Two randomized phase III clinical trials evaluating
`anti-inflammatory dose doxycycline (40-mg
`doxycycline, USP capsules) administered
`once daily for treatment of rosacea
`
`James Q. Del Rosso, DO,a Guy F. Webster, MD, PhD,b Mark Jackson, MD,c Marta Rendon, MD,d
`Phoebe Rich, MD,e Helen Torok, MD,f and Mark Bradshaw, PhDg
`Las Vegas, Nevada; Philadelphia, Pennsylvania; Louisville, Kentucky;
`Miami, Florida; Portland, Oregon; and Medina, Ohio
`
`Background: Doxycycline monotherapy at antimicrobial doses has been shown to be effective for the
`treatment of rosacea.
`
`Objective: To evaluate the efficacy and safety of once-daily anti-inflammatory dose doxycycline for the
`treatment of rosacea.
`
`Methods: In two phase III, parallel-group, multicenter, randomized, double-blind, placebo-controlled
`studies (studies 301 and 302), patients received 40-mg of controlled-release doxycycline (n = 269) or
`placebo (n = 268) for 16 weeks. The primary efficacy end point was the mean change from baseline in
`facial inflammatory lesion count.
`
`Results: The mean lesion count at baseline was approximately 20 in each study arm. At week 16, the mean
`change from baseline in lesion count in the active-treatment groups was e11.8 in study 301 and e9.5
`in study 302 compared with e5.9 and e4.3, respectively, in the placebo groups (P \ .001 for both
`comparisons). Anti-inflammatory dose doxycycline was well tolerated; the most common adverse events
`were nasopharyngitis (4.8%), diarrhea (4.4%), and headache (4.4%).
`
`Limitations: In both studies, the reduction of inflammatory lesion counts did not plateau within the
`16-week time frame in either treatment group. Rosacea is often treated for a period of months or years.
`
`From the Department of Dermatology, University of Nevada
`School of Medicine, and Touro University College of Osteo-
`pathic Medicine, Las Vegasa; Jefferson Medical College, Phila-
`delphiab; the Division of Dermatology, University of Louisvillec;
`University of Miami School of Medicined; Oregon Health
`Sciences University, Portlande; Trillium Creek Dermatology &
`Surgery Center, Medinaf; and Global Consulting Partners in
`Medical Biometrics, Princeton.g
`Supported by CollaGenex Pharmaceuticals, Inc.
`Disclosure: Dr Del Rosso has received grant/research support/
`honoraria from and has been a consultant/advisor and speaker
`for CollaGenex Pharmaceuticals, Galderma,
`Intendis, Medicis,
`Stiefel and has been a consultant/advisor, speaker and received
`honoraria from Bradley, Dermik, Warner-Chilcott, Ranbaxy.
`Dr Webster has received grant/research Support and honoraria
`from and been a consultant and speaker
`for CollaGenex
`Pharmaceuticals, Dermik, Galderma. Dr Jackson has received
`grant/research support and/or honoraria support from and
`been a consultant for Abbott Laboratories, Amgen, Inc. Biogen,
`Centocor, Inc, Collagenex Pharmaceuticals, Connetics Corpora-
`tion, Genentech, Novartis Pharmaceuticals, Roche Laboratories,
`3M. Dr Rendon has received grant/research support and
`honoraria from and been a consultant/advisor and speaker
`for Dermik, Galderma and received grant/research support
`from CollaGenex Pharmaceuticals. Dr Rich has been a
`
`consultant for and received grant/research support and/or
`honoraria support from Altana, Anacor, Bailer, CollaGenex
`Pharmaceuticals, Dow, DUSA, Galderma,
`Intendis, Topical
`Solutions, and Xoma. Dr Torok has received grant/research
`support or honoraria from and been a consultant/advisor and
`speaker for CollaGenex Pharmaceuticals, Galderma,
`Intendis,
`Stiefel and has received honoraria from and been a consultant/
`advisor and speaker for Amgen, Dermik, Medicis, Novartis,
`Ranbaxy., and has equity/interest in Medicis. Dr Bradshaw has
`been a consultant for CollaGenex Pharmaceuticals.
`A summary of the data found in this article has been presented
`at Academy ’05, July 21, 2005, Chicago,
`Ill; the Fall Clinical
`Dermatology Conference, October 21 and 22, 2005, Las Vegas;
`and the Coastal Dermatology Seminar, October 1, 2005, Napa,
`Calif. Highlights of the data were presented in poster format at
`the 64th Annual Meeting of the American Academy of Derma-
`tology, March 3-7, 2006, San Francisco, Calif.
`Accepted for publication November 15, 2006.
`Reprint requests: James Q. Del Rosso, DO, PMB-531, 3230 E
`Flamingo Rd #8, Las Vegas, NV 89121. E-mail:
`jqdelrosso@
`yahoo.com.
`Published online March 20, 2007.
`0190-9622/$32.00
`ª 2007 by the American Academy of Dermatology, Inc.
`doi:10.1016/j.jaad.2006.11.021
`
`791
`
`

