Download the ESMO 2012 Abstract Book - Oxford Journals

Annals of Oncology
Methods: We did a PubMed query using keywords simultaneously (lung neoplasm,
TKI, EGFR mutation, survival). We also searched for relevant abstracts in
proceedings of ASCO, ESMO, WCLC annual meetings. We cross-checked all
references from all articles. Only phase III randomized controlled trials comparing
TKI and chemotherapy were included. We used EasyMA software.
Results: The 6 eligible studies included 2223 patients (516 males, 1688 females,
mostly Asian, median age 60 years, 2155 adenocarcinomas (97 %), 996 mutated
tumors, 389 stage IIIB, 1572 Stage IV (71 %), 1989 never smokers (89.5 %). Four
studies assessed gefitinib, 2 erlotinib. Chemotherapies were doublets including a
platinum salt. Four studies included only mutated patients. Compared to
chemotherapy, EGFR TKIs significantly improved PFS (HR = 0.39, 95 % CI
0.28-0.53, random effect model). Conversely, OS was similar among patients who
first received TKI or chemotherapy (HR = 1.00, CI 0.83-1.22, fixed effect model). PFS
was significantly worse among non mutated patients in the 2 studies including both
mutated and wild-type EGFR patients, whereas OS was unchanged. Concerning
side-effects, rash, diarrhoea and interstitial lung disease were significantly more
frequent after TKI (RRs 5.00; 2.40 and 6.07). As expected, fatigue (RR 0.41), nausea/
vomiting (0.19) and haematological disorders were all significantly more frequent
after chemotherapy (RRs for neutropenia, thrombocytopenia and anaemia 0.08; 0.18
and 0.26).
Conclusions: The major discrepancy between a markedly improved PFS and a
similar OS after TKI compared with chemotherapy could be related to the high level
of crossing-over between the 2 groups. We found no detrimental effect on OS of
TKIs among wild-type patients.
Disclosure: All authors have declared no conflicts of interest.
1266P RETROSPECTIVE STUDY OF CLINICOPATHOLOGIC
FACTORS ASSOCIATED WITH ALK REARRANGEMENT AND
SURVIVAL OUTCOME IN CHINESE PATIENTS (PTS) WITH
NSCLC
X. Zhang 1 , C. Blanckmeister 2 , Y. Cheng 3 ,D.Wu 4 , J. Yang 1 , H. Tian 1 , J. Chen 5 ,
Z. Xie 1 ,H.Yan 1 ,Y.Wu 6
1 Guangdong General Hospital, Guangdong Academy of Medical Sciences,
Guangdong Lung Cancer Institute,, Guangdong, CHINA, 2 Pfizer, New York, NY,
UNITED STATES OF AMERICA, 3 Oncology, Jiling Province Tumor Hospital, Jilin,
CHINA, 4 Jilin Provincial Cancer Hospital, Jilin Province, CHINA, 5 Pfizer China
Medical Affairs, Shanghai, CHINA, 6 Guangdong Lung Cancer Institute,
Guangdong General Hospital (GGH) & Guangdong Academy of Medical
Sciences, Guangzhou, CHINA
Background: We tested 400 NSCLC pts for ALK, EGFR and KRAS status and
correlated clinical factors with ALK status and overall survival (OS).
Methods: ALK status was assessed with RACE-PCR, IHC and FISH; EGFR and
KRAS status were assessed by direct DNA sequencing. OS was estimated by the
Kaplan-Meier method.
