4. Diagnostik – Aktinische Keratose und Plattenepithelkarzinom der Haut

Study

Aims

Design

Population

Outcomes

Results

Comments and methodological assessment

LoE

Chen et al 2013

To assess the reliability of four different methods used to quantify AKs and to investigate whether a consensus meeting affects the reliability.

Single-blinded, cross-sectional study of 12 experienced dermatologist raters counting AKs on the face and ears of nine subjects before and after a consensus meeting

To compare the degree of agreement of (i) AK count of discrete lesions > 0.25 cm (big), (ii) AK count of discrete lesions < 0.25 cm (small); (iii) total count, irrespective of size, and (iv) the body surface area method

N=12 board-certified dermatologists

67% were women

Mean age: 47 years ± 9

N=9 consecutive subjects with extensive AKs on their face and ears were invited to participate in a research kick-off meeting for AK assessment,

100% male

intraclass correlation coefficient (ICC) among raters for pre- and post-consensus evaluations

total count’ method had the greatest ICC for both pre- (0.18, P=0.04) and post-consensus (0.66, P≤ 0.0001) assessments.

Total count was also the only pre-consensus ICC for which the null hypothesis of no association among assessments was rejected.

Small sample size

No histological confirmation of AK was obtained

Patients with AK to be rated were only male

2

Cockerell 2000

To propose criteria for grading keratinocytic intraepidermal (malignant) neoplasia.

Modification of an existing classification system based on a literature review

à expert review/opinion

-

Classification system

Proposed criteria for grading keratinocytic intraepidermal (malignant) neoplasia.

Grade I:

Clinical: Flat, pink macule or patch on solar-damaged skin; background mottling; no roughness or hyperkeratosis.

Histologic: Focal atypia of basal keratinocytes of lower one third of the epidermis.

Grade II:

Clinical: Pink to red papule or plaque with rough, hyperkeratotic surface; variable induration. Histologic: Focal atypia of keratinocytes of at least the lower two thirds of the epidermis; focal hyperkeratosis, alternating orthokeratosis and parakeratosis with sparing of acrotrichia and acrosyringia; prominent acanthosis and buds of keratinocytes into the upper papillary dermis; may see some involvement of upper acrotrichia and acrosyringia.

Grade III:

Clinical: Red, scaly indurated plaques on sun-damaged skin; may be pigmented. Histologic: Diffuse atypical keratinocytic proliferation involving the full thickness of the epidermis diffusely; parakeratosis, acanthosis, papillomatosis, the involvement of adnexal structures.

No validation with real-world data of the system

No inter-rater reliability or other measure were investigated

Only expert review

5

Cockerell et al 2005

To propose a classification for AK analogous to that used for cervical intraepithelial neoplasia.

Modification of an existing classification system based on a literature review

à expert review/opinion

-

Classification system

Criteria for grading keratinocytic intraepidermal (malignant) neoplasia.

Grade I:

Clinical: Flat, pink macule or patch on solar-damaged skin; background mottling; no roughness or hyperkeratosis.

Histologic: Focal atypia of basal keratinocytes of lower one third of the epidermis.

Grade II:

Clinical: Pink to red papule or plaque with rough, hyperkeratotic surface; variable induration. Histologic: Focal atypia of keratinocytes of at least the lower two thirds of the epidermis; focal hyperkeratosis, alternating orthokeratosis and parakeratosis with sparing of acrotrichia and acrosyringia; prominent acanthosis and buds of keratinocytes into the upper papillary dermis; may see some involvement of upper acrotrichia and acrosyringia.

Grade III:

Clinical: Red, scaly indurated plaques on sun-damaged skin; may be pigmented. Histologic: Diffuse atypical keratinocytic proliferation involving the full thickness of the epidermis diffusely; parakeratosis, acanthosis, papillomatosis, involvement of adnexal structures.

No validation with real-world data of the system

No inter-rater reliability or other measure were investigated

Only expert review

5

Dirschka et al 2017

To develop and perform an initial pilot validation of a new easy-to-use quantitative tool for assessing AK severity on the head.

