Targeting Tropomyosin Receptor Kinase in Cutaneous CYLD Defective Tumors With Pegcantratinib: The TRAC Randomized Clinical Trial.

Importance: There are no medical interventions for the orphan disease CYLD cutaneous syndrome (CCS). Transcriptomic profiling of CCS skin tumors previously highlighted tropomyosin receptor kinases (TRKs) as candidate therapeutic targets. Objective: To investigate if topical targeting of TRK with an existing topical TRK inhibitor, pegcantratinib, 0.5% (wt/wt), is safe and efficacious in CCS. Design, Setting, and Participants: A phase 1b open-label safety study, followed by a phase 2a within-patient randomized (by tumor), double-blind, placebo-controlled trial (the Tropomyosin Receptor Antagonism in Cylindromatosis [TRAC] trial). The setting was a single-center trial based at a tertiary dermatogenetics referral center for CCS (Royal Victoria Infirmary, Newcastle, United Kingdom). Patients who had germline mutations in CYLD or who satisfied clinical diagnostic criteria for CCS were recruited between March 1, 2015, and July 1, 2016. Interventions: In phase 1b, patients with CCS applied pegcantratinib for 4 weeks to a single skin tumor. In phase 2a, allocation of tumors was to either receive active treatment on the right side and placebo on the left side (arm A) or active treatment on the left side and placebo on the right side (arm B). Patients were eligible if they had 10 small skin tumors, with 5 matched lesions on each body side; patients were randomized to receive active treatment (pegcantratinib) to one body side and placebo to the other side once daily for 12 weeks. Main Outcomes and Measures: The primary outcome measure was the number of tumors meeting the criteria for response in a prespecified critical number of pegcantratinib-treated tumors. Secondary clinical outcome measures included an assessment for safety of application, pain in early tumors, and compliance with the trial protocol. Results: In phase 1b, 8 female patients with a median age of 60 years (age range, 41-80 years) were recruited and completed the study. None of the participants experienced any adverse treatment site reactions. Three patients reported reduced pain in treated tumors. In phase 2a (15 patients [13 female; median age, 51 years], with 150 tumors), 2 tumors treated with pegcantratinib achieved the primary outcome measure of response compared with 6 tumors treated with placebo. The primary prespecified number of responses was not met. The incidence of adverse events was low. Conclusions and Relevance: In this study, pegcantratinib, 0.5% (wt/wt), applied once daily appeared to be well tolerated and to penetrate the tumor tissue; however, the low tumor drug concentrations demonstrated are likely to account for the lack of response. Dose-escalation studies to assess the maximal tolerated dose may be beneficial in future studies of CCS. Trial Registration: isrctn.org Identifier: ISRCTN75715723.

Tropomyosin Receptor Antagonism in Cylindromatosis (TRAC), an early phase trial of a topical tropomyosin kinase inhibitor as a treatment for inherited CYLD defective skin tumours: study protocol for a randomised controlled trial.

BACKGROUND: Patients with germline mutations in a tumour suppressor gene called CYLD develop multiple, disfiguring, hair follicle tumours on the head and neck. The prognosis is poor, with up to one in four mutation carriers requiring complete surgical removal of the scalp. There are no effective medical alternatives to treat this condition. Whole genome molecular profiling experiments led to the discovery of an attractive molecular target in these skin tumour cells, named tropomyosin receptor kinase (TRK), upon which these cells demonstrate an oncogenic dependency in preclinical studies. Recently, the development of an ointment containing a TRK inhibitor (pegcantratinib - previously CT327 - from Creabilis SA) allowed for the assessment of TRK inhibition in tumours from patients with inherited CYLD mutations. METHODS/DESIGN: Tropomysin Receptor Antagonism in Cylindromatosis (TRAC) is a two-part, exploratory, early phase, single-centre trial. Cohort 1 is a phase 1b open-labelled trial, and cohort 2 is a phase 2a randomised double-blinded exploratory placebo-controlled trial. Cohort 1 will determine the safety and acceptability of applying pegcantratinib for 4 weeks to a single tumour on a CYLD mutation carrier that is scheduled for a routine lesion excision (n = 8 patients). Cohort 2 will investigate if CYLD defective tumours respond following 12 weeks of treatment with pegcantratinib. As patients have multiple tumours, we intend to treat 10 tumours in each patient, 5 with active treatment and 5 with placebo. Patients will be allocated both active and placebo treatments to be applied randomly to tumours on the left or right side. The target is to treat 150 tumours in a maximum of 20 patients. Tumour volume will be measured at baseline and at 4 and 12 weeks. The primary outcome measure is the proportion of tumours responding to treatment by 12 weeks, based on change in tumour volume, with secondary measures based on adverse event profile, treatment compliance and acceptability, changes in tumour volume and surface area, patient quality of life and pain. DISCUSSION: Interventions for rare genetic skin diseases are often difficult to assess in an unbiased way due to small patient numbers and the challenges of incorporating adequate controls into trial design. Here we present a single-centre, randomised, placebo-controlled trial design that leverages the multiplicity of tumours seen in an inherited skin tumour syndrome that may inform the design of other studies in similar genetic diseases. TRIAL REGISTRATION: International Standard Randomised Controlled Trial Number Registry, ISRCTN75715723 . Registered on 22 October 2014.

