POPLAR: NCT01903993 [N = 142]). Patients had asymptomatic
untreated brain metastases or stable previously treated brain metastases at
baseline. Results: 27 (3%) of 843 patients in the pooled cohort had baseline
brain metastases; 23 of whom were previously treated with radiation to the
brain. Among the 27 patients, mean age was 60 years, 41% were male, 85% had
non-squamous NSCLC, and 70% had received 3L+ therapy. Median number of
atezolizumab cycles (21d/cycle) was 4 (range, 1-39). Neurological AEs occurred
in 12 (44%) patients with and 229 (28%) patients without baseline brain
metastases (Table). The incidence of treatment-related neurological AEs
was 4 (15%) in patients with and 77 (9%) in patients without baseline brain
metastases, including the most common treatment-related AE of headache
in 2 (7%) and 27 (3%) patients, respectively. The most common all-cause
AEs in patients with baseline brain metastases were fatigue, nausea, and
vomiting (7 [26%] each); 3 (11%) patients developed new brain lesions on
study, none during treatment. No treatment discontinuations occurred due
to AEs. Conclusion: The current analyses indicate that atezolizumab has an
acceptable safety profile in patients with NSCLC who have asymptomatic or
previously treated stable brain metastases. Further investigation is needed
to fully assess the efficacy of atezolizumab in this patient population.
Summary of safety data in patients with advanced NSCLC with and without
baseline brain metastases following atezolizumab as 2L+ treatment
Pooled Cohort (N = 843)
Patients Without Baseline
Brain Metastases (n = 816)
n (%)
Any AE 779 (96%) 26 (96%)
Any neurological AE 229 (28%) 12 (44%)
Treatment-related
AEs
548 (67%) 16 (59%)
Treatment-related
neurological AEs
77 (9%) 4 (15%)
Serious AE 311 (38%) 9 (33%)
Serious neurological
AEs
21 (3%) 1 (4%)
Treatment-related
SAEs
78 (10%) 1 (4%)
Discontinued
treatment due to AE
47 (6%) 0 (0%)
Patients With Baseline
Brain Metastases (n =
27) n (%)
Keywords: Immunotherapy, atezolizumab, brain metastases, NSCLC
S492 Journal of Thoracic Oncology • Volume 12 Issue S1 January 2017