1. review of other severity scoring systems in dermatology, in particular the psoriasis area and severity index (PASI)

2. Initial validation was performed by 13 physicians assessing AK severity in 18 AK patients and two controls using a physician global assessment (PGA) and AKASI.

To determine an AKASI score, the head was divided into four regions (scalp, forehead, left/right cheek ear, chin and nose). In each region, the percentage of the area affected by AKs was estimated, and the severities of three clinical signs of AK were assessed: distribution, erythema and thickness.

N=18 patients with AK on the head [mean age (range): 73 years (60–80); 10 men and eight women] and two controls without AK (62-year old man and 70-year-old woman)

N=13 physicians were involved in the rating

AKASI & PGA scores

Correlation between AKASI and PGA (Pearson correlation coefficient)

AKASI & PGA scores

mean (SD) PGA score was 0.08 (0.27) for the controls and 1.74 (0.80) for the patients. The mean (SD) AKASI score was 0.11 (0.38) for the controls and 4.75 (2.51) for the patients

strong correlation between AKASI and PGA scores (Pearson correlation coefficient: 0.86). AKASI was able to discriminate between different PGA categories: mean (SD) AKASI increased from 2.88 (1.18) for ‘light’ to 5.33 (1.48) for ‘moderate’, 8.28 (1.89) for ‘severe’, and 8.73 (3.03) for ‘very severe’ PGA classification. The coefficient of variation for AKASI scores was low and relatively constant across all PGA categories.

This work was funded by Centroderm GmbH.

An initial 30-min training on AKASI was provided,

and the physician participants were provided with a handout containing detailed instructions on AKASI.

Small-medium sample size of both patients and physicians.

2

Dréno et al 2017

To develop, test, and validate an Actinic Keratosis Field Assessment Scale (AK-FAS) based on photographic clinical cases.

Development:

initial draft of the AK-FAS was based on a combination of Olsen criteria for AK. This draft was then dynamically modified in a face-to-face meeting on 15 July 2016.

Validation:

108 standardized photographs of patients representing the full range of AK severity were collected.

Six investigators (involved in the development) and 2 untrained investigators independently rated each photograph according to 3 criteria (AK-FAS scale): AK area (total skin area affected by AK lesions), hyperkeratosis and sun damage.

The final AK-FAS was tested separately on face and scalp areas (on 66 and 30 photographic cases, respectively) by 6 trained (testing) and 2 untrained investigators (validation).

Inter-rater reproducibility

Validation of the AK-FAS by investigators

Inter-rater reproducibility: For face and scalp combined, the inter-rater κ scores were 0.69, 0.71 and 0.51, respectively

substantial agreement for 2 criteria (AK area and hyperkeratosis) and moderate agreement for the third criteria (sun damage)

Intra-rater reproducibility: good for all criteria (AK area, hyperkeratosis and sun damage)

substantial agreement range for all investigators for AK area, almost perfect or substantial agreement range for all but one of the investigators for hyperkeratosis and moderate agreement range for the majority of the investigators for sun damage

Validation of the AK-FAS by untrained investigators

κ scores of 0.59, 0.54 and 0.38, respectively, for AK area, hyperkeratosis and sun damage, indicating moderate agreement (good reproducibility) for AK area and hyperkeratosis, and fair agreement for sun damage; s imilar results were obtained for the face and scalp analysed separately

àValidation of the AK-FAS showed good reproducibility for AK area and hyperkeratosis, even for dermatologists untrained on use of the scale.

Assessment for each area was repeated (with at least 1 h between assessments) to allow evaluation of inter- and intra-rater agreement.

The order of the photographic cases was randomly changed in the repeat assessment, in order to minimize grading by memory. The investigators had not previously seen any of the cases provided by their colleagues and were not allowed to confer during the assessment.

This study was funded by LEO Pharma.

2

Fernández-Figueras et al 2015

To evaluate the prevalence of classic and differentiated pathways in the development of cutaneous iSCC.

