Welcome to this July twenty twenty-six review from the Journal of the American Academy of Dermatology. In this episode we're covering four articles spanning genomic risk prediction in cutaneous squamous cell carcinoma, a population-based look at melanoma follow-up and survival, a retrospective series on capecitabine for field cancerization, and a practical guideline on perioperative management of patients on systemic cancer therapy. Let's get into it.

Our first article is titled "Genomic Predictors of Perineural Invasion in Cutaneous Squamous Cell Carcinoma: Insights from an MD Anderson Cohort," by King, Lee, Davis, Hayden, Kang, Barker, Lee, Connolly, and LeBoeuf. The clinical question here is whether specific somatic mutations correlate with perineural invasion, or P-N-I, in high-risk primary cutaneous squamous cell carcinoma, since P-N-I is currently identifiable only on final pathology, which limits preoperative risk stratification. This was a retrospective analysis of a publicly available MD Anderson Cancer Center cohort hosted on c-BioPortal, using targeted sequencing via the M-S-K-I-M-P-A-C-T five-hundred-five gene panel. Thirty-nine patients with known P-N-I status were included, and the cohort was enriched for aggressive features such as regional or distant metastasis, tumor diameter of four centimeters or greater, and subcutaneous invasion. Genes mutated in at least five patients were analyzed, and logistic regression assessed associations with P-N-I, with an OncoKB annotator used to filter for oncogenic or likely oncogenic variants.

Nineteen of the thirty-nine patients, or forty-eight point seven percent, had histologically confirmed P-N-I. Mean tumor size was three point three centimeters and mean depth of invasion was nine point two millimeters. On univariate analysis, four genes stood out. D-D-R-2 mutations were present in forty-two point one percent of P-N-I-positive tumors versus five percent of P-N-I-negative tumors, giving an odds ratio of thirteen point eight-two, with a ninety-five percent confidence interval of one point five-two to one hundred twenty-five point six-five, and a p-value of zero point zero-two-zero. F-L-T-3 mutations showed forty-two point one percent versus ten percent, odds ratio six point five-five, confidence interval one point one-seven to thirty-six point six-one, p equals zero point zero-three-two. NOTCH-2 mutations were seen in seventy-three point seven percent versus thirty percent, odds ratio six point five-three, confidence interval one point six-one to twenty-six point four-seven, p equals zero point zero-zero-nine. And NOTCH-1 mutations appeared in seventy-eight point nine percent versus forty percent, odds ratio five point six-two, confidence interval one point three-six to twenty-three point two-seven, p equals zero point zero-one-seven. Strikingly, every single patient with P-N-I harbored at least one mutation among these four genes, compared to only forty-five percent of patients without P-N-I. On the flip side, MAP-three-K-fourteen mutations were associated with the absence of P-N-I, odds ratio zero point one-four, confidence interval zero point zero-three to zero point seven-nine, p equals zero point zero-two-six. When the analysis was restricted to functionally oncogenic variants, the NOTCH-2 association held, odds ratio eight point one-zero, confidence interval one point four-six to forty-five point zero-six, p equals zero point zero-one-seven, but the NOTCH-1 association lost significance, odds ratio three point six-zero, confidence interval zero point eight-seven to fourteen point eight-seven, p equals zero point zero-seven-seven, which the authors attribute to limited statistical power.

For surgical practice, the key takeaway is that P-N-I may represent a distinct genomic phenotype involving differentiation, signaling, and invasion pathways, and genomic profiling could eventually help flag tumors at higher risk for aggressive perineural spread before final pathology returns. That said, this is clearly exploratory data. Limitations include the small sample size of thirty-nine patients, retrospective design, a limited set of clinicopathologic covariates, absence of outcomes data such as recurrence or survival, and the fact that multivariable adjustment for established P-N-I risk factors was not adequately powered. Validation in larger, prospectively characterized cohorts is needed before this translates into clinical decision-making.

Moving to our second article, this is a JAAD Game Changers commentary by Jane Grant-Kels, summarizing the original study "Association between dermatology follow-up and melanoma survival: a population-based cohort study" by Huang, Fleming, Fung, and Chan. The clinical question addressed by the original authors was whether adherence to annual dermatology follow-up after invasive melanoma diagnosis affects melanoma-specific survival, given that current guidelines on follow-up frequency and duration remain inconclusive. This was a retrospective, population-based cohort study of adults diagnosed with primary invasive melanoma between twenty ten and twenty thirteen, followed through the end of twenty eighteen.

