A systematic review of the frequency of features of the seven‐point checklist in proven cutaneous melanoma: The importance of change

Abstract Background Pigmented skin lesions in human adults can present with several different visible features that may indicate signs of malignancy, particularly melanoma. Patient and clinician awareness of these features can aid the early recognition and melanoma diagnosis improving patient outcomes. The seven‐point checklist (7PCL) is a clinical prediction rule advocated by the National Institute for Health Care Excellence to aid the assessment of pigmented skin lesions in primary care to indicate referral for specialist opinion. Objectives Assess the current evidence to establish which features of the 7PC present more frequently, so public education and clinician assessment can be focused to maximise early diagnosis and minimise referrals of benign lesions. Methods A systematic review of published evidence identified studies that assessed the seven features of the 7PCL in histologically proven melanomas. Two independent reviewers screened eligible studies and independently extracted data and assessed quality. Results 112 studies were screened, 20 were assessed in full, seven met the inclusion criteria. 1184 histologically diagnosed melanomas were assessed using the 7PCL. Four studies involved patients assessing 335 melanomas, and three involved clinicians who assessed 849 melanomas. The most common feature identified was a change in size of the lesion, and the least common was inflammation. Conclusions The most frequently occurring features of melanoma involve shape, size and colour, however focussing on changes in features, rather than irregularity, is more likely to identify early melanoma and increase the accuracy of referrals.

naked eye before closer examination with dermoscopy.This systematic review (SR) will focus on naked eye (clinical) assessment.
Clinical prediction rules (CPRs) are tools to guide clinicians' decision-making. 5There are CPRs for clinical and dermoscopic assessment of skin lesions.A 2017 SR found 4 clinical CPR's: ABCD; ABCDE; the seven-point checklist (7PCL) and the revised/weighted 7PCL. 6The ABCD mnemonic (Asymmetry Border; Colour; Diameter) was developed in 1985 7 and is still used by primary care clinicians in the USA and Australia. 8,9he original 7PCL, known as the Glasgow Checklist, was introduced by MacKie 10 in 1986, following a study demonstrating many patients with melanoma delayed presentation due to lack of information regarding lesion recognition.The checklist was revised in 1989, dividing the features into three major and four minor features. 11ince 2015, the National Institute for Health Care Excellence (NICE) advocated the weighted 7PCL for primary care clinicians' assessment of pigmented skin lesions 12 based on the evidence of two studies.Appendix 1. 13 The first study prospectively studied 688 people at increased risk of melanoma over 10 years. 14esions were screened by naked eye examination and dermoscopy, but the study authors centre their conclusion on results from the dermoscopic algorithm version of the 7PCL, which has different criteria based on pattern analysis devised by Argenziano et al. 15 The second study analysed the original and weighted 7PCL in 1436 lesions in primary care using naked eye examination only, however only 36 lesions proved to be melanoma. 16NICE updated its evidence in July 2022 but continues to advocate the 7PCL.The evolution of the 7PCL can be seen in Table 1.
In dermatology, the ABCD/E and 7PCL are both commonly accepted CPRs for assessing pigmented skin lesions and have overlapping features.For this review, only the 7PCL is included as it includes all the features of the ABCDE CPR as well as three unique features as shown in Table 2.

| Objectives
Evaluate the frequency that each feature of the 7PCL presents in histologically proven melanomas in human adults and consider if the checklist is a robust CPR to assess pigmented skin lesions in adult humans.

| Methods
The Preferred Reported Items for SRs and Meta-Analysis (PRISMA) 17 was applied to ensure robust, transparent reporting. 18The protocol was registered to the International prospective register of SRs (PROSPERO), 19 registration number CRD42022 339359.To minimise errors, data extraction was managed using Covidence, 20 an online SR management programme.

| Eligibility criteria
Eligible studies evaluated adult human patients with clinically suspicious pigmented skin lesions that were: assessed using the 7PCL or included assessment of features of the 7PCL and referred for specialist opinion; assessed by patient and/or clinician; provided histological diagnoses.Excluded studies did not have histological diagnoses; used dermoscopy or other digital assessment tools; assessed acral lesions; included cosmetic lesions; focused on other skin cancers/ conditions.

| Information sources
A systematic literature search was conducted Feb 2022 of the following databases: CINAHL Plus; AMED -The Allied and Complementary Medicine Database; MED-LINE; ASSIA; Cochrane; Scopus.Prospero was included to search for grey literature and avoid work

What is already known about this topic?
� There are currently 2 lesion recognition tools in frequent use.ABCDE for the public and the seven-point check list for General practitioners.Some features included in the tools overlap and some have been called into question.� Studies have assessed the sensitivity and specificity of the tools as a whole, but no systematic review has assessed the frequency that individual features present in histologically proven melanoma.

