Welcome back to the journal review. This is our June twenty twenty-six sweep of the Journal of the American Academy of Dermatology, and we've got four brief reports worth your time today — a retrospective cohort on proliferating pilar tumors, a SEER population-based study on melanoma transection, a systematic review of keratinocyte carcinomas arising on tyrosine kinase and Janus kinase inhibitors, and a randomized trial testing whether masking scents can cut down on that unmistakable electrosurgery smell in clinic. Let's get into it.

First up is a retrospective cohort study out of UT Southwestern looking at proliferating pilar tumors, or PPTs — those follicular neoplasms arising from the outer root sheath that everybody remembers seeing maybe once or twice in fellowship and then panicking slightly about, because the literature has always been murky on how aggressive these things really are. The background here is worth sitting with: a prior systematic review had classified something like a third of reported PPTs as malignant, with roughly one in thirteen showing metastatic spread — numbers that would justify real anxiety. But this group's own prior work suggests those alarming numbers may reflect misclassification, with aggressive squamous cell carcinomas being called PPTs when they weren't. So the gap they're addressing is basically: what does a real-world, carefully adjudicated cohort of PPTs actually look like, and how do they behave when you follow them?

Methodologically, this is a single-institution retrospective chart review spanning two full decades, capturing adults with biopsy-proven PPT. That twenty-year window makes sense here — this is a rare tumor, fewer than five hunder reported cases exist in the entire literature, so you simply cannot power anything prospective; you take what your institution has accumulated and describe it honestly. The authors don't frame it this way explicitly, but the long lookback also lets them capture meaningful follow-up duration, which matters enormously for a tumor whose main clinical question is local recurrence and metastatic potential over time.

They identified seventeen patients, mean age fifty-six, and this cohort was overwhelmingly female and overwhelmingly scalp-located — echoing prior epidemiology almost exactly. Mean lesion size was about two and a half to three centimeters, but with a wide range up to nine centimeters, so these can get large. About seven in ten patients went on to definitive treatment, and when they did, Mohs was the dominant modality, used roughly twice as often as standard excision, and the vast majority of procedures were done by dermatologic surgeons. When Mohs was used, most cases cleared in a single stage, which is reassuring from a margin-control standpoint. Histologically, every case was reviewed by board-certified dermatopathologists specifically to distinguish PPT from squamous cell carcinoma, and only two cases showed any coexisting well-differentiated SCC arising within the PPT — and even those were dermally confined without clear epidermal origin. Critically, over a mean dermatology follow-up approaching two and a half years, and all-specialty follow-up averaging nearly four years, there were zero recurrences and zero metastases in this entire cohort.

The authors' interpretation is that this more reassuring course likely reflects accurate diagnosis — in other words, once you rigorously exclude the SCC mimickers, true PPT may behave in a fairly banal fashion. They're appropriately honest that both excision and Mohs provided durable control here, and that current evidence still isn't sufficient to declare one modality definitively superior, even though Mohs is theoretically appealing for margin assessment in a tumor with reported aggressive variants. The limitations are the obvious ones for any rare-tumor study: retrospective, single-center, small numbers — inherent to studying something this uncommon, not a flaw of execution.

Practically, this isn't practice-changing in the sense of mandating a new algorithm, but it is reassuring and it is useful. If you get a clean dermatopathology read confirming PPT without SCC features, this data supports that either excision or Mohs can achieve durable local control, and it reinforces that meticulous histologic distinction from SCC — rather than the treatment modality itself — may be the thing that actually predicts a benign course. Worth keeping in your back pocket next time a scalp nodule referral says "rule out proliferating pilar tumor."

Next, a population-based study using the Surveillance, Epidemiology, and End Results program — SEER — addressing a question that comes up constantly in Mohs and surgical dermatology clinics: does melanoma transection during biopsy actually harm patients. We all know the theoretical concern — tangential biopsy is the dominant real-world technique despite guideline preference for excisional biopsy with narrow margins, and transection through the tumor means you can't reliably assess true Breslow depth, which drives staging and management. Prior data addressing outcomes had been single-center and reassuring, but genuinely underpowered population-level evidence was missing until now.

