Dose Optimization
Dashboard Systems: Pharmacokinetic/ Pharmacodynamic Mediated Dose Optimization for Monoclonal Antibodies
The Journal of Clinical Pharmacology 2015 55(S3) S51–S59 © 2015, The American College of Clinical Pharmacology DOI: 10.1002/jcph.370
Diane R. Mould, PhD1 and Marla C. Dubinsky, MD2
Abstract Many marketed drugs exhibit high variability in exposure and response. While these drugs are efficacious in their approved indications, finding appropriate dose regimens for individual patients is not straightforward. Similar dose adjustment problems are also seen with drugs that have a complex relationship between exposure and response and/or a narrow therapeutic window. This is particularly true for monoclonal antibodies, where prolonged dosing at a sub-therapeutic dose can also elicit anti-drug antibodies which will further compromise safety and efficacy. Thus, finding appropriate doses quickly would represent a substantial improvement in healthcare. Dashboard systems, which are decision-support tools, offer an improved, convenient means of tailoring treatment for individual patients. This article reviews the clinical need for this approach, particularly with monoclonal antibodies, the design, development, and testing of such systems, and the likely benefits of dashboard systems in clinical practice. We focus on infliximab for reference.
Keywords population modeling, pharmacokinetics, pharmacodynamics, inflammatory bowel disease, monoclonal antibodies
Background Clinical Need Therapeutic monoclonal antibodies (MAbs) targeting the tumor necrosis alpha pathway (anti-TNFa, anti-TNF) in the treatment of immune diseases such as rheumatoid arthritis, psoriasis, and inflammatory bowel diseases (IBD) have improved short and long term clinical outcomes. Crohn’s Disease (CD) and ulcerative colitis (UC), two main subtypes of IBD, are chronic diseases resulting from immune dysregulation in genetically susceptible individuals.1 CD and UC are conventionally treated using anti-inflammatory agents including aminosalicylate based therapies (mesalamines), corticosteroids, and antimetabolites such as purine analogs (azathioprine and 6-mercaptopurine) and methotrexate.2 A high percentage of patients fail to respond or are intolerant to these therapies and require treatment with anti-TNF. However, despite their therapeutic efficacy, approximately 25–30% of patients show no or limited response during induction therapy (primary non-responders) and in up to 50% of responders, treatment becomes ineffective during maintenance therapy despite initial response (secondary non-responders).3 Recent publications have underscored substantial variability in patient exposure and response when anti-TNF therapies are administered at the labeled induction and maintenance dose,4,5 supporting the need to individualize dosing to account for variability and ensure safe and sustainable efficacy.6,7 Suboptimal exposure can be attributed to under-dosing, rapid drug clearance and/or
the development of anti-drug antibodies (ADA) and can result in primary or secondary loss of response (LOR). Identifying an individual’s effective dose and adjusting the doses of anti-TNF over the course of treatment to maintain effective concentrations is not intuitive. Software-guided dosing has been shown to effectively control doses for individual patients and to increase efficiency in clinics.8 Individualized adaptive dosing using PK models has been undertaken9 but was a laborintensive process prior to using dashboard systems. Several dashboard systems already exist 10,11 to improve dosing in pediatric patients. Clinical use of such systems is still limited, in part because of a lack of familiarity with dashboards, ineffective communication to practicing clinical staff on the use and benefits of such systems to facilitate decision making,12 and the resources required to use modeling to fully individualize treatment. However in the case of pediatric patients, particularly for those
1
Projections Research, Phoenixville, PA 19460, USA Department of Pediatrics, IBD Center, Cedars-Sinai Medical Center, Los Angeles, CA, USA 2
