The Lymphatic System Plays a Major Role in the ... - ACS Publications

18 Dec 2013 - This study highlights for the first time the significant role of the lymphatic ... Josefin Koehn , Loren Kinman , Jianguo Sun , Ann C. C...
1 downloads 0 Views 2MB Size
Subscriber access provided by DUESSELDORF LIBRARIES

Article

The lymphatic system plays a major role in the intravenous and subcutaneous pharmacokinetics of trastuzumab in rats Annette Dahlberg, Lisa M Kaminskas, Alanna Smith, Joseph A. Nicolazzo, Christopher J.H. Porter, Jurgen B Bulitta, and Michelle P McIntosh Mol. Pharmaceutics, Just Accepted Manuscript • Publication Date (Web): 18 Dec 2013 Downloaded from http://pubs.acs.org on December 24, 2013

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Molecular Pharmaceutics is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

The lymphatic system plays a major role in the intravenous and subcutaneous pharmacokinetics of trastuzumab in rats Annette M. Dahlberg 1†, Lisa M Kaminskas1†, Alanna Smith1, Joseph A. Nicolazzo1, Christopher J. H. Porter1, Jürgen B. Bulitta2, Michelle P. McIntosh1* 1

Drug Delivery Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences, Monash University, 381 Royal Pde, Parkville, VIC, Australia, 3052

2

Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, 381 Royal Pde, Parkville, VIC, Australia, 3052



AMD and LMK contributed equally to this manuscript

Corresponding Author Dr Michelle P. McIntosh Drug Delivery Disposition Dynamics Group Monash Institute of Pharmaceutical Sciences 381 Royal Pde Parkville, VIC Australia, 3052 Ph: +61 (3) 9903 9531 Fax: +61 (3) 9903 9583 Email: [email protected] Keywords: trastuzumab, lymphatic, pharmacokinetics, monoclonal antibody, subcutaneous TOC: SC PHARMACOKINETICS OF TRASTUZUMAB Vascular capillary

X%

Central comp

Y% Lymph capillary

Redistribution to interstitium

1

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 38

Abstract: Therapeutic monoclonal antibodies are currently delivered mainly via the intravenous route, since large volumes are often required to deliver a therapeutic dose. Administration via the subcutaneous route would have several therapeutic advantages, the absorption mechanisms for antibodies dosed subcutaneously are poorly understood. This study was conducted to develop a better understanding of the mechanisms governing the subcutaneous absorption and trafficking of monoclonal antibodies. Specifically, the role of the lymphatic system in the absorption and prolonged plasma exposure of trastuzumab was explored in thoracic lymph duct cannulated rats after SC and IV dosing. A population pharmacokinetic model was developed in S-ADAPT to simultaneously fit all plasma and lymph concentrations and to predict the pharmacokinetics in non-lymph-cannulated animals. The estimated absolute bioavailability of trastuzumab after SC administration in rats was 85.5%. Following SC administration, 53.1% of the trastuzumab dose was absorbed via a first-order process (mean absorption time: 99.6 h) into the peripheral lymph compartment and 32.4% of the dose was absorbed by a Michaelis-Menten process into the central compartment. Recovery in thoracic lymph over 30 h was 26.7% after SC and 44.1% after IV administration. This study highlights for the first time the significant role of the lymphatic system in maintaining the long plasma exposure of trastuzumab, with the model predicting an extensive distribution of this monoclonal antibody into the lymph following SC and IV administration. This extensive direct absorption from the SC injection site into lymph may enable novel therapeutic strategies for the treatment of lymph resident metastatic cancer.

