Correspondence
www.thelancet.com Vol 366 July 23, 2005
all new cases of lung cancer across the globe, this would result in a sample size of more than 2500 patients in a week, which is a large enough sample to detect a 5% survival difference and to answer a key treatment question almost instantly. Many new agents for lung cancer are now starting to emerge, but unless we adopt innovative new ways of working to assess each one properly, we may be no closer to completing the jigsaw in another 25 years than we are now. I declare that I have no conflict of interest.
Richard Stephens
[email protected] MRC Clinical Trials Unit, 222 Euston Road, London NW1 2DA, UK 1 2
3
4
5
The Lancet. A tax to prevent the epidemic of lung cancer. Lancet 2005; 365: 1663. Breathnach OS, Freidlin B, Conley B, et al. Twenty-two years of phase III trials for patients with advanced non-small cell lung cancer: sobering results. J Clin Oncol 2001; 19: 1734–42. Brundage MD, MacKillop WJ. Locally advanced non-small cell lung cancer: do we know the questions? A survey of randomized trials from 1966-1993. J Clin Epidem 1996; 49: 183–92. Non-small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials. BMJ 1995; 311: 899–909. Stephens R. The need for a world strategy for clinical trials. Lung Cancer 2003; 1 (suppl 3): S96.
Aseptic peritonitis due to peptidoglycan Leo Martis and colleagues (Feb 12, p 588)1 document that the outbreak of aseptic peritonitis seen in peritoneal dialysis patients given icodextrin between November, 2001, and July, 2002, was due to a contamination of some icodextrin dialysate batches by peptidoglycans released from Alicyclobacillus acidocaldarius during the manufacturing process. We would like to make several points. First, Martis and colleagues mention several times that peptidoglycan contaminated “pharmacopoeia standard dialysis solutions”. This statement is incorrect: the contamination occurred in icodextrin dialysates only, and not in
standard glucose-based solutions. This point is of importance because icodextrin, unlike glucose-based dialysates, is derived from the hydrolysis of corn starch during a manufacturing process that can lead to contamination, as documented here. Second, very little new clinical information is provided by the review of 186 cases of aseptic peritonitis. We would be interested to know the proportion of patients to whom unnecessary antibiotics had been given or in whom the peritoneal catheter had to be removed while clinicians thought they were facing relapsing infectious peritonitis. In addition, we and others2,3 have seen a relapse of aseptic peritonitis in some patients reexposed to non-contaminated icodextrin (ie, peptidoglycan concentration below the detection limit of 7·4 g/L), a finding also mentioned by Martis and colleagues. This suggests an individual susceptibility to peptidoglycan, as also substantiated by the great variability Martis and colleagues saw in the interleukin 6 response of donor peripheral blood mononuclear cells exposed to peptidoglycan. Alternatively, this could indicate an immunisation against peptidoglycan and a subsequent relapse of aseptic peritonitis in the presence of very low concentrations of peptidoglycan in the icodextrin dialysates. Third, Martis and colleagues found a striking increase in interleukin 6 concentrations in the peritoneal effluent of a patient with aseptic peritonitis and also in peritoneal fluid of rats given intraperitoneal infusions of icodextrin with various concentrations of peptidoglycan. Peritoneal biopsies of these animals would have been of interest because biopsies from human beings with aseptic peritonitis have shown various degrees of peritoneal inflammation with mesothelial desquamation, fibrin clusters on the peritoneal surface, lymphocytic infiltration of venular walls, and the presence of mast cells in the peritoneal tissue.4
Rights were not granted to include this image in electronic media. Please refer to the printed journal.
Science Photo Library
know relatively little.2 One of the reasons for this may be the lack of a global strategy. Trials are often designed to address local issues, rather than seeing the bigger picture, and although all research is valuable, this lack of strategy has left us with numerous pieces of the jigsaw that do not fit together, or do not allow the combination of trials into meta-analyses.3 This lack of strategy can be illustrated by the fact that 73 trials of chemotherapy in advanced non-smallcell lung cancer (NSCLC) identified in a literature search used 16 different drugs in 57 different drug combinations (not counting different drug doses or methods of administration). In the future, at the very least, we should be designing trials with consistent control groups, which link or build on previous work. In addition we need to run much larger trials. Since the addition of cisplatin-based chemotherapy to supportive care improves 1-year survival by about 10% in patients with NSCLC,4 it is unlikely that tweaking chemotherapy regimens will result in more than another 5% improvement. To reliably detect such a difference requires a trial with at least 1000 patients per group, but the largest trial of chemotherapy in advanced NSCLC involved less than 400 patients per group. As a result we still do not know which drugs to use, in which combinations, at what doses, or how to schedule their administration. What other strategies could we employ? Sequential two-group trials are a very ineffective way of progressing our knowledge about treatment, whereas multigroup trials can address several questions at the same time and can be very efficient in terms of time and recruitment of patients, since the same patients are used as controls for all the experimental groups. Another proposal is the short-duration global trial.5 Such a trial would aim to accrue as many patients as possible in as short a time as possible to address a simple but important question. For instance, if it were possible to randomise 10% of
289
Correspondence
AP
Rights were not granted to include this image in electronic media. Please refer to the printed journal.