Submitted for publication 14 April 2014; Revised 18 July 2014; accepted 18 July 2014. Corresponding Author: Diane R. Mould, PhD, FCP, Projections Research Inc., 535 Springview Lane, Phoenixville PA 19460 Email: [email protected]
S52 patients dosed based on body size (e.g. mg/kg or mg/m2), the drug exposure in pediatrics is often substantially lower than adult patients13 making these dosing metrics particularly difficult for patients with low body weight or pediatrics, as has been shown for infliximab.14 This suggests that pediatric patients would potentially garner the greatest benefit from individualized dosing. Pharmacokinetic variability for MAbs One potential cause of therapeutic failure and LOR is between-subject variability in exposure, which can arise from several sources, particularly MAb clearance and, for subcutaneously administered MAbs, from variability in the extent of absorption. Thus, it is important to understand these processes. MAb Clearance. The primary route of IgG clearance is intracellular proteolytic catabolism via the reticuloendothelial system (RES),15 although sites of catabolism have not been identified.16 Renal elimination does not contribute to MAb clearance, owing to their high molecular weight which limits glomerular filtration.17 While biliary is an elimination pathway for IgA antibodies, it does not contribute to the elimination of IgG antibodies.15 Target mediated clearance, a saturable route of elimination, can comprise a substantial component of MAb clearance. The contribution of receptor-mediated endocytocis to overall clearance depends on MAb concentration and target receptor expression, including the process of MAb-receptor internalization and receptor turnover rate,18 which can result in both nonlinear and time dependent changes in clearance.19 While frequently associated with cell surface antigens, target-mediated elimination has been reported for soluble antigens. The elimination process of large multimeric immune complexes may partly explain the nonlinear elimination of omalizumab, which interacts with soluble IgE.20 Antigens on cell surfaces may also be released into the serum, circulating as free antigens. Monoclonal antibodies can bind to shed receptors, which may accelerate clearance or decrease the concentration of free MAb through competitive binding.21 After binding to antigen, endogenous antibodies are salvaged by the Brambell receptor (FcRn) which also maintains albumin homeostasis22 by protecting IgG antibodies and albumin from catabolism, prolonging half-life.23 FcRn is primarily expressed in vascular endothelial cells or the RES, but is also present in lower levels on monocyte cell surfaces, tissue macrophages, and dendritic cells.24 FcRn receptors can be saturated at high IgG concentrations, shortening half-life, and reducing circulating levels of endogenous IgG and albumin.25 In some patients, such as those with multiple myeloma, high production of IgG antibodies results in a shortened halflife.26 Similarly, chronic inflammatory diseases which
The Journal of Clinical Pharmacology / Vol 55 No S3 (2015)
can present with elevated IgG levels, could reduce the half-life of exogenously administered MAbs. FcRn receptor affinity is species specific, which helps to explain the fact that murine MAbs have a short half-life in humans.27 Thus, half-life generally increases as the MAb structure becomes more human, with murine MAbs having a very short half-life (1–2 days); chimeric MAbs having a half-life of approximately 10–14 days, and humanized and fully human antibodies having half-lives generally greater than 10 days.28 Three other classes of Fc gamma receptors (FcgRI, FcgII, and FcgIII) have been identified.29 These receptors are expressed by macrophages, natural killer cells, B and T cells, and platelets. Binding to these receptors elicits complement or antibody dependent cell cytotoxicity.30 MAb clearance through the RES may be partly regulated through interactions by FcgRs. Thus, FcgR may contribute to elimination of circulating antibodies as reducing MAb binding affinity to FcgR slows clearance.31 Furthermore, polymorphisms in FcgR have been associated with clinical response to antibodies directed against tumor necrosis factor (TNF) in