The study, conducted by Joseph McGrath, Evelyn Fagan, Kaeli Samson, Yue Zhan, and Adam Sutton, was approved by the University of Nebraska Medical Center’s institutional review board and carried out in accordance with the Declaration of Helsinki, with written informed consent obtained from every participant. The research team set out to systematically compare what patients say about their own skin lesion histories against clinical verification, a comparison that matters enormously because the entire architecture of skin cancer surveillance in many health systems rests on self-reported data gathered in intake questionnaires and verbal histories.

Actinic keratoses are rough, scaly patches that develop on chronically sun-exposed skin, most often in fair-skinned, older adults. They are technically keratinocyte dysplasias rather than frank malignancies, but they occupy a critical position on the continuum of cutaneous carcinogenesis. A subset of them can progress to squamous cell carcinoma, and their presence signals what dermatologists call field cancerization, a state in which broad swathes of skin have sustained enough ultraviolet damage that multiple lesions can arise simultaneously and independently across the affected area. Keratinocyte carcinomas, meanwhile, encompass basal cell carcinoma and squamous cell carcinoma, the two most frequent cancers in fair-skinned populations worldwide and cancers whose incidence continues to climb as populations age and cumulative ultraviolet exposure accumulates.

The problem with relying on patient recall for these conditions is baked into their biology and their clinical course. Actinic keratoses are frequently treated with destructive methods such as cryotherapy, in which liquid nitrogen obliterates the lesion without any tissue being sent to a pathology laboratory. A patient who has had a dozen such treatments over twenty years may genuinely not know whether any given rough patch was confirmed as an actinic keratosis, whether it was something benign like a seborrheic keratosis, or whether it was an early squamous cell carcinoma that was frozen off preemptively. The diagnostic label often exists only in billing codes or in the memory of a clinician the patient saw years ago, if it exists at all.

Keratinocyte carcinomas present a related but distinct recall challenge. Basal cell carcinomas are rarely life-threatening but frequently recur, and patients with one history of nonmelanoma skin cancer face substantially elevated odds of developing subsequent lesions, a relationship established in meta-analytic work going back to Marcil and Stern’s widely cited 2000 review in the Archives of Dermatology. Squamous cell carcinomas carry greater metastatic potential, making accurate history-taking even more consequential. If a patient underreports a prior squamous cell carcinoma, a clinician may underestimate risk and calibrate follow-up too loosely. If a patient overreports, the result can be unnecessary anxiety, redundant biopsies, and wasted health system resources. Either direction of error degrades the precision of epidemiological studies that depend on self-reported cancer histories as their exposure variable.

This is not the first time the validity of patient self-report in dermatology has been scrutinized. A landmark 2004 study by Ming and colleagues, published in the same Archives of Dermatology, examined the validity of patient self-reported history of skin cancer and found meaningful discordance between what patients reported and what medical records could substantiate. That work, along with the more recent effort by Kitrell, Crew, Wysong, and Sutton in 2023 to refine the classification of field cancerization, frames the conceptual backdrop against which the new Nebraska-led study operates. The field cancerization concept itself, originally coined in the head and neck oncology literature and since extended to sun-damaged skin, implies that patients with field disease are precisely the ones most likely to have long, complicated lesion histories that strain the limits of lay recall.

Health literacy adds another layer of complexity. A 2024 systematic review by Chang and colleagues in the Journal of Prevention examined the role of health literacy in skin cancer preventive behavior and highlighted how unevenly patients understand even basic distinctions between lesion types. Many patients use the phrase skin cancer loosely, applying it to anything their doctor froze, burned, or scraped. Others with genuine histories of malignancy may not register the diagnosis because the encounter was brief, the terminology was unfamiliar, or the lesion was treated in a setting where pathology confirmation never occurred. In epidemiological surveys, in genetic studies recruiting participants on the basis of skin cancer history, and in clinical trials of field therapies, these misclassifications propagate quietly through the data.

The methodological stakes extend beyond the clinic into the research enterprise itself. Countless studies of keratinocyte carcinoma risk factors, from tanning bed use to immunosuppression to genetic susceptibility loci, rely on questionnaires asking participants whether a physician has ever told them they had skin cancer. Validation studies like the one from the University of Nebraska team provide the sensitivity and specificity estimates that allow researchers to correct for misclassification statistically. Without such grounding, effect estimates can be biased toward or away from the null in unpredictable ways, and the direction of bias often differs for actinic keratoses, which are over-reported when patients conflate them with benign barnacles of aging, versus keratinocyte carcinomas, which may be under-reported when pathology confirmation was never communicated clearly.

