Richard R. Winkelmann, DO; Gregory Nikolaidis, MD; Darrell S. Rigel, MD, MS; Natalie Tucker, BS; Laura Speck, MD
Dr. Winkelmann is from Rigel Dermatology, New York, New York.
Dr. Nikolaidis and Dr. Speck are from Westlake Dermatology, Austin, Texas.
Dr. Rigel is from the Department of Dermatology, New York University School of Medicine, New York, New York.
Ms. Tucker is from MELA Sciences Inc., Irvington, New York.
Disclosure: Dr. Winkelmann’s research fellowship is funded in part by MELA Sciences Inc. Drs. Nikolaidis and Speck have worked as consultants for MELA Sciences Inc. Ms. Tucker is employed by MELA Sciences Inc. Dr. Rigel is a consultant to MELA Sciences Inc.
Abstract
Objective: To observe how a multispectral digital skin lesion analysis device was used by dermatologists in a community-based clinical setting and determine differences from a university-based environment. Design: Use of multispectral digital skin lesion analysis was incorporated into a community-based practice by 12 dermatologists across six clinics over seven consecutive days with the data provided by the device integrated as an adjuvant to their clinical evaluation for their pigmented lesion management decisions. Multispectral digital skin lesion analysis results were collected electronically for lesions prior to biopsy, and histopathological evaluation was performed for the biopsied lesions. Multispectral digital skin lesion analysis and pathology results were then compared to assess the degree of morphological disorganization. Setting/participants: Study of 160 consecutive patients in community-based clinical setting. Measurements: Proportion of “low” and “high” disorganization lesions identified by multispectral digital skin lesion analysis. Results: Of the 344 pigmented skin lesions analyzed by multispectral digital skin lesion analysis, 255 were high disorganization, 113 of which were biopsied. Of the 89 lesions evaluated by multispectral digital skin lesion analysis to be low disorganization, seven were biopsied and all pathology was benign. Data demonstrate a higher rate of multispectral digital skin lesion analysis low disorganization readings for pigmented skin lesions (32% for single use per patient lesions, p<0.0001; 26% for all lesions, p<0.0001) than observed in the pigmented lesions clinics providing data for the university-based clinical study (10%). Conclusion: Multispectral digital skin lesion analysis in the community-based clinical setting may outperform specificity results from the university-based clinical trial study, perhaps because of a higher proportion of subtle lesions encountered at high-risk pigmented lesion clinics of participating major academic centers as compared with those in a community-based practice setting. (J Clin Aesthet Dermatol. 2015;8(2):16–18.)
Early diagnosis of melanoma is critical for patient survival. New technology may improve the accuracy of dermatologist decisions to biopsy suspicious pigmented lesions.[1] Multispectral digital skin lesion analysis (MSDSLA) (MelaFind®; MELA Sciences Inc., Irvington, New York) provides objective information about the morphological disorganization of clinically ambiguous pigmented skin lesions. Dermatologists then incorporate the additional information provided by MSDSLA into their biopsy decision. Objective probability information provided by this device has been studied for its potential to enhance the accuracy of dermatologists evaluating suspicious lesions for biopsy.[2]
MSDSLA uses narrow bands of visual and near-infrared light (430–950nm) to capture data from the skin surface down to 2.5mm in depth. At a 20-micron resolution (~diameter of 3 melanocytes), individual cellular atypia is not evaluable by MSDSLA. Alternatively, the device uses 75 analytical parameters to determine the level of morphological disorder within a clinically atypical pigmented skin lesion and generate a classifier score (CS).[3] Using a binary output, a CS greater than or equal to 0 is considered to have “high” disorganization and scores less than 0 have “low” disorganization. High disorganization was associated with 98-percent sensitivity to melanoma and severely dysplastic nevi. Low disorganization was associated with significantly higher specificity than study dermatologists (10% vs. 4%, p=0.022) and a 98 percent negative predictive value.[4]
The safety and effectiveness of MSDSLA were originally established from data analyzing 1,632 skin lesions collected by physicians at multiple major academic centers.[4] In this primarily academic pigmented lesion center based study, 10 percent of the lesions had low disorganization results. However, the sampling of benign lesions from that study may not be representative of what would be seen in the general population. The use of MSDSLA and the distribution of high and low disorganization lesions in the community-based practice clinical setting have not yet been studied. The purpose of this study is to observe how the MSDSLA device was used by dermatologists in a multi-physician, multi-clinic, community-based clinical setting and to measure the distribution of high and low morphological disorganization to determine if a difference exists from what was found in academic-based centers.
