An Open-Label Pilot Study to Evaluate the Efficacy of Tofacitinib in Moderate to Severe Patch-Type Alopecia Areata, Totalis, and Universalis.

INTRODUCTION
AA is a common autoimmune disease, with a lifetime risk of approximately 2%,
affecting an estimated 5.3 million individuals in the US (Safavi et al), (McMichael
et al). Persistent moderate/severe AA causes significant disfigurement
and psychological distress to affected individuals (Colon et al). Clinical development of innovative therapies in AA has
lagged far behind other autoimmune conditions.
Alopecia areata results from autoimmune attack on the hair follicles. Using
comparative genomics of the transcriptional profiles of skin from both AA model
mice, and humans with AA, we have found that cytotoxic CD8 (+) NKG2D (+) T cells are
both necessary and sufficient for the induction of AA in mouse models of disease. On
the basis of our preclinical findings (Xing et
al), we initiated a Phase 2 efficacy signal-seeking clinical trial in
moderate to severe AA, assessing the clinical and immunopathological response to
treatment with oral tofacitinib, a JAK 1,3 inhibitor which also inhibits JAK 2.
Presently, tofacitinib is FDA-approved for the treatment of adult patients with
moderate-to-severe rheumatoid arthritis (RA) and is under study for many other
autoimmune conditions (Schwartz et al).
Tofacitinib has been shown to prevent the onset of, and reverse, AA in the C3H-HeJ
animal model of AA. Thus far, several studies have demonstrated clinical efficacy of
oral tofacitinib in treating patients with AA (Jabbari et al) or AU (alopecia universalis) (Kennedy et al), (Liu et
al). In all reported cases, clinical response was achieved with minimal
or no adverse events.

RESULTS

Primary Efficacy Endpoint
This study was an open-label, clinical trial to investigate tofacitinib 5
mg to 10 mg PO twice daily in the treatment of moderate/severe AA.
Eight out of twelve patients met the study’s primary efficacy
endpoint of 50% or greater hair re-growth from baseline as assessed by SALT
index at the end of treatment. The duration of treatment ranged from 6 to 18
months, at the discretion of the investigator and dependent on the individual
subject’s response as well as safety considerations. The length of time
from baseline for patients to reach primary efficacy endpoint was on average 32
weeks, with the time period ranging from as little as 8 weeks to as much as 64
weeks (Figure 1), (Figure 2), (Table
1).
Four out of the five patients with either AT or AU successfully achieved
50% or greater response in hair regrowth. All patients who had reached the
study’s primary efficacy endpoint, with the exception of one patient,
failed to respond to lower doses, in some cases despite prolonged treatment, but
responded with onset of regrowth within 4 weeks of initiation of the higher dose
of tofacitinib, at 10 mg BID.

Secondary Efficacy Outcomes
As secondary endpoints, efficacy was measured by changes in hair
re-growth as a continuous variable as determined by physical exam and Canfield
photography, as well as patient and physician global evaluation scores.

Global overall improvement in SALT score at end of treatment
Eleven out of twelve patients attained a global overall improvement
in SALT score at the end of treatment with results ranging from 12.1% to
100% regrowth, with an average 56.8% regrowth. Baseline SALT scores for the
twelve patients ranged from 46% to 100% and at the end of treatment SALT
scores ranged from 0% to 99%. The average baseline SALT score of 81.3%
decreased to 40.8% at the end of treatment. Only one subject had no response
to the study medication after 36 weeks of administration, having experienced
a negligible decrease in SALT score with approximately 1% hair regrowth that
consisted of 0.5–1mm depigmented fine terminal hairs throughout the
scalp and facial area. Vellus hair growth was not used in SALT score
calculations. No patients experienced worsening of AA from baseline at the
time of treatment discontinuation, with eleven patients exhibiting varying
degrees of hair regrowth.

