Pharmacokinetics of Dactinomycin in a Pediatric Patient Population: a United Kingdom Children’S Cancer Study Group Study Gareth J

Cancer Therapy: Clinical

Pharmacokinetics of Dactinomycin in a Pediatric Patient Population: a United Kingdom Children’s Cancer Study Group Study Gareth J. Veal,1 Michael Cole,1 Julie Errington,1 Annie Parry,2 Juliet Hale,2 Andrew D.J. Pearson,2 Karen Howe,3 Julia C. Chisholm,3 Carol Beane,4 Bernadette Brennan,4 Fiona Waters,5 Adam Glaser,5 Sue Hemsworth,6 Heather McDowell,6 Yvonne Wright,7 Kathy Pritchard-Jones,7 Ross Pinkerton,7 Gail Jenner,8 James Nicholson,8 Ann M. Elsworth,9 Alan V. Boddy,1 and UKCCSG Pharmacology Working Group

Abstract Purpose: Dactinomycin (actinomycin D) is an antitumor antibiotic used routinely to treat certain pediatric and adult cancers. Despite concerns over the incidence of toxicity, little is known about the pharmacology of dactinomycin. A study was done to investigate dactinomycin pharmacokinetics in children. Experimental Design: Dactinomycin was administered to 31patients by bolus i.v. infusion, at doses of 0.70 to1.50 mg/m2. Plasma concentrations were determined by liquid chromatography- mass spectrometry up to 24 hours after drug administration and National Cancer Institute Com- monToxicity Criteria was assessed. Results: Pharmacokinetic data analysis suggested that a three-compartment model most accurately reflected dactinomycin pharmacokinetics. However, there was insufficient data available to fully characterize this model. A median peak plasma concentration (Cmax)of25.1ng/mL (range, 3.2-99.2 ng/mL) was observed at 15 minutes after administration. The median exposure (AUC0-6), determined in 16 patients with sampling to 6 hours, was 2.67 mg/L.min (range, 1.12-4.90 mg/L.min). After adjusting for body size, AUC0-6 and Cmax were positively related to dose (P =0.03andP = 0.04, respectively). Patients who experienced any level of Common Toxicity Criteria grade had a 1.46-fold higher AUC0-6, 95% confidence interval (1.02-2.09). AUC0-6 was higher in patients <40 kg, possibly indicating a greater toxicity risk. Conclusions: Data presented suggest that dosing of dactinomycin based on surface area is not optimal, either in younger patients in whom the risk of toxicity is greater, or in older patients where doses are capped.

Dactinomycin(actinomycinD;Fig.1)isanantitumor both adults and children. The drug has been a key component antibiotic used routinely to treat certain forms of cancer in in the successful treatment of Wilms tumor patients over the past 30 years, with >80% of patients treated now being cured of the disease (1–3). However, despite its long history of clinical Authors’ Affiliations: 1Northern Institute for Cancer Research; 2School of Clinical use, limited knowledge exists concerning the relationship Medical Sciences (Child Health), University of Newcastle upon Tyne, Newcastle between dactinomycin dosing and pharmacokinetics. uponTyne; 3Department of Haematology/Oncology, Great Ormond Street Hospital, London; 4Department of Paediatric Oncology, Manchester Children’s Hospital, Dactinomycin is most commonly administered to pediatric Manchester; 5Paediatric Oncology Unit, St. James’s Hospital, Leeds; 6Royal patients as a single bolus i.v. injection. The dose is based on a Liverpool Children’s Hospital, Alder Hey, Liverpool; 7Royal Marsden Hospital, child’s calculated body surface area, with a 3- to 9-week interval 8 Surrey; Department of Paediatrics, Addenbrooke’s Hospital, Cambridge; and between courses depending on the tumor type and disease stage. 9UKCCSG, University of Leicester, Leicester, United Kingdom Received 12/13/04; revised 5/16/05; accepted 5/27/05. The dosing regimen for infants is modified, with patients <1 year Grant support: Supported in part by Cancer Research U.K. G.J. Veal was a old at diagnosis, or weighing <10 kg, frequently receiving a recipient of an AACR-GlaxoSmithKline Outstanding Clinical Scholar award. reduced dose based on body weight (4). For example, a standard The costs of publication of this article were defrayed in part by the payment of page dactinomycin dose of 1.5 mg/m2 is commonly adjusted to charges. This article must therefore be hereby marked advertisement in accordance 25 Ag/kg from birth to 6 months of age and to 1.0 mg/m2 from with 18 U.S.C. Section 1734 solely to indicate this fact. Note: This work was presented in part at the 95th Annual AACR meeting, March 27 6 months to 1 year or <10 kg body weight. This is further to 31, Orlando, FL. complicated in some clinical protocols by the fact that once a Requests for reprints: Alan V. Boddy, Northern Institute for Cancer Research, child reaches 10 kg, doses are recalculated as dose/m2 but kept at Paul O’Gorman Building, Medical School, Framlington Place, University of 50% (0.75 mg/m2) until their first birthday. Alternatively, a Newcastle uponTyne, Newcastle uponTyne NE2 4HH, United Kingdom. Phone: A 44-191-246-4412;Fax: 44-191-246-4301;E-mail: [email protected]. dose of 50 g/kg may be prescribed for children aged 1 to 3 2 F 2005 American Association for Cancer Research. years, with the conventional dose of 1.5 mg/m only being doi: 10.1158/1078-0432.CCR-04-2546 used for children >3 years of age (5).

