Welcome back to the journal review. This is the May twenty twenty-six issue of the Journal of the American Academy of Dermatology, and we've got four pieces worth your time this episode — two of them speak directly to disparities in how we manage aggressive cutaneous tumors, one is a nice piece of visual-science brief report on basal cell carcinoma detection, and one introduces a promising new dermoscopic biomarker for melanoma. Let's get into it.

First up is a letter to the editor — a response piece — addressing sebaceous carcinoma. You'll recall the original brief report by Zhou and colleagues used the National Cancer Database to look at Mohs utilization trends in sebaceous carcinoma and found no survival benefit over wide local excision. This response, from Nolasco, Kabarriti, and Nadelmann, doesn't dispute that main finding — it extends it, specifically restricting the cohort to stage one through three patients undergoing definitive surgery with known margin status, deliberately excluding stage four disease, since metastatic sebaceous carcinoma is managed with systemic or palliative intent rather than curative surgery. That's a sensible methodological choice: it isolates the population where surgical modality is actually the clinically relevant question.

Within that refined cohort of about fifteen hundred patients, the authors reproduce the expected pattern — Mohs was used more often at academic centers and for smaller, earlier-stage tumors. But the new contribution here is twofold. First, they show a persistent racial disparity: Black patients had less than half the odds of receiving Mohs compared to non-Hispanic white patients, even after adjusting for tumor stage, size, insurance, and facility type. That's a meaningful, adjusted, statistically significant gap — not explained away by case mix. Second, they looked at outcomes nobody had examined yet in this database — margin status and receipt of adjuvant radiation — and found that Mohs was not associated with lower odds of positive margins, nor with less radiation. What did predict margin positivity and adjuvant therapy was tumor stage and size, not surgical modality. And consistent with the original report, there was still no survival advantage for Mohs over wide local excision.

The authors' takeaway is important for how you counsel patients and refer: in sebaceous carcinoma, tumor biology — stage and size — is driving outcomes far more than which surgical technique is used. That doesn't mean Mohs has no role; margin control still matters practically, especially periocularly. But for prognosis, standard excision appears to be a reasonably effective alternative, particularly where Mohs access is limited. The real actionable message is the disparity itself — the persistent gap in who gets referred for and receives Mohs likely reflects downstream effects of provider geography, referral patterns, and delayed presentation, and it's worth examining your own referral pathways. Do keep in mind the limitations inherent to any National Cancer Database analysis — no recurrence data, no disease-specific survival, no high-risk histology captured, and some very small subgroup cell counts on the margin analysis, so treat the margin and radiation findings as hypothesis-generating rather than definitive.

Next, a brief report tackling a deceptively simple question: how small does a basal cell carcinoma have to be before the naked eye can actually catch it? This is a retrospective cross-sectional study out of Memorial Sloan Kettering, comparing basal cell carcinomas excised by a single expert dermoscopist a decade apart — two hundred fifteen lesions in 2013, two hundred eighty-three in 2023 — looking specifically at lesion diameter on dermoscopy images. The single-observer design is deliberate: it controls for inter-rater variability in what counts as a suspicious lesion, though the authors are upfront that this limits generalizability to how a broader range of clinicians would perform.

The finding is fairly elegant. In both time periods, the number of tumors detected jumped sharply once lesion diameter crossed roughly four millimeters — a two-hundred-sixty-five percent jump in 2013 and a one-hundred percent jump in 2023 between the three-to-four millimeter and four-to-five millimeter bins. A Gaussian mixture model formally supported a two-component distribution — essentially a "hard to see" cluster and an "easy to see" cluster — with the small-lesion component's lower detection bound sitting around four and a half millimeters when pooled across both years. That threshold barely moved in ten years, despite meaningful advances in dermoscopy technology over that decade. The proportion of sub-four-millimeter basal cell carcinomas detected did roughly double, from about four percent to seven percent, suggesting some incremental gain from better imaging and more thorough exams — but the clustering above four millimeters persisted regardless.

There's also a nice observation about pigmentation: over half of the small lesions detected were pigmented, even though pigmented basal cell carcinoma is a minority subtype overall in fair-skinned populations. The authors reasonably interpret this as a contrast-detection effect — pigmented tumors simply stand out better against surrounding skin, consistent with basic visual science on contrast sensitivity for small targets.

What's the practical takeaway here? This isn't practice-changing in the sense of altering your surgical technique, but it's a genuinely useful cognitive calibration point. It reinforces that the rate-limiting step in early basal cell carcinoma detection is still unaided visual recognition, not diagnostic technology — which argues for continued emphasis on total body skin exam training and perhaps lower threshold dermoscopic screening of non-pigmented, subtle papules under four millimeters, since those are precisely the lesions your eye is most likely to miss.