The epidermis adjacent to and overlying iSCC, assumed to be representative of pre-existing lesions, was histologically studied in 196 skin biopsy specimens showing iSCC by three pathologists.

N=196 skin biopsy specimens showing iSCC

(79 women and 117 men)

mean age:77.3 years (±10.1)

Most lesions (108 out of 196) were located on the face.

mean diameter of the selected biopsy specimens was

19.30 mm [median, 17 mm ± 13.01 mm; interquartile range (IQR), 10.00–27.50 mm]. The mean width of iSCC areas was 9.05 mm (median, 7 mm; ±7.17 mm; IQR, 3.50–13.00 mm). The anatomic level of infiltration was II in two cases (1.0%), III in 14 cases (7.1%), IV in 127 cases (64.8%), and V in 53 cases (27.0%). The mean maximum depth of infiltration was 3.83 mm (median, 3.00 mm; ±2.85 mm; IQR 2.00–5.00 mm).

Features of the epidermis overlying iSCCs and at the edge of iSCCs, distributed according to the degree of involvement and the presence of ulceration (number of cases and corresponding percentages of total)

Adnexal involvement by atypical cells (proliferative AK) according to the thickness of atypical changes in the epidermis overlying iSCCs

AK I, AK II and AK III lesions overlying iSCC were present in 63.8%, 17.9% and 18.4% of cases respectively. The corresponding percentages in the epidermis adjacent to iSCC were 77.9%, 6.6% and 8.3% respectively (stage could not be assessed in 8.1% of cases). Focal epidermal ulceration overlying iSCC was seen in 32% of AK I, 28.6% of AK II and 33.3% of AK III instances.

Adnexal involvement by atypical keratinocytes (proliferative AK) was present more frequently in cases with overlying AK I (39/125, 31.2%) than with AK II (8/35, 22.9%) and AKII I (5/36, 13.9%).

Almirall has provided financial support for this study but has not participated in the elaboration and discussion of results.

Baseline characteristics of included participants were unbalanced.

2

Fernández-Figueras et al 2018

To demonstrate that follicular extension of an AK is associated with the depth of invasion (Breslow) of iSCC.

retrospective histologic review of 193 biopsy specimens of iSCC with an associated AK by three dermatopathologists

assessment of the presence and depth of follicular extension of atypical keratinocytes in the AK, using tumour (Breslow) thickness and the follicular unit level (infundibular, isthmic and subisthmic), as well as iSCC being present directly adjacent to the follicular basalis

N=193 biopsy specimens

Follicular extension

Follicular extension was present in 25.9% of the cases (50 cases), usually extending into the lower follicular segment. . In 12 of them, the atypia reached the upper portion of the follicle (infundibulum), and in three of them (25%), there was evidence of invasive squamous cell carcinoma (iSCC) originated from the follicular basalis. In 33 cases, the atypical cells reached the isthmic portion, and in 21 of them (63.6%), iSCC originated from the follicular basalis. In five cases, the atypical cells reached the isthmic portion and in all of them iSCC originated from the follicular basalis.

The iSCC was present directly adjacent to the follicular basalis in 58% of the cases (29 cases), correlating highly with the depth of follicular extension (infundibular: 3/12; isthmic: 21/33; subisthmic 5/5).

all biopsy specimens smaller than 3 mm and larger than 2.5 cm in size and those specimens with poor architectural preservation (i.e. superficial or fragmented specimens and tumours with extensive exulceration) were excluded

2

Ianhez et al 2013

To evaluate the actinic keratoses counting by various raters and suggest approaches to increase the reliability.

Cross-sectional study: forty-three patients were evaluated by four raters (inter- and intra-rater assessment) on the face and forearms.

rating:

two board-certified dermatologists and two residents of Dermatology

N=43 patients

Age: 63.1 years (range: 50–80 years)

25 females

mean actinic keratoses counts on the face and forearms were 7.7 and 9.1

Inter-rater agreement

Intra-rater agreement

The overall agreement among the raters for the facial and forearm actinic keratoses was 0.74 and 0.77. The intra-rater assessment showed high rates of agreement for the face (ICC = 0.93) and forearms (ICC = 0.83).

