ArticlePDF Available

Anti-Tumour Necrosis Factor Therapy for Dupuytren's Disease: A Randomised Dose Response Proof of Concept Phase 2a Clinical Trial

July 2018
EBioMedicine 33

DOI:10.1016/j.ebiom.2018.06.022

License
CC BY 4.0

Authors:

Jagdeep Nanchahal

University of Oxford

Catherine E Ball

University of Oxford

Lynn Williams

University of Oxford

Show all 17 authorsHide

Background: Dupuytren's disease is a common fibrotic condition of the hand that causes irreversible flexion contractures of the fingers, with no approved therapy for early stage disease. Our previous analysis of surgically-excised tissue defined tumour necrosis factor (TNF) as a potential therapeutic target. Here we assessed the efficacy of injecting nodules of Dupuytren's disease with a TNF inhibitor. Methods: Patients were randomised to receive adalimumab on one occasion in dose cohorts of 15 mg in 0.3 ml, 35 mg in 0.7 ml, or 40 mg in 0.4 ml, or an equivalent volume of placebo in a 3:1 ratio. Two weeks later the injected tissue was surgically excised and analysed. The primary outcome measure was levels of mRNA expression for α-smooth muscle actin (ACTA2). Secondary outcomes included levels of α-SMA and collagen proteins. The trial was registered with ClinicalTrial.gov (NCT03180957) and the EudraCT (2015-001780-40). Findings: We recruited 28 patients, 8 assigned to the 15 mg, 12 to the 35 mg and 8 to the 40 mg adalimumab cohorts. There was no change in mRNA levels for ACTA2, COL1A1, COL3A1 and CDH11. Levels of α-SMA protein expression in patients treated with 40 mg adalimumab (1.09 ± 0.09 ng per μg of total protein) were significantly lower (p = 0.006) compared to placebo treated patients (1.51 ± 0.09 ng/μg). The levels of procollagen type I protein expression were also significantly lower (p < 0.019) in the sub group treated with 40 mg adalimumab (474 ± 84 pg/μg total protein) compared with placebo (817 ± 78 pg/μg). There were two serious adverse events, both considered unrelated to the study drug. Interpretation: In this dose-ranging study, injection of 40 mg of adalimumab in 0.4 ml resulted in down regulation of the myofibroblast phenotype as evidenced by reduction in expression of α-SMA and type I procollagen proteins at 2 weeks. These data form the basis of an ongoing phase 2b clinical trial assessing the efficacy of intranodular injection of 40 mg adalimumab in 0.4 ml compared to an equivalent volume of placebo in patients with early stage Dupuytren's disease. Funding: Health Innovation Challenge Fund (Wellcome Trust and Department of Health) and 180 Therapeutics LP.

Trial profile.

…

Box and whiskers plot of log RNA concentration by treatment received. The figure is divided into separate plots for each gene assessed. The box represents the inter-quartile range (IQR) and whiskers extend to 1.5 relevant IQR (Tukey boxplot). The x-axis shows individual patients, grouped by treatment received, with the three repeat measures performed for each patient represented by points stacked within the same column.

…

Box and whiskers plot of ?-SMA protein concentration by treatment received. The box represents the inter-quartile range (IQR), the horizontal line represents the median and whiskers extend to 1.5 relevant IQR. The x-axis shows individual patients, grouped by treatment received. Samples from each patient were analysed in duplicate on three separate plates and the values for each plate are shown as circles, triangles or squares.

…

Box and whiskers plot of pro-collagen protein concentration by treatment received. The box represents the inter-quartile range (IQR), the horizontal line represents the median and whiskers extend to 1.5 relevant IQR (Tukey boxplot). The x-axis shows individual patients, grouped by treatment received, with the three repeat measures performed for each patient represented by points stacked within the same column. The plot includes pro-collagen concentrations reported below the minimum detectable range as zero, 3 of 9 in 35 mg and 2 of 6 in 40 mg adalimumab cohorts.

…

characteristics for treatment groups. Age of onset of Dupuytren's disease and age on entry into trial shown by treatment group. Patients receiving placebo are pooled.

…

Figures - uploaded by Jagdeep Nanchahal

Content may be subject to copyright.

Content uploaded by Jagdeep Nanchahal

Content may be subject to copyright.

Available via license: CC BY 4.0

Content may be subject to copyright.

Research Paper

Anti-Tumour Necrosis Factor Therapy for Dupuytren's Disease:

A Randomised Dose Response Proof of Concept Phase 2a Clinical Trial

Jagdeep Nanchahal

⁎, Catherine Ball

, Dominique Davidson

,LynnWilliams

, William Sones

Fiona E. McCann

,MarisaCabrita

, Jennifer Swettenham

,NeilJ.Cahoon

, Bethan Copsey

, E. Anne Francis

Peter C. Taylor

, Joanna Black

, Vicki S. Barber

, Susan Dutton

,MarcFeldmann

, Sarah E. Lamb

Kennedy Institute of Rheumatology, Nufﬁeld Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Oxford, UK

Edinburgh Department of Plastic Surgery, St John's Hospital, Livingston, UK

Oxford Clinical Trials Research Unit, Centre for Statistics in Medicine, Nufﬁeld Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Oxford, UK

abstractarticle info

Article history:

Received 26 April 2018

Received in revised form 8 June 2018

Accepted 19 June 2018

Available online 6 July 2018

Background: Dupuytren's disease is a common ﬁbrotic condition of the hand that causes irreversible ﬂexion con-

tractures of the ﬁngers, with no approved therapy for early stage disease. Our previous analysis of surgically-

excised tissue deﬁned tumour necrosis factor (TNF) as a potential therapeutic target. Here we assessed the

efﬁcacy of injecting nodules of Dupuytren's disease with a TNF inhibitor.

Methods: Patients were randomised to receive adalimumab on one occasion in dose cohorts of 15 mg in 0.3 ml,

35 mg in 0.7 ml, or 40 mg in 0.4 ml, or an equivalent volumeof placebo in a 3:1 ratio. Two weekslater the injected

tissue was surgically excised and analysed. The primary outcome measure was levels of mRNA expression for α-

smooth muscle actin (ACTA2). Secondary outcomesincluded levels of α-SMAand collagen proteins.The trial was

registered with ClinicalTrial.gov (NCT03180957) and the EudraCT (2015-001780-40).

Findings:We recruited 28 patients, 8 assigned to the 15 mg, 12 to the 35 mg and 8 to the 40 mg adalimumab cohorts.

There was no change in mRNA levels for ACTA2, COL1A1, COL3A1 and CDH11. Levels of α-SMA protein expression in

patients treated with 40 mg adalimumab (1.09 ± 0.09 ng per μg of total protein) were signiﬁcantly lower (p=

0.006) compared to placebo treatedpatients(1.51±0.09ng/μg). The levels of procollagen type I protein expression

were also signiﬁcantly lower (pb0.019) in the sub group treated with 40 mg adalimumab (474 ± 84 pg/μgtotal

protein) compared with placebo (817 ± 78 pg/μg). There were two serious adverse events, both considered unre-

lated to the study drug.

