J AM ACAD DERMATOL
VOLUME 76, NUMBER 4
UV exposure. This may have previously overlooked
clinical implications, particularly for patients with
photosensitive dermatoses.
Our findings show the variability in spectral
irradiances among light sources found in daily
ambient indoor surroundings, highlighting the
concern that the ever-increasing artificial light may
have overt or clandestine effects on skin biology.
Teo Soleymani, MD,a Lorcan M. Folan, PhD,b
Nicholas A. Soter, MD,a Nada Elbuluk, MD,
MSc,a and David E. Cohen, MD, MPHa
The Ronald O. Perelman Department of
Dermatology,a New York University School of
Medicine, and The Tandon School of
Engineering,b New York University, New York,
New York.
Funding sources: None.
Conflicts of interest: None declared.
Reprint requests: Teo Soleymani, MD, The Ronald
O. Perelman Department of Dermatology, NYU
Langone Medical Center, New York University
School of Medicine, 240 E 38th St, 11th fl, New
York, NY 10016.
E-mail: Teo.Soleymani@gmail.com
http://dx.doi.org/10.1016/j.jaad.2016.11.016
Antimalarial therapy for granuloma
annulare: Results of a retrospective
analysis
To the Editor: Granuloma annulare (GA) is a rela-
tively common granulomatous skin disease that most
often presents as erythematous annular plaques with
central clearing. Lesions may be localized (localized
GA), widespread (generalized GA), or present as
subcutaneous nodules (subcutaneous GA).
GA is benign and largely asymptomatic. However,
its epidemiology, comorbidities, and treatment pro-
tocols are poorly defined.1 Published work often
lacks comprehensive evaluation of GA cohorts, his-
tologic confirmation,1,2 or combines GA variants.
Whereas antimalarials have been proposed in the
treatment of GA, reports of their use are scarce, with
the largest studies conducted decades ago.1
To address these limitations, we retrospectively
reviewed charts of all GA patients seen within the
granulomatous skin disease clinic at the University of
Pennsylvania (Penn) from 2009 to 2016 (by M.R.).
Initial search criteria within PennSeek, a tool allow-
ing querying of Penns electronic medical records,
required that patients either 1) received an
Research Letters 765
International Classification of Diseases code for
GA or 2) had GA or granuloma annulare written
in their chart. This yielded 63 patients. A manual
chart review identified 35 patients who were ulti-
mately diagnosed with GA. All cases were histolog-
ically confirmed.
For each patient, the following information was
retrieved (Table I): demographics, GA clinical
variant, antimalarial response, and the presence of
comorbidities previously reported in association
with GA. Treatment response was determined from
clinical notes.
Table I describes the characteristics of our GA
cohort. Eighty percent were female; the mean age at
diagnosis was 54 years (range, 28-79 years). More
than three-quarters (77.1%) of patients had general-
ized GA (n 27), 17.1% had localized GA (n 6),
and 5.7% had subcutaneous GA (n 2). Five patients
had features of annular elastolytic giant cell
granuloma.
The prevalence of diabetes, dyslipidemia, and
hypothyroidism is the United States is estimated at
17.5% ( for patients 45-64 years of age),3 29.4% ( for
patients 55-64 years of age without cardiovascular
disease),4 and 6.1% ( for patients 50-59 years of age),5
respectively. In contrast, our GA cohort was of a
similar age group, yet had a higher prevalence of
these comorbidities. More than half (63.2%) had
some form of glucose intolerance; 31.6% were
diabetic. Eighty percent had dyslipidemia and 37%
had thyroid disease. Ten of 18 patients (55.6%)
improved on hydroxychloroquine (Fig 1). Nearly
all patients prescribed hydroxychloroquine had
previously failed or experienced inadequate
response from alternative GA treatments, the most
common of which were topical steroids (n 12),
intralesional triamcinolone acetonide (n 10),
topical tacrolimus (n 5), and minocycline
(n 5). All 6 patients prescribed chloroquine
improved (100%), 5 of whom had earlier failed
treatment with hydroxychloroquine. The average
treatment duration with hydroxychloroquine and
chloroquine before noting improvement was 3.6
and 3 months, respectively.
Our study is limited by its small size, the lack
of standardized tools available to assess GA
severity and treatment efficacy, and its retrospec-
tive cross-sectional nature. To our knowledge, this
is the largest cohort of GA patients treated with
hydroxychloroquine, and further emphasizes the
potential association of GA with thyroid disease.1
Notably, our tertiary referral cohort included
predominantly patients with generalized GA,
while 70% of GA cases are localized in general
practice.1
766 Research Letters J AM ACAD DERMATOL
APRIL 2017