`

`792 Del Rosso et al
`
`J AM ACAD DERMATOL
`MAY 2007
`
`The duration of the studies did not allow for assessment of safety beyond 16 weeks or whether the
`progressive improvement seen with active treatment would continue beyond 16 weeks. Neither study
`assessed the effect of treatment in patients with only erythematotelangiectatic (subtype 1) rosacea.
`
`Conclusion: Once-daily anti-inflammatory dose doxycycline appears to be effective and safe for the
`treatment of rosacea. ( J Am Acad Dermatol 2007;56:791-802.)
`
`Rosacea is a common chronic facial dermatosis
`characterized by intermittent periods of exacerbation
`and remission.1,2 Epidemiologic data suggest there is
`a genetic predisposition for this disease, with several
`intrinsic and extrinsic factors potentially correlating
`with the phenotypic expression of rosacea.2-7 Clinical
`subtypes and grading of rosacea have been defined in
`the literature.8,9
`Although there is no curative therapy for rosacea,
`recommended treatment strategies used to control
`its signs and symptoms include a combination of
`proper skin care, avoidance of recognized trigger
`factors, photoprotection, topical agents, oral ther-
`apy, and light-based physical modalities, such as
`intense-pulsed light.1,2,10-13 The most widely used
`systemic agents for the treatment of rosacea are
`oral tetracycline derivatives, including tetracycline,
`doxycycline, and minocycline.
`The use of oral tetracyclines for rosacea is based
`primarily on widespread clinical experience and a
`limited collection of placebo-controlled, compara-
`trials.2,10,12,13 The dosing ranges of
`tive clinical
`tetracycline (250-1000 mg/d) and doxycycline (100-
`200 mg/d) used to treat rosacea produce selection
`pressure against sensitive commensal flora, transient
`flora, and selected bacterial pathogens.12,14,15 Based
`on single-dose pharmacokinetics,
`conventional
`formulations of doxycycline that are not controlled-
`release and administered at a dose of 40 mg or higher
`achieve serum levels that may produce selection
`pressure against susceptible bacterial strains based on
`minimum inhibitory concentration evaluations.14,15
`As the pathogenesis of rosacea appears to be multi-
`factorial and is not definitively associated with erad-
`ication or reduction of a bacterial pathogen, it is
`scientifically plausible to use tetracycline agents as
`therapy for rosacea in a manner that does not
`exert antibiotic activity, avoids development of drug-
`resistant microbes, and exploits only their anti-
`inflammatory activities.
`The following details the results of two phase III,
`parallel-group, multicenter,
`randomized, double-
`blind, placebo-controlled trials (studies 301 and
`302) that evaluate the efficacy and safety of 40-mg
`doxycycline monohydrate in a formulation with
`30-mg immediate-release and 10-mg delayed-release
`
`Abbreviations used:
`
`adverse event
`AE:
`ANOVA: analysis of variance
`CEA:
`Clinician’s Erythema Assessment (scale)
`CMH:
`Cochran-Mantel-Haenszel (test)
`IGA:
`Investigator’s Global Assessment (scale)
`ITT:
`intent to treat
`
`beads, once daily (anti-inflammatory dose doxycy-
`cline) versus placebo once daily for the treatment
`of adults with rosacea. The studies were collectively
`inclusive of 269 patients who received anti-inflam-
`matory dose doxycycline and 268 patients who
`received placebo. Both studies included patients
`with a marked number of total inflammatory lesions
`(10-40 papules and pustules and \2 nodules),
`moderate-to-severe erythema, and presence of
`telangiectasia.
`
`METHODS
`Overall design of the studies
`Two 16-week, phase III, parallel-group, multicen-
`ter (14 sites for each study), randomized, double-
`blind, placebo-controlled studies (study 301 and
`study 302) were conducted in parallel (Fig 1).
`Study protocols for both studies were virtually iden-
`tical with the exception of a posttherapy assessment
`in study 302 that evaluated the persistence of efficacy
`and safety profile 4 weeks after discontinuation of
`study medication. Both studies enrolled a similar
`number of patients (n = 251 in study 301 and n = 286
`in study 302) and were conducted between June
`2004 and April 2005 in the United States and Puerto
`Rico.
`
`Study populations
`Patients were eligible for enrollment if they were
`healthy adults, at least 18 years of age with moderate-
`to-severe rosacea, which was defined as the presence
`of 10 to 40 papules and pustules and 2 or fewer
`nodules. At study entry, patients scored higher than
`2 on the Investigator’s Global Assessment (IGA)
`scale, a subjective 5-point measure of overall disease
`severity. IGA scores range from 0 to 4: 0 = no signs
`
`

