Asparaginase pharmacokinetics and implications of therapeutic drug monitoring

Asparaginase is widely used in chemotherapeutic regimens for the treatment of acute lymphoblastic leukemia (ALL) and has led to a substantial improvement in cure rates, especially in children. Optimal therapeutic effects depend on a complete and sustained depletion of serum asparagine. However, pronounced interpatient variability, differences in pharmacokinetic properties between asparaginases and the formation of asparaginase antibodies make it difficult to predict the degree of asparagine depletion that will result from a given dose of asparaginase. The pharmacological principles underlying asparaginase therapy in the treatment of ALL are summarized in this article. A better understanding of the many factors that influence asparaginase activity and subsequent asparagine depletion may allow physicians to tailor treatment to the individual, maximizing therapeutic effect and minimizing treatment-related toxicity. Therapeutic drug monitoring provides a means of assessing a patient's current depletion status and can be used to better evaluate the potential benefit of treatment adjustments.


Introduction
Over the past 50 years, asparaginase (ASP) has become a key component of treatment protocols for acute lymphoblastic leukemia (ALL). Lymphoblastic leukemic cells lack or have low levels of the enzyme asparagine synthetase; therefore, they are unable to produce asparagine on their own and heavily depend on extracellular sources [1]. At suffi cient activity levels, ASP depletes serum asparagine and eventually leads to leukemic cell death [1]. With the addition of ASP to existing chemotherapeutic regimens, long-term survival rates in children with ALL have improved; however, signifi cant challenges still exist in optimizing treatment for many patients [2,3].
While asparagine depletion is considered the pharmacological goal of ASP therapy, there is no universally accepted dose or treatment schedule for all patients. Asparagine levels vary greatly between individuals, with several factors infl uencing the relationship between ASP dose and serum asparagine concentrations. Optimal treatment necessitates practitioners to be knowledgeable of the basic pharmacological principles underlying ASP therapy and utilize this knowledge to eff ectively guide patient care. Th e goal of this review is to give a practical overview of the pharmacology underlying ASP therapy in the treatment of patients with ALL. assay used [4 -6]. However, the critical level of asparagine depletion in serum required for in vivo leukemic cell death is unknown.
Results from several studies highlight the importance of asparagine depletion in both pediatric and adult patients with ALL [7 -9]. Jarrar et al . showed an inverse relationship between day 14 asparagine concentrations and remission rate on day 35 in 214 children with fi rst relapse of ALL [7]. Patients with serum asparagine levels Ͻ 1 μ M were more likely to achieve second remission compared with patients with higher asparagine levels.
Currently available ASP formulations exhibit preferential selectivity for asparagine, but also hydrolyze glutamine to a lesser extent ( Figure 1) [10]. Given that glutamine can be used as an amino-group donor in the synthesis of asparagine [11], it has been suggested that depletion of both asparagine and glutamine are necessary to achieve optimal outcomes [12,13]. Recent experimental evidence in acute myeloid leukemia cell lines suggests that the therapeutic eff ect of ASP may largely be due to its eff ects on intracellular glutamine and inhibition of downstream mammalian target of rapamycin complex 1 (mTORC1) signaling pathways [14]. It has been reported that Ͼ 90% deamination of glutamine levels is needed to achieve optimal asparagine depletion in vivo [13]. Current formulations of ASP diff er in their glutamine pharmacokinetics. While both Erwinia chrysanthemi -and Escherichia coli -derived ASP formulations show similar binding affi nities (K m ) for glutamine, the maximal conversion rate at saturation (K cat ) is greater with ASP E. chrysanthemi (Table I) [10, 15 -17]. For all formulations, however, selectivity for glutamine is markedly weaker than is found with asparagine. Furthermore, glutamine levels in blood are much higher than asparagine levels, and a relatively greater ASP activity level is needed to suffi ciently reduce levels of both amino acids [12].

