The Martin Baker (MB) SCARD Melanoma Project – A detailed overview

Introduction

The Skin Cancer Audit Research Database (SCARD) project has been developed by the Skin Cancer College of Australia and New Zealand (SCCANZ). This project has accumulated a large amount of data which is being analysed in partnership with the University of Queensland for research purposes.

A paper, "Measuring Performance in Skin Cancer Practice: The SCARD Initiative" has been published by the International Journal of Dermatology. To view a demonstration version of SCARD including the current pooled report go to https://skincanceraudit.com/demo

In this new proposed project, data will be collected using the SCARD infrastructure. This study will take the quality of data collection one step further. Completeness of data will be verified as will the entry of the Provisional Diagnosis (PD) prior to knowledge of the Histological Diagnosis (HD). Also accuracy of entered HD will be independently verified by the pathology provider.

 

Participant acknowledgement by name

Every participating practitioner who substantially fulfils the requirements of this study will be offered acknowledgement by name in any papers submitted for publication which use this data.

Date of commencement and conclusion of data collection.

This project will access ethics compliant data collected as from March 1 2012 to February 28 2013 or for an alternative 12 month period commencing at a date to be determined during 2012

 

Questions that this study will endeavour to answer

1. What is the percentage of final melanoma diagnoses which were preceded by a benign diagnosis?

2. What percentage of diagnoses where “Exclude melanoma?” is selected, is subject to a request for review and what percentage of these have the diagnosis upgraded or downgraded?

3. What are the features that lead to a request for review of the Histological Diagnosis (HD) and which features are most likely and least likely to be associated with an upgraded diagnosis of melanoma?

4. What is the incidence of recommendation by the pathologist for “Melanoma margins” when the diagnosis provided is benign and what is the exact pathology report in these circumstances?

There are many other studies possible from this data including but not to be limited to:-

1. Number of melanomas diagnosed when the intention was „Exclude non-melanoma skin cancer (NMSC)?

2. NNT (number needed to treat to find one melanoma) related to various practitioner variables

3. Diagnostic accuracy related to practitioner and body location variables

4. Proportion of various melanomas detected by the doctor/patient/spouse/other related to body location, melanoma thickness etc.

 

Statistical details of the study

This will be an observational cohort study. There is no intervention as a result of this study.

The group of people studied will be patients of the participating doctors who have surgical treatment on suspicion of skin malignancy. The treatment will be according to normal clinical guidelines and will not be influenced by the study.

The 2 primary outcome variables will be:-

1. Request for review of histological diagnosis

2. Change of histological diagnosis by the reporting pathologist.

These 2 primary outcome variables can be associated with

1. Independent variables such as patient sex and age

2. Clinical factors (e.g. ugly duckling sign) or dermatoscopic features

“Change of histological diagnosis” could be a consequence of sampling error (changed diagnosis as a result of deeper levels), misinterpretation (changed diagnosis as a result of a second opinion) or because of ambiguity of the histopathological presentation that may or may not be resolved by dermatoscopic/dermatopathological correlation.

 

Participating in the MB SCARD Melanoma Project – Step by Step Instructions

If after reading this you wish to participate follow these steps:-

1. If you have not done so you should first „Register for SCARD? at the SCARD home page www.skincanceraudit.com

2. Also If you have not done so you need to read the „Consent Information Sheet? and sign the „Consent Form? available at the „Downloads? section of the SCARD home page

3. Activate the „Martin Baker SCARD Melanoma Project? in the preferences section of your SCARD database. When you do this you can start recording data in the format of the project but your data will not be accessed by researchers until they have received your signed consent form for the MB SCARD Melanoma Project.

4. Contact your pathology provider(s) and give them the „The Martin Baker SCARD Melanoma Project: Letter to Pathology Provider?. Solicit and secure their agreement to participate as requested in the letter.

5. Set up a protocol with your reception staff to give each patient seen for the treatment of suspected skin cancers the opportunity to sign consent to participate. This only needs to be done once per patient. The „Patient Consent Form? is available at the „Downloads? section of the SCARD home page under „Documents for the Martin Baker SCARD Melanoma Project?

6. We suggest you have a rubber stamp made for your pathology forms titled „MB SCARD Melanoma Project? with a space in which you can write the „MB SCARD Specimen number? and below this a space where you can write the „Provisional Diagnosis?.

7. When all of the above are in-place you can sign and forward the „MB SCARD Melanoma Project Participant (Practitioner) Consent form? also available at the „Downloads? section of the SCARD home page under „Documents for the Martin Baker SCARD Melanoma Project?

