Occupational lung cancer generates high mortality and is the most common cancer compensated in France (1). In 1997, the Helsinski criteria for identifying individuals with a high risk of asbestos exposure at work were accepted (2). Older studies found inconsistent results regarding the lobe of origin and histology of asbestos-related lung cancer (ARLC) (3). Some studies showed an upper lobe location similar to tobacco-related lung cancer, whereas other investigators found a lower lobe location (4–11). Although adenocarcinoma was found to be the most prevalent in some studies, a recent review of the literature by Nielsen et al (3) showed that there was no difference in location and cell type between ARLC and non-ARLC. They concluded that cell type and location of lung cancer were not useful for differentiating ARLC from other lung cancers. Prognosis of ARLC was not different from that of other lung cancers (3).

Presence of pleural plaques demonstrated a previous asbestos exposure but was not a precancerous condition. Asbestosis argued for high asbestos exposure and was associated with an increased risk of lung cancer (3). All asbestos types were associated with lung cancer (3).

Asbestos-exposed smokers had a higher risk of developing lung cancer compared with asbestos-exposed non-smokers. In fact, smoking was found to be the main risk factor but its interaction with asbestos was not totally clear (3). Some studies showed a multiplicative effect, whereas other studies suggested an additive model (12–15).

The present study investigated the occupational data of patients with compensated lung cancer, their smoking habits and histological subtype of lung cancer. The aim of the present study was also to discuss the criteria for warding occupational cancer compensation.

After ethics approval was obtained from our institutional review board (CHRU Brest, 2016. CE10), we led a retrospective study based on financially compensated ARLCs identified in occupational disease center between 2003 and 2013. Data on occupational history and smoking habits were collected. Ex-smokers were defined as those who had ceased smoking at the time they consulted in our occupational disease center. Non-smokers were defined as those who had never smoked in their lifetime. Tumor location and histological subtype of lung cancer were noted according to the last pathological data. Patients who had two occupations were recorded separately.

The present study included 146 male patients of whom the majority (90%) were heavy current or former smokers (mean, 30.4 packs/year).

To the best of our knowledge, this is the first comprehensive study conducted in France on occupational asbestos exposure and lung carcinoma pathological features. The link between lung cancer and asbestos exposure is sometimes difficult to attest (16). Two criteria for this recognition are used in many countries (Germany, Austria, Belgium, Finland, Norway, Sweden and Switzerland): proved intense exposure (Helsinki criteria or threshold of 25 fibers/ml/year) and/or asbestos-related conditions (asbestosis and/or extensive modification of the pleura) (17). In France, occupational disease tables with specific criteria are used to determine whether or not lung cancer is asbestos-related (Tables 30 and 30 bis) or not. The diagnosis must be based on histological data or, if not available, on the basis of a clinical course and suggestive radiological data (17). Many differences regarding criteria exist between European countries for the recognition of ARLC (Table III) (17).

Shipbuilding and repair are the main industry in Brest, a French port located at the Department of Occupational and Environmental Diseases. The workers most exposed to asbestos were involved in this activity. Approximately 90% of compensated lung cancer patients were current or ex-smokers in the present study. Very few studies described smoking habits in this population (3). A synergistic interaction between asbestos and cigarette smoking is commonly accepted. A review of 23 studies was in favor of a multiplicative interaction (13). However, the multiplicative effect claimed by Hammond et al (12) has never been replicated and the hypothesis of a model somewhere between additive and multiplicative is the most probable (3). In most European countries, smoking is not involved in the recognition procedure of ARLC (17). However, in Denmark, any doubt regarding exposure of asbestos, heavy smoking leads to rejection of the procedure (17). If the patient consumed more than 7 g of tobacco per day or more than 10 packs a year, smoking was taken into account for the compensation of lung and laryngeal cancer. Those considerations also applied to cases of asbestosis if obstructive airways disease or chronic bronchitis was revealed in the past clinical history (17). However, the smoking factor is never retained in cases of mesothelioma or pleural plaques (17).

Adenocarcinoma was the most prevalent histological subtype in the present study (45.9%). Excess of adenocarcinoma in ARLC was shown in some studies (3). Nevertheless, recent well-controlled studies failed to show any significant differences between ARLC and non-ARLC (3). Adenocarcinoma is in fact the most frequent type of lung cancer. According to a recent review of the literature, histology has no significant value in determining whether or not lung cancer can be due to asbestos (3).

Notably, thoracic sarcoma was identified in the present study. To the best of our knowledge, if the link between lung carcinoma and asbestos exposure is well established, this is not true about thoracic sarcoma (6,8–11,18–20). Further investigations also identified two non-primary lung origin carcinomas of esophageal and laryngeal origin in the present study with controversial data regarding the association with asbestos exposure in the literature. In addition, there was no histological proof for 6.8% of the compensated lung cancers, diagnosis being solely based on radiologic data. Metastases of non-ARLC or non-carcinoma tumors cannot be excluded for these cases. In our opinion, histological analysis is a major parameter to evaluate the relationship between asbestos exposure and a lung malignant tumor. It may permit the distinction between a carcinoma of lung origin in which the link with asbestos exposure is well established, and other more doubtful asbestos-related primary or metastatic tumors.

In conclusion, the present results are globally in accordance with other studies on ARLC. Even if the relationship between smoking and asbestos exposure is not clear, the cases reported in the present study strongly argue for the crucial importance of smoking cessation in asbestos-exposed workers.

Thus, histological data have to be considered to evaluate the potent relationship between asbestos exposure and lung malignancy. Indeed, although the relationship between asbestos and lung carcinoma is well established, it is insufficiently proven regarding non-carcinoma type and/or cancer from extra-pulmonary primitive origin, except for mesothelioma.