The intra-rater agreement for 0–5, 6–10 and 11–20 counted lesions were: 0.66, 0.29 and 0.30.

All evaluators received a specific training session on counting AKs for the study.

2

Jiyad et al 2017

To provide a detailed evaluation of the consistency of AK counts on digital photographs with clinical examination in renal transplant recipients, using defined areas of skin.

Skin sites of renal transplant recipients were examined clinically and on digital photographs by independent dermatologically-trained examiners.

N=28 patients with 138 skin sites

Mean age: 57 ± 9 years,

the majority were male (67%)

mean ± SD length of time since transplantation was 9 ± 7 years

number of AKs per skin site ranged from 0 to 14 and overall 305 AKs were diagnosed in total across all skin sites

mean ± SD number of AKs on any single skin site was 2 ± 3

Specificity, sensitivity, and Kendall’s tau-b correlation coefficient were calculated based on exact photographic AK counts as well as counts with ± 1 AK tolerance.

When 138 skin sites with 305 clinical AK counts were examined for total count ± 1 AK, the sensitivity and specificity of photography was 95% and 100%, respectively.

There was a significant positive correlation between AK counts on photographs and clinical examination (Tb = 0.537) and correlation was even higher for total count ± 1 AK (Tb

= 0.758).

àmoderate to strong concordance between AK counts on digital photographs and clinical examination

Blinding was performed

2

Röwert-Huber et al 2007

To propose a system that classifies an AK as an SCC, in conjunction with an atypia grading system.

Review based on existing classification systems; modification of the existing system of Cockerell

Expert opinion

-

-

The authors recommend an AK classification system that describes these lesions as squamous cell carcinomas (SCCs), using the terminology ‘early in situ SCC Type AK I’, ‘early in situ SCC type AK II’ and ‘in situ SCC type AK III’.

No validation of the system; only expert opinion

4

Schmitz et al 2018a

To determine the association between chronically UV-induced tumours such as basal cell carcinomas or squamous cell carcinomas and AKASI.

retrospective analysis of patients who had undergone oncological surgery due to UV-induced tumours and who were assessed for AKASI and Physician’s global assessment (PGA) before surgery.

AKASI assessment and PGA were performed by one investigator.

Statistical analysis of the correlation between AKASI, PGA and invasive carcinomas

N=210 patients were included

Median age: 77 years (42–95)

66.2% were male

61.9%: FST II

52.4%: history of NMSC

N=626 lesions

Histological evaluation revealed AKs in 298 (46.9%), BD in 88 (13.8%), SCC in 32 (5.0%), BCC in 118 (18.6%) and other tumours such as seborrhoeic keratosis in 104 (16.4%) cases

AKASI

PGA

Statistical differences

Correlation (Spearman)

patients exclusively presenting AKs (n = 106) showed a median (range) AKASI of 4.6 (0–15.5) and PGA of 2.0 (0–4.0)

Patients with solely noninvasive tumours (n = 43) such as AKs and BD had a median (range) AKASI of 5.0 (0.6–11.2).

Significantly higher AKASI in patients with SCC compared to patients with non-invasive lesions like AK and Bowen disease (BD) (P=0.0275).

Spearman’s coefficient of rank correlation between AKASI and PGA indicates that these measures of AK severity were strongly correlated (P < 0.0001; r = 0.90; 95% CI 0.865–0.920).

Large sample size; retrospective design

AKASI and PGA have been evaluated by only one investigator

2

Schmitz et al 2018b

To determine whether there is a correlation between the commonly used histological classification scheme for AK lesions proposed by Röwert-Huber and basal epidermal growth patterns of AK.

To investigate accompanying factors such as vascular density, inflammation and involvement of adnexal structures.

Retrospective analysis of histologically confirmed AK lesions occurring on the head/face from patients seen in routine practice

Determination of histological grade (AK I-III), basal growth patterns of atypical keratinocytes (crowding, budding and papillary sprouting) and accompanying parameters by two investigators independently from each other.

