Oral cyclosporine A in neonatal swines for transplantation studies.

The purpose of this study is to define the optimal dose of oral cyclosporine A (CsA) microemulsion in newborn swine for transplantation studies and to describe its pharmacokinetics and acute renal effects in short-term administration. Thirteen neonatal pigs were randomized into four groups: one control and three groups with CsA administration at 4, 8 and 12 mg/kg/d for 15 days (D). Blood samples were collected on D 0, 2, 4, 9 and 14 to determine the changes of the CsA trough concentrations, the creatinine (Cr) and blood urea nitrogen (BUN) serum concentrations. On D 14, blood samples were collected every hour from 1 h to 10 h after CsA administration to determine the area under the curve (AUC). On D 15, kidneys were removed for histological analysis. We observed a stabilization of CsA trough concentrations from D 4 to D 14. On D 14, in the three treated groups, CsA trough concentrations were 687 ± 7, 1200 ± 77 and 2211 ± 1030 ng/ml, respectively; AUC (0–10 h) were 6721 ± 51 ng·h/ml in group 4 mg/kg/d, 13431 ± 988 ng·h/ml in group 8 mg/kg/d and 28264 ± 9430 ng·h/ml in group 12 mg/kg/d. Cr concentrations were not significantly different among the four groups; but compared to control group, BUN concentrations of the three treated groups increased significantly. CsA was well tolerated; neither acute, severe adverse event nor renal histological abnormality was observed. In conclusion, a 15-d course of oral CsA treatment ranged from 4 to 12 mg/kg/d is safe for newborn pigs, which need much lower CsA dose than adult pigs to reach comparable trough level and AUC. As immunosuppressive therapy in newborn pigs, we recommend a CsA dose of 4 mg/kg/d to achieve a trough blood concentration between 400 and 800 ng/ml.

Keywords: Acute renal effects; cyclosporine A; newborn swine; pharmacokinetics; transplantation

Neonate stage is an immune "window phase" susceptible for induction of transplantation tolerance. In mice, it has been known for decades that the injection of allogeneic spleen cells during neonate period under certain conditions can result in neonatal tolerance and chimerism without any conditioning regimen[1],[2]. Although the relevant immunological mechanisms were not precisely understood, neonatal tolerance could be regarded as the result of immature and plastic immune system[[3]]. However, neonatal tolerance induction protocols, which have been proved to be effective in rodent species, are rarely, if ever, as effective as in large animals and humans, which are born with almost entirely functional T cell immune system[6],[7].

Nevertheless, in many animals and humans, neonatal immune system still demonstrates several immature features: (i) the cellular immune system of neonates is still antigen-inexperienced and consists of high percentage of naïve helper T cells and low percentage of memory T cells and cytotoxic T cells[8],[9]; (ii) ABO-incompatible organ transplantation and B cell tolerance are achievable in human infants[10]; (iii) neonatal innate immune cells including neutrophils, dendritic cells, macrophages and NK cells, are qualitative or quantitative incompetent[[11]]; and (iv) recipients of umbilical cord blood cells experience less graft-versus-host disease than recipients of hematopoietic stem cell transplantation from an adult donor[15]. In such circumstances, exploring of central tolerance induction protocol in neonate large animal might give essential indications to transplantation program in very young human recipient[7].