`

`J AM ACAD DERMATOL
`VOLUME 56, NUMBER 5
`
`Del Rosso et al 793
`
`Fig 1. Flow diagram.
`
`or symptoms present (clear); 1 = 1 to 2 small,
`noninflammatory papules (near clear); 2 = 3 to 10
`papules/pustules (mild); 3 = 11 to 19 papules/
`pustules (moderate); and 4 = 20 or more papules/
`pustules and nodules (severe). Patients were also
`required to have telangiectasia and moderate to
`severe erythema as determined with the use of the
`Clinician’s Erythema Assessment (CEA) scale. Scores
`on the CEA scale range from 0 to 4: 0 = none (no
`redness present); 1 = mild (slight pinkness); 2 =
`moderate (definite redness); 3 = significant (marked
`erythema); and 4 = severe (fiery redness). Total CEA
`scores are derived by summing scores over 5 facial
`areas (forehead, chin, nose, and right and left cheek)
`and ranged from 0 to 20. In the two studies reported
`
`herein, moderate to severe erythema was defined as
`at least one area-specific CEA score of 2 or higher and
`a total CEA score of 5 or higher. Female patients of
`childbearing potential were eligible for enrollment
`only if they were using birth control, were not
`nursing, and had a negative pregnancy test at entry
`into the studies.
`Patients were not eligible for enrollment in the
`studies if they met any of the following criteria:
`initiation of or change in hormonal method of con-
`traception within 4 months of baseline or during the
`study; use of topical acne treatments or topical or
`systemic antibiotics within 4 weeks of baseline; use
`of an investigational drug within 90 days of base-
`line; known hypersensitivity to tetracyclines; use
`
`