Interpretation: In this dose-ranging study, injection of 40 mg of adalimumab in 0.4 ml resultedindownregulationof

the myoﬁbroblast phenotype as evidenced by reduction in expression of α-SMA and type I procollagen proteins at

2 weeks. These data form the basis of an ongoing phase 2b clinical trial assessing the efﬁcacy of intranodular injec-

tion of 40 mg adalimumab in 0.4 ml compared to an equivalent volume of placebo in patients with early stage

Dupuytren's disease.

Funding: Health Innovation Challenge Fund (Wellcome Trust and Department of Health) and 180 Therapeutics LP.

(http://creativecommons.org/licenses/by/4.0/).

Keywords:

Dupuyten's disease

Anti-TNF

Fibrosis

Adalimumab

Myoﬁbroblast

1. Introduction

Dupuytren's disease (DD) is a common ﬁbrotic disease conﬁned to

the hand that affects approximately 4% of the general UK and US popu-

lations [1]. The early stages of the disease are manifest as nodules that

are typically quiescent for a period and thenbecome active, progressing

to cords and ﬂexion deformities of the ﬁngers in approximately 50% of

patients [2] and result in loss of hand function [3]. Whilst the mainstay

of treatment remainssurgical excision (fasciectomy) of the diseased tis-

sue or cords [4], approximately 40% of patients in the USA are treated by

disruption of the cords using collagenase or needle fasciotomy [5].

Generally patients undergo these treatments when digits are ﬂexed to

30° or more and hand function is impaired [6]. The recurrence rate

following surgery is 21% within 5 years [7] and these individuals may

require more extensive surgery involving excision of the diseased tissue

and overlying skin (dermofasciectomy). Post-operatively, some pa-

tients require prolonged hand therapy and splintage. Complications

occur in approximately 20% of patients undergoing surgery [8]. Alterna-

tive, less invasive techniques to disrupt the cords with a needle orcolla-

genase digestion are associated with rapid recovery of hand function

with minimal therapy. However, recurrence rates are high, affecting

EBioMedicine 33 (2018) 282–288

⁎Corresponding authorat: Kennedy Institute of Rheumatology, Nufﬁeld Department of

Orthopaedics, Rh eumatology and Musculoskeletal Sc iences, Universi ty of Oxford,

Roosevelt Drive, Oxford OX3 7FY, UK.

E-mail address: jagdeep.nanchahal@kennedy.ox.ac.uk (J. Nanchahal).

https://doi.org/10.1016/j.ebiom.2018.06.022

Contents lists available at ScienceDirect

EBioMedicine

journal homepage: www.ebiomedicine.com

85% of patients treated with percutaneous needle aponeurotomy [7]

and 32% of those treated with collagenase [9] at 5 years. The complica-

tion rate is 20% following needle aponeurotomy [8]andover70%after

collagenase injection, the majority being minor and mostly transient

[10].

The ideal therapy would be directed towards patients with early

stage disease to prevent progression to development of cords and sub-

sequent ﬂexion contractures of the digits. Our systematic review [11]

highlighted the lack of robust evidence for treatments proposed for

early stage DD for which there is currently no approved therapy. Studies

reporting the efﬁcacy of intranodular injection of steroids or radiother-

apy are limited by a lack of quality, with no blinding or randomisation

and the use of subjective outcome measures [11]. Fifty percent of pa-

tients receiving steroid therapy developed transient subcutaneous atro-

phy or depigmentation. Approximately 20–30% of patients receiving

radiotherapy developed long-term adverse effects, including dry skin,

desquamation, skin atrophy, telangiectasia, erythema, and altered heat

and pain sensation. A recent trial reported that injection of collagenase

resulted in nodules becoming smaller and softer, with approximately

50% of patients experiencing bruising and pain [12]. Therefore, there is

a need to develop an effective therapy to retard progression of early

DD and also prevent the development of recurrent disease following

surgery, needle fasciotomy or collagenase injection in patients with

established ﬁnger contractures.

The cell responsible for deposition of the excessive collagenous

matrix and contraction in all ﬁbrotic conditions, including DD, is the

myoﬁbroblast [13,14], which is characterised by the expression of α-

smooth muscle actin (α-SMA) [15]. Unlike the ﬁbrotic diseases of

visceral organs such as the kidney, lung and liver, tissue from patients

with DD is readily accessible, allowing identiﬁcation of potential novel

therapeutic targets [16]. Using surgically excised tissue from patients

we found that the myoﬁbroblasts in DD are aggregated in nodules in

the vicinity of the affected joints, and nodules were absent in patients

with more advanced stage disease [17]. Interspersed through the nod-

ule are immune cells, including macrophages, T cells and mast cells,

and the nodular cells secrete a variety of cytokines [18]. Comparison

of the effects of each of these cytokines showed that only tumour necro-

sis factor (TNF) converted palmar ﬁbroblasts from patients withDD into

myoﬁbroblasts at the low concentrations seen ex vivo, but not non-

palmar ﬁbroblasts from the same patients or palmar ﬁbroblasts from

normal individuals. In contrast, TGF-βindiscriminately converts all ﬁ-

broblasts into myoﬁbroblasts [18]. This is important as the ﬁbrotic pro-

cess seen in DD is limited to the palm of genetically susceptible

individuals. Genome-wide association studies have highlighted the

role of Wnt signalling in DD [19,20] and we found that TNF acted via

the Wnt signalling pathway only in palmar dermal ﬁbroblasts from pa-

tients with DD [18]. Myoﬁbroblasts from DD showed a dose-dependent

reduction in contractility on treatment with anti-TNF, with a concomi-

tant reduction in expression of α-SMA [18]. All the clinically approved

anti-TNF agents assessed were effective in down regulating DD

myoﬁbroblast contractility in vitro, with the two fully human IgG mole-

cules, adalimumab and golimumab being the most efﬁcacious at the

doses tested [18].

Here we report the effects of injection of escalating doses of

adalimumab or corresponding volume of placebo directly into the nod-

ules of patients who then underwent surgery two weeks later. Markers

of myoﬁbroblast phenotype and collagen production were assessed in

the excised samples.

