For personal use only. Not to be reproduced without permission of The Lancet. 2014 THE LANCET • Vol 356 • December 9, 2000 CORRESPONDENCE 1 Bolli GB, Owens DR. Insulin glargine. Lancet 2000; 356: 443–44. 2 Kurzthals P, Schäffer L, Sorensen A, et al. Correlations of receptor binding and metabolic and mitogenic potencies of insulin analogs designed for clinical use. Diabetes 2000; 49: 999–1005. Don’t blaze the trailblazer TRAIL too early Sir—Robert Nitsch and colleagues (Sept 2, p 827) 1 describe that the death receptor ligands, rcombinant human tumour necrosis factor  (rhTNF-) and FLAG-tagged recombinant human tumour-necrosis-factor-related apoptosis-causing ligand (rhTRAIL) induce apoptosis after incubation of brain-tissue slices with these compounds. TRAIL, in contrast to TNF-, was previously judged safe for intravenous use in the clinic. rhTRAIL has apoptosis-inducing capacity in tumour cells in vitro in various cancer cell lines; and in vivo in athymic nude mice. 2,3 Repeated rhTRAIL administration to non- human primates (cynomolgus monkeys) seemed safe and non- immunogenic, 3 and chimpanzees showed no overt toxic effects after 5 mg/kg TRAIL. 4 A histidine-tagged rhTRAIL led to apoptosis of cultured human hepatocytes and has caused concern about the clinical applicability of rhTRAIL. 5 This finding and that of Nitsch and colleagues prompted us to reappraise these results. The cultured hepatocytes were obtained from livers not used for transplantation. Cultured hepatocytes lack the natural microenvironment, which can lead to increased mRNA expression of receptors, for example, as was seen for the TRAIL receptor DR4. 5 This mechanism might have facilitated apoptosis by TRAIL in the cultured hepatocytes. The environment was better in the brain slices experiments. TRAIL concentrations used in the hepatocyte and brain slice experiments were, however, relatively high (>200 ng/mL), since exposure of many human tumour cell lines to 10–100 ng TRAIL/mL resulted in 50–100% apoptosis. In the animal experiments daily intravenous TRAIL doses up to 10 mg/kg for 7 days resulted in serum concentrations of more than 200 ng/mL, whereas lower doses could slow growth and, in some cases, induce regression of tumour-cell xenografts. 2,3 All the in-vitro toxic effects experiments were done with modified soluble rhTRAIL, despite the native soluble rhTRAIL molecule being non- toxic in cynomolgus monkeys and chimpanzees. 3,4 We observed, with immunohistochemistry in paraffin- sections, that the TRAIL receptor DR4 was clearly and comparably expressed in livers of chimpanzees and humans. Nitsch and colleagues also used 200 ng/mL rhTNF-, which induced apoptosis in brain slices similar to the TRAIL effect. In our hospital, after hyperthermic isolated limb perfusions with rhTNF- (3–4 mg) and melphalan in 45 patients, the mean maximum TNF- plasma con- centration due to systemic leakage was 76·7 ng/mL (range 1·39–574·0). However, despite plasma concen- trations of more than 200 ng/mL in five of 45 patients, no signs of neurotoxic effects were seen; brain toxicity is not, therefore, a limiting factor. TRAIL might still exert toxic effects to normal tissues, especially when used in doses higher than those required for antitumour activity. However, oncologists accept absence of tissue selectivity for widely applied anticancer drugs and are used to working with agents with a limited therapeutic window. The potentially valuable native anticancer-molecule TRAIL, therefore still deserves serious consideration for use in the clinic. *E G E de Vries, W T A van der Graaf, F J Heijenbrok, H J Hoekstra, S de Jong Departments of *Medical and Surgical Oncology, University Hospital Groningen, PO Box 30.001, 9700 RB Groningen, Netherlands (e-mail: e.g.e.de.vries@int.azg.nl) 1 Nitsch R, Bechman I, Diesz RA, et al. Human brain-cell death induced by tumour- necrosis-factor-related apoptosis-inducing ligand (TRAIL). Lancet 2000; 356: 827–29. 