2

ACS Paragon Plus Environment

Page 3 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Introduction: Breast cancer is the most common cancer among the female population, with an incidence of 1.4 million new diagnoses per year and an annual mortality rate of 450,000 women worldwide 1. Prognosis is significantly poorer for patients with a cancer that overexpresses human epidermal growth factor receptor 2 (HER-2). This receptor is overexpressed in approximately 15-30% of breast cancers, and is indicative of highly aggressive and metastatic cancer 2-4. The anti-HER-2 monoclonal antibody trastuzumab (150 kDa, marketed as Herceptin®) was developed as an adjunct to chemotherapy with small molecule cytotoxic drugs to improve the treatment of HER-2-positive breast cancer. The antibody specifically targets the extracellular domain of the HER-2 receptor5 and has been shown to dramatically improve therapeutic outcomes in HER-2-positive breast cancer patients 6. Trastuzumab is currently approved for intravenous (IV) administration, however, there is significant interest in formulating trastuzumab to allow for administration via the subcutaneous (SC) route 7, 8. This route of delivery has the potential to facilitate greater ease of administration, reduced administration time and reduced infusion related side effects 9, 10 To date, however, little is known about how monoclonal antibodies, such as trastuzumab, are absorbed into the systemic circulation and lymphatic system following SC administration. Studies conducted in sheep have shown that as the molecular weight of proteins increase, the proportion of the dose absorbed by the lymphatics following SC injection increases. In general, proteins with molecular weights exceeding 16 kDa are preferentially absorbed by the lymphatic system11. This occurs as a result of the reduced vascular permeability of large proteins, promoting uptake via the lymphatic system, the vessels of which are inherently more leaky than the peripheral circulatory vessels12. In these studies the thoracic or popliteal lymph ducts were cannulated to enable the continuous collection of central or peripheral lymph in order to determine the proportion of a SC dose that was absorbed via the lymph11,

13-17

. The largest

protein investigated in terms of lymphatic absorption had a molecular weight of 84 kDa 12 While there is a wealth of information on the lymphatic disposition of SC-administered smaller proteins, little is known about the lymphatic uptake of very large proteins such as monoclonal antibodies. The nanoparticle and colloid literature suggests, however, that very large particles display reduced lymphatic absorption by virtue of reduced convection from the 3

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 38

injection site 18. It is not known whether a similar phenomenon also exists for very large proteins. It is known that the biodistribution and trafficking of antibodies is influenced by the neonatal Fc receptor (FcRn) which is expressed on various cells throughout the body, including macrophages, smooth muscle cells and epithelial cells 19. The FcRn receptor is not expressed on vascular capillary or lymph endothelial cells

19

. Additionally, epithelial transcellular transport

may be mediated by the binding of antibodies to cell surface antigens or the Fcγ

20, 21

. These

targets for antibody binding may therefore influence the pathways by which trastuzumab is absorbed from an SC injection site and biodistributed within the body. A further consideration in these studies was that trastuzumab targets metastatic breast cancers overexpressing Her-2 and the lymphatic system is a major pathway for the metastatic spread of cancers. Therefore, if circulating lymphatic concentrations of trastuzumab are elevated in the region of SC injections, future studies will be warranted to investigate if this can be exploited for a therapeutic advantage. Given the limited data on the absorption and disposition of trastuzumab in lymph, the main objective of this study was to characterize the lymphatic uptake and disposition of trastuzumab following SC and IV administration in rats. The second objective was to develop a novel population pharmacokinetic model based on the physiology of the lymphatic system that can quantitatively describe the time-course of trastuzumab concentrations in plasma and lymph after administration. This model enabled, for the first time, the prediction of the long-term pharmacokinetics of a mAb in both plasma and lymph.

4

ACS Paragon Plus Environment

Page 5 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Materials and Methods: Materials Trastuzumab (Herceptin®) was purchased as the lyophilized product from Roche Pty Ltd (Dee Why, NSW, Australia). Costar® 96 well microplates, Tween-20, 30% hydrogen peroxide and 3,3’,5,5’-tetramethylbenzidine (TMB) chromogen solution were purchased from Sigma (NSW, Australia). Goat anti-human IgG affinity purified polyclonal antibody and horseradish peroxidase-conjugated anti-human IgG raised in goats were obtained from Millipore (Kilsyth, VIC, Australia). Sterile saline was purchased from Baxter (Toongabbie, NSW, Australia). All other reagents were AR grade. Medical grade polyvinyl, polyethylene and silastic tubing (0.58 mm internal diameter, 0.96 mm external diameter) was purchased from microtube extrusions (NSW, Australia).