What does this study add?
� The most frequent features seen in biopsy proven melanomas are changes in size, colour and shape.The least frequent features are itch, inflammation, and crusting -also seen in many benign lesions.� Focussing on changes in features rather than 'irregular' features may improve melanoma recognition and reduce the number of benign lesions referred for specialist opinion.

| Data items
The main data outcome was the frequency of occurrence of each of the 7PCL features in histologically proven

| Data synthesis
Extracted data is presented as percentages of features seen, with the number of melanomas in the study shown as (n = ).Statistical information will be presented where this was provided by the study.
Due to the small number of studies identified and the vast heterogeneity, there was insufficient data for meta-analysis, so findings will be presented in a narrative synthesis. 24

| Literature search
Seven studies met the eligibility criteria for inclusion.The PRISMA flow chart (Figure 1) includes the rationale for why some papers that initially met the criteria were ultimately excluded.

| Study characteristics
2457 participants were included (mean: 405.85, median: 159) -excluding one study not stating the number of participants, only the number of lesions assessed. 2594 were females (mean: 249, median: 74.5) and 874 males (mean: 124.86, median: 49).Four studies collated ethnicity 16,[26][27][28] of these, participants were white or Caucasian, which is consistent with typical epidemiology of melanoma. 29Only one study provided skin type data. 27Age data was challenging to analyse as studies presented this data differently.Results are included in Table 3.

| Study design
All studies were observational without interventions.Two stated their methodological approach. 16,31One was a retrospective modified case-control, 31 the other a diagnostic validation study. 16The remaining studies were considered to be case-control studies 25,26 or case studies. 27,28,30Ethical approval was only sought for two studies. 26,31Four studies received funding or support from cancer charities. 16,27,28,31One study declared competing interests. 16

| Study aims
The study aims can be seen in Table 3.One study did not aim to assess the 7PCL but was included as it provided data for five of the seven features. 28

| Study methodologies
Of the four patient-assessed studies, one utilised questionnaires 28 and three used interviews. 26,27,31The study using questionnaires 28 extracted data from a larger study investigating delayed cancer presentation among 15 different cancer types. 32Results of this study must be interpreted with caution as the only skin-related symptom was "Mole Itchiness Rash or red spots," 32 (p.584)The questionnaire included a free text section, but results were a low frequency of occurrences for some features compared to other studies.One study interviewing patients asked questions directly related to the 7PCL. 27Two studies used open-ended nondirective questions that were recorded and categorised to features of the 7PCL. 25,31wo clinician-assessed studies were prospective. 16,25One examined lesions identified from a previous randomised control trial assessing a novel computerised diagnostic tool. 16The other examined lesions presenting to a pigmented skin lesion clinic. 25he third retrospectively examined the case notes of patients with histologically proven melanoma from a computerised database over a given period. 30ue to the low numbers of studies, results were not divided for subgroup analysis.

| RESULTS ACROSS STUDIES
Results of frequency of occurrence of each feature of the 7PCL, including statistical analysis where provided, can be seen in Table 4.
Whilst the included studies demonstrated vast heterogeneity and inherent bias, the average frequency of occurrence of the 7PCL was calculated to identify trends in results, which are displayed in Figure 2. No adjustments were made for the study that only included five of the seven features of the 7PCL. 28Further interpretation of these results is considered in the following discussion.
T A B L E 3 Study characteristics.T A B L E 4 Results of frequency of occurrence of each feature of the seven-point checklist (7PCL) including statistical analysis where provided.

