Total and Active Rabbit Anti-Thymocyte globulin (rATG;Thymoglobulin®) Pharmacokinetics in Pediatric Patients Undergoing Unrelated Donor Bone Marrow Transplantation

Abstract

Rabbit anti-thymocyte globulin (rATG; Thymoglobulin®) is currently used to prevent or treat graft-versus-host disease (GvHD) during hematopoietic stem cell transplantation. The dose and schedule of rATG as part of the preparative regimen for unrelated donor (URD) bone marrow transplantation (BMT) have not been optimized in pediatric patients. We conducted a prospective study of 13 pediatric patients with hematologic malignancies undergoing URD BMT at St. Jude Children’s Research Hospital from October 2003–March 2005 to determine the pharmacokinetics and toxicities of active and total rATG. The conditioning regimen comprised total body irradiation, thiotepa, and cyclophosphamide; cyclosporine and methotrexate were administered as GvHD prophylaxis. Patients received a total dose of 10 mg/kg rATG, and serial blood samples were assayed for total rATG by ELISA (Enzyme Linked ImmunoSorbent Assay) and active rATG by FACS (Florescence Activated Cell Sorting). We found that our weight-based dosing regimen for rATG was effective and well tolerated by patients. The half-lives of total and active rATG were comparable to those from previous studies, and despite high doses our patients had low maximum concentrations of active and total rATG. There were no occurrences of Grade 3–4 GvHD even in patients having low peak rATG levels and the overall incidence of Grade II GvHD was only 15%. None of the patients had serious infections following transplantation. These data support the use of a 10 mg/kg dose of rATG in children with hematologic malignancies since it can be administered without increasing the risk of graft rejection, or serious infection in pediatric patients with a low rate of GvHD. These conclusions may not apply to patients with non-malignant disorders.

Introduction

Rabbit anti-thymocyte globulin (rATG), a polyclonal antibody that targets human T lymphocytes, is currently approved for treatment of acute rejection in renal transplant recipients. Because of its activity against T cells, it is frequently used for hematopoietic stem cell transplantation centers to prevent [1–7] or treat [8,9] graft-versus-host disease (GvHD) and to improve engraftment of donor stem cells during hematopoietic transplantation [1–7].

Currently, 2 polyclonal antibodies – Thymoglobulin® (rATG) and Atgam® (equine ATG, eATG) – are commercially available in the United States and differ in their biological activities. Compared with eATG, rATG shows activity at lower doses (15 mg/kg vs 1.5 mg/kg, respectively). rATG also has a considerably longer half-life (5.7 vs 30 days, package insert) and higher specificity for human T lymphocytes than eATG, thereby favoring its use for hematopoietic stem cell transplant. Our institution chose to use rATG because of its longer half-life and activity at lower doses.

There have been few pharmacokinetic studies of rATG in adults [2, 10, 11, 12] and only one study [13] in children. The optimal dose and schedule of rATG as part of the preparative regimen for bone marrow transplantation (BMT) have not yet been defined. High or low doses of rATG can alter the risk of GvHD, graft rejection, or infectious complications. [1, 11, 13]. At this time, low and high doses have not been firmly established.

To evaluate the pharmacokinetics of active and total rATG and toxicities associated with the dosing regimen, we performed a prospective clinical trial in children who underwent an unrelated donor (URD) BMT for hematologic malignancies at St. Jude Children’s Research Hospital from October 2003 to March 2005. We assayed total and active rATG serum concentrations at multiple defined time points and assessed the effectivity and tolerance of the dose and schedule of rATG in this pediatric group.

Materials and Methods

Patients

Thirteen pediatric patients undergoing an unrelated donor BMT for hematologic malignancy were enrolled in a prospective clinical trial at St. Jude Children’s Research Hospital from October 2003 to March 2005. All patients received non-T-cell–depleted bone marrow grafts from unrelated donors. High-resolution HLA typing was performed for the loci A, B, C, DR, and DQ. Donors were considered acceptable if they matched the recipient at the allele level at 7 of 8 or 8 of 8 loci. The trial was approved by the St. Jude Institutional Review Board, and all research participants gave informed consent and assent (as appropriate) for participation.

