Intrapleural administration of tumour necrosis factor-alpha (TNF alpha) inpatients with mesothelioma: cytokine patterns and acute-phase protein response

Background Tumour necrosis factor-alpha (TNF alpha) has been found to be ve ry effective in the isolated limb perfusion setting for advanced extremity tumours. In a phase I study of intrapleural administration of TNF alpha 5 p atients were followed for inflammatory response patterns. Patients and methods Malignant mesothelioma patients were treated with repe ated intrapleural administration of 0.1-0.2 mg recombinant TNF alpha. Sampl es of serum and pleural fluid were taken at different time-points before an d after TNF alpha-administration. Levels of TNF alpha, interleukin-6 (IL-6) , interleukin-8 (IL-8), C-reactive protein (CRP) and secretory phospholipas e A(2) (sPLA(2)) were measured using enzyme-linked immunosorbent assays (EL ISAs). Alpha 1-acid glycoprotein (alpha 1-AG) was measured by nephelometry. Results In pleural fluid TNF alpha and IL-8 reached peak levels, up to 50-7 00 ng mL(-1) and 6-60 ng mL(-1), respectively, 24 h after administration of TNF alpha. IL-6 (peak levels up to 250 ng mL(-1)) and sPLA(2) peaked after 48 h. A slower and less dramatic pattern was observed for the levels of CR P and alpha 1-AG. In serum no detectable levels of TNF alpha and no IL-8 we re observed, whereas serum levels of IL-6, sPLA(2) and CRP showed a clear i ncrease after intrapleural administration of TNF alpha. Cytokines and acute -phase proteins showed the same pattern during subsequent cycles even up to 12 cycles. Tumour regression was not observed. Conclusions In the setting of a phase I study of repetitive intrapleural ad ministration of TNF alpha in mesothelioma patients, we studied the characte ristics of the inflammatory response. Intrapleural administration was follo wed by a clear inflammatory response locoregionally. In spite of TNF alpha peak levels as high as 700 ng mL(-1) systemic levels were never detectable. The secondary cytokine response led to very high intrapleural IL-6 and IL- 8 levels. Systemically IL-8 levels were never detectable whereas high IL-6 levels were induced systemically initially, with a decreased response to ea ch intrapleural TNF alpha administration over time. The acute-phase respons e in contrast remained remarkably constant throughout the course of repeate d intrapleural administrations of TNF alpha. Intrapleural administration of TNF alpha is well tolerated but associated with inconsistent and rather mo derate impact on production of pleural fluid. This can be achieved by other simpler and cheaper treatment, thus we see no justification for further st udies.