Table I. Granuloma annulare cohort characteristics: Demographics, comorbidities, and response to

antimalarials
Antimalarial response Demographics Comorbid conditions*
Mean
age at GA Glucose Thyroid
y z
n (%) Hydroxychloroquine Chloroquine diagnosis, y Female intolerance Dyslipidemia disease HCV
x
n 35 18 6 35 35 19 25 27 13
GA overall 35 (100) 10 (55.6) 6 (100) 54 28 (80) DM: 6 (31.6); 20k (80) 10{ (37) 2 (15.4)
pre-DM: 6 (31.6)
nx 2 1 6 6 4 4 4 3
Localized 6 (17.1) 1 (50) 1 (100) 54.5 4 (66) DM: 2 (50); 4 (100) 2 (50) 0 (0)
GA# pre-DM: 1 (25)
nx 14 4 27 27 14 20 21 10
Generalized 27 (77.1) 9 (64.3) 4 (100) 54.3 22 (81.5) DM: 4 (28.6); 16 (80) 8 (38.1) 2 (20)
GA** pre-DM: 5 (35.7)
nx 2 1 2 2 1 1 2 0
Subcutaneous 2 (5.7) 0 1 (100) 50.5 2 (100) DM: (0); 0 (0) 0 (0) N/A
GA pre-DM: 0 (0)

DM, Diabetes mellitus; GA, granuloma annulare; HCV, hepatitis C virus.

*Columns for HIV, hepatitis B virus, serum protein electrophoresis, and urine protein electrophoresis are excluded because no cases of HIV,
hepatitis B virus, or paraproteinemias were detected in patients who underwent testing.
y
Four hundred mg daily (maximum 6.5 mg/kg ideal body weight); some patients were treated with 200-300 mg/daily. Of the 18 patients
prescribed hydroxychloroquine, 11 received 400 mg/daily, 3 received 300 mg/daily, and 4 received 200 mg/daily. Three of the 18 patients
prescribed hydroxychloroquine chose to discontinue because of diarrhea (n 2) and hair loss (n 1). A fourth patient reported mild
gastrointestinal symptoms that eventually self-resolved.
z
Two hundred fifty mg/day either 5 days a week (n 5) or 3 days a week (n 1). Of the 6 patients prescribed chloroquine, 1 discontinued
because of development of bronchiolitis.
x
n reflects the denominator for each calculation. In order to be included in prevalence calculations for comorbidities, a given patient with GA
must have either (1) received a diagnostic code for the particular comorbidity OR (2) been screened at least once for the relevant comorbid
condition.
k
Of 20 patients with dyslipidemia, the number of patients with pure hypercholesterolemia, pure hypertriglyceridemia, and mixed
dyslipidemia were 16, 2, and 2, respectively. Sixteen of the patients with dyslipidemia were also assessed for glucose intolerance. Of these 16
patients, 5 had diabetes, 6 were prediabetic, and 5 had isolated dyslipidemia.
{
All 10 patients with thyroid disease were hypothyroid. Three were hypothyroid as a result of thyroidectomies performed for various
indications ( follicular adenoma, benign thyroid cyst, and papillary thyroid cancer). Of the remaining 7 cases, the etiology was only noted for
1 patient, who had Hashimoto disease.
#
Two of the 6 patients with localized GA (33%) had features of annular elastolytic giant cell granuloma.
**Three of the 27 patients with generalized GA (11.1%) had features of annular elastolytic giant cell granuloma.