www.aacrjournals.org 5893 Clin Cancer Res 2005;11(16) August 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2005 American Association for Cancer Research. Cancer Therapy: Clinical

Patients and Methods

Patient eligibility and treatment. The study protocol was approved by the U.K. Trent Multicentre Ethics Committee and participating centers obtained local ethical approval; written informed consent was required, either from patients or parents as appropriate, for all patients entered onto the study. Patients 21 years or younger, receiving dactinomycin chemotherapy as part of their standard clinical treatment, were eligible (Table 1). All patients were required to have central venous access, in the form of single or double lumen central venous catheters or Port-a-Caths, in order to participate in this pharmacokinetic study. Dactinomycin (0.70-1.50 mg/m2) was administered by bolus i.v. infusion as part of the standard chemotherapy regimen that each patient was currently receiving. Toxicity was assessed by the National Cancer Institute Common Toxicity Criteria (CTC), version 2.0, Fig. 1. Chemical structure of dactinomycin (MW1,255). following dactinomycin treatment. Blood sampling and analysis. Blood samples for measurement of Dosage regimens for the treatment of patients with Wilms dactinomycin concentration were obtained from a central line prior to tumor in the U.K. have been revised on more than one occasion administration and at 15 and 30 minutes, and 1, 2, 4, 6, and 24 hours due to concerns regarding potential underdosing or a possible post-administration. Actual sampling times were recorded for all patients studied and were as follows (median value and range given): link between toxicity from venoocclusive disease of the liver 15 minutes, 17 (15-25); 30 minutes, 32 (26-50); 60 minutes, 60 and dactinomycin dosing. Similar amendments to dosing were (57-68); 120 minutes, 122 (112-132); 240 minutes, 241 (215-297); required during a Children’s Oncology Group study for the 360 minutes, 360 (330-385); 1,440 minutes, 1,440 (1,295-1,531). All treatment of rhabdomyosarcoma (5). Venoocclusive disease, a eight samples were obtained in a total of seven children, with between severe hepatic toxicity with a reported incidence of 2% to two and seven samples obtained from the remaining patients; most 13.5% across different clinical studies (6–8), is the main frequently with the omission of the 24 hours post-administration drawback to the clinical use of dactinomycin. However, despite sample. The actual number of samples obtained at each of the specified the many changes to the dosing of dactinomycin, optimum time points are given in Table 2. Samples were obtained from 17 patients dosage regimens for the treatment of infants and young with double lumen central venous catheters, 9 patients with single children remain unclear. As dactinomycin is metabolized in lumen lines, and 5 patients with Port-a-Cath access. All lines were the liver, with both the liver and kidneys involved in drug flushed thoroughly following administration of dactinomycin and prior to the withdrawal of samples for pharmacokinetic analysis and separate excretion, development of renal and hepatic function in young drug infusion and sampling sites were used whenever possible. Blood children may be an important factor in influencing its clinical samples (2 mL) were collected in heparinized tubes and centrifuged at pharmacology. This could potentially lead to significant 1,200 g for 10 minutes at 4jC. Plasma was separated and frozen at variations in drug pharmacokinetics between infants and older 20jC prior to analysis using a validated liquid chromatography-mass children. spectrometry assay, with a limit of quantitation of 1.0 ng/mL (13). As the dose intensity of dactinomycin treatment has been clearly defined as a significant risk factor for the development of hepatotoxicity in Wilms tumor, it would seem logical to carry out pharmacokinetic studies in these patients. This would allow Ta b l e 1. Patient characteristics potential correlations to be investigated between the extent of interpatient variation in exposure to dactinomycin and clinical Characteristic No. of patients (%) response and toxicity. However, to date, there is a paucity of Age (y) information regarding the clinical pharmacology of dactino- <511(36) mycin, with no pharmacokinetic studies having been carried 5-9 8 (26) out in pediatric patients, and a very limited number of studies 10-14 6 (19) in adults. Indeed, the only publications in this area are those 15 -20 6 (19) focusing on the analytical methods developed, with data from Sex one or two patients to show a clinical application (9–12). No Male 18 (58) dactinomycin pharmacokinetic study has been published using Female 13(42) these methods, possibly due to concerns over a lack of Diagnosis specificity for the parent drug, or due to the labor-intensive Wilms tumor 9 (29) nature of the methods, making the analysis of large numbers of Ewing sarcoma 7 (23) clinical samples unrealistic. Soft tissue sarcoma 6 (19) We have recently developed a liquid chromatography-mass Rhabdomyosarcoma 5 (16) spectrometry assay which allows the quantification of dactino- Primitive neuroectodermal tumor 4 (13) mycin concentrations in plasma obtained from patients Concomitant chemotherapy receiving dactinomycin in the clinic (13). In the current study, Vincristine 28 (90) we have used this assay to investigate the pharmacokinetics of Ifosfamide 18 (58) dactinomycin in a pediatric patient population, including Cyclophosphamide 3(10) patients receiving treatment for several different types of Doxorubicin 3(10) malignancy.