Conditioning regimen

The conditioning regimen comprised 12 Gy total body irradiation (TBI; fractionated into 8 doses on days –8 to –5), thiotepa (5 mg/kg/dose for 2 doses on day –4), and cyclophosphamide (60 mg/kg/dose on days –3 and –2). GvHD prophylaxis consisted of cyclosporine initiated on day −2 at 2.5 mg/kg/dose intravenously (IV) every 12 h and adjusted to maintain trough concentrations between 170 and 250 ng/mL by the enzyme multiplied immunoassay technique (EMIT), and methotrexate IV (15 mg/m²/dose on day +1; 10 mg/m²/dose on days +3, +6, and +11). rATG was administered intravenously via a central venous access device. All patients received 1 mg/kg rATG on day –4 over 1 h and 3 mg/kg on days –3, –2, and –1 (total dose 10 mg/kg) over 6 h each day. To minimize side effects, all patients received premedications (methylprednisolone, diphenhydramine, and acetaminophen) before each rATG dose and a continuous IV infusion of hydrocortisone (5mg/kg) during rATG infusion.

Collection of information on adverse events

Information on adverse events was collected prospectively and scored according to the Common Toxicity Criteria (CTC) of the National Cancer Institute (NCI) Version 2 [http://ctep.cancer.gov/reporting/ctc_archive.html] and reported at least annually as appropriate to institutional committees and the Food and Drug Administration. Adverse events were assigned by the treating physician or principal investigator. Relationship to treatment and classification of event (expected or unexpected) for Grade III and IV adverse events were detailed. Acute and chronic GvHD were graded according to established criteria. [14] To assure patient safety and data validity and integrity, an independent Data Safety Monitoring Board (DSMB) at St. Jude conducted regular reviews of the study’s clinical and safety-related data.

Measurement of total and active rATG

Blood samples (2–3 mL) were obtained at baseline (before administering the test dose of rATG), before each 3 mg/kg dose, and before reinfusion of stem cells. After the transplant, blood samples were obtained on days +1, +3, +5, +7 and at weeks 1, 2, 4, 8, 16, and 24. Serum was separated within 2 h of blood collection and stored at −70°C (Translational Trials Unit Core Pharmacology Laboratory, St. Jude Children’s Research Hospital) until they were shipped on dry ice to Genzyme (Cambridge, MA) for processing. Total and active plasma rATG were determined by Genzyme by the method of Regan et al. [12] Total plasma rATG was determined by ELISA (limit of quantification 3.9 µg/mL) and active plasma rATG by FACS (limit of quantification 0.2 µg/mL).

Pharmacokinetic methods

We used a 2-compartment pharmacokinetic model with first-order elimination (ADVAN 3 subroutine in NONMEM version V, double precision, level 1.1 using the first-order conditional estimation (FOCE) method with INTERACTION] [15]) to describe total and active rATG. The POSTHOC option was used to estimate individual parameters. Pharmacokinetic parameters estimated by the model were systemic clearance (CL), volume of the central compartment (V), and intercompartmental rate constants (k₁₂, k₂₁). Intersubject variability was modeled with an exponential model.

Results

Of the 13 patients enrolled, 11 were male. The median age at BMT was 10 years (range, 2–16 years). The median weight was 36.7 kg (range, 11.1–86.6 kg). Three patients had chronic myelogenous leukemia (CML; 2 in first chronic phase and 1 in the second chronic phase), 3 had acute lymphoblastic leukemia (ALL; 1 CR1 and 2 CR3), 4 had acute myelogenous leukemia (AML; 3 CR1 and 1 CR2), 2 had juvenile CML (JCML; 1 CR1 and 1 PD), and 1 had myelodysplastic syndrome (MDS; stable disease). (Table 1) Gender, diagnosis and age did not affect the results (data not shown).

