U.S. patent application number 15/762913 was filed with the patent office on 2018-09-20 for microprojection arrays with enhanced skin penetrating properties and methods thereof.
The applicant listed for this patent is Vaxxas Pty Limited. Invention is credited to Robert William GODDARD, Mark Anthony Fernance KENDALL, Stefano MELIGA.
Application Number | 20180264244 15/762913 |
Document ID | / |
Family ID | 58422498 |
Filed Date | 2018-09-20 |
United States Patent
Application |
20180264244 |
Kind Code |
A1 |
MELIGA; Stefano ; et
al. |
September 20, 2018 |
MICROPROJECTION ARRAYS WITH ENHANCED SKIN PENETRATING PROPERTIES
AND METHODS THEREOF
Abstract
An apparatus for delivering an active ingredient into the skin
of an animal at a defined depth, the apparatus including: a
microprojection array including a plurality of microprojections
having a density of at least 2,000 projections per cm.sup.2; and an
applicator that drives the microprojection array towards the skin
in use so that the microprojection array impacts on the skin with a
mass-to-velocity ratio of between 0.0005 g/m/s and 0.1 g/m/s per
cm.sup.2.
Inventors: |
MELIGA; Stefano; (West End,
Queensland, AU) ; KENDALL; Mark Anthony Fernance;
(Chelmer, Queensland, AU) ; GODDARD; Robert William;
(Capalaba, Queensland, AU) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Vaxxas Pty Limited |
Sydney, New South Wales |
|
AU |
|
|
Family ID: |
58422498 |
Appl. No.: |
15/762913 |
Filed: |
September 28, 2016 |
PCT Filed: |
September 28, 2016 |
PCT NO: |
PCT/AU2016/050907 |
371 Date: |
March 23, 2018 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62233607 |
Sep 28, 2015 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
A61M 2210/04 20130101;
A61M 37/0015 20130101; A61M 2202/30 20130101; A61M 2037/0061
20130101; A61D 7/00 20130101; A61B 17/205 20130101; A61M 2037/0023
20130101; A61M 2207/00 20130101; A61M 2037/0046 20130101 |
International
Class: |
A61M 37/00 20060101
A61M037/00; A61B 17/20 20060101 A61B017/20 |
Claims
1. An apparatus for delivering an active ingredient into the skin
of an animal at a defined depth, the apparatus including: a) a
microprojection array including a plurality of microproj ections
having a density of at least 2,000 projections per cm.sup.2; and,
b) an applicator that drives the microprojection array towards the
skin in use so that the microprojection array impacts on the skin
with a mass-to-velocity ratio of between 0.0005 g/m/s and 0.1 g/m/s
per cm.sup.2.
2. The apparatus of claim 1, wherein the microprojection array
impacts on the skin with a mass-to-velocity ratio of at least one
of: a) less than 0.05 g/m/s; b) less than 0.005 g/m/s; and, c)
between 0.033 g/m/s and 0.0008 g/m/s.
3. The apparatus of claim 1, wherein the microprojection array
impacts the skin with a mass between at least one of: a) 0.001 g
and 5g; b) 0.005 g and 2 g; and, c) 0.02 g and 0.5 g.
4. The apparatus of claim 1, wherein the microprojection array
impacts the skin at velocities between: a) 5 m/s and 50 m/s; b) 10
m/s and 30 m/s; and, c) 15 m/s and 25 m/s.
5. The apparatus of claim 1, wherein the microprojection array has
an area between at least one of: a) 16 mm.sup.2 and 400 mm.sup.2;
b) 36 mm.sup.2 and 225 mm.sup.2; and, c) 64 mm.sup.2 and 100
mm.sup.2.
6. The apparatus of claim 1, wherein the microprojection array has
a microprojection density between 5,000 and 20,000 projections per
cm.sup.2.
7. The apparatus of claim 1, wherein the microprojections are at
least one of: a) solid; b) non-porous; and, c) non-hollow.
8. The apparatus of claim 1, wherein the microprojections are at
least one of: a) tapered; b) substantially conical; c)
substantially flattened; d) hexagonal; and, e) octagonal.
9. The apparatus of claim 1, wherein the microprojections have a
length of at least one of: a) more than 100 .mu.m; b) more than 200
.mu.m; c) less than 1000 .mu.m; d) less than 5000 .mu.m; and, e)
between 200 .mu.m and 300 .mu.m.
10. The apparatus of claim 1, wherein the microprojections include:
a) a base having a width of about 5 .mu.m to about 50 .mu.m; and,
b) a tip having a width of about 0.5 .mu.m to about 2 .mu.m.
11. The apparatus of claim 1, wherein the applicator includes a
driver that drives the microprojection array towards the skin and
wherein the microprojection array is releasably mounted to the
driver so that the microprojection array is released from the
driver prior to the microprojections contacting the skin.
12. The apparatus of claim 11, wherein the driver abuts against a
stop to thereby release the microprojection array.
13. The apparatus of claim 12, wherein the stop includes an annular
shoulder.
14. The apparatus of claim 12, wherein the applicator includes: a)
a housing containing the driver; and, b) a substantially tubular
spacer that in use is positioned with an open end in contact with a
surface of the skin to thereby space the housing from the skin, the
stop being provided proximate the open end of the spacer.
15. The apparatus of claim 14, wherein the driver is urged from a
retracted to an extended position using a biasing mechanism, and
wherein the biasing mechanism and engagement between the driver and
housing define a driver velocity in use.
16. The apparatus of claim 15, wherein the driver is a piston.
17. The apparatus of claim 15, wherein the biasing mechanism
includes at least one of: a) a spring; and, b) a pneumatic
actuator.
18. The apparatus of claim 15, wherein the engagement is frictional
engagement between a piston and piston chamber within the
housing.
19. The apparatus of claim 1, wherein the microprojection array
impacts on the skin with a mass-to-velocity ratio sufficiently high
to effect at least one of: a) fracture the skin; b) concentrate
mechanical stress in superficial layers of the skin; c) invoke
strain-rate dependent skin stiffening; d) cause consistent
penetration independent of variations in subcutaneous properties of
the skin; e) dissipate inertia so as to avoid mechanical stress on
body parts underlying the skin; and, f) cause a controlled amount
of mechanical stress for immune-enhancing inflammation.
20. The apparatus of claim 1, wherein at least tips of the
microprojections are coated.
21. The apparatus of claim 1, wherein the active ingredient is one
or more vaccine antigens.
22. A method of determining the design of a microprojection array
and the velocity for delivering the microprojection array to a
predetermined range of skin depth comprising calculating the
microprojection array density, microprojection array area,
microprojection array mass and microprojection velocity to mass
ratio to deliver the microprojection array to the predetermined
depth range.
Description
BACKGROUND OF THE INVENTION
[0001] The invention is generally directed to devices and methods
for intradermal delivery of active agents into the skin, more
particularly the invention is directed to devices and methods for
improving the immunogenicity of vaccine preparations by intradermal
delivery of the vaccine via a microprojection array in which the
parameters for delivery of the active agents have been developed to
achieve appropriate depth penetration and efficient delivery of the
active agent.
DESCRIPTION OF THE PRIOR ART
[0002] Next-generation healthcare increasingly relies on
minimally-invasive biomedical devices capable of negotiating skin
mechanical properties to mediate intracutaneous and transcutaneous
tasks like administering therapeutics, extracting diagnostic
biomarkers and performing surgical procedures. For instance,
epidermal and dermal targeted delivery of vaccines is a promising
candidate for increasing global vaccine coverage, due to ease of
access as well as unique immunological properties of the skin.
Passive permeation of the antigen is impractical due to the large
molecular size of most antigens, therefore, the payload is actively
transported to the viable-cell strata by mechanically breaching
through the skin's outer barriers. This transport is typically
achieved by either: 1) high-pressure jet injectors that fire the
payload in liquid or powder form (microparticles) or 2) penetrator
tips that deposit payload through a channel in the skin (e.g.
intradermal syringe needles and hollow microneedles), or that embed
the payload in a matrix/coating that dissolves in the skin (e.g.
dissolvable/coated microneedle and microprojection arrays). Some
studies have reported improved immune responses compared to
standard syringe injection. In addition, the mechanisms underlying
the low-dose efficacy or increased potency are not yet fully
understood thereby limiting the potential of cutaneous
vaccination.
[0003] Precise penetration to the targeted depth for vaccine uptake
by site-specific cells is of fundamental importance and relies on
negotiating the unique elastic and failure properties of the skin
which is a multilayer composite `material`. Despite the many
published mechanical characterization and underlying linear and
non-linear elastic models, there is a paucity of investigations
focusing on skin elastic and failure behavior in mechanical
conditions relevant for epidermal and dermal delivery of active
agents including vaccines. There are reasons beyond the skin's
intrinsic structural complexity, and inter-species (e.g. mouse vs
human), inter-individual (ethnicity, gender) and intra-individual
(age, body site) variabilities for this failure. Firstly, the
persistent assumption of skin homogeneity and isotropicity resulted
in different elastic moduli depending on the loading mode.
Secondly, the Young's moduli extrapolated from indentations showed
a marked inverse dependence with the probe diameter. Thirdly,
although the extensive literature on skin viscoelasticity provides
solid evidence of the rate-dependence of skin elasticity, there
appear to be no published out-of-plane tests where the load was
applied at velocities >1 m s.sup.-1 or strain rates >1
.mu.s.sup.-1.
[0004] While underlying linear-elastic and hyperelastic
descriptions are corroborated by empirical data, skin also lacks
established constitutive models of failure. Skin penetration by
individual needles has typically been described using either: 1)
stress-based failure criteria extend the traditional yield criteria
such that the skin fails when the stress (typically the von Mises
component) exceeds a threshold strength; as such, this framework
does not account for the irrecoverable energy dissipated into
material damage and, thus, for example, cannot be used to predict
the depth achieved by penetrators fired at a given velocity; or 2)
energy-based fracture propagation extends the concept of fracture
toughness to ductile materials, i.e. an energy per unit area
representing the cost to create crack interfaces. This model,
though, does not specify if an initial notch forms at all (failure
initiation), how the crack propagates (e.g. direction and speed),
and what fraction of the penetrator energy is utilized in the
fracture (as opposed of being elastically stored or dissipated in
viscous or plastic phenomena). Rather, the prediction of skin
penetration requires a complete description of the spatial
stress-strain distributions to detect the instant and coordinates
of failure initiation, and the energy repartition among various
reversible and irreversible phenomena.
[0005] Skin out-of-plane mechanical properties of skin at the
microscale are typically measured ex vivo using indentation (e.g.