Clin Cancer Res 2005;11(16) August 15, 2005 5894 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2005 American Association for Cancer Research. Dactinomycin Pharmacokinetics in Children

Ta b l e 2 . Summary of dactinomycin pharmacokinetic data

Sample time (h) No. of patients Dactinomycin plasma concentration (ng/mL) Mean Median SD Minimum Maximum 0.25 30 29.3 25.1 22.6 3.2 99.2 0.5 27 13.8 13.8 8.1 2.4 43.9 1 24 9.0 7.6 4.4 3.4 20.4 2 237.1 6.4 3.4 2.9 16.1 4 17 5.7 5.7 2.4 2.39.9 6 16 4.4 4.6 1.8 1.37.9 24 7 2.4 2.4 0.84 1.4 3.4

NOTE: Number of patients indicates those samples taken at the indicated times and where dactinomycin was detectable in the sample.

Pharmacokinetics. Pharmacokinetic modeling and pharmacokinetic study between January 2002 and October 2003. The study parameter estimation were done using WinNonlin Professional Version population had a median age of 7 years (range 1-20) and 3.1 software (Pharsight Corp., Mountain View, CA). The area under the included 18 male and 13 female patients. Children were treated plasma concentration versus time curve from drug administration to at seven United Kingdom Children’s Cancer Study Group 6 hours (AUC ) was estimated for those patients with valid samples 0-6 (UKCCSG) centers for a range of different malignancies, with obtained at all sampling times up to the 6-hour time point (n = 16) using the logarithmic trapezoidal rule. For those patients with dactinomycin administered according to the clinical protocol satisfactory data to 24 hours (n = 7), the AUCs from 0 to 24 hours on which the patient was being treated. Vincristine was the

