Page 1 of 21 Combination Antiviral–Anti-Inflammatory Treatment for Recurrent Herpes Simplex 1 ...

24 Pages·2013·1.45 MB·English

AAC Accepts, published online ahead of print on 16 December 2013
Antimicrob. Agents Chemother. doi:10.1128/AAC.02150-13
Copyright © 2013, American Society for Microbiology. All Rights Reserved.
1 Combination Antiviral–Anti-Inflammatory Treatment for Recurrent Herpes Simplex
2 Labialis: Rationale and Efficacy Demonstrated with a Novel Composite Efficacy Measure
3 Christopher M. Hull, MD (corresponding author): University of Utah, Salt Lake City, UT;
4 [email protected]
D
5 Myron J. Levin, MD: University of Colorado, Anschutz Medical Campus, Aurora, CO; o
w
n
6 [email protected] lo
a
d
7 Stephen K. Tyring, MD, PhD: University of Texas Medical School at Houston, Houston, TX; e
d

f
r
8 [email protected] o
m

h
9 Spotswood L. Spruance, MD: University of Utah, Salt Lake City, UT; [email protected] t
t
p
:
/
/
a
a
c
10 Running Title: Combination Treatment for Recurent HSL .
a
s
m
.
o
r
11 Keywords: acyclovir 5%/hydrocortisone 1% cream; cumulative lesion area; ulcerative lesion g
/

o
n
12 rate; recurrent herpes simplex labialis; efficacy endpoints; topical antiviral therapy
A
p
r
il
13 Research Areas: antimicrobial chemotherapy, host-microbial interactions, virology 6
,

2
0
1
9

b
y

g
u
e
s
t
Page 1 of 21
14 CONFLICT OF INTEREST
15 Dr. Christopher Hull is a full-time employee of University of Utah. He has received financial
16 compensation from Valeant for work as an advisory board member and consultant. He has
17 received financial compensation from Medivir AB for work as a study board member and from
18 Novartis Pharmaceuticals Corporation for work as speakers’ bureau member. He has also
D
o
19 received financial compensation for participation in contract research for Medivir,
w
n
20 GlaxoSmithKline, Transport Pharmaceuticals, Inc., and Nanobio Corporation. lo
a
d
e
d
21 Dr. Myron Levin is a full-time employee of University of Colorado. He has received financial f
r
o
m
22 compensation for participation in contract research for GlaxoSmithKline, Merck & Co., and
h
t
t
p
23 Sanofi Pasteur. He shares a patent with Merck and has received financial compensation from the :
/
/
a
a
24 company for consultant work. He has also received financial compensation from
c
.
a
s
25 GlaxoSmithKline for work as an adjudication board member. m
.
o
r
g
26 Dr. Stephen Tyring is a full-time employee of University of Texas Medical School at Houston. /
o
n

27 He has participated in contract research for Medivir; financial compensation for this work was A
p
r
28 paid directly to his employer. il
6
,

2
0
29 Dr. Spotswood Spruance is a retired employee of University of Utah. He has received financial 1
9

b
30 compensation from Valeant for work as a scientific advisor. y

g
u
e
s
t
Page 2 of 21
31 ABSTRACT
32 Historically, the primary target for research and treatment in recurrent herpes simplex labialis
33 (HSL) has been limited to inhibiting herpes simplex virus (HSV) replication. Antiviral
34 monotherapy, however, has proven only marginally effective in curtailing the duration and
35 severity of recurrent lesions. Recently, the role of inflammation in the progression and resolution
D
o
36 of recurrences has been identified as an additional target. This was evaluated in a randomized
w
n
37 study comparing combination topical acyclovir 5%/hydrocortisone 1% cream (AHC) with lo
a
d
e
38 acyclovir 5% alone (AC; in AHC vehicle) and vehicle. The efficacy of each topical therapy was d

f
r
o
39 evaluated with cumulative lesion size, a novel composite efficacy endpoint incorporating episode m

h
40 duration, lesion area, and proportion of nonulcerative lesions. In that study, cumulative lesion tt
p
:
/
/
41 area was significantly decreased with AHC compared with AC (25% decrease; p<0.05) and a
a
c
.
42 vehicle (50% decrease; p<0.0001). As research continues in this arena, cumulative lesion area a
s
m
43 should be included as a measure of efficacy in clinical trials of recurrent HSL therapies. .o
r
g
/

o
n

A
p
r
il
6
,

2
0
1
9

b
y

g
u
e
s
t
Page 3 of 21
44 INTRODUCTION
45 The seroprevalence of herpes simplex virus 1 (HSV-1), the HSV type most often associated with
46 herpes simplex labialis (HSL) or ‘cold sores’, is estimated to be 58% among immunocompetent
47 Americans 14-49 years of age (1) and 31% among children 6-13 years of age (2). Approximately
48 15-40% of these seropositive individuals will experience a symptomatic HSL recurrence, the
D
o
49 frequency of which depends on genetic susceptibility, immune status, age, site of infection, and
w
n
50 viral subtype (HSV-1 vs HSV-2) (3, 4). In most cases, patients with recurrent HSL experience lo
a
d
e
51 minimal discomfort, pain, and disfigurement, but frequent outbreaks (>5 episodes/year; d