AFM) at velocities up to .about.100 .mu.m s.sup.-1; however,
vaccines are delivered in vivo across the skin's superficial
barriers using penetrators applied (by hand or impact applicators)
at velocities >>mm s.sup.-1; strain-rate effects and
subcutaneous layers play an important mechanical role during skin
penetration.
[0006] The limited understanding of skin elastic response to high
strain rates, mechanisms of failure and fracture, and interaction
with multiple penetrators have prevented the rational design of
epidermal and dermal targeted vaccination devices. Some
microprojection arrays are silicon chips containing, on one side,
thousands of densely-arranged (>>1,000 cm.sup.-2)
microprojections, i.e. solid cone-like structures measuring
.about.100 .mu.m in length. Notably, application of vaccine-coated
microprojection arrays to mouse skin elicited immune response using
.about.1/100 of the dose required by intramuscular injection. The
precise and consistent targeting of specific strata within the skin
is important and achieved by applying the array against the skin at
controlled velocities (.about.1 m s.sup.-1). Therefore, there is a
need for in-depth understanding of the skin mechanical interaction
with microneedles/microprojections which would allow the tailoring
of an array design and application conditions to achieve customized
antigen placement and to increase the targeting consistency across
patients and minimize the penetration energy of the array while
controlling skin inflammation, tolerability and acceptability.
[0007] The reference in this specification to any prior publication
(or information derived from it), or to any matter which is known,
is not, and should not be taken as an acknowledgment or admission
or any form of suggestion that the prior publication (or
information derived from it) or known matter forms part of the
common general knowledge in the field of endeavour to which this
specification relates.
SUMMARY OF THE PRESENT INVENTION
[0008] In a broad form the present invention seeks to provide an
apparatus for delivering an active ingredient into the skin of an
animal at a defined depth, the apparatus including: [0009] a) a
microprojection array including a plurality of microprojections
having a density of at least 2,000 projections per cm.sup.2; and,
[0010] b) an applicator that drives the microprojection array
towards the skin in use so that the microprojection array impacts
on the skin with a mass-to-velocity ratio of between 0.0005 g/m/s
and 0.1 g/m/s per cm.sup.2.
[0011] Typically the microprojection array impacts on the skin with
a mass-to-velocity ratio of at least one of: [0012] a) less than
0.05 g/m/s; [0013] b) less than 0.005 g/m/s; and, [0014] c) between
0.033 g/m/s and 0.0008 g/m/s.
[0015] Typically the microprojection array impacts the skin with a
mass between at least one of: [0016] a) 0.001 g and 5g; [0017] b)
0.005 g and 2 g; and, [0018] c) 0.02 g and 0.5 g.
[0019] Typically the microprojection array impacts the skin at
velocities between: [0020] a) 5 m/s and 50 m/s; [0021] b) 10 m/s g
and 30 m/s; and, [0022] c) 15 m/s and 25 m/s.
[0023] Typically the microprojection array has an area between at
least one of: [0024] a) 16 mm.sup.2 and 400 mm.sup.2; [0025] b) 36
mm.sup.2 and 225 mm.sup.2; and, [0026] c) 64 mm.sup.2 and 100
mm.sup.2.
[0027] Typically the microprojection array has a microprojection
density between 5,000 and 20,000 projections per cm.sup.2.
[0028] Typically the microprojections are at least one of: [0029]
a) solid; [0030] b) non-porous; and, [0031] c) non-hollow.
[0032] Typically the microprojections are at least one of: [0033]
a) tapered; [0034] b) substantially conical; [0035] c)
substantially flattened; [0036] d) hexagonal; and, [0037] e)
octagonal.
[0038] Typically the microprojections have a length of at least one
of: [0039] a) more than 100 .mu.m; [0040] b) more than 200 .mu.m;
[0041] c) less than 1000 .mu.m; [0042] d) less than 5000 .mu.m;
and, [0043] e) between 200 .mu.m and 300 .mu.m.
[0044] Typically the microprojections include: [0045] a) a base
having a width of about 5 .mu.m to about 50 .mu.m; and, [0046] b) a
tip having a width of about 0.5 .mu.m to about 2 .mu.m.
[0047] Typically the applicator includes a driver that drives the
microprojection array towards the skin and wherein the
microprojection array is releasably mounted to the driver so that
the microprojection array is released from the driver prior to the
microprojections contacting the skin.
[0048] Typically the driver abuts against a stop to thereby release
the microprojection array.
[0049] Typically the stop includes an annular shoulder.
[0050] Typically the applicator includes: [0051] a) a housing
containing the driver; and, [0052] b) a substantially tubular
spacer that in use is positioned with an open end in contact with a
surface of the skin to thereby space the housing from the skin, the
stop being provided proximate the open end of the spacer.
[0053] Typically the driver is urged from a retracted to an
extended position using a biasing mechanism, and wherein the
biasing mechanism and engagement between the driver and housing
define a driver velocity in use.
[0054] Typically the driver is a piston.
[0055] Typically the biasing mechanism includes at least one of:
[0056] a) a spring; and, [0057] b) a pneumatic actuator.
[0058] Typically the engagement is frictional engagement between a
piston and piston chamber within the housing.
[0059] Typically the microprojection array impacts on the skin with
a mass-to-velocity ratio sufficiently high to effect at least one
of: [0060] a) fracture the skin; [0061] b) concentrate mechanical
stress in superficial layers of the skin; [0062] c) invoke
strain-rate dependent skin stiffening; [0063] d) cause consistent
penetration independent of variations in subcutaneous properties of
the skin; [0064] e) dissipate inertia so as to avoid mechanical
stress on body parts underlying the skin; and, [0065] f) cause a
controlled amount of mechanical stress for immune-enhancing
inflammation.
[0066] Typically at least tips of the microprojections are
coated.
[0067] Typically the active ingredient is one or more vaccine
antigens.
[0068] In another broad form the present invention seeks to provide
a method of determining the design of a microprojection array and
the velocity for delivering the microprojection array to a
predetermined range of skin depth comprising calculating the
microprojection array density, microprojection array area,
microprojection array mass and microprojection velocity to mass
ratio to deliver the microprojection array to the predetermined
depth range.
BRIEF DESCRIPTION OF THE DRAWINGS
[0069] An example of the present invention will now be described
with reference to the accompanying drawings, in which:
[0070] FIG. 1A is a schematic drawing of various modes of
penetrating the skin;
[0071] FIG. 1B is a schematic diagram of design specifications for
individual and arrays of penetrators (e.g.
microneedles/microprojections);
[0072] FIG. 1C is a schematic drawing of a mouse ear section and
skin layer thickness;
[0073] FIGS. 2A-2H are graphical representations of the
hyperelastic properties for the skin layers (SC=stratum corneum,
VE=viable epidermis, dermis) of mouse ear as a function of
indentation velocity (or peak logarithmic strain rate): FIG. 2A is
a plot of Young's moduli versus the strain rate and velocity for
the stratum corneum; FIG. 2B is a plot of Young's moduli versus the
strain rate and velocity for the viable epidermis; FIG. 2C is a
plot of Young's moduli versus the strain rate and velocity for the
dermis; FIG. 2D is a plot of the stretch exponent versus the strain
rate and velocity for the stratum corneum; FIG. 2E is a plot of the
stretch exponent versus the strain rate and velocity for the viable
epidermis; FIG. 2F is a plot of the stretch exponent versus the
strain rate and velocity for the dermis; and FIGS. 2G and 2H are
bar diagrams of Young's modulus and stretch exponent extrapolated
for a probe measuring 1 .mu.m in diameter indenting the skin layers
in the velocity range 0.3-10 m s.sup.-1 (or strain-rate range
0.3-10 .mu.s.sup.-1);
[0074] FIGS. 3A-3D are graphical representations of skin stress and
energy transfers during the penetration by arrays with different
densities applied with equal energy per projection (.about.1/2*35
g* (2 m s.sup.-1).sup.2/3000): FIG. 3A shows VM stress in the skin
during the penetration of arrays characterized by projection
densities of .about.0 proj cm.sup.-2 (infinitely-spaced
projections); FIG. 3B shows VM stress in the skin during the
penetration of arrays characterized by projection densities of
5,000 proj cm.sup.-2; FIG. 3C shows VM stress in the skin during
the penetration of arrays characterized by projection densities of
10,000 proj cm.sup.-2; and FIG. 3D shows VM stress in the skin
during the penetration of arrays characterized by projection
densities of 20,400 proj cm.sup.-2;
[0075] FIG. 3E is a diagram of symmetric FE geometry and mesh used
to simulate the penetration of arrays with .gtoreq.5,000 proj
cm.sup.-2, in which the inset shows the fundamental skin unit
simulated (red) and the planes of symmetry (dashed lines) on a
top-view schematics of the array;
[0076] FIG. 3F is a diagram of the fraction of application energy
(mean.+-.range) utilized during the penetration of the .about.0
proj cm.sup.-2 array into mouse ear when the tip reaches the bottom
of the (ventral) dermis as calculated using FEM; the range
represent the variation between successive time points (.+-.0.5
.mu.s);
[0077] FIG. 3G is a diagram of the energy utilized as function of
projection density/spacing as calculated using FEM;
[0078] FIG. 3H is a diagram of the energy fraction (mean.+-.sd)
transferred (utilized) to the ear as measured experimentally from
the difference between the impact energies transmitted across the
backing and the ear+ backing to an underlying force sensor.