(AUC0-24) and from 0 to infinity (AUC0-1) were estimated, again using most frequent concomitant chemotherapy, with 90% of the logarithmic trapezoidal rule. Extrapolation to infinity was done patients receiving these two drugs in combination. Patient using an estimate of the terminal phase half-life obtained from the characteristics including age, sex, tumor type, and concomitant slope of the concentration versus time curve using the 6- and 24-hour chemotherapy are given in Table 1. samples. Clearance was estimated as dose divided by AUC0-1. Pharmacokinetics. A total of 144 plasma concentrations Two- and three-compartment models were fitted to data from the were obtained from 31 patients as shown in Fig. 2. Summary patients with 24-hour samples. During estimation, observations were statistics on plasma concentrations of dactinomycin at each of weighted according to the inverse of the squared expected plasma the sampling times are presented in Table 2. Although the concentrations. The Akaike Information Criterion was used to assess goodness of fit. majority of the early samples were available for analysis, only The development of a population model for dactinomycin was 16 valid samples were obtained at 6 hours, further reduced to 7 undertaken using the first-order conditional estimation method with D/ samples after 24 hours. Failure to obtain full sampling for all q interaction implemented as part of the NONMEM version V level 1.1 patients was due to the outpatient nature of the treatment. software (14). This was carried out using data from all patients studied. Sampling was not limited by considerations of patient safety or The nonlinear nature of the loge plasma concentration versus time curve due to toxicity. The median plasma concentration of dactino- suggested that a two-compartment model or a higher order model mycin at 15 minutes (peak concentration, Cmax) was 25.1 would be appropriate. Parameter estimates for the two-compartment ng/mL (range, 3.2-99.2 ng/mL), with a median concentration model were obtained using ADVAN3 with the TRANS4 reparameteriza- of 2.4 ng/mL (range, 1.4-3.4 ng/mL) at 24 hours. No tion, whereas ADVAN5 was used for the three-compartment model. A relationship was observed between the type of central venous composite error model was used to describe the intrasubject variation. Statistical analysis. The primary pharmacokinetic end points for access used for blood sampling and dactinomycin plasma the study were peak dactinomycin plasma concentration (concentra- pharmacokinetic parameters. tion at 15 minutes) and AUC0-6. Linear regression was used to For those seven patients with samples taken up to the determine the relationship between log10 peak concentration and log10 24-hour time point, estimates of noncompartmental pharma- AUC0-6 and dose of dactinomycin, patient weight and body surface cokinetic parameters were obtained. Following doses of 0.7 to 2 area. Differences in mean log10 peak concentration and log10 AUC0-6 2.0 mg (0.7-1.5 mg/m ) in these patients, half-life ranged between patient groups were assessed using the t test (two groups) or from 14 to 43 hours, clearance values of 48 to 345 mL/min one-way ANOVA (more than two groups). Adjustment for weight and/ (68-203 mL/min/m2) were determined and the volume of or dose was made using ANCOVA. The following variables were distribution (V ) ranged from 59 to 714 L. However, these considered: sex, tumor type, line type, concurrent chemotherapy, CTC z values should be interpreted with caution as plasma grade 1 or higher treatment-related toxicity for infection, fever, aspartate transaminase, alanine transaminase, platelets. Also considered were concentrations of dactinomycin were only measured up to grade 1 or higher, grade 2 or higher, and grade 3 or higher for any of the 24 hours and a considerable degree of data extrapolation was abovementioned toxicities. Statistical analysis was done using MINITAB required to obtain parameter estimates. These data clearly release 13.1. indicate that more prolonged sampling would be needed for definitive determination of pharmacokinetic parameters. Results Two- and three-compartment models were fitted to the data from patients with 24-hour samples. For two of the seven Patient characteristics and treatment. Thirty-one children patients, it was not possible to fit the three-compartment and adolescents receiving dactinomycin were entered onto the model despite using numerous different starting values for the

www.aacrjournals.org 5895 Clin Cancer Res 2005;11(16) August 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2005 American Association for Cancer Research. Cancer Therapy: Clinical

Fig. 2. Dactinomycin plasma concentration-time profiles for all patients. Data are shown for all patients and illustrate the degree of variability in plasma concentration.