Altogether, the points we raise do not detract from the major clinical benefits seen in patients on peritoneal dialysis with the advent of icodextrin.5 We simply regret that, despite their impressive clinical database and laboratory investigations, Martis and colleagues do not provide more details about the nature of the peritoneal inflammatory response induced by peptidoglycan and its potential influence on the long-term preservation of the peritoneal membrane. Finally, we completely agree that European and US pharmacopoeia standards for parenteral products should be reassessed to avoid further similar events. EG and OD have received lecturer fees from Baxter Healthcare, Belgium.
*Eric Goffin, Michel Tintillier, Olivier Devuyst goffi
[email protected] Department of Nephrology, Université Catholique de Louvain, Cliniques Universitaires St Luc, Avenue Hippocrate 10, 1200 Brussels, Belgium (EG, OD); and Department of Nephrology, Cliniques Ste Elisabeth, Namur, Belgium (MT) 1
2
3
4
5
Martis L, Patel M, Giertych J, et al. Aseptic peritonitis due to peptidoglycan contamination of pharmacopoeia standard dialysis solution. Lancet 2005; 365: 588–94. Williams PF, Foggensteiner L. Sterile/allergic peritonitis with icodextrin in CAPD patients. Perit Dial Int 2002; 22: 89–90. Tintillier M, Pochet JM, Christophe JL, et al. Transient sterile chemical peritonitis with icodextrin: clinical presentation, prevalence, and literature review. Perit Dial Int 2002; 22: 534–37. Goffin E, Cosyns JP, Pirson F, Devuyst O. Icodextrin-associated peritonitis: what conclusions thus far? Nephrol Dial Transplant 2003; 18: 2482–85. Davies SJ, Woodrow G, Donovan K, et al. Icodextrin improves the fluid status of peritoneal dialysis patients: results of a double-blind randomized controlled trial. J Am Soc Nephrol 2003; 14: 2338–44.
Authors’ reply Peritoneal dialysis solutions, whether they are icodextrin-based or glucosebased, must meet pharmacopoeial standards for sterility and pyrogenicity. It is true that excess cases of aseptic peritonitis were seen with icodextrin-based solution and not with standard glucose-based dialysates; our remark regarding “pharmacopoeia standard dialysis solutions” 290
refers to the sterility and pyrogenicity standards that are relevant to all peritoneal dialysis solutions. As Eric Goffin and colleagues correctly point out, our report does not contain new clinical information, such as the proportion of patients to whom unnecessary antibiotics had been given or in whom the peritoneal catheter had to be removed. Since the cases were spontaneously reported by the dialysis centres, we did not have accurate and complete information on the use of antibiotics and catheter removal for them to be included in the report. The major objective of our investigation was to determine the root cause of aseptic peritonitis and implement corrective actions to minimise future occurrences. The pharmacovigilance data gathered after the implementation of corrective actions provide the evidence to support our hypothesis that peptidoglycan contamination of icodextrin was the cause of excess cases of aseptic peritonitis, and that the steps taken to minimise this contaminant are effective in reducing the incidence of aseptic peritonitis. Some of these data are reported in our original article. These data are periodically reported to the health authorities of European countries and provided to the public on request. We agree with Goffin and colleagues that additional details on the nature of peritoneal inflammatory and immunogenic responses induced by peptidoglycan and the potential influence on the long-term health of the peritoneal membrane are lacking. LM is employed by Baxter Healthcare (Deerfield, IL, USA), which manufactures and distributes peritoneal dialysis solution. CV and WFO declare that they have no conflict of interest.
*Leo Martis, Christian Verger, William F Owen Jr
[email protected] Baxter Healthcare, 1620 Waukegan Road, McGaw Park, IL 60085, USA (LM); Hospital de Pontoise, Service de Nephrologie, Pontoise, France (CV); and University of Tennessee Health Sciences Center, Memphis, TN, USA (WFO)
Emergency contraception Paul Bissell and Claire Anderson (May 14, p 1668)1 make the case that in concentrating on links between easier access to emergency contraception and contraceptive behaviours or sexually transmitted infections (STIs), researchers may have overlooked the need to address whether the current configuration of emergency contraception services are appropriate for all women. They proceed to make the important case for improved access for younger women and those in the lower socioeconomic groups. However, they stop short of situating the discussion in the context of the factors that have challenged the process of reaching the indicated target groups, especially in sub-Saharan Africa. Promoting access to emergency contraception continues to be dogged by the need to assuage sceptics about the perceived effect on condom use and hence vulnerability to STIs.2 These concerns, however unfounded, are likely to gain greater currency as research extends into finding ways to reach the young and poor. This is because these groups also constitute those at greatest risk of acquiring STIs, especially in sub-Saharan Africa. For many women in this part of the world, sexual encounters that result in unintended pregnancies occur in the context of relationships that put them at high risk of acquiring STIs, including HIV.3,4 In both research and practice, the two issues would therefore be difficult to decouple. I propose that the two issues should be taken in tandem, but not, as most researchers have sought, to disprove the latter and use that to make the case for more liberalised access to emergency contraception. The goal of research aimed at increasing access to emergency contraception should be to come out with designs that address both issues simultaneously. Promoting emerwww.thelancet.com Vol 366 July 23, 2005
