Malignant mesothelioma has a very dismal prognosis with very few patients surviving one year after diagnosis. Early multimodal treatment, however, is expected to improve the outcome. Today, there is a strong need to have disease markers which could be used for screening, diagnosing, and/or monitoring tumour response to treatment. Old markers such as hyaluronic acid, various cytokeratin fragments (CYFRA 21.1, TPA) and other cancer antigens (CA 15.3, CA 125 or CA 19.9 or CEA) are not sensitive or specific enough and cannot be used in practice. More recently new molecules, such as soluble mesothelin and osteopontin, have been proposed for diagnostic purposes. Soluble mesothelin has a good specificity but has a suboptimal sensitivity being negative in all sarcomatoid and in up to one half of epithelioid mesothelioma. On the contrary osteopontin has an inadequate specificity. Combining different markers together does not lead to an improvement in diagnostic accuracy. Neither marker can be used for screening purposes, the main limitation being the very low incidence of the disease in the at-risk, asbestos exposed population. Mesothelin is also a promising marker for monitoring response to treatment but published data is still insufficient to make recommendations. There is still a strong need for research is this area both in order to discover new markers as well as to correct the positioning of each existing molecule (alone or in combination) is the evaluation of the patients with a mesothelioma.