Hand or face vascularized composite allograft (VCA) is a potential therapy for severe congenital upper extremity or craniofacial defects in newborn infants; however, an effective tolerance induction protocol to composite allografts is mandatory to justify this high-risk intervention in these vulnerable neonatal patients[16],[17]. Swine is an excellent model for transplantation immunology research program because of its similarities to human genetics and body size[18]. In a previous study, we have established a vascularized composite tissue allograft model in non-weaned newborn pigs[19]. Besides, preclinical model of lung allotransplantation conducted in neonatal swines was also reported[20]. However, knowledge of immunosuppressive therapy and relevant side-effects in newborn pigs remains unknown. Cyclosporine A (CsA) is currently the predominant immunosuppressive drug for pediatric transplantation, even in very young infants[21]. However, the use of CsA is mainly limited by its nephrotoxicity, which is characterized by a reversible, dose-dependant reduction of the renal blood flow (RBF) and the glomerular filtration rate (GFR), manifested clinically by an increase in serum creatinine (Cr) and blood urea nitrogen (BUN)[22]. Swine serves as a suitable animal model in pharmacological and toxicological studies of CsA due to its close organ structure[23] and activity of the drug-metabolizing cytochrome P450 3A enzyme system to human analogs[24],[25]. In human and pig neonates (<4 weeks of age), the pharmacokinetic profiles and acute renal effects of CsA still remain unknown.

In this study, our principal objective was to describe the pharmacokinetics and acute renal effects of oral microemulsion CsA in a newborn swine model, in order to define the appropriate dose and monitoring strategy of CsA for immunosuppressive treatment for VCA in newborn piglets. This study may also help other preclinical transplantation studies such as ABO-incompatible heart, lung or heart–lung transplantations in neonatal swines to perform immunosuppressive therapy with CsA.

Thirteen newborn domestic piglets with a patrilineal pedigree of Large White and a matrilineal pedigree of Youli, the same age of five days (D), weights ranged from 1.39 to 4.06 kg, 11 males and 2 females, were housed together with the sow in one cage under the standard procedure for neonatal pigs in the Institute Claude Bourgelat, animal research center of the national veterinary school of Lyon (VetAgro Sup-Campus Vétérinaire de Lyon, Marcy l'Etoile, France). The sow was fed with a standard pig diet. Newborn piglets were breastfed exclusively and unrestrictedly. All animal experimentations were approved by the Ethical Committee of the Veterinary Campus of Lyon (Number – 2012/1257) and conformed to the European Guideline 86/609/EEC of 1986 for the care and use of animals for scientific purposes.

Considering that the difference in body weight might cause a change in drug distribution, piglets were weighed at D 0 and classified into three classes according to their body weights. Then, we randomly allocated piglets according to CsA doses: 0, 4, 8 and 12 mg/kg/d during 15 d, the weight was balanced within groups (Table 1).

Table 1. Distribution of piglets was according to body weights and CsA doses.

3 Body weights on D 0 were balanced among four groups with CsA doses of 0, 4, 8, and 12 mg/kg/d.

Cyclosporine (Neoral® 100 mg/ml, oral solution) was purchased from Novartis Pharmaceuticals S.A.S (Rueil-Malmaison, France). The piglets were weighed every morning at 9:00 AM, and CsA doses were calculated according to their weights. Appropriate doses of CsA oral solution were diluted with 0.5 ml of normal saline solution in a 1 ml syringe and released directly into deep buccal cavity of the piglets to make sure cyclosporine was swallowed down and not spat out. CsA was administrated orally twice daily at 9:30 AM and 5:30 PM. On the last day (D 14), only a single dose of CsA was administered at 9:30 AM.

Anesthesia was induced by 5% isoflurane (Laboratoires Belamont, Paris, France) in O2, and then maintained with 1–2% isoflurane in O2. The blood samples were drawn from the axillary vein of the anterior limbs. Collected samples were not more than 1% of circulating blood volume according to the practice guide to the administration of substances and removal of blood published by Diehl et al.[26]. This study mainly focused on the concentration versus time data of different doses of CsA and on acute renal function side effects. CsA total blood concentrations were measured just before drug administrations in the morning of D 0, 2, 4, 9 and 14, to determine the changes of the trough concentrations during the period of drug exposure. By D 14, to determine the area under the curve (AUC), blood samples were collected every hour from 1 to 10 h after CsA administration. In addition, at random timing, spare samples were also performed from D 7 to D 12 in order to complete the information on pharmacokinetic profiles. CsA whole blood concentrations were measured by Biomnis (Lyon, France) using the antibody-conjugated magnetic immunoassay method on the Dimension® RxL Max® Integrated Chemistry System (Siemens Healthcare Diagnostics S.A.S., Saint Denis, France). The coefficient of variation for this measurement was ± 5%. In parallel, blood samples at D 0, 2, 4, 7, 9, 11 and 14 were used to evaluate renal function based on the serum Cr and BUN concentrations measured with a KONELAB KL20 ISEND automatic analyzer (Thermo Clinical Labsystems, Vantaa, Finland). The GFR evaluation was not explored in this study because of the examination difficulty in neonatal piglets in which long-term anesthesia and surgery should be required.