The key finding was that only twenty-eight percent of patients were adherent to annual dermatology visits. Predictors of adherence included younger age, female sex, higher income, greater access to dermatologic care, stage two or three disease, a prior history of keratinocyte carcinoma, fewer comorbidities, and having had any outpatient visit in the twelve months preceding melanoma diagnosis. Most importantly, greater adherence to annual dermatology follow-up was associated with reduced melanoma-specific mortality compared to patients with lower adherence. The commentary does not report specific hazard ratios or confidence intervals from the original study within this Game Changers summary, but the directional finding is clear and consistent.

The evidence-based takeaway, as Doctor Grant-Kels frames it, is validation for continued annual dermatologic surveillance of melanoma patients, even years after initial diagnosis, given the observed association with improved disease-specific survival. For practice, this reinforces the value of structured long-term follow-up programs and may support efforts to improve adherence particularly among lower-income patients and those with reduced access to care, since these were identified as vulnerable groups for non-adherence.

Our third article is "Low Dose Long-Term Capecitabine for Refractory Field Cancerization: A Retrospective Cohort Study," by Neff, Mortelliti, Solomon, Karn, Stump, Ran, Silk, and Ruiz. The clinical question is whether long-term, low-dose capecitabine—the oral prodrug of five-fluorouracil—is effective and tolerable for patients with topical-refractory cutaneous field cancerization, a population where repeated surgery carries high morbidity and prior capecitabine data have been largely limited to solid organ transplant recipients with reportedly poor tolerability.

This was a retrospective cohort and survey study from Dana-Farber Cancer Institute and Brigham and Women's Hospital, covering patients treated between January twenty sixteen and September twenty twenty-five. Twenty patients were included, mean age seventy-two point five years, with ten patients, or fifty percent, immunosuppressed, including six solid organ transplant recipients, three with hematologic malignancy, and one with rheumatoid arthritis. Dosing ranged from five hundred to fifteen hundred milligrams twice daily, with a mean daily dose of one thousand six hundred sixty-eight milligrams, given on days one through fourteen of a twenty-one day cycle. The treatment strategy incorporated planned holidays: mean total treatment duration was twenty point eight months, ranging from zero point eight to fifty-seven point five months, comprising an average of twelve point eight months of active cycling and eight months of treatment holiday.

For efficacy, the primary endpoint was the change in annual keratinocyte carcinoma surgeries pre- versus post-capecitabine using a matched-pairs design. In year one, among eighteen evaluable patients, mean annual keratinocyte carcinoma surgeries fell from three point eight-nine to two point nine-four, a twenty-four percent reduction, which did not reach significance, p equals zero point two-nine. In year two, among eleven evaluable patients, surgeries dropped from a mean of three point four-six to zero point nine-three, a seventy-three percent reduction, which was statistically significant, p equals zero point zero-two-one. In year three, among only five evaluable patients, surgeries went from a mean of three point zero to two point two, a twenty-seven percent reduction, not significant, p greater than zero point nine-nine, likely reflecting the small remaining sample.

On tolerability, only two patients, or ten percent, discontinued therapy due to toxicity—a marked contrast to prior transplant-population data cited in the article, where discontinuation rates ranged from twenty to forty-three percent. Three deaths occurred during the study period, all unrelated to capecitabine or to keratinocyte carcinoma, and none of these patients had taken capecitabine in the month preceding death. On the patient-reported side, of seventeen patients offered the survey, twelve responded, a seventy point six percent response rate. Seventy-five percent felt capecitabine improved their skin lesions at least moderately, and eighty-three point three percent preferred capecitabine over alternative treatments.

The evidence-based takeaway for practice is that low-dose capecitabine with planned treatment holidays appears effective in meaningfully reducing keratinocyte carcinoma surgical burden, particularly by year two, and is far better tolerated in this immunocompetent-and-immunosuppressed mixed cohort than prior transplant-focused literature would suggest, with high patient satisfaction and preference. Limitations are the retrospective design and small sample sizes that shrink further at each successive year of analysis, eighteen, then eleven, then five patients, which limits statistical power especially at the three-year mark and introduces survivorship and selection effects among patients who remained on therapy long enough to be evaluable.

Our fourth and final article is "Perioperative Management of Dermatologic Surgery Patients on Systemic Cancer Therapy," by Trager, Gordon, Handfield, Kriplani, Smithy, Rossi, Lee, Nehal, and Connolly from Memorial Sloan Kettering. This is not a primary data study but an expert-based clinical guideline addressing a genuine practice gap: as targeted therapies and immunotherapy expand survival for oncology patients, more of these patients are presenting for dermatologic surgery while actively on systemic treatment, and perioperative management in this setting has not been well defined. The methodology combined multidisciplinary meetings between four Mohs surgeons and two medical oncologists held between October twenty twenty-five and January twenty twenty-six, review of FDA package inserts for each drug class, and a PubMed search from July to September twenty twenty-five for perioperative considerations specific to Mohs micrographic surgery and dermatologic surgery.