f
r
o
52 approximately 35% of patients with recurrent HSL) are associated with significant impact on m

h
53 physical, emotional, and social well-being (5). tt
p
:
/
/
a
a
54 Recurrences of HSL differ from the primary/initial infection with respect to the roles that viral
c
.
a
s
55 replication and inflammation play in the course of a lesion. In primary infection, lesion m
.
o
56 development, progression, and resolution are closely correlated with viral replication. In rg
/

o
57 recurrent disease, lesion development and progression depend primarily on the pro-inflammatory n

A
p
58 host response (6) (Figure 1). The central role of inflammation in lesion progression in recurrent r
il
6
59 disease may explain the limited efficacy of current antiviral monotherapy and suggests a new ,
2
0
1
60 approach for more effective treatment. The value of one such approach, antiviral—anti- 9

b
y
61 inflammatory combination treatment, has been demonstrated via several endpoints including
g
u
e
62 cumulative lesion area, a new efficacy endpoint. s
t
Page 4 of 21
63 DUAL DISEASE PROCESSES IN HERPES SIMPLEX LABIALIS RECURRENCES
64 Recurrent Lesion Development: Correlations with Viral Processes
65 In recurrent HSL, HSV periodically reactivates (by mechanisms that are not completely
66 understood) and moves from the cell body of latently infected sensory neurons toward the
67 epidermis, where the virus begins to replicate. Death of infected cells produces an inflammatory
D
o
68 response (discussed below) leading to papule and subsequent vesicle formation, generally within
w
n
69 8 hours of onset of prodromal symptoms (7). Given this short interval and the correlation of lo
a
d
e
70 lesion area with the neuron sensory field, it is thought that lesions form through coalescence of d

f
r
o
71 multiple foci of infection, rather than lateral spread within the skin (6). Viral replication is m

h
72 evident as early as the prodromal stage, reaching maximal levels at the vesicle stage and tt
p
:
/
/
73 declining rapidly thereafter (8, 9). The level of viral replication, or viral load, correlates with a
a
c
.
74 episode severity as measured by lesion area, symptoms, and healing time (8, 10). a
s
m
.
o
75 The well-recognized importance of HSV replication in lesion pathogenesis has established it as rg
/

o
76 the main target of treatment. Antivirals are effective in shortening episode duration in primary n

A
p
77 HSV infections; however, in recurrent HSL, topical treatments have only moderate effect on r
il
6
78 lesion healing time, and rarely any effect on lesion area or the proportion of nonulcerative lesions ,
2
0
1
79 (recurrences that do not progress beyond the papule stage; sometimes referred to as macular, 9

b
y
80 papular, or aborted lesions) (11, 12). Based on results from clinical trials, the benefit of topical
g
u
e
81 antiviral therapy on episode duration in recurrent HSL appears to be limited to about 10–15% s
t
82 (11, 12) despite a robust effect on viral replication (13). This may be due in part to the timing of
83 viral replication and treatment initiation. Since viral replication is rapidly controlled (1-2 days)
84 by host immune responses (8), if the antiviral drug is applied after viral replication reaches its
85 maximum, there is little opportunity for the drug to have a significant effect as demonstrated by
Page 5 of 21
86 the differences in efficacy observed when antivirals are initiated early (during prodrome or
87 erythema stages) versus late (during vesicle or later stages) (14-18).
88 Recurrent Lesion Progression and Resolution: Correlations with Inflammatory Processes
89 The importance of robust HSV-specific cell-mediated immunity to the maintenance of HSV
D
90 latency has long been appreciated. More recently, the contribution of proinflammatory processes o
w
n
91 to the progression and resolution of HSL recurrences has been established (19-24). Inflammatory lo
a
d
92 infiltrates are apparent soon after the onset of symptoms (even in those with no visible lesions) e
d

f
r
93 and persist during vesicle and crust formation (19-21). In addition, there is evidence that o
m

h
94 interferon and/or other proinflammatory mediators play a role in stimulating cellular responses t
t
p
:
/
95 and preventing lesion area enlargement (23). Histopathologically, vesicles form when HSV- /a
a
c
96 infected cells lyse inducing cytokine and chemokine production that leads to inflammatory .a
s
m
97 infiltrates of neutrophils, monocytes, and T lymphocytes that extend through the epidermis and .
o
r
g
98 basement membrane to the dermis (19-22). In later lesion stages (ie, ‘crusted’ lesions), /
o
n