2 Walczak H, Miller RE, Ariail K, et al. Tumoricidal activity of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat Med 1999; 5: 157–63. 3 Ashkenazi A, Pai RC, Fong S, et al. Safety and antitumor activity of recombinant soluble Apo2 ligand. J Clin Invest 1999; 104: 155–62. 4 Kelley SK, Harris LA, DeForge LE, et al. The pharmacokinetics of Apo2Ligand/TRAIL(Apo2L/TRAIL) following a single iv bolus dose to chimpanzees. Proc Am Assoc Cancer Res 2000; 41: 5167. 5 Jo M, Kim TH, Seol DW, et al. Apoptosis induced in normal human hepatocytes by tumor necrosis factor-related apoptosis- inducing ligand. Nat Med 2000; 6: 564–67. Visualisation of cell death Sir—Johan De Sutter and colleagues (Oct 21, p 1439) 1 comment on our finding of increased Technetium-99- labelled annexin V ( 99m Tc-AnxV) uptake in at-risk patients with acute myocardial infarction. 2 99m Tc-AnxV binds to phosphatidylserine exposed by cells with an activated cell-death programme and activated platelets. We concluded that this uptake in the infarct area reflects programmed cell death, possibly inhibited by cell-death-blocking drugs. De Sutter and colleagues suggest that increased uptake of 99m Tc-AnxV reflects binding to activated platelets trapped in the microcirculation of the infarct zone and not to apoptotic cardiomyocytes. Substantial evidence from animal studies supports the view that the signal partly reflects binding of 99m Tc-AnxV to cardiomyocytes having an activated cell- death programme. 3 We saw no increased uptake of 99m Tc-anxV in a large thrombus in an infarct-related artery, which embolised to the microcirculation, which is not expected if increased radiolabel uptake is a consequence of binding to activated platelets. De Sutter and colleagues further argue that their results in four dogs strengthen their argument against the presence of large numbers of apoptotic cardiomyocytes in patients with acute myocardial infarction. They did imaging, however, 3 weeks after myocardial infarction and the finding is consistent with the time frame of the apoptotic process; 4,5 after 3 weeks, cardiomyocytes triggered to activate the cell-death programme during ischaemia or reperfusion have done their job and are probably removed from the tissue by infiltrated macrophages and are no longer detected. This situation must not be confused with the acute phase of myocardial infarction, which is the subject of our studies. The latter situation is well within the time frame of appearance of apoptotic cardiomyocytes exposing phosphatidylserine on the outer leaflet of the cell membrane, as seen in animal models. 3 *Chris Reutelingsperger, Leo Hofstra Departments of Biochemistry and Cardiology, Cardiovascular Research Institute Maastricht, University Maastricht, 6200 MD Maastricht, Netherlands (e-mail: c.reutelingsperger@bioch.unamaas.nl) 1 De Sutter J, Lahorte C, Taeymans Y, Dierckx R, Slegers G. Cell death in myocardial infarction. Lancet 2000; 356: 1439–40. 2 Hofstra L, Liem IH, Dumont E, et al. Visualisation of cell death in vivo in patients with acute myocardial infarction. Lancet 2000; 356: 209–12. 3 Dumont E, Hofstra L, Van Heerde WL, et al. Cardiomyocyte death induced by myocardial ischemia and reperfusion: detection with human recombinant Annexin-V in a mouse model. Circulation 2000; 102: 1564–68. 4 Green DR. Apoptotic pathways: the roads to ruin. Cell 1998; 94: 695–98. 5 Goldstein JC, Waterhouse NJ, Juin P, Evan GI, Green DR. The coordinate release of cytochrome c during apoptosis is rapid, complete and kinetically invariant. Nat Cell Biol 2000; 2: 156–62.