parameter estimates. Figure 3 shows the fitted curves for a Toxicity of dactinomycin. Table 3 summarizes the toxicities representative patient. Although the three-compartment model observed during courses of dactinomycin in which patients seems to fit well, the two-compartment model underestimates undertook the pharmacokinetic study. The most common plasma concentrations at around 2 hours and overestimates at adverse event was elevated concentrations of alanine trans- 6 hours. This pattern of under- and overestimation was seen aminase, occurring in 42% of patients, but in no cases was in the remaining patients. The Akaike Information Criterion this classified as CTC grade 3 or 4 toxicity. Grade 3 to 4 for the three-compartment model was lower than that of the thrombocytopenia (platelet count <50,000/AL) was observed in two-compartment model for each of the patients reflecting the 4 of 31 patients (13%) receiving dactinomycin, and grade 3 to 4 improved fit of the three-compartment model. Differences in infection occurred in 3 of 31 patients (10%). Overall, 9 of 31 Akaike Information Criterion ranged from 5.8 to 25.4 with a patients (29%) experienced no adverse events and 13 patients median difference of 18.6. (42%) experienced one or more CTC grade 1 or 2 toxicities; A population pharmacokinetic analysis using data from all grade 3 or 4 toxicities were observed in 9 patients (29%). patients was conducted with NONMEM as described. All data Patients with low platelet count as defined by a grade 1 points were included in the analysis. Although it was or higher CTC tended to have a lower dactinomycin Cmax possible to estimate parameters for a two-compartment model (data not shown), visual examination of individual fitted plasma concentration versus time curves and residuals showed a poor fit to the data particularly at the 24-hour time point (seven patients) which were systematically under- estimated. It was not possible to obtain population parameter estimates for a three-compartment model even after setting population variances to zero for most of the parameters. The median AUC0-6 was 2.67 mg/L.min (range, 1.12-4.90 mg/L.min) in the 16 patients with valid 6-hour samples. Cmax and AUC0-6 were statistically significantly related to patient weight and body surface area (P < 0.01 for each). An increase in body weight of 10 kg resulted in an estimated 1.32-fold decrease in Cmax, 95% confidence interval (1.19, 1.47) and a 1.19-fold decrease in AUC0-6, 95% confidence interval (1.09- 1.31). After adjusting for patient weight (or body surface area), using linear regression, Cmax was found to be positively related to dose (P = 0.03) and after adjusting for body surface area, AUC0-6 was found to be positively related to dose (P = 0.04). Figure 4 shows the relationship between dose of Fig. 3. Fitting of two-compartment (broken line) and three-compartment dactinomycin administered and the peak concentration, (solid line) models for the analysis of dactinomycin data in patient 5. .,actual dactinomycin plasma concentrations determined at specific time points.TheAkaike Cmax (A) and dactinomycin systemic exposure as defined by InformationCriterionwas20.0(two-compartment)and2.1 (three-compartment) AUC0-6 (B). reflecting the improved fit of the three-compartment model.

Clin Cancer Res 2005;11(16) August 15, 2005 5896 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2005 American Association for Cancer Research. Dactinomycin Pharmacokinetics in Children

Fig. 4. Relationship between maximum plasma concentration (A) observed (Cmax; n = 31) or area under the dactinomycin plasma concentration-time curve (B)to6 hours (AUC0-6; n = 16) and the dose of dactinomycin administered. x, values from individual patients.

(P < 0.01, t test), but also tended to be larger in terms of body for weight and dose, Cmax and tumor type were no longer weight. Consequently, after adjusting for weight and dose, Cmax significantly related (P = 0.701). and low platelet count were no longer significantly related (P = 0.123, ANCOVA). Patients with any CTC grade toxicity had a Discussion 1.46-fold higher AUC0-6, 95% confidence interval (1.02-2.09) than those patients with no toxicity, even after adjusting for The current study was driven by clinical concerns regarding patient weight. Figure 5 shows a plot of AUC0-6 versus patient how best to dose dactinomycin to avoid clinical toxicity, weight highlighting those patients with grade 1 or higher particularly relating to the incidence of the life-threatening toxicities. venoocclusive disease. As a starting point to address this issue, Concurrent chemotherapy and line type. Patients who the primary aim of this study was to obtain pharmacokinetic received concurrent ifosfamide tended to have a lower Cmax data concerning the anticancer drug dactinomycin in a pediatric than those patients not receiving ifosfamide (P = 0.025, t test) patient population. These data may be used to design more but also tended to be older children. No other coadministered focused studies to answer specific questions concerning the chemotherapy had a significant effect on dactinomycin clinical importance of dactinomycin exposure and how doses pharmacokinetics. After adjusting for body weight neither could be optimized in particular patient populations. concurrent ifosfamide administration or line type (single Thirty-one children and young adolescents, receiving dacti- lumen, Port-a-Cath or double lumen) were associated with nomycin chemotherapy as part of their standard clinical Cmax (P = 0.267 and P = 0.224, respectively, ANCOVA). treatment, were studied over a period of 22 months in seven Tumor type. There was significant variation in Cmax with U.K. centers. Patients were being treated for a wide range of tumor type (P < 0.01, ANOVA). Body weight was again a malignancies including Wilms tumor, Ewing sarcoma, and soft confounding factor as patients with Ewing sarcoma tended to tissue sarcoma. Samples for pharmacokinetic analysis were be older and consequently larger, whereas those with Wilms obtained at defined time points up to 24 hours following drug tumor tended to be younger and smaller. Again after adjusting administration. The complete data set obtained from all patient