N=246 lesions were included

Median age: 79 (56– 94) years

92.3% male

30.9% of lesions located on the scalp

histological grade (AK I-III)

basal growth patterns of atypical keratinocytes (crowding, budding and papillary sprouting)

accompanying parameters

correlation of basal growth patterns and histological AK grades

Of the 246 lesions included, 28.0% were histologically classified as AK I, 46.7% as AK II and 25.2% as AK III.

Approximately 26.4% of the basal growth patterns were classified as crowding (pro I), 49.6% as budding (pro II), 17.9% as papillary sprouting (pro III) and 6.1% without basal directed growth.

No significant correlation of the histological AK IIII grading and underlying growth patterns was observed (P = 0.4666). However, adnexal structure involvement (OR = 2.37; 95% CI 1.21–4.65), infiltration (OR = 2.53; 95% CI 1.31–4.90) and increased number of vessels (OR = 2.56; 95% CI 1.42–4.65) were independent positive predictive markers for pro II and pro III basal growth patterns.

-

2

Schmitz et al 2019a

To investigate a possible relationship between basal growth patterns of AKs adjacent to iSCC.

Retrospective study

Histological assessment of the epidermis overlying and adjacent to iSCCs

Determination of the histological grade (AK I–III), basal growth pattern (PRO I–III) and accompanying parameters such as adnexal involvement.

Assessments were performed by two investigators independently from each other.

N= 307 iSCC lesions were included

73.6% male

Median age: 81 (76–86)

Tumour thickness (mm), median: 1.9 (IQR 1.1–3.9)

59.6% of patients with a history of more than one iSCC

Most of the tumours were well differentiated (85.3%) and invaded into the reticular dermis (75.9%)

histological grade (AK I–III), basal growth pattern (PRO I–III) and accompanying parameters

Among 307 lesions, 52.4% of AKs were histologically classified as AK grade I, 38.1% as AK II and 6.8% as AK III (v2-test, P < 0.001). Only 2.6% of adjacent epidermal samples did not show any atypical keratinocytes. The epidermis adjacent to iSCCs was classified as having a PRO I basal growth pattern in 25.7%, PRO II in 31.9% and PROIII in 39.4% of cases. Only 2.9% of AKs showed no basal growth (v2-test, P < 0.001). In total 118 AKs (48.8%) showed extension into adnexal structures. These AKs were graded as PRO I in 18.6% of cases, PRO II in 30.5% and PRO III in 50.8%. The epidermis above iSCCs could be assessed only for upwards-directed growth and showed no significant differences in the three AK grades (P = 0.42).

adjacent and overlying AKs

were assessed, but the investigators were not able to evaluate the original AK from which the iSCC had originated. Changes at the mutational level most probably lead to the final progression of this distinct subtype of AK.

2

Schmitz et al 2019b

To compare the inter-rater reliability of the established classification of upwards directed growth of AKs (AK I–III) and the recently established classification of basal growth pattern of AKs (PRO I–III).

Retrospective study

Histological images of 54 AKs were classified by 21 independent dermatopathologists with regard to basal proliferation (PRO I–III), histological grade (AK I–III) and assumed risk of progression into invasive carcinoma.

N=54 AKs

Raters: 21 expert dermatopathologists, thereof 12 dermatologists (57.1%) and nine pathologists (42.9%)

mean age: 43.4 (±9.6) years and a mean experience in dermatopathology of 10.9 (±8.6) years

Classification of AK images with regard to basal proliferation (PRO I–III), histological grade (AK I–III) and assumed risk of progression into invasive carcinoma

Inter-rater reliability

16.7% of AK (9/54) were classified as AK I, 66.7% (36/54) as AK II, and 16.7% (9/54) as AK III. With regards to basal growth pattern, 25.9% (14/54) were classified as PRO I, 42.6% (23/54) as PRO II, and 31.5% (17/ 54) as PRO III. We observed a highly significant inter-rater reliability for PRO-grading (P < 0.001) which was higher than for AK-grading (Kendall’s W coefficient: AK = 0.488 vs. PRO = 0.793). We found substantial agreement for assumed progression risk for AKs with worsening basal proliferation (k = 0.759) compared to moderate agreement (k = 0.563) for different AK-gradings.