Animals Sprague Dawley rats (male, 260-320 g) were obtained from the Animal Resources Centre (Perth, WA, Australia). Animals were individually housed in metabolic cages following surgery and during the experimental period. Rats were supplied with water and food ad libitum with the exception of overnight following surgery and for the 8 hours proceeding dosing when food was withheld. All experiments involving rats were approved by the Monash Institute of Pharmaceutical Sciences Animal Ethics Committee (Monash University, VIC, Australia) and were performed in accordance with the Australian National Health and Medical Research Council (NHMRC) guidelines for the care and use of animals for scientific purposes.

Dosing groups and experimental design The

experimental

design

involved

the

parallel

examination

of

trastuzumab

pharmacokinetics in six groups of animals: (1) an IV dosed control group (without thoracic lymph duct cannulation) in carotid artery cannulated rats over 5 days, (2) an IV control group (without thoracic lymph duct cannulation) in non-carotid artery cannulated rats over 2 months, (3) a SC dosed control group (without thoracic lymph duct cannulation) in carotid artery cannulated rats over 5 days, (4) a SC dosed control group (without thoracic lymph duct cannulation) in non-carotid artery cannulated rats over 2 months, (5) an IV dosed lymph5

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 38

cannulated group (containing a thoracic lymph duct cannula) over 30 h and (6) a SC dosed lymph-cannulated group (containing a thoracic lymph duct cannula) over 30 h. IV and SC control rats bearing carotid artery cannulae enabled frequent blood sampling in conscious, freely moving rats to determine the plasma pharmacokinetics of trastuzumab over the initial 5 day postdose period. Institutional animal ethics guidelines preclude experimentation in individually housed, cannulated rats beyond a period of one week, and so long term trastuzumab plasma pharmacokinetics in control rats were determined in a second cohort of rats where blood was sampled once or twice per week for 2 months via the tail vein. Samples were collected from lymph cannulated rats for a maximum of 30 h post-dose as ethical considerations precluded continuous collection of lymph beyond this time.

Surgical implantation of cannulae Indwelling polyethylene cannulae were surgically implanted into the right carotid artery of rats in groups 1, 3, 5 and 6 to enable blood sample collection as previously described

22

.

Polyethylene cannulae were also surgically implanted into the right jugular vein of rats in groups 1, 5 and 6 to enable IV dosing and the replacement of fluid lost via the thoracic lymph duct cannula as previously described

22, 23

. Polyvinyl cannulae were surgically implanted into the

thoracic lymph duct immediately below the diaphragm of rats in groups 5 and 6 as previously described

23

. All cannulae were exteriorized to the back of the neck and cannulated rats were

connected to a swivel-tether apparatus to allow blood sampling from conscious freely moving animals. Rats were transferred into individual metabolism cages following surgery. Animals were allowed to recover overnight prior to drug administration. Long term rats (in groups 2 and 4) were group housed in microisolator cages.

Trastuzumab administration and sampling Rats in the short term IV control and IV lymph cannulated groups (groups 1 and 5 above) were administered trastuzumab via the jugular vein cannula at a dose of 2 mg/kg as a 1 ml bolus in saline over 2 min. Rats in the long term IV control group (group 2 above) were administered 2 mg/kg trastuzumab via a lateral tail vein in a volume of 0.5 ml saline. In order to determine if trastuzumab displays dose linearity at doses that result in plasma concentrations similar to those 6

ACS Paragon Plus Environment

Page 7 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

obtained after SC administration of the antibody, rats were also administered 0.2 (n=3) or 0.02 mg/kg (n=4) trastuzumab IV respectively and plasma pharmacokinetics measured over 5 days. Rats in the short term and long term SC control and SC lymph cannulated groups (groups 3, 4 and 6 above) were administered 2 mg/kg trastuzumab in a volume of 0.15 ml saline into the inner left hind leg approximately 0.5 cm above the ankle. Blood (0.15 ml) was collected into heparinized eppendorf tubes via the carotid artery cannula (for short term rats) or via an alternate lateral tail vein (for long term rats) and centrifuged at 1150 g for 10 min to isolate plasma. Lymph was collected continuously via the thoracic lymph duct cannula into heparinized tubes and the volume of lymph collected was recorded. Plasma and lymph samples were stored at -80 °C until analysis.

