Interferon‐gamma administration to patients after major surgery influences cellular immunity without pro‐inflammatory response

Severe tissue injury is associated with suppression of cellular immunity, as reflected by decreased human leucocyte antigenDR (HLA-DR) expression on monocytes [1,2]. In trauma patients, reduced HLA-DR expression correlates with complications and mortality [3]. Decreased HLA-DR expression on monocytes may be related to a shift of the Th1/ Th2 balance towards a Th2 type immune response, which occurs in humans after endotoxin challenge and trauma [4,5]. Administration of the Th1 cytokine interferonγ (IFNγ ) restores expression of HLA-DR on monocytes in septic or trauma patients [6,7]. It is unclear whether this effect of IFNγ is necessarily favourable in situations of severe physical stress. In contrast, systemic derangements as seen during sepsis may result from increased cytokine release, among which is IFNγ [8–10]. Previously we have studied the acute effects of recombinant human (rh-)IFNγ administration on inflammatory response mediators in healthy subjects [11]; rhIFNγ exerted stimulatory effects on neutrophilic granulocytes and on acute-phase protein and chemokine levels, but lacked effects on circulating cytokines. These effects, under noninflammatory conditions, may not be representative of the effects that could be induced under inflammatory conditions, where this cytokine might work synergistically with other inflammatory mediators. To study the short-term effects of IFNγ on host inflammatory response parameters during inflammatory conditions, we selected 13 patients scheduled for pylorus-preserving pancreaticoduodenectomy, aimed at curation of a malignant tumour. At the second postoperative day we randomized them into rhIFNγ (Immukine, 100 μ g/ m 2 subcutaneously) vs. control (saline), to evaluate host inflammatory parameters and cellular immunity, just before and at regular intervals during the first 24 h after injection of rhIFNγ or saline. Seven patients received rhIFNγ and six other patients received saline. Of all measured parameters, preoperative levels and levels on the second postoperative day, just prior to rhIFNγ administration, did not differ between the groups. Cytokines: rhIFNγ did not affect plasma TNFα , IL-6 and IL-10 levels as compared to the controls (Fig. 1). Chemokines: rhIFNγ did not affect plasma IL-8 levels, but significantly increased IFNγ -inducible protein 10 (IP-10) levels as compared to the controls (Fig. 1). Acute-phase proteins: rhIFNγ did not affect plasma CRP and sPLA2 levels as compared to the controls (not shown). Granulocyte and monocyte activation: rhIFNγ did not affect plasma elastase levels, but induced an increase in plasma lactoferrin and neopterin levels as compared to the controls (Fig. 2). The present data show that rhIFNγ has no major influences on the host inflammatory response, whereas cellular immunity remains sensitive to rhIFNγ . The increased IP-10 levels may stimulate lymphocyte traffic. The administered dose of rhIFNγ was indeed relevant and biologically active because it increased monocyte HLA-DR expression [12] and induced plasma levels of IFNγ in range with those reported in septic-shock patients [10]. Previously, rhIFNγ administration to severely injured or septic patients was shown not to result in a definitive improvement of clinically relevant end points [7,13,14]. Nevertheless, there is reason to believe that high-risk patients might benefit from adjuvant IFNγ therapy [15]. Our data show that the stimulatory effects of IFNγ on the cellular immune system of postoperative patients can be achieved in the absence of severe pro-inflammatory effects. Department of Endocrinology and Metabolism (J. de Metz, H. P. Sauerwein), Department of Clinical Immunology and Rheumatology (I. J. M. ten Berge), and Renal Transplant Unit (I. J. M. ten Berge), Division of Internal Medicine; Division of Surgery (D. J. Gouma), Clinical Immunology Laboratory (T. A. Out, R. M. R. Reijneke and Department of Pathophysiology of Plasma Proteins of Sanquin and the Landsteiner Laboratory (C. E. Hack), Academic Medical Center of the University of Amsterdam, Amsterdam; Department of Endocrinology, Leiden University Medical Center, Leiden (J. A. Romijn), the Netherlands.