-

2

Schmitz et al 2016

Determine whether there is a correlation between the commonly used clinical classification scheme for AK lesions proposed by Olsen et al. with that of the histological classification system developed by Roewert-Huber et al.

Retrospective analysis

One AK lesion from patients in three pivotal clinical studies and routine practice was assessed clinically and histologically. A match in grading was defined as Olsen grade 1 being classified histologically as AK I, Olsen grade 2 as AK II and Olsen grade 3 as AK III.

Patient population for this study came from three randomized clinical trials (ALA-AK-CT002, ALA-AK-CT003 and ALA-AK-CT007) and routine clinical practice.

N=892 patients

Mean age: 71.6±7.3

83.4% male

64.0% of lesions located on face/forehead

Classification of AK according to Olsen and Röwert-Huber

correlation between clinical and histological classification

29.0% of lesions were classified as Olsen grade 1, 59.6% as Olsen grade 2 and 11.3% as Olsen grade 3; 19.2% were histologically classified as AK I, 69.6% as AK II and 11.2% as AK III.

Only 480 lesions (53.8%) had a matching clinical and histological classification. Of these matches, most were ‘Olsen grade 2 = AK II’ (83.1%).

The Spearman’s rank correlation coefficient for clinical and histological classification was r=0.0499 (P=0.137).

-

2

Sinnya et al 2015

To compare the inter-observer agreement between trained observers for AK counts based on photographs compared with clinical AK counts.

Clinical AK counting was carried out in 2 sessions by 4 dermatologically trained clinicians with 6 months or more of dermatological experience and a senior consultant dermatologist with over 25 years of experience, whose counts served as practical reference standard.

The observers repeated counting from the same photographic images, 2 weeks later.

N=6 patients (n=3 immunocompetent, n=3 OTRs)

84% male

Mean age: 60 years±15

Agreement of AK counts:

interclass and intraclass correlation coefficients (ICC)

ICC for agreement across 5 observers for the photographic AK counts was 0.63 (95% CI 0.48–0.78) compared with 0.79 (95% CI 0.68–0.88) for clinical counts

ICC for agreement between photographic AK counts and clinical counts ranged from 0.66 (95% CI 0.48–0.81) to 0.84 (95% CI 0.73–0.91) across the 5 observers

interclass correlation coefficient for agreement between AK counts based on the same photographs but counted on 2 separate occasions ranged from 0.86 to 0.99

Small sample size;

Reference standard was the assessments of the senior consultant

3

Yantsos et al 1999

To propose a classification system for AK.

Expert review/opinion

-

Classification system

Criteria for Grading KIN

Grade I:

Clinical: Undetectable or fiat, pink macule or patch on solar damaged skin; background mottling; no roughness or hyperkeratosis. Histological: Focal atypia of basal keratinocytes of lower one-third of the epidermis.

Grade IIa:

Clinical: Pink to red papule or plaque with rough, hyperkeratotic surface; minimal induration. Histological: Focal atypia of keratinocytes of the lower two thirds of the epidermis; alternating orfho and parakeratosis with sparing of acrotrichia and acrotrichia.

Grade IIb:

Clinical: Similar to Ila but more induration, more hyperkeratosis and/or more erythema; all KIN lesions (other than KIN III) on lip and conjunctiva. Histological: Focal atypia of keratinocytes of at least the lower two-thirds of the epidermis; focal hyperkeratosis, often greater than Ila; one or more additional features such as acantholysis, the involvement of adnexal structures, prominent acanthosis and buds of keratinocytes into the upper papillary dermis present.

Grade III:

Clinical: Red, scaly indurated plaques on sun-damaged skin; may be pigmented; seen on other sites such as mucosa in Bowenoid papulosis and erythroplasia of Queyrat; non sun-damaged skin in squamous cell carcinoma in situ induced by arsenic.

