Topical Therapy for Cutaneous Precancerous Lesions: a Digitally-enabled Randomized Controlled Trial of Fluorouracil compared with Fluorouracil/Calcipotriene

Actinic keratoses (AKs) are common pre-cancerous skin lesions that can undergo transformation to cutaneous squamous cell carcinoma (cSCC), the second most common cancer in the United States. When there are multiple AKs in one body area, field treatment with the topical chemotherapy 5-fluorouracil (5-FU) for 2-4 weeks can be used, which reduces the risk of cSCC by 75%. 5-FU has substantial downsides, however: patients apply 5-FU at home for 2-4 weeks and expected effects include weeks to months of intense redness, irritation, crusting, and erosions, which can negatively impact utilization. Our prior work has shown that 5-FU is used in only 20% of patients who may have been candidates for field treatment. To address issues of duration and symptoms, 5-FU combined with topical calcipotriene, a vitamin D analog used in psoriasis, for a shorter duration of 4 days (combination treatment; ComboTx) has been proposed. A proof-of-concept trial of ComboTx showed a decrease in AKs. It is unknown whether this shorter-course ComboTx is non-inferior to the lengthier single-agent 5-FU regimen. Furthermore, the goals of using ComboTx (improved patient satisfaction, decreased symptoms) are unstudied. Comparing 5-FU and ComboTx head-to- head is critical for evidence-based AK treatment and cSCC prevention. To best center the patient experience in treating AKs to prevent cSCC, we also need rapid assessment of response, convenient to patients. Quantifying AKs using digital photography and artificial intelligence (AI)- supported algorithms could greatly improve our ability assess clinical response when treating AKs and support cSCC prevention remotely. However, it is unknown whether digital photography can be used to quantify AKs, and unclear whether an AI-supported image segmentation algorithm could automate AK quantification for telemonitoring. In Aim 1 we will perform a dual-site randomized non-inferiority clinical trial comparing 5-FU to ComboTx in patients with multiple AKs. We will evaluate both medical outcomes (e.g., change in AK count, number of cSCCs diagnosed) and patient reported outcomes (e.g., patient satisfaction, adverse events, adherence). In Aim 2 we will compare in-person counts of AK lesions to counts using standardized digital photography, explore whether an AI-supported algorithm could automate AK counts, and determine the potential of vision language models to reliably grade the severity of precancerous conditions. The results of this trial will have clinically significant implications for AK treatment and cSCC prevention, materially influence healthcare practice, and support development of novel methods for efficient clinical trial deployment around AK counts, which will be extendable to other remote monitoring applications in other skin cancer types and dermatology more broadly. Project Number: 1R01CA303720-01A1 | Fiscal Year: 2026 | NIH Institute/Center: National Cancer Institute (NCI) | Principal Investigator: Mackenzie Wehner | Institution: UNIVERSITY OF TX MD ANDERSON CAN CTR, HOUSTON, TX | Award Amount: $704,028 | Activity Code: R01 | Study Section: Cancer Prevention Study Section[CPSS] View on NIH RePORTER: https://reporter.nih.gov/project-details/11364262