Time to Regrowth of Scalp and Body Hair
Regrowth was seen in responders as soon as four weeks after the
effective dose of study medication was initiated. All twelve patients,
experienced between 0 and less than 25% of hair regrowth by week 4 as
assessed by physician global assessment (PGA), a static evaluation of scalp
regrowth rated as “worse”, “same”, or
“improved”. The degrees to which hair regrowth presented
itself at 4 weeks were highly varied and individual responses ranged from
less than 1% with the introduction of few fine terminal hairs to at most
approximately 45% regrowth with depigmented and pigmented, terminal and fine
terminal hairs. Three patients displayed mild shedding of scalp hair while
on tofacitinib, but at end of treatment, remained improved compared to
baseline.
All 12 enrolled patients exhibited varying degrees of body hair
regrowth. Regrowth of body hair was documented as soon as four weeks after
effective dose of study medication was initiated. Body hair regrowth was
mixed, with some patients experiencing minimal to full facial hair regrowth,
including eyelashes and eyebrows. Other areas of body hair regrowth noted in
patients included the arms, legs, axillary and groin area.

Durability of Responses
To assess the durability of responses, patients who achieved 50%
regrowth from baseline during the first 6 to 18 months, were followed for an
additional 6 months off treatment or until it was determined that relapse
had occurred. Of the eight patients who achieved 50% regrowth, one patient
dropped out of the observation period in order to continue the medication
outside of the study. Of the seven patients who were followed
observationally, six patients exhibited variable hair shedding after
completion of the study treatment, with two patients showing initial signs
of shedding approximately 3–4 weeks after end of treatment, and four
patients showing initial signs of shedding approximately 8 weeks after end
of treatment. Hair shedding was initially slight, but accelerated at
4–6 months off tofacitinib. The final patient did not exhibit any
hair shedding throughout the observational period (24 weeks / 6 months off
tofacitinib). Excellent durability of response was seen in three out of the
eight responders, maintaining lower SALT scores as compared to baseline SALT
scores at nearly 24 weeks off tofacitinib. Four patients experienced
worsening of AA compared to baseline at the conclusion of the study at
nearly 24 weeks off the study medication. Shedding of body hair coincided
with the timeline of scalp hair loss.
Overall, eleven out of the twelve patients who were initially
enrolled in this study completed the intended 24 to 72 weeks of study
treatment. One patient underwent early termination of the study treatment at
week 12 due to experiencing hypertensive urgency as an adverse event.

Change in Patient Quality of Life Assessment
Change in patient quality of life assessment was compared from
baseline to selected visits during the treatment period (Weeks 12 and 24).
Quality of life measures were based on changes in the Dermatology Life
Quality Index (DLQI) (Basra et al).
Seven out of the twelve patients experienced a decrease in their DLQI score
as measured from baseline to week 24 (see table 4). The mean baseline DLQI
score of 6.5 ± 5 decreased to 5.2 ± 6.7 at 3 months of
treatment and later increased to 6 ± 6.9 at the end of 6 months of
treatment.

Differences in regrowth between patients with patch type AA vs patients
with alopecia totalis or universalis
At the end of treatment, the five subjects who were either AT or AU
had experienced hair regrowth ranging from 1.0% to 84%, with an average of
52.2% regrowth. This is in comparison to subjects with moderate to severe
patchy AA who at the end of treatment, experienced hair regrowth ranging
from 12.1% to 100%, with an average 52.1% regrowth. Overall, patients with
either patch type AA or AT or AU had, on average, very similar percentage
hair regrowth at the end of study treatment. Given the small sample of these
patients, it is not known if the observation of similar regrowth rates among
AA, AT, and AU patients will continue to hold in future studies.

Biomarker and Clinical Correlative Studies
Gene expression profiling was performed on skin biopsies taken at
baseline and up to 24 weeks of treatment, with additional optional biopsies
performed if indicated by clinical considerations.
We applied both naïve and supervised clustering to this dataset
in order to assess two features: the overall molecular effect of tofacitinib
treatment on patient samples, and the concordant molecular response of the
disease. The former was assessed by an unbiased, unsupervised differential
expression analysis and the latter was measured by response to previously
published gene expression signatures defining AA pathology. To ensure parity of
the data, for this analysis we only included patient samples that had matched
pre-treatment, Tx-00, and 24 weeks of treatment, Tx-24, samples.
Figure 3A represents the overall
molecular impact of tofacitinib treatments observed in the patient biopsies.
Molecular response in this instance was defined as molecular divergence between
patient samples taken at pre-treatment and >24 weeks of treatment.
Unsupervised clustering was able to produce a biomarker panel that robustly
segregated pre- and post-treatment samples with no cross-over in the clustering
(4A dendrogram). The gene list that comprises this signature is available in
supplemental
data.
These molecular effects coincide with significant shifts in ALADIN score
in most patients (4B heatmap). In this instance, unaffected control patient
biopsies were included as a reference, (indicated as AAB-N-Cx). All initial
patient biopsies cluster away from unaffected controls using ALADIN. Treated
patient samples that cluster with the control samples indicate significant
suppression of AA molecular pathology and suggest response to tofacitinib, while
patient samples that remain clustered with pre-treated samples suggest
resistance or recalcitrance to treatment. This is more simply represented in the
3D ALADIN plot. Prior to treatment, patients start at a steady-state with
significant distance from healthy, unaffected samples (T0 vs NC, normal
control). During treatment, patients that respond will shift in this space
towards the NC samples (T24). One patient (a clinical nonresponder) showed
virtually no molecular shift following treatment (T0 and T24 datapoints are
proximal to each other). The pharmacodynamic reduction in ALADIN score during
treatment in the favorable responders and lack thereof in the patients in whom
there was little or no clinical response, confirm the utility of the ALADIN
scores as a physiological relevant dynamic biomarker that positively correlates
with clinical response.

