Reproducibility of the clinical criteria (ABCDE rule) and dermatoscopic features (7FFM) for the diagnosis of malignant melanoma.

Dermatoscopy improves the sensitivity and the specificity in the diagnosis of melanoma. Although the reproducibility of dermatoscopic features has been the subject of research, no study up to now has compared the reproducibility of dermatoscopic features to the reproducibility of the clinical criteria of the ABCDE rule. For this reason we decided to examine the reproducibility of the clinical ABCDE rule and of our diagnostic dermatoscopic method 7FFM, as well as of the individual criteria of both. A total of 73 dermatologists attended three dermatoscopic courses and examined a set of clinical and dermatoscopic slides of 50 pigmented skin tumors. Agreement % and K value for a kappa statistical analysis have been calculated to evaluate inter-rater reliability. The clinical and the dermatoscopic methods showed similar values of concordance: clinical score 2 mean agreement = 68%, mean K = 0.44; clinical score 3 mean agreement = 73%, mean K = 0.61; 7FFM mean agreement = 83%, mean K = 0.64. The clinical criteria A, B, and C and the dermatoscopic features of our method presented similar values of concordance as well: clinical criteria mean K range 0.35-0.25, dermatoscopic features mean K range 0.62-0.25. The dermatoscopic features of our method 7FFM show a good reproducibility after a short training program, similar to the reproducibility of the clinical criteria of the ABCDE rule for the diagnosis of melanoma.

The dermoscopic (7FFM) versus the clinical (ABCDE) diagnosis of small diameter melanoma.

The diagnosis of small diameter melanoma, that is early melanoma, is particularly difficult. For this reason we decided to evaluate the improvement given from our diagnostic dermoscopic method 7FFM to the clinical diagnosis, ABCDE rule, of small diameter melanoma. A retrospective study evaluating the clinical and the dermoscopic slides of 76 small diameter melanomas observed from January 1 1993 to December 31 1998, and of 524 small melanocytic nevi consecutively observed from September 1 1997 to September 30 1999, has been undertaken. The sensitivity and the specificity of the ABCDE rule and of our diagnostic dermoscopic method 7FFM in the diagnosis of small diameter melanoma have been calculated. The difference of diagnostic power between the two methods has been calculated with chi square test. The sensitivity and the specificity of the ABCDE rule in the diagnosis of small melanomas were 47.3% and 56%, while the sensitivity and the specificity of our method 7FFM were 68.8% and 86%. The difference of diagnostic power between the two methods was statistically significant: P<0.01 for both sensitivity and specificity. The sensitivity of the two methods together was 81.5% while the specificity of the two methods together was 50.6%. Our results show that our diagnostic dermoscopic method 7FFM improves both sensitivity and specificity in the clinical diagnosis of small diameter melanomas. Anyway the clinical and the dermoscopic diagnosis are not mutually exclusive as the best sensitivity is obtained with the two methods together.

Reproducibility of a dermoscopic method (7FFM) for the diagnosis of malignant melanoma.

Dermoscopy improves sensitivity in the diagnosis of melanoma. We have developed a new diagnostic dermoscopic method which evaluates only seven dermoscopic features or criteria. We called the method Seven features for melanoma (7FFM). To present and to evaluate the reproducibility of our dermoscopic diagnostic method (7FFM) we held eight dermoscopic courses from 10/4/96 to 04/03/98 in various Italian cities. In fact the reliability of a diagnostic test or method mainly depends on the level of agreement in the interpretation of results among different observers. Only methods with good agreement can be used in clinical practice for the diagnosis of melanoma. Many dermatologists (207) attended one of the eight dermoscopic courses: each course was one-day in length and at the end of the course the participants evaluated a set of 25 dermoscopic slides using our dermoscopic method. Percentages of concordance and K values for a kappa statistical analysis to evaluate inter-rater reliability have been calculated. The method showed a mean percentage of concordance of 85.7%, median 88%, a mean K value of 0.699, median 0.684. These data point to a good agreement level. Our method shows good reproducibility after a short training program.

The dermoscopic versus the clinical diagnosis of melanoma.

Dermoscopy is an useful technique for the diagnosis of melanoma. All the diagnostic dermoscopic methods developed until now have been tested on dermoscopic slides and the real improvement given by this technique to the clinical diagnosis based upon ABCDE criteria is still unknown. For this reason, we decided to undertake a prospective study to compare the clinical diagnosis made with ABCDE criteria, to the dermoscopic diagnosis, made with the method 7FFM, developed by us. A total of 401 lesions were evaluated clinically and dermoscopically. On the basis of the number of the clinical criteria considered sufficient to diagnose a melanoma, various clinical scores, ranging from 1 to 5, can be obtained; data about the sensitivity of the clinical diagnosis of melanoma suggest that the most often used score is score 3. Our method 7FFM presents a sensitivity and a specificity better than those obtained with score 3: 80% versus 66.6% and 89.1% versus 79.3%. Besides this, we have evaluated if the sensitivity of our method 7FFM in the diagnosis of melanoma can be improved with the adjunct of the clinical criteria or of the clinical scores. The best values of sensitivity 93.3% and predictive value negative 97.3% have been obtained with score 2 plus 7FFM. Our results confirm that our method can be used in the screening of pigmented skin lesions in daily office practice and that an improvement in sensitivity without an excessive sacrifice of specificity can be obtained with the adjunct of the clinical score 2.

