Malignant Mesothelioma

Bruce WS Robinson, Arthur W Musk, Richard A Lake. The Lancet. Volume 366, Issue 9483. July 30-August 5, 2005.


The fact that asbestos can cause cancer is now widely known in western countries, and the public are increasingly familiar with the word mesothelioma, especially since the deaths of some well known individuals with the disease, such as actor Steve McQueen and scientist and author Stephen Jay Gould.

Why this change in awareness? One reason is that mesothelioma has climbed the league table of male cancers over the past 30 years, and is now roughly as common as cancers of the liver, bone, and bladder, especially in Europe and Australia. Its incidence is expected to continue to increase for the next decade or so. Secondly, almost everyone who lives in industrialised areas of the western world has asbestos fibres in their lungs, and many can remember being exposed to asbestos incidentally (eg, carpenters, plumbers, military personnel, school teachers, and students who handled asbestos samples, mats, or blankets; home renovators; and people in many other situations). Media interest in asbestos has produced in many of these individuals a level of awareness—even anxiety—about mesothelioma that does not exist for most other sporadic cancers of comparable incidence. These issues, combined with the complex medical-legal aspects of the disease, have led to a lot of interest. For example, a simple Google search of the common cancers at the time of writing identified nearly 3 million webpage results for mesothelioma, second only to breast cancer and substantially more than the number of results for other well known cancers such as lung cancer, leukaemia, lymphoma, and bowel or colon cancer.

In this Seminar, we aim to review mesothelioma, highlighting key clinical features plus some controversies, recent developments, and important questions for future research into this disease.


There are few disease processes for which the distribution and determinants of occurrence in the community are as well known as they are for mesothelioma of the pleura and peritoneum. In fact, mesothelioma owes its entire existence as a disease entity to its relation with asbestos, first recognised by an astute pathologist working in South Africa’s Cape Province in the 1950s. As an industrial disease, it is one of the most accurately predictable. Of all the outcomes from the production and use of asbestos, mesothelioma is likely to be the one that will be most remembered.

Few common cancers have such a direct causal relation with an exposure to a defined carcinogen as mesothelioma has with asbestos exposure—even lung cancer with cigarette smoking. The future occurrence of mesothelioma can be predicted from the pattern of asbestos use around the world. The only other generally recognised causes of mesothelioma are: endemic erionite exposure in Turkey; ionising radiation, especially in patients for whom thorotrast was used as a radiographic contrast material in the 1950s; and chest injuries. The potential role for the SV40 virus is discussed later.

Mesothelioma remains a universally fatal disease of increasing incidence worldwide. Median survival from presentation is 9-12 months. Based on World Health Organization reports, mesotheliomarates from different countries show large differences by sex and country. Male rates are much higher than female rates in virtually all countries, rising from the sporadic background rate of around 1 per million to over 40 per million in some countries. Industrialised countries have much higher rates than non-industrialised countries, reflecting the past production and use of asbestos in industry.

The developed world has seen an epidemic of mesothelioma in the 60 years since World War II because of the demand for asbestos of all varieties that accompanied industrialisation. This epidemic did not take place immediately because of the long incubation period between onset of disease and time since first exposure. This is typically longer than 30 years, although incubation periods of around 15 years have been described. Unlike bronchogenic carcinoma, mesothelioma is not related to cigarette smoking, presumably because the carcinogens in cigarette smoke do not reach the mesothelium. By contrast with the relation between tobacco smoke exposure and lung cancer, in which the risk of developing the disease declines after quitting and approaches that of someone who has never smoked after 25-30 years, the risk of mesothelioma—virtually zero for the first 10-15 years after first exposure—progressively increases with time. Even though clearance of crocidolite, the most durable of all forms of commercially used asbestos, from the lungs is about 10% per year, a rate estimated from lung-fibre-content studies and epidemiological studies, the risk of mesothelioma increases throughout life.


How does asbestos cause mesothelioma?

A single layer of mesothelial cells covers the entire surface of the pleura; these cells aid free movement of the pleural surfaces during respiration by enmeshing lubricating glycoproteins, and are readily able to proliferate in response to injury and growth factors. At least four plausible explanations have been proposed as to how a mineral fibre such as asbestos causes malignant change in these cells.