For VEGF inhibitors, used in colorectal, renal cell, thyroid, and hepatocellular carcinoma, the concern is impaired angiogenesis leading to poor wound healing, dehiscence, and bleeding. The recommendation is to hold therapy for three to five half-lives preoperatively depending on procedure scale. For long half-life agents like bevacizumab and cabozantinib, the FDA recommends holding three to four weeks preoperatively; the group recommends the same three-to-four week hold for bevacizumab specifically, individualized in discussion with oncology. For shorter half-life agents such as pazopanib, sorafenib, and sunitinib, no formal trials exist, but seven to ten days preoperatively is considered sufficient. Postoperatively, the FDA label suggests holding two weeks after major surgery for most VEGF inhibitors, up to four weeks for bevacizumab, though minor procedures like biopsy may allow resumption after just one week.

For MEK inhibitors, used mainly in BRAF or NRAS-mutated melanoma, major hemorrhagic events including gastrointestinal and intracranial bleeding have been reported in trials, along with wound healing complications; a study of neoadjuvant dabrafenib and trametinib in anaplastic thyroid carcinoma found venous thromboembolism and wound infections led to treatment interruption. Given limited dermatologic-surgery-specific literature, the authors recommend holding forty-eight hours prior to surgery and restarting seven days postoperatively or once wounds are healed.

For BRAF inhibitors, used in BRAF V600E and V600K mutated unresectable metastatic melanoma, the main perioperative concern is paradoxical MAPK activation causing hyperproliferative and squamoproliferative lesions, including keratoacanthomas and cutaneous squamous cell carcinomas—not impaired wound healing; in fact, BRAF inhibitors may enhance wound healing through improved epithelialization and angiogenesis. The group recommends baseline dermatologic exam, exams every two months while on therapy and up to six months after discontinuation, photosensitivity counseling, and generally safe continuation through the perioperative period in coordination with oncology.

For combination BRAF and MEK inhibitor therapy, given its superior progression-free survival over BRAF inhibition alone, the recommendation is to hold both agents forty-eight hours preoperatively. Because dabrafenib alone has no data suggesting impaired wound healing or bleeding risk, it can be resumed as early as one day postoperatively, while trametinib, given its wound-healing and bleeding signal, should wait seven days post-surgery, again in conjunction with oncology.

For the BCL-2 inhibitor venetoclax, used in chronic lymphocytic leukemia, small lymphocytic lymphoma, and acute myeloid leukemia, the dominant perioperative concern is neutropenia and infection, with febrile neutropenia occurring in four to eight percent of patients across trials. The recommendation is preoperative complete blood count review using existing interval labs and close postoperative monitoring for infection signs and symptoms.

For tyrosine kinase inhibitors, cabozantinib, which targets VEGFR2, carries VEGF-inhibitor-like risks including thrombosis, hypertension, and gastrointestinal perforation; given its ninety-nine hour half-life, the manufacturer recommends stopping at least twenty-eight days before surgery, and safety of postoperative resumption timing is not established. Axitinib, targeting VEGFR one through three, carries bleeding and wound-healing risk with no formal wound-healing studies conducted; the drug label recommends withholding at least twenty-four hours before scheduled surgery, with resumption based on clinical judgment of wound healing. The provided text transitions into lenvatinib at this point but does not include full detail in this excerpt; the full manuscript also covers cytotoxic chemotherapies, immunomodulators, proteasome inhibitors, and antibody-drug conjugates per its abstract, though those sections are not detailed in the material reviewed here.

Across all agent classes, the unifying theme is individualized, multidisciplinary decision-making: preoperative laboratory assessment, drug-specific timing of interruption based on half-life and documented wound-healing or hematologic risk, and close postoperative monitoring, weighed against the oncologic risk of delaying systemic therapy. As a guideline rather than a controlled trial, its main limitation is that it reflects institutional experience and literature synthesis rather than prospective outcomes data, so recommendations should be applied with clinical judgment and direct oncology coordination on a case-by-case basis.

That closes out this July twenty twenty-six review. Across these four articles, the throughline is precision: genomic signatures that may one day flag perineural invasion risk before final pathology, follow-up adherence data reinforcing the survival value of routine dermatologic surveillance in melanoma, real-world evidence supporting long-term low-dose capecitabine as a tolerable chemoprevention strategy for field cancerization, and a practical framework for safely sequencing dermatologic surgery around an increasingly complex landscape of systemic cancer therapies. Thanks for listening.