Analytical method Trastuzumab concentrations in rat plasma and lymph were determined using a Generic Immunoglobulin Pharmacokinetic (GRIP) Enzyme Linked Immunosorbent Assay (ELISA) modified from Yang et al 24 which specifically targets human monoclonal antibodies and can be used to quantify human or humanized IgG in non-human biological matrices. Microtiter plates were coated with 100 µl of a goat anti-human IgG capture antibody at a concentration of 1.25 µg/mL in 50 mM sodium carbonate solution (pH = 9.6) overnight at 4 °C. Following washing with 5 changes of wash solution (200 µl of PBS containing 0.05% Tween-20 (v/v)), plates were subsequently blocked with 200 µl of PBS containing 0.05% (v/v) Tween-20 and 5% (w/v) skim milk powder for 2 h at 37 °C to prevent non-specific binding. Plates were then washed 5 times before the addition of samples or standard solutions. Standard curves and quality control samples in either plasma or lymph were prepared on the day of dosing and frozen together with the study samples to match the storage conditions of the samples. Standards comprised plasma or lymph spiked with trastuzumab at concentrations ranging from 30 to 100,000 ng/mL and were diluted either 1:10 or 1:100 (to 100 µl) in sample diluent (PBS, 0.05% (v/v) Tween-20, 1% (w/v) skim milk powder). Intra- and inter-day accuracy and precision for the assay was also determined and is reported in the supplementary information. Plasma or lymph samples were diluted in sample diluent to fall within the linear range of the standard curve (300 to 3,000 ng/mL) and the linearity of the dilution was validated. 7

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 38

Standards, quality control samples or study samples (100 µl) were added to each well and incubated for 1 h at 37 °C. After washing 5 times, the detection antibody (goat anti-human IgG conjugated to horseradish peroxidase, diluted to 1:32,000 in sample diluent, 100 µl) was then added to each well. The plates were then washed again and TMB chromogen solution (100 µl) was added for 10 min and the reaction was stopped by the addition of a 100 µl aliquot of 1 M phosphoric acid. The absorbance was read at 450 nm on a Fluostar Optima microplate reader (BMG Labtech, Mornington, VIC, Australia).

Non-compartmental pharmacokinetic analysis The peak concentrations (Cmax) and the times for their occurrence (Tmax) were directly read from the observed plasma concentration versus time curves. The terminal half-life was determined by linear regression of the plasma concentration-time profiles on a semi-logarithmic scale. Since trastuzumab is known to have a prolonged circulatory half life, groups of control animals sampled for 2 months were used to assess the terminal phase. Non-compartmental analysis was performed in WinNonlin® Pro (Version 5.3, Pharsight Corp., Mountain View, CA). The area under the plasma concentration-time curve from time zero to the last quantifiable concentration was calculated using the linear up / log down rule as implemented in WinNonlin®. For extrapolation to time infinity via non-compartmental analysis, the plasma concentrations from the group of non-cannulated rats sampled over two months were used. In the lymph cannulated groups, the recovery in lymph was determined by multiplying the measured concentration in each sampling interval by the volume of lymph. The cumulative recovery in lymph (Flymph) was subsequently calculated as the total amount collected divided by the administered dose.