www.aacrjournals.org 5897 Clin Cancer Res 2005;11(16) August 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2005 American Association for Cancer Research. Cancer Therapy: Clinical

values should be interpreted with caution. As plasma concen- Ta b l e 3 . Treatment-related toxicity trations of dactinomycin were only measured up to 24 hours, a substantial degree of data extrapolation was required to obtain Toxicity No. of patients (%) parameter estimates. Further studies incorporating more All CTC Grades All CTC Grades intensive and prolonged sampling times are needed for grades 3/4 grades 3/4 definitive determination of dactinomycin pharmacokinetic parameters. Alanine transaminase 130 42 0 These data indicate that some of the variability in dactino- Platelets 11 4 35 13 mycin clearance results from factors other than body size, as Fever 6 1 19 3 indicated by the observed variation in clearance normalized to Infection 5 316 10 surface area, thus suggesting that surface area–based dosing is Aspartate transaminase 2 0 6 0 not optimal. Both peak drug concentration and drug exposure Hepatic enlargement 1 1 33 (AUC0-6) were inversely related to patient weight. Thus, the Skin 1 0 30 exposure to dactinomycin in larger or older patients is lower Hemoglobin 1 0 30 than in younger patients. Based on the current dosing guide- Eye infection 1 0 30 lines for dactinomycin, the data presented here indicate that the Rash 1 0 30 practice of capping the dose of dactinomycin at 2 mg in larger patients may result in underdosing. NOTE: Based on National Cancer Institute Common Toxicity Criteria (version No clear impact of concomitant chemotherapy on the 2.0). pharmacokinetics of dactinomycin was observed in the current study. Although patients receiving ifosfamide tended to have lower Cmax values, this relationship was influenced by the fact samples was used to investigate the fitting of various that these tended to be the older patients studied. The effect of pharmacokinetic models. The characterization of a meaningful ifosfamide was not statistically significant after adjusting for population pharmacokinetic model was not possible due to the body weight. Vincristine was coadministered in 28 of the 31 small number of patients for whom a full set of pharmacoki- patients studied and it was therefore not possible to discern any netic samples was available; only 7 of the 31 patients having interaction between these two drugs. No relationship was plasma samples available beyond the 6-hour time point. observed between line type and dactinomycin pharmacokinetics However, it was evident that a two-compartment population in this study, despite sampling from single lumen cannulas. model was not appropriate for these data. Individual models However, it may be prudent to further investigate the potential fitted to the patients with complete data suggest that a three- for specimen contamination with infused drug in single lumen compartment model most accurately reflects the pharmacoki- lines prior to the initiation of a larger clinical study. netics of dactinomycin in children. A modification in the Treatment with dactinomycin was associated with limited sampling schedule with the addition of later time points up to adverse effects, with 29% of patients experiencing no observable 72 hours, as well as an earlier time point at 5 minutes after toxicity and 42% reporting CTC grade 1 or 2 toxicities only. The administration, would have been beneficial. However, predict- most common toxicities reported were elevated concentrations ing the optimal sampling times prior to the opening of the of alanine transaminase and thrombocytopenia. No significant study was difficult due to the lack of published data. The correlations were observed between individual toxicities and practical issue of obtaining samples up to 48 or 72 hours, pharmacokinetic parameters after allowing for differences in which would be required to accurately characterize the clearance of dactinomycin, is likely to present problems with access to patients who are not being treated as study center inpatients. This type of problem is a common consideration in the planning of clinical pharmacology studies in pediatric oncology but may be of particular concern with the long half- life of dactinomycin and the patient group being studied. Population pharmacokinetic approaches will therefore play an important role in the planning of future studies. A wide range of dactinomycin plasma concentrations were observed in the 31 patients studied, with a Cmax range of 3.2- 99.2 ng/mL. These concentrations are equivalent to those shown to inhibit RNA synthesis in tumor cell lines (15). Noncompartmental pharmacokinetic analysis was carried out on data from seven patients where a full set of samples were taken at regular time intervals up to 24 hours following dactinomycin administration. Estimated clearance of dactinomycin varied from 68 to 203 mL/min/m2, with the terminal phase half-life (t1/2) ranging from 14 to 43 hours, suggesting extensive extravascular distribution. Although it is encouraging Fig. 5. Relationship between area under the dactinomycin plasma concentration- time curve (AUC0-6) and patient body weight. 4, patients with observable toxicity that these parameters exhibited a similar range of variation (CTC grades 1-4) following dactinomycin treatment; o, patients experiencing no across the seven patients with data to 24 hours, these parameter toxicity.

