FDA Approves AI-Powered Detection Device for Skin Cancers

The FDA has granted clearance for the first real-time, non-invasive skin cancer evaluation system which will be utilized to detect melanoma, basal cell carcinoma, and squamous cell carcinoma, according to a press release from DermaSensor Inc.¹

The approval of this handheld, wireless device was based on data from both the blinded validation DERM-SUCCESS study (NCT05126173), which included 1,579 lesions biopsied in 1,005 patients, and the supplemental melanoma DERM-ASSESS III validation study.^2,3 Investigators reported that the device had a sensitivity rate of 95.5% across all skin cancers, improving primary care providers’ (PCPs) cancer prediction rates by 12.5% (P < 0.0001). With the device the sensitivity rates were 87.5%, 97.8%, and 98.7% when detecting melanoma, basal cell carcinoma, and squamous cell carcinoma, respectively.^1,3 The overall specificity rate of the device was 20.7%.³

“We are entering the golden age of predictive and generative artificial intelligence in healthcare, and these capabilities are being paired with novel types of technology, like spectroscopy and genetic sequencing, to optimize disease detection and care,” Cody Simmons, cofounder and chief executive officer of DermaSensor, stated in the press release. “Equipping PCPs, the most abundant clinicians in the country, to better evaluate the most common cancer in the country has been a major, long-standing unmet need in medicine.”

One in 5 Americans will experience skin cancer by the age of 70, with approximately 5.5 million new cases annually in the United States. Although early detection improves cure rates, access to dermatology remains a challenge, especially in rural areas. Because of this it is important to give PCPs the ability to identify cases in need of referral. Therefore, the DermaSensor AI-powered spectroscopy technology was designed to evaluate suspicious moles in an attempt to enhance diagnostic capabilities beyond traditional methods.¹

“While dozens of companies have attempted to address this problem in recent decades, we are honored to be the first device cleared by the FDA that provides PCPs with an automated tool for evaluation of suspicious lesions,” Simmons added.

The population for this observational study included adults who attended a dermatology clinic for 1 or more skin lesions that were biopsied due to a suspicion of skin cancer.⁴

Exclusion criteria included the presence of any other significant disease or disorder that may have put the patient or study at risk; presence of acral or ungal lesions, as well as skin lesions on mucosal surfaces; lesions greater than the diameter of the dermoscopic lenses; lesions located on an anatomical site unsuitable for photographing, including on the surface of genitals and hair-bearing areas; lesions that have been previously biopsied, excised, treated or otherwise traumatized; or lesions located in an area of visible scarring or tattooing.⁴

This study was conducted across 22 primary care study centers with 18 in the United States and 4 in Australia.³ Notably, clinical care for patients enrolled was provided according to the standard of care for suspicious lesions, and all lesions included in the study were biopsied. Furthermore, statistical analyses included standard diagnostic test parameters of the device for skin cancer detection.³

Of the 1,005 patients included in the study, a majority were female (51.4%) and the average age of patients included was 59 years. Furthermore, the majority of lesions that were biopsied were Fitzpatrick Skin Type III (35.0%), and on average patients had 1 (65.4%) or 2 (20.6%) lesions.³

Prior to biopsy, PCPs predicted that 822 of the lesions would be malignant, of which 322 would be melanoma and 757 would be benign. However, traditional dermatopathology found 224 high-risk lesions, of which 48 were melanoma, 90 were basal cell carcinoma, and 86 were squamous cell carcinoma.³

“Achieving this medical milestone is a testament to the 12 years and tens of millions of dollars our company has invested in research and development to bring this powerful technology to market,” Maurice R. Ferre, MD, cofounder and chairman of DermaSensor, said. “We are incredibly grateful to the FDA for their collaboration and dedication to this area starting with our first FDA pre-submission meeting in 2016. Having begun patient enrollment in our FDA pivotal study in mid-2020, we are now ecstatic to have clearance of our FDA-breakthrough designated de novo submission.”

References

1. FDA clearance granted for first AI-powered medical device to detect all 3 common skin cancers (melanoma, basal cell carcinoma and squamous cell carcinoma). News release. DermaSensor. January 17, 2024. Accessed, February 15, 2024. https://www.dermasensor.com/fda-clearance-granted-for-first-ai-powered-medical-device-to-detect-all-three-common-skin-cancers-melanoma-basal-cell-carcinoma-and-squamous-cell-carcinoma/
2. Device label: indications for use, contraindications, warnings and precautions. DermaSensor. Accessed February 15, 2024. https://support.dermasensor.com/labeling-guidance
3. Merry SP, Chatha K, Croghan I, et al. Clinical performance of novel elastic scattering spectroscopy (ESS) in detection of skin cancer: a blinded, prospective, multi-center clinical trial. J Clin AesthetDermatol. 2023;16(4 suppl 1). https://pub-press.mydigitalpublication.com/publication/?i=787933
4. DERM US and EU validation study. ClinicalTrials.gov. Updated June 28, 2023. Accessed February 15, 2024. https://classic.clinicaltrials.gov/ct2/show/NCT05126173