2. Methods

2.1. Study Design and Patients

Repurposing anti-TNF for Dupuytren's disease (RIDD) is a two-part

phase 2 randomised, double-blinded placebo-controlled dose response

study to assess the efﬁcacy of local injection of adalimumab in patients

with DD. The protocol was reviewed by the South Central Oxford B

Research Ethics Committee (Reference number 15/SC/0259) and the

Medicine and Healthcare products Regulatory Authority (EudraCT no:

2015-001780-40) and has been published [21] (details of subsequent

amendments in appendix). The phase 2a dose escalation study reported

here was performed at a single centre in the UK at the Edinburgh De-

partment of Plastic Surgery at St John's Hospital, NHS Lothian. Patients

referredto the hand surgery service atSt John's Hospitalby their general

practitioner with a diagnosis of DD were screened for entry to the trial.

Eligibility criteria included no prior treatment for Dupuytren's disease to

the affected hand, a clinically distinct nodule of DD, suitability for injec-

tion of adalimumab and fasciectomy. All potential participants were

screened for TB, HIV, hepatitis B and C using serological testing and

chest X-ray in accordance with local standard procedures for anti-TNF

screening.

During the study a new preparation of adalimumab became avail-

able with 40 mg formulated in 0.4 ml. The lower volume and absence

of excipients including citrate was expected to result in reduced pain

and improve participant acceptability. Therefore, the trial design was

modiﬁed to include a cohort at a dose of 40 mg using this preparation.

The 40 mg in 0.4 ml formulation is currently only available in a pre-

ﬁlled single use syringe and so only the full syringe dose of 40 mg

could be administered with this formulation. Therefore, the trial design

was modiﬁed to include a cohort of 40 mg in 0.4 ml. Since publication,

the protocol has been amended to introduce an endpoint to assess

whether adalimumab affects the healing of the surgical incisions or

subsequent scarring. This visual assessment of the surgical wounds

was carried out by blinded review of the hand photographs taken at

baseline and then at 2 and 4 weeks post-surgery.

2.2. Procedures

Patients were randomised (3:1) to receive either adalimumab or sa-

line in 3 dose cohorts (15 mg in 0.3 ml, 35 mg in 0.7 ml), both using the

40 mg in 0.8 ml preparation, or 40 mg in 0.4ml using the more recently

introduced formulation, or an equivalent volume of placebo (normal sa-

line) on one occasion by intra-nodular injection two weeks (±3 days)

prior to scheduled surgery. Nodule hardness was measured using a du-

rometer (RX-1800-00, Rex Gauge Company, Illinois, USA) and nodule

size was assessed using an ultrasound scan before injection and again

before surgery. Blood was collected at baseline and again immediately

before surgery, and the serum stored at -80 °C prior to analysis. ELISA

kits from Promonitor were used according to the manufacturer's in-

structions to measure serum levels of adalimumab (Cat. No. 728552)

and anti-adalimumab (Cat. No. 728533). All samples were measured

in duplicate and repeated on 3 plates using a FLUOstar Omega Spectro-

photometer (BMG Labtech) and MARS™software. Pain related to the

adalimumab injection was rated by the participant and the injection

site assessed by visual inspection. Following surgery, the excised

Dupuytren's tissue was transported to the lab, where the nodule was

dissected to retain a central piece for histological examination whilst

the remainder was frozen and pulverised. The resulting powder was

split in two for protein or RNA extraction and stored at −80 °C. RNA

was extracted and following the generation of cDNA, absolute levels of

ACTA2, COL1A1, COL3A1and CDH11 were determined using thestandard

curve method (appendix). Procollagen 1A1 protein levels were deter-

mined by DuoSet ELISA reagents (DY6220-05, R&D systems, Oxon,

UK) in triplicate following the manufacturer's instructions. α-SMA

protein levels were determined using a custom developed MSD

plate

(appendix 2). Tissue samples for histology were ﬁxed in 4%

paraformaldehyde, longitudinally bisected, embedded in parafﬁn wax

and 7 μm sections obtained from the cut surface. Sequential sections

were stained with hematoxylin-eosin, mouse anti–α-SMA antibody

(Abcam 7817) or a mouse isotype control. Antibodies were detected

using biotinylated anti-mouse antibody and avidin/biotin/HRP complex

reagent (Vectastain ABC, Vector Lab, UK). Patients completed their

283J. Nanchahal et al. / EBioMedicine 33 (2018) 282–288

standard care following surgery, returning at 2 weeks (±1 week) for as-

sessmentof the wound, change of dressing and commencement of hand

therapy. Final clinical assessment was at 12 weeks (±4 weeks), when a

photograph was obtained for ﬁnal blinded assessment of wound healing

and scarring.

2.3. Outcomes

The primary outcome was mRNA expression of ACTA2 (α-SMA). Sec-

ondary outcomes comprised mRNA expression for COL1A1, COL3A1 and

CDH11, and protein levels of α-SMA. Protein levels of type I procollagen,

type III procollagen, acid and pepsin soluble and insoluble collagen,

alongside nodule hardness and size of the nodule on ultrasound scan

were also assessed. Adverse events were assessed using visual inspec-

tion of the injection site and site of surgery, photographs and laboratory

reports. Tertiary outcomes comprised circulating levels of adalimumab,

antibodies to adalimumab and participant injection experience

recorded as pain during injection using a 5-level Likert scale (1 =

Not at all), and immediately after injection also using a Likert scale

(1–10, 1 = no pain).

2.4. Randomisation and Masking

Participants were randomised for treatment with placebo or

adalimumab using a 1:3 ratio with a block size of four. The dose cohort

was determined by the progression of the trial, with initial patients allo-

cated to the lowest dose cohort (15 mg adalimumab) and doses esca-

lated (35 mg then 40 mg adalimumab) following a review of patient

wound healing two weeks after surgery. The volume of saline solution

for the placebo treatment matched the relevant volume associated

with the dose cohort to ensure blinding. The allocation log, generated

by the trial statistician, using the statistical software STATA version

13.0, was stored securely within the hospital pharmacy. Due to the dis-

tinctive packaging of the 40 mg in 0.4 ml preparation of adalimumab in

pre-ﬁlled syringes, blinding of the person administering the injection

was not feasible and therefore this healthcare professional was inde-

pendent and not involved with routine patient care, recruitment, out-

come assessment, surgery or follow-up. A shield prevented patients in

this dose cohort from seeing the syringe. For all cohorts, participants,

healthcare and laboratory professionals involved in recruitment, sur-

gery, follow-up and outcome assessment were blinded to treatment

allocation.

2.5. Statistical Analysis

This study was designed as a phase 2a clinical trial to enable

estimation of potential dose and efﬁcacy of adalimumab in inhibiting

progression of DD. As no previous studies have explored the effect of

adalimumab on in vivo expression of ACTA2 mRNA, sample size was de-

termined using in vitro ﬁndings and inﬂated to account for the non-

clinical nature of the research upon which estimation was performed.

Descriptive statistics, complemented using exploratory data analyses,

were used to assess the demographics across intervention groups for

each dose based on an intention-to-treat population. Additional

analyses were performed using either linear mixed modelling or nested

analysis of variance to account for trial structure, which included a

random-effects component to allow for the values from three plates

and duplicates per participant. Where appropriate, Tukey's HSD adjust-

ment for multiplicity was applied when assessing contrasts. For the

basis of most contrasts the patients allocated to placebo were pooled.

