We thank Drs Grigoriu and Scherpereel for commenting on our recent publication examining the diagnostic value of soluble mesothelin in malignant pleural mesothelioma.1 We agree with their comments that the numbers in our study may not have been sufficient to address the question of the prognostic value of this marker. While we have found that mesothelin levels reflect tumour burden and would therefore be expected to have prognostic value, the patients did not receive a standardised treatment regime. At our centre, patients are offered a range of surgery, chemotherapy, radiotherapy, novel immunotherapies or best supportive care treatment options. The numbers of patients in each category therefore further reduces the power of the survival analysis. However, given that patients with sarcomatoid mesothelioma have low mesothelin levels and a poor prognosis, one would not a priori anticipate a close correlation unless patients are stratified according to histology—that is, elevated mesothelin levels could indicate greater tumour bulk (worse prognosis) or greater epithelial differentiation (better prognosis). We are currently evaluating the prognostic value on patients enrolled in a standardised treatment regime. In both studies pleural effusion levels of mesothelin were not related to survival.1 2 While serum mesothelin levels may indeed have prognostic value with Grigoriu and colleagues showing in their analysis of 76 patients that high serum mesothelin levels (>3.5 nM) had prognostic significance, it is unclear at this stage whether an individual’s mesothelin level will have a strong clinically relevant predictive value that adds to that of the currently used prognostic indicators.3

As emphasised by Lee4 in his editorial published with our paper and by Grigoriu and Scherpereel, we support the need for an international multicentre investigation into the value of soluble mesothelin in the management of patients with mesothelioma.