Population Pharmacokinetic Modeling Population modeling was performed to gain a more thorough understanding of the lymphatic disposition of trastuzumab and to better evaluate the impact of the point of lymph collection relative to the distribution of lymphatic vessels in rats. This modeling approach enables one to predict the time-course of plasma concentrations and amounts in lymph over time in non-lymph-cannulated (ie ‘normal’) animals. All plasma concentrations and amounts 8

ACS Paragon Plus Environment

Page 9 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

recovered in lymph were simultaneously fitted via population pharmacokinetic modeling in the parallelized S-ADAPT software

25

using maximum likelihood estimation. The analysis was

facilitated by the SADAPT-TRAN pre- and post-processor

26

. Competing models were

distinguished by comparison of the objective function, plausibility of parameter estimates, curve fits in plasma and lymph, visual predictive checks, normalized prediction distribution errors

27

and other standard diagnostic plots as described previously 28. Structural model: Candidate models contained a central compartment with or without a peripheral compartment for distribution into peripheral sites (excluding lymph; Fig 1). After SC dosing into the injection site compartment (amount: A1), drug entered the peripheral lymph compartment (A4) via a first-order process and the central (‘blood’) compartment (A2) by a Michaelis-Menten process with a maximum rate of transfer (Vmax) and Michaelis-Menten constant (Km). The initial conditions of all differential equations were zero:

dA1  Vmax  = −  + k 14  ⋅ A1 dt  Km + A1 

(1)

For a short-term IV infusion, the time-delimited zero-order input rate is denoted as R(1). The differential equation for the amount of drug in the central (A2) and peripheral (non-lymph) compartment (A3) was:

dA2 Vmax = R(1) + ⋅ A1 − (CL + CLd) ⋅ C2 + CLd ⋅ C3 − k 24 ⋅ A2 − k 26 ⋅ A2 dt Km + A1 + k 51 ⋅ (1− Cann) ⋅ A5 + k 71 ⋅ A7

(2)

dA3 = (C2 − C3) ⋅ CLd − k 34 ⋅ A 3 − k 36 ⋅ A 3 dt

(3)

C2 is the trastuzumab concentration in the central compartment and C3 is the trastuzumab concentration in the peripheral (non-lymph) compartment. The variable Cann denotes a switch for cannulated animals (Cann=1) and non-cannulated animals (Cann=0) as described below. Based on the physiology of the lymphatic system, total lymph was separated into lymph draining the ‘posterior’ and ‘anterior’ aspects of the rat (Fig 1). Distribution from the central compartment into the anterior and posterior peripheral lymph compartments was assumed to occur via extravasation of the antibody across capillary beds throughout the body and reabsorption into lymph. Drug in peripheral lymph was assumed to flow into central lymph and subsequently to enter the central compartment via the junction of the thoracic duct with the internal jugular vein. 9

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

Distribution from the central compartment (A2) to the posterior (k24) and anterior (k26) peripheral lymph compartments was modeled via first-order kinetics. The same was assumed for the transfer from the peripheral compartment (A3) to the posterior (k34) and anterior (k36) peripheral lymph compartments. The transfer from peripheral lymph (A4) to central lymph (A5) was also modeled as a first-order process (k45). This yields the following differential equations for the amount of trastuzumab in the posterior peripheral lymph (A4) and posterior central lymph (A5) compartments:

dA4 = k14 ⋅ A1 + k 24 ⋅ A2 + k 34 ⋅ A3 − k 45 ⋅ A4 dt

(4)

dA5 = k 45 ⋅ A4 − k 51 ⋅ A 5 dt

(5)

