Patient management by cyclosporine C2 monitoring: not enough science yet to justify the practice.

Abstract

The authors wish to express their concern regarding the recently published recommendations for Neoral C2 monitoring (supplement to Transplantation, May 15, 2002), which have been made in the absence of any supportive evidence from controlled clinical trials (1–3). The editorial accompanying the three articles concluded that “there is clear evidence for the benefits of C2 monitoring compared to C0 monitoring,” citing observed associations between C2 cyclosporine levels and short- to medium-term transplant outcomes in cohort studies. However, although such studies are hypothesis-generating, they do not prove causality. Controlled clinical trials comparing C2 versus C0 monitoring are ultimately required to determine whether these statistical associations lead to improved patient outcomes. To the authors’ knowledge, there have been no such published trials of C2 versus C0 cyclosporine monitoring in renal transplant recipients. The International Neoral Renal Transplantation Study Group recently reported the results of a multicenter, randomized, controlled trial comparing area-under-the-curve (AUC) cyclosporine monitoring with trough monitoring in 204 renal transplant patients over a follow-up period of 3 months (4). This study failed to show any significant clinical benefit of AUC measurement with respect to rejection rate (30.3% vs. 32.6%, respectively), rejection severity, graft survival (92.7% vs. 97.7%), serum creatinine (168±86 vs. 167±85 μM), adverse events (34% vs. 21%), or histologically documented cyclosporine nephrotoxicity (9.1% vs. 5.3%). The study was powered (80% probability) to detect a 36% reduction in rejection rate (from 33% to 21%) and a 23-μM difference in plasma creatinine levels between the two groups. Thus, if any clinical benefit of AUC monitoring existed, it was likely to have been relatively small. The possible, but as yet unproven, benefit of C2 monitoring on hard clinical endpoints must be weighed against its practical disadvantages. C2 monitoring requires blood samples to be taken within a more critical time period than does trough monitoring. In large centers during multicenter drug trials, this appears to be feasible, but it may be more difficult under less optimal circumstances. Furthermore, a significant number of patients take cyclosporine in association with drugs known to affect its pharmacokinetic or even pharmacodynamic properties, such as diltiazem, ketoconazole, sirolimus, or everolimus. It is difficult to know whether the described statistical relationships between C2 levels and transplant outcomes will hold up in these circumstances, or whether other time points would be preferable for monitoring. C2 monitoring may prove to be a superior means of monitoring cyclosporine. The statistical associations presented in the May 15, 2002 supplement and elsewhere are promising and give indirect support for this interesting hypothesis. The authors strongly feel that cyclosporine C2 monitoring warrants further testing in prospective randomized trials before enthusiastic authors and specially convened groups of eminent clinicians recommend its widespread adoption. After 20 years of using cyclosporine, the optimal means of therapeutic drug monitoring remains to be determined. It would be unfortunate if the resolution of this important clinical issue was further delayed by the unquestioning adoption of the latest novel monitoring technique without accumulating the data needed to demonstrate a clinical advantage. Scott B. Campbell David W. Johnson