First, there is pleural irritation. Good epidemiological evidence shows that the fibre shape and length-to-width ratio is an important physical attribute that determines how deeply into the lung the fibres are inhaled and whether they then have the capacity to penetrate the lung epithelium and enter or irritate the pleural space. The most dangerous fibres are long and thin and can penetrate the lung, repeatedly scratching the mesothelial surface and causing prolonged cycles of damage, repair, and local inflammation. This process can lead to scarring (plaques) or cancer (mesothelioma). Second is interference with mitosis. Asbestos fibres have the capacity to interfere with the mitotic process. Fibres can sever or pierce the mitotic spindle, disrupting mitosis, which has the potential to lead to aneuploidy and the other forms of chromosome damage that characterise mesothelioma. Generation of toxic oxygen radicals is a third explanation. Asbestos-induced cell damage is mediated to some extent by iron-related reactive oxygen species, which induce DNA damage and strand breaks. And fourth is persistent kinase-mediated signalling. Asbestos fibres induce phosphorylation of the mitogen—activated protein kinases and extracellular signal—regulated kinases 1 and 2 and elevate expression of early response protooncogenes (FOS or JUN or activator protein 1 family members) in mesothelial cells. Although several growth factors and their receptors drive mesothelioma proliferation, the only factors whose blockade has been shown to arrest mesothelioma growth are transforming growth factor β and platelet-derived growth factor A chain.

What is the role of SV40 in mesothelioma?

SV40 has been implicated as a cofactor in the causation of malignant mesothelioma. The polyoma viruses are a group of small double-stranded DNA viruses, and SV40, which blocks rumour suppressor genes, is a potent oncogenic virus for human and rodent cells. SV40 DNA sequences have been found in brain and bone tumours, lymphomas, and malignant mesotheliomas as well as in atypical mesothelial proliferations and superficial non-invasive lesions of the mesothelium. SV40 infection is semipermissive in human cell lines and up to 60% of human mesotheliomas contain SV40 DNA. According to microdissection experiments, SV40 is present in the malignant cells and sometimes in reactive mesothelial cells but not in normal adjacent tissues or in lung cancers. SV40 was disseminated widely throughout the world in the Salk polio vaccine in the 1950s and 1960s and can possibly also be contagiously transmitted horizontally in human beings, although convincing evidence connecting vaccine-derived infection and the occurrence of mesothelioma has not been forthcoming. Overall, any role for SV40 in the pathogenesis of mesothelioma remains unclear and unproven. Asbestos remains the major causative agent.

Understanding the molecular pathology of mesothelioma

Molecular studies in animals and people suggest a genetic susceptibility to asbestos-induced tumour formation. Conventional cytogenetic analysis shows that most mesotheliomas have abnormal karyotypes, often with extensive aneuploidy and structural rearrangements. Loss of chromosome 22 is the most common gross change, but structural rearrangement of 1p, 3p, 9p, and 6q are often noted. Neurofibromatosis type 2 (NF2) is present at 22q12 and loss of heterozygosity has been reported in 100% of mesothelioma cell lines.52 Overall, the commonest abnormalities seen are loss of P16^sup INK^sub 4^A^, P14^sup AKF^, and NF2, which implies that a particular pattern of tumour-suppressor gene loss is necessary for mesothelioma development. New animal models of mesothelioma could help determine the sequence of these in mesothelioma development.

Clinical presentation

Typically, a patient with mesothelioma presents with a pleural effusion, which is often associated with chest wall pain. In fact, mesothelioma should be suspected in any individual with an unexplained pleural effusion associated with chest wall pain. Constitutional symptoms such as weight loss and fatigue can be present, but these generally appear later in the course of disease-the presence of such symptoms at presentation is associated with a poor prognosis. Occasionally, patients have no symptoms and their pleural disease is found on a routine chest radiograph undertaken for insurance or preoperative reasons. Patients with peritoneal mesothelioma tend to present with abdominal distension, pain, and occasionally bowel obstruction. Mesotheliomas of the pericardium or tunica vaginalis are rare, but tend to present with features of pericardial effusion or tamponade, or blood-stained hydrocoele, respectively.