The impact of CsA doses on the bodyweight increase, CsA trough concentrations, Cr and BUN was tested through an analysis of variance with an interactive model including days as a numeric variable. To assess the proportionality of CsA dose on trough concentrations and AUC of CsA at D14, a Pearson correlation test has been performed. This test was also used to determine the correlation between the increase of body weight and Cr, and the correlation between the increase of BUN and Cr. The significant threshold was set at 5% for all tests. Statistics has been carried out using R statistical program (version 2.15.2, Boston, MA).

CsA was tolerated by the piglets in different groups without severe adverse events. Hirsutism, inflation in testicles and mammary glands were observed in all piglets since the second week of treatment, but no sign of dermatitis was observed. All piglets survived except the smallest piglet (Pig 1C), which presented osteoarthritis of the ankle and very slow body weight growth and was euthanized during experiment on D 9. Bone biopsy has shown severe infection and necrosis in the anterior left ankle joint. Among the other piglets independently of CsA dose, a very significant increase (p < 0.00001) in body weight was observed with growing rates between 108% and 159% from the initial body weight, with a median rate of 137% on D 14. There was no significant difference of body weight increase between the three CsA groups of piglets whatever the dose was (4, 8 and 12 mg/kg/d) and between the three CsA groups together versus the control group.

In our experiment, the CsA blood concentration began to be detected after two days of drug administration. Figure 1 shows the results of CsA trough concentrations on D 2, 4, 9 and 14. We observed a significant higher CsA trough concentrations at D 14 compared to D 2 (p = 0.019) in 7 of 8 cases with an increase ranging from 7% to 106%. In all cases, except piglet 3C, we observed a stabilization of CsA trough concentrations from D 4 onwards. On D 14, trough concentrations were significantly different within the three CsA dose groups (p = 0.04). Moreover, the trough concentrations on D 14 were correlated and proportional (r = 0.75; p = 0.03) to the CsA doses (4, 8 and 12 mg/kg/d, Table 2).

Graph: Figure 1. CsA trough concentrations in three groups of eight piglets on D 2, 4, 9 and 14. A significant higher CsA trough concentrations at D 14 was observed comparing to D 2 in seven of eight cases (p = 0.019). Stabilization of CsA trough levels were observed since D 4 to D 14 (p = 0.18).

Table 2. CsA pharmacokinetics results in neonatal piglets.

4 The trough levels and AUC(0–10 h) on D 14 were correlated and proportional (trough levels: r = 0.75, p = 0.03; AUC(0–10 h): r = 0.86, p = 0.006) with the CsA doses among three groups (data was expressed in mean ± SD).

CsA concentrations were detected from 0 to 10 h on D 14. The Figure 2 shows the results of CsA concentration versus time curves on D 14. AUC(0–10 h) were significantly different within the three CsA doses groups (p = 0.04) and they were also correlated and proportional (r = 0.86; p = 0.006) to the CsA doses (4, 8 and 12 mg/kg/d; Table 2).

Graph: Figure 2. CsA blood concentration versus time curves in three groups of eight piglets monitored for 10 h after 14 d treatment with CsA. The concentration versus time curves (C–T curves) were flat. No classic period containing an obvious, single absorption peak and a subsequent consecutive clearance trough was observed.