99 eosinophilic infiltration is apparent (19). A
p
r
il
6
100 Based on the observation that peak interferon levels correlate with time-to-next recurrence (24). ,

2
0
101 Bernstein et al (25) tested the effects of topical imiquimod, a toll-like receptor 7 agonist that 1
9

b
102 induces interferon and other cytokines, in patients with recurrent disease. They found that y

g
u
103 imiquimod significantly delayed time-to-next recurrence compared with vehicle (50 vs 91 days; e
s
t
104 p=0.018). However, the enhanced immune reaction also caused a significant increase in
105 cutaneous lesion symptoms at the time of application (burning, erythema, edema, scabbing
106 and/or flaking, pain, burning; p<0.05 for all) and enlarged the maximum lesion area (median 84
107 vs 30 mm2; p=0.006), demonstrating the important contribution of cellular immune responses to
Page 6 of 21
108 lesion progression. In contrast, when local cellular responses were suppressed using oral
109 levamisole, patients experienced dose-dependent reductions in lesion pain (correlation
110 coefficient [r] = -0.34; p=0.033) and duration (r= -0.32; p=0.051), but with increased recurrence
111 frequency (r= 0.43; p=0.007) compared with placebo (26).
D
112 Molecular Mechanisms in the Natural History of Recurrent Herpes Simplex Labialis o
w
n
113 After reactivation, HSV travels down the nerve to infect and replicate within keratinocytes (27) lo
a
d
114 (Figure 2). This results in upregulation and secretion of β-chemokines, interleukins, and adhesion e
d

f
r
115 molecules (via NFκB activation), which attract monocytes, neutrophils, and CD4+ T cells to the o
m

h
116 site of infection. Resident dendritic cells take up HSV antigens and migrate to the draining t
t
p
:
/
117 lymph node to present antigen to specific central memory cells. Antigen presentation also /a
a
c
118 stimulates CD4+ lymphocytes at the infection site to release IFNγ and dendritic cells to release .a
s
m
119 IL-12. The IFNγ restores major histocompatability complex (MHC) class I and II expression in .
o
r
g
120 infected keratinocytes and stimulates inducible nitric oxide synthase (iNOS) expression; both /
o
n
121 IFNγ and IL-12 activate CD4+ T lymphocyte cytotoxicity. A
p
r
il
6
122 As a result of the actions of monocytes, neutrophils, and cytotoxic T cells, virus-infected ,

2
0
123 keratinocytes lyse, releasing viral particles along with cellular contents, which causes erythema 1
9

b
124 and edema. Proinflammatory cytokines released from keratinocytes and inflammatory cells (via y

g
u
125 iNOS) along with prostaglandins (via NFκB-induced expression of cyclooxygenase 2 [COX2]) e
s
t
126 dilate blood vessels leading to greater fluid accumulation and inflammatory cell migration (27).
127 Eventually, these fluids and inflammatory cells coalesce into a vesicle, which as more and more
128 cells lyse, ruptures to form an ulcer. At this (ulcer) stage, neutrophil infiltration increases as a
129 result of increased IL-8 and E-selectin expression while CD8+ cells that are local or have
Page 7 of 21
130 migrated from the circulation (due to increased E-selectin expression) are activated by IFNγ and
131 IL-12 (19, 27). Lesion resolution is characterized by fewer and fewer inflammatory cell
132 infiltrates until the wound heals and skin returns to normal.
133 TOWARD MORE EFFECTIVE TREATMENTS FOR RECURRENT HERPES SIMPLEX LABIALIS
D
134 Given the contributions of both viral and inflammatory processes depicted in Figure 2, several o
w
n
135 authors have proposed a role for immunomodulators in mitigating the clinical impact of recurrent lo
a
d
136 HSL (12, 28, 29) by limiting lesion area enlargement and/or and preventing ulcerative lesions by e
d

f
r
137 reducing edema, protease activity, and/or inflammation. Using a newly developed mouse model, o
m

h
138 Harmenberg et al (30) tested various antiviral-immunomodulator combinations (including t
t
p
:
/
139 corticosteroids and NSAIDs) and found that combinations with corticosteroids (especially /a
a
c
140 hydrocortisone) were most effective at reducing lesion duration and severity. The value of .a
s
m
141 corticosteroids in combination with an anti-infective agent has been demonstrated in the .
o
r
g
142 management of herpetic stromal keratitis (31, 32), bacterial meningitis (33), Pneumocystis /
o
n

143 carinii pneumonia (34, 35), herpes zoster (36), and others (37). A
p
r
il
6
144 These potential benefits are based on the actions of corticosteroids in the skin as depicted in ,