The validity of samples for circulating levels of adalimumab could not

be veriﬁed for three patients. To address this, a per-protocol approach

was adopted and the corresponding patients were excluded from

analysis.

2.6. Role of the Funding Source

The study was funded by the Health Innovation Challenge Fund

(Wellcome Trust and Department of Health). Funding for purchase of

the adalimumab was provided by 180 Therapeutics LLP. The funders

had no involvement in study design or data analyses. The study was

sponsored by the University of Oxford. Raw unblinded data were

analysed by the trial statisticians (WS and SD) and subsequently made

available to the remainder of the trial team. The corresponding author

drafted the manuscript in conjunction with all the other authors and

all share the responsibility for submission for publication.

3. Results

3.1. Patient Demographics

Between November 2015 and November 2016, 85 participants were

screened and 28 were randomised tosuccessive dose escalation cohorts,

8 to the 15 mg adalimumab cohort, 12 to the 35 mg adalimumab cohort

and 8 to the 40 mg adalimumab cohort. In each cohort, patients were

randomised in a ratio of 3:1 to receive either adalimumab or an equiva-

lent volume of placebo (saline) (Fig. 1). Baseline characteristics were

similar between different cohorts (Table 1).

3.2. Outcomes

There was no difference between any of the groups, including pla-

cebo, in mRNA levels for genes that encode ACTA2,COL1A1,COL3A1 or

CDH11 (Fig. 2). Levels of α-SMA protein expression in patients treated

with 40 mg adalimumab (1.09 ± 0.09 ng per μgoftotalprotein)were

statistically signiﬁcantly lower(p= 0.006) compared toplacebo treated

patients (1.51 ± 0.09 ng per μg of total protein) and compared to those

treated with 15 mg (1.60 ± 0.09 ng per μg of total protein; pb0.001) or

35 mg (1.44 ± 0.08 ng per μg of total protein; p=0.024)adalimumab,

whilst 35 mg and 15 mg adalimumab cohorts showed no difference

compared to placebo (Fig. 3). Qualitative assessment of immunohisto-

chemical staining for α-SMA showed reduced intensity for the re-

sponders in the 40 mg cohort compared to placebo (appendix,

Supplementary Fig. 1).

The levels of procollagen type I protein expression were signiﬁcantly

lower (p= 0.019) in the subgroup treated with 40 mg adalimumab

(474 ± 84 pg/μg total protein) compared with placebo (817 ±

78 pg/μg total protein) (Fig. 4). Pro-collagen type I levels were below

the minimum limit for quantiﬁcation of 250 pg/μg in three patients

injected 35 mg adalimumab and two injected with 40 mg adalimumab.

Tissue fromnodules surgically excised two weeksafter injection was

also assessed for levels of procollagen type III and collagen using the

Biocolor (UK) Sircol Soluble and Insoluble Collagen Assays. There were

no differences between treatment groups and placebo for either assay.

Plasma levels of adalimumab and antibodies to adalimumab were

assayed prior to intra-nodular injection of adalimumab and repeated

2 weeks (±3 days) after injection,prior to surgery. Prior to nodule injec-

tion, circulating blood adalimumab levels were below the minimum de-

tectable limit for all patients. At two weeks after injection, patients

within the placebo treatment group had undetectable levels of blood

adalimumab whilst all patients within adalimumab treatment groups

demonstrated detectable levels (appendix, Supplementary Table 1).

Two weeks after treatment, no antibodies to adalimumab were detected

amongst placebo treated patients, demonstrating assay speciﬁcity and

the absence of cross reactivity. Antibodies were detected in one patient

treated with 15 mg adalimumab, 5 patients treated with 35 mg

adalimumab and two patients treated with 40 mg adalimumab. How-

ever, antibody levels only exceeded the minimum threshold for clinical

signiﬁcance [22] of 12 AU/ml in two of the three repeat measurements

(mean±SD, 12.4±0.63 AU/ml) in one patient treated with 35 mg

adalimumab.

284 J. Nanchahal et al. / EBioMedicine 33 (2018) 282–288

Ultrasound scans performed prior to injection and two weeks later

prior to surgery showed no signiﬁcant changes in any of the dose

cohorts in nodule size based on the area of the nodule, the length of

the perimeter and the feret (maximum length across the nodule)

(appendix, Supplementary Fig. 2). Similarly, there was no change in

nodule hardness measured using a durometer in any of the dose cohorts

when comparing before and two weeks after injection (appendix,

Supplementary Fig. 3).

All but one patient reported pain during the injection, with higher

levels of pain experienced with larger volumes and with the more dilute

formulation of adalimumab. The Likert scores immediately after injec-

tion showed that high volume of injection (0.7 ml) was associated

with more pain than lower volumes (0.3–0.4 ml) and that the 40 mg

in 0.4 ml adalimumab formulation was associated with lower pain

scores (appendix, Supplementary Fig. 4).

3.3. Adverse Events

Two patients suffered serious adverse events (SAE) after injection.

One patient fell two months before injection of placebo whilst running

and presented 5 days after surgery for DD with impaired motor function

due to bilateral subdural haematomas that were treated non-

operatively. He made a full recovery with spontaneous resolution of

Table 1

Baseline characteristics for treatment groups. Ageof onset of Dupuytren's disease and age

on entry into trial shown by treatment group. Patients receiving placebo are pooled.

Treatment Mean SD Min Max Median n

Age upon onset (years)

15 mg adalimumab 47.0 10.5 30 60 48.5 6

35 mg adalimumab 53.8 11.9 28 67 57.0 9

40 mg adalimumab 53.8 5.6 45 60 55.0 6

Sodium chloride 0.9% 56.1 8.5 42 69 57.0 7

Age upon trial entry

15 mg adalimumab 57.2 9.1 48 71 53.0 6

35 mg adalimumab 63.7 9.9 48 78 64.0 9

40 mg adalimumab 63.3 5.7 56 69 65.0 6

Sodium chloride 0.9% 62.9 8.3 46 72 63.0 7

Total

Age upon onset 52.9 9.8 28 69 56.0 28

Age upon trial entry 62.0 8.5 46 78 63.0 28

Assessed for eligibility (n = 85)

Excluded (n=57)

Not meeting inclusion criteria (n=13)

Declined to participate (n=34)

Screened ineligible (n=7)

Screened declined/not recorded (n=3)

Withdrew consent (n=0)

Allocated to adalimumab injection (n = 6)

Received adalimumab (n = 6)

Randomised to

15 mg (n = 8)

Allocated to saline injection (n = 2)

Received saline (n = 2)

Randomised to

35 mg (n = 12)

Allocated to adalimumab injection (n = 9)

Received adalimumab (n = 9)

Randomised to

40 mg (n = 8)

Randomised (n=28)

Allocated to saline injection (n = 3)

Received saline (n = 3)

Allocated to adalimumab injection (n = 6)

Received adalimumab (n = 6)

Allocated to saline injection (n = 2)

Received saline (n = 2)

Fig. 1. Trial proﬁle.