Histological: Diffuse atypical keratinocytic proliferation involving the full thickness of the epidermis; parakeratosis, acanthosis, papillomatosis, involvement of adnexal structures.

No validation of the system with real-world data or assessment of inter-rater reliability

Only expert opinion

5

Study

Aims

Design

Population

Outcomes

Results

Comments and methodological assessment

LoE

Câmara et al 2016

To evaluate comparatively the influence of histopathological features on epithelial dysplasia (ED) and the effectiveness in usage of WHO and binary grading systems in actinic cheilitis (AC).

Cytological and architectural alterations established by WHO for ED were evaluated in 107 cases of AC.

Epithelial dysplasia was graded using WHO (no, mild, moderate, and severe ED) and binary systems (low risk and high risk for malignant transformation).

The comparisons were performed using kappa, chi-square, and phi coefficient tests (P < 0.05).

Lower lip biopsies from 107 cases of AC were retrieved from Universidade Federal Fluminense.

Sample was composed of 60 males (56.1%) and 47 females (43.9%) with ages ranging from 21 to 86 years (mean of 59 years). There are only light-skinned types (phototypes I, II, or III)

presence and grade of ED

Most cases were classified as mild ED (44.5%) in the WHO system and as low risk for malignant transformation (64.5%) in the binary system. There was a positive correlation between WHO and binary systems (k = 0.33; P < 0.0002). Loss of basal cell polarity (P < 0.001) was associated with severity of ED grade in the WHO system. Anisonucleosis (P < 0.0001), nuclear pleomorphism (P < 0.0001), anisocytosis (P = 0.03), cell pleomorphism (P = 0.002) increased nuclear/cytoplasm ratio (P < 0.0001), increased nuclear size (P < 0.0001), increased number of mitotic figures (P = 0.0006), and dyskeratosis (P = 0.008) were associated with severity of ED grade in the binary system.

No calibration exercises were attempted by the evaluators.

Unclear whether blinding was performed.

3

Cavalcante et al 2008

To analyze the clinical and histological features of actinic cheilitis (AC).

Clinical evaluation of 29 patients with AC, incisional biopsies for confirmation of clinical diagnosis.

Histological features were analyzed, and dysplasia was classified as mild, moderate, or severe. The x² test was used for the following variables: gender, age, race, and smoking habits. The degree of dysplasia was related to these variables (Fisher’s test) to test for independence between them (P<0.05).

N=29 patients

72.41% male,

75.86% were over age 40 years

93.10% were white

72.41% were nonsmokers

Histological features

Classification of dysplasia (mild, moderate, severe)

Associations with baseline demographic features

Clinical findings:

Dryness (100%), atrophy (72.41%), scaly lesions (65.52%), swelling of the lip (62.07%), erythema (58.62%), ulceration (58.62%), blurred demarcation between the lip vermilion border and the skin (58-62%), marked folds along with the lip vermilion (48.28%), white spots or plaque (41.3%)s, crusts (34.48%), blotchy areas (27.59%), and areas of pallor (17.24%).

Histological findings:

Dysplasia (100%), Elastosis (100%), inflammatory infiltrate (100%), vasodilatation (100%), Hyperplasia and/or acanthosis (86.21%), hyperparakeratosis (58.62%), hyperorthokeratosis (55.17%), epithelial atrophy (55.17%), granulosis (37.92%, hypergranulosis (37.93%), parakeratosis (17.24%), orthokeratosis (3.45%)

Dysplasia was mild in 10.34% of the patients, moderate in 27.59%, and severe in 62.07%.

Gender, age, race, or smoking habits were not related to the degree of dysplasia in the sample.

Unclear whether blinding was performed.

3

Pilati et al 2008

To determine the histopathologic findings in actinic cheilitis (AC) and lip squamous cell carcinomas (LSCC) to attempt to predict the evolution from AC to LSCC based on the comparison of two dysplasia classification systems.

Histopathologic features were evaluated according to the World Health Organization classification of dysplasia and binary system of classification. Also, in LSCC, pattern, stage of invasion, and degree of keratinization were evaluated. A total of 58 cases of AC and 70 cases of LSCC were studied, and data correlation was performed using statistical analysis.