Clin Cancer Res 2005;11(16) August 15, 2005 5898 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2005 American Association for Cancer Research. Dactinomycin Pharmacokinetics in Children patient weight and dose. However, a relationship was observed infants versus older children in the dactinomycin pharmacoki- between dactinomycin exposure (AUC0-6) and incidence of any netic study presented here was unfortunately not possible as in CTC grade toxicity, in the 16 patients for whom a 6-hour sample the patient population studied only one patient was under was available. Although further data is required to confirm these 2 years of age, with only one additional patient under the age of observations, these preliminary results suggest that the higher 3 years. plasma concentrations of dactinomycin achieved in younger There is a potential benefit in using pharmacologic data to patients may result in greater risk of toxicity. guide dose modifications in the relevant patient population. Although many agents currently used for the treatment of The current study provides preliminary pharmacokinetic data children with cancer have been with us for many years, there for dactinomycin in children. Although the extent of charac- remain considerable gaps in our knowledge relating to the terization of the pharmacokinetics of the drug was restricted by optimal clinical use of these drugs. This is particularly relevant the number of patients for whom samples were obtained over to the dosing of infant patients, where modifications to dosing the defined 24-hour sampling schedule, these data provide the regimens are often based on empirical evidence and may vary first insight into dactinomycin pharmacokinetics in a clinical considerably across protocols for different tumor types. Of study and serve as a guide to conducting future studies in this relevance to the use of dactinomycin in pediatric patients is the area. Extended sampling times will be imperative in order for recent publication of results from the Children’s Oncology pharmacokinetic parameters to be estimated more precisely in Group, concerning the treatment of children with rhabdomyo- future studies. This is likely to present considerable practical sarcoma (5). For patients treated with vincristine, dactinomy- challenges as previously discussed. However, providing that cin, and cyclophosphamide, the risk of chemotherapy-induced these problems can be overcome, studies may be designed to hepatopathy was 4% in children aged 3 years and older, as provide a scientific rationale for the variable dosing of this drug compared with 15% in children under 36 months of age. These in infants, as well as addressing issues such as the impact of results clearly indicate that dose modifications are required for dose capping in older children. The fact that many pediatric younger children receiving this particular drug regimen and tumors now have excellent prognoses should not be a deterrent support previous publications concerning the vulnerability of to further improvements in chemotherapy, including the infants being treated with dactinomycin for the treatment of avoidance of life-threatening toxicities such as venoocclusive Wilms tumor (16, 17). A meaningful comparison of data from disease.