2
0
145 Figure 2; namely, inhibiting the expression, release, and/or pericellular translocation of 1
9

b
146 proinflammatory mediators including lipocortin 1 (LC1), thereby reducing neutrophil infiltration y

g
u
147 and proinflammatory cytokine expression (including IL-6 and IL-8). These effects occur directly e
s
t
148 via glucocorticoid response elements or indirectly via inactivation of various transcription factors
149 (38). Corticosteroids also reduce vasodilation, inflammation, and oxidative damage within the
150 skin by inhibiting the cytokine-induced expression of iNOS and thereby reducing NO
151 production. In addition, corticosteroids can interrupt NFκB-stimulated inflammatory cytokine
Page 8 of 21
152 and AP-1 adhesion molecule expression, and reduce COX2 expression and prostaglandin
153 production. Corticosteroids can also block prostaglandin production via LC1-mediated effects on
154 phospholipase A2.
155 Ultimately, the efficacy of any treatment will depend on its dermal penetration. Recently, a new
156 topical formulation of acyclovir with better dermal penetration showed improved antiviral D
o
w
157 efficacy in a randomized clinical trial (18). In addition, adding a corticosteroid to an antiviral n
lo
a
158 may have additive or synergistic effects including enhanced effective dermal concentration of the d
e
d
159 antiviral component via steroid-mediated vasoconstriction. In a randomized trial of combination f
r
o
m
160 oral famciclovir and topical fluocinonide compared with famciclovir alone, combination therapy
h
t
t
p
161 reduced lesion healing time to normal skin (median 5.9 vs 8.9 days; p=0.06), reduced maximum :
/
/
a
162 lesion area (median 48 vs 162 mm2; p=0.02), and increased proportion of nonulcerative lesions a
c
.
a
s
163 (41 vs 8%; p=0.09) (39). m
.
o
r
g
/

o
164 IMPROVING UPON TRADITIONAL MEASURES OF EFFICACY IN RECURRENT HERPES SIMPLEX n

A
165 LABIALIS pr
il
6
166 Historically, episode duration and proportion of nonulcerative lesions have been the main ,

2
0
167 endpoints used in trials of antiviral treatments; however, these endpoints have limitations and 1
9

b
168 several alternative measures have been suggested. For example, Spruance et al (14) suggested y

g
u
169 using duration of the most painful and disfiguring lesion stages (vesicle, ulcer, and hard crust). e
s
t
170 They showed that the combined durations of these stages was reduced by 43% (p<0.001) by
171 idoxuridine among patients with ulcerative recurrences, whereas total healing time was
172 unaffected, thereby demonstrating that antiviral treatment can reduce the duration of the most
173 bothersome stages of HSL lesions even if total healing time is unchanged. Similar endpoints
Page 9 of 21
174 have been utilized by Evans et al (40) and Crane et al (41). However, lesion stage-duration
175 endpoints do not capture the ability of an antiviral to prevent ulcerative lesions, which may be
176 the most important goal of HSL therapy.
177 Several studies have addressed this by assigning nonulcerative lesions a value of zero for episode
178 duration and lesion area (values can be set to zero because, by definition, nonulcerative lesions D
o
w
179 have 'zero' duration of vesicle, ulcer, and crust stages (42)). For example in combined data from n
lo
a
180 two randomized trials, high-dose oral valacyclovir increased the proportion of nonulcerative d
e
d
181 recurrences by 6-7% (p<0.05) over placebo (42). When this effect was accounted for in the f
r
o
m
182 calculation of episode duration, the apparent efficacy of oral valacyclovir versus placebo
h
t
t
p
183 increased from a 21% (p<0.001) reduction in episode duration to loss of hard crust :
/
/
a
a
184 (nonulcerative excluded) to a 31%% (p<0.001) reduction (nonulcerative included). In another
c
.
a
s
185 randomized trial, when only ulcerative lesions were assessed, topical (via iontophoresis) m
.
o
186 acyclovir reduced episode duration to loss of hard crust by 26 hours compared with placebo (139 rg
/

o
187 vs 165 h; p=0.03). However, when aborted lesions were included, the difference in median n

A
p
188 episode duration was even more pronounced (35 hour difference; 113 vs 148 h; p=0.02). r
il
6
,

2
189 Rodu et al (43) has argued that a better measure may be efficacy in containing a lesion 0
1
9
190 (measured by lesion area) rather than ability to alter the repair process (episode duration). To this b
y

g
191 end, they devised a multi-step method of measuring lesion area and compared values at Days 2, u
e
s
192 3, 4, and 5. However, because they chose to compare lesion areas by day, strict enrollment t
193 criteria were needed so that baseline lesion areas were as homogenous as possible. When they
194 tested the efficacy of a combination of tannic, boric, and salicylic acids in patients with
195 ulcerative lesions, they showed significant reductions in lesion area on Days 2-5 of 42.6 – 53.4%
196 over placebo.
Page 10 of 21