Transition from cylindroma to spiradenoma in CYLD-defective tumours is associated with reduced DKK2 expression.

Patients carrying heterozygous germline truncating mutations in the CYLD gene develop multiple primary hair follicle-related tumours. A highly patterned tumour, termed cylindroma, and a highly disorganized tumour, termed spiradenoma, may both develop in the same patient. Furthermore, histological features of both tumour types have been described within the same tumour specimen. We used three-dimensional computer-aided reconstruction of these tumours to demonstrate contiguous growth of cylindromas into spiradenomas, thus suggesting a transition between the two tumour types. To explore factors that may influence cutaneous tumour patterning, genome-wide transcriptomic analysis of 32 CYLD-defective tumours was performed. Overexpression of the Wnt/ß-catenin signalling pathway was observed relative to normal perilesional tissue. Morphometric analysis was used to investigate the relationship between Wnt pathway-related gene expression and tumour organization. This revealed an association between reduced Dickkopf 2 (DKK2-a negative regulator of the Wnt/ß-catenin signalling pathway) expression and loss of tumour patterning. Reduced DKK2 expression was associated with methylation of the DKK2 gene promoter in the majority of tumour samples assayed. RNA interference-mediated silencing of DKK2 expression in cylindroma primary cell cultures caused an increase in colony formation, cell viability, and anchorage-independent growth. Using these data, we propose a model where epigenetic programming may influence tumour patterning in patients with CYLD mutations.

Tumor mapping in 2 large multigenerational families with CYLD mutations: implications for disease management and tumor induction.

OBJECTIVES: To comprehensively ascertain the extent and severity of clinical features in affected individuals from 2 large families with proven heterozygous mutations in the CYLD locus and to correlate these findings with the 3 appendageal tumor predisposition syndromes (familial cylindromatosis, Brooke-Spiegler syndrome, and multiple familial trichoepitheliomas) known to be associated with such germline mutations. DESIGN: Interfamilial and intrafamilial observational study. SETTING: Tertiary genetic and dermatology referral center. PARTICIPANTS: Thirty-four individuals recruited from 2 large multigenerational families with CYLD mutations. Clinical details, history, and tumor maps were obtained from all participants; in 18, the information was corroborated by detailed clinical examination. MAIN OUTCOME MEASURES: Tumor density, distribution and histologic findings, associated medical conditions, patient symptoms, and impact of disease on quality of life. RESULTS: The severity of penetrance and phenotype varied within families. Although an approximately equal female to male predisposition was noted, 5 women and 1 man (of 26 patients surveyed [23%]) had undergone total scalp removal. The average age at onset was 16 years (range, 8-30 years). Symptoms reported by affected patients included painful tumors (in 12 of 23 patients [52%] who answered the question), conductive deafness, and sexual dysfunction. Of the 26 surveyed patients, tumors were noted on the scalp in 21 (81%), on the trunk in 18 (69%), and in the pubic area in 11 (42%). Tumor mapping provided clinical evidence that correlated with hormonally stimulated hair follicles being particularly vulnerable to loss of heterozygosity and tumor induction. CONCLUSIONS: The burden of disease at sites other than the head and neck appears to be underreported in the literature and greatly affects quality of life. Differentiation between the clinical diagnoses has little prognostic or clinical utility in genetic counseling, even within individuals from the same family. Thus, we suggest an encompassing diagnosis of "CYLD cutaneous syndrome." Finally, the clinical distribution of tumors suggests that hormonal factors may play an important role in tumor induction in these patients.