The use of asparaginase in acute lymphoblastic leukemia
Currently, three diff erent types of ASP are available for use in the treatment of ALL. Native E. coli ASP and pegylated (PEG)-ASP are derived from the bacteria E coli . ASP E. chrysanthemi is derived from E. chrysanthemi. It has a distinct immunogenic profi le, making ASP E. chrysanthemi an appropriate treatment option for patients who experience hypersensitivity to E. coli -derived formulations [18,19]. Native E. coli ASP is no longer available in the United States [20], and is being replaced by PEG-ASP and ASP E. chrysanthemi in new protocols.
Activity levels of ASP inversely correlate with serum asparagine concentrations, and are commonly used as a proxy measure to estimate asparagine depletion [4,9,11,21 -23]. Early experiments in non-human primates indicate that asparagine depletion in the serum and central nervous system consistently occurs at ASP activity Ն 0.1 IU/mL [21]. Th is 0.1 IU/mL target has subsequently received support from a number of human trials [4,24,25], and is generally accepted as the activity level necessary to achieve therapeutic depletion of asparagine [2].
Several studies show an association between ASP activity and positive outcomes in patients with ALL [26 -29]. A study conducted in adults by the Cancer and Leukemia Group B compared outcomes between patients treated with PEG-ASP with ASP activity Ͼ 0.03 IU/mL and patients with activity Ͻ 0.03 IU/mL [26]. Overall, the 63 patients with ASP activity Ͼ 0.03 IU/mL showed greater median survival compared with the 22 patients with reduced activity, 31 vs. 13 months, respectively ( p ϭ 0.001). A prolonged course of high-dose intensity, likely resulting in prolonged asparagine depletion, has also been shown to improve outcomes in children with ALL [27 -29].

Relationship between dose, asparaginase activity and depletion of asparagine
A number of factors infl uence ASP activity and asparagine concentrations following a given ASP dose. Th e formulation of ASP, degree of interpatient variability, formation of ASP antibodies, concomitant medications and even the method of administration can have an important impact on ASP activity dynamics and patient outcomes.

Asparaginase formulations
All three ASP formulations show equivalent leukemic cell kill in vitro [23]. However, the pharmacokinetic properties of each ASP diff er greatly (Table II) [30,31]. PEG-ASP shows the longest half-life of the three formulations, which has been estimated at 5.7 days following intramuscular (IM) administration. ASP E. chrysanthemi shows the shortest half-life at approximately 15.6 h [31]. Th ese diff erences carry practical implications for constructing optimal dose schedules, as formulations with longer half-lives are cleared at a slower  Figure 1. Mechanism of action of asparaginase [1]. Adapted with permission from Muller and Boos, 1998 [1].
rate, and therefore provide relatively longer exposure to the enzyme and subsequent duration of asparagine depletion. For this reason, diff erent ASP preparations are not readily interchangeable. Identifying the appropriate dose schedule to achieve therapeutic levels of ASP activity for the diff erent ASP formulations has been the focus of numerous studies (Table III) [4,5,8,9,22,32 -36]. Identical dose schedules of E. coli-derived formulations and ASP E. chrysanthemi can result in signifi cantly diff erent ASP activity levels and may lead to a worse outcome, thus giving the misleading perception of reduced effi cacy [37,38]. Duval et al. compared outcomes in 700 children treated on identical dose schedules of ASP E. chrysanthemi or native E. coli ASP (10 000 IU/m 2 administered twice weekly) [38]. ASP E. chrysanthemi was associated with signifi cantly inferior 6-year event-free survival (EFS) and inferior overall survival. A study by Boos et al . compared trough asparagine and ASP activity levels in 49 children given the same 10 000 IU/ m 2 dose of three ASP preparations (ASP medac ® , Crasnitin ® and ASP E. chrysanthemi ) [4]. Th e investigators found lower mean ASP activity for patients receiving ASP E. chrysanthemi compared with those treated with the same dose of native E. coli ASP (ASP medac or Crasnitin) [4]. Correspondingly, the percent of subjects achieving complete asparagine depletion (defi ned as Յ 0.1 μ M) was also lower in patients receiving ASP E. chrysanthemi [4]. Additional evidence suggests  Table III. Selected pharmacokinetic studies [4,5,8,9,22,32 -36]. largest group of patients ( n ϭ 416) were administered 1000 IU/m 2 PEG-ASP IV [42]. While median ASP activity for these patients was well above the 0.1 IU/mL criterion, the researchers observed pronounced interpatient variability, with ASP activity levels between 0 IU/mL and 3.3 IU/mL within 7 days of dose administration. Overall, the researchers found 30% of ASP activity samples Ͻ 0.1 IU/mL. A second study focused on ASP activity in patients switched to 20 000 IU/m 2 ASP E. chrysanthemi IV administered on a Monday/ Wednesday/Friday schedule [5]. Th e researchers reported substantial interpatient variability in ASP activity at days 2 and 3 post-treatment. Trough ASP activity ranged between Ͻ 0.02 IU/mL and Ͼ 0.5 IU/mL in the 21 patients for which data were available. Recently reported results of 25 000 IU/ m 2 ASP E. chrysanthemi administered IV in patients with hypersensitivity to E. coli -derived ASP highlight pronounced variability in ASP activity between patients [43]. In this study, the coeffi cient of variation of ASP activity measured 48 and 72 h post-dose ranged from 72% to 109%.