FEM=finite-element modeling, exp=experiment, inf=infinite;
[0079] FIG. 4A is a schematic of a model used to simulate
projection array penetration into skin backed by soft tissue; mouse
ear layers were modeled using an axisymmetric FE geometry with a
symmetric boundary; the soft backing material was modeled using a
parallel spring-damper-mass lumped element;
[0080] FIG. 4B is a schematic of the penetration depth resulting
from standard treatment, i.e. firing the array with an energy of
.about.13 mJ (.about.35 g piston at .about.0.85 m s.sup.-1) on a
PDMS-backed ear (left), and .about.1.3 mJ (.about.5 g at
.about.0.75 m s.sup.-1) on ear alone; a +15% correction factor was
considered to account for the tissue shrinking due to histology
treatment; the mean.+-.se (n=4) is represented for the experimental
groups, whereas the error-bars of the model group represent the
uncertainty due to FE parameterization as in FIG. 4C;
[0081] FIG. 4C is a schematic of the sensitivity of the numerical
solution to model parameterization when the standard treatment
condition (35 g, .about.0.85 m s.sup.-1) is used; the bars indicate
the penetration depth resulting varying the model parameters; the
direction of the depth change when the specific model parameter
increases is indicated by the black curves;
[0082] FIG. 4D is a schematic of the numerical and experimental
variations of penetration depth; the depth range originating from
the skin variability has been represented using the deviation of
the penetration measurements across biological repeats, and
compared to the widest numerical variability deriving from skin
properties, i.e. skin stiffness;
[0083] FIGS. 5A-5E are plots of numerical (FEM result.+-.FE error)
and experimental (exp mean.+-.se) penetration depths as a function
of varying application conditions and array designs: FIG. 5A is a
plot of penetration depth versus application velocity; FIG. 5B is a
plot of penetration depth versus piston mass; FIG. 5C is a plot of
penetration depth versus array size; FIG. 5D is a plot of
penetration depth versus projection density; FIG. 5E is a plot of
penetration depth versus energy/projection, in which the
significant Spearman correlation (p<0.0001) found between
penetration depth pd and application energy per projection U was
modeled with power laws pd=A U.sup.B, i.e. straight (dotted) lines
in Log-Log scale, horizontal error-bars represent the standard
deviation of the measurement of application velocity and number of
microprojections on the array following wafer dicing, and vertical
error-bars were obtained as in FIG. 4B;
[0084] FIG. 5F is a schematic representation of applicator function
and main parameters;
[0085] FIG. 6A is a plot of penetration depth versus piston mass
under conditions where 1) constant spring load; 2) constant energy
and 3) constant velocity;
[0086] FIG. 6B is a plot of penetration depth versus array size
under conditions where 1) constant projection density and 2)
constant projection number;
[0087] FIG. 6C is a plot of penetration depth versus
energy/projection comparing experimental and FEM for velocity sets,
mass sets, array-size-sets and density sets, in which the
penetration depth escapes the Log-Log linear dependence with
application energy per projection for very low piston masses and
large array sizes; error-bars were omitted for clarity;
[0088] FIG. 6D is a plot of the percentage of application energy
transferred to the skin versus piston mass;
[0089] FIG. 6E is a plot of the percentage of application energy
transferred to the skin versus array size;
[0090] FIG. 7 is a flowchart of skin failure model, in which the
clockwise flow describes the approach used in the present
application; whereas the anti-clockwise flow (in grey) shows the
simplified implementation used in previous work (VM=von Mises);
[0091] FIG. 8 is a plot of force measured by piezoelectric load
cell placed under the PDMS following .about.2 m s.sup.-1 impact of
a microprojection array on the PDMS-backed skin (`PDMS+ear`) and
flat patch on PDMS backing only `(PDMS`);
[0092] FIG. 9 is a plot of force versus compression displacement
for impact tests that were performed on carbon tab-topped PDMS
firing a 5mm-diameter flat-ended piston; piston mass and impact
velocity (and relative theoretical peak engineering strain rate)
are indicated; the green datasets have .about.constant kinetic
energy; the vertical error-bars indicate the sd of the measurements
across the different PDMS samples; the horizontal error-bars of the
impact tests show the uncertainty (sd over the different PDMS
samples) of the compression displacement measures using the
high-speed camera; and the full and dashed lines show the stiffness
curves selected after PDMS model validation for the brass and
plastic pistons, respectively; and,
[0093] FIG. 10 is a schematic diagram of model geometry of uncoated
(full) and coated (dashed) microprojection.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0094] In-depth understanding of skin elastic and rupture behaviors
is important for next-generation biomedical devices because it
enables targeted delivery of vaccines, as well as
minimally-invasive extraction of diagnostic biomarkers and
robotic/haptic surgery. Penetration of the skin's superficial
barriers and precise targeting of strata rich in antigen-presenting
cells is critical to elicit potent low-dose immunogenicity.
However, the paucity of relevant skin mechanical characterization
and lack of established fracture models has limited the rational
design of cutaneous devices. The present invention exploits
experimental and numerical studies of skin mechanics during dynamic
interaction with individual and arrays of microscopic penetrators
to provide improved methods and devices for delivering active
agents into the skin. Micro-indentation of individual strata
reveals that the hyperelastic moduli are dramatically
rate-dependent, and allows extrapolation of the stiffness
properties at velocity regimes (>mm s.sup.-1) relevant for
dynamically-actuated cutaneous devices. These are used to
parameterize a layered finite-element (FE) representation of skin
that includes a novel implementation of ductile failure. Iterative
refinement to match empirical penetration assays yields
characteristic fracture energies (.about.10 pJ
.mu.m.sup.-2)significantly lower than previously reported
(>>100 pJ .mu.m.sup.-2) The resulting FE simulations
satisfactorily predict the penetration depth of microprojection
arrays across a diverse range of designs and application
conditions, and shows limited sensitivity to the parameterization
choice. The knowledge and numerical tools developed provide
guidelines to rationally engineer skin penetrators. Specific array
design and application conditions can be developed to increase the
targeting consistency across patients and minimize the penetration
energy while controlling skin inflammation, tolerability and
acceptability.
[0095] Both experiments and theoretical models were used to develop
an understanding of the skin's mechanical properties relative to
the dynamic penetration of individual and multiple microscopic
penetrators. These properties are particularly relevant to the skin
treatment by microneedles/microprojections for vaccine delivery as
well as minimally-invasive extraction of diagnostic biomarkers.
Starting from micro-indentation experiments on mouse skin (FIG.
1C), the hyperelastic properties of the epidermal and dermal layers
at high strain-rates (>1 .mu.s.sup.-1) were derived. These were
utilized in conjunction with finite-element simulations to further
investigate the rate-dependent skin mechanical response to the
impact of individual and arrays of penetrator tips.
[0096] The complete model schematized in FIG. 4A was used to
simulate skin mechanical interaction with the microprojections in
the conditions used for mouse vaccination experiment (G. J. P.
Fernando, X. F. Chen, T. W. Prow, M. L. Crichton, E. J. Fairmaid,
M. S. Roberts, I. H. Frazer, L. E. Brown, M. A. F. Kendall, PLoS
One 2010, 5, c10266). Penetration was studied for varying array
designs and application parameters. For validation, the calculated
penetration depths were compared with experimental measurements
from histological sections of skin treated with dye-coated arrays
according to an established protocol (M. L. Crichton, A. Ansaldo,
X. F. Chen, T. W. Prow, G. J. P. Fernando, M. A. F. Kendall,
Biomaterials 2010, 31, 4562).
[0097] FIG. 4B shows the simulation and experimental results for a
4.times.4 mm.sup.2 array containing .about.3000 microprojections
spaced of L=70 .mu.m (i.e. .about.20 kproj cm.sup.-2) applied on
PDMS-backed skin at 0.85 m s.sup.-1 with the 35 g piston (i.e.
.about.13 mJ), the `standard treatment` condition. The resulting
penetration depth, 48 .mu.m from the model, is in good agreement
with the experimental measurement, 41.+-.2 .mu.m (mean.+-.se). This
simulation indicated that 6.2% of the energy is transferred to the
skin. The model was revised by removing the backing and applying
the array to the ear alone using (conservatively) .about.10% of the
energy (.about.5 g at .about.0.75 m s.sup.-1, i.e., .about.1.3 mJ).
FIG. 4B shows that this reduced-energy condition penetrates
un-backed skin to a depth comparable with the standard treatment on
backed skin, which further validates the skin and PDMS
parameterizations.
[0098] The sensitivity of the numerically-derived penetration depth
to the variation of the model parameters was assessed with a set of
limit analyses. In brief, the standard treatment simulation was
repeated assigning upper and lower boundary values to each
individual parameter, one at a time. The input-parameter intervals
are summarized in Table 1 and are representative of the range of FE
parameters, variation of skin properties as reported in the
literature and possible array design tolerances or modifications.
For simple reference to FIGS. 5A-5E, the top, respectively bottom,
row in Table 1 shows the condition resulting into shallower,
respectively deeper, penetration.
TABLE-US-00001 TABLE 1 Summary of parameter variation ranges used
to assess the sensitivity of the numerical solutions. st = standard
value. FE Skin Skin Array Array Array FE mass Skin Skin Skin epider
dermis Skin- Array proj proj tip proj tip mesh scaling elastic
Poisson's fracture thickn thickn proj proj Array diamet angle
diamet density [ps] moduli ratio strain [.mu.m] [.mu.m] friction
location coating [.mu.m] [deg] [.mu.m] -50% ~100 +70% 0.35 >20%
9 80 center 29 5 st.sup.a) +25%. ~50 st.sup.b) 0.45
.ltoreq.20%.sup.c) 20 60 0.4 mean.sup.d) yes.sup.e) 23 30 1 none
-50% 0.49 .sup. 0% 27 40 0.7 edge no 17 50 0.05 10
[0099] FIG. 4C shows that no significant difference resulted after
refining the mesh, which indicates that the mesh of choice is
appropriate. Among other skin characteristics, penetration depth
was most sensitive to strata stiffness (FIG. 4C); and,
interestingly, the resulting numerical depth range is in close
agreement with the measurement variation across biological repeats
(FIG. 4D). On the other hand, the experiments revealed a
significantly deeper penetration depth towards the edges of the
array, likely due to the larger force exerted by peripheral
microprojections. By scaling the microprojection momentum the
increasing penetration depth caused by projections located at
increasing distance from the array center could be reasonably
predicted (FIG. 4D).
[0100] The penetration resulting from different array application
conditions (FIGS. 5A-5B) and designs (FIGS. 5C-5D) was investigated
numerically and empirically by further applying the computational
and experimental methods. Increasing microprojection velocity
resulted in deeper penetration due to the larger energy.
Separately, lower piston masses (using the same application spring
load) resulted in slightly decreasing penetration, despite the
theoretically-constant application potential energy. In fact,
applicator characterization revealed lower than expected
application velocities for the lower masses (<35 g), possibly
due to a greater friction of the lighter plastic piston against the
applicator housing compared to the standard brass piston. The
simulations were run using the measured velocities, rather than the
theoretically-calculated ones. Decreasing the array size or the
microprojection density (constant array size) resulted in deeper
penetration mostly because the same application energy is shared
among fewer projections. The numerical prediction and the
experimental measurement were in reasonable agreement.
Specifically, the model appears to overestimate the depth
especially when the projections are widely spaced and approach the
deep dermis. This is possibly due to two reasons: 1) the deeper
penetration of the peripheral projections (FIG. 4D) might allow
contact between the SC and the base of the array, especially for
sparse arrays; and 2) the projection interacts with the cartilage,
which mechanical properties were not accurately established.
[0101] There is significant Spearman correlation (p<0.0001)
between the penetration depth pd and the application energy per
projection U (FIG. 5E). The power (1.30.+-.0.04)
U.sup.(0.38.+-.0.04) (mean.+-.se) fitted the experimental data
satisfactorily (R.sup.2=0.931). An analogous non-linear regression
for the numerical dataset yielded (1.43.+-.0.05)
U.sup.(0.44.+-.0.05) with similar goodness-of-fit (R.sup.2=0.932).