`

`794 Del Rosso et al
`
`J AM ACAD DERMATOL
`MAY 2007
`
`of clinically significant concomitant drug therapy
`(eg, long-term use of nonsteroidal anti-inflamma-
`tory drugs); use of systemic anti-inflammatories or
`corticosteroids in the 4 weeks before baseline or
`during the study; use of vasodilators or a-adrenergic
`receptor-blocking agents 6 weeks before baseline
`or during the study; or ocular
`rosacea and/or
`blepharitis/meibomianitis requiring treatment by an
`ophthalmologist.
`These studies were conducted in accordance
`with applicable good clinical practice guidelines
`and in accordance with the ethical principles de-
`scribed in the Declaration of Helsinki. The original
`protocol, protocol amendments, and patient’s in-
`formed consent form were approved by the appro-
`priate institutional review boards for each of the 28
`study sites before the screening and enrollment of
`study participants. Before screening, patients were
`fully informed both verbally and in writing of the
`conduct and consequences of the study and signed
`an institutional review boardeapproved consent
`form.
`
`Randomization and blinding procedures
`For each study site, a master randomization list
`in blocks of 4 was prepared by the sponsor for all
`study sites. With the use of a computer-generated
`randomization scheme, patients were assigned in
`equal proportions (1:1) to receive the study drug or
`placebo. Study drug and placebo capsules were
`identical in size, shape, and color, and investiga-
`tors,
`study site personnel, and patients were
`blinded with respect to identity of the study med-
`ication being taken. All employees of the sponsor
`and its affiliates who were involved in data mon-
`itoring, data entry, or data analysis were blinded as
`well.
`
`Study medications and treatment regimens
`Patients were randomized to receive anti-inflam-
`matory dose doxycycline capsules or placebo once
`daily in the morning for 16 weeks. Patients were
`cautioned about exposure to sunlight and were
`encouraged to apply sunscreen with a sun protec-
`tion factor value of at least 30 whenever they were
`outdoors during daylight hours. In addition, pa-
`tients were instructed that the following medica-
`tions were prohibited during the studies: long-term
`use ([14 days) of sulfonamide drugs, erythromycin,
`cephalosporins, quinolones, and nonsteroidal anti-
`inflammatory drugs;
`tetracycline antibiotics; acne
`treatments, including spironolactone; antimicrobial
`soaps; penicillin antibiotics; and niacin at a dose of
`500 mg or more per day. Antacids and vitamins
`containing aluminum, calcium, or magnesium were
`
`allowed only if taken at least 1.5 hours before or
`3 hours after the patient took study medication.
`In study 302, patients in both treatment arms
`were required to discontinue study medication at
`week 16 and were instructed not
`to take any
`systemic or topical medications for the treatment
`of rosacea or acne or any of
`the medications
`prohibited at entry into the study. Patients were
`re-evaluated at week 20.
`
`Clinical evaluations
`Before entry into one of the studies, a complete
`medical history was obtained from each patient.
`Baseline evaluations also included a count of inflam-
`matory lesions (papules 1 pustules 1 nodules), vital
`signs, and routine laboratory tests. In addition to
`these measures, baseline scores were obtained on
`the CEA scale and on the IGA scale (both described
`above). Female patients of childbearing potential
`were given urine pregnancy tests.
`In both trials, study medication was distributed
`at baseline and again at week 12. Patients in both
`studies returned for evaluations at weeks 3, 6, 12,
`and 16. In study 302, patients also returned at
`week 20 for a 4-week posttreatment evaluation.
`At each visit, patients were evaluated for number
`and types of lesions, concomitant medication usage,
`adverse events (AEs), vital signs, height, and weight.
`IGA and CEA scores were obtained at each visit.
`At the week-16 visit, female patients of childbearing
`potential were again given a urine pregnancy test.
`
`Efficacy and safety evaluations
`In both studies, the primary efficacy end point was
`the mean change from baseline in total inflammatory
`lesion count (papules 1 pustules 1 nodules) at week
`16. Secondary end points included the mean change
`from baseline in CEA and IGA scores at week 16. In
`addition, the static dichotomized IGA score (yes/no),
`defined as patients who achieved a score of 0 (clear)
`or 1 (near clear), was analyzed at week 16.
`The efficacy variables for the 4-week posttherapy
`assessment conducted in study 302 included the
`mean change in total
`inflammatory lesion count
`(papules 1 pustules 1 nodules) and the mean
`changes in CEA and IGA scores from week 16 to
`week 20.
`Safety was evaluated at each study visit by
`recording AEs, concomitant medication use, and
`vital signs at each study visit and by routine labora-
`tory tests conducted at week 16. AEs were evaluated
`as mild, moderate, or severe, and the relationships
`to study medication were determined by the study
`investigator.
`
`

