Trial Review

Please note the ANZCTR will be unattended from Friday 20 December 2024 for the holidays. The Registry will re-open on Tuesday 7 January 2025. Submissions and updates will not be processed during that time.

Registering a new trial?

To achieve prospective registration, we recommend submitting your trial for registration at the same time as ethics submission.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12623000275662
Ethics application status
Approved
Date submitted
28/02/2023
Date registered
15/03/2023
Date last updated
15/03/2023
Date data sharing statement initially provided
15/03/2023
Type of registration
Prospectively registered

Titles & IDs
Public title
One Million Skin Checks: A Feasibility Study of Teledermatology and Artificial Intelligence in General Practice
Scientific title
One Million Skin Checks: A Feasibility Study of Teledermatology and Artificial Intelligence in General Practice For Skin Cancer Detection in Adults
Secondary ID [1] 309082 0
None
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Skin cancer diagnosis 329144 0
Condition category
Condition code
Cancer 326125 326125 0 0
Malignant melanoma
Cancer 326126 326126 0 0
Non melanoma skin cancer

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
The study involves 3 interventions made directly to GP's. All GP's will have access to all 3 interventions. The interventions will not affect patient care.

Intervention 1: Provision of a 2 hour online education module for GP's, produced by MetaOptima, to teach GP's how to use the DermEngine software to obtain dermoscopic images of suspicious skin lesions using a smartphone and dermatoscope, and to interpret the artificial intelligence report. Each GP will access the education module via the DermEngine Academy online training website on one occasion, at the start of the trial. The module involves text and images demonstrating correct and incorrect ways to use the DermEngine software, take dermoscopic images as well as labelled images of appropriate lesions and a selection of images of examples of conditions which would not be suitable for imaging (eg inflammatory skin disorders). There is an evaluation quiz at the end of the module.


Immediately after successful completion of the training module, GP's will be able to begin enrolling sequential patients and accessing interventions 2 and 3, as below. Access to the DermEngine app and cloud based web portal will be granted automatically by the software on successful completion of the training module. The study team (Drs Anderson and Koutsis) will check that the information has been entered and accessed on DermEngine on a weekly basis and contact GP's where there are significant delays.

Intervention 2: Provision of a teledermatology opinion by an Australian Consultant Dermatologist to the GP's on skin lesions from patients that they have enrolled into the study and imaged, after they have finalised standard care clinical management plan. The Consultant Dermatologist will log in to a dedicated clinical trial version of the secure online (cloud based) DermEngine site and view trial images and enter their diagnosis. The teledermatology opinion is generated for every patient and will be accessed within the DermEngine or website by the GP. Teledermatology opinions will be delayed by at least 24hrs to ensure that they do not affect patient care. The DermEngine app and site are already in common use in clinical care in dermatology services across Australia and meet relevant IT data security/governance requirements.

Intervention 3: Provision of an artificial intelligence (AI) assessment to the GP's on skin lesions from patients that they have enrolled into the study and imaged, after they have finalised standard care clinical management plan for patients. The AI opinion is generated for every patient and provided within the cloud based DermEngine site. AI results will be delayed by at least 24 hrs to ensure that they do not affect patient care.

DermEngine is a commercial product produced by MetaOptima Technology Australia Pty Ltd
Sydney, NSW. It consists of a smartphone app and a cloudbased platform, accessed from any web browser. The product has been modified for this trial to incorporate produce a bespoke study version. Modifications include identifying patients only by their study ID (ie DermEngine is anonymised for trial patients) and the inclusion of teledermatology and AI opinions. which will only be available to those engaged in this clinical trial.
Intervention code [1] 325541 0
Diagnosis / Prognosis
Intervention code [2] 325546 0
Treatment: Devices
Comparator / control treatment
no control group
Control group
Uncontrolled

Outcomes
Primary outcome [1] 334013 0
Proportion of GP's screened for the study who agree to participate, as measured by: Number of GPs who agree to participate in the study divided by number of GPs who are assessed as suitable during the screening process
Timepoint [1] 334013 0
Prior to patient recruitment, at screening of GP expression of interest forms by investigators
Primary outcome [2] 334014 0
Proportion of lesion images that are of adequate quality for assessment by the teledermatologist. The teledermatologist will access images for review by logging into their DermEngine account, using their study login, which allows them to access the modified and restricted anonymised study version of DermEngine.

