INK4A1 p16 inactivation mechanisms in non small-cell lung cancer patients
2 occupationally exposed to asbestos
a-cde-gh,i4 Pascal Andujar, Jinhui Wang, Alexis Descatha, Françoise Galateau-Sallé,
jkl,m5 Issam Abd-Alsamad, Marie-Annick Billon-Galland, Hélène Blons, Bénédicte
h,no,pa-cl,m6 Clin, Claire Danel, Bruno Housset, Pierre Laurent-Puig, Françoise Le
q,rh,na7 Pimpec-Barthes, Marc Letourneux, Isabelle Monnet, Jean-François
sp,rp,ra-c8 Régnard, Annie Renier, Jessica Zucman-Rossi, Jean-Claude Pairon,
p,r* 9 Marie-Claude Jaurand.
a 11 HôpitalIntercommunal de Créteil, Service de Pneumologie et de Pathologie
12 Professionnelle, Créteil, F-94010, France
b13 INSERM, U955, Equipe 4, Créteil, F-94010, France
c14 Université Paris 12, Faculté de Médecine, Créteil, F-94010, France
d15 INSERM, E0337, Créteil, F-94010, France
e16 AP-HP, GHU Ouest, Hôpital Poincaré, Unité de Pathologie Professionnelle,
17 Environnementale et d'Insertion, Garches, F-92380, France
f18 INSERM, U687, Villejuif, F-94807, France
g19 Université de Versailles Saint-Quentin-en-Yvelines, Faculté de médecine
20 Paris-Ile-de-France-Ouest, Guyancourt, F-78280, France
h21 INSERM, ERI 3, Caen, F-14000, France
i22 CHU Caen, Service d’Anatomo-Pathologie, Caen, F-14033, France
j23 Hôpital Intercommunal de Créteil, Service d’Anatomo-Pathologie, F-94010,
24 Créteil, France
k25 Laboratoire d’Etude des Particules Inhalées, Paris, F-75013, France
l26 INSERM, UMR-S775, Paris, F-75006, France
m27 Université Paris Descartes, Paris, F-75005, France
n28 CHU Caen, Service de Médecine du Travail et Pathologie Professionnelle, Caen,
29 F-14033, France
o30 AP-HP, GHU Ouest, Hôpital Européen Georges Pompidou, Service
31 d’Anatomo-Pathologie, Paris, F-75015, France
p32 Université Paris 7, Paris, F-75005, France
q33 AP-HP, GHU Ouest, Hôpital Européen Georges Pompidou, Service de Chirurgie
34 Thoracique, Paris, F-75015, France
r35 INSERM, U674, Fondation Jean Dausset-CEPH, IFR105, Paris, F-75010,
s37 Département de Chirurgie Thoracique et Vasculaire, et de Transplantation
38 Cœur-Poumon, Centre Chirurgical Marie Lannelongue, Le Plessis-Robinson, 39 F-92350, France
41 *Corresponding author: Phone: +33 1 53 72 51 88; Fax: +33 1 53 72 51 92
42 Email address: firstname.lastname@example.org (MC. Jaurand)
45 Present addresses:
47 J. Wang: Division of Biology, Beckman Research Institute, City of Hope National
48 Medical Center, Duarte, CA 91010, California, United States of America.
49 C. Danel: AP-HP, GHU Nord, Hôpital Bichat-Claude Bernard, Service
50 d’Anatomie et de Cytologie Pathologiques, 46 rue Henri Huchard, F-75018 Paris,