The proposed model structure (Fig. 1) assumes that all lymph from the posterior aspect of the rat drains into the thoracic duct and is collected in all lymph cannulated animals. In contrast, lymph from the anterior half drains into the right lymphatic duct and the anterior part of the thoracic lymph duct, and therefore enters the central compartment without being collected (Fig. 1). This assumption enabled the assessment of the transfer of the SC dose into blood vs lymph, since lymph from the inner leg drains almost exclusively via the popliteal lymph node towards the iliac node and the thoracic lymph duct posterior to the cannula. This model is a slight simplification as we realize that some lymph from the posterior half of the rat drains into the thoracic lymph above the cannulation point and is therefore not collected. Lymph concentrations in anterior lymph were not measured, since the right thoracic duct is not easily accessed in a rat and the thoracic lymph duct can only be cannulated posterior to the diaphragm. Transfer parameters were therefore assumed to be the same for the posterior and anterior aspects of the lymphatic system and the differential equations for the anterior peripheral (A6) and anterior central lymph (A7) were similar to the equations for A4 and A5, with the exception of input from the SC injection site (since the lymph draining the injection site was assumed to be posterior lymph) :

dA6 = k 26 ⋅ A2 + k 36 ⋅ A3 − k 67 ⋅ A6 dt

(6)

dA7 = k 67 ⋅ A6 − k 71 ⋅ A7 dt

(7)

10

ACS Paragon Plus Environment

Page 11 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

For lymph cannulated rats, the central lymph flowing into the central compartment was quantitatively collected via cannulation of the thoracic duct and therefore did not enter the central compartment, if the switch variable Cann was set to 1. Instead, this drug in lymph entered the collected lymph compartment (A8):

dA8 = Cann ⋅ k 51 ⋅ A5 dt

(8)

Parameter variability and residual error model: An exponential parameter variability model was used for all parameters. Plasma concentrations were fitted using an additive plus proportional residual error model. The individual amounts collected in lymph during each dosing interval were fitted using an additive plus proportional residual error following standard procedures for modeling of urinary excretion data via population methods 29.

Statistics Non-compartmental pharmacokinetic parameters calculated for the IV control group were compared to those for the SC control group via unpaired Student’s t-tests (Prism 6 for Windows, version 6.00, GraphPad Software, Inc.). Similarly, pharmacokinetic parameters for control groups were compared to the pharmacokinetic parameters for lymph cannulated groups via unpaired Student’s t-tests. For a descriptive comparison, plasma concentration vs. time curves in control and lymph cannulated groups were compared via two way ANOVA with Bonferroni test for significant differences at each time point. Significance was determined using an α of 0.05.

11

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

Results: Non-compartmental pharmacokinetics The plasma concentration-time profiles of trastuzumab at three IV dose levels (Fig. 2) suggest a linear increase of AUC0-120h with dose. The AUC0-120h was 15.2 ± 4.38 mg·h/L for the 0.02 mg/kg dose group, 211 ± 15.2 mg·h/L for the 0.2 mg/kg dose group and 2113 ± 444 mg·h/L for the 2 mg/kg dose group. For all dose levels plasma concentrations declined slowly between approximately 24 and 120 h. The Tmax following SC injection of trastuzumab in control rats was approximately 3 days, suggesting slow absorption after SC administration (Fig 3A).The plasma concentrations following SC and IV administration of 2 mg/kg (Fig. 3, Panels A and B) were lower for cannulated compared to non-cannulated animals in agreement with a significant fraction of dose being recovered in lymph from 0 to 30 h (Fig 3C; 44.1 ± 12.6% for IV dosing and 26.7 ± 10.4% for SC dosing; Table 1). Due to the large between animal variability, however, the fraction recovered in lymph did not differ significantly between the IV and SC group. The terminal half-life in control animals was calculated based on the groups of animals sampled for two months (see supplementary material) and was 285 ± 111 h for the IV group and 262 ± 200 h for the SC group. Due to the availability of data after IV dosing, population PK modeling was able to estimate whether this long terminal half-life was caused by the elimination or absorption phase after SC dosing. The collection of thoracic lymph caused a reduction in AUC0-30h from 776 ± 193 to 700 ± 133 mg/L.h and 143 ± 112 to 44.9 ± 17.9 mg/L.h in IV and SC dosed rats respectively compared to control (Table 1 and Figure 3A and B), suggesting a substantial role of the lymphatic system in the absorption and plasma disposition of trastuzumab. The extensive lymphatic recovery observed following IV administration implies that the antibody can readily access the lymphatics via redistribution from the systemic circulation.