Author and year;
Walter 2013 16 Osborne 1999 30 Healsmith 1994 25 Bränström   NICE advocated the 7PCL to assess pigmented skin lesions, 28 so the terminology did not always match NICE definitions, therefore affecting results.The original 1986 7PCL 10 preceded the more formal introduction of evidence-based medicine 33 and appears based on expert-led anecdotal evidence and case studies.
The book published as an update in 1989 11 included a reference list, but no direct references to supporting evidence.
The first study that assessed the individual features of the original 7PCL was published 4 years later. 34ther studies have since reviewed the 7PCL, 14,[35][36][37] but this is the first known SR assessing the frequency at which features of the 7PCL appear in histologically proven melanoma.Figure 1 includes the reasons these studies were excluded from this SR.
This SR found size change was the most frequently reported feature.Each study, apart from one, 25 used different definitions to NICE, but were comparable enough to assess a change in lesion size.One study also included 'new lesions' in their definition, 31 but no data was provided to differentiate between a change in size and new lesions.
Colour was the second most frequent feature, however, none of the studies used "irregular colour" as defined by NICE.A common benign mimic of melanoma -seborrhoeic keratoses (SK) -frequently present with irregular colours 38 and are often a cause of concern to patients.SK can arise from solar lentigos, so do change appearance over a long time period, so the patient lesion history is vital to clinical assessment.Two of the included studies 26,27 reported that 'darkening' or 'very dark' lesions were a significant observation among patients.Bränström 26 concluded 'very dark' might be added to future recommendations for identifying melanoma.
Whilst most melanomas are pigmented, amelanotic melanomas are typically red or pink 39 This less common subtype is difficult to clinically diagnose due to the pigment absence.None of the studies in this SR recording subtypes of melanoma reported amelanotic melanomas, likely due to their rarity and absence of colour.
Irregular shape and border were the third most common feature.All studies used only shape or border in their definitions.Another benign mimic of melanoma, solar lentigos, frequently present with irregular shapes or borders, 40 which may increase benign referrals.
Nodular melanomas -another infrequent melanoma subtype 41 -tend to stay symmetrical but enlarge rapidly.They are also associated with higher mortality 42 so it is important that any CPR assessing skin lesions can also identify nodular melanomas.
Diameter ≥7 mm was the fourth most commonly occurring feature, however, only 4 studies specifically reported lesions ≥7 mm.For this review, the studies recording lesions 'larger than others' 16 and 'up to 1 cm' 27 were removed from the average calculations for diameter, as including them would not accurately reflect the NICE recommended 7PCL of ≥7 mm.The study including "larger than others" 16 as their diameter definition reported the second highest frequency of occurrence, but this definition may fit better with the 'ugly duckling' sign discussed later.
The ABCDE checklist includes diameter of 6 mm or more.One study assessing both checklists, found the ABCDE tool missed 5 melanomas (7.7%) that had changed in size because they were less than 6 mm. 25 These lesions were picked up by the 7PCL because of the 'change in size'.Whilst some studies argue diameter of ≤6 mm is a useful parameter, 43,44 another proposes diameter of ≥7 mm can be called into question. 45I G U R E 2 Average frequency of occurrence for each feature of the seven-point checklist.
CONGDON and DAVIS Definitions used for sensory changes also differ throughout the studies.One study included 'itch' and 'altered sensation' in their definition. 16Three used only itch; in their definition [26][27][28] whilst the others referred to sensory changes 25,31 or altered sensations. 16,30he original 7PCL booklet states itch is a common feature of malignant and benign lesions, 10 however, a study of biopsy-proven skin cancers and SK in elderly males found only 5.3% (n = 39) of patients with melanoma reported itch, compared to 25.5% (n = 224) of patients with SK. 46 du Vivier et al. 27 also observed many benign lesions were referred because of itch alone.
Inflammation definitions were consistent with NICE.MacKie describes inflammation as "a red inflammatory flare around the edge"10 (p, 6) of a lesion not seen in benign lesions.Conversely, du Viver et al. 27 noted inflammation was a common reason for referral of benign lesions.
When assessing oozing and crusting, 'crusting' was used in all definitions except one that used only 'bleeding'. 26.MacKie reasoned bleeding is "very common" 10 (p, 6) in early melanomas.Other studies argue bleeding (and itch) are usually late signs of malignancy. 26,47The National Cancer Institute supports this, 48 yet the NICE advocated 7PCL does not include bleeding and only assigns one point to oozing and crusting.
Throughout these studies, change is persistently regarded as the most significant feature of malignancy in pigmented skin lesions.Osborne reported 90% of lesions examined reported a history of change. 30A study examining patients' use of the ABCD rule suggests using only 'evolution' in public campaigns, 49 an opinion previously argued by Weinstock et al. 50acKie et al. incorporated change into the 7PCL in 1990, 51 yet the NICE version of the 7PCL only includes change in size.Shape, border, and colour are assessed as 'irregular'.This may explain the large number of benign lesions referred for specialist opinion such as solar lentigos that can present with irregular shapes or borders, 40 and SK, which often have irregular colours, can itch, bleed, and become easily inflamed. 52hilst change is key to identifying melanoma, patients have expressed difficulty in recalling how lesions looked previously. 53,54To help identify changes, patients are encouraged to photograph lesions of concern and engage the assistance of a friend or relative to assist self-monitoring of difficult-to-see areas.A study of skin examination in partnerships compared to solo examinations, found patients in partnerships were more likely to engage in skin examination with greater efficiency. 54Smartphone apps have also been developed to assist the public in monitoring skin lesions, 55 however many older patients who are at higher risk of melanoma 56 do not own smartphones and the accuracy of apps is debatable. 57,58tients are also educated observe for new lesions.O'Shea et al. 28 reported change in pre-existing moles in 76% of participants.Conversely, du Vivier et al. 27 found ¾ of the lesions assessed were new.A meta-analysis of the prevalence and implications of nevusassociated melanoma, found 70.9% of melanomas developed as new lesions. 59'New' lesions are not currently a feature of the 7PCL.
Patients are also advised to recognise lesions that look different from the rest, often referred to as 'ugly ducklings'.Walters assessment of diameter as "larger than others" can be viewed as differentiating lesions from others.Liu 31 observed that some participants commented on 'ugly' lesions, but Bränström 26 believes these features are subjective.In clinical practice it is usually the lesions that stand out as different from the rest that are examined more closely with dermoscopy.
This review has focused on the clinical examination of pigmented lesions.After a suspicious lesion has been detected by the naked eye, dermoscopy can aid making a clinical diagnosis.While some primary care clinicians use dermoscopy to assess pigmented skin lesions, it requires training and regular practice to maintain competence, and is thus not available for use by the public.It is therefore fundamental that any CPR is reliable and robust to help identify early melanoma, something which can improve the quality of life of patients and result in cost savings. 4The findings of this SR may influence the revision of the current 7PCL or the development of a new CPR to assist in the assessment of pigmented skin lesions by both the public and primary care clinicians.