This is a retrospective, registry-based cohort study, and the design choice here is straightforward to explain: you cannot randomize biopsy technique in a way that induces transection, so a large administrative database is the only feasible way to get adequately powered survival data on a relatively uncommon scenario. SEER added a specific Breslow tumor thickness variable in twenty eighteen that actually flags transection status, which is what made this analysis possible — the authors used SEER-17 data from twenty eighteen through twenty twenty-two, restricting to histologically confirmed melanomas with known Breslow depth, and building multivariate Cox models adjusted for the usual prognostic suspects: age, sex, site, subtype, thickness, and ulceration.

Out of roughly a hundred five thousand patients, about three percent had transected melanomas. Who gets transected more often is intuitive once you hear it: older patients, head and neck primaries, lentigo maligna melanoma and nodular melanoma subtypes, ulcerated tumors, and anything deeper than the thinnest T1a category. The authors reasonably speculate that head and neck and lentigo maligna transection rates may reflect clinicians intentionally limiting biopsy depth in cosmetically sensitive areas. But here's the number that actually matters for your practice: in the fully adjusted model, transection was not associated with worse melanoma-specific survival — the hazard ratio was right around one, essentially flat, and not statistically significant. That confirms, now at a population level, what those single-center studies had already suggested.

There's a second finding worth flagging for shared decision-making around sentinel lymph node biopsy. Among the thinnest tumors, pT1a, transected melanomas underwent SLNB more than twice as often as non-transected ones — about one in ten versus one in twenty-five — presumably because clinicians worry the true depth might actually be pT1b or deeper once you account for the missing tissue. But despite that increased testing, the actual SLNB positivity rate was essentially identical between transected and non-transected pT1 tumors, right around six percent either way, and that difference was not significant. So the biopsy behavior — ordering more sentinel node biopsies out of transection anxiety — isn't being validated by a correspondingly higher yield of positive nodes.

Limitations are appropriately disclosed: no recurrence data, and no information on true Breslow depth after re-excision, so residual depth misclassification is possible even in this large dataset. Practically, I'd call the survival finding close to practice-affirming rather than practice-changing, since it essentially validates what most of us already believed and practiced around — transection itself is not a survival-altering event once you account for standard prognostic factors. The more actionable piece is the SLNB signal: this data offers a legitimate talking point when counseling patients with transected thin melanomas that transection status alone, absent other high-risk features, may not need to independently push you toward sentinel node biopsy.

Third, a systematic review looking at keratinocyte carcinomas reported in patients on tyrosine kinase inhibitors and Janus kinase inhibitors — TKIs and JAKis. Background here is that both drug classes are increasingly common across oncology and inflammatory disease, and there's accumulating case-level suspicion linking them to basal cell carcinoma, squamous cell carcinoma, and keratoacanthomas, but nobody had pulled the literature together systematically.

This followed standard PRISMA methodology, searching MEDLINE and Embase, and grading study quality with the Oxford levels-of-evidence framework — sensible for a review, since the evidence base here is inherently heterogeneous, mostly case reports and case series rather than controlled trials, and grading quality helps the reader calibrate how much weight to put on any single signal. They pulled in forty-seven studies totaling just over a thousand patients — a mix of case reports, clinical trials, retrospective studies, case series, and one prospective study.

Breaking down by agent: ruxolitinib accounted for the largest single group, roughly a third of all patients, with the majority of lesions being basal cell carcinomas, but with a meaningful chunk of squamous cell carcinomas too, and — importantly — this was the only agent associated with metastatic SCC, with three reported deaths. Tofacitinib and upadacitinib each contributed sizable numbers of keratinocyte carcinomas as well, split fairly evenly between basal cell and squamous cell carcinoma. On the TKI side, sorafenib stood out for a different pattern — its lesions skewed heavily toward keratoacanthoma-like squamous cell carcinomas, developing notably fast, with a mean time to onset under four months, versus a more spread-out average of about five months across the whole cohort. Overall lesion distribution across the entire dataset was roughly forty percent basal cell carcinoma and just under thirty percent squamous cell carcinoma, with patients averaging around four lesions each.

The authors propose distinct mechanisms for the two drug classes rather than lumping them together, which is a nice bit of biological reasoning worth relaying: TKIs may drive keratinocyte proliferation through paradoxical MAPK-ERK pathway activation via RAF kinase inhibition, which is the same mechanism long described with BRAF inhibitors and their keratoacanthoma-like eruptions — so there's a plausible class effect, though the phenotype seems to vary depending on kinase selectivity. JAK inhibitors, on the other hand, are hypothesized to impair cutaneous immune surveillance by blocking JAK-STAT signaling that normally supports interferon and T-cell antitumor responses, which would explain why ruxolitinib in particular shows up with the more aggressive, occasionally metastatic phenotype.