285J. Nanchahal et al. / EBioMedicine 33 (2018) 282–288

the haematomas. A patient with type 2 diabetes who received an injec-

tion of 35 mg of adalimumab into a nodule situated over the proximal

phalanx ofthe little ﬁnger presented four days aftersurgical fasciectomy

with a suture abscess at the distal palmar crease in the axis of the ring

ﬁnger. He was initially treated with oral antibiotics by his general prac-

titioner but due to lack of improvement after 5 days he wasadmitted to

hospital for i.v. antibiotics. Inspection of the wounds revealederythema

around the sutures and a small amount of pus was expressed following

suture removal. He was discharged home the following day on oral

antibiotics and the wound healed uneventfully. Blinded assessment of

photographs of scars obtained 12 weeks (±4 weeks) after surgery

Fig. 2. Box and whiskers plotof log RNA concentration by treatment received. The ﬁgureis divided into separate plots for each gene assessed. The box represents the inter-quartile range

(IQR) and whiskers extend to 1.5 relevant IQR (Tukey boxplot). The x-axisshows individual patients, grouped by treatment received, with the three repeatmeasures performedfor each

patient represented by points stacked within the same column.

Fig. 3. Box and whiskers plot of α-SMA protein concentration by treatment received. The

box represents the inter-quartile range (IQR), the horizontal line represents the median

and whiskers extend to 1.5 relevant IQR. The x-axis shows individual patients, grouped

by treatment received. Samples from each patient were analysed in duplicate on three

separate plates and the values for each plate are shown as circles, triangles or squares.

Fig. 4. Box and whiskers plot of pro-collagen protein concentration by treatmentreceived.

The box represents the inter-quarti le range (IQR), the horizontal line represents the

median and whiskers extend to 1.5 relevant IQR (Tukey boxplot). The x-axis shows

individual pati ents, grouped by tr eatment received, with the three repeat measures

performed for each patient represented by points stacked within the same column. The

plot includes pro-collagen concentrations reported be low the minimum detectable

range as zero, 3 of 9 in 35 mg and 2 of 6 in 40 mg adalimumab cohorts.

286 J. Nanchahal et al. / EBioMedicine 33 (2018) 282–288

showed well-healed mature scars in all patients except the patientwho

developed the post-operative wound infection, where the scars were

more prominent. Both events were considered by the local investigator

and an independent reviewer not to be related to the investigational

medicinal product and the second SAE was considered to be related to

the surgical procedure.

Swelling at the injection site was only apparent in 3 patients receiving

35 mg adalimumab and two receiving 0.7 ml saline. Transient redness

was noted at the injection site in 7 patients distributed across the differ-

ent treatment groups. No patient experienced blistering, haematoma,

bruising, local itching or signs of nerve injury. No patient reported any ad-

verse events one week after injection.

4. Discussion

Our results show that two weeks following administration of 40 mg of

adalimumab in 0.4 ml into Dupuytren's nodules there was down regula-

tion of the myoﬁbroblast phenotype as evidenced by lower expression of

α-SMA and pro-collagen type I proteins. These ﬁndings represent the

clinical translation of our in vitro data based on human tissue where we

showed that tumour necrosis factor (TNF) selectively converts precursor

palmar ﬁbroblasts from Dupuytren's patients to myoﬁbroblasts via the

Wnt signalling pathway, and anti-TNF is inhibitory [18]. Like all ﬁbrotic

conditions, Dupuytren's disease is characterised by the deposition of ex-

cessive collagenous extracellular matrix which is remodelled and

contracted by α-SMA-expressing myoﬁbroblasts [15], that aggregate in

nodules [17]. Collagen is synthesised and secreted in the precursor state

as procollagen and the N- and C-terminal globular domains are cleaved

at the plasma membrane to produce tropocollagen that self-aggregates

to form ﬁbrils, which grow and are eventually cross-linked by lysyl oxi-

dase to form insoluble collagen [23]. The soluble Sircol assay detects all

the uncross-linked forms and the insoluble assay only detects the cross-

linked collagens. The short half-life of procollagen due to its rapid conver-

sion to ﬁbrillar collagen may explain the reduction in type I procollagen

we observed at the two-week primary endpoint and it is possible that

soluble and insoluble collagens may also change but at later time points.

Although type III collagen comprises 35–49% of the total [24]in

Dupuytren's nodules, which represent the earlier stages of disease, we

were unable to detect any changes in type III procollagen due to the

poor sensitivity of the only commercially available assay, with levels fall-

ing below the threshold of detection in half of samples in all treatment

groups. The downregulation of both procollagen and the contractile pro-

tein α-SMA by local inhibition of tumour necrosis factor (TNF) provides

the ﬁrst successful proof of concept for targeted treatment of the patho-

logical mechanisms underlying this common localised ﬁbrotic disease

of the hand that affects 4% of the general UK and US populations [1].

There was nochange in the levels mRNA levels for genes that encode

for ACTA2,COL1A1,COL3A1 and CDH11. The most likely explanation for

the down regulation in protein expression for α-SMA and type I

procollagen without a concomitant change in the respective mRNA is

that expression of these proteins is predominantly regulated at the

post-transcriptional level. We have previously shown to be the case

for α-SMA in Dupuytren's myoﬁbroblasts [25], but did not select α-

SMA protein levels as the primary outcome measure as the assay for

the protein was optimised over the course of the trial. Collagen 1 ex-

pression is also mainly regulated post translationally in ﬁbrotic condi-

tions [26]. Another potential but less likely explanation would be

different half-lives of mRNA and collagen proteins of hours and days re-

spectively [26], which would potentially allow mRNA levels to recover

at 2 weeks whilst corresponding protein levels remained suppressed.

The downregulation of α-SMA and procollagen type I only occurred

in patients where 40 mg of adalimumab in 0.4 ml was injected directly

into the nodule, but not following administration of 35 mg in 0.7 ml.

This could be explained by some of the adalimumab in the more dilute

preparation extravasating out of the nodule and hence not available to

act at high concentrations locally on the aggregates of myoﬁbroblasts

and immune cells. An alternative but less likely explanation is that the

excipients in the more dilute preparation of adalimumab rendered it

less efﬁcacious for this application. Our data would suggest that only

high local concentrations of adalimumab are effective in down regulat-

ing the myoﬁbroblast phenotype in DD and it is unlikely adalimumab or

other anti-TNF drugs would be efﬁcacious if administered systemically.

