Monday August 24 2009
A magnified asbestos fibre in a human lung
Scientists may have developed “a more sensitive test for the asbestos-related cancer mesothelioma”, according to a story in BBC News. This devastating cancer is usually diagnosed by looking for cancerous cells in the fluid surrounding the lungs (cytology), but this method is not a very sensitive test and does not distinguish well between mesothelioma and other cancers. See BBC News' article on the new test for mesothelioma.
In a well-conducted diagnostic study, researchers in Oxford assessed the accuracy of combining cytology with their new test, which measures the amount of the protein mesothelin in fluid around the lungs. They demonstrated that this is a valuable addition to the usual cytology tests given to people with possible mesothelioma.
Importantly, the researchers suggest that this test should be used alongside cytology, although it also performed well alone. This potential new method has the benefit of being performed on the same fluid samples currently used in diagnosis by cytology, so it could easily be added to current testing programmes.
Where did the story come from?
This research into pleural effusion and mesothelioma was conducted by Dr Helen Davies and colleagues from the Oxford Centre for Respiratory Medicine and other institutions in the UK. The research was funded by several organisations including the British Lung Foundation, the Department of Health Clinical Lecturer Award and the Medical Research Council. It was published in the American Journal of Respiratory and critical Care Medicine, a peer-reviewed medical journal.
What kind of scientific study was this?
This was a diagnostic study looking at the accuracy of a new form of test to detect the cancer mesothelioma by examining the presence of proteins in lung fluid. This method was compared against the gold standard test for the cancer, which relies on cytology of the fluid (looking for cancerous cells). Mesothelioma is a form of cancer related to asbestos exposure.
People with mesothelioma often have pleural effusion (excess fluid in their lungs), which can affect breathing. However, not everyone with effusion will have mesothelioma, and some will have other forms of cancer or benign pleuritis (non-cancerous lung disease). Cytology of pleural effusions is good at identifying cancers, but it is not very specific to mesothelioma.
Mesothelin is a protein produced by cancerous mesothelioma cells and released into the fluid around the lungs. Measuring the level of mesothelin in the blood is already used to monitor, and sometimes diagnose, this cancer, but several other studies have suggested that measuring mesothelin directly from the pleural fluid may be a better method. The researchers set out to describe the accuracy and use of such a test.
In total, the researchers used 429 samples of pleural fluid collected from 209 patients at the Oxford Pleural Unit. In their first experiment they collected pleural fluid samples from 167 patients with symptoms of pleural effusion who were being investigated for possible malignancy. Analysis was possible in 166 of these samples.
Based on cytology, tissue samples or clinical diagnosis samples were classified as malignant or benign. The type of cancer was also identified. The researchers compared the levels of mesothelin in the pleural fluid of people with mesothelioma, people with metastatic cancers in the lung and those with benign disease. They used statistical methods common in diagnostic testing to determine how accurate a test this was and the optimal concentration for detecting mesothelioma.
They also compared the value of the mesothelin test with that of using pleural fluid cytology alone. In other experiments, they assessed the effect of pleurodesis (sticking the pleural membranes together to prevent further liquid building up) on levels of mesothelin in lung fluid. They also measured levels of mesothelin over time in 33 patients with malignancy, seven of whom had mesothelioma, to see how they changed. Some patients were inoculated with bacteria in their lungs to assess the effect of infection on mesothelin levels.
What were the results of the study?
The study has many results and a selection is presented here. Combined diagnostic methods showed 24 of the 166 patients had mesothelioma, 67 had a non-mesothelioma cancer and 75 had benign pleural effusion. The researchers found that the mesothelin levels were 6.6 times higher in patients with mesothelioma compared to those with metastatic cancers, and 10.9 times higher than those with benign disease. Only two people with benign disease had elevated mesothelin.
The researchers calculated various measures of effectiveness for the mesothelin lung fluid test as a method of distinguishing mesothelioma from all other causes of pleural effusion. These measures were:
- Sensitivity (proportion of people with mesothelioma who test positive) of 71%.
- Specificity (proportion of people without mesothelioma who test negative) of 90%.
- A positive predictive value (probability that someone with a positive test result truly has mesothelioma) of 53%.
- A negative predictive value (probability that someone with a negative test result is truly free of mesothelioma) of 95%.
Of the 13 false positives, 12 were other lung cancers (adenocarcinoma).
Compared with pleural fluid cytology on its own, using mesothelin levels was a better test with greater sensitivity (71% compared with 35%). When the tests were used together, mesothelin levels improved the diagnosis of mesothelioma by cytology. A concentration greater than 20nM correctly identified the eight mesothelioma cases out of the 47 that cytology identified as malignant.
In the 105 patients with negative cytology results, mesothelin levels increased confidence that underlying mesothelioma was correctly excluded: negative cytology and negative mesothelin had a specificity of 97% and a negative predictive value of 94%.
Pleurodesis led to a statistically significant reduction in the levels of mesothelin in pleural fluid at 24 and 48 hours, but had no effect on the levels of mesothelin in the blood. The presence of bacteria had no notable effect on mesothelin levels.
What interpretations did the researchers draw from these results?
The researchers conclude that detection of mesothelin levels in pleural fluid contributes valuable information to the use of pleural fluid cytology alone, especially when the cytology results are inconclusive or suspicious. They say that, as many patients with mesothelioma fall into the ‘suspicious’ category following standard diagnostic tests, adding another assessment of mesothelin to the examination process may benefit up to 3,000 patients in western Europe every year. As current investigations usually involve sampling of the pleural fluid, mesothelin analysis can easily be included within existing programmes.
What does the NHS Knowledge Service make of this study?
This well-conducted study demonstrates that diagnostic accuracy for mesothelioma is improved by combining a test assessing the levels of mesothelin in the pleural fluid with diagnosis through cytology (looking for cancerous cells).
This study was conducted in people with a high background risk of having mesothelioma. In fact, the participants’ chance of having mesothelioma before entering the study was 24/167 or 14%. This means that the findings need to be confirmed in other populations, preferably people at different levels of risks of the disease, to ensure that the predictive values obtained in this study can be more widely replicated.
It is important to consider what resources are required when discussing the introduction of any new diagnostic test. With mesothelin detection, pleural fluid is already being collected for analysis, so this would not involve any further sampling or invasive procedures for patients. However, consideration must be given to the laboratory and reporting facilities required.
Overall, the test improves the accuracy of methods for diagnosing mesothelioma, a cancer which, unfortunately, offers a very poor rate of long-term survival. This improvement is greatest when the method is combined with cytology, offering a high specificity that means that a positive result effectively confirms the diagnosis.
The researchers, importantly, are therefore advocating that this test be used in addition to the usual explorations, and not as a replacement for them, and they say that it will be particularly useful in those patients who have ‘suspicious’ pleural effusions diagnosed by cytology.