IBD patients.32 MAbs are exogenous proteins, and regardless of extent of humanization, can induce an immune response leading to ADA production.33 Numerous factors can impact ADA formation, including the formulation, its stability, extent of humanization, dose regimen, and treatment duration.34 Formation of aggregates is an important predictor of ADA.35 While the intravenous route of administration is generally least likely to induce an ADA response, this is not always true.36 The presence of ADA appears to occur more frequently following administration of low doses than with high doses of MAbs.37 ADAs can be neutralizing, in which MAb binding is impaired, or can increase MAb clearance;38 ADA formation usually compromises efficacy.39 MAb Absorption. Following subcutaneous or intramuscular administration, MAbs are thought to be absorbed primarily via convective transport of antibody through lymphatic vessels and into the blood,40 although other research suggests that diffusion of the MAb across blood vessels distributed near the site of injection may play a substantial role in absorption.41 In addition to its role in absorption, convection can play a substantial role for MAb distribution. Furthermore, MAb clearance from the lymph and the vascular system can be quite different (CLlymph >> CLvascular). Following administration, MAb absorption is slow, incomplete, and variable, with the time to reach maximal plasma concentrations generally ranging between 2 and 8 days. Approximately 50%– 100% of the administered dose is absorbed 42 with the remaining dose being lost through presystemic catabolism (e.g., proteolysis, endocytosis, or target mediated clearance). This loss is balanced with FcRn recycling.15 Other factors have been identified as impacting absorption via
S53
Mould et al
effecting lymph and blood flow, including age, body weight (particularly in obese subjects), and injection site.43 Impact of Variability on Safety and Efficacy. Sustaining a durable response to anti-TNF therapy beyond the 1 year clinical trial outcome measure is a priority for clinicians and patients with IBD. Research efforts have focused on both genetic predictors as well as the influence of pharmacokinetics and immunogenicity on the safety and efficacy of anti-TNF therapies.44,45 Infliximab was first approved for induction therapy in 1998 and initially was used on an episodic basis, which was very immunogenic. Clinical studies have confirmed that regularly scheduled maintenance therapy is a superior dosing strategy.46 Baert et al.47 first reported that those patients who developed anti-infliximab antibodies (ATI, a specific form of ADA) lost response to therapy faster than those with none or lower levels of ATI. Specifically, ATI concentrations 8 mg/mL were predictive of shorter duration of response, whereas ATI concentrations
Dashboard Systems: Pharmacokinetic/ Pharmacodynamic Mediated Dose Optimization for Monoclonal Antibodies
The Journal of Clinical Pharmacology 2015 55(S3) S51–S59 © 2015, The American College of Clinical Pharmacology DOI: 10.1002/jcph.370
Diane R. Mould, PhD1 and Marla C. Dubinsky, MD2
Abstract Many marketed drugs exhibit high variability in exposure and response. While these drugs are efficacious in their approved indications, finding appropriate dose regimens for individual patients is not straightforward. Similar dose adjustment problems are also seen with drugs that have a complex relationship between exposure and response and/or a narrow therapeutic window. This is particularly true for monoclonal antibodies, where prolonged dosing at a sub-therapeutic dose can also elicit anti-drug antibodies which will further compromise safety and efficacy. Thus, finding appropriate doses quickly would represent a substantial improvement in healthcare. Dashboard systems, which are decision-support tools, offer an improved, convenient means of tailoring treatment for individual patients. This article reviews the clinical need for this approach, particularly with monoclonal antibodies, the design, development, and testing of such systems, and the likely benefits of dashboard systems in clinical practice. We focus on infliximab for reference.