This would be consistent with the low-grade inﬂammation that is con-

ﬁned to the nodules of DD [18]. The half-life of adalimumab is 2 weeks

[27], and trough levels at this time for patients on regular systemic ther-

apy of 40 mg every 2 weeks averaged ~5μg/ml in patients with rheuma-

toid arthritis and 6–10μg/ml in patients with psoriatic arthropathy not

treated with concomitant methotrexate [27]. This would suggest that

the kinetics of absorption of adalimumab when injected into the nod-

ules of DD may be similar to when the drug is administered subcutane-

ously. We would postulate that formulations that retard dissipation of

adalimumab from the site of injection are likely to be more effective

and may need to be administered less frequently for the treatment of

localised ﬁbrotic conditions characterised by low-grade inﬂammation

such as DD. Neutralising antibodies to adalimumab are likely to result

in reduced efﬁcacy [28]. Only one patient, who received 35 mg in

0.7 ml, had antibody levels (12.4±0.63 AU/ml) just above the threshold

of signiﬁcance of 12Au/ml [22], which are unlikely to be of clinical sig-

niﬁcance. Neutralising antibodies may occur more commonly after re-

peated administration of adalimumab, especially in patients who are

not on concomitant methotrexate.

Whilst almost all patients reported pain during injection, pain scores

immediately after injection were higher in participants receiving volumes

of either placebo or adalimumab N0.3–0.4 ml and in those injected with

the formulation of adalimumab comprising 40 mg in 0.8 ml of carrier

that was used for the 15 mg and 35 mg cohorts. This formulation differs

signiﬁcantly from the more concentrated form of 40 mg in 0.4 ml in

that it contains a variety of excipients [27], including citrate, which is as-

sociated with higher pain scores when injected subcutaneously [29]. The

40 mg in 0.4 ml formulation was not available commercially when the

trial opened to recruitment and therefore the study was initially based

on dose cohorts using the 40 mg in 0.8 ml formulation. Transient swelling

at the injection site was apparent in some patients receiving 0.7 ml of ei-

ther placebo or adalimumab, most likely due to extravasation of the

injected material outside the conﬁnes of the nodule. The safety proﬁle

of adalimumab is well-known and the commonest adverse events in

patients with autoimmune disorders on long-term therapy are related

to infection [30]. One patient in our study, who suffered from type 2

diabetes mellitus received 35 mg of adalimumab developed a wound

infection 4 days following surgery that resolved following removal of

sutures and antibiotics. The infection was considered to be unrelated to

the adalimumab injection as it was a localised suture abscess away

from the site of the injected nodule. Wound infection has been reported

to occur in approximately 3.6% of patients undergoing Dupuytren's

fasciectomy [8].

Unsurprisingly, there was no difference in nodule hardness or sizeas

assessed by tonometry and ultrasound scanrespectively at 2 weeks post

injection. This trial enabled us to optimisethe methods for obtaining the

data and analysing them in preparation for follow up studies.

5. Conclusions

This phase 2a randomised trial shows that a single intranodular

injection of 40 mg adalimumab in 0.4 ml in patients with Dupuytren's

disease is safe and leads to down regulation of the myoﬁbroblast

phenotype as evidenced by reduced expression of α-SMA and type I

procollagen proteins. Having deﬁned the most efﬁcacious dose and

preparation and based on these positive proof of concept data we are

now proceeding with a phase 2b trial in 138 patients with early

stage Dupuytren's disease randomised 1:1 to receive 4 injections of

adalimumab or placebo at 3 month intervals and followed for a total

of 18 months from baseline [21]. Our study also illustrates the utility

287J. Nanchahal et al. / EBioMedicine 33 (2018) 282–288

of using early stage ﬁbrotic human tissue to elucidate novel therapeutic

targets [18] that can be translated to the clinic.

Conﬂicts of Interest

Mrs. Ball, Dr. Davidson, Dr. Williams, Dr. Sones, Dr. McCann, Ms.

Cabrita, Dr. Swettenham, Dr. Cahoon, Dr. Copsey, Dr. Francis, Dr. Black,

Dr. Barber, Mrs. Dutton report grants from Wellcome Trust, grants

from Department of Health, other from 180 Therapeutics LP, during

the conduct of the study.

Professor Taylor reports grants from Wellcome Trust, grants from

Department of Health, other from 180 Therapeutics LP, during the con-

duct of the study; grants from UCB, personal fees from UCB, AbbVie,

Pﬁzer, outside the submitted work.

Professor Lamb reports grants from Wellcome Trust, grants from

Department of Health, during the conduct of the study; Professor

Sarah Lamb reports grants from NIHR Health Technology Assessment

Programme during the conduct of this study.

Professor Nanchahal and Professor Feldmann report grants from

Wellcome Trust, grants from Department of Health, other from 180

Therapeutics LP, during the conduct of the study; In addition, Professor

Nanchahal has a patent PCT/EP2011/069147 issued to 180 Therapeutics,

and with Professor Feldmann a patent PCT/US2017/049691 pending, a

patent PCT/US2017/026382 pending, a patent 16759325.0 EP2017

pending, and a patent 16759326.8 EP2017 pending.

Funding Sources

We would like to acknowledge funding from the Health Innovation

Challenge Fund (HICF) (HICF-R8-433), a parallel funding partnership

between the Wellcome Trust and the Department of Health, 180 Thera-

peutics, and the National Institute for Health Research (NIHR) Oxford

Biomedical Research Centre (BRC). The views expressed are those of

the authors and not necessarily those of the NHS, the NIHR or the

Department of Health.

Author Contributions

JN, SL, MF, CB, SD and PT conceived and designed the study. DD and

NC recruited the patients. CB, DDand NC collected the clinical data. LW,

FM and MC performed the laboratory analyses. WS, SD and BC per-

formed the statistical analyses. JS, VB, JB, EAF and SL managed the trial.

Appendix A. Supplementary Data

Supplementary data to this article can be found online at https://doi.

org/10.1016/j.ebiom.2018.06.022.

References

[1] Hindocha S, McGroutherDA, Bayat A. Epidemiological evaluation of Dupuytren's dis-

ease incidence and prevalence rates in relati on to etiology. Hand (N Y) 2009;4:

256–69.

[2] Reilly RM, Stern PJ, Goldfarb CA. A retrospective revie w of the management o f

Dupuytren's nodules. J Hand Surg 2005;30:1014–8.

[3] Engstrand C, Krevers B, Kvist J. Factors affecting functional recovery after surgery

and hand therapy in patients with Dupuytren's disease. J Hand Ther 2015;28:

255–259; quiz 260.

[4] Davis TR. Surgical treatment of primary Dupuytren's contractures of the ﬁngers in

the UK: surgeons' preferences and research priorities. J Hand Surg Eur Vol 2013;

38:83–5.