These curves pd=A U.sup.B appear as straight lines in Log-Log scale
(FIG. 5E) where A is the intercept, B is the slope, the depth pd is
measured in .mu.m and U in .mu.J. FIG. 5E also suggests that the
penetration depth of arrays with custom design and application
conditions can be simplistically predicted from the application
energy (per projection) using this empirical relationship.
[0102] The computational model was applied to investigate
alternative designs and application conditions and challenge the
trend of FIG. 5E. Interestingly, decreasing piston mass (FIG. 6A)
or increasing the array size (FIG. 6B) resulted in increased
penetration depth although the energy per projection was held
constant. These conditions, as well as 10 m s.sup.-1 applications
for masses below 0.2 g (FIG. 6A) markedly violated the Log-Log
linear relationship between depth and energy per projection (FIG.
6C). Specifically, the results indicate that isoenergetic
applications achieve a .about.2-fold deeper penetration using a
mass <0.05 g or spreading the microprojections over a 10-fold
larger area. Equivalently, the energy required to reach a
mid-dermal depth (.about.50 .mu.m) can be reduced by over 80% by
lowering the mass from 35 g to 0.05 g. Key for this `energy
sparing` phenomenon is the increasing application velocity required
to maintain a constant energy while decreasing the mass. In fact,
the simulations of velocities <3 m s.sup.-1 showed that skin
fracture starts after a large compression of the backing and
terminates after 0.5-1 ms. In contrast, the fracture process is
completed in .about.10 .mu.s at 10 m s.sup.-1, before the backing
has started to deform Likely, these different penetration regimes
arise because the projection motion competes with the transmission
of the deformation to the backing through the stress waves. Such
behavior suggests that an efficiency around 55% can be
theoretically achieved by reducing the moving mass down to the
array itself (.about.0.03 g). In addition, the energy transfer
efficiency linearly correlated with array size (FIG. 6E; Pearson's
r=0.966, p<0.0001, slope=(0.126.+-.0.013)% mm.sup.-2,
intercept=(5.78.+-.0.62)%). This is likely to be because
distributing the impact over a larger surface increases the overall
backing elastic force response, thus results in an effectively
stiffer substrate.
[0103] The results of FIGS. 6A-6E indicate that penetration depth
is not a unique function of the energy per projection. Rather, the
application energy required to target a specific depth can be
modulated by varying the velocity-to-mass ratio. This represents an
important degree of freedom to seek immunologically-beneficial
levels of inflammation (e.g. cell stress/death via mechanical
perturbation) without compromising treatment tolerability and
acceptability by the patient. On the other hand, high-velocity,
low-mass applications allow the microprojections to interact mainly
with the superficial layers (i.e. the skin). This effectively
reduces the dependence of penetration on the skin backing
properties, hence potentially improves the targeting consistency
across patients with different subcutaneous tissue composition
(e.g. different body-mass index).
[0104] The skin dynamic behavior is the main cause of such a
diverse mechanical response. Firstly, the heterogeneous layered
composition favored fracture in the early impact stages for large
application velocities. Specifically, the stress was effectively
retained at the surface due to the slow stress-wave propagation of
the deep strata (cartilage, PDMS, fat or muscle), comparatively
lower in stiffness. Secondly, the equivalent strain required to
initiate failure (i.e. meet the yield criterion) decreased with
increasing velocity because skin elasticity (i.e. the stress
response to a specific strain) has a steeper rate-dependent
increase compared to the yield strength. As a consequence,
penetration is more difficult in quasi-static conditions, as the
Young's modulus-to-yield strength ratio decreases below 1, due to
the resulting strata softness (compliance).
[0105] The resulting penetration model satisfactorily reproduced
the experimental behavior for a wide range of conditions, and
further proved robust to variations in parameterization. However,
the utilized elastic moduli were derived from indentations using
constant probe velocity, and are relative to the peak strain rates
at impact. Hypothetically, the resulting skin stress relaxation
should result in lower penetration depths that match the
experimental measurements more closely.
[0106] While significant differences in skin behavior are expected
if the dynamic regime is changed (e.g. from impact to quasi-static
or vibratory application), penetration of other microneedle array
designs (typically characterized by sparser, larger penetrators)
will likely follow the trends showed in FIGS. 5A-5E and FIGS.
6A-6E. This is justified by the low variation between the relative
energetic contributions (e.g. fracture, deformation and friction)
(FIG. 3G) and the approximately constant stress generated as tip
radius and spacing increase. As can be seen in FIGS. 4A-4D a
variety of parameters may affect the depth of penetration of
microprojections into the skin: skin stiffness, skin fracture
strain, epidermis thickness, dermis thickness, skin-microprojection
friction, distance of projections from the array center, amount of
coating on microprojection, microprojection tip angle,
microprojection shape, velocity of microprojection array into the
skin, mass of microprojection array, velocity to mass ratio of the
microprojection area, area of the microprojection array, density of
microprojection array, backing used behind skin target.
[0107] When administered to the skin the microprojection array may
have a velocity which is greater than about 5 m/s or about 6 m/s,
or about 7 m/s, or about 8 m/s, or about 9 m/s, or about 10 m/s, or
about 15 m/s, or about 20 m/s, or about 25 m/s, or about 30 m/s, or
about 40 m/s, or about 45 m/s, or about 50 m/s, or about 55 m/s.
When administered to the skin the microprojection array may have a
velocity which is about 5 m/s to about 50 m/s, or from about 5 m/s
to about 45 m/s, or from 5 m/s to about 40 m/s, or from about 5 m/s
to about 35 m/s, or from about 5 m/s to about 30 m/s, or from 5 m/s
to about 25 m/s, or from about 5 m/s to about 20 m/s, or from about
5 m/s to about 15 m/s, or from 5 m/s to about 10 m/s, or from about
10 m/s to about 50 m/s, or from about 10 m/s to about 45 m/s, or
from 10 m/s to about 40 m/s, or from about 10 m/s to about 35 m/s,
or from about 10 m/s to about 30 m/s, or from 10 m/s to about 25
m/s, or from about 10 m/s to about 20 m/s, or from about 10 m/s to
about 15 m/s, or from about 15 m/s to about 50 m/s, or from about
15 m/s to about 45 m/s, or from 15 m/s to about 40 m/s, or from
about 15 m/s to about 35 m/s, or from about 15 m/s to about 30 m/s,
or from 15 m/s to about 25 m/s, or from about 15 m/s to about 20
m/s, or from about 20 m/s to about 50 m/s, or from 20 m/s to about
45 m/s, or from about, or from 20 m/s to about 40 m/s, or from
about 20 m/s to about 35 m/s, or from about 20 m/s to about 30 m/s,
or from about 20 m/s to about 25 m/s, or from about 25 m/s to about
50 m/s, or from about 25 m/s to about 45 m/s, or from 25 m/s to
about 40 m/s, or from about 25 m/s to about 35 m/s, or from about
25 m/s to about 30 m/s, or from about 30 m/s to about 50 m/s, or
from about 30 m/s to about 45 m/s, or from about 30 m/s to about 40
m/s, or from about 30 m/s to about 35 m/s.
[0108] The microprojection arrays may have a mass of less than 1
gram, or less than 0.9 grams, or less than 0.8 grams, or less than
0.7 grams, or less than 0.6 grams, or less than 0.5 grams, or less
than 0.6 grams, or less than 0.5 grams, or less than 0.4 grams, or
less than 0.3 grams, or less than 0.2 grams, or less than 0.1
grams, or less than 0.05 grams, or less than 0.01 grams, or less
than 0.005 grams, or less than 0.001 grams. The microprojection
array may have a mass of from about 0.001 grams to about 5 grams of
about 0.001 grams to about 2 grams, or from about 0.001 grams to
about 1.5 grams, or from about 0.001 grams to about 1.0 grams, or
from about 0.001 grams to about 0.9 grams, or from about 0.001
grams to about 0.8 grams, or from about 0.001 grams to about 0.7
grams, or from about 0.001 grams to about 0.6 grams, or from about
0.001 grams to about 0.5 grams, or from about 0.001 grams to about
0.4 grams, or from about 0.001 grams to about 0.3 grams, or from
about 0.001 grams to about 0.2 grams, or from about 0.001 grams to
about 0.1 grams from about 0.01 grams to about 5 grams of about
0.01 grams to about 2 grams, or from about 0.01 grams to about 1.5
grams, or from about 0.01 grams to about 1.0 grams, or from about
0.01 grams to about 0.9 grams, or from about 0.01 grams to about
0.8 grams, or from about 0.01 grams to about 0.7 grams, or from
about 0.01 grams to about 0.6 grams, or from about 0.01 grams to
about 0.5 grams, or from about 0.01 grams to about 0.4 grams, or
from about 0.01 grams to about 0.3 grams, or from about 0.01 grams
to about 0.2 grams, or from about 0.01 grams to about 0.1 grams, or
from about 0.05 grams to about 5 grams of about 0.05 grams to about
2 grams, or from about 0.05 grams to about 1.5 grams, or from about
0.05 grams to about 1.0 grams, or from about 0.05 grams to about
0.9 grams, or from about 0.05 grams to about 0.8 grams, or from
about 0.05 grams to about 0.7 grams, or from about 0.05 grams to
about 0.6 grams, or from about 0.05 grams to about 0.5 grams, or
from about 0.05 grams to about 0.4 grams, or from about 0.05 grams
to about 0.3 grams, or from about 0.05 grams to about 0.2 grams, or
from about 0.05 grams to about 0.1 grams, or from about 0.1 grams
to about 1.0 grams, or from about 0.1 grams to about 5 grams, or
from about 0.1 grams to about 2 grams, or from about 0.1 grams to
about 0.9 grams, or from about 0.1 grams to about 0.8 grams, or
from about 0.1 grams to about 0.7 grams, or from about 0.1 grams to
about 0.6 grams, or from about 0.1 grams to about 0.5 grams, or
from about 0.1 grams to about 0.4 grams, or from about 0.1 grams to
about 0.3 grams, or from about 0.1 grams to about 0.2 grams.