52 JF Régnard: AP-HP, GHU Ouest, Hôpital Hôtel-Dieu, Service de Chirurgie
53 Thoracique, 1 place du Parvis de Notre-Dame, F-75004 Paris, France.
57 Epidemiological studies have shown that asbestos fibers constitute the major
58 occupational risk factor and that asbestos acts synergistically with tobacco
59 smoking to induce lung cancer. Although some somatic gene alterations in lung
60 cancer have been linked to tobacco smoke, few data are available on the role of
61 asbestos fibers. P16/CDKN2A is an important tumor suppressor gene that is 62 frequently altered in lung cancer via promoter 5’-CpG island hypermethylation 63 and homozygous deletion, and rarely via point mutation. Many studies suggest
64 that tobacco smoking produces P16/CDKN2A promoter hypermethylation in lung 65 cancer, but the status of this gene in relation to asbestos exposure has yet to be
66 determined. The purpose of this study was to investigate the mechanism of
67 P16/CDKN2A alterations in lung cancer in asbestos-exposed patients.
68 P16/CDKN2A gene status was studied in 75 human non-small-cell lung cancer
69 (NSCLC) cases with well-defined smoking habits, and detailed assessment of
70 asbestos exposure, based on occupational questionnaire and determination of
71 asbestos bodies in lung tissue. The results of this study confirm published data
72 on the effect of tobacco smoke on P16/CDKN2A gene alterations, characterized 73 by significantly higher P16/CDKN2A promoter hypermethylation in heavy 74 smokers (more than 40 Pack-Years (P-Y)) than in smokers of less than 40 P-Y.
75 These results also demonstrate a higher incidence of loss of heterozygosity and
76 homozygous deletion in asbestos-exposed cases, after adjustment for age and
77 cumulative tobacco consumption, than in unexposed cases (P = 0.0062). This 78 study suggests that P16/CDKN2A gene inactivation in asbestos-exposed
79 NSCLC cases mainly occurs via deletion, a feature also found in malignant
80 mesothelioma, a tumor independent of tobacco smoking but associated with
81 asbestos exposure, suggesting a possible relationship with an effect of asbestos
84 Key words
85 Lung cancer; asbestos; occupation; tobacco smoking; P16/CDKN2A; INK4A; 86 deletion; hypermethylation.
88 1. Introduction
90 Lung cancer is still the leading cause of cancer-related death in the world .
91 Environmental and occupational factors, and genetic susceptibility interact to
92 influence lung carcinogenesis . About 90% of lung cancer risks are
93 attributable to tobacco smoking . Other environmental, occupational and 94 genetic factors also contribute to the development of lung cancer.
95 Epidemiological studies have shown that asbestos fibers constitute the major
96 occupational risk factor and that asbestos acts synergistically with tobacco
97 smoking to induce lung cancer [4-6]. Molecular analyses of lung cancer cells 98 have demonstrated that some alterations in oncogenes and tumor suppressor
99 genes can be associated with the risk factor, especially tobacco smoke. In
100 contrast, other molecular changes were not associated with exposure to certain
101 risk factors. For instance, mutations in TP53 and KRAS genes and P16CDKN2A
102 promoter 5’-CpG island hypermethylation in lung cancer were associated with
103 tobacco smoking, while EGFR mutations were mainly found in lung 104 adenocarcinoma in never smokers [7-9]. In this context, it has been suggested
105 that some gene alterations could be markers of exposure to specific
106 carcinogenic factors. The nature of genetic alterations attributable to exposure to
107 asbestos fibers remains to be defined in lung cancer.
108 P16/CDKN2A is an important tumor suppressor gene showing both genetic or
109 epigenetic alterations frequently found in cancers. In lung cancer, P16/CDKN2A
110 is inactivated via promoter 5’-CpG island hypermethylation and homozygous
111 deletion, and rarely via point mutations [10-14]. According to published data,
112 P16/CDKN2A promoter hypermethylation in NSCLC is observed in about 20 to
113 40% of cases [15-20]. Gene deletion and point mutations also contribute to loss
INK4A114 of p16 expression, as these alterations have been reported to occur with a
115 frequency of 10 to 30% and less than 5%, respectively [21-23].