Asparaginase antibody formation
ASP is a non-human protein, and has the potential to elicit an immune response in patients. Th e formation of ASP antibodies can substantially reduce ASP activity and negatively impact outcomes [18,44]. Th e presence of ASP antibodies is often associated with symptoms of clinical hypersensitivity [18,45]. However, a number of patients develop antibodies without any outward signs of allergy, referred to as silent inactivation or subclinical hypersensitivity [5,30]. Subclinical hypersensitivity was found to occur in 29% of patients with ALL during front-line treatment with native E. coli ASP in one large monitoring study [18]. Th e risk of subclinical hypersensitivity at the time of relapse is even higher; therefore, it should be considered in designing treatment regimens for relapsed disease [46,47]. Th is condition presents a serious challenge to achieving ASP depletion in many patients, with a number of reports recommending the regular monitoring of ASP activity as a means of identifying patients with subclinical hypersensitivity [5,42,48]. Diff erent ASP formulations each have a distinct immunogenic profi le and are associated with a diff erent risk of antibody formation. Th e incidence of antibodies in patients treated with native E. coli ASP has been estimated as high as 60% during the entire course of therapy [18]. Antibody formation or hypersensitivity with either ASP E. chrysanthemi or PEG-ASP is less prevalent, with studies reporting antibodies in 2 -18% of patients treated with PEG-ASP [22,49] and 8 -33% with ASP E. chrysanthemi [50,51]. PEG-ASP is derived from the same bacterial strain as native E. coli ASP, and patients often exhibit signifi cant cross-resistance between these formulations [18,47,52,53]. In a recent study, researchers evaluated the cross-reactivity of samples drawn from 16 patients who developed native E. coli ASP antibodies during reinduction [53]. None of these samples showed any reactivity with ASP E. chrysanthemi ; however, 63% of samples (10 of 16) showed a signifi cant reaction to PEG-ASP.
A number of strategies have been suggested to reduce the risk of antibody formation. Th ese strategies include: (1) pretreatment with glucocorticoids; (2) treating patients on that ASP E. chrysanthemi may be associated with favorable results if administered at a higher dose and greater frequency in specifi c treatment paradigms [33,34]. In the Associazione Italiana Ematologia Oncologia Pediatrica (AIEOP) ALL 95 study protocol, standard-risk (SR) patients received reduced-intensity treatment compared with patients in the intermediate-risk (IR) group. Both risk groups were randomized to receive either standard treatment or extended intensifi cation with ASP E. chrysanthemi (25 000 IU/m 2 given IM weekly for 20 weeks) during maintenance in SR and during reinduction and maintenance in IR patients. Patients treated with prolonged high-dose ASP therapy showed signifi cantly improved outcomes compared with those treated according to the standard ASP regimen in the SR group (10-year disease-free survival of 78.7% and 87.5%, respectively) but not in the IR group [28,39]. Taken together, these results suggest that earlier reports of suboptimal outcomes with ASP E. chrysanthemi are most probably the result of inadequate dosing and the subsequent lower levels of ASP activity in those patients.
Due to its shorter half-life, ASP E. chrysanthemi must be administered at higher doses and/or more frequently than PEG-ASP or native E. coli ASP in order to maintain consistent asparagine depletion. Albertsen et al. found that when ASP E. chrysanthemi was administered at 30 000 IU/ m 2 intravenously (IV) or IM twice weekly, target ASP activity levels of Ն 0.1 IU/mL were achieved in approximately two-thirds of samples [36]. Results of the Children ' s Oncology Group (COG) ALL07P2 trial showed that 25 000 IU/m 2 ASP E. chrysanthemi administered IM three times a week achieved ASP activity Ͼ 0.1 IU/mL in 92.7% of evaluable patients at 48 h and 88.4% of evaluable patients at 72 h postdose [33]. Based on these results, the substitution dose of ASP E. chrysanthemi recommended by the Food and Drug Administration (FDA) in patients with hypersensitivity to E. coli -derived ASP is 25 000 IU/m 2 administered IM for each scheduled dose of native E. coli ASP three times a week (Monday/Wednesday/Friday) for six doses for each planned dose of PEG-ASP [33,40]. Th e current AIEOP -Berlin, Frankfurt, M ü nster (BFM) ALL 2009 trial includes a randomized intensifi cation study of extended-use PEG-ASP. A dose of 2500 IU/m 2 has been carefully chosen to adequately replace the dose intensity provided by previously used native E. coli products [41]. Furthermore, the strict implementation of therapeutic drug monitoring (TDM) to identify patients with suboptimal ASP activity has been incorporated into this study since 2009. Patients with no observable ASP activity or overt signs of clinical hypersensitivity are switched to ASP E. chrysanthemi . In the AIEOP -BFM ALL 2009 protocol, the substitution dose for a single administration of PEG-ASP is seven doses of 20 000 IU/m 2 ASP E. chrysanthemi administered IV every other day.