Adverse Events
In this study, an Adverse Event (AE) is defined as any new untoward
medical occurrence or worsening of a pre-existing medical condition in a patient
or clinical investigation subject administered an investigational (medicinal)
product and may or may not have a causal relationship with this treatment.
Specific frequencies for adverse events are noted in (Table 2). Of note, tofacitinib was discontinued for
1 patient with a history of hypertension who experience a hypertensive urgency
while on the study medication.
Clinical laboratory evaluation was performed on all subjects during the
screening period, at baseline and as otherwise deemed necessary to monitor for
abnormal values and for normalization of those values. Clinical laboratory
evaluation consisted of complete blood count, basic metabolic profile, hepatic
panel and urinalysis with microscopic exam, hepatitis B and C screening panel,
HIV test, fasting lipid profile, TB and serum pregnancy test for women of child
bearing potential. No patient met protocol-defined clinical laboratory
discontinuation criteria. Overall, there were no major abnormalities in
laboratory evaluations. The laboratory adverse events that were observed are
listed in Table 2. Notably, tofacitinib
was discontinued for 1 patient after experiencing persistent 1+ blood on
urinalysis. The patients primary care doctor was concerned and requested that we
discontinue the study drug.

Discussion
Tofacitinib is known to effectively treat rheumatoid arthritis by modulating
the interferon response inflammatory pathway by inhibition of JAK1/JAK3 (Keisuke et al). AA and RA share the same
interferon-gamma response pathway, which provided the rationale for selecting
tofacitinib for evaluation in AA (Xing et
al). Already, tofacitinib has been shown to prevent the onset of, and
reverse, AA in the C3H-HeJ animal model of AA (Xing
et al).
Tofacitinib was effective in achieving a global overall improvement in SALT
score at the end of treatment for eleven out of twelve patients. The effective dose
for the majority of responders was 10mg BID. Only 1 subject had a full response to
5mg BID. In contrast to previous work on this subject (Kennedy et al) the majority of our patients did not
respond to 5mg BID after at least 1 month of therapy. Some patients remained on 5mg
BID for 4 to 6 months prior to dose escalation yet had absolutely no hair growth.
However, once the dose was increased to 10mg BID, regrowth onset was seen within one
month. Further studies are needed to explore the difference in response between our
studies versus prior reported studies. Baseline SALT scores for the twelve patients
ranged from 46% to 100% and at the end of treatment SALT scores ranged from 0% to
99%. The mean baseline SALT score of 81.3% decreased to 40.8% at the end of
treatment. Overall, patients with either patch type AA, or AT or AU had on average
very similar percentage hair regrowth at the end of study treatment. This suggests
that the severity of disease may not be the determining factor for response versus
no response. It may be possible that a long duration of current episode of AA
decreases the probability of therapeutic response, but based on our observations
longer duration of disease does not appear to preclude response to therapy.
Tofacitinib was well tolerated in all twelve patients. One patient
discontinued the study due to hypertension. There were no reported serious adverse
events. Observed adverse effects were infrequent and clinical laboratory
abnormalities were uncommon. This study adds to the existing literature supporting
the potential of Janus kinase inhibitors to halt hair loss and allow regrowth in
some patients with alopecia areata. Although this therapy is not a cure, as
evidenced by relapse of hair loss after treatment ended, tofacitinib may potentially
be a therapeutic option for the treatment of hair loss in some AA patients. Further
work is needed to elucidate optimal treatment strategies for maintenance of response
and minimization of risks. Limitations in this study consist of those inherent in an
open label study. This includes observer bias as a result of physicians and study
participants being un-blinded during treatment. To limit this bias, objective tools,
such as SALT scoring, and photography were used to measure the primary and secondary
efficacy of treatment. Another limitation of this study is the small sample size.
This decreases the study’s external validity and as a result may not be
generalizable to the population of interest i.e. patients with moderate to severe
AA.