Results: Consecutive, unselected cases from southern (n = 294) and northern (n =
106) China were tested; 31 (7.7%), 105 (26.2%) and 33 (8.2%) pts were ALK + ,
EGFR+ and KRAS + , respectively. ALK and EGFR aberrations were largely exclusive
(P = 0.009); 2 cases were ALK + /EGFR+. ALK+ status was associated with female sex
(P = 0.023), non/light-smoking (P = 0.001) and adenocarcinoma (P = 0.017), and was
more prevalent in southern (9.2%; 27/294) than northern (3.8%; 4/106) pts. The
ALK fusion partner was EML4 in all 31 cases; predominantly variants V1–3 (90.3%,
28/31). In 326 modeled pts, female sex (HR = 2.84; P = 0.022), wild type EGFR (HR
= 15.87; P = 0.001), non/light-smoking (HR = 3.95; P = 0.021) and adenocarcinoma
(HR = 4.12; P = 0.031) were associated with ALK+. In 187 cases evaluated by IHC
and RACE-PCR, IHC sensitivity was 100% (19/19; [5 IHC + ; 10 IHC + +; 4 IHC + +
+ ]) and specificity was 79.8% (134/168). All 19 ALK+ cases by RACE-PCR were also
ALK+ by FISH. In the first 303 pts, there were no significant OS differences between
ALK + , EGFR + , KRAS+ and triple-negative pts, between ALK+ and ALK– pts, or
between ALK+ cases matched with non-TKI-treated controls balanced for disease
stage, sex, histology and EGFR/KRAS status. In ALK+ pts, non/light-smoking status
(P = 0.016) and prior surgery (P = 0.024) were associated with increased OS by
log-rank test. In a Cox model, disease stage (P ≤ 0.001 for both stage I and II), and
surgical treatment (P = 0.001) were prognostic for OS.
Conclusion: ALK fusion prevalence was 7.7% overall, and was higher in southern
pts. ALK+ was associated with non/light-smoking, adenocarcinoma and female sex.
ALK fusion variants were mostly V1–3, which may inform RT-PCR screening. IHC
for ALK protein and FISH for ALK rearrangement may be useful for patient
selection. ALK+ was not a favorable prognostic factor, although disease stage and
smoking history may influence OS in ALK+ NSCLC.
Disclosure: C. Blanckmeister: Employment: Pfizer. Myself. Compensated. Stock
ownership: Pfizer. Myself. J. Chen: Employment: Senior Medical Affairs Therapeutic
Area Manager. Pfizer. Myself. Compensated. All other authors have declared no
conflicts of interest.
1267P A LARGE RETROSPECTIVE ANALYSIS OF PEMETREXED
(PEM) ACTIVITY IN PATIENTS (PTS) WITH ALK-POSITIVE
(ALK+) NON-SMALL CELL LUNG CANCER (NSCLC) PRIOR TO
CRIZOTINIB (CRIZ) TREATMENT
G. Scagliotti1 , D.W. Kim2 , A.T. Shaw3 , S.I. Ou4 , G.J. Riely5 , S. Gettinger6 ,
B. Besse7 , K. Wilner8 ,Y.Tang8 , C.H. Bartlett9 1 2
S. Luigi Hospital, University of Turin, Turin, ITALY, Department of Internal
Medicine, Seoul National University Hospital, Seoul, KOREA, 3 Cancer Center,
Massachusetts General Hospital, Boston, MA, UNITED STATES OF AMERICA,
4
Cancer Centre, University of California at Irvine, Irvine, CA, UNITED STATES OF
AMERICA, 5 Medical Oncology, Memorial Sloan-Kettering Cancer Center,
New York, NY, UNITED STATES OF AMERICA, 6 Thoracic Oncology, Yale
University School of Medicine, New Haven, CT, UNITED STATES OF AMERICA,
7 8
Department of Medicine, Institute Gustave Roussy, Villejuif, FRANCE, Reseach
and Development, Pfizer Oncology, La Jolla, CA, UNITED STATES OF AMERICA,
9
Medical Oncology, Pfizer Oncology, New York, NY, UNITED STATES OF
AMERICA
Background: Retrospective small cohort studies evaluating multiple lines of therapy
suggest ALK+ NSCLC may be particularly sensitive to PEM treatment.
Methods: PROFILE 1005 (NCT00932451; Pfizer) is a large phase 2 multinational,
single-arm study of CRIZ in previously treated ALK+ NSCLC. We retrospectively
assessed best overall response rate (ORR) and time to progression (TTP; 1st dose to
objective progression) in pts who received PEM-based regimens prior to CRIZ in the
1st-line (1L) and 2nd-line (2L) settings. ORR to CRIZ post-PEM-based regimens was
assessed.
Results: Of the 901 ALK+ pts enrolled as of Jan 2012, 711 (79%) received prior PEM
(single-agent or combination; any line advanced/metastatic) with an ORR to PEM of
19%. Pts who received 1L PEM combinations (n = 308) were predominately of young
age (median 53 y, 82%