Interpatient variability
A number of studies report pronounced variability in ASP activity measurements between patients on identical treatment schedules [5,42,43]. One study found widespread interpatient variability in the ASP activity of 763 patients administered three diff erent ASP preparations. Th e Asparaginase pharmacokinetics and monitoring 2277 an intense and continuous dose schedule; and (3) switching to an alternative ASP when appropriate.
A comparison across studies shows that clinical hypersensitivity is most common when ASP is given without concurrent steroids, suggesting that the immunosuppressive eff ect of steroids might decrease the immunogenic risk [18,22,51,54]. However, the use of corticosteroids may also mask the presence of subclinical hypersensitivity, which is associated with signifi cantly reduced ASP activity levels.
Th e consistency and intensity of ASP therapy may also infl uence the risk of antibody formation. Studies show a reduced incidence of ASP antibodies in protocols where ASP was administered at a greater frequency and with fewer gaps between treatment periods [36,52]. It has been suggested that large concentrations of ASP may overwhelm the immune response and lower the likelihood of future hypersensitivity [12,36].
When clinical hypersensitivity or subclinical hypersensitivity is observed, patients should be immediately switched to a diff erent ASP formulation. Improved outcomes are seen when subclinical hypersensitivity can be rapidly identifi ed and appropriate treatment adjustments are made [48].
A recent study by Tong et al . focused on improving therapeutic benefi ts from ASP therapy by reducing the risk of hypersensitivity or subclinical hypersensitivity in children with ALL [32]. Patients were treated with PEG-ASP during consolidation, after receiving native E. coli ASP during induction. Th e authors reported high PEG-ASP activity levels in patients without clinical hypersensitivity (70% of the study population); however, 22% of patients developed allergy, and 8% showed subclinical hypersensitivity [32]. Importantly, patients with hypersensitivity or subclinical hypersensitivity showed no asparagine depletion and substantially lower ASP activity levels compared with patients who did not show an immune response [32]. Th ese fi ndings partially confi rm the results of an earlier investigation using a diff erent treatment schedule, and highlight the utility of trough ASP activity measurement to identify patients with subclinical hypersensitivity [55].

Route of administration
In Europe, ASP is commonly administered IV, while IM is the approved method of administration in the United States. Early studies indicated that IM administration was associated with a lower incidence of immune reaction [56]; however, a number of recent reports show comparable rates with both methods of administration [9,12,22,27,45]. In fact, IV administration may be preferable in some instances, such as in the case of a severe anaphylaxis, as infusion of the drug can be stopped immediately, sparing the patient from full exposure, whereas this is not possible with IM injections. Administration by IV also allows the use of high-continuousdose schedules without the pain and anxiety of numerous IM injections.
Following IV administration, the observed time to peak activity of ASP is much shorter, and may result in decreased time of exposure (area under the curve) compared with IM injections [57]. Th e elimination half-life of IV-administered ASP E. chrysanthemi was reported as 6.4 h in one study [57], while the half-life following IM injection of ASP E. chrysanthemi has been shown to be over 15 h [31]. Similar diff erences have been found with E. coli -derived formulations, with estimates ranging from 24 to 32 h [31,58]. Th e longer half-life reported with IM administration is largely a factor of the slow absorption rate following injections, limiting the rate of enzyme elimination. Due to this, the decline in plasma activity following IM injections represents both the absorption and elimination of the enzyme [1,57]. Furthermore, IV administration of ASP is associated with signifi cantly higher peak enzymatic activity levels and lower trough activity compared with an identical IM dose [42,57]. Th ese diff erences can have practical implications for asparagine depletion. A recent study evaluated trough ASP activity levels following an identical dose of ASP E. chrysanthemi 10 000 IU/m 2 administered either IV or IM in patients with ALL during reinduction [42]. Th e researchers found a greater percentage of patients with trough activity Ն 0.1 IU/mL in the IM group compared with the IV group (IM, 85%; IV, 55%). Th ese results highlight the need to account for pharmacokinetic diff erences between routes of drug administration when constructing treatment schedules. When ASP is administered IV, the frequency or dose may need to be increased to ensure continuous asparagine depletion.