MATERIALS AND METHODS

Study Design, Oversight, and Participants
The study was conceived and conducted by the investigative team at
Columbia University. This study was conducted in accordance with Good Clinical
Practice (GCP), as defined by the International Conference on Harmonization
(ICH) and in accordance with the ethical principles underlying European Union
Directive 2001/20/EC and the United States Code of Federal Regulations, Title
21, Part 50 (21CFR50). Monitoring for regulatory compliance and adherence to the
IRB approved protocol was performed by the Columbia University Clinical Trials
Office and the Department of Surgery Regulatory Team. The study was registered
on clinicaltrials.gov prior to initiation. IRB
approval was granted for all experiments performed in this study and written
informed patient consent was obtained for all experiments, procedures, and
treatments involved in this study. In addition, permission was granted from
patients for the publication of their photographs. This was an open-label pilot
study of tofacitinib 5mg BID to 10mg orally BID, for 6 to 18 months in the
treatment of moderate to severe AA, and AT or AU, followed by 6 months follow-up
off drug to assess for delayed response to treatment and/or the incidence and
timing of recurrence of disease. The initial treatment dose was tofacitinib 5mg
PO BID for at least 1 month, which was increased to 10mg + 5mg QD for at least 1
month, and then to 10mg PO BID if the patient had absence of any terminal hair
regrowth on the scalp. End of treatment was defined as at least 6–12
months on a dose that appeared to be having a positive response i.e. an
effective dose or ending of treatment if no regrowth occurred within 3 months of
the highest tolerable dose, and at least 6 months total tofacitinib treatment. A
full course of treatment was defined as at least 6–12 months on a dose
that appeared to be having a positive response i.e. an effective dose or ending
of treatment if no regrowth occurred within 3 months of the highest tolerable
dose and at least 6 months total tofacitinib treatment. We enrolled 12 adult
patients, including 7 patients with moderate to severe AA (30–95% hair
loss) and 5 patients with alopecia totalis (AT) or AU.

Study Assessments and Outcomes
The study’s primary efficacy endpoint was the
proportion of responders at the end of treatment, (6 to 18 months of treatment),
with response defined as 50% or greater hair re-growth from baseline as assessed
by the Severity of Alopecia Tool (SALT) score, a standardized, validated method
for estimating hair loss in AA (Olsen et
al). Secondary efficacy endpoints included hair re-growth as a
continuous variable as determined by physical exam and Canfield photography, as
well as patient and physician global evaluation scores. Additionally, Quality of
Life measures (Dermatology Quality of Life Index – DLQI) were done at
regular pre-specified intervals. To assess the durability of responses, patients
were followed for an additional 6 months after treatment was completed.
Additionally, safety analysis was included as a secondary endpoint for all
subjects who received at least one dose of tofacitinib and was monitored at
monthly visits.

Biomarker Studies
We were able to accurately recapitulate the molecular assessment of
patient recovery during treatment (see figure
3) using gene expression and the ALADIN signature. Moreover, we were
able to leverage this data to construct biomarkers corresponding to patient
response to treatment and tofacitinib mechanism-of-action (MoA) (see figure 3). Rather than being derived from the
pathogenic definitions of the disease, this signature is specifically derived to
measure patients’ individual molecular responses in their scalp skin
biopsies as a function of tofacitinib treatment over time, facilitated by the
sequential time points. Statistically significant overlap of this signature and
the pathogenic signature provides a quantitative framework for the a
priori prediction of drug efficacy (Petukhova et al). Furthermore, these quantitative molecular profiles
can be used to define the molecular regulatory mechanism of response to
tofacitinib using network analysis (Chen et
al), (Margolin et al), and to
predict non-responders a priori - patients that exhibit low
phenotypic response to treatment also show incomplete or impartial molecular
response based on the tofacitinib MoA exhibited in responder patients, (see
figure 3).

Supplementary Material
1