[0109] The density of the microprojection on the microprojection
arrays may be about 2000 microprojections/cm.sup.2, or about 2500
microprojections/cm.sup.2, or about 3000 microprojections/cm.sup.2,
or about 3500 microprojections/cm.sup.2, or about 4000
microprojections/cm.sup.2, or about 4500 microprojections/cm.sup.2,
or about 5000 microprojections/cm.sup.2, or about 5500
microprojections/cm.sup.2, or about 6000 microprojections/cm.sup.2,
or about 6500 microprojections/cm.sup.2, or about 7000
microprojections/cm.sup.2, or about 7500 microprojections/cm.sup.2,
or about 8000 microprojections/cm.sup.2, or about 8500
microprojections/cm.sup.2, or about 9000 microprojections/cm.sup.2,
or about 9500 microprojections/cm.sup.2, or about 10000
microprojections/cm.sup.2, or about 11000
microprojections/cm.sup.2, or about 12000
microprojections/cm.sup.2, or about 13000
microprojections/cm.sup.2, or about 14000
microprojections/cm.sup.2, or about 15000
microprojections/cm.sup.2, or about 16000
microprojections/cm.sup.2, or about 17000
microprojections/cm.sup.2, or about 18000
microprojections/cm.sup.2, or about 19000
microprojections/cm.sup.2, or about 20000
microprojections/cm.sup.2. The density of the microprojection on
the microprojection arrays may be from about 2000 to about 20000
microprojections/cm.sup.2, or from about 2000 to about 15000
microprojections/cm.sup.2, or from about to about 10000
microprojections/cm.sup.2, or from about 2000 to about 9000
microprojections/cm.sup.2, or from about 2000 to about 8000
microprojections/cm.sup.2, or from about 2000 to about 7500
microprojections/cm.sup.2, or from about 2000 to about 7000
microprojections/cm.sup.2, or from about 2000 to about 6000
microprojections/cm.sup.2, or from about 2000 to about 5000
microprojections/cm.sup.2, or from about 2000 to about 4000
microprojections/cm.sup.2, or from about 3000 to about 20000
microprojections/cm.sup.2, or from about 3000 to about 15000
microprojections/cm.sup.2, or from about to about 10000
microprojections/cm.sup.2, or from about 3000 to about 9000
microprojections/cm.sup.2, or from about 3000 to about 8000
microprojections/cm.sup.2, or from about 3000 to about 7500
microprojections/cm.sup.2, or from about 3000 to about 7000
microprojections/cm.sup.2, or from about 3000 to about 6000
microprojections/cm.sup.2, or from about 3000 to about 5000
microprojections/cm.sup.2, or from about 3000 to about 4000
microprojections/cm.sup.2, or from about 4000 to about 20000
microprojections/cm.sup.2, or from about 4000 to about 15000
microprojections/cm.sup.2, or from about to about 10000
microprojections/cm.sup.2, or from about 4000 to about 9000
microprojections/cm.sup.2, or from about 4000 to about 8000
microprojections/cm.sup.2, or from about 4000 to about 7500
microprojections/cm.sup.2, or from about 4000 to about 7000
microprojections/cm.sup.2, or from about 4000 to about 6000
microprojections/cm.sup.2, or from about 4000 to about 5000
microprojections/cm.sup.2, or from about 5000 to about 20000
microprojections/cm.sup.2, or from about 5000 to about 15000
microprojections/cm.sup.2, or from about to about 10000
microprojections/cm.sup.2, or from about 5000 to about 9000
microprojections/cm.sup.2, or from about 5000 to about 8000
microprojections/cm.sup.2, or from about 5000 to about 7500
microprojections/cm.sup.2, or from about 5000 to about 7000
microprojections/cm.sup.2, or from about 5000 to about 6000
microprojections/cm.sup.2.
[0110] At least a portion of the projections may be coated.
Accordingly, one way of providing material for delivery to the
biological subject is by providing the material within the coating.
For example, the coating may include a vaccine for providing an
immunological response within the subject. The coating may be
provided in liquid or non-liquid forms, and may further include
ingredients other than the material to be delivered, such as an
adjuvant. Suitable coating formulations for use with projections
patches and methods of applying such coatings to the projections
are known, as described, for example, in WO/2010/042996 and
WO/2009/079712.
[0111] Although any type of coating may be used, particularly
advantageous embodiments of the microprojection arrays are provided
with at least a portion of the projections coated with a non-liquid
coating. In this regard, the term "non-liquid" coating will be
understood to include a coating that is applied in a liquid form
and allowed to dry or otherwise solidify to thereby form a
non-liquid coating.
[0112] The non-liquid coating may act as an additional
substantially solid layer of material which can be used to even
further adjust the geometry of the projections by optionally
causing the projections to have an effective profile of a different
shape to the underlying uncoated profile of the projections as
initially fabricated.
[0113] The microprojections of the array of the present invention
may be of any shape including cylindrical or conical. Other
geometries are also possible. The microprojection arrays may have
substrate with a plurality of microprojections protruding from the
substrate wherein the microprojections have a tapering hexagonal
shape and comprise a tip and a base wherein the base has two
substantially parallel sides with a slight draught angle of
approximately 1 to 20 degrees up to a transition point at which
point the angle increases to from about 20 degrees to about 70
degrees. A sharp blade-like tip will allow for enhanced penetration
of the microprojections into the skin while also generating an
enhanced localized cell death/bystander interaction in the skin
with a different profile than conical microprojection arrays. The
sharp blade-like tips of the microprojections may also increase the
level of danger signals and antigen to more live cells thereby
increasing the physical adjuvant effect of microprojections and
thereby improving immune responses. The tip of the microprojections
of the present invention may have a width of about 0.5 .mu.m, or
about 1.0 .mu.m, or about 1.5 .mu.m, or about 2.0 .mu.m, or about
2.5 .mu.m, or about 3.0 .mu.m, or about 3.5 .mu.m, or about 4.0
.mu.m, or about 4.5 .mu.m, or about 5.0 .mu.m. The tip of the
microprojections of the present invention may have a width of from
about 0.5 .mu.m to about 5.0 .mu.m, or from about 0.5 .mu.m to
about 4.5 .mu.m, or from about 0.5 .mu.m to about 4.0 .mu.m, or
from about 0.5 .mu.m to about 3.5 .mu.m, or from about 0.5 .mu.m to
about 3.0 .mu.m, or from about 0.5 .mu.m to about 2.5 .mu.m, or
from about 0.5 .mu.m to about 2.0 .mu.m, or from about 0.5 .mu.m to
about 1.5 .mu.m, or from about 0.5 .mu.m to about 1.0 .mu.m, or
from about 1.0 .mu.m to about 5.0 .mu.m, or from about 1.0 .mu.m to
about 4.5 .mu.m, or from about 1.0 .mu.m to about 4.0 .mu.m, or
from about 1.0 .mu.m to about 3.5 .mu.m, or from about 1.0 .mu.m to
about 3.0 .mu.m, or from about 1.0 .mu.m to about 2.5 .mu.m, or
from about 1.0 .mu.m to about 2.0 .mu.m, or from about 1.0 .mu.m to
about 1.5 .mu.m, or from about 1.5 .mu.m to about 5.0 .mu.m, or
from about 1.5 .mu.m to about 4.5 .mu.m, or from about 1.5 .mu.m to
about 4.0 .mu.m, or from about 1.5 .mu.m to about 3.5 .mu.m, or
from about 1.5 .mu.m to about 3.0 .mu.m, or from about 1.5 .mu.m to
about 2.5 .mu.m, or from about 1.5 .mu.m to about 2.0 .mu.m, or
from about 2.0 .mu.m to about 5.0 .mu.m, or from about 2.0 .mu.m to
about 4.5 .mu.m, or from about 2.0 .mu.m to about 4.0 .mu.m, or
from about 2.0 .mu.m to about 3.5 .mu.m, or from about 2.0 .mu.m to
about 3.0 .mu.m, or from about 2.0 .mu.m to about 2.5 .mu.m, or
from about 2.5 .mu.m to about 5.0 .mu.m, or from about 2.5 .mu.m to
about 4.5 .mu.m, or from about 2.5.mu.m to about 4.0 .mu.m, or from
about 2.5 .mu.m to about 3.5 .mu.m, or from about 2.5 .mu.m to
about 3.0 .mu.m.
[0114] The microprojection array when applied to the skin may have
a mass-to-velocity ratio of less than about 0.0005 g/m/s, or less
than about 0.001 g/m/s/or less than about 0.002 g/m/s, or less than
about 0.003 g/m/s, or less than about 0.004 g/m/s/or less than
about 0.005 g/m/s, or less than about 0.006 g/m/s, or less than
about 0.007 g/m/s/or less than about 0.008 g/m/s, or less than
about 0.009 g/m/s, or less than about 0.01 g/m/s/or less than about
0.02 g/m/s, or less than about 0.03/m/s, or less than about 0.04
g/m/s/or less than about 0.05 g/m/s, or less than about 0.06 g/m/s,
or less than about 0.07 g/m/s/or less than about 0.08 g/m/s, or
less than about 0.09/m/s, or less than about 0.10 g/m/s/or less
than about 0.20 g/m/s, or less than about 0.30 g/m/s, or less than
about 0.40 g/m/s/or less than about 0.50 g/m/s. The microprojection
array when applied to the skin may have a mass-to-velocity ratio of
about 0.0005 g/m/s to about 0.50 g/m/s, or from about 0.0005 g/m/s
to about 0.40 g/m/s, or from about 0.0005 g/m/s to about 0.30
g/m/s, or from about 0.0005 g/m/s to about 0.20 g/m/s, or from
about 0.0005 g/m/s to about 0.10 g/m/s, or from about 0.0005 g/m/s
to about 0.009 g/m/s, or from of about 0.0005 g/m/s to about 0.008
g/m/s, or from about 0.0005 g/m/s to about 0.007 g/m/s, or from
about 0.0005 g/m/s to about 0.006 g/m/s, or from about of about
0.0005 g/m/s to about 0.005 g/m/s, or from about 0.0005 g/m/s to
about 0.004 g/m/s, or from about 0.0005 g/m/s to about 0.003 g/m/s,
or from about of about 0.0005 g/m/s to about 0.002 g/m/s, or from
about 0.0005 g/m/s to about 0.001 g/m/s, or from about 0.001 g/m/s
to about 0.50 g/m/s, or from about 0.001 g/m/s to about 0.40 g/m/s,
or from about 0.001 g/m/s to about 0.30 g/m/s, or from about 0.001
g/m/s to about 0.20 g/m/s, or from about 0.001 g/m/s to about 0.10
g/m/s, or from about 0.001 g/m/s to about 0.009 g/m/s, or from of
about 0.001 g/m/s to about 0.008 g/m/s, or from about 0.001 g/m/s
to about 0.007 g/m/s, or from about 0.001 g/m/s to about 0.006
g/m/s, or from about of about 0.001 g/m/s to about 0.005 g/m/s, or
from about 0.001 g/m/s to about 0.004 g/m/s, or from about 0.001
g/m/s to about 0.003 g/m/s, or from about of about 0.001 g/m/s to
about 0.002 g/m/s, or from about 0.005 g/m/s to about 0.50 g/m/s,
or from about 0.005 g/m/s to about 0.40 g/m/s, or from about 0.005
g/m/s to about 0.30 g/m/s, or from about 0.005 g/m/s to about 0.20
g/m/s, or from about 0.005 g/m/s to about 0.10 g/m/s, or from about
0.005 g/m/s to about 0.009 g/m/s, or from of about 0.005 g/m/s to
about 0.008 g/m/s, or from about 0.005 g/m/s to about 0.007 g/m/s,
or from about 0.005 g/m/s to about 0.006 g/m/s, or from about 0.033
g/m/s to about 0.0008 g/m/s.