The calculation for this study outcome = number of lesions of adequate quality for assessment by teledermatologist (i.e lesions not rejected by the teledermatology (TD) divided by total number of lesions recruited. Recorded by Consultant Dermatologist
Timepoint [2] 334014 0
At point of teledermatology assessment by Consultant Dermatologist
Primary outcome [3] 334015 0
Proportion of lesion images that are of adequate quality for assessment by the AI.
As measured by: Number of lesions of adequate quality for assessment by the AI (i.e lesions not rejected by the AI) divided by total number of lesions recruited. Determined by the AI algorithm within DermEngine.
Timepoint [3] 334015 0
At point of AI assessment of images within the DermEngine app
Secondary outcome [1] 419082 0
Proportion of lesions that are correctly selected for analysis by GPs (ie. not lesions in the exclusion criteria.
As measured by: Number of lesions diagnosed by histopathology which are not excluded lesions (Eg: not lesions in the excluded clinical scenarios - inflammatory, infective, etc) divided by total number of lesions recruited
Timepoint [1] 419082 0
At point that GP uploads histopathology result to DermEngine app
Secondary outcome [2] 419083 0
GP's views and experiences of using the AI.
As measured by a questionnaire assessing feasibility, acceptability and appropriateness measures based on validated intervention tools with answers on a 5 point Likert scale.
Timepoint [2] 419083 0
At the point the GP logs into DermEngine app to access the AI result for each lesion
Secondary outcome [3] 419084 0
GP views and experiences of using the teledermatology service.
As measured by a questionnaire assessing feasibility, acceptability and appropriateness measures based on validated intervention tools with answers on a 5 point Likert scale.
Timepoint [3] 419084 0
At the point the GP logs into the DermEngine app to access the teledermatology result for each lesion
Secondary outcome [4] 419085 0
GP’s views and experiences of using the online training modules.
As measured by questionnaires for each module assessing feasibility, acceptability and appropriateness measures based on validated intervention tools with answers on a 5 point Likert scale.
Timepoint [4] 419085 0
Immediately after GP's complete each of the modules of the 2 hour online training modules within the DermEngine Academy.
Secondary outcome [5] 419375 0
GP's views and experiences of using the AI as measured by a semi-structured video interview based on constructs from the consolidated framework for implementation of research (CFIR).
Timepoint [5] 419375 0
Within 2 months of recruitment of each GP's 10th and final patient
Secondary outcome [6] 419376 0
GP views and experiences of using the teledermatology service, as measured by a semi-structured video interview based on constructs from CFIR.
Timepoint [6] 419376 0
Within 2 months of recruitment of each GP's final patient
Secondary outcome [7] 419377 0
GP's views and experiences of using the online training modules, as measured by a semi-structured video interview based on constructs from CFIR
Timepoint [7] 419377 0
Within 2 months of recruitment of each GP's final patient
Secondary outcome [8] 419378 0
GP sensitivity in diagnosing any form of skin cancer

Sensitivity = (number of cancerous lesions correctly identified as such by GP)/(total number of cancerous lesions)
Data collection: GPs will enter provisional diagnosis for each lesion at time of image capture, which may include cancer. GPs will biopsy each lesion and submit the histopathology report for each lesion which is gold standard for diagnosing cancer.
Timepoint [8] 419378 0
Calculated within 2 months following recruitment of last patient
Secondary outcome [9] 419379 0
GP specificity in diagnosing any form of skin cancer.

Specificity = (number of benign lesions correctly identified as such by GP)/(total number of benign lesions) x 100
Data collection: GPs will enter provisional diagnosis for each lesion at time of image capture, which may include cancer. GPs will biopsy each lesion and submit the histopathology report for each lesion which is gold standard for diagnosing cancer.
Timepoint [9] 419379 0
Calculated within 2 months following recruitment of last patient

Eligibility
Key inclusion criteria
STUDY PARTICIPANTS:
Patients with a skin lesion where the GP is concerned about possible skin cancer
Lesions:
• For surgical biopsy and submission for histopathological assessment
• Suspicious for one of the three target cutaneous malignancies of DermDx
o Melanoma
o Basal Cell Carcinoma
o Squamous Cell Carcinoma or Actinic Keratosis (Intraepithelial SCC/Bowen’s Disease)
• Amenable for imaging with a dermatoscope (or dermoscopic attachment) and digital imaging device (E.g.: digital camera or smartphone)
• Amenable for contact or non-contact dermoscopy
• Size: entire lesion must fit within field of view of the dermatoscope
Biopsy type:
• All types of surgical biopsy or excision suitable for histological assessment. These include partial biopsy or complete excision, shave, punch, curettage.
Patients:
• Types 1 to 3 Fitzpatrick skin phototype
• Aged 18 years or older
• ability to consent to digital image acquisition, analysis of the image by teledermatologist and artificial intelligence image algorithm and surgical biopsy/excision. If they are unable to consent an authorised person may provide consent on their behalf (As outlined in the Health Privacy Act 1988).