Third, a brief report introducing a new dermoscopic sign worth knowing: ochre fluorescence under ultraviolet-induced fluorescence dermatoscopy — a newer modality that layers UV excitation onto standard dermoscopy. This international retrospective cross-sectional study pooled pathology-confirmed pigmented lesions suspicious for melanoma across four centers in Poland, France, Ukraine, and Chile, imaging all of them under the same polarized and UV-fluorescence conditions before correlating fluorescence findings with the final pathology.

Across roughly two hundred thirty lesions — half invasive melanoma, half in situ, plus a comparison group of dysplastic nevi and other borderline and benign lesions — ochre fluorescence showed up in about one in four melanomas but only rarely, in about three percent, of everything else. That difference was statistically significant and clinically substantial: the odds ratio for melanoma given ochre fluorescence was around ten. More importantly for risk-stratification, ochre fluorescence was far more common in invasive than in situ melanoma — about forty percent versus six percent — and correlated with greater Breslow thickness. In other words, when this sign is present, it's flagging the more concerning end of the spectrum. Interobserver agreement was excellent, essentially near-perfect, even between raters naive to dermoscopy, which speaks to how reproducible this sign is once you know what you're looking for.

The catch, and the authors are candid about it, is sensitivity — only about twenty-three percent of melanomas showed the sign, so its absence tells you nothing. But specificity was extremely high, around ninety-seven percent, with a positive predictive value around eighty-five percent in this dataset. The authors propose lipofuscin or melanolipofuscin as the likely fluorophore, drawing an analogy to orange pigment overlying choroidal melanomas, though they acknowledge this needs histochemical confirmation. Limitations are the usual retrospective, selection-bias caveats, a European-predominant population, and reliance on photographed rather than directly visualized fluorescence, plus unknown camera-to-camera variability.

Practically, this is interesting and promising rather than immediately practice-changing — ultraviolet-induced fluorescence dermatoscopy isn't standard equipment in most Mohs and dermatologic oncology practices yet. But if you have access to this modality or are considering it, ochre fluorescence is a rule-in sign worth incorporating into your equivocal-lesion workup: seeing it should raise your suspicion for invasive disease and lower your threshold for biopsy, while its absence should not reassure you at all.

Finally, another retrospective cohort study, this time on dermatofibrosarcoma protuberans, from a single academic center — Weill Cornell — looking at diagnosis, management, and outcomes over fifteen years in a racially diverse population of one hundred forty-four patients. The rationale for a single-center design here is straightforward, even though the authors don't spell it out explicitly: dermatofibrosarcoma protuberans is rare enough that you need a reasonably high-volume referral center to accumulate meaningful numbers, and a single site allows consistent chart-level detail on things like initial misdiagnosis and managing specialty that a multicenter or database study typically can't capture.

The headline findings are sobering. Only about a third of these patients were actually managed by a dermatologist — most went through general surgery or surgical oncology. Overall misdiagnosis before biopsy occurred in about one in six patients. Wide local excision outnumbered Mohs roughly two to one among those who had surgery. Cutaneous recurrence occurred in about one in ten, metastasis was rare. Then, restricting to the subset with documented race, Black patients had roughly four-fold higher odds of being misdiagnosed overall, with especially strong associations for misdiagnosis as a keloid — nearly eight-fold higher odds — or as a cyst. Black patients also had four-fold higher odds of being treated with wide local excision rather than Mohs, and — this is the finding that matters most clinically — nearly a five-fold higher odds of cutaneous recurrence, all statistically significant.

The authors' interpretation ties these threads together logically: if most dermatofibrosarcoma protuberans is being managed by non-dermatologists, and Black patients are disproportionately misdiagnosed as benign entities like keloids or cysts, that delays and possibly changes management toward less margin-precise excision — which would plausibly explain the elevated recurrence signal. They're careful to note this study doesn't show a difference in Mohs utilization specifically between Black and non-Black patients, only in wide local excision utilization — so the story is more about diagnostic delay and excision type than direct denial of Mohs access. They also contextualize against prior literature, noting one previous study found lower Mohs use specifically in Black patients while another found no racial difference in local recurrence, so this is not a fully settled picture across studies.

Limitations are the standard single-center retrospective ones — incomplete race documentation limiting the disparity analysis to a subset of patients, small numbers within that subset, and no ability to compare Black patients against individual other racial groups separately. So treat the specific effect sizes as preliminary.

The practical takeaway, though, has an immediately actionable piece: maintain a low threshold for biopsy on keloidal or cyst-like lesions, particularly on the trunk and extremities in Black patients, since this is precisely the pattern driving misdiagnosis in this cohort. And more broadly, this reinforces the case for early dermatology referral and dermatology-driven surgical planning in dermatofibrosarcoma protuberans, given the margin-control advantages Mohs offers for this infiltrative, subclinically extensive tumor.

That wraps up this issue. Across these four pieces, a common thread emerges — whether it's sebaceous carcinoma, basal cell carcinoma detection, melanoma imaging, or dermatofibrosarcoma protuberans, the gap between what our tools can detect and what actually gets diagnosed, referred, and treated appropriately remains the real frontier. Thanks for listening, and we'll see you next month.