[0115] The area of the microprojection arrays in area may be
between about 10 mm.sup.2 to about 1000 mm.sup.2, or from about 10
mm.sup.2 to about 900 mm.sup.2, or from about 10 mm.sup.2 to about
800 mm.sup.2, or from about 10 mm.sup.2 to about 700 mm.sup.2, or
from about 10 mm.sup.2 to about 600 mm.sup.2, or from about 10
mm.sup.2 to about 600 mm.sup.2, or from about 10 mm.sup.2 to about
500 mm.sup.2, or from about 10 mm.sup.2 to about 400 mm.sup.2, or
from about 10 mm.sup.2 to about 300 mm.sup.2, or from about 10
mm.sup.2 to about 200 mm.sup.2, or from about 10 mm.sup.2 to about
100 mm.sup.2, or from about 10 mm.sup.2 to about 90 mm.sup.2, or
from about 10 mm.sup.2 to about 80 mm.sup.2, or from about 10
mm.sup.2 to about 70 mm.sup.2, or from about 10 mm.sup.2 to about
60 mm.sup.2, or from about 10 mm.sup.2 to about 50 mm.sup.2, or
from about 10 mm.sup.2 to about 40 mm.sup.2, or from about 10
mm.sup.2 to about 30 mm.sup.2, or from about 10 mm.sup.2 to about
20 mm.sup.2, or from about 20 mm.sup.2 to about 1000 mm.sup.2, or
from about 20 mm.sup.2 to about 900 mm.sup.2, or from about 20
mm.sup.2 to about 800 mm.sup.2, or from about 20 mm.sup.2 to about
700 mm.sup.2, or from about 10 mm.sup.2 to about 600 mm.sup.2, or
from about 20 mm.sup.2 to about 500 mm.sup.2, or from about 20
mm.sup.2 to about 400 mm.sup.2, or from about 20 mm.sup.2 to about
300 mm.sup.2, or from about 20 mm.sup.2 to about 200 mm.sup.2, or
from about 20 mm.sup.2 to about 100 mm.sup.2, or from about 20
mm.sup.2 to about 90 mm.sup.2, or from about 20 mm.sup.2 to about
80 mm.sup.2, or from about 20 mm.sup.2 to about 70 mm.sup.2, or
from about 20 mm.sup.2 to about 60 mm.sup.2, or from about 20
mm.sup.2 to about 50 mm.sup.2, or from about 20 mm.sup.2 to about
40 mm.sup.2, or from about 20 mm.sup.2 to about 30 mm.sup.2, or
from about 30mm.sup.2 to about 1000 mm.sup.2, or from about 30
mm.sup.2 to about 900 mm.sup.2, or from about 30 mm.sup.2 to about
800 mm.sup.2, or from about 30 mm.sup.2 to about 700 mm.sup.2, or
from about 10 mm.sup.2 to about 600 mm.sup.2, or from about 30
mm.sup.2 to about 500 mm.sup.2, or from about 30 mm.sup.2 to about
400 mm.sup.2, or from about 30 mm.sup.2 to about 300 mm.sup.2, or
from about 30 mm.sup.2 to about 200 mm.sup.2, or from about 30
mm.sup.2 to about 100 mm.sup.2, or from about 30 mm.sup.2 to about
90 mm.sup.2, or from about 30 mm.sup.2 to about 80 mm.sup.2, or
from about 30 mm.sup.2 to about 70 mm.sup.2, or from about 30
mm.sup.2 to about 60 mm.sup.2, or from about 30 mm.sup.2 to about
50 mm.sup.2, or from about 30 mm.sup.2 to about 40 mm.sup.2, or
from about 40 mm.sup.2 to about 1000 mm.sup.2, or from about 40
mm.sup.2 to about 900 mm.sup.2, or from about 40 mm.sup.2 to about
800 mm.sup.2, or from about 40 mm.sup.2 to about 700 mm.sup.2, or
from about 10 mm.sup.2 to about 600 mm.sup.2, or from about 40
mm.sup.2 to about 500 mm.sup.2, or from about 40mm.sup.2 to about
400 mm.sup.2, or from about 40 mm.sup.2 to about 400 mm.sup.2, or
from about 40 mm.sup.2 to about 200 mm.sup.2, or from about 40
mm.sup.2 to about 100 mm.sup.2, or from about 40 mm.sup.2 to about
90 mm.sup.2, or from about 40 mm.sup.2 to about 80 mm.sup.2, or
from about 40 mm.sup.2 to about 70 mm.sup.2, or from about 40
mm.sup.2 to about 60 mm.sup.2, or from about 40 mm.sup.2 to about
50 mm.sup.2, or from about 50 mm.sup.2 to about 1000 mm.sup.2, or
from about 50 mm.sup.2 to about 900 mm.sup.2, or from about 50
mm.sup.2 to about 800 mm.sup.2, or from about 50 mm.sup.2 to about
700 mm.sup.2, or from about 10 mm.sup.2 to about 600 mm.sup.2, or
from about 50 mm.sup.2 to about 500 mm.sup.2, or from about 50
mm.sup.2 to about 400 mm.sup.2, or from about 50 mm.sup.2 to about
300 mm.sup.2, or from about 50 mm.sup.2 to about 200 mm.sup.2, or
from about 50 mm.sup.2 to about 100 mm.sup.2, or from about 50
mm.sup.2 to about 90 mm.sup.2, or from about 50 mm.sup.2 to about
80 mm.sup.2, or from about 50 mm.sup.2 to about 70 mm.sup.2, or
from about 50 mm.sup.2 to about 60mm.sup.2, or from 60 mm.sup.2 to
about 1000 mm.sup.2, or from about 60 mm.sup.2 to about 900
mm.sup.2, or from about 60mm.sup.2 to about 800 mm.sup.2, or from
about 60 mm.sup.2 to about 700 mm.sup.2, or from about 10 mm.sup.2
to about 600 mm.sup.2, or from about 60 mm.sup.2 to about 500
mm.sup.2, or from about 60 mm.sup.2 to about 400 mm.sup.2, or from
about 60 mm.sup.2 to about 300 mm.sup.2, or from about 60 mm.sup.2
to about 600 mm.sup.2, or from about 60 mm.sup.2 to about 100
mm.sup.2, or from about 60 mm.sup.2 to about 90 mm.sup.2, or from
about 60 mm.sup.2 to about 80 mm.sup.2, or from about 60 mm.sup.2
to about 70 mm.sup.2, or from about 16 mm.sup.2 to about 400
mm.sup.2, or from about 36 mm.sup.2 to about 225 mm.sup.2, or from
about 64 mm.sup.2 to about 100 mm.sup.2
[0116] The microprojections of the microprojection arrays of the
present invention may be solid or non-porous or contain hollow
portions therein. In some embodiments the microprojection as solid
and non-porous and do not contain hollow portion therein. In
preferred embodiments the devices of the present invention do not
contain reservoirs.
[0117] In view of the above, it will be appreciated that the
present invention is generally directed to devices and methods for
intradermal delivery of active agents into the skin. The invention
is directed to devices and methods for improving the immunogenicity
of vaccine preparations by intradermal delivery of the vaccine via
a microprojection array in which the parameters for delivery of the
active agents have been developed to achieve appropriate depth
penetration and efficient delivery of the active agent.
[0118] The methods of the present invention may be used to design
vaccination devices as well as develop the parameters for delivery
of vaccines efficiently and minimize the penetration energy of the
array while controlling skin inflammation, tolerability and
acceptability. The present methods further enable investigation of
the application of other cutaneous devices (e.g. solid, hollow, or
dissolvable penetrators of custom size, possibly arranged in
linear, rectangular or round arrays of arbitrary density) to
different skin types.
[0119] The present invention relates to microprojection arrays
wherein the physical parameters of the arrays such as but not
limited to array mass, microprojection density, microprojection
diameter, array size, microprojection tip angle, microprojection
base diameter are determined for a given application.
[0120] The present invention relates to microprojection arrays
wherein the physical parameters of the arrays can be determined for
a given penetration depth range.
[0121] The present invention relates to methods of designing the
physical parameters of microprojection arrays for a given
penetration depth range.
EXAMPLES
Example 1
Microprojection Array Application to Mouse Skin
[0122] Microprojection arrays were fabricated using a deep-reactive
ion etching approach and diced from silicon wafers by the
Australian National Fabrication Facility (ANFF) at The University
of Queensland as previously described (D. Jenkins, S. Corrie, C.
Flaim, M. Kendall, RSC Advances 2012, 2, 3490). Arrays were first
cleaned in 70% ethanol for 10 min, flushed with an excess of water,
then dried under ambient conditions. Prior application to skin, the
arrays were coated with fluorescent nanoparticles
(Fluospheres.RTM., 0.2 mm, Yellow Green Fluorescent 505/515 nm, 2%
Solids, Molecular Probes.RTM., Oregon, USA) as described by Coffey
et al (J. W. Coffey, S. R. Corrie, M. A. Kendall, Biomaterials
2013, 34, 9572). In brief, 8 .mu.L of solution containing
Fluospheres.RTM. with 0.2% solids and 1% methylcellulose (w/v
methylcellulose, Sigma-Aldrich, USA) was deposited onto a 4.times.4
mm.sup.2 array and dried using a rotating nitrogen jet to evenly
distribute the solution on the whole array while simultaneously
localizing the respective payload on the projection (X. Chen, T. W.
