Welcome to this month's journal review — we're covering the August 2026 issue of the Journal of the American Academy of Dermatology, and I've got four items for you today. Two are substantive letters engaging with recent original research, one is a retrospective cohort brief report, and one is a letter-format exchange on a wound care technique. Let's get into it.

First up is a letter to the editor titled "Beyond Surveillance, Epidemiology, and End Results Data: Clinical Trial Landscape and Biological Considerations in Merkel Cell Carcinoma." This is correspondence responding to an ongoing exchange — Fakult and colleagues had published a SEER-based epidemiologic analysis showing improved Merkel cell carcinoma survival from 2016 to 2021, attributing it to immune checkpoint inhibitors, and Sheu and colleagues had pushed back, noting that SEER data lack treatment-level granularity, so that attribution is speculative. This letter from Peng and colleagues offers a complementary angle: they systematically catalogued the global Merkel cell carcinoma clinical trial landscape — 113 unique trials across 12 countries from 1996 to 2025, pulled from Trialtrove and ClinicalTrials.gov.

What they found is a clear surge in trial initiation between 2016 and 2020 — right when that survival improvement window opens — with the United States hosting about two-thirds of all trials. Checkpoint inhibitors dominate, making up roughly six in ten immunotherapy trials, with pembrolizumab, avelumab, and nivolumab as the workhorses targeting the PD-1/PD-L1 axis. The authors frame this temporal coincidence — trial surge preceding survival improvement — as circumstantial support for the therapeutic-innovation hypothesis, while being appropriately careful to say correlation isn't causation here.

The more clinically interesting material, honestly, is what they say about resistance. About half of advanced Merkel cell carcinoma patients show primary or acquired resistance to checkpoint inhibitor monotherapy. And dual checkpoint blockade isn't a clean answer — in the Merkel cell carcinoma cohort of CheckMate 358, nivolumab plus ipilimumab showed essentially no improvement over nivolumab alone, with objective response rates around 58 versus 60%, but with more toxicity and shorter response duration. Yet a separate randomized trial by Kim and colleagues showed the same combination improving outcomes. The authors' read — and I think this is the key teaching point — is that biological heterogeneity, particularly Merkel cell polyomavirus status and tumor microenvironment, likely determines who benefits from intensified immunotherapy versus who just gets added toxicity. Trials still rarely stratify beyond basic polyomavirus positivity, which limits our ability to sort this out.

They also flag emerging resistance-directed strategies, all early phase and small in number — alternative checkpoint targets like LAG-3 and TIM-3, epigenetic modulators to restore MHC class I expression, radioimmunotherapy pairing pembrolizumab with peptide receptor or alpha-particle therapy, and virus-targeted adoptive T-cell therapies. Only about a dozen trials total, concentrated in the US, Germany, and Australia.

For your practice, this isn't practice-changing — it's a landscape piece, not a trial readout. But it's a useful literacy update: dual checkpoint blockade should not be reflexively assumed superior to monotherapy in Merkel cell carcinoma, the biology likely needs to guide combination selection, and the field is actively moving toward biomarker-stratified approaches that may eventually matter for how you counsel patients with treatment-refractory disease.

Next, a retrospective cohort study: "Integrated Multidisciplinary Care Reduces Time to Treatment in Patients with Advanced Cutaneous Squamous Cell Carcinoma." This comes out of the Center for Cutaneous Oncology at Brigham and Women's and Dana-Farber, and it addresses a genuinely practical question — does formalizing multidisciplinary care actually speed up treatment, and does it move the needle on survival, in advanced cutaneous squamous cell carcinoma specifically? Prior data support multidisciplinary care shortening time to treatment in other cancers, and a meta-analysis has linked multidisciplinary case conferences to improved survival broadly, but nobody had looked at this in cutaneous squamous cell carcinoma before.

The design is a natural experiment created by an institutional change: before April 2020, specialists coordinated informally, mostly by email; after April 2020, the clinic implemented a formal integrated model — unified triage, shared workspaces, weekly case conferences. The authors compared patients by which era they entered care in — nonintegrated from 2015 to early 2020, integrated from 2020 to 2024 — a before-and-after cohort design. This is really the only feasible design here: you can't randomize a clinic's organizational structure, and a formal structural change like this happens once, at one point in time, so a historical control is the natural comparator. The obvious weakness, which the authors themselves flag, is that "before and after" comparisons can't cleanly separate the effect of the intervention from everything else that changed over that same calendar time — treatment paradigms evolved, and this period also straddles the COVID-19 pandemic.

