A 9-year-old girl was admitted to our hospital with juvenile metachromatic leukodystrophy (arylsulfatase A deficiency). Symptoms of this lysosomal storage disease, such as decreased school performance and compromised motor skills, had started 1 year earlier. She underwent bone marrow transplantation from a matched unrelated donor after receiving conditioning with fludarabine, treosulfan, and thiotepa, together with thymoglobulin, a rabbit antithymocyte globulin. Conditioning was well tolerated, and the posttransplantation period was uneventful except for 1 febrile episode. Graft-vs-host disease (GvHD)4 prophylaxis consisted of 3 methotrexate doses in combination with a starting daily cyclosporin A (CsA) dosage of 3 mg · kg−1 · day−1. Trough CsA concentrations were monitored with the antibody-conjugated magnetic immunoassay (ACMIA) for CsA (RxL Dimension; Siemens). CsA concentrations entered the therapeutic interval after 3 days, and the dosage was adjusted to achieve trough concentrations of 120–150 μg/L (Fig. 1). The patient received CsA for 16 weeks. During this period, the CsA concentration was measured 31 times, with results from 98 μg/L to 219 μg/L (mean, 156 μg/L). Four weeks after transplantation, the patient developed a mild GvHD of the skin, which disappeared immediately after commencement of prednisolone treatment. Because B lymphocytes were absent or low early after transplantation, the patient received immunoglobulins for 5 months (Fig. 1). Endogenous immunoglobulin production started only in the later phase of immune reconstitution. Six weeks after discontinuation of CsA therapy, a concentration of 147 μg/L was still detected in whole blood. Although the patient was not supposed to have received CsA and the parents had denied the administration of CsA or drugs other than those prescribed, CsA concentrations between 116 μg/L and 174 μg/L were obtained over the next 4 months.
QUESTIONS TO CONSIDER
Has the CsA therapy really been discontinued?
Can delayed drug elimination explain the continued increased CsA concentrations?
What tests could be done to identify any potential analytical interference?
The introduction of CsA to transplantation medicine dramatically reduced the incidence of acute rejection and improved long-term graft and patient survival (1). Blood CsA concentrations, however, must be routinely monitored for numerous reasons, including a narrow therapeutic index (nephrotoxicity), highly variable absorption, extensive metabolism via the cytochrome P450 enzyme CYP3A4, metabolic (hepatic) elimination, numerous drug interactions related to both CYP3A4 and the P-glycoprotein cellular efflux pump [encoded by the gene ABCB1, ATP-binding cassette, sub-family B (MDR/TAP), member 1; formerly known as MDR1)], risk of poor compliance because of potential lifetime administration, and difficulties in clinically distinguishing adverse effects from an inadequate pharmacologic effect (underimmunosuppression). Most transplantation centers, including ours, use the whole-blood trough concentration to adjust CsA dosing.
Most laboratories use immunoassays to quantify the CsA concentration in whole blood. Immunoassays are attractive for laboratories because they can be automated, have low startup costs, do not require highly skilled staff, and allow 24-h service (2, 3). Immunoassays, however, have major drawbacks, including varying degrees of metabolite cross-reactivity and interferences that produce falsely higher or lower results than those obtained with chromatographic procedures that resolve metabolites from the parent drug (4, 5). Currently, 1 RIA (DiaSorin) and 6 different nonisotopic immunoassays are available for quantifying CsA concentrations in whole blood: fluorescence polarization immunoassay (TDx/FLx; Abbott), enzyme multiplied immunoassay technique (EMIT V-Twin; Siemens), ACMIA (Dimension RxL, Siemens), cloned enzyme donor immunoassay (Microgenics/Roche Modular), chemiluminescent microparticle immunoassay (ARCHITECT; Abbott Laboratories), and chemiluminescence immunoassay (ADVIA Centaur; Siemens). Except for the ACMIA assay, all of the immunoassays require a manual pretreatment step, which causes higher imprecision.
In our patient, whole-blood CsA concentrations remained in the therapeutic interval, even months after discontinuation of the therapy. The CsA elimination half-life of 19 h can be prolonged in elderly patients and patients with hepatic dysfunction. Because our patient was 9 years old with an unimpaired liver function, we suspected an analytical problem. To test for possible soluble interfering agents, we analyzed the plasma of the same sample. Because CsA accumulates in red blood cells, the CsA concentration should be lower in plasma than in whole blood. In 10 of our bone marrow transplant patients, the mean (SD) percentage of the CsA concentration in plasma was 16% (7%) of that in whole blood, a finding in agreement with published results (6). In this patient's sample, however, the ACMIA CsA signal was 40% higher in plasma than in whole blood, suggesting that an interfering substance present in plasma was responsible for the measurable (false-positive) ACMIA CsA results. Indeed, we detected no CsA when we analyzed the patient's samples with another immunoassay (chemiluminescence immunoassay; Siemens) or by liquid chromatography–tandem mass spectrometry.