8. At this point your data will be accessed by the researchers prospectively from the date after you signed consent.

 

MAKING ENTRIES

When you make an entry on your SCARD database a unique number is generated for each entry. This needs to be entered on the pathology form along with the provisional diagnosis you entered on your SCARD database. If you enter your SCARD data later in the day you can print out a list at the end of the day (see „MB SCARD pathology report: day-sheet? on the home page of your database – print date-range) and send this through to your pathologist.

 

VERIFYING COMPLETENESS AND ACCURACY OF DATA

1. From time to time as agreed between you and your pathologist(s) (e.g. weekly) print out a „MB SCARD pathology report: verification version?. This version prints all specimens with a HD listing since the last printing and it will have specimens listed sequentially with date, name, location on the left and PD, HD, specimen ID number and tick-box on the right. At the bottom of the page is a statement: “This list includes all specimens referred for exclusion of skin cancer in this time-interval” with a space for the pathologist?s signature.

Forward this to your pathologist(s)

2. Your pathologist (or delegated staff member) will be asked to place a tick beside all entries where the PD matches that on the form and HD matches the signed out result, and to sign the form.

3. A replied paid envelope will be provided for the pathologist to forward the right hand part of the document (without patient?s name or date of birth) to the SCARD principal researcher (Cliff Rosendahl)

 

Details of data to be recorded

Already collected on SCARD

Practitioner details – Sex, year of graduation, whether metropolitan, provincial or rural, country and state of practice, use of dermatoscopy, whether GP, GP with a special interest in skin cancer, primary care doctor exclusively practicing in the field of skin cancer, specialist dermatologist or other specialist.

Patient details – Age and sex

Specimen details – Date of procedure, Body location, New or Previously biopsied, whether the procedure is to exclude melanoma, exclude Non-melanoma Skin Cancer (NMSC) or Not Applicable (N/A), The Provisional Diagnosis, The Biopsy or Definitive Management type, whether dermatoscopy has been used and whether the procedure is being done because of digital dermatoscopic monitoring change. After the HD is received information is added including the HD, whether surgical margins are adequate, and whether further action is required. There is the optional ability to record complications and to make (unlimited) notes.

Additional data to be collected

Practitioner details – Postcode, any additional academic qualifications after basic medical degree including year of granting, membership status of the Skin Cancer College of Australia and New Zealand as at January 1 2011

Patient details – No change

Specimen details:-

1. For every new specimen there will be a tick box “Has this patient had a total body skin examination in the last 2 years?

2. For every new specimen there will be a field (located under the exclusion field) where the participant will be asked to record “Certainty that this is a melanoma” This will vary from zero (for most cases of “Exclude NMSC?”) to 100%. There will be drop down options of 0,10,20,30,40,50,60,70,80,90,100%.

3. For every lesion with a HD of melanoma (any type) there will be a drop down box activated with the choices of „Superficial spreading, Lentiginous, Nodular or other?. Participants will need to ensure their pathologists are aware of the need for this information.

4. For every lesion with a HD of melanoma (any type) there will be a drop down box where the diameter of the melanoma can be recorded

5. If an HD of „Melanoma? (any type) is entered there will be a field to record who detected the lesion. There will be drop-down options of “Treating doctor, Another doctor, A Paramedical professional (e.g. nurse, physiotherapist, podiatrist), Another (paid) professional (e.g. hairdresser), Patient, other person”

6. There will be a tick-box to be ticked if the pathologist recommends that “melanoma margins” be applied where the HD is benign, with a notes section beneath this where the participant will be asked to “paste in” the complete pathology report (without the patient?s or pathologist?s name).

7. There will be a tick box to be ticked if a review of the HD is requested and that will activate the following further fields

a) “Reason for requested review” with a “notes” section where exact details can be typed in.

b) “Nature of review performed” (e.g. 2nd opinion, deeper levels cut etc) with a “notes” section where exact details can be typed in.

c) “Outcome of review” with a drop down box where the diagnosis after review can be recorded (same options as in the HD drop down box)

Request for review will be at the discretion of the treating practitioner with the exception that a review request will be mandatory if the level of certainty for melanoma recorded is 90% or 100% and the HD is benign.

Participating practitioners will be encouraged to take dermatoscopic images of lesions strongly suspected of being melanomas and to make these available to the researchers upon specific request.