N= 58 cases of AC

Histological features according to the WHO classification and binary system classification

presence of dyskeratosis and keratin pearls was found to be strongly associated with severe dysplasia and could represent higher proximity between the severe dysplasia in AC

changes related to the nuclei, such as hyperchromasia, nuclear pleomorphism, anisonucleosis, increase in the number and size of nucleoli, increased number of mitoses, and atypical mitoses, indicate progression in dysplasia spectrum

Unclear whether blinding was performed.

3

Poitevin et al 2017

To make a proposition of a clinical score to actinic cheilitis (Grade I starting from dryness of vermilion to endured ulcers representing Grade IV) and to assess its reproducibility.

Fifty subjects were assessed

Obtained data were analyzed by means of descriptive statistics and by Kappa test to assess the interexaminer and the clinical Golden-Pattern concordance (95% CI).

AC Grade I. Dryness and desquamation on the vermilion of lips.

AC Grade II. Atrophy on the vermilion’s border, presenting soft superficies and pallid areas with eruptions. Blurred limit between the lip’s vermilion border and the skin, or a dark line demarking that limit can be seen. This melanotic line should be different from ephelides or other pigmented lesions.

AC Grade III. Rough and squamous areas on the drier parts of the vermilion and hyperkeratotic areas, especially when they spread to the wet lip’s mucosa (border between mucosa and semimucosa).

AC Grade IV. Ulceration present in one or more sites of the lip’s vermillion or Leukoplakia, mainly in more traumatic places, due to the history of pipe or cigarettes consumption. These lesions could suggest that a malignization process would be in progress, especially when they are accompanied by endured areas on palpation.

N=35 patients

20 men and 15 women, 32 Caucasians, 2 mixed ethnicities and 1 black.

28 were farmers, 2 were business people, 2 were homemakers and 3 were students

average age of 46.12 (18–74) years

interexaminer and the clinical Golden-Pattern concordance

During calibration, 15 patients were examined three times a week by each examiner (4) until Kappa test observed k= 0.8 or more. In the main experiment, the inter-examiner concordance was classified between good (k= 0.779; Po0.05) and very good (k= 0.925; Po0.05) from the 35 examined subjects. With the Golden-Pattern, it was considered very good (k= 0.812; Po0.05 to k= 0.925; Po0.05).

Calibration: four examiners studied and discussed it before its application. The examiners were composed of two professors of oral medicine, one student finishing dental School and one dentist with 2 years of general dentistry practice.

The study was conducted in Brazil.

2

Santana et al 2020

To summarize the results of published studies on immunohistochemical biomarkers in lip carcinogenesis, to evaluate if there is a marker that can distinguish the different histological grades of AC.

Systematic review of retrospective studies that investigated immunohistochemical biomarkers in AC defined on standardised histological assessment

Systematic literature search in 5 databases until 25 April 2017: PubMed, Scopus, Web of Science, ScienceDirect and Scielo

Partial grey literature search was conducted as well.

Meta-analysis of protein Ki-67

Qualitative appraisal using the Critical Appraisal Tools from SUMARI.

N=27 retrospective studies were included in the systematic review and n=3 in the meta-analysis

number of AC cases in each study varied from 10 to 70

type of histological grading performed; immunohistochemical biomarkers that were analyzed; expression of biomarkers in each subgroup

Among the studied biomarkers, the ones that were most investigated were DNA repair proteins, with 12 antibodies assessed. The inflammatory markers were the second most assessed group. Other groups of proteins were also analyzed, including apoptosis markers, metalloproteins, cell cycle markers, growth factors, neural and muscle markers:

The proliferation marker Ki-67 was the most studied biomarker and we observed, through meta-analysis, that it was differently expressed between AC and lip cancer, but not in AC subgroups. Ki-67 mean expression

was similar in control groups and was higher in LSCC than in AC.

N=5 articles used the binary classification system

Most studies had a high risk of bias.

Studies from Brazil were overrepresented.

High heterogeneity among studies.

2