`

`J AM ACAD DERMATOL
`VOLUME 56, NUMBER 5
`
`Del Rosso et al 795
`
`Statistical analysis
`The sample size determination for each study
`was based on a previous placebo-controlled study.
`It was anticipated that 16 weeks of active treatment
`would result in a mean change from baseline of
`e7.0 in total lesion count, whereas treatment with
`placebo would result in a mean change of e3.5
`lesions, with a common standard deviation of 8.0.
`Hence a total of 111 patients per treatment group
`would be sufficient to ensure a power of higher
`than 90% for detecting a significant
`treatment
`difference at the two-sided alpha = 0.05 level of
`significance.
`An analysis of variance (ANOVA) was performed
`on the primary efficacy data (mean change from base-
`line in number of inflammatory lesions [papules 1
`pustules 1 nodules]) from the intent-to-treat (ITT)
`populations in each study to test the null hypothesis
`of no treatment effect. The dependent variable was
`the difference from baseline in total lesion count
`at each study time point. Treatment and center were
`the main effects examined in the ANOVA model.
`Differences between the two treatment groups were
`considered statistically significant at the P \.05 level
`using two-tailed tests. Additionally,
`the residuals
`from the ANOVA analyses were studied for devia-
`tions from normality at week 16. If the test for
`normality failed at the .05 level of significance, the
`Van Elteren test,16,17 stratified by center, was used
`to further evaluate the differences between the two
`treatment groups.
`Changes from baseline in IGA scores were evalu-
`ated at each visit using frequency distribution data.
`The distribution of IGA scores was analyzed using
`the Cochran-Mantel-Haenszel (CMH) test,18,19 strati-
`fied by center, to test for differences between treat-
`ment groups. Additional analyses were performed on
`dichotomized IGA data (yes/no). These data were
`also analyzed by using the CMH test, stratified by
`center.
`The 4-week posttherapy assessment of efficacy
`data in study 302 included only those patients who
`had data available from study visits at week 16 and
`at week 20. For both total lesion count and CEA
`scores, within-treatment data from week 16 and
`from the 4 week posttreatment follow-up visit at
`week 20 were compared by paired t test. Analysis of
`between-treatment data was performed by using the
`ANOVA method described above. For IGA scores,
`within-treatment frequency distribution data were
`analyzed by using the Wilcoxon signed rank test
`and between-treatment data were analyzed by using
`the CMH test, stratified by center. For dichotomized
`IGA scores, within-treatment data were analyzed by
`means of McNemar’s test and between-treatment
`
`data were analyzed by using the CMH test stratified
`by center.
`AEs were coded using MedDRA20 and incidences
`were calculated for each treatment group by system
`organ class and MedDRA preferred term. In this
`report, safety data (eg, incidence rates of AEs) have
`been pooled from both studies. Pooling of
`the
`safety data is justified as both studies were mul-
`ticenter
`studies
`that
`followed identical
`study
`protocols
`(with the exception of
`the 4-week
`posttreatment assessment in study 302), both stud-
`ies were conducted over the same time frame (ie,
`between June 2004 and April 2005), and the inci-
`dence rates of AEs by type were similar between the
`two studies as well as similar between treatment
`arms in both studies.
`
`RESULTS
`Patient population
`A total of 537 patients were enrolled in the two
`studies with 251 patients in study 301 (127 from the
`active-treatment arm and 124 from the placebo arm)
`and 286 patients from study 302 (142 from the active-
`treatment arm and 144 from the placebo arm). A total
`of 160 patients were enrolled in the 4-week post-
`therapy assessment conducted in study 302 (84 from
`the active-treatment arm and 76 from the placebo
`arm).
`Patient demographics and disposition and base-
`line data for efficacy variables were similar between
`treatment groups in the two studies and were also
`similar between the two studies (Table I). More than
`40% (239/537) of the patients in each study were 36
`to 50 years of age (47% [118/251] in study 301 and
`42% [121/286]
`in study 302). Seventeen percent
`(42/251) and 20% (57/286) of the patients in studies
`301 and 302, respectively, were 18 to 35 years of age,
`and 32% (79/251) and 33% (95/286) of the patients
`were 51 to 70 years of age. Seventy percent (375/537)
`of the patients in the two studies were women
`and 91% (491/537) were Caucasian. The mean total
`inflammatory lesion count was 19.9 for the patients
`participating in study 301 and 20.8 for the patients
`participating in study 302. The mean CEA score was
`9.6 for the patients participating in study 301 and
`9.3 for the patients participating in study 302.
`Approximately half of the patients in each study
`had an IGA score of 3 (moderate rosacea) and about
`90% of the patients had an IGA score of 3 to 4
`(moderate to severe rosacea). There were no statis-
`tically significant differences between treatment
`arms in either study in terms of mean lesion counts
`(papules, pustules, nodules, total lesions) or mean
`CEA and IGA scores.
`
`