Compartmental Pharmacokinetic analysis The developed population PK model yielded unbiased and reasonably precise individual and population fits for the plasma concentrations and amounts excreted into lymph (Fig 3, panels

12

ACS Paragon Plus Environment

Page 13 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

D to I). The final model included a central and a peripheral compartment as well as two sets of compartments for the posterior and anterior loop to describe distribution in lymph (Fig 1). For a SC injection of 588 µg (equivalent to a dose of 2 mg/kg in the rats), 53.1% of the dose was absorbed via a first-order process (mean absorption time: 99.6 h) from the injection site into the peripheral lymph compartment and 32.4% of dose was absorbed by a Michaelis-Menten process into the central compartment. A Michaelis-Menten process was significantly better (p 100 nm sized) nanomaterials display hindered lymphatic uptake as a result of reduced convection through the interstitium 18, monoclonal antibodies are considerably smaller (maximum diameter of 10 nm) and the data reported here suggest that this is not a significant limitation to lymphatic access. In order to confirm the validity of the bioavailability calculations for trastuzumab in rats, pharmacokinetic linearity was examined at dosage levels that resulted in plasma concentrations 14

ACS Paragon Plus Environment

Page 15 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

across the range expected after SC administration. After IV administration, trastuzumab was found to display dose-linear pharmacokinetics within the range 0.02 to 2 mg/kg (Fig 2). In contrast, rituximab displays non-linear SC pharmacokinetics in rats within the range 1-10 mg/kg. 31

Rituximab bioavailability was shown to decrease as the dosage increased31. The authors

speculated that the non-linear pharmacokinetics was a result of antibodies binding to FcRn at the injection site which prevented degradation with lower doses, whereas FcRn binding could be saturated at higher dosages and hence more degradation was possible, contributing to a reduction in bioavailability. In contrast to the linear pharmacokinetics observed in healthy rats at doses of 0.02-2 mg/kg, trastuzumab clearance is non-linear in humans with HER-2 positive breast cancer and faster at IV doses lower than 100 mg (approximately 1.6 mg/kg,

32

). This is consistent with

reports that monoclonal antibodies are eliminated via saturable target mediated mechanisms in humans

33

. This is not, however, consistently observed for other monoclonal antibodies. For

instance, Abila

34

recently reported that a monoclonal antibody that targets myelin-associated

glycoprotein (GSK249320) displays linear IV pharmacokinetics within the range of 0.04 to 25 mg/kg in humans. In light of the linearity in trastuzumab pharmacokinetics after IV injection in rats, the SC bioavailability of trastuzumab in rats was confidently estimated to be 85.5%. This is moderately higher than the SC bioavailability of 1 mg/kg rituximab in rats (~ 70%) after injection into the foot, abdomen or back

35

. The bioavailability of these human monoclonal antibodies in rats is

relatively consistent with the bioavailability of monoclonal antibodies in humans which have been reported to be between 52-80%

30

. Although SC bioavailability data for the commercial

formulation of trastuzumab is not available, bioavailability is approximately 84-99% in humans when coinjected with hyaluronidase, where hyaluronidase disrupts the extracellular matrix at the injection site and increase the maximum injectable volume

36

. Further studies are planned to

investigate the influence of hyaluronidase on lymphatic absorption of trastuzumab from SC injection sites. The current study also sought to examine the lymphatic access of trastuzumab after intravenous dosing in rats. This is important since monoclonal antibodies are typically administered via the intravenous route, and until now, it was not known what proportion of an 15