The concentration versus time curves (C–T curves) were unusual, flat with no peak or Tmax observed. There was no classic period containing an obvious, single absorption peak and a subsequent consecutive clearance trough in these piglets, instead, multi-waved curves can be observed in most of them.

The serum Cr and BUN concentrations on D 0, 2, 4, 7, 9, 11 and 14 are shown, respectively, in Figures 3 and 4. At D 0, the baselines of serum Cr before drug administration were between 40 and 50 μmol/L. During the experiment, the values of Cr increased significantly from D 0 to D 14 (p = 0.00002) except in Pig 1C. The increase of Cr concentrations was equivalent among these groups (p = 0.15, Figure 3). In pig 1C, the serum Cr was 45 μmol/l on D 0; however, this value decreased to 38 μmol/l on D 4 and to 40 μmol/l on D 9. In all animals, the increases of serum Cr concentrations appeared to be correlated with the body weight increase (r = 0.4 and p = 0.0002).

Graph: Figure 3. Means of the serum creatinine (Cr) concentration in four groups on D 0, 2, 4, 9 and 14. The increasing of Cr level were equivalent among four groups (p = 0.035). However, there was a significant difference of Cr serum concentrations within the four groups (p = 0.035) on D 14.

Graph: Figure 4. Means of the blood urea nitrogen (BUN) concentration in four groups on D 0, 2, 4, 9 and 14. BUN level increasing of three CsA groups was significantly different (p < 0.00001) from the control group with a stable BUN level.

The baselines of BUN in most of the pigs were less than 3.0 mmol/L. In control group, serum BUN concentrations remained stable throughout the experiment. Compared to control group, significant increase of serum BUN concentrations in CsA-treated groups were observed from D 4, and the values continued to increase during the 15-d course of experiment (p < 0.00001). On D 14, the mean values of serum BUN in CsA-treated groups were approximately 3–4 folds as that of control group. Thus, the ratio of BUN/Cr increased during CsA treatment. Among three CsA-treated groups, the increases of serum BUN were not dose-dependent; however, from D 7 to D 14, higher values of BUN were consistently observed in groups 8 and 12 mg/kg/d compared to group 4 mg/kg/d.

In kidney histological analysis (Figure 5), we did not observe any renal histological abnormality such as glomerulosclerosis, arteriolar hyalinosis, interstitial fibrosis or tubular atrophy after 15 d of CsA treatment.

Graph: Figure 5. Kidney histological examination. Kidney examination of all groups (HE staining, × 100) showed no abnormal changes such as glomerulosclerosis, arteriolar hyalinosis, interstitial fibrosis or tubular atrophy.

Our study described the different pharmacokinetic profiles and relevant acute renal effects of oral CsA in newborn pigs. It is interesting to compare the observed results to those obtained from adult or juvenile pigs. It has been described by Frey et al.[27] that in order to obtain the same plasma concentration of CsA measured by high-performance liquid chromatography as in adult human patients, adult pigs require about two times higher IV dose and about four times higher oral dose of CsA than adult human, because of a higher volume of distribution and a lower availability of CsA in pigs than in humans. This conclusion was confirmed by the study of Cibulskyte et al.[22], in which normal juvenile pigs (aged 80 d, weighed about 30 kg) received oral CsA microemulsion (Sandimmun Neoral Oral Solution, Novartis) and for which CsA whole blood concentration was measured by an enzyme-multiplied immunoassay technique assay. In juvenile pigs, five days of CsA treatment at 30 mg/kg/d was necessary in order to reach a trough concentration of 475 ng/ml, a peak concentration of 914 ng/ml and an AUC0–11.5 h level of 6275 ngh/ml. Our results in newborn pigs showed that two days at 4 mg/kg/d were necessary to reach a trough concentration of 480 ng/ml followed by 750 ng/ml on D 4 and 687 ng/ml on D 14 and a AUC0–10 h level of 6721 ng·h/ml. Thus, to reach a comparable CsA trough level and AUC, newborn pig might require at least 7.5-fold less CsA than juvenile pigs.