Table 1.

Patient Demographics

Patient	Diagnosis	Age (years)	Weight (kg)	Disease Status at Transplant	Days to ANC >500/mm3	Days to Platelets >20× 103/mm3 without Transfusion × 7 days
001	AML	12	40.8	CR2	17	49
002	JCML	2	13.4	PD	22	32
003	AML	16	60.4	CR1	23	48
004	AML	10	38.7	CR1	23	38
005	ALL	3	14.5	CR3	16	28
006	CML	10	36.7	CP1	19	27
007	CML	16	86.6	CP1	19	26
008	MDS	12	40	SD	19	33
009	CML	8	29.3	CP2	16	18
010	JCML	3	14.1	CR1	21	29
011	ALL	10	34.9	CR3	17	25
012	AML	2	11.1	CR1	17	34
013	ALL	14	48.5	CR2	15	25

Note: AML, acute myeloid leukemia; ALL, acute lymphoblastic leukemia; JCML, juvenile chronic myeloid leukemia; CML, chronic myeloid leukemia; MDS, myelodysplastic syndrome; CR, complete remission; CP, chronic phase; PD, progressive disease; SD, stable disease; ANC, absolute neutrophil count.

We observed a biphasic elimination of total and active rATG in our patients (Figure 1). Mean CL for the population for total and active rATG was 198 and 4530 mL/day, respectively. The intersubject variability [coefficient of variation (CV%)] in the CL for total and active rATG were 47% and 61%, respectively. When accounting for the positive relation between CL and weight (kg) (Figure 2), intersubject variability (CV%) significantly decreased (p < 0.0001) to 37% and 50%, respectively. On the basis of post hoc estimates, the median beta half-lives for total and active rATG were 27.3 days (range, 25.7–30.4 days) and 12.5 days (range, 5.8–22.4 days), respectively. In addition, the median actual C_max values were 57.7 (range, 23.7–80.7) and 4.0 (range, 1.6–8.0) µg/mL, respectively.

Figure 1.

Total rATG (top) and active rATG (bottom) versus time. Squares represent data and the solid curve represents the population average model fitted curve. LOQ: limit of quantification (3.9 mg/ml for Total ATG and 0.2 mg/ml for active ATG). Since the assay only indicated that a level was below the LOQ rather than assigning a specific value to it, we randomly assigned it a level below the LOQ to improve the visual interpretation of the results. Time line represents time from first rATG dose. No patient had measurable rATG concentrations prior to the first dose.

Figure 2.

Total rATG clearance (top) and active rATG clearance (bottom) versus weight.

Total rATG was measurable in all patients until week 4 and in 11/13 patients until week 8. Mean total rATG on day 0 was 53 µg/mL (range, 23.7–80.7 µg/mL) and on day 28 was 15.9 µg/mL (range, 5.1–20.6 µg/mL). Active rATG was measurable in all patients through week 2, 10 patients at week 4, and undetectable (<0.2 µg/mL) in 7 patients at week 8. Mean active rATG on day 0 was 4 µg/mL (range, 1.6–8.0) and on day 28 was 1.13 µg/mL (range <0.2 to 4.64). Corresponding median concentrations for total and active rATG showed similar elimination curves (data not shown).

All patients achieved neutrophil engraftment at a median of day +19 (range 15 to 23 days) and platelet engraftment at a median of day +32 (range 15 to 34 days). There were no episodes of grade 3 or 4 GvHD; 2 patients developed grade 2 GvHD. There were no grade 3 or 4 non hematologic serious adverse events, no grade 3 or 4 serious adverse events related to rATG administration and no serious infections. One of the patients developed cGVHD (overall incidence of 7.6%). Patients were tested weekly by PCR (polymerase chain reaction) in peripheral blood for cytomegalovirus (CMV), adenovirus (ADV) and Epstein Barr Virus (EBV). All detectable copy numbers for CMV were indications for treatment with ganciclovir or foscarnet. EBV was treated with rituximab if copy numbers exceeded 2000 copies/ug total cellular DNA. Three patients reactivated CMV in the first 45 days post-transplant. One patient developed a positive PCR for adenovirus and there were 6 episodes of positive PCR for EBV all with copy numbers less than our threshold for treatment.