References 1. Green DM. Wilms’ tumour. Eur J Cancer 1997;33: with vincristine and dactinomycin using single dose 12. Saunders GC, Martin JC, Jett JH, Perkins A. Flow 409^18. or divided-dose schedules: a report from the National cytometric competitive binding assay for determina- 2. Green DM, Breslow NE, Beckwith JB, et al. Effect of Wilms’ Tumor Study. J Clin Oncol 1990;8:1525^ 30. tion of actinomycin D concentrations. Cytometry duration of treatment on treatment outcome and cost 7. Davidson A, Pritchard J. Actinomycin D, hepatic 19 9 0;11:311 ^ 3. of treatment for Wilms’ tumor: a report from the Na- toxicity and Wilms’ tumouröa mystery explained? Eur 13. Veal GJ, Errington J, Sludden J, et al. Determination tional Wilms’ Tumor Study Group. J Clin Oncol 1998; JCancer 1998;34:1145^7. of anticancer drug actinomycin D in human plasma by 16:374 4 ^ 51. 8. D’Antiga L, BakerA, PritchardJ, Pryor D, Mieli-Vergani liquid chromatography-mass spectrometry. J Chroma- 3. Mitchell C, Jones PM, KelseyA, et al.The treatment of G.Veno-occlusive disease with multi-organ involve- togr B 2003;795:237^ 43. Wilms’ tumour: results of the United Kingdom Child- ment following actinomycin D. Eur J Cancer 2001;37: 14. Boeckmann A, Beal SL, Sheiner LB.Technical report ren’s Cancer Study Group (UKCCSG) second Wilms’ 1141 ^ 8 . of the Division of Clinical Pharmacology, in NONMEM tumour study. BrJ Cancer 2000;83:602^ 8. 9. Tattersall MHM, Sodergren JE, Sengupta SK, Trites users manualV5; 1997. 4. Tournade MF, Com-Nougue C, de Kraker J, et al. DH, Modest EJ, Frei E. Pharmacokinetics of actinomy- 15. Yung BY-M, Bor AM-S, Chang P-K. Short exposure Optimal duration of preoperative therapy in unilateral cin D in patients with malignant melanoma. Clin Phar- to actinomycin D induces reversible translocation of and nonmetastatic Wilms’ tumor in children older than macol Ther 1975;17:701 ^ 8. protein B23as well as reversible inhibition of cell 6 months: results of the Ninth International Society of 10. Brothman A. Development of an antibody to actino- growth and RNA synthesis in HeLa cells. Cancer Res Pediatric OncologyWilms’ TumorTrial and Study. JClin mycin D and its application for the detection of serum 1990;50:5987^91. Oncol 2001;19:488^500. levels by radioimmunoassay. Cancer Res 1982;42: 16. Jones B, Breslow NE,Takashima J. Toxic deaths in 5. Arndt C, Hawkins D, AndersonJR, Brietfeld P, Womer 118 4 ^ 7. the second National Wilms’ Tumour Study. J Clin R, Meyer W. Age is a risk factor for chemotherapy- 11. Fujiwara K, Saita T, Takenawa N, Matsumoto N, Oncol1984;2:1028^33. induced hepatopathy with vincristine, dactinomycin, Kitagawa T. Enzyme-linked immunosorbent assay 17. Morgan E, Baum E, Breslow N,Takashima J, D’Angio andcyclophosphamide.JClinOncol2004;22:1894^901. for the quantification of actinomycin D using h-D- G. Chemotherapy-related toxicity in infants treated 6. Green DM, Norkool P, Breslow NE, Finklestein JZ, galactosidase as a label. Cancer Res 1988;48: according to the Second NationalWilms’ Tumor Study. D’Angio GJ. Severe hepatic toxicity after treatment 4843^7. J Clin Oncol 1988;6:51 ^ 5.

www.aacrjournals.org 5899 Clin Cancer Res 2005;11(16) August 15, 2005 Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2005 American Association for Cancer Research. Pharmacokinetics of Dactinomycin in a Pediatric Patient Population: a United Kingdom Children's Cancer Study Group Study

Gareth J. Veal, Michael Cole, Julie Errington, et al.

Clin Cancer Res 2005;11:5893-5899.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/11/16/5893

Cited articles This article cites 13 articles, 9 of which you can access for free at: http://clincancerres.aacrjournals.org/content/11/16/5893.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/11/16/5893.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/11/16/5893. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.