Prow, M. L. Crichton, D. W. Jenkins, M. S. Roberts, I. H. Frazer,
G. J. Fernando, M. A. Kendall, J Control Release 2009, 139, 212).
The volume was 4.5 .mu.L and 18 .mu.L for the 3.times.3 mm.sup.2
and 6.times.6 mm.sup.2 arrays, respectively, to maintain a constant
coating volume per unit array area. Coated arrays were stored in
sealed Petri dishes protected from light until used. Scanning
electron Microscopy (SEM) was performed before and after coating to
ensure microprojection integrity and shape consistency. The arrays
selected measured (uncoated) 90-110 .mu.m in length, 16-20 .mu.m in
width at the base, and tapered a 15.degree. -25.degree. angle
terminating in a tip of .about.1 .mu.m in diameter. Coating
increased base width increase of .about.4 .mu.m and the tip angle
to .about.35.degree.. Female BALB/c mice aged 6 to 8 weeks were
chosen because commonly used for immunology experiments and due to
the reduced speckling during tissue imaging. The mice were
anesthetised prior to array application with a solution of 60 .mu.L
of 25 mg/mL ketamine and 5 mg/mL xylazine in saline via
intraperitoneal injection and were treated according to the
protocol approved by the University of Queensland Animal Ethics
Committee. Arrays were applied to the inner earlobe of the ears
using an applicator device consisting of a sprung piston. Different
impact velocities and energies were generated firing pistons of
different masses and varying the initial spring compression through
holes drilled in the cylinder housing. The mass was decreased from
the standard 35 g of the brass piston, using a plastic piston
jointly with .about.9 g incremental weights screwed on its top end.
During application, the ear rested on a 3 mm-PDMS backing slab.
Adhesive carbon tabs fixed the ear to the PDMS and the PDMS to the
bench support. The array was left in place for 2 min and then
carefully removed. The animals were euthanized immediately after
treatment through cervical dislocation and the ears excised for
experimental characterization.
Example 2
Experimental Characterization of Skin Penetration
[0123] The excised ear specimen was immediately fixed by immersion
into in 2% paraformaldehyde in phosphate buffer saline (PBS) for
.about.2 hours, and then frozen in Optimal Cutting Temperature.RTM.
(OCT) compound (Tissue Tek, QLD, Australia). 10 .mu.m-thick
sections of frozen ear were sectioned normal to the skin surface
and approximately parallel to projection holes rows using a Leica
Ultracut UCT cryo-microtome (Leica Microsystems, Wetzlar, Germany)
at the HistoTechnology facility of the QIMR Berghofer Medical
Research Institute. Sections were imaged under a Zeiss LSM510
confocal microscope (Carl Zeiss Inc., Germany), using excitation
and collection wavelengths of 488 nm and 500-550 nm nm,
respectively. The fluorescent tracks left by fluorescent
microsphere-coated projections were measured using imageJ (NIH,
USA, http://imagej.nih.gov/ij/) for a minimum of 3 slides
(distributed uniformly across the treated area) per ear sample,
resulting in over 100 measurements per application condition.
Because penetration depth varied across the array, the measurements
taken for each slides were divided in an edge group, including up
to 10 tracks from each side, and a center group, including all
other tracks. For each slide the mean and standard deviation of the
depth measurements was calculated for the edge group and center
group independently. A weighted average was performed on the center
group means and standard deviation for each slide within a sample,
with weights equal to the number of track measured per slide. This
allowed the measure to rely more on slides with a larger amount of
tracks. The standard deviation was also calculated across the
slides within a sample. An identical procedure was followed for the
edge group. For each one of the n=4 ear samples, the mean and
standard deviation between the center and edge group means gave the
sample mean and error. The overall mean (across the repeats of each
penetration condition) penetration depth (FIGS. 4B, 5A-5E and 6A-B)
was further calculated as weighted average across sample means with
weights equal to the number of tracks measured in each ear, to
allow the result to rely more on samples where more tracks were
measured. The standard deviation across the samples means was taken
as measure of overall standard error (se) of the mean depth and
plotted as error-bars (FIGS. 4B, 5A-5E and 6A-B). To quantify the
penetration depth variation due to skin (and application)
variability across subjects (mice), the standard deviation (of the
population) was estimated by multiplying the se of the mean depth
by the square root of the number of terms n.sub.t in each average
step performed, according to the Bienayme's formula
se=sd/(n.sub.t).sup.0.5 (see any inferential statistics textbook).
Note that this is a rough approximation because statistical
independence of the values in the sample cannot be strictly
assumed. This factor is .+-.2.sup.0.540.sup.6.5 (where `2` derives
from the step where center and edge means were averaged, and `40`
is (conservatively) the largest number of tracks measured in an
edge or center group). To quantify the penetration depth variation
due to microprojection position across the array, the depths of the
10 most peripheral tracks were averaged across slides, and then
again across samples. The maximum of such 10 mean depths was taken
to be the upper end of the bar in FIG. 4D. Similarly, the depths of
10 center tracks were averaged across slides, and then across
samples. The minimum of such 10 mean depths was taken to be the
lower end of the bar in FIG. 4D. Separately, cryogenic SEM of
penetrated skin was performed in accordance with Coffey et al. (J.
W. Coffey, S. R. Corrie, M. A. Kendall, Biomaterials 2013, 34,
9572).
Example 3
Indenter/Microprojection Model
[0124] The microprojection geometry was drawn according to the SEM
measurements (FIG. 10). The coated profile was considered for the
penetration-depth study to accurately reproduce the characteristics
of the arrays used for the experimental validation. The
microprojections (or indenters) were assumed to be undeformable
because silicon (E.sub.Si>100 GPa) is over 100-fold stiffer than
the skin (M. A. Hoperoft, W. D. Nix, T. W. Kenny, J
Microelectromech S 2010, 19, 229). Euler buckling theory (R. C.
Hibbeler, in Statics and Mechanics of Materials, Prentice Hall,
Singapore 2004) was used to estimate the critical axial load of
microprojections .about.40 mN, which is above the maximum force
acting on axially on the tip for the application conditions used in
this work. Post-application examination of the arrays showed
negligible or no microprojection failure.
[0125] The motion of the rigid analytical surface that modeled the
projection was characterized by an initial velocity (i.e. the
velocity generated by the applicator) and a bound mass (determined
by the piston mass). The movement was restricted to translation
along the vertical axis x=0, y=0, i.e. orthogonal indentation
respect to the skin surface. Normal contact interactions were
implemented in the FEA using the kinematic contact method because
the penalty method was occasionally observed to allow cross-over of
the master (microprojection) and slave (skin) surfaces. This
happened although the skin elements in contact with the
indenter/microprojection were always much smaller than the tip
radius (<<0.5 .mu.m). In contrast, the simpler penalty method
was used to model tangential friction contact. A friction
coefficient of 0.4 was chosen according to the experimental
measurement of Bhushan and colleagues (B. Bhushan, J Colloid Interf
Sci 2012, 367, 1; B. Bhushan, S. Chen, S. R. Ge, Beilstein J
Nanotech 2012, 3, 731).
Example 4
FE Parameterization of Skin Fracture
[0126] Ultimate and yield strength, and plastic strain at damage
were derived from previous works (R. C. Haut, Journal of
Biomechanical Engineering-Transactions of the Asme 1989, 111, 136).
The properties measured for the SC in high humidity conditions
(.about.90% RH) where used to parameterize the VE, because the
corneocytes are essentially flattened and dried epidermal cells.
The properties measured for whole skin were used to parameterize
the dermis because this layer dominates the skin overall
composition and mechanical properties (R. Reihsner, B. Balogh, E.
J. Menzel, Med Eng Phys 1995, 17, 304). For simulations including
fracture, the vertical mesh pitch (i.e. element length) was
increased in the SC and VE and decreased in the deep dermis to
allow larger element deformation and better accuracy in the
simulation of dermal penetration.
Example 5
Experimental Characterization of Impact Velocity and PDMS Backing
Behavior
[0127] To characterize the impact response of the backing alone,
the applicator was fired (n=5) without array on the PDMS+carbon tab
(no ear) using different masses and spring compressions (resulting
in 1-7 m s.sup.-1). The movement of the piston was filmed using a
Photron SA4 high-speed camera (HSC) at 20,000 frames s.sup.-1
(Photron Inc., San Diego, Calif., USA). We tracked the motion of
the piston with the HSC software to obtain piston displacement,
velocity and acceleration over time before and after contact with
PDMS. The dynamic compression displacement of the backing was then
the combined with the transient impact force measured (n=5) with a
quartz force sensor (model 208CO2, PCB piezoelectronic, Depew,
N.Y., USA) placed under the PDMS slab and recorded using a labview
program (National Instrument Corp., Austin, Tex., USA). The
resulting force-displacement characteristic (FIG. 9) was non-linear
with a small-strain stiffness .about.20 N mm.sup.-1. This was in
agreement with dynamic mechanical analysis (DMA) tests (not shown)
using an Instron Testing System 5543 (Instron, Norwood, Mass., USA)
equipped with a 5.times.5 mm.sup.2 probe driven at 50 Hz with
peak-to-peak amplitude of .about.0.8 mm (i.e. peak displacement
velocity .about.0.1 m s.sup.-1). The loss tangent was tan
.delta.=0.23.+-.0.06 and in the typical range for elastomers and
viscoelastic rubbers. Separately, the impact energy U (FIG. 3H) was
calculated from the momentum p=(2 U m).sup.0.5, which was obtained
integrating the load-cell force-time curves (FIG. 8) up to the
peak.
Example 6
Backing Lumped-Parameter Model
[0128] The backing was modeled as a viscoelastic material using the
lumped-parameter Kelvin-Voigt-like element consisting of a mass
connected to ground through a spring-damper parallel, and
implemented in Abaqus using a connector element. The non-linear
stiffness k measured with the impact tests (FIG. 9) was implemented
in tabular form. The effective mass m* accounts for the inertia of
the mass distributed across the thickness of PDMS itself, hence was
approximated to 1/3 of the mass of the PDMS volume covered by the
piston according to E. Linder-Ganz, A. Gefen, Mechanical
compression-induced pressure sores in rat hindlimb: muscle
stiffness, histology, and computational models, Vol. 96, 2004. The
damping coefficient is c=tan .delta. (k m*).sup.0.5, where k was
approximated to the small-strain value. This model (backing only)
was employed to simulate the backing impact test and the
parameterization iteratively refined until the numerical force
response matched the results of the backing impact tests. All
lumped parameters were scaled according to the area simulated when
used in conjunction with the skin FE model, i.e. m*, k and c
relative to the piston impact tests where divided by the piston
cross-sectional area and multiplied by the square of the
microprojection spacing.