The practical implications for clinical practice are equally significant. Dermatologists performing surveillance on patients with a history of field cancerization already know that these individuals require more frequent examinations, and the new findings underscore that the history a patient volunteers should be treated as a starting point rather than a settled fact. Wherever possible, clinicians and researchers seeking accurate keratinocyte carcinoma histories should corroborate patient report with pathology records, tumor registries, or prior biopsy reports. For actinic keratoses, where confirmation is often impossible because destructive treatment leaves no tissue to examine, the study’s findings argue for building explicit uncertainty into both clinical documentation and research instruments, perhaps by distinguishing between patient-recalled lesions and provider-diagnosed, pathology-confirmed ones.

What makes this research letter notable is not that it overturns any single dogma but that it quantifies a vulnerability running through a vast body of dermatological science and everyday clinical care. The authors, who declare no competing interests and report no external funding, designed the study so that statistical analysis was performed by dedicated biostatisticians within the University of Nebraska Medical Center’s Department of Biostatistics, with McGrath and Fagan contributing equally to the work. Their data cannot be shared openly, a restriction that protects participant privacy under the informed consent agreements patients signed. As skin cancer incidence continues to rise and as health systems increasingly depend on patient-reported outcomes and self-administered intake tools, including digital questionnaires and telehealth screening algorithms, the gap between what patients believe about their own skin and what has actually occurred on it becomes a variable that medicine can no longer afford to ignore. This study is a measured, careful reminder that in dermatology, as in much of medicine, the patient’s memory is a diagnostic instrument, and like any instrument, it needs calibration.

Subject of Research: Accuracy of patient-reported histories of actinic keratoses and keratinocyte carcinomas

Article Title: Assessing the accuracy of patient-reported actinic keratoses and keratinocyte carcinomas

Article References: McGrath, J., Fagan, E., Samson, K., Zhan, Y., & Sutton, A. (2026). Assessing the accuracy of patient-reported actinic keratoses and keratinocyte carcinomas. Archives of Dermatological Research, 318(1), Article 489. https://doi.org/10.1007/s00403-026-04942-8

Image Credits: AI Generated

DOI: 10.1007/s00403-026-04942-8

Keywords: actinic keratosis, keratinocyte carcinoma, skin cancer, patient self-report, dermatology, field cancerization, basal cell carcinoma, squamous cell carcinoma, health literacy, misclassification, skin cancer screening, epidemiology

The concept of a field effect dates back to the 1940s and early 1950s, when Slaughter and colleagues observed that tissue appearing benign next to oral cancers was often histologically abnormal, and that surgically resected patients frequently developed new tumors at the same site. They coined the terms field effect and field cancerization to describe a broad swath of predisposed tissue from which malignancies can arise. Since then, the phenomenon has been documented in gastrointestinal, lung, head and neck, breast, skin, and urological cancers, though research has shifted from microscopic evaluation toward molecular signatures. Contiguous epithelial tumors such as those of the head and neck, skin, gastrointestinal tract, and bladder tend to fit a monoclonal expansion model, whereas glandular tumors of the prostate and breast are often, though not always, linked to polyclonal expansion. More recently, computational modeling and spatial omics approaches have begun to map these altered fields with unprecedented resolution.

DNA methylation, a chemical modification of cytosine bases that can silence gene expression without changing the underlying sequence, has emerged as one of the most tractable molecular readouts of field cancerization. Prior work in RCC offered suggestive evidence: a comparison of renal tumors, matched normal renal tissues, and kidneys from healthy volunteers found that the average number of methylated CpG islands was highest in tumors, but matched normal tissue showed more methylation than tissue from healthy individuals. Notably, the degree of methylation in matched normal samples correlated with a higher histological grade of the corresponding tumors. If a methylation field effect exists in RCC, it could potentially flag patients at risk of local recurrence or metastasis, which occurs in up to 20 percent of patients curatively treated with partial or radical nephrectomy, and might even serve as a molecular check on surgical resection margins.

The Maastricht team, led by Kim Lommen and Kim M. Smits, focused on five candidate genes previously implicated in RCC diagnosis or field effects: ANGPTL6, ANKRD34B, CAIX, NHLH2, and ZIC1. ANGPTL6 encodes angiopoietin-like protein 6, linked to angiogenesis and tissue remodeling, while CAIX encodes carbonic anhydrase IX, a well-established RCC-associated protein regulated by the hypoxia sensor HIF-1α. NHLH2 has been associated with metastatic progression in RCC, and ANKRD34B and ZIC1 have been flagged in age-related methylation studies of normal kidney tissue. The researchers quantified promoter methylation using quantitative methylation-specific PCR (qMSP) with hydrolysis probes, normalizing results against ALU repeat elements and expressing methylation as the percentage of methylated reference (%PMR), a standard approach validated in prior MethyLight methodology studies.