`

`796 Del Rosso et al
`
`J AM ACAD DERMATOL
`MAY 2007
`
`Table I. Patient demographics and disposition, baseline values for efficacy assessments, and study
`medication usage
`
`Study 301
`
`Study 302
`
`Parameter
`
`Age, y (SD)
`Female, No. (%)
`Lesion counts, mean (SD)
`Papules
`Pustules
`Nodules
`Total
`CEA score, mean (SD)
`IGA, No. (%)
`0 = Clear
`1 = Near clear
`2 = Mild
`3 = Moderate
`4 = Severe
`Exposure to study
`medication, days,
`mean (SD)
`Completers, No. (%)*
`Discontinuations, No. (%)
`AEs
`Insufficient efficacy
`Loss to follow-up
`Protocol violation
`y
`Other
`Total
`
`Anti-inflammatory dose
`doxycycline
`(n = 127)
`
`46.8 (13.2)
`91 (71.7)
`
`15.2 (7.9)
`4.1 (5.2)
`0.2 (0.6)
`19.5 (8.8)
`9.7 (3.0)
`
`0
`0
`8 (6.3)
`67 (52.8)
`52 (40.9)
`103.1 (30.1)
`
`Placebo
`(n = 124)
`
`47.6 (11.5)
`95 (76.6)
`
`16.4 (9.2)
`3.7 (4.7)
`0.2 (0.5)
`20.3 (10.4)
`9.5 (2.7)
`
`0
`0
`10 (8.1)
`65 (52.4)
`49 (39.5)
`106.9 (24.2)
`
`Anti-inflammatory dose
`doxycycline
`(n = 142)
`
`46.3 (12.7)
`94 (66.2)
`
`17.4 (10.8)
`3.0 (4.5)
`0.1 (0.5)
`20.5 (11.7)
`0.5 (2.9)
`
`0
`0
`17 (12.0)
`77 (54.2)
`48 (33.8)
`101.2 (29.8)
`
`Placebo
`(n = 144)
`
`47.6
`95 (66.0)
`
`17.8 (10.9)
`3.3 (6.0)
`0.1 (0.5)
`21.2 (12.5)
`9.1 (2.5)
`
`0
`0
`7 (4.9)
`80 (55.6)
`57 (39.6)
`106.9 (28.7)
`
`101 (79.5)
`
`103 (83.1)
`
`115 (81.0)
`
`118 (81.9)
`
`10 (7.9)
`2 (1.6)
`4 (3.1)
`2 (1.6)
`8 (6.3)
`26 (20.5)
`
`4 (3.2)
`2 (1.6)
`2 (1.6)
`2 (1.6)
`11 (8.8)
`21 (16.8)
`
`9 (6.3)
`1 (0.7)
`5 (3.5)
`4 (2.8)
`8 (5.6)
`27 (18.9)
`
`7 (4.9)
`4 (2.8)
`5 (3.5)
`5 (3.5)
`5 (3.5)
`26 (18.2)
`
`AE, Adverse event; CEA, Clinician’s Erythema Assessment score (range, 0-20); IGA, Investigator’s Global Assessment.
`*Percent of patients taking $ 80% of the assigned study medication.
`y
`Other reasons for withdrawal
`include illness not related to study drug,
`and administrative reasons.
`
`loss to follow-up, patient withdrawal for personal reasons,
`
`The mean exposure to anti-inflammatory-dose
`doxycycline was 103.1 days in study 301 and 101.2
`days in study 302 for a combined exposure to active
`treatment of 26,377 person-days. Compliance with the
`per-protocol treatment regimen was generally very
`good; 474 of the 537 (88%) patients in the two studies
`took at least 80% of the assigned study medication.
`The majority of patients (437/537 [81%]) com-
`pleted the two studies. The rates of discontinuation
`because of AEs were higher among patients in the
`active-treatment groups
`in both studies
`(7.9%
`[10/127] and 6.3% [9/142], in studies 301 and 302,
`respectively) than among patients in the placebo
`arms of the studies (3.2% [4/124] and 4.9% [7/144],
`respectively). Discontinuations because of insuffi-
`cient efficacy were low in both studies among
`patients in either treatment arm. Overall, 3 of 269
`(1.1%) patients discontinued treatment with doxy-
`cycline compared with 6 of 268 (2.2%) patients
`receiving placebo.
`
`Efficacy assessments
`Total inflammatory lesion counts. In both
`studies, patients in the ITT population who received
`active treatment demonstrated significantly greater
`reductions from baseline in total
`inflammatory
`lesions at week 16 compared with patients who
`received placebo. The mean change from baseline
`in total inflammatory lesions in the active-treatment
`group was e11.8 in study 301 and e9.5 in study 302
`compared with e5.9 and e4.3, respectively, in the
`placebo arms. These results were statistically signif-
`icant in both trials (P \.001). Figs 2 and 3 depict the
`reduction in total inflammatory lesion counts that
`was observed throughout the 16-week study period
`in both studies. There was a significantly greater
`decrease in lesion count
`in the active-treatment
`group when compared with the placebo group
`starting at the initial 3-week follow-up visit (P =
`.005) that continued at week 6 (P \ .001), week 12
`(P \.001), and week 16 (P \.001). To illustrate the
`
`