Although occasionally patients have no physical signs (eg, those with a localised pleural mass only), the most common signs are those typical of a pleural effusion. A fixed hemithorax (lack of expansion of one side of the chest due to extensive tumour) is a relatively late sign. Clubbing is rare. Patients with peritoneal mesothelioma present with signs of ascites, tenderness, and, especially later in their course, palpable masses. Subcutaneous masses are rare in the absence of surgical intervention, and are almost always associated with operative procedures or the insertion of intercostal drainage tubes. They can grow to an enormous size. Signs of local invasion such as superior vena caval obstruction and Horner’s syndrome are rare at presentation.

Although metastatic deposits of mesothelioma are fairly common at post mortem, these deposits rarely manifest clinically. The most common sites of spread are the hilar, mediastmal, internal mammary, and supraclavicular lymph nodes. Metastasis occurs to major organs, such as bone, and, occasionally, miliary spread is apparent. Local invasion involves other contiguous organs such as the spinal cord (resulting in back pain and paralysis), the pericardium (resulting in pericardial effusions and tamponade), and the contralateral lung (eg, presenting as contralateral pleural effusion).


Accurate diagnosis of mesothelioma is important for several reasons, not least for clinical management of the patient. Issues of compensation also exist and, because the likelihood of survival a year beyond diagnosis is less than 50%, there may be some urgency to expedite the diagnosis for litigation purposes, especially if the jurisdiction will only accept cases brought on behalf of a living victim.

The most common diagnostic problem is distinguishing mesothelioma from adenocarcinoma, especially when tumour has invaded the pleura. Both these diseases tend to arise in elderly men and both often present with chest pain, pleural effusion, and respiratory distress. Mesothelioma usually presents as multiple greyish tan nodules in a diffusely thickened pleura, and this presentation is also common in adenocarcinoma.


Can cytopathology be sufficient to diagnose mesothelioma?

There has been a gradual shift over the past decade towards acceptance of the diagnosis of mesothelioma from cytological specimens, given the quality of current immunohistochemical and ultrastructural methods of assessment. Closed-needle biopsy (eg, with the Abrams needle) is used less commonly nowadays because it causes substantial discomfort and yet often produces inadequate or inconclusive biopsy specimens. Although cytology cannot always provide a definitive diagnosis, it is often sufficient in the appropriate clinical context.

Typically, fluid from a serous effusion is sent to the cytology laboratory and centrifuged for smears and a cell block is prepared, with material set aside for ultrastructural examination by electron microscopy. Fine-needle aspiration cytology is useful for sampling pulmonary, chest wall, or lymph node masses, especially in the absence of an effusion. Thin-core biopsy samples are generally used if there is a substantial pleural-based lesion, especially in the absence of an accompanying effusion. It is well tolerated and provides sufficient tissue for electron microscopy as well as histopathology.

The immunocytochemical stains used to distinguish mesothelioma from adenocarcinoma and other diseases are listed later, but, importantly, calretinin identifies cells as being of mesothelial origin, and epithelial membrane antigen staining in cytological samples in a thick membrane distribution is highly suggestive of mesothelioma.


When cytological examination of pleural effusion to distinguish adenocarcinoma and mesothelioma is inconclusive, a biopsy sample of tumour tissue is usually taken for histopamological analysis. Since, however, adenocarcinoma resembles mesothelioma macroscopically and microscopically, biopsy can also be unhelpful. Microscopically, both tumours can form papillary structures. The tumour cells in an epithelioid mesothelioma are usually more uniform, cuboidal, and less crowded than those in adenocarcinoma, which are more often pleomorphic, columnar, and crowded with nuclear moulding. However, even with these criteria, it is often difficult to differentiate mesothelioma from adenocarcinoma in slides stained with haematoxylin and eosin. A suite of immunohistochemical markers have been developed to aid differential diagnosis of mesothelioma.

Examination of mesothelioma tissue does help to determine the extent of mesothelial proliferation, the degree of cytological atypia, and the degree of invasion. Genuine invasion due to mesothelioma can be distinguished from entrapment of reactive mesothelial cells by its extent and by the presence of a so-called raining down pattern, with insinuation of malignant cells into adjacent structures. Sarcomatoidmesothelioma has keratin positive malignant spindle cells sometimes with extensive collagen deposition. If the latter is predominant, the term desmoplastic is applied.