Discussion

Our weight-based dosing regimen (total dose 10 mg/kg) of rATG was effective and well tolerated by pediatric patients undergoing an unrelated donor BMT for hematologic malignancy. There was no Grade 4 non-hematologic toxicity, the overall incidence of acute GvHD (maximum Grade II) was 15% and there were no occurrences of Grade 3–4 GvHD, despite 11 of 13 patients having peak total rATG levels less than 70 µg/ml on Day 0. Also, there were no serious infections in any patient after transplant. Only 1 of 13 patients relapsed: his day 0 concentration of rATG was less than 70 µg/mL and he had an active rATG of more than 1 µg/mL for 2 weeks (average follow-up 3 years). [2,11]

We observed 2 phases of clearance for both total and active rATG in our patients, which is consistent with the results of Waller et al. [11] and Regan et al. [12] Although the dose and schedule given to our patients were different from those of previously published studies (Waller et al. 6 mg/kg; Guttman, et al 7 mg/kg; Seidel, et al dose range 7.5mg/kg – 40 mg/kg), the half-lives for total and active rATG were similar to those found in previous adult studies [2, 10–12]. Conversely, despite receiving higher doses, the peak total concentrations of rATG for our patients were 67% and 28% lower than in studies by Waller et al. [11] and Guttman et al. [10], respectively. In the only other pediatric study Seidel and colleagues reported concentrations of active rATG approximately 60% higher than our median peak concentration of 4 mcg/ml at a dose 25% lower than the dose used in our study. [13] These differences might be attributed to differences in sampling times between their study and ours. Remberger et al.[2] have suggested that peak concentrations of total rATG less than 70 µg/mL before and 1 week after BMT increases patient’s risk of developing Grade II–IV acute GvHD. In our study, only 2 patients had total rATG concentrations of more than 70 µg/mL at day 0 and neither developed GvHD. On the other hand, only 2 of 11 patients who had day 0 concentrations less than 70 µg/mL developed Grade II GvHD (overall incidence 15%).

Waller et al. [11] have suggested that less than 1 µg/mL of active rATG is subtherapeutic. The 2 patients who developed acute GvHD had less than 1 µg/mL active rATG concentrations by week 2. The concentrations of active rATG were measurable to day 28 in 10 patients who had low incidence of GvHD. This low incidence in our patients might have been because active rATG persisted in the serum; however, inhibition of T-cell stimulation by phytohemagglutinin (PHA) in vitro by rATG (as described by Remberger et al. [5]) was not performed as samples were not assayed in real time and thus t-cell function could not be assessed.

The primary limitation of our study is the low patient numbers yet we were able to describe the pharmacokinetics of rATG in pediatric patients in a prospective study. These data support a hypothesis that a total rATG dose of 10 mg/kg can be safely administered without increasing the risk of GvHD, graft rejection, or serious infection in pediatric patients with heamtologic malignancies. The definition of low or high dose rATG remains to be determined in the BMT population, although an attempt has been made by Remberger and colleagues [16] to delineate differences in outcome in patients who received different doses of rATG. Using retrospective data from 1989–2003 they evaluated BMT outcomes from 4 different doses of rATG (4 mg/kg, 6 mg/kg, 8 mg/kg and 10 mg/kg). They found a higher incidence of infections in the 10mg/kg patients, who received transplants in the 1990’s. Our data did not show an increased risk of infections in our patients, this may be due to differences in infection prophylaxis over the past 18 years. Whether total thymoglobulin or active thymoglobulin concentrations are more predictive of GvHD, engraftment or risk of relapse also remain to be answered.