[5] Zhao JZ, Hadley S, Floyd E,Earp BE, Blazar PE. The impact of collagenase Clostridium

histolyticum introduction on Dupuytren treatment patterns in the United States.

J Hand Surg 2016;41:963–8.

[6] Rayan GM. Dupuytren's disease: anatomy, pathology, presentation, and treatment.

Instr Course Lect 2007;56:101–11.

[7] van Rijssen AL, ter Linden H, Werker PM. Five-year results of a randomized clinical

trial on treatment in Dupuytren's disease: percutaneous needle fasciotomy versus

limited fasciectomy. Plast Reconstr Surg 2012;129:469–77.

[8] Crean SM, Gerber RA, Le Graverand MP, Boyd DM, Cappelleri JC. The efﬁcacy and

safety of fasciectomy and fasciotomy for Dupuytren's contracture in European pa-

tients: a structu red review of pu blished studies. J Hand Surg Eur Vol 2011;36:

396–407.

[9] Peimer CA, Wilbrand S,Gerber RA, Chapman D, Szczypa PP. Safety and tolerability of

collagenase Clostridium histolyticum and fasciectomy for Dupuytren's contracture. J

Hand Surg Eur Vol 2015;40:141–9.

[10] Hurst LC, Badala mente MA, Hentz VR, et al. Injecta ble collagenase clostridium

histolyticum for Dupuytren's contracture. N Engl J Med 2009;361:968–79.

[11] Ball C, Izadi D, Verjee LS, Chan J, Nanchahal J. Systematic review of non-surgical

treatments for early dupuytren's disease. BMC Musculoskelet Disord 2016;17:345.

[12] CostasB, Coleman S, Kaufman G, James R, Cohen B,Gaston RG. Efﬁcacy and safety of

collagenase clostridium histolyticum for Dupuytren disease nodules: a randomized

controlled trial. BMC Musculoskelet Disord 2017;18:374.

[13] Darby IA, Zakuan N, Billet F, Desmouliere A. The myoﬁbroblast, a key cell in normal

and pathological tissue repair. Cell Mol Life Sci 2016;73:1145–57.

[14] Wynn TA, Ramalingam TR. Mechanisms of ﬁbrosis: therapeutic translation for ﬁ-

brotic disease. Nat Med 2012;18:1028–40.

[15] Hinz B, Phan SH, Thannickal VJ, Galli A, B ochaton-Piall at ML, Gabbiani G. Th e

myoﬁbroblast: one function, multiple origins. Am J Pathol 2007;170:1807–16.

[16] Nanchahal J, Hinz B. Strategies to overcome the hurdles to treat ﬁbrosis, a major

unmet clinical need. Proc Natl Acad Sci U S A 2016;113:7291–3.

[17] Verjee LS, Midwood K, Davidson D, Essex D, Sandison A, Nanchahal J. Myoﬁbroblast

distribution in Dupuytren's cords: correlation with digital contracture. J Hand Surg

2009;34:1785–94.

[18] Verjee LS, Verhoekx JS, Chan JK, et al. Unraveling the signaling pathways promoting

ﬁbrosis inDupuytren's diseasereveals TNF as a therapeutic target.Proc Natl Acad Sci

U S A 2013;110:E9 28–37.

[19] Dolmans GH, Werker PM, Hennies HC, et al. Wnt signaling and Dupuytren's disease.

N Engl J Med 2011;365:307–17.

[20] Ng M, Thakkar D, Southam L, et al. A genome-wide association study of dupuytren

disease reveals 17 additional variants implicated in ﬁbrosis. Am J Hum Genet

2017;101:417–27.

[21] Nanchahal J, Ball C, Swettenham J, et al. Study protocol: a multi-Centre, doubleblind,

randomised, placebo-controlled, parallel group, phase II trial (RIDD) to determine

the efﬁcacy of intra-nodular injection of anti-TNF to control disease progression in

early Dupuytren's disease, with an embedded dose response study. Wellcome

Open Res 2017. https://wellcomeopenresearch.org/articles/2-37/v2.

[22] Bartelds GM, Krieckaert CL, Nurmohamed MT, et al. Development of antidrug anti-

bodies against adalimumab and association with disease activity and treatment fail-

ure during long-term follow-up. JAMA 2011;305:1460–8.

[23] Canty EG, Kadler KE. Procollagen trafﬁcking, processing and ﬁbrillogenesis. J Cell Sci

2005;118:1341–53.

[24] Lam WL, Rawlins JM, Karoo RO, Naylor I, Sharpe DT. Re-visiting Luck's classiﬁcation:

a histological analysis of Dupuytren's disease. J Hand Surg Eur Vol 2010;35:312–7.

[25] Verjee LS, Midwood K, Davidson D, Eastwood M, Nanchahal J. Post-transcriptional

regulation of alpha-smooth muscle actin determines the contractile phenotype of

Dupuytren's nodular cells. J Cell Physiol 2010;224:681–90.

[26] Stefanovic B. RNA protein interactions governing expression of the most abundant

protein in human body, type I collagen. Wiley Interdiscip Rev RNA 2013;4:535–45.

[27] Abbvie. http://www.rxabbvie.com/pdf/humira.pdf;2018.

[28] ChenDY, Chen YM, Tsai WC, et al. Signiﬁcantassociations of antidrug antibodylevels

with serum drug trough levels and the rapeutic respon se of adalimumab and

etanercept treatment in rheumatoid arthritis. Ann Rheum Dis 2015;74:e16.

[29] Laursen T, Hansen B, FiskerS. Pain perception aftersubcutaneous injections of media

containing different buffers. Basic Clin Pharmacol Toxicol 2006;98:218–21.

[30] Burmester GR, Panaccione R, Gordon KB, McIlraith MJ, Lacerda AP. Adalimumab:

long-termsafety in 23 458 patients fromglobal clinical trialsin rheumatoid arthritis,

juvenile idiopathic arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis and

Crohn's disease. Ann Rheum Dis 2013;72:517–24.