Immunohistochemistry is important for determining that the tissue of origin is mesothelial (eg, calretinin) and that it is malignant mesothelioma, especially epithelial membrane antigen, expression on the luminal aspects of the tumour. Cytokeratin stains are important for confirming invasion and distinguishing mesothelioma from sarcomas and melanoma. More than 85% of epithelioid mesotheliomas stain for epithelial membrane antigen, calretinin, Wilms’ tumour 1 (WT1), cytokeratin 5/6, human mesothelial cell 1 (HBME-1), or mesothelin. Mesothelioma must be distinguished from adenocarcinoma. CEA, CD15, TTF-1, and B72 are almost never expressed in malignant mesothelioma, whereas calretinin and epithelial membrane antigen suggest mesothelioma. Where results are equivocal by immunohistochemistry, electron microscopy helps distinguish mesothelioma from adenocarcinoma, and that approach is of greater diagnostic value than additional immunohistochemical analyses. Electron microscopy is also occasionally necessary to distinguish desmoplastic or sarcomatoid mesothelioma from fibrous pleuritis.

Mesothelioma in situ (atypical mesothelial proliferation) is assumed to be the earliest lesion similar to cellular atypia in cervical dysplastic lesions, but this notion has not been proven.


At presentation, the conventional chest radiograph shows either pleural effusion or, occasionally, a pleural-based mass. In the few patients who initially present with an advanced tumour, an encircling rind of tumour and/or extensive lobulated pleural-based tumour masses are evident.

CT scanning often shows only pleural effusion or pleural-based masses with or without thickening of interlobular septae at presentation. A CT scan is also used for identification of signs of asbestos exposure (eg, calcined plaques). It is unknown why some forms of mesothelioma produce localised lobular thickening whereas in other cases the thickening produces an essentially uniform rind of tumour that encases the lung. CT scanning of patients with peritoneal mesothelioma tend to show ascites and scattered mesothelioma masses.

A rounded atelectasis, sometimes known as folded lung sydrome, is a benign pleurally-based mass lesion associated with asbestos, but it can mimic mesothelioma or even lung cancer. It forms as a result of the organisation of local effusions and fibrosis, which fix a portion of the lung to the pleura and entrap the lung when the lung rotates after re-expansion. Typically, it is a rounded peripheral mass with a comet-shaped tail. MRI is useful for determination of the extent of mesothelioma, especially invasion of local structures such as rib and diaphragm.

Positron emission tomography (PET) is used as an adjunct to diagnosis in some centres, with avid uptake suggesting malignant disease. It is also useful for identification of extrathoracic disease, especially lymph node involvement. Hypermetabolic lymph node involvement is often evident when the lymph nodes appear normal on a CT scan.

Staging of malignant mesothelioma

Although staging is of little use for medical management, it has implications for surgical management. Because various staging systems have been developed in the past few decades, the International Mesothelioma Interest Group developed a new staging system in 1994 based on the TNM system of lung cancer. Their staging system parallels prognosis. Accurate preoperative assessment requires CT, MRI, PET, and often thoracoscopy and mediastinoscopy. Final staging is only possible at surgery.

Serum mesothelin-related protein (SMRP) for mesothelioma diagnosis

Until recently, no reliable serum marker of mesothelioma had been identified. SMRP is the circulating product of mesothelin, a surface protein thought to be important in mesothelial cell adhesion and possibly signalling; it is possibly the protein product of an alternatively spliced variant of mesothelin mRNA. In a study of 44 mesothelioma patients, 84% had elevated concentrations of SMRP compared with fewer than 2% of patients with other pleural or pulmonary inflammatory or malignant diseases. Thus SMRP had a sensitivity of 84% and specificity approaching 100% compared with other lung tumours or pleural diseases, and a specificity over 80% compared with other people who had been exposed to asbestos. SMRP concentrations parallel disease progression and regression, and, importantly, are elevated in more than 60% of mesothelioma patients at diagnosis. SMRP will probably be used as an adjunct for mesothelioma diagnosis.

There are also clues that serum SMRP could prove useful in the early detection of mesothelioma in individuals who are at risk-three of seven healthy asbestos-exposed individuals who had elevated SMRP concentrations presented 1-6 years later with mesothelioma.

