1192PD - Lace-bio pooled analysis of the prognostic and predictive value of TP53 mutations in completely resected non small cell lung cancer (NSCLC)

Date 29 September 2012
Event ESMO Congress 2012
Session NSCLC - Immunotherapy, SCLC and Mesothelioma
Topics Non-Small Cell Lung Cancer
Translational Research
Basic Principles in the Management and Treatment (of cancer)
Presenter Pierre Hainaut
Authors P. Hainaut1, X. Ma2, B. Lacas3, M. Tsao4, J. Douillard5, V. Rousseau6, A. Dunant6, L.K. Seymour7, M. Filipits8, S. Graziano9
  • 1International Prevention Research Institute, 69006 - Lyon/FR
  • 2Jinan Central Hospital, Shandong University, Jinan/CN
  • 3Meta-analysis Unit, Institut Gustave-Roussy, Villejuif/FR
  • 4Hematology And Oncology, Princess Margaret Hospital, Toronto/CA
  • 5Medical Oncology, Centre René Gauducheau, Saint-Herblain/FR
  • 6Biostatistics, Institut Gustave Roussy, Villejuif/FR
  • 7Ncic Clinical Trials Group, Queen's University, Kingston/CA
  • 8Medicine, Medical University of Vienna, Vienna/AT
  • 9Hematology And Internal Medicine, SUNY Upstate Medical University Hospital, Syracuse/US



Mutations in the Tumour Suppressor Gene TP53 occur in about 50% of Stage II-III NSCLC. However, prognostic and predictive value of survival after adjuvant therapy not defined. We report here on the pooled analysis of 4 randomized trials of platinum-based Adjuvant ChemoTherapy (ACT) or observation.


Trials included IALT, JBR10, CALGB-9633 and ANITA. Mutations in exons 5-8 of TP53 (encompassing hotspots and over 90% of known mutations) were detected by PCR/sequencing in DNA extracted from formalin-fixed paraffin-embedded NSCLC. Mutations were classified into 3 structure/function groups: non-missense; missense outside DNA Binding Notifs (missense non-DBM); missense DBM. Main endpoint was overall survival (OS). Hazard ratio (HR) and 95% confidence interval (CI) were estimated with a Cox model stratified on trial and adjusted on gender, age and clinico-pathological variables.


1209 NSCLC patients (pts) (5.5 years median follow-up) were analysed. Mutations were detected in 434 pts (36%; IALT: 42%; JBR10: 28%; CALGB-9633: 39%; ANITA: 30%). Mutant TP53 had a borderline significant prognostic effect (HR for OS, mutation vs. wild-type (WT): 1.19; CI [1.00-1.41]; p = 0.05). Missense non-DBM mutations, likely to alter p53 protein structure, had a marginally worse prognostic effect (HR for OS, missense non-DBM vs. WT: 1.38; CI [1.08-1.75]), but effects in the 3 function groups were not significantly different (p = 0.23). Among pts with WT TP53, OS was better in the ACT group (HR for OS, ACT vs. no ACT: 0.77; CI [0.62-0.96]; p = 0.02), whereas ACT had no effect among pts with mutation. Conversely, in the ACT group, OS was significantly shorter in pts with TP53 mutation than in WT pts (HR for OS, Mutant vs. WT TP53: 1.39; CI [1.09-1.77]; p = 0.008), whereas TP53 had no prognostic effect in the control group. The ACT effect in the mutated and WT groups and the prognostic effect in the ACT and control groups were not different (interaction test p = 0.07).


TP53 mutation status had a borderline prognostic effect and was not predictive of adjuvant chemotherapy effect on OS in NSCLC.

Supported by French Ligue against Cancer (LNCC) and by unrestricted grant from Sanofi Aventis. XM was supported by a fellowship from International Agency for Research on Cancer


All authors have declared no conflicts of interest.