| Strengths and limitations of review
This review was limited to a small number of observational studies without interventions.Including 'signs' and 'symptoms' of melanoma in the search criteria may have found further studies examining features of melanoma that could be aligned to features of the 7PCL.
Only features of the 7PCL were included in this review on the basis that it encompassed the features of the ABCDE CPR, however, some studies contend the two CPR's differ in their ability to identify melanoma. 26,31,37A SR of both CPR's also concluded differences between them, 60 however, this was based on the overall CPR's, rather than analysis of individual features.
Whilst averages were used to identify trends in results, heterogeneity between study methods and varying definitions for features of the 7PCL used means these findings should be interpreted with caution.Due to the limited studies available, only the frequency of presentation of features has been reviewed, while this may assist in the assessment of pigmented lesions, further studies providing sensitivities and specificities for each feature would be more likely to minimise This SR identified the most frequently occurring features of melanoma involve the shape size and colour, however a change in these features are more likely to indicate melanoma than 'irregular' features as defined by the 7PCL advocated by NICE.Features involving sensation, oozing, crusting, and inflammation present less frequently and are commonly found in benign mimics of melanoma.They can also be signs of advanced melanomas, so require serious consideration when assessing lesions using the 7PCL.Further studies designed to assess the sensitivities and specificities of each feature of the 7PCL would provide better guidance for melanoma identification and may guide revision of the 7PCL or development of a new CPR for clinical melanoma recognition.

Note:
Definitions used by each study are included where they differ from those of the NICE definitions.a % calculated from numbers given.b Up to 10 mm (<0.5 =

Author MacKie 1985 10 MacKie 1989 11 MacKie 1990 51 NICE 2015 12
T A B L E 1 Evolution of the Seven Point.Checklist This table shows the changes in terminology used and the order the features were presented as the checklist evolved.

assessed by: patient/clinician Total number of lesions examined Number of histologically proven melanomas Findings. Frequency of features are shown greatest to smallest including any statistical analysis where given. Definitions are shown as used in each study Quality assessment (% of 'yes' answers)
-CONGDON and DAVIS T A B L E 3 (Continued)