The seven features for melanoma: a new dermoscopic algorithm for the diagnosis of malignant melanoma.

The clinical diagnosis of melanoma has a mean sensitivity of 67%, dermoscopy or dermatoscopy is a non invasive technique which improves this sensitivity. Our purpose was to create a simple dermoscopic method for the diagnosis of melanoma useful in daily office practice. For this reason a training set of 218 cutaneous pigmented lesions was used and scored for 16 dermoscopic features: for each feature sensitivity, specificity and statistical significance were evaluated. The results were used to create a simple dermoscopic diagnostic method of only seven dermoscopic features (7FFM). The method was used to evaluate a test set of 713 pigmented skin lesions consecutively observed. The diagnostic dermoscopic method developed gave a sensitivity of 94.6%, a specificity of 85.5% and an efficiency of 87.6%. Our method improves the sensitivity in the diagnosis of melanoma and can be used for the screening of pigmented skin lesions.

Molecular analysis of cutaneous B- and T-cell lymphomas.

Among extranodal non-Hodgkin's lymphomas, primary cutaneous lymphomas (CLs) represent a consistent group of B- and T-cell malignancies. We investigated the arrangement of Ig and T-cell receptor (TCR) genes, together with the involvement of several oncogenes and the tumor-suppressor gene p53, in a panel of primary cutaneous B- and T-cell lymphomas (CBCLs and CTCLs). Southern blot analysis was performed to detect rearrangements of the Ig, c-myc, bcl-1, bcl-2, bcl-3, bcl-6, and the NFKB2/lyt-10 genes in 52 cases of CBCLs and of the TCR, bcl-3, and NFKB2/lyt-10 genes in 38 cases of CTCLs. tal-1 gene deletions were analyzed in CTCLs by means of polymerase chain reaction (PCR). p53 gene mutations were assayed using PCR, single-strand conformation polymorphism analysis, and direct DNA sequencing in CBCL and CTCL cases. Clonal rearrangements of Ig genes or oncogenes were found in 25 of the 52 CBCLs. In particular, we detected rearrangements of the bcl-1 locus (2 cases), the bcl-2 gene (2 cases), the NFKB2/lyt-10 gene (2 cases), and the bcl-6 gene (1 case); interestingly, 4 of these cases showed a germline arrangement of the Ig genes. Clonal rearrangements of TCR genes were detected in 37 of the 38 CTCLs. Rearrangements of the NFKB2/lyt-10 gene were present in 2 cases and tal-1 gene deletions in 3 CTCL cases; p53 gene mutations were detected in 1 CTCL case. Overall, our data indicate that (1) clonal rearrangement of Ig genes is frequently undetectable by means of Southern blot in CBCLs (60%); (2) genetic lesions are involved in a limited but significant fraction of primary CLs showing a molecular marker of clonality (13/62; 20%); and (3) rearrangements of the bcl-1, bcl-2, or bcl-6 loci, associated with specific subsets of nodal lymphoid neoplasias, are rarely observed in CBCLs. Moreover, our results suggest that tal-1 gene deletions may play a pathogenetic role in non-acute T-cell malignancies and that, in the context of lymphoid malignancies, CLs may represent a favorable target for the possible oncogenic potential of the NFKB2/lyt-10 gene.

bcl-1, bcl-2, p53, c-myc, and lyt-10 analysis in cutaneous lymphomas.

In the present study we investigated the pathogenetic role of c-myc, bcl-2, and lyt-10 oncogenes, bcl-1 locus, and p53 suppressor gene in a representative panel of cutaneous lymphomas, including 25 cases of cutaneous B cell lymphoma (CBCL) and 29 cases of cutaneous T cell lymphoma (CTCL). In our analysis four cases of CBCL were found rearranged for bcl-2 and two for the bcl-1 locus. Two cases of CTCL and one case of CBCL were found rearranged for lyt-10. No rearrangements of c-myc oncogene were found in CBCL. Analysis of p53 gene showed mutation only in one case of mycosis fungoides in tumoral stage, at codon 163 of p53 gene (TAC-->CAC; Tyr--> Asp). Our data suggest that in primary CBCL bcl-2 oncogenes and bcl-1 locus are rarely involved. Furthermore, in primary CTCL p53 gene is not affected at significant frequency. The occurrence of p53 mutation in a patient affected by mycosis fungoides in tumoral stage may represent an involvement of p53 gene in tumor progression of CTCL, a finding observed in several types of human cancer.

Delta chain-positive T-cell lymphoma of the skin.

Cutaneous T-cell lymphomas (CTCL) represent a heterogeneous group with respect to clinical presentation, histology, and phenotype. The most frequent CTCL represent a proliferation of x/x+ T-cell with a helper inducer memory phenotype (CD4+, CD29+, CD45RO+).