For the survival endpoint, they used a competing-risks framework — cumulative incidence with subdistribution hazard ratios — rather than a standard Kaplan-Meier or Cox model. That's the right statistical choice here, and it's worth spending a moment on why. In a population of older patients with advanced squamous cell carcinoma, plenty of patients die of other causes before they could ever die of their skin cancer — those deaths are competing risks. If you naively censor those patients in a standard Kaplan-Meier curve, you actually inflate the apparent probability of the cancer-specific event, because censoring assumes those patients remained equally at risk, which isn't true once they've died of something else. A competing-risks model instead tracks the actual proportion of the population experiencing disease-specific death over time, properly accounting for the fact that other-cause death removes patients from being able to have that event at all. It's the more honest analysis whenever your outcome is cause-specific mortality in a population with meaningful competing mortality.

Onto results: 127 patients total, about 30% treated under the old nonintegrated model and 70% under the new integrated one. Time to treatment initiation dropped from a median of 36 days to 17 days — that's clinically meaningful, cutting time-to-treatment roughly in half, and it held up after adjusting for age, sex, prior treatment, and disease stage, with an adjusted decrease of about 17 days. Patients under integrated care also needed fewer visits before starting treatment — a median of 2 versus 3 — and notably, over a quarter of integrated-care patients got started after just a single visit, something that never happened in the old model. Interestingly, the number of specialists seen was unchanged — so the efficiency gain is about compressing the process, not about seeing fewer people.

On survival, disease-specific death occurred numerically less often in the integrated group — about 6% versus 23% — but the 3-year cumulative incidence estimates were essentially the same, around 22 to 23% in both groups, and not statistically different. After adjusting for disease stage and neoadjuvant immunotherapy use — which, importantly, was only available to patients in the integrated era — there was no independent association between integrated care and disease-specific death. The authors are upfront that the study is underpowered for this endpoint, and that neoadjuvant immunotherapy availability itself is a confounder baked into the era comparison, not something integration itself caused.

Practical takeaway: the time-to-treatment finding is a genuinely useful, actionable data point if you're involved in building or advocating for multidisciplinary infrastructure for advanced cutaneous squamous cell carcinoma — cutting nearly three weeks off time to treatment initiation with fewer visits is a real operational win you could cite to administrators. The survival finding, however, is not practice-changing — it's underpowered and confounded, so don't take this as evidence that multidisciplinary care structure itself extends survival; that question remains open.

Third, a letter to the editor: "Response to Navarrete-Dechent et al's 'Ultraviolet-Induced Fluorescent Dermoscopy for the Diagnosis of Skin Tumors: A Multicenter Study.'" The original multicenter study looked at over 550 skin tumors under ultraviolet-induced fluorescent dermoscopy and reported that an "ochre color" fluorescence pattern was seen predominantly in invasive melanomas — present in roughly 44% of invasive melanomas versus under 9% of melanoma in situ, a significant difference, though ochre color didn't correlate with Breslow thickness among the invasive cases, and it wasn't seen in other pigmented lesions in their cohort.

This letter from Erol Mart and Akay pushes back on the specificity claim. They report five pediatric cases of atypical Spitz tumors — clinically and dermoscopically classic spitzoid nodules — that also displayed ochre fluorescence under the same 365 nanometer UV dermatoscope. Their point is straightforward: if a benign, albeit atypical, melanocytic proliferation can produce the identical fluorescence signature, then ochre color cannot be considered a melanoma-restricted marker. They walk through the proposed biochemical mechanisms — lipofuscin accumulation, melanin degradation byproducts, or the "porphyrin overdrive" phenomenon the original authors invoked — and note that none of these are biologically exclusive to melanoma; Spitz tumors have their own heightened metabolic activity and turnover that could plausibly generate the same fluorophores.

Their conclusion is a reinterpretation: ochre color is likely a nonspecