Study of AI-Powered Spectroscopy Device Shows Promise for Skin Cancer Detection

DermaSensor Inc. is touting encouraging results from a study that evaluated the safety and efficacy of an Elastic Scattering Spectroscopy (ESS) device (DermaSensor) in assessing skin lesions suggestive of skin cancer. The study was conducted to provide clinical evidence for European regulatory authorities and initial effectiveness evidence for the U.S. Food and Drug Aministration (FDA). The device has been under review by the FDA since early 2023; it is not currently available for sale in the United States.
 
The prospective, multi-center clinical validation study, conducted at four investigational sites in the United States, focused on patients with skin lesions suggestive of both melanoma and non-melanoma skin cancers. These lesions underwent clinical evaluation by expert dermatologists and by the DermaSensor device, a handheld artificial intelligence (AI)-powered spectroscopy device that is non-invasive and wireless. Key statistical metrics, such as sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and Area Under the Curve (AUC), were used to assess performance.
 
The study demonstrated an overall sensitivity of 97% (n=169) for the study’s test dataset. When examining specific skin cancer types, subgroup sensitivity rates were 96.7% for melanoma (n=30), 97% for basal cell carcinoma (n=72), and 97% for squamous cell carcinoma (n=67). Importantly, the device’s sensitivity and overall accuracy (i.e., AUC) were each within 1% of the sensitivity and accuracy of the expert study dermatologists’ in-person clinical assessments when enrolling the lesions. The device exhibited a 22% specificity for benign lesions biopsied by the dermatologists, and the device’s overall NPV was 90%, while PPV was 58%. A PPV of 58% equates to a number needed to biopsy (NNB) of 1.7 for detecting one skin cancer.¹
 
“This study shows that the DermaSensor device has high performance in detecting malignancies and should help PCPs by improving their decision-making regarding appropriate referrals of skin cancer,” said Dr. Armand Cognetta, program director and professor at the FSU College of Medicine and senior study author. 

Skin cancer is a global health concern, with increasing incidence rates worldwide. Artificial Intelligence (AI) is expanding its presence within dermatology with several tools available abroad, although no FDA-approved skin cancer AI imaging tools are currently available in the United States. 
 
The study’s findings reinforce results from other recent analyses, which found the device improves PCPs’ detection of skin cancer,² and the device performance was similar across patient skin types.³ With the high sensitivity in detecting all skin cancers in this study, DermaSensor may play a crucial role in assisting primary care clinicians in evaluating suspicious skin lesions. This 
 
“This study was a key early comparative effectiveness study showing that our device can bring dermatologist-level triage to primary care. We have completed eight clinical studies for review by regulatory bodies in the U.S. and Europe, and we hope to soon be the first-ever device on the U.S. market that provides PCPs with an immediate, objective result for skin cancer,” DermaSensor Co-Founder/CEO Cody Simmons stated. “While dermatoscopes are in widespread use by dermatologists and have been in dermatology clinical guidelines since 1989, only 8% of PCPs currently use one given the significant training time required. We hope to soon have available our first-in-class device for America’s 300,000 PCPs, and that its simple point-and-click method of use will result in widespread adoption.”
 
References
¹ Manolakos D, Patrick G, Geisse JK, Rabinovitz H, Buchanan K, Hoang P, Rodriguez-Diaz E, Bigio IJ, Cognetta AB, Use of an Elastic-Scattering Spectroscopy and Artificial Intelligence Device in the Assessment of Lesions Suggestive of Skin Cancer: A Comparative Effectiveness Study. Journal of the American Academy of Dermatologists International (2023), doi: https://doi.org/10.1016/j.jdin.2023.08.019.
² Seiverling EV, Agresta T, Cyr P, Caines L, Nguyen VL, Chatha K, Siegel DM. Clinical Utility of an Elastic Scattering Spectroscopy Device in Assisting Primary Care Physician’s Detection of Skin Cancers. Journal of Clinical and Aesthetic Dermatology (2023), 16(4 Supple): s16-17.
³ Patrick G, Leffell D, Merry SP, Croghan I, Rabinovitz H. Cognetta A. Clinical Capabilities of a Handheld Elastic Scattering Spectroscopy- Artificial Intelligence Device as an Adjunctive Tool for Evaluating Skin Cancer in Skin of Color. Presentation at the American Academy of Dermatology Summer Meeting, Tampa Bay, Florida (Aug 2023).