DNA-microarray-based strategies for mesothelioma diagnosis

Microarrays offer great capacity for parallel measurement of gene expression and can reveal the complexity of the molecular processes that underlie the genesis and progression of cancer. There have been only a handful of studies about microarray profiling of mesothelioma. A recent analysis of 16 tumours and four normal pleural samples using relatively simple microarrays (4K) provided information about gene expression differences between the paired normal and tumour tissues in vivo. A simple technique, based on the expression levels of a few genes, has been reported to accurately discriminate between mesothelioma and lung cancer. The method depends upon quantitative PCR to measure gene expression ratios of paired markers.

Prognostic factors

Prognostication for mesothelioma is dependent on somewhat crude criteria, such as tumour extension and differentiation, rather than on genetic information. Poor prognosis is associated with poor performance status, high white blood cell count, being male, chest wall pain and the sarcomatoid histological subtype.

High throughput methods, such as cDNA and oligonucleotide microarrays, are now being used to systematically investigate the molecular changes that accompany clinical features used to stratify cancers. Therefore, it seems likely that future mesothelioma treatment will be individualised, based on our ability to accurately predict a patient’s response to specific drugs, generate cost savings, but importantly spare individuals who are not likely to respond from the morbidity associated with the therapy.

A small study of outcome prediction in mesothelioma involving 17 cases identified four genes which were suggestive of outcome. Although this work is encouraging, only 29 independent cases were used for validation of the predictive profile by RT-PCR of the candidate genes and there is a need to replicate and expand this analysis.

Management of mesothelioma

In the past few years, there have been several major developments in the management of mesothelioma, especially the development of more effective therapies plus new discoveries that could improvemesothelioma diagnosis and new insights into the pathobiology of the disease. These discoveries are producing new approaches to diagnosis and prognostication, such as DNA microarray patterns to predict outcome, and genetic information to develop novel therapies such as gene therapy.


There are three sorts of surgery involved in mesothelioma management. First, diagnostic surgery might be required to establish a diagnosis—eg, video thoracoscopy and open pleural biopsy and/or mediastinoscopy or laparoscopy. Second, palliative surgery can include partial pleurectomy with pleurodesis, thoracoscopy with pleurodesis, and, rarely, pleuroperitoneal shunting. Third, potentially curative surgery, which has been used in several centres, involves extrapleural pneumonectomy with a goal of removing all gross tumour. Typically, this operation is followed by some form of adjuvant therapy aimed at eliminating residual microscopic disease. Only a few patients are suitable for extrapleural pneumonectomy, and cardiac and general medical status is routinely assessed preoperatively to reduce the risk of major postoperative complications. Operative mortality is now around 6% for this procedure. In some centres this approach has led to median survival times of more than 2 years with good control of local thoracic disease. Selection bias, the difficulty of the surgery, and the perioperative mortality rate combine to make the role of radical surgery controversial. There is general agreement that extrapleural pneumonectomy should only be done in specialised expert centres and that adjunct therapy is necessary after surgery.


No chemotherapy regimen for mesothelioma has proven curative, but several regimes are valuable for palliation. These treatments not only decrease tumour burden but also improve symptoms ofmesothelioma such as pain, breathlessness, and chest wall masses. Until 5 years ago, comprehensive reviews of chemotherapy and mesothelioma were unable to recommend any particular therapy as a standard of care because of the low response rates. Two therapeutic regimens have since been shown to be of value.

Pemetrexed plus cisplatin represents a combination of a multitargeted antifolate and a platinum compound. Pemetrexed is a potent inhibitor of thymidylate synthase, which is required for DNA synthesis. A multicentre phase III study in 448 patients compared this drug combination with cisplatin alone, and showed an improvement in overall survival of nearly 3 months with the combination, with an objective response rate of 41%.

Gemcitabine plus cisplatin has also been shown to offer similar levels of palliation. Gemcitabine is a false nucleotide that inhibits DNA synthesis and, when incorporated into DNA, terminates DNA polymerisation and inhibits repair. Excision repair does not occur because an additional nucleotide is added which masks the fraudulent nucleotide, though gemcitabine also has several other mechanisms of action. Two trials of this combination have yielded objective response rates of 48% and 33% in a total of 74 patients, with symptomatic improvement and quality-of-life benefits.

The knowledge that the platelet-derived growth factor and epidermal growth factor signalling pathways can be involved in mesothelioma has led to trials using imatinib mesilate and gefitinib to block each of these respective pathways, though no convincing evidence of response is evident in the early studies.