GP'S
• GPs must have general, unrestricted medical practitioner registration with AHPRA
• They must work and recruit patients in a primary care environment (mixed general practice, GP skin cancer practice)
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Lesions
• Previously biopsied
• Sites: mucosal, acral, ocular, umbilicus
Clinical scenarios
• Emergency or Life-Threatening conditions, e.g.: Steven-Johnson Syndrome, Toxic-Epidermal Necrolysis
• Infections, e.g.: folliculitis
• Infestations, e.g.: scabies
• Inflammatory conditions, e.g.: eczema, psoriasis
• Blistering disorders, e.g.: pemphigus vulgaris
• Hair loss, e.g.: alopecia areata
• Traumatic injuries
Patient
• Pregnant women (due to known physiological changes that occur in naevi during pregnancy)
• Unable to tolerate dermoscopy (e.g.: sitting still, hypersensitivity to any contact materials)

GP
Practitioners
• Other primary care clinicians who do not hold medical registration with AHPRA. E.g.: nurse practitioners

• Practitioners working in a specialist practice or environment. E.g.: GPs working within a surgical oncology or dermatology practice

Study design
Purpose of the study
Diagnosis
Allocation to intervention
Non-randomised trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Single group
Other design features
This is a feasibility study and so there is no control group
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
Sample size for the process of using the AI and TD is 300 lesions. For all skin malignancies (melanoma and keratinocyte cancers) the number number needed to biopsy (NNB) is estimated at 1.9 to 2.1. 300 cases would allow sensitivity and specificity for detecting any kind of skin malignancy (i.e. melanoma, SCC, or BCC) to be estimated with a precision of +/- 7% or less, assuming the proportion of cases with any form of skin cancer is 50%, an expected sensitivity of 95%, expected specificity of 80%, 5% missing data, type I error of 5%, and power of 80%.

The primary outcomes will be reported using statistical proportions with 95% confidence intervals.

Results from the questionnaires will be summarised with descriptive statistics such as proportions and means, with 95% Confidence Intervals. Thematic analysis will be used to analyse the interview data.

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
1/04/2023
Actual
Date of last participant enrolment
Anticipated
30/09/2023
Actual
Date of last data collection
Anticipated
31/10/2023
Actual
Sample size
Target
300
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
ACT,NSW,NT,QLD,SA,TAS,WA,VIC

Funding & Sponsors
Funding source category [1] 313286 0
Commercial sector/Industry
Name [1] 313286 0
MetaOptima Technology Australia Pty Ltd
Country [1] 313286 0
Australia
Primary sponsor type
Other Collaborative groups
Name
Melanoma Institute Australia
Address
The Poche Centre
40 Rocklands Rd,
Wollstonecraft
Sydney
NSW 2065
Country
Australia
Secondary sponsor category [1] 315040 0
None
Name [1] 315040 0
Address [1] 315040 0
Country [1] 315040 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 312516 0
Sydney Local Health District Ethics Review Committee (RPAH Zone)
Ethics committee address [1] 312516 0
Ethics committee country [1] 312516 0
Australia
Date submitted for ethics approval [1] 312516 0
Approval date [1] 312516 0
16/09/2022
Ethics approval number [1] 312516 0
X22-0214 & 2022/ETH01500

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 124938 0
Dr James Koutsis
Address 124938 0
Melanoma Institute of Australia
40 Rocklands Rd, Wollstonecraft, NSW, 2065
Country 124938 0
Australia
Phone 124938 0
+61 29911 7200
Fax 124938 0
Email 124938 0
james.koutsis@melanoma.org.au
Contact person for public queries
Name 124939 0
James Koutsis
Address 124939 0
Melanoma Institute of Australia
40 Rocklands Rd, Wollstonecraft, NSW, 2065
Country 124939 0
Australia
Phone 124939 0
+61 29911 7200
Fax 124939 0
Email 124939 0
james.koutsis@melanoma.org.au
Contact person for scientific queries
Name 124940 0
Pascale Guitera
Address 124940 0
Melanoma Institute of Australia
40 Rocklands Rd, Wollstonecraft, NSW, 2065
Country 124940 0
Australia
Phone 124940 0
+61 9515 8537
Fax 124940 0
Email 124940 0
pascale.guitera@melanoma.org.au

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
Sensitive commercial data


What supporting documents are/will be available?

No Supporting Document Provided



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.