`

`J AM ACAD DERMATOL
`VOLUME 56, NUMBER 5
`
`Del Rosso et al 797
`
`Fig 2. Mean change from baseline in total inflammatory
`lesion count (papules 1 pustules 1 nodules) through
`week 16 in study 301.
`
`Fig 3. Mean change from baseline in total inflammatory
`lesion count (papules 1 pustules 1 nodules) through
`week 16 in study 302.
`
`clinical changes seen with use of anti-inflammatory
`dose doxycycline throughout
`the 16-week study
`period, Fig 4 shows the inflammatory lesions and
`facial erythema of a patient at baseline and then
`again at week 16.
`Total erythema scores. In study 301,
`the
`reduction from baseline in the mean total erythema
`score (defined as the CEA score) was significantly
`(P = .017) greater at week 16 in the active-treatment
`arm when compared with the placebo arm. At week
`16,
`the mean change from baseline in the total
`erythema score was e2.7 and e1.8 for the active-
`treatment and placebo groups, respectively (Fig 5).
`In study 302,
`the change from baseline in total
`erythema scores
`indicated that
`facial
`redness
`decreased in patients in the active-treatment group
`(Fig 6); however, the between-group difference did
`not reach statistical significance.
`Investigator global assessment. The active-
`treatment group demonstrated significantly greater
`improvement in IGA scores by study end point when
`compared with the placebo group in both studies.
`In study 301, 45.7% (58/127) of the patients in the
`
`Fig 4. Change in inflammatory lesions throughout the
`study period (A) at baseline in a male patient; (B) at week
`16 in this same patient.
`
`active-treatment group achieved a 2-point or greater
`improvement in IGA scores at week 16 compared
`with 25.8% (32/124) of the patients in the placebo
`group (P \ .001). In this same study, a significantly
`larger percentage of actively treated patients (30.7%
`[39/127]) achieved an IGA score of 0 (clear) or 1 (near
`clear) when compared with placebo-treated patients
`(19.4% [24/124], P = .036). In study 302, 22.5%
`(32/142) of the patients in the active-treatment arm
`achieved a 2-point or greater improvement in IGA
`
`

`

`798 Del Rosso et al
`
`J AM ACAD DERMATOL
`MAY 2007
`
`Table II. Pooled data of treatment-emergent
`adverse events reported from studies 301 and 302
`
`Adverse event*
`
`Nasopharyngitis
`Diarrhea
`Headache
`Upper respiratory
`tract infection
`Hypertension
`Sinusitis
`" AST
`Abdominal pain,
`upper
`Fungal infection
`Influenza
`Nausea
`
`Anti-inflammatory dose
`doxycycline
`(n = 269)
`No. (%)
`
`13 (4.8)
`12 (4.4)
`12 (4.4)
`9 (3.3)
`
`8 (2.9)
`7 (2.6)
`6 (2.2)
`5 (1.8)
`
`y
`5 (1.8)
`5 (1.8)
`5 (1.8)
`
`Placebo
`(n = 268)
`No. (%)
`
`9 (3.3)
`7 (2.6)
`16 (5.9)
`20 (7.4)
`
`2 (0.7)
`2 (0.7)
`2 (0.7)
`1 (0.3)
`
`1 (0.3)
`3 (1.1)
`8 (2.9)
`
`AST, Aspartate aminotransferase.
`*Reported adverse events not necessarily determined to be
`probably or possibly related to study drug.
`y
`There were no cases of vaginal candidiasis or photosensitivity
`in the active-treatment arm.
`
`identified during the course of either study, includ-
`ing assessments of reported AEs, vital signs, weight,
`and laboratory values. Table II lists pooled data of
`AEs reported during the 16-week treatment period in
`two or more patients in either treatment group (ITT
`analysis). No cases of photosensitivity were reported
`in either treatment group and no cases were sus-
`pected by investigators. Among female patients in
`both studies, vaginal mycotic infections, including
`candidiasis, were reported in 4 patients in the
`placebo group and in none of the patients in the
`active-treatment group.
`Adverse events. In study 301, 44.1% (56/127) and
`38.7% (48/124) of patients in the active-treatment
`and placebo groups,
`respectively,
`reported AEs
`over the 16-week study period. Most of these AEs
`were rated as mild or moderate in severity in both the
`active-treatment arm (82.1% [46/56]) and placebo
`arm (87.5% [42/48]). AEs considered by the investi-
`gator to be possibly or probably related to study drug
`were experienced by 19.7% (25/127) of patients in
`the active-treatment group and 13.7% (17/124) in the
`placebo group. In study 302, 65.5% (93/142) of the
`patients in the active-treatment arm and 51.4%
`(74/144) in the placebo arm noted AEs over the 16-
`week study period. The majority of these AEs were
`rated as mild or moderate in severity in both the active
`treatment group (93.5 % [87/93]) and placebo group
`(95.9% [71/74]). AEs judged by the investigator to be
`possibly or probably related to study drug were
`experienced by 21.8% (31/142) of patients in the
`
`Fig 5. Mean change from baseline in CEA score through
`week 16 in study 301.
`
`Fig 6. Mean change from baseline in CEA score through
`week 16 in study 302.
`
`scores at week 16 compared with 16.0% (23/144)
`of
`the patients in the placebo arm (P = .004).
`A significantly larger proportion of patients in the
`active-treatment arm in study 302 (14.8% [21/142])
`achieved an IGA score of 0 (clear) or 1 (near clear)
`when compared with patients in the placebo arm
`(6.3% [9/144], P = .012).
`Four-week posttherapy assessment. Of the
`patients who consented to participate in this study,
`those in the active-treatment arm maintained a
`greater overall treatment benefit through week 20.
`The mean total lesion count at week 20 was 10.3 in
`the active-treatment group and 15.3 in the placebo
`group, representing a mean treatment difference of 5
`lesions at 4 weeks after discontinuation of therapy.
`Significant differences in CEA and IGA scores within
`both study arms recorded at weeks 16 and 20 were
`not observed.
`Safety analysis. Both anti-inflammatory-dose
`doxycycline administered once daily and placebo
`were well tolerated throughout both studies 301
`and 302. No major safety issues or concerns were
`
`