Other chemotherapy combinations are being tested, including large multicentre studies. None has yet achieved the response rates described above.

In patients with good performance status, we consider the standard of care for first line palliative chemotherapy to be pemetrexed and cisplatin in the dose and schedule listed in reference 101. Prophylactic dexamethasone plus supplemental folic acid and vitamin B12 are also recommended. There is no second line chemotherapy regimen that could be described as a standard of care at present, but the most commonly used second line drugs are gemcitabine plus cisplatin.


Radiotherapy has been studied in mesothelioma for 31 years and, overall, the results have been largely disappointing; the exception is for local postsurgical radiotherapy, which is effective in the prevention of seeding of tumour cells in the wound site. The biggest limitation to radical radiotherapy is the diffuse nature of the tumour covering most of the pleural surface as well as the interlobular fissures. It is difficult to undertake radical radiotherapy in that situation without causing radiation pneumonitis. Various fractionation methods have been used but the most recent, and probably the most successful, is intensity modulated radiotherapy. This technique has been used for mesothelioma patients after extrapleural pneumonectomy, with radiation fields accurately defined by placement of markers during surgery. The combined use of extrapleural pneumonectomy and intensity modulated radiotherapy has been largely successful at controlling local disease, although these patients then tend to die of metastatic disease. Use of local radiotherapy—eg, radioactive colloids and other forms of brachytherapy—have a compelling logic, given the local nature of disease, but have been largely disappointing.


Mesothelioma seems to be sensitive to destruction by tumour immunotherapies. This finding has been confirmed in animals and has been supported by studies in human beings that show that substantial tumour regression can be induced, though not predictably enough to warrant standard use in clinical practice. For several reasons, mouse models of mesothelioma have been studied extensively from the immunotherapy perspective and a large body of preclinical information is available.

Patients with mesothelioma mount an antimesothelioma immune response, albeit one that is weak and unable to destroy the tumour. The goal of immunotherapy is to boost that weak response and induce tumour regressions. Early studies with crude immunotherapy approaches (eg, BCG) recorded little in the way of responses. Local cytokine therapy with recombinant interleukin 2 or interferon gamma into the pleural cavity induced some responses in patients with early disease but were ineffective in patients with advanced disease. Recombinant interferon alfa given systemically as a single drug has produced partial response rates of around 10-15%, complicated by the usual effects of interferon therapy-lethargy, weight loss, and fevers. We have noted that some patients (around 10%) had striking responses lasting 5-13 years (unpublished). The reason that these patients respond well and the others do not is unclear. In a pilot study of six patients, continuous delivery of the recombinant cytokine granulocyte-macrophage colony stimulating factor molgramostim into mesothelioma tumours induced several part responses despite substantial local catheter-related difficulties.

In view of the apparent sensitivity of mesothelioma to immunotherapy, new immunotherapy approaches will probably continue to be studied in this disease, almost certainly in combination with other therapeutic modalities. It is possible that in the future some form of immunotherapy will enter standard practice for treatment of mesothelioma.

Gene therapy

Gene therapy generally involves administration of engineered viruses into which a gene of interest has been inserted, with the aim of inducing prolonged expression in the tissues of the protein product of the inserted gene. Mesothelioma has been the target of two separate gene therapy approaches. The first involves what is known as suicide gene therapy. This approach involves transfer into tumour cells of a DNA molecule encoding an enzyme (herpes simplex virus thymidine kinase) that by itself has no effect but, after administration of ganciclovir, is capable of converting that drug to a toxic metabolite that destroys the tumour cells and, via a bystander effect, neighbouring tumour cells. Several long-term responses have been achieved; however, this type of therapy is in early stages of development.

The second type of approach involves immunomodulatory gene therapy. The rationale, supported by preclinical data in animal models of mesothelioma, is that local delivery of a vector producing a cytokine within the tumour over prolonged periods would mimic the inflammatory or immune processes that occur in organs undergoing autoimmune destruction. Direct intratumoural injection of a vector comprising a vaccinia virus transduced with an interleukin-2 transgene induced a lymphocytic infiltrate of the tumour; however, tumour regressions were not evident. Gene therapy continues to offer promise as a form of cancer treatment, but has no established place in standard mesothelioma therapy.