 

Copy Slides

No copy slides will be accessed by the researches during the data collection period but to facilitate a possible follow-on study participating pathologists will be asked to put aside a copy slide and tissue block in the following circumstances:-

1. If review of HD is requested

2. If a HD of melanoma is made.

3. If the diagnosis is benign but “melanoma margin” re-excision is recommended

Any follow-on study using this material would require the agreement of all pathologists and approval by the ethics committee. In preparation for this possibility patients will be requested to give signed consent when they are seen for the first time at the commencement of the study.

Consent issues

As with the current (Ethics approved) SCARD project patient identity is not revealed to the researchers and the anticipated patient and specimen numbers will be large.

Patient consent will be required specifically because histology slides containing their body tissue may be provided to a 3rd party (panel of expert pathologists) for an opinion in a further extension of this study subject to agreement by pathology providers and ethics approval. A consent information sheet incorporating a consent form for the provision of slides and also possibly actual tissue collected for examination should be provided to each patient at the first encounter with them during the 12 months of the study. In the event of a patient declining to sign the consent form their data should still be entered on SCARD but there will be a field to tick “Exclude from MB Study” and that patient?s data would be excluded from the research pool for the Martin Baker SCARD Melanoma Project. The participating practitioner should hold consent forms in a secure place for a period of 5 years then they can be destroyed.

Practitioners who choose to participate will need to have signed the consent form for the current SCARD project. They will also be asked to read this detailed overview and a “Consent information Sheet” for The Martin Baker SCARD Melanoma Project and sign a specific consent form for this project

Literature Review

It has been established that the use of dermatoscopy increases sensitivity 1 and specificity 2 of melanoma diagnosis. There remain cases that remain difficult to diagnose histopathologically either as a result of specimen sectioning/sampling errors 3 or overlap of the histopathological morphologies of some benign and malignant melanocytic lesions 4.

It has been demonstrated in the histopathological review process that the addition of dermatoscopic information or the highlighting of specific doubtful areas by the clinician 5,6,8 increases histopathological diagnostic sensitivity for melanoma. Thus, addition of clinical dermatoscopic information to the review process of lesions suspected to be melanoma may reduce false negative histopathological reporting, the occurrence of which is potentially fatal to the patient and medico-legally of concern for the clinician and the histopathologist.7

There has not previously been any examination of the interaction between histopathologists and primary care skin cancer practitioners with respect to resolving discordant dermatoscopic-histopathological correlation in cases of suspected melanoma. We intend to assess the frequency and cause, as well as the outcome of such requests.

1. Westerhoff K, McCarthy WH, Menzies SW. Increase in the sensitivity for melanoma diagnosis by primary care physicians using skin surface microscopy . Br J Dermatol ,2000; 143: 1016-1020 )

2. Carli P, de Giorgi V, Crocetti E, Mannone F, Massi D, Chiarugi , Giannotti B Improvement of malignant ?benign ratio in excised melanocytic lesions in the dermoscopy era: a retrospective study 1997–2001. Br J Dermatol ,2004; 150: 687–692.

3. Ferrara G, Argenziano G, Cerroni L, et al. A pilot study of a combined dermoscopic pathological approach to the telediagnosis of melanocytic skin neoplasms. J Telemed Telecare. 2004;10:34-38

4. Ruiter D J, van Dijk M, Ferrier CM. Current Diagnostic Problems in Melanoma Pathology. Seminars in Cutaneous Medicine and Surgery, 2003;22: 33-41

5. Bauer J, Metzler G, Rassner G, Garbe C, Blum A. Dermatoscopy turns histopathologist?s attention to the suspicious area in melanocytic lesions. Arch Dermatol. 2001;137:1338-1340

6. Soyer HP, Kenet RO, Wolf IH, Kenet BJ, Cerroni L. Clinicopathological correlation of pigmented skin lesions using dermoscopy. Eur J Dermatol. 2000;1022-28.

7. Leong AS, Braye S, Bhagwandeen B. Diagnostic 'errors' in anatomical pathology: relevance to Australian laboratories. Pathology. 2006 Dec;38(6):490-7.

8. Gerardo Ferrara et al., “The Influence of Clinical Information in the Histopathologic Diagnosis of Melanocytic Skin Neoplasms,” PLoS ONE 4, no. 4 (April 30, 2009): e5375.

Conclusion

The Martin Baker SCARD Melanoma Project follows on from the current SCARD project but the extra effort by participants will provide data which will give more validity to the results of studies. The amount of extra data collection required for routine entries is minimal with more substantial extra data recorded only where a melanoma is diagnosed or when a review of HD is requested.

We anticipate that this study will provide new information in an important area of skin cancer and melanoma diagnosis and management.