Example 7
The Out-of-Plane Hyperelastic Properties of Skin Layers for Varying
Strain Rates
[0129] The strain-rate dependence of skin elasticity by indenting
individual strata of freshly-excised mouse ear (SC, VE and dermis)
with spherical tips (1.9 .mu.m and 6.6 .mu.m in diameter) at
different velocities was investigated. This experimental procedure
and the extrapolation hyperelastic 1st-order Ogden parameters was
performed as described by M. L. Crichton, B. C. Donose, X. F. Chen,
A. P. Raphael, H. Huang, M. A. F. Kendall, Biomaterials 2011, 32,
4670 (FIG. 2A-2F). In FIGS. 2A-2F, the purple data were collected
with a 1.9 .mu.m probe and the green data were collected with a 6.6
.mu.m probe. The approximate logarithmic strain-rate generated is
indicated by the top abscissa. A dotted line indicates that a
statistically significant Spearman correlation was found between
the hyperelastic parameter and the velocity/strain rate, and
represents a linear regression in Log-Log scale. A horizontal
dashed line indicates that the correlation was not significant
(p>0.05). A square bracket indicates a statistically significant
variation of the hyperelastic parameter with probe size;
****p<0.0001, ***p<0.001. Young's modulus E of the SC (both
probe sizes; FIG. 2A) and dermis (small probe only; FIG. 2C), and
the stretch exponent a of the VE (small probe only; FIG. 2E)
significantly correlated (Spearman r.gtoreq.0.95, p<0.001) with
the indentation velocity. This further implicates correlation with
the peak strain rate at contact because of its defining linear
relationship with the probe impact velocity. Power relationships,
i.e. the dotted straight lines in Log-Log scale, fitted these
datasets better (adjusted R.sup.2>0.83 except for SC 6.6
.mu.m-probe E that scored 0.62) than logarithmic, linear and
exponential curves. This rate dependency is in general agreement
with the elastic properties previously extrapolated from in-plane
uniaxial stretch tests on pig skin up to .about.10.sup.-2
.mu.s.sup.-1[45] and rat skin up to .about.10.sup.4 .mu.s.sup.-1.
For the parameters that correlated with velocity non-linear
regressions were used to predict the layer hyperelastic properties
at larger strain rates (0.3-10 .mu.s.sup.-1), i.e. relevant for the
application microprojection arrays (0.3-10 m s.sup.-1). For
example, FIG. 2G shows that the Young's modulus of the SC and
dermis increase with strain rate and is expected to exceed 100 MPa
above 1 .mu.s.sup.-1, whereas it remains approximately constant and
below 5 MPa for the VE. FIG. 2H indicates that the stretch exponent
(a) of the VE may increase over 100 at strain rate >1
.mu.s.sup.-1. No previous report of such effect was found for the
skin. In FIGS. 2G and 2H, both the column height and the numbers
indicate the means; the error bars represent the se for the
experimental measurement at 10.sup.-4 m s.sup.-1, whereas show the
90% prediction band for the values extrapolated at 0.3-10 m
s.sup.-1.
[0130] Separately, the smaller tip resulted in a statistically
significant (Wilcoxon p<0.0001) larger E for the VE (FIG. 2B)
and lower a for the dermis (FIG. 2F), compared to the larger tip.
Recent measurements of whole mouse ear skin showed an inverse
Log-Log linear trend (E.sub.skin=29.times.(2r).sup.-1; E.sub.skin
in MPa, r in .mu.m) between the Young's modulus and the probe
radius r across .mu.m to mm scales. The analogous curve (not shown)
intercepting our two scale-dependent values of VE Young's modulus
(averaged over the velocities) was
E.sub.VE=2.7=2.7.times.(2r).sup.-0.9. SC stretch exponent did not
show significant scale or rate dependence (FIG. 2D), thus the
overall mean across the velocities for the small probe was reported
in FIG. 2H.
Example 8
Skin Failure and Fracture Mechanics during Penetration: Model and
Properties
[0131] Characterization of skin penetration following penetrator
impact was accomplished by numerically modeling microprojection
application to skin and comparing against experimental
observations. FIG. 7 illustrates the descriptive framework used to
capture skin failure and fracture mechanics. In brief, 1) a skin
element deforms reversibly according to the hyperelastic
properties; 2) when the von Mises (VM) stress exceeds the yield
strength, it starts deforming irreversibly (plastically) according
to a linear curve (dotted) that intercepts the stress-strain
coordinate defining the onset of damage (breaking strength and
strain at damage); 3) when the plastic strain exceeds a damage
threshold, the skin element progressively loses stiffness (material
damage) linearly with the plastic energy dissipated (dashed line);
4) the element is completely inactivated when this plastic energy
reaches a characteristic fracture energy.
[0132] The initial values for the failure properties were
determined starting from previous skin mechanical tests and then
refined to validate the fracture model against the penetration
experiments. The puncture and tearing energy of whole skin and
isolated SC has been reported to exceed 600 pJ .mu.m.sup.-2.
Initially, simulation of a 2 m s.sup.-1 microprojection impact
using the threshold strengths and strains and fracture energy of
600 pJ .mu.m.sup.-2 for all skin layers resulted in failure
initiation and plastic deformation of the elements. However, no
element inactivation occurred above 6 .mu.m displacement of the tip
into the skin, with a maximum stiffness degradation <10%. This
indicated that the fracture energy had been overestimated, possibly
because previous measurements could not isolate fracture
dissipation from other energetic contributions (e.g. elastic strain
or yielding). Hence, we varied the layer fracture energies in the
range 0-200 pJ .mu.m.sup.-2 (0, 0.2, 1, 6, 35, 100 and 200 pJ
.mu.m.sup.-2 were used) until the simulations matched the fracture
behavior observed experimentally. For example, the SC optimal
energy was approximately 35 pJ .mu.m.sup.-2 suggesting that its
rupture occurs through a combination of delamination (energetically
`cheaper` 1-10 pJ .mu.m.sup.-2) and tear (energetically more
`costly`.about.10.sup.3 pJ .mu.m.sup.-2). Using the layers optimal
energies, the total irreversible strain energy (i.e. plastic and
damage dissipations) when the projection has penetrated to the
bottom boundary of the dermis (i.e. 4.45 .mu.s after the contact)
was about 100 nJ. The simulations showed that this value was most
sensitive to the dermis fracture energy, probably due to its larger
thickness. The dissipation error bounds were taken to be 50 nJ and
170 nJ, which resulted when the dermis was parameterized with 1 pJ
.mu.m.sup.-2 and 35 pJ .mu.m.sup.-2, respectively. Such error range
is reasonably tight compared to the total energy of the system (the
application energy per projection is 21 .mu.J) and is satisfactory
for the purpose of this work considering the limited literature
about rupture energy measurements, especially for penetration-like
fracture modes.
[0133] SC flaps partially overlap with the VE. This non-physical
behavior occurs because, for simplicity, no `self`-contact
interaction properties were defined for the skin elements. However,
the overlap involves skin portions that have already failed and
have little or no load-bearing capacity; therefore, the errors in
strain energy and stress were assumed to be negligible.
Interestingly, stiffness degradation and fracture (element
inactivation) originated .about.1 .mu.m off the microprojection
axis, i.e. where the dilatational strain peaked, rather than
immediately below the tip where the VM stress and compressive
strain peaked. This also indicates that this fracture approach
captures, at least in part, the different rupture behaviors in
tension and compression, in contrast with fracture models solely
based on a VM stress threshold. Note that the cartilage was not
assigned failure mechanisms because this work focuses on skin
targeting and cartilage penetration is avoided. Rather, to avoid
bias of the numerical results due to artificial cartilage
resistance to penetration, the projection was allowed to penetrate
the cartilage with at zero energy cost by deactivating contact
interactions of its FE nodes with the microprojections. Having
established the optimal skin fracture parameters, this failure
implementation is used in the next section to simulate the
penetration by arrays of microprojections.
Example 9
Energy Contributions to Skin Penetration: Elastic Deformation,
Fracture and the Role of Subcutaneous Backing Layers
[0134] FIG. 3A represents a snapshot along the penetration
trajectory of a .about.3000-microprojection array impacting the
skin at 2 m s.sup.-1 with a bound mass (applicator piston) of
.about.35 g. According to this simulation, when a microprojection
has penetrated to the dermis bottom boundary its velocity has
decreased negligibly (<2%) and penetration would continue across
the cartilage. FIG. 3F shows that less than 3% of the initial
application energy is transferred to the skin, while the majority
remains array kinetic energy. In contrast, experiments showed that
similar application velocities (.about.2 m s.sup.-1) result in
mid--to deep-dermal penetration. This means that the current model
does not account for several mechanisms that absorb a major
fraction (>90%) of the application energy. One possible reason
could be attributed to the linked assumption that microprojections
are largely spaced and do not influence each other. Hence, the
penetration of arrays with finite microprojection
densities/spacings (FIG. 3B-D) was simulated using the 3D symmetric
FE geometry schematized in FIG. 3E. Interestingly, densities around
10 kproj cm.sup.-2 (i.e. 10,000 proj cm.sup.2) appeared to decrease
the friction dissipation in favor of an increased energy
contribution to failure and fracture (FIG. 3G). The elastic strain
energy was approximately constant with the projection density;
however, VM stress above 1 MPa concentrates at the penetration site
in the ventral (top) skin layers when the projections are largely
spaced, while it progressively spreads to the cartilage, dorsal
(bottom) dermis, VE and SC as the density approaches 20 kproj
cm.sup.-2. Most importantly, the total energy transferred to the
skin when the projection has penetrated to the bottom of the dermis
is essentially independent of the microprojection density (at least
up to 20 kproj cm.sup.-2). Rather, the remaining kinetic energy may
be transferred to the backing layer, i.e. a 3 mm-thick PDMS slab
placed under the ear during the microprojection array application.
This is employed to cushion the impact and avoid ear tissue damage
while allowing applications at high velocities (.about.m s.sup.-1).
The force transmitted across the backing was measured by placing a
piezoelectric load cell below the PDMS slab (i.e. on the bench
support; FIG. 8). FIG. 3H shows that this energy is approximately
5% lower than the energy transmitted when a flat (projection-less)
patch is applied on the backing alone (without mouse ear). This
means that only a small amount of energy (.about.5%) is transferred
to the ear, which explains the excess of energy in the simulation
(.about.95% to the backing). Hence, accurate modeling of skin
penetration requires accounting for possible compliant backing
layers like our PDMS or subcutaneous fat and muscle found in vivo
(less stiff than skin).
[0135] Throughout this specification and claims which follow,
unless the context requires otherwise, the word "comprise", and
variations such as "comprises" or "comprising", will be understood
to imply the inclusion of a stated integer or group of integers or
steps but not the exclusion of any other integer or group of
integers.
[0136] Persons skilled in the art will appreciate that numerous
variations and modifications will become apparent. All such
variations and modifications which become apparent to persons
skilled in the art, should be considered to fall within the spirit
and scope that the invention broadly appearing before
described.
* * * * *
References