Other therapies

Photodynamic therapy involves generation of toxic oxygen radicals when light converts a sensitising drug in the presence of oxygen. These radicals damage the cells and induce cellular necrosis. Intracavitary treatment of mesothelioma with this form of therapy involves administration of sensitising drug followed by intrapleural or intra-abdominal delivery of light to the tumour by dye lasers. A light-scattering medium is used to overcome the problem of uneven distribution. This approach is laborious and time-consuming but is an effective method of cytoreduction. Long term responses have not been noted.

Several clinical trials of other novel agents are underway. The antiangiogenic agents, bevacizumab, thalidomide, and PTK/ZK 787 have been or are being tested. They target, in part, the vascular endothelial growth factor pathway. Because copper is a necessary cofactor for angiogenesis, a study is underway of tetrathiomolybdate, a copper-lowering agent that is effective in preclinical models of tumour growth, after cytoreductive surgery. Other agents that specifically target essential biological pathways of mesothelioma include the histone deacetylase inhibitor superoylanilide and hydroxamic acid. Other agents that will soon be evaluated in phase II clinical trials include a proteosome inhibitor (PS-341), another histone deacetyalase (PXD101), and another antagonist of vascular endothelial growth factor (AZD2171). Finally, the group that originally identified mesothelin has generated monoclonal antibodies labelled with toxins for use in immunotoxin trials, with phase II studies about to commence.

Optimum palliation for mesothelioma

Recurrent pleural effusions are controlled by pleurodesis, usually involving talc instillation, and occasionally surgery. Pain is a major issue for these patients and there are several types of pain inmesothelioma. Invasion of the chest wall can cause localised somatic pain; intercostal nerve invasion or vertebral invasion can cause neuropathic pain; and lung invasion can cause diffuse visceral pain. Also common is dyspnoea due to pleural effusion or more often tumour spread.

There is no upper limit of dose for opioid to control pain in mesothelioma. The dose chosen should give adequate pain relief for the duration of action of the drug (4 h for liquid morphine, 12 h for sustained release morphine) without causing unnecessary side-effects. Somatic pain often responds to an additional nonsteroidal anti-inflammatory drug. Neuropathic pain requires addition of an anticonvulsant such as carbamazepine or sodium valproate. Some patients require procedural pain relief such as intrathecal analgesia or nerve block. Weight loss and anorexia might respond to dietary advice (eg, small frequent meals, visually appealing food, odour-free dining) and to drug therapy (eg, dexamethasone).

The importance of attention to the psychosocial aspects of this disease cannot be overemphasised. Patients often carry a substantial burden of fear and anger with this disease, often complicated by difficulties associated with medicolegal process. Multidisciplinary teams are very effective at addressing the various issues, as are community support networks.

Prevention of mesothelioma in asbestos-exposed populations

Since the epidemic of mesothelioma began, the notion that the disease could be prevented in high-risk individuals has been suggested. In some highly exposed cohorts the lifetime risk can approach 25%. Since these individuals can be identified by their high exposure, much research has been aimed at identifying possible ways of preventing this disease. The main approach has involved dietary methods. In a study aimed at determining whether vitamin A could be protective, daily retinol was associated with a lowering of the rate of mesothelioma in crocidolite-exposed individuals, compared with daily betacarotene, which seemed to have no significant effect.

Although several mesothelioma antigens have been described, the possibility of generating a vaccine that could be given to all high-risk individuals remains remote, probably because of the risk of side-effects-tumour cells are almost exclusively self, and the potential risk of continuing autoimmunity means that studies of antimesothelioma vaccines with mesothelial proteins are unlikely to be undertaken.

Since mesothelioma usually begins in the parietal pleura, it is conceptually possible to undertake prophylactic bilateral pleurectomies in patients at very high risk of mesothelioma, though such an approach has not been published. Future research will need to apply modern microarray, immunological, and proteomics approaches to accurately map the key events in mesothelioma development and progression. Then, both pre-clinical and clinical studies will need to target these processes to find effective therapies or preventions.

Asbestos is predicted to cost the economy of the western world around $US 300 billion in compensation in the coming decades, in addition to health-care costs associated with the disease. This figure is a strong driving force for research aimed at preventing and curing mesothelioma.