Keywords population modeling, pharmacokinetics, pharmacodynamics, inflammatory bowel disease, monoclonal antibodies
Background Clinical Need Therapeutic monoclonal antibodies (MAbs) targeting the tumor necrosis alpha pathway (anti-TNFa, anti-TNF) in the treatment of immune diseases such as rheumatoid arthritis, psoriasis, and inflammatory bowel diseases (IBD) have improved short and long term clinical outcomes. Crohn’s Disease (CD) and ulcerative colitis (UC), two main subtypes of IBD, are chronic diseases resulting from immune dysregulation in genetically susceptible individuals.1 CD and UC are conventionally treated using anti-inflammatory agents including aminosalicylate based therapies (mesalamines), corticosteroids, and antimetabolites such as purine analogs (azathioprine and 6-mercaptopurine) and methotrexate.2 A high percentage of patients fail to respond or are intolerant to these therapies and require treatment with anti-TNF. However, despite their therapeutic efficacy, approximately 25–30% of patients show no or limited response during induction therapy (primary non-responders) and in up to 50% of responders, treatment becomes ineffective during maintenance therapy despite initial response (secondary non-responders).3 Recent publications have underscored substantial variability in patient exposure and response when anti-TNF therapies are administered at the labeled induction and maintenance dose,4,5 supporting the need to individualize dosing to account for variability and ensure safe and sustainable efficacy.6,7 Suboptimal exposure can be attributed to under-dosing, rapid drug clearance and/or
the development of anti-drug antibodies (ADA) and can result in primary or secondary loss of response (LOR). Identifying an individual’s effective dose and adjusting the doses of anti-TNF over the course of treatment to maintain effective concentrations is not intuitive. Software-guided dosing has been shown to effectively control doses for individual patients and to increase efficiency in clinics.8 Individualized adaptive dosing using PK models has been undertaken9 but was a laborintensive process prior to using dashboard systems. Several dashboard systems already exist 10,11 to improve dosing in pediatric patients. Clinical use of such systems is still limited, in part because of a lack of familiarity with dashboards, ineffective communication to practicing clinical staff on the use and benefits of such systems to facilitate decision making,12 and the resources required to use modeling to fully individualize treatment. However in the case of pediatric patients, particularly for those
1
Projections Research, Phoenixville, PA 19460, USA Department of Pediatrics, IBD Center, Cedars-Sinai Medical Center, Los Angeles, CA, USA 2
Submitted for publication 14 April 2014; Revised 18 July 2014; accepted 18 July 2014. Corresponding Author: Diane R. Mould, PhD, FCP, Projections Research Inc., 535 Springview Lane, Phoenixville PA 19460 Email: [email protected]
S52 patients dosed based on body size (e.g. mg/kg or mg/m2), the drug exposure in pediatrics is often substantially lower than adult patients13 making these dosing metrics particularly difficult for patients with low body weight or pediatrics, as has been shown for infliximab.14 This suggests that pediatric patients would potentially garner the greatest benefit from individualized dosing. Pharmacokinetic variability for MAbs One potential cause of therapeutic failure and LOR is between-subject variability in exposure, which can arise from several sources, particularly MAb clearance and, for subcutaneously administered MAbs, from variability in the extent of absorption. Thus, it is important to understand these processes. MAb Clearance. The primary route of IgG clearance is intracellular proteolytic catabolism via the reticuloendothelial system (RES),15 although sites of catabolism have not been identified.16 Renal elimination does not contribute to MAb clearance, owing to their high molecular weight which limits glomerular filtration.17 While biliary is an elimination pathway for IgA antibodies, it does not contribute to the elimination of IgG antibodies.15 Target mediated clearance, a saturable route of elimination, can comprise a substantial component of MAb clearance. The contribution of receptor-mediated endocytocis to overall clearance depends on MAb concentration and target receptor expression, including the process of MAb-receptor internalization and receptor turnover rate,18 which can result in both nonlinear and time dependent changes in clearance.19 While frequently associated with cell surface antigens, target-mediated elimination has been reported for soluble antigens. The elimination process of large multimeric immune complexes may partly explain the nonlinear elimination of omalizumab, which interacts with soluble IgE.20 Antigens on cell surfaces may also be released into the serum, circulating