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

intravenous dose reaches the lymphatic circulation. It is of interest to note that HER-2 positive breast cancers metastasize via the lymph and that efficient lymphatic transfer of trastuzumab may promote anticancer efficacy against lymph-metastasizing cells. Whilst the SC administration of trastuzumab is expected to provide high localized concentrations in lymph draining the injection site, this approach is not expected to target cancer cells that metastasize via lymphatic vessels distant from the injection site. The results of this work demonstrate that approximately 45% of an IV dose of trastuzumab is recovered in thoracic lymph over 30 h (Table 1). The placement of the thoracic lymph duct cannula in rats, however, precludes the collection of lymph from the upper half of the body and from some areas of the lower half of the body (which empty into the thoracic duct above the cannula or which drain into the right lymph duct). The recovery of 45% of the IV dose over 30 h therefore underestimates the actual proportion of an IV dose of trastuzumab that accesses the lymph, since the anterior lymph loop is not sampled. A population pharmacokinetic model was therefore developed using the plasma and lymph concentration time profiles from SC control and lymph cannulated groups, to predict the total proportion of an IV dose that accesses the lymph over 30 h and longer. The model assumed that the lymphatic transfer rate constants were the same in both the anterior and posterior lymph loops. The model predicts that trastuzumab distribution into the lymphatic system is extensive after both SC and IV dosing (Table 4). This is significant, since it suggests that intravenously administered monoclonal antibody is able to efficiently access the lymph throughout the body and therefore to reach lymph-resident diseases in multiple sites (which in this case would be metastasizing cancer cells). While the mechanism for the Michaelis-Menten kinetics for absorption from the SC injection site into blood is unknown, a solubility limitation of the drug before it reaches the central circulation could contribute to saturable absorption kinetics as described previously 37. The Michaelis-Menten constant (Km) was estimated to be much smaller than the administered dose which may have contributed to its large relative standard error (Table 2). Both the maximum rate of absorption (Vmax) and Km had considerable between animal variability. A potential limitation of the modeling analysis is the lack of an external validation using a new dataset that was not used for model development.

16

ACS Paragon Plus Environment

Page 17 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Large proteins were previously thought to evade extravasation and be retained in systemic circulation for prolonged periods of time. Recent work, however, has suggested that high molecular weight (> 30 kDa) PEGylated proteins and polymers that circulate for long periods of time ultimately extravasate across capillary beds and are reabsorbed via the lymph and returned to systemic circulation 23, 38. This was originally believed to be a function of improved interstitial convection of highly PEGylated macromolecules that efficiently diffuse from the site of extravasation, through interstitial water channels and access lymphatic capillaries. This work, however, shows that high molecular weight proteins and polymers display good lymphatic access irrespective of the presence of PEG. These results also suggest that the extensive lymphatic exposure of monoclonal antibodies is a result of the prolonged circulation time, but also suggests that the pharmacokinetic behavior of antibodies may be altered in patients with compromised lymphatic function. In summary, the results of this work have shown that the lymphatic system is an integral part of the absorption profile of trastuzumab from interstitial injection sites and is a conduit for the continued circulation of proteins with prolonged plasma exposure profiles in rats. A novel population pharmacokinetic model has been developed that simultaneously predicts concentrations of a monoclonal antibody in plasma and lymph compartments. The model performed well and enabled predictions of the time-course of drug in lymph to better understand the pharmacokinetic consequences of SC compared to IV dosing of trastuzumab in rats. However, these data should be extrapolated conservatively from rat to human due to potential additional complexities related to species differences.

Acknowledgements: AMD was supported by a Faculty of Pharmacy Postgraduate Research Scholarship. LMK was supported by an Australian National Health and Medical Research Council (NHMRC) Career Development fellowship (APP1022732) and by NHMRC project funding (APP1044802). JBB is the recipient of an Australian Research Council Discovery Early Career Research Award (DE120103084).

17

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 38

Supporting Information: Supporting information is included to show validation data for the ELISA and the plasma pharmacokinetic profile of trastuzumab in rats over two months. This information is available free of charge via the internet at http://pubs.acs.org/.

18

ACS Paragon Plus Environment

Page 19 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Tables: Table 1. Non-compartmental pharmacokinetic parameters for trastuzumab administered SC or IV to Sprague-Dawley rats at 2 mg/kg. Data represent average ± SD (n=4-7 rats). *p