The study drew on two patient cohorts. The first, a hospital-based series collected between 1995 and 2008 from archives in Leuven and Maastricht, comprised 86 RCC tumor samples, 28 matched adjacent normal (AN) tissue samples from the same patients, and 47 normal kidney (NK) samples obtained from non-cancerous kidneys, including autopsy material. The second was a prospective pathological series of 11 patients who underwent nephrectomy in 2019 and 2020, from whom five formalin-fixed, paraffin-embedded blocks were sampled per patient: the tumor core (C0), the tumor-to-normal transition zone (C1), and three additional samples spaced one centimeter apart moving away from the tumor (C2 through C4). An experienced pathologist blindly confirmed the histological status of every sample using hematoxylin and eosin staining.

The results were striking. Methylation was present in 26 to 59 percent of RCC samples depending on the gene, and in 0 to 54 percent of matched adjacent normal samples, but in only 2 percent of normal kidney samples across all five markers. For ANKRD34B, no adjacent normal sample showed methylation at all, while 26 percent of tumors did. Concordance between tumor and adjacent tissue varied by gene: every AN sample methylated for NHLH2 came from a patient whose tumor was also methylated, whereas for ZIC1 only 42.8 percent of methylated adjacent samples corresponded to methylated tumors. In the spatial series, mean %PMR across all genes declined gradually from 12.4 percent at the tumor core to 5.8 percent at the sample farthest from the tumor, and methylation outside the tumor core appeared across all stages, grades, tumor diameters, and histological subtypes, suggesting the effect was not driven by aggressive disease alone.

The most practically important finding emerged when the team recalculated diagnostic biomarker performance using different control tissues. When ROC curve analysis compared RCC cases against adjacent normal tissue, estimated sensitivities ranged from 11.6 percent for CAIX to 55.8 percent for ANGPTL6. When the same tumors were compared against normal kidney tissue from non-cancerous patients, sensitivities rose to between 25.6 and 59.3 percent. The largest gap appeared for ZIC1, a 30.2 percentage point difference. Specificities were similar in both analyses except for ANGPTL6, which jumped from 57.1 to 97.9 percent when normal kidney tissue replaced adjacent tissue as the control. The methylation present in adjacent normal tissue effectively shifts the cutoff for test positivity, masking the marker’s true discriminatory power.

The implications for the biomarker field are considerable. In a previous systematic review, the same group found that two-thirds of studies evaluating diagnostic DNA methylation biomarkers in kidney tissue used adjacent normal tissue as controls, and only one-third used tissue from non-cancerous patients. Given the new data, many published estimates of biomarker sensitivity may have been artificially depressed, not because the markers were weak but because the comparison group was molecularly contaminated. The authors advocate that diagnostic biomarker studies use normal tissue from non-cancerous patients as controls rather than tumor-adjacent tissue, a methodological correction that could reshape how candidate markers are triaged for further development.

The study’s authors are careful to frame the spatial findings as exploratory. The pathological series included only 11 patients, too few to analyze histological subtypes separately, and the RCC and normal kidney cohorts differed in age, a confounder given that age-dependent methylation of ANKRD34B and ZIC1 in normal kidney tissue has been reported. Although all extra-tumoral samples were histologically normal, the presence of scattered tumor cells cannot be entirely excluded, so methylation detected in the C2 through C4 samples cannot be attributed to a true epigenetic field effect with certainty. No follow-up data on recurrence or metastasis were yet available, leaving open whether the observed methylation patterns carry prognostic or predictive value.

Future work, the team argues, should scale up sample sizes, apply genome-wide methylation profiling to the same spatial sampling design, and integrate computational and spatial omics approaches to determine whether the observed patterns extend across the genome and reflect biologically functional changes. If validated, field methylation markers could eventually be monitored through liquid biopsies of blood or urine during post-nephrectomy surveillance, offering a non-invasive window onto residual molecular risk. For now, the study delivers a dual message: kidney tumors may leave an epigenetic fingerprint in the tissue around them, and the way researchers choose their control tissue can make the difference between a biomarker that looks mediocre and one that reveals its true potential.

Subject of Research: A DNA methylation field effect in tissue surrounding renal cell carcinoma and its impact on diagnostic biomarker evaluation

Article Title: Exploring a DNA methylation field effect in renal cell carcinoma and its implications for biomarker research

Article References: Exploring a DNA methylation field effect in renal cell carcinoma and its implications for biomarker research. (n.d.). https://doi.org/10.1186/s43682-026-00052-8

Image Credits: AI Generated

DOI: 10.1186/s43682-026-00052-8

Keywords: DNA methylation, renal cell carcinoma, field effect, field cancerization, epigenetics, biomarker, qMSP, kidney cancer, ANGPTL6, CAIX, ZIC1, control tissue