`

`J AM ACAD DERMATOL
`VOLUME 56, NUMBER 5
`
`Del Rosso et al 799
`
`active-treatment group and 14.6% (21/144) in the
`placebo group. In the 4-week follow-up period from
`week 16 through week 20, AEs were experienced
`by 4.8% (4/84) of the patients initially randomized to
`the active-treatment arm and 9.2% (7/76) of patients
`initially randomized to the placebo arm.
`Vital
`signs/weight. Vital
`signs and weight
`assessments demonstrated only minimal mean
`changes from baseline with no apparent between-
`group differences in either study. No changes in
`blood pressure levels were considered to be AEs,
`except for one patient in the active-treatment group
`in study 302. This patient experienced a marked
`increase in blood pressure that was reported as an AE
`and was not considered to be related to study drug.
`Laboratory evaluations. In both trials, all ran-
`domized patients underwent hematology and serum
`chemistry panels at baseline and week 16. Overall,
`in both studies, there were no notable changes or
`emergent trends in abnormal laboratory values in
`either treatment group from baseline to end point in
`any hematologic or serum chemistry indices.
`
`DISCUSSION
`The results from the two phase III 16-week trials
`reviewed above demonstrate that anti-inflammatory
`dose doxycycline administered once daily appears to
`be effective and safe for the treatment of moderate
`to severe papulopustular rosacea. Both studies dem-
`onstrated the statistically significant superiority of
`anti-inflammatory dose doxycycline when compared
`with placebo after 16 weeks based on multiple
`efficacy end points,
`including reduction in facial
`inflammatory lesions as early as week 3, improve-
`ments in IGA scores ( $ 2 grade improvement), and
`proportion of patients rated as clear or near clear
`(dichotomized IGA). In both studies, actively treated
`patients achieved lower erythema scores than the
`placebo groups at weeks 12 and 16, thus displaying
`a trend toward improvement. The trend in erythema
`reduction in the active-treatment groups continued
`progressively through study end point without dem-
`onstration of a plateau effect (Figs 5 and 6).
`In addition to the results reported herein with
`anti-inflammatory dose doxycycline,
`large-scale,
`randomized, vehicle-controlled, phase III
`trials
`have been reported with topical metronidazole21
`and topical azelaic acid.22 Similar to the present
`studies of anti-inflammatory dose doxycycline, these
`topical therapy trials enrolled patients with mostly
`moderate to severe rosacea (the mean number of
`lesions was approximately 18 in the 3 trials, com-
`pared with approximately 20 lesions in the present
`studies). In the present studies,
`the mean total
`inflammatory lesion count decreased by 61% and
`
`46% in patients receiv