as free antigens. Monoclonal antibodies can bind to shed receptors, which may accelerate clearance or decrease the concentration of free MAb through competitive binding.21 After binding to antigen, endogenous antibodies are salvaged by the Brambell receptor (FcRn) which also maintains albumin homeostasis22 by protecting IgG antibodies and albumin from catabolism, prolonging half-life.23 FcRn is primarily expressed in vascular endothelial cells or the RES, but is also present in lower levels on monocyte cell surfaces, tissue macrophages, and dendritic cells.24 FcRn receptors can be saturated at high IgG concentrations, shortening half-life, and reducing circulating levels of endogenous IgG and albumin.25 In some patients, such as those with multiple myeloma, high production of IgG antibodies results in a shortened halflife.26 Similarly, chronic inflammatory diseases which
The Journal of Clinical Pharmacology / Vol 55 No S3 (2015)
can present with elevated IgG levels, could reduce the half-life of exogenously administered MAbs. FcRn receptor affinity is species specific, which helps to explain the fact that murine MAbs have a short half-life in humans.27 Thus, half-life generally increases as the MAb structure becomes more human, with murine MAbs having a very short half-life (1–2 days); chimeric MAbs having a half-life of approximately 10–14 days, and humanized and fully human antibodies having half-lives generally greater than 10 days.28 Three other classes of Fc gamma receptors (FcgRI, FcgII, and FcgIII) have been identified.29 These receptors are expressed by macrophages, natural killer cells, B and T cells, and platelets. Binding to these receptors elicits complement or antibody dependent cell cytotoxicity.30 MAb clearance through the RES may be partly regulated through interactions by FcgRs. Thus, FcgR may contribute to elimination of circulating antibodies as reducing MAb binding affinity to FcgR slows clearance.31 Furthermore, polymorphisms in FcgR have been associated with clinical response to antibodies directed against tumor necrosis factor (TNF) in IBD patients.32 MAbs are exogenous proteins, and regardless of extent of humanization, can induce an immune response leading to ADA production.33 Numerous factors can impact ADA formation, including the formulation, its stability, extent of humanization, dose regimen, and treatment duration.34 Formation of aggregates is an important predictor of ADA.35 While the intravenous route of administration is generally least likely to induce an ADA response, this is not always true.36 The presence of ADA appears to occur more frequently following administration of low doses than with high doses of MAbs.37 ADAs can be neutralizing, in which MAb binding is impaired, or can increase MAb clearance;38 ADA formation usually compromises efficacy.39 MAb Absorption. Following subcutaneous or intramuscular administration, MAbs are thought to be absorbed primarily via convective transport of antibody through lymphatic vessels and into the blood,40 although other research suggests that diffusion of the MAb across blood vessels distributed near the site of injection may play a substantial role in absorption.41 In addition to its role in absorption, convection can play a substantial role for MAb distribution. Furthermore, MAb clearance from the lymph and the vascular system can be quite different (CLlymph >> CLvascular). Following administration, MAb absorption is slow, incomplete, and variable, with the time to reach maximal plasma concentrations generally ranging between 2 and 8 days. Approximately 50%– 100% of the administered dose is absorbed 42 with the remaining dose being lost through presystemic catabolism (e.g., proteolysis, endocytosis, or target mediated clearance). This loss is balanced with FcRn recycling.15 Other factors have been identified as impacting absorption via
S53
Mould et al
effecting lymph and blood flow, including age, body weight (particularly in obese subjects), and injection site.43 Impact of Variability on Safety and Efficacy. Sustaining a durable response to anti-TNF therapy beyond the 1 year clinical trial outcome measure is a priority for clinicians and patients with IBD. Research efforts have focused on both genetic predictors as well as the influence of pharmacokinetics and immunogenicity on the safety and efficacy of anti-TNF therapies.44,45 Infliximab was first approved for induction therapy in 1998 and initially was used on an episodic basis, which was very immunogenic. Clinical studies have confirmed that regularly scheduled maintenance therapy is a superior dosing strategy.46 Baert et al.47 first reported that those patients who developed anti-infliximab antibodies (ATI, a specific form of ADA) lost response to therapy faster than those with none or lower levels of ATI. Specifically, ATI concentrations 8 mg/mL were predictive of shorter duration of response, whereas ATI concentrations