116 The relationship between P16/CDKN2A alteration and tobacco smoking has
117 been largely investigated. According to Kim et al. (2001), who studied a large
118 series of 185 non-small-cell lung cancer (NSCLC) cases, P16/CDKN2A
119 promoter hypermethylation was more likely found in current smokers than in
120 non-smokers, and was associated with pack-years (P-Y) and duration of
121 smoking, suggesting that the mechanism of action of tobacco smoke involves
122 P16/CDKN2A promoter hypermethylation . Similarly, P16/CDKN2A promoter
123 hypermethylation was statistically associated with tobacco smoking in a series of
124 51 NSCLC cases . In another study, allelic loss at chromosome 9p21 was
125 associated with tobacco smoking in 47 NSCLC, but no association was
126 observed between tobacco smoking and P16/CDKN2A homozygous deletion or
127 promoter hypermethylation . In contrast, a link between homozygous
128 deletion and never smoking status was reported by Kraunz et al. .
129 In contrast with the numerous studies on tobacco smoking, few data are
130 available on P16/CDKN2A gene status in lung cancer in asbestos-exposed
131 subjects. No statistically significant difference was observed between
132 asbestos-exposed and unexposed patients in four studies investigating
133 P16/CDKN2A gene inactivation [13, 16, 18, 25]. Nevertheless, one study found
134 that P16/CDKN2A promoter hypermethylation was significantly enhanced in
135 asbestos-exposed cases compared to unexposed cases . More details on 136 P16/CDKN2A gene status are available in malignant mesothelioma, a cancer
137 related to asbestos exposure but not to tobacco smoking. In this disease,
138 P16/CDKN2A gene is generally altered via homozygous deletion [27-29].
139 The purpose of this study was to investigate the mechanism of P16/CDKN2A 140 alterations in lung cancer of asbestos-exposed patients. P16/CDKN2A gene 141 status was studied in 75 cases of human NSCLC with well-defined smoking
142 habits, and detailed assessment of asbestos exposure, based on both
143 occupational questionnaire and determination of asbestos bodies in lung tissue.
144 After adjustment for age and cumulative tobacco consumption, the results
145 showed higher P16/CDKN2A promoter hypermethylation in unexposed cases
146 and loss of heterozygosity (LOH) and homozygous deletion (HD) of
147 P16/CDKN2A in asbestos-exposed cases.
149 2. PATIENTS AND METHODS
151 2.1. NSCLC population
153 Patients: Patients were selected from consecutive cases of primary NSCLC after 154 surgical resection at 4 hospitals (Centre Hospitalier Intercommunal, Créteil; 155 Hôpital Européen Georges Pompidou, Paris; Centre Hospitalier Universitaire, 156 Caen; Centre Chirurgical Marie Lannelongue, Plessis-Robinson) from January 157 1994 through June 1999. The study was approved by the local Ethics Committee 158 and all patients provided written informed consent. The following eligibility 159 criteria were required: (a) lung tumor histology; (b) absence of neoadjuvant 160 chemotherapy or radiotherapy; (c) both normal and tumor lung tissue available; 161 (d) data on asbestos exposure history including quantification of asbestos 162 bodies (AB) in lung tissue and interviewer-administered questionnaire; and (e) 163 data on smoking habits. A total of 75 patients, 34 asbestos-exposed and 41 164 unexposed, were selected.
165 Tissue samples were snap-frozen in liquid nitrogen after surgical resection and 166 stored at -80?C until use. Detailed information describing the tumors was 167 obtained from pathology reports.
168 Smoking habits: Information regarding smoking status was obtained from an 169 interviewer-administered questionnaire: smoking class, i.e never smokers, 170 current smokers and former smokers (quitting smoking at least 1 year before 171 diagnosis); age at onset of smoking, smoking duration and tobacco consumption 172 (cigarettes, cigars and pipes), expressed as pack-years (P-Y). Heavy smokers 173 were defined as patients who had smoked more than 40 P-Y. This value 174 represents the median of the overall study population.