Therapeutic drug monitoring and clinical implications
TDM in ALL refers to the regular measurement of asparagine concentrations, ASP activity or ASP antibodies as a means of evaluating a patient ' s current treatment schedule. Subclinical hypersensitivity and interpatient variability can result in suboptimal ASP activity and negatively impact outcomes, yet are not directly addressed in many treatment protocols [34,48]. TDM provides a means of identifying patients with suboptimal ASP activity and off ers physicians the necessary information to make informed adjustments in treatment.
A number of recent publications highlight the use of TDM in patients with ALL [5,36,42,48]. Of particular interest, Vrooman et al . report the results of a randomized clinical trial using TDM to guide individualized dose adjustment in patients undergoing ASP therapy [48]. In this study, patients were randomized into either a fi xed-dose (FD) or an individualized-dose (ID) group. Patients in the FD group were administered 25 000 IU/ m 2 native E. coli ASP, while patients in the ID group began therapy on a lower dose, but were evaluated every 3 weeks for the need to adjust treatment. Th e goal of ID was to maintain trough ASP activity between 0.1 IU/mL and 0.14 IU/mL. Th e study found signifi cantly higher 5-year EFS in the ID group (90%) compared with the FD group (82%, p ϭ 0.04), likely due to the ability to identify and adjust treatment for ID patients who developed subclinical hypersensitivity, as patients with clinical hypersensitivity switched preparations in both groups. A total of 19 patients (10%) in the ID group showed subclinical hypersensitivity and were subsequently switched to a diff erent ASP formulation (EFS, 95%). Following this change, 89% showed at least one trough ASP activity measurement Ն 0.1 IU/mL. In the FD group, 18 patients (9%) showed no measurable ASP activity Ն 0.1 IU/mL, with no overt signs of clinical assistance was provided by Cory Hussar, PhD, of Th e Curry Rockefeller Group, LLC, Tarrytown, NY, USA, and supported by Jazz Pharmaceuticals plc or its subsidiaries.

Potential confl ict of interest:
Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal. hypersensitivity, and therefore were not switched to an alternative ASP (EFS, 76%).
Several European treatment protocols recommend the monitoring of ASP activity; however, TDM is not yet commonplace in the United States [2]. Th e FDA approval of ASP E. chrysanthemi as a component of multiagent chemotherapeutic regimens in patients with ALL who have developed hypersensitivity to E. coli -derived ASP was largely based on results from TDM studies [33,59]. In Europe, the Dutch Children ' s Oncology Group, the AIEOP and the BFM group recommend the periodic measurement of ASP activity and antibody levels in real time to modulate ASP treatment [32,41]. In these protocols, TDM is primarily used to determine subclinical hypersensitivity and evaluate the need to switch ASP formulations; however, the use of TDM as a doseadjustment tool is increasing [32,41]. In the United States, a commercially available, Clinical Laboratory Improvement Amendments (CLIA)-certifi ed, ASP activity assay has recently become available for use in patients undergoing ASP therapy with all currently approved ASP formulations [60].

Conclusion
Th e complete and sustained depletion of asparagine is critical to achieving optimal outcomes with ASP therapy in the treatment of patients with ALL. Interpatient variability, the development of ASP antibodies, pharmacokinetic diff erences in ASP preparations and a host of other factors can aff ect the enzymatic activity following a given dose of ASP. To ensure optimal treatment, physicians must remain vigilant of these sources of variability and be ready to make any necessary treatment adjustments. As success rates in the treatment of ALL continue to increase, the goal can begin to shift toward improving the short-term and long-term quality of life of patients with ALL. Minimizing the number of long-term side eff ects, possibly through individually tailored treatment regimens, will be a key component toward achieving that goal.
Ongoing clinical studies promise to further clarify the relationship between ASP activity levels, outcomes and toxicity [41,61]. Th e increasing use of TDM will help identify patients with suboptimal enzyme activity and potentially inadequate asparagine depletion status. While there is a general consensus supporting the use of TDM to identify subclinical hypersensitivity, additional studies are needed to evaluate the benefi ts of TDM to guide dose adjustments in patients with measurable ASP activity. Additionally, a number of recent studies have reported promising results with ASP in the treatment of adolescent and young adult patients with ALL, a group with historically poor outcomes [62 -66]. Future studies are needed to evaluate ASP pharmacokinetics and optimal treatment schedules in these older patients. Th ese eff orts will further contribute to shed light on what has recently been termed the " asparaginase galaxy " [67].