Methods
Twelve dermatologists in one community-based practice group across six clinics were trained to use MSDSLA. Dermatologists were instructed to identify suspicious or atypical pigmented skin lesions for which the necessity of biopsy was borderline. Following analysis with MSDSLA, dermatologists were instructed to include the additional MSDSLA information into their clinical evaluation and subsequent biopsy decision. Data were collected over a seven-day period. All lesions analyzed with MSDSLA were required to meet the United States Food and Drug Administration (FDA)-approved labeling of the device. Not all eligible lesions analyzed with MSDSLA were necessarily biopsied as the final biopsy decision was made by the clinician or refused by the patient. For example, a clinician may have elected to biopsy an atypical pigmented lesion in a patient with a strong personal or family history of skin cancer despite a “low” disorganization MSDSLA result. Degree of morphological disorganization was obtained for all evaluated lesions and pathology reports were subsequently collected for the lesions selected for biopsy. Lesions were considered “positive” for biopsy diagnoses of melanoma, moderate-to-severe dysplastic nevi, and atypical melanocytic proliferation. Findings of this study were compared to data observed in the academic center based MSDSLA clinical trial using chi-square analysis.[4]
Results
A total of 353 skin lesions from 160 patients were selected for evaluation with MSDSLA. Nine lesions did not meet MSDSLA evaluation criteria and were excluded. Of the 344 skin lesions evaluated with MSDSLA over seven consecutive days, 255 were classified as ‘‘high” disorganization, 113 of which were biopsied. Pathology results of these lesions included one melanoma, one atypical melanocytic proliferation, 63 dysplastic nevi (57 mild, 5 moderate, 1 severe), 43 nevi, two seborrheic keratoses, and three solar lentigines. Eighty-nine lesions were found to have “low” disorganization by MSDSLA and the seven lesions selected for biopsy were all benign (1 mild dysplastic nevus, 6 nevi) (Table 1). The distribution of CSs was similar to that seen in the university-based clinical study (-2.6 – +6.2 vs. -5.2 – +9.0) (Figure 1). Data showed a higher rate of MSDSLA low disorganization readings of pigmented skin lesions (32% for single-use per patient lesions [p<0.0001], 26% for all lesions [p<0.0001]) than observed in the university-based clinical study [10%]).
Discussion
A higher proportion of suspicious pigmented lesions in this community-based clinical study was found to have low disorganization readings by MSDSLA than was found in the academic-based clinical trial (26% vs. 10%, p<0.0001). In the community-based dermatology clinical setting, benign lesions may be encountered more often and may be easier to differentiate than those presenting to pigmented lesion academic centers catering to high-risk patients. The observed high:low disorganization ratio is consistent with the significant findings of a recent MSDSLA study performed on a different patient set.5 These findings suggest that the actual specificity of MSDSLA when used in a community-based setting may be higher than what was reported in the university-based clinical trial.
Conclusion
Using objective information provided by MSDSLA as an adjuvant to the clinical evaluation of suspicious pigmented lesions may enhance the accuracy of biopsy decisions by dermatologists in community-based settings. Additionally, if the specificity of MSDSLA is indeed higher than reported in the university-based clinical trial settings, the implications of MSDSLA’s impact on our healthcare system may be greater than previously thought. Findings of this study support that MSDSLA has the potential to further improve the accuracy of dermatologists while reducing additional costs due to unnecessary biopsies of benign pigmented lesions.
references
1. Ferris LK, Harris RJ. New diagnostic aids for melanoma. Dermatol Clin. 2012;30(3):535–545.
2. Gutkowicz-Krusin D, Elbaum M, Jacobs A, et al. Precision of automatic measurements of pigmented skin lesion parameters with a MelaFind multispectral digital dermoscope. Melanoma Res. 2000;10(6):563–570.
3. Winkelmann RR, Yoo J, Tucker N, et al. Assessment of a diagnostic predictive probability model provided by a multispectral digital skin lesion analysis device for melanoma and other high risk pigmented lesions and its impact on biopsy decisions. J Clin Aesthet Dermatol. 2014;7(12):16–18.
4. Monheit G, Cognetta AB, Ferris L, et al. The performance of MelaFind: a prospective multicenter study. Arch Dermatol. 2011;147(2):188–194.
5. Winkelmann RR, Rigel DS, Kollmann E, et al. Negative predictive value of pigmented lesion evaluation by multispectral digital skin lesion analysis in a community practice setting. 2014. Accepted: J Clin Aesthet Dermatol.