The pharmacokinetic profiles of CsA in newborn pigs were also different from the results observed in juvenile pigs[22] or in pediatric patients[28] where the distribution and clearance of CsA are well described by a two-compartment model. In neonatal pigs, the C–T curves on D 14 appeared flat and we did not observe any peak. The ranges between the maximum concentrations and trough concentrations were less different. For this reason, the AUC0–10 h was highly correlated with the trough levels, and thus we had difficulties to analyze other pharmacokinetic parameters (CsA absorption time, time to peak and half-life).

These pharmacokinetic profiles should not be caused by CsA accumulation after administration of a high dose during a long period of time because CsA trough concentrations in the three groups were highly relevant and proportional to CsA doses. Moreover, a stabilization of CsA trough concentrations since D 4 was achieved. In addition, at random timing, spare samples were performed from D 7 to D 12 that confirmed this CsA level stabilization.

Probable explanations of these special pharmacokinetic profiles include the following: (a) newborn pigs with breast-feeding may have improved CsA absorption compared to juvenile/adult pigs with formula feeding (CsA mixed with food)[29],[30]; (b) lipid profile of breast-feeding newborn pigs may have changed the distribution of CsA and improved CsA blood concentration[31],[32]; and (c) newborn pigs on unrestricted breast-feeding may have altered absorption and clearance phases of CsA.

Renal impairment is a well-known complication of CsA treatment. In juvenile pigs, acute nephrotoxicity has been reported after five days of CsA treatment at a dose of 30 mg/kg/d, but not at 15 mg/kg/d[22]. However, this conclusion is not convincible because in this study, significant increase of serum Cr was observed in Group 30 mg/kg/d, but values remained within normal range, while decrease of GFR in the same group was revealed as non-significant. Furthermore, no significant correlation between serum Cr, AUC, peak and trough CsA concentration were observed.

In our study, acute renal effect was evaluated only by serum Cr and BUN concentrations. RBF and GFR were not assessed because of the difficulty to perform these examinations and to compare these parameters in piglets with high body weight variability[22]. In all piglets, the values of Cr increased significantly from D 0 to D 14 with no significant difference within the four groups. Considering that the increasing muscle mass of piglets can also contribute to the change of serum Cr, the ratios of Cr/weight were also compared among the different groups and no significant difference was observed. In parallel, during 14 d of experiment, a very significant growth of body weight was observed in all piglets, except Pig 1C, but no significant difference was observed within the four groups. These observations allowed us to conclude that CsA treatment had no impact on body weight growth and serum Cr increase. In newborn pigs, the increase of serum Cr might probably be dominated by muscle growth and did not reflect an acute renal toxicity of CsA. On the other hand, after 2–4 days of CsA administration, BUN increased gradually during time in all CsA-treated piglets, including Pig 1C. The increase of BUN was significant comparing to control group in which the BUN level remained stable around 3.0 mmol/L. No evidence showed that higher dose of CsA would result in higher BUN levels. Similar results have been shown in human patients with no correlation between higher CsA Cmax and AUC and renal side effects[33].

CsA causes more pronounced increase in serum BUN than in the Cr concentration, this is a common finding in clinical and experimental studies[34],[35]. Similar to our study, in a rat model, CsA was administered at 100 mg/kg/d for 10 d[34]; serum BUN significantly increased by four days of administration and continued to rise, whereas serum Cr was not elevated above control level until D 10. These results suggest that cyclosporine may have selectively reduced urea clearance. However, underlying mechanisms of this disproportionate BUN increase are relatively not clear. Laskow et al. explained this phenomenon by decrease in GFR and more significantly, CsA causes increase in urea reabsorption[36]. As analyzed by Laskow et al., the CsA-induced increased urea reabsorption may be due to at least two mechanisms: (1) CsA enhances renal sympathetic activity, which subsequently improves proximal tubular sodium chloride and water reabsorption, and may thereby increases urea reabsorption by enhanced passive diffusion and solvent drag[37],[38] and (2) CsA causes increased renal vascoconstriction[39], mediated by renal sympathetic nervous system. Increased renal vascular resistance may alter the net peritubular factors, for example, peritubular hydrostatic pressure may decrease[40], thus results in an increase in proximal tubular reabsorption of urea. Authors also concluded that the acute nephrotoxicity of CsA is due to a hemodynamic event rather than to a tubular toxicity.