Human antianimal antibodies and the lower-affinity heterophile antibodies are the most common cause of test interferences in immunological assays. Pretreatment of the patient's plasma sample with a commercial Heterophilic Blocking Tube (Scantibodies Laboratory), which can remove test interferences produced by these antibodies, eliminated the false-positive plasma CsA signal for the ACMIA assay. In samples from other patients with and without CsA therapy, plasma CsA measurements were not affected by the Heterophilic Blocking Tube treatment. This result demonstrated that interfering antibodies caused the persistent high CsA values in the ACMIA assay. Such interferences may cause both positive and negative deviations from the true concentration, depending on the nature of the antibody and the principle of the assay (7). Possibly, the thymoglobulin given in the conditioning phase induced human antianimal antibodies. Thymoglobulin contains a rabbit antithymocyte IgG, which may have induced antibodies that, despite the species difference, recognized the mouse anti-CsA antibodies used in the ACMIA assay. The protein sequence homology of IgG molecules from different animal species is considerable, and cross-reactivity of antianimal antibodies between species does occur (8). No other therapeutic animal antibodies had been administered to this patient. The ADVIA Centaur chemiluminescence immunoassay assay, which also uses mouse monoclonal anti-CsA antibodies, was not affected by the interference. Even if the same antibody is used, interferences can be platform specific. A case of an antibody assay interference directed only against the antibody–enzyme complex has recently been reported (9).
Such antibodies usually develop within 2 weeks of treatment and may persist up to 30 months thereafter (8). Prevention of this complication includes the humanizing of therapeutic antibodies, the use of antibody fragments only, or PEGylation. In addition, concomitant immunosuppressive therapy by itself diminishes the development of endogenous antibodies. We believe that interfering antibodies started to emerge in our case after reduction of the CsA dose, concomitant increase in B-lymphocyte counts (Fig. 1), and endogenous antibody production. In fact, assay interference was not detectable in a plasma sample collected and frozen before CsA treatment, and the first measurement after the second dose revealed a low CsA concentration in whole blood. These results demonstrate that the interference was not present initially. Because immune reconstitution (B-lymphocyte count) was proceeding as expected and in light of the initial GvHD in a patient with a nonmalignant disease, we did not reduce the CsA dose more rapidly, despite constant high concentrations. The development of GvHD, however, suggests that CsA concentrations might have been overestimated, at least for part of the time during treatment. Thus far, only 1 case of false-positive measurement of CsA with the immunological ACMIA assay has been described in the literature (10). In that case, HPLC detected an interfering peak with a retention time close to that of CsA, but the peak could not be identified. Our case demonstrates the importance of both a careful evaluation and knowledge of the method. An unusual distribution of CsA concentrations in plasma and whole blood that does not fit the pharmacokinetic characteristics or unexpected results in a multiple-sampling strategy can suggest assay interference. If a test result is unexpected and an interference by endogenous antibodies is suspected, several strategies to detect and eventually remove the interference can be used: (a) methods to detect or remove protein interferents (dilution linearity study, sample pretreatment with blocking reagent, protein precipitation, ultrafiltration, chromatographic immunoglobulin purification); (b) analysis with an unaffected method (optimally the reference method); and (c) careful examination of the patient's history (exposure to immunogenic products, history of hyperactive immune system). Our case demonstrates that interference in therapeutic drug monitoring does occur and may be very difficult to detect if the results are within the therapeutic interval and correlate with the administered dosing.
POINTS TO REMEMBER
Continuously high CsA concentrations after discontinuation of the drug can be caused by an interfering antibody, such as a human antianimal antibody induced by therapeutic application of rabbit IgG.
Interfering antibodies can cause false-positive or false-negative results in immunoassays.
To investigate a potential analytical problem, one can do the following: dilution linearity studies, treatment of the sample with a blocking reagent, protein precipitation, ultrafiltration, chromatographic immunoglobulin purification, analysis with an unaffected method (preferably the reference method), careful examination of the patient's history (exposure to immunogenic products, history of hyperactive immune system, and so on). In this case, the plasma CsA concentration was measured to detect analytical interferences.
Communication between the laboratory and the physician is essential to resolve these types of problems.
↵4 Nonstandard abbreviations:
- graft-vs-host disease;
- cyclosporin A;
- antibody-conjugated magnetic immunoassay.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
- Received for publication April 14, 2010.
- Accepted for publication September 7, 2010.
- © 2011 The American Association for Clinical Chemistry