288 J. Nanchahal et al. / EBioMedicine 33 (2018) 282–288

Commentary

Cytokine Targeted Therapy for Dupuytren's Disease

Latha Satish ⁎

Research Department, Shriners Hospitals for Children, Cincinnati, OH-45229, USA

Department of Pathology and Laboratory Medicine, University of Cincinnati, OH-45229, USA

Dupuytren's disease (DD) is a complex ﬁbroproliferative disease of

the hand and a difﬁcult condition to live with for the patients and a puz-

zling disease for the treating physicians. It has been morethan 180 years

since Baron Dupuytren was credited with discovering the disease, yet

there is no deﬁnitive treatment to combat DD. Varying numbers with

regards to prevalence rate are reported but a recent systematic review

estimated 12% those aged 55 years are affected, rising to 29% in those

aged 75 years in the general population in Western countries [1]. DD

is a benign condition and early stage disease manifests itself as pits or

nodules in the palm that develop into cords [2]. Non-invasive treat-

ments for early stage disease are of limited efﬁcacy [3] and the mainstay

of treatment for late stage disease for years has been surgery. Recently,

collagenase Clostridium histolyticum (CCH) injection has gained popu-

larity as an alternative to surgery for treating DD. However, it is associ-

ated with a high rate of adverse effects [4] and patient satisfaction

decreases over time asthe disease recurs[5]. Cu rrent research is focused

on targeting factors (mainly cytokines) that are responsible for mediat-

ing matrix production in DD, speciﬁcally collagen. Several growth fac-

tors mainly TGF-β1, PDGF, IGF, CTGF, bFGF and the pro-inﬂammatory

cytokine TNF are found to play a prominent role in the progression of

DD. TNF has been shown to act via the Wnt signaling pathway to pro-

mote contraction and proﬁbrotic signaling in DD cells. In vitro studies

also showed that neutralizing antibodies to TNF downregulates

myoﬁbroblast activity [6]. In EBioMedicine, Nanchahal and colleagues

identiﬁed TNF as a potential target for clinical translation to treat DD [7].

The study reports a phase 2a double-blind, randomized placebo-

controlled clinical trial on the efﬁcacy of injecting nodules of DD with

adalimumab, a TNF inhibitor [7]. The authors should be applauded for

choosing a local delivery route via intra-nodular injection to assess the

efﬁcacy of a biologic, which might become routine in the future to

avoid the necessity for surgical procedure. In this trial, the authors

chose to inject the nodules two weeks prior to the scheduled surgery

to test three different doses of adalimumab, which are 15 mg in 0.3 ml

(n=6;placebon= 2), 35 mg in 0.7 ml (n=9;placebon=3)or

40 mg in 0.4 ml (n = 6; placebo n = 2). Surgically excised tissues

were subjected to mRNA and protein analyses. The primary outcome

measure was to determine the levels of mRNA expression for α-SMA.

Secondary outcomes were to determine the mRNA expression for

collagen types I, III, and cadherin 11, as well as levels of α-SMA and col-

lagen proteins. The authors found no changes in the mRNA expression

for all genes assayed. Expression of α-SMA protein was reduced in pa-

tients who received 40 mg of adalimumab compared to those injected

with placebo, or 35 mg or 15 mg of adalimumab. These data are in con-

cordance with the previous in vitro ﬁnding that inhibiting TNF levels can

downregulate the myoﬁbroblast phenotype [6]. Previous reports sug-

gest an increased proportion of type III collagen in earlier lesions,

which changes to a greater proportion of type I collagen at later stages

of the disease [8]. In the present study, the authors reported that TNF in-

hibition decreased the protein expression of procollagen type I but no

differences were noted in procollagen type III levels due to the low sen-

sitivity of the assay for procollagen type III [7]. TNF inhibition also did

not reduce nodule size or hardness, which is not unexpected as the in-

jection was administered only on one occasion.

The study also has a few limitations, which hopefully be addressed

by the authors in their upcoming trial. An interesting ﬁnding would

have been if the levels of the growth factors mainly TGF-β1andPDGF,

along with TNF in the excised tissue was reported, which would have

added more strength on the utility of adalimumab for DD. Another in-

teresting observation would have been if authors had reported the

changes in the expression level of the ECM protein namely ﬁbronectin

a known contributor in DD pathogenesis along with type I and type III

collagens.

Overall, with this trial, the authors haveestablished safety and deter-

mined the effective dose and volume (40 mg in 0.4 ml) of adalimumab

to proceed to a phase 2b clinical trial with a larger cohort of patients

with early-stage DD to investigate the efﬁcacy of intra-nodular injec-

tions every 3 months over a 12-month period. The ongoing phase 2b

clinical trial should allow the authors to ascertain whether multiple in-

jections can decrease nodule size and hardness. If a signiﬁcant decrease

in nodule size and hardness is noticed, it indirectly reﬂects that there

might have been a signiﬁcant decrease in myoﬁbroblast formation and

collagen accumulation. The authors should also consider a futureclinical

trial of patients with the advanced form of the disease to investigate the

potential of adalimumab in preventing recurrence.

Disclosure

The author declared no conﬂicts of interest.

EBioMedicine 34 (2018) 14–15

DOI of original article: https://doi.org/10.1016/j.ebiom.2018.06.022.

⁎Corresponding author at: ShrinersHospitals for Children, Research Department, 3229

Burnet Avenue, Cincinnati, OH 45229, USA.

E-mail addresses: lsatish@shrinenet.org,satishla@uc.edu (L. Satish).

Contents lists available at ScienceDirect

EBioMedicine

journal homepage: www.ebiomedicine.com

https://doi.org/10.1016/j.ebiom.2018.07.016

References

[1] Lanting R, Broekstra DC, Werker PM, van den Heuvel ER. A systematic review and

meta-analysis on the prevalence of Dupuytren disease in the general population of

Western countries. Plast Reconstr Surg 2014;133:593–603.

[2] Warwick D. Dupuytren's disease: my personal view. J Hand Surg Eur 2017;42E(7):

665–72.

[3] Ball C, Izadi D, Verjee LS, Chan J, Nanchahal J. Systematic review of non-surgical treat-

ments for early dupuytren's disease. BMC Musculoskelet Disord 2016 Aug 15;17(1):

345.

[4] Peimer CA, Wilbrand S, Gerber RA, Chapman D, Szczypa PP. Safety and tolerability of

collagenase clostridium histolyticum and fasciectomy of dupuytren's contracture. J

Hand Surg Eur 2014;40(2):141–9.

[5] BradleyJ, Warwick D. Patient satisfaction with collagenase. J Hand Surg [Am] 2016;41

(6):689–97.

[6] Verjee LS, Verhoekx JS, Chan JK, Krausgruber T, Nicolaidou V, Izadi D, Davidson D,

Feldmann M, Midwood KS, Nanchahal J. Unraveling the signaling pathways promot-

ing ﬁbrosis in Dupuytren's disease reveals TNF as a therapeutic target. Proc Natl Acad

Sci U S A 2013;110:E928–37.

[7] Nanchahal J, Ball C , Davidson D, et al . Anti-tumour n ecrosis factor therapy for

Dupuytren's disease: a randomised dose response proof of concept phase 2a clinical

trial. EBioMedicine 2018. https://doi.org/10.1016/j.ebiom.2018.06.022.

[8] Lam WL, Rawlins JM, Karoo RO, Naylor I, Sharpe DT. Re-visiting Luck's classiﬁcation: a

histological analysis of Dupuytren's disease. J Hand Surg Eur 2010;35:312–7.

15L. Satish / EBioMedicine 34 (2018) 14–15