In addition, we did not observe any abnormal change in the renal histological examination after 15 d of CsA treatment, it might be due to the non-significant CsA nephrotoxicity, or was perhaps because of the short-term drug exposure. Similarly, Dean et al.[41], who have treated pigs with high dose of CsA at 100–140 mg/kg/d given orally for two weeks to reach whole blood trough concentrations of CsA at approximately 1500 ng/mL, did not observe any changes in renal histology; the serum Cr concentrations were not adversely affected in spite of the high dosage. Cibulskyte et al. administered juvenile pigs with orally CsA at 30 mg/kg/d for four weeks, and observed that the serum Cr concentrations remained within normal range, and no abnormal changes were found in renal histological examination[22].

In summary, our study showed that CsA caused significantly increase in serum BUN but without any change in serum Cr and renal histological examination. Increase in serum BUN should be due to increased urea reabsorption induced by CsA, and probably, mild, compensable decrease in RBF and GFR, which need to be confirmed in future study. We then conclude that in our study, oral CsA treatment with the doses range from 4 to 12 mg/kg/d for a 15-d course is safe for newborn pigs in spite of very high blood concentrations of CsA were reached.

Another main purpose of this study was to define the appropriate dose and monitoring strategy of CsA for immunosuppressive treatment in newborn pigs, which can serve as an excellent large animal model for neonate and pediatric transplantation. In practical use, CsA is often dosed by target trough levels, correlated to a range of CsA dosage according to weight. However, in previous studies, there was no definitive knowledge on the appropriate trough level, dose and toxicity of CsA for transplantation in newborn pig model. Concerning the immunosuppressive treatment, we only have data in (1) adult or juvenile swine models[22],[42],[43] and (2) in pediatric transplantation patients[[44]]. The oral CsA doses we choose for newborn pigs was in the range of 4–12 mg/kg/d. This choice was based on the CsA doses used in adult pigs and in human pediatric patients. We discovered that newborn pigs required much lower oral dose of CsA achieved comparable trough level or AUC levels than adult pigs. In newborn pig, CsA blood level required by immunosuppressive treatment can be achieved with an oral dose of only 4 mg/kg/d. Although during 14 d of treatment, higher doses of CsA at 8 and 12 mg/kg/d did not result in more serious acute renal effects; considering long-termed side effects on kidney and other organs such as testicle and mammary glands, we recommend a dose of CsA not superior to 4 mg/kg/d for immunosuppressive treatment in newborn piglets. For CsA monitoring strategy in newborn pigs, because of the flat C–T curves, trough levels were highly correlated with AUC. Thus, CsA trough level wonderfully indicates the drug exposure during 24 h.

The authors thank Mr. R. Roume and Mr. R. Lasseur from Institut Claude Bourgelat, VetAgro Sup-Campus Vétérinaire de Lyon for logistical support and animal care; Pr. J-M. Bonnet from Unité de Physiologie, Pharmacodynamie et Thérapeutique, VetAgro Sup-Campus Vétérinaire de Lyon; and Pr. L. Juillard from Néphrologie, Hôpital Edouard Herriot, Hospices Civils de Lyon for their help on the analysis of CsA nephrotoxicity in neonatal pigs; Dr. D. Sachs from Transplantation Biology Research Center, Massachusetts General Hospital, and Harvard Medical School, Boston, for his help on the analysis of CsA nephrotoxicity and immunosuppressive therapy in neonatal pigs.

The authors declare that there are no conflicts of interest. This study was supported by the Fondation Centaure (France) and the National Natural Science Foundation, no. 30901365 (China).