The honest limitations here are substantial and the authors don't shy away from them: wildly heterogeneous study designs, incomplete reporting of onset timing and follow-up in a large fraction of cases, and — critically — no denominator data, meaning they cannot calculate incidence or relative risk, only describe what's been published. So this is hazard identification, not risk quantification.

For practice, I'd frame this as clinically important but not yet actionable in a protocol sense — there's no threshold here that changes your surveillance intervals. What it should change is your index of suspicion: any patient referred to you on ruxolitinib, tofacitinib, upadacitinib, or a TKI like sorafenib deserves a more deliberate skin exam and lower threshold for biopsy of new or changing lesions, and if you're the one managing a rapidly growing keratoacanthoma-like nodule in a sorafenib patient, know that the literature supports treating it with real urgency given how fast these can arise and, at least with ruxolitinib, the documented potential for aggressive behavior.

Last one today is lighter but genuinely useful for your day-to-day clinic experience — a double-blinded randomized controlled trial testing whether a masking scent reduces patients' perception of electrosurgery odor during outpatient skin surgery. The premise is simple: even with a smoke evacuator running per NIOSH standards, that cautery smell is still very much present, and it's a recurring, if under-studied, source of patient discomfort.

Design-wise, this was a three-arm trial — intervention, placebo, and untreated control — randomized two-to-one-to-one, so twice as many patients got the active masking scent as either comparator, which is a sensible allocation when your primary interest is characterizing the intervention's effect rather than spending equal power on two control comparisons. Patients in the intervention and placebo arms chose their own scent from a menu — eucalyptus, lavender, lemongrass, peppermint, sweet orange, or tea tree — and held either a scented or an identical-looking unscented device near the nostrils during the procedure; control patients got nothing. That patient-selected-scent design is a nice methodological touch, since it controls for individual scent preference, which could otherwise be a huge confounder in an odor-perception study.

Just over a hundred ten patients completed it, mean age around sixty-seven, majority male, and the bulk of procedures were on the head and neck, which is exactly where cautery odor is most noticeable to the patient. The headline result: perceived electrosurgery odor was significantly lower with the masking scent compared to both control and placebo, and — importantly — placebo and control were statistically indistinguishable from each other, meaning the effect wasn't just about holding a device near your face or the ritual of participating in a study; the actual fragrance did the work. Subgroup analyses suggested the effect was more pronounced in women, in patients seventy and younger, in those undergoing a single procedure rather than multiple, and in trunk or extremity cases rather than head and neck. Overall patient-reported experience also improved significantly with the active scent compared to placebo. But — and this is the honest, clinically important nuance — pain and anxiety scores did not differ across any of the arms. No serious adverse events; two patients reported transient lightheadedness they attributed to first-time-surgery nerves rather than the scent itself, and both chose to continue.

The authors' interpretation is that this is a genuine odor-masking effect, possibly compounded by a psychological sense of control or distraction from a self-selected sensory intervention, and they draw a nice parallel to prior work on guided imagery and music during Mohs surgery, which similarly improved experience without moving the needle on measured pain or anxiety — suggesting these adjunctive comfort measures work through a different channel than analgesia or anxiolysis per se.

This one I'd call low-effort, practically actionable, and reasonably practice-changing for something this trivial to implement. It's a low-cost, safe, essentially zero-downside intervention — a scented device is a much easier lift than, say, arranging music or guided imagery protocols — and the data support a real, patient-perceptible benefit, particularly for your female patients, younger patients, and trunk or extremity cases. It won't reduce your patients' pain or anxiety, so don't oversell it as such, but if patient experience and satisfaction scores matter to your practice — and they increasingly do — this is about as easy a win as you'll find in the surgical literature this year.

That wraps our four articles for June. To recap the through-line: reassuring, mostly non-practice-changing confirmatory data on both proliferating pilar tumor behavior and melanoma transection outcomes, a signal-generating but still hypothesis-level review on keratinocyte carcinoma risk with kinase inhibitors that should raise your suspicion without yet changing protocols, and a genuinely simple, low-cost intervention in the masking-scent trial that you could reasonably start using in clinic tomorrow. Thanks for listening, and we'll see you next month.