Fig 1. Granuloma annulare. Erythematous papules and confluent, vaguely annular plaques on
the dorsal surface of the hand of a patient with generalized granuloma annulare. Note complete
clearance of all lesions 6 months after hydroxychloroquine therapy (400 mg daily).
J AM ACAD DERMATOL
VOLUME 76, NUMBER 4
Well-controlled population-based cohort studies
are needed to investigate potential associations
between GA and comorbidities, particularly
thyroid disease. Dermatologists should consider
antimalarials as first-line treatment for genera-
lized GA.
Sungat K. Grewal, BS, Courtney Rubin, MD, and
Misha Rosenbach, MD
Department of Dermatology, Perelman School of
Medicine at the University of Pennsylvania,
Philadelphia, Pennsylvania
Dr Rosenbach has received support from the
Dermatology Foundations Medical Dermatology
Career Development Award.
Conflicts of interest: None declared.
Reprints not available from the authors.
Correspondence to: Misha Rosenbach, MD, Depart-
ment of Dermatology, University of Pennsylva-
nia, 2 Maloney Bldg, 3600 Spruce St,
Philadelphia, PA 19104-4208
E-mail: misha.rosenbach@uphs.upenn.edu
REFERENCES
1. Piette EW, Rosenbach M. Granuloma annulare: pathogenesis,
disease associations and triggers, and therapeutic options. J
Am Acad Dermatol. 2016;75:467-479.
2. Piette EW, Rosenbach M. Granuloma annulare: clinical and
histologic variants, epidemiology, and genetics. J Am Acad
Dermatol. 2016;75:457-465.
3. Menke A, Casagrande S, Geiss L, Cowie CC. Prevalence of and
trends in diabetes among adults in the United States,
1988-2012. JAMA. 2015;314:1021-1029.
4. Goff DC Jr, Bertoni AG, Kramer H, et al. Dyslipidemia
prevalence, treatment, and control in the Multi-Ethnic Study
of Atherosclerosis (MESA): gender, ethnicity, and coronary
artery calcium. Circulation. 2006;113:647-656.
5. Hollowell JG, Staehling NW, Flanders WD, et al. Serum TSH,
T(4), and thyroid antibodies in the United States population
(1988 to 1994): National Health and Nutrition Examination
Survey (NHANES III). J Clin Endocrinol Metab. 2002;87:489-499.
http://dx.doi.org/10.1016/j.jaad.2016.11.044
Practical management of the adverse
effects of Hedgehog pathway inhibitor
therapy for basal cell carcinoma
To the Editor: The Hedgehog pathway inhibitors
(HPIs), sonidegib and vismodegib, are indicated for
locally advanced BCC or metastatic BCC but have a
high incidence of adverse effects (AE) resulting in
therapy interruption or discontinuation. We conduct-
ed a literature review according to the Preferred
Items for Systematic Review and Meta-Analysis
(PRISMA) guidelines on the management of AEs to
Research Letters 767
improve knowledge gaps in therapy. Combined with
our own 5-year experience,1 we identified the major
AEs leading to management challenges: alopecia,
dysgeusia, weight loss, elevated creatine phospho-
kinase (CPK), muscle spasms, ovarian dysfunction,
and new onset squamous cell carcinoma.
Myopathies and elevations of CPK are the most
common AEs. Ally et al,2 in a small clinical trial
(n 17), noted that patients experienced a modest
improvement in symptoms within 2 weeks of amlo-
dipine treatment. A study is in progress to evaluate
levocarnitine in alleviating HPI-associated muscle
spasms (NCT01893892). Elevated CPK has been
described for both sonidegib and vismodegib, with
monitoring for required for sonidegib but not
vismodegib. Significantly, CPK levels were not
monitored during vismodegib trials. HPI patients at
risk for CPK abnormalities or myopathies, such as
those concurrently using a statin, may benefit from
periodic monitoring.
Alopecia is the second most common AE. Unlike
with chemotherapy, alopecia usually involves less
than 50% of the scalp.3 MacDonald et al3 recommend
concealment measures and minoxidil 2-5%
continued for 6 months after therapy.
Dysgeusia is the third most common AE. Le
Moigne et al4 demonstrated a benefit of early
nutritional screening for those taking vismodegib,
with less weight loss in the intervention group that
received nutritional counseling. Significantly, a
greater than 5% weight loss can occur without
dysgeusia, possibly due to diarrhea, nausea, xero-
stomia, oral thrush, or abdominal pain.4 Addressing
oral hygiene, reflux disease, low-grade oral infec-
tion, and postnasal drip are reported to improve taste
sensation in non-HPI patients.3
While patients who undergo HPI therapy are
generally older adults, premenopausal women of
childbearing age may experience amenorrhea likely
due to reversible follicle-stimulating hormone
receptor inhibition.1 It is essential to counsel
primary care physicians on both the teratogenicity
and sequelae of ovarian failure because non-
dermatologists are largely unfamiliar with the side
effects of HPIs. As use of HPIs for basal cell nevus
syndrome expands to a younger population, a
larger number of patients may experience
amenorrhea.
Recently, new onset squamous cell carcinomas
have been reported during HPI therapy. Zhu et al5
recommend that a new or persistent ulceration,
nodule or erythema in the BCC tumor site should
be biopsied if present after 3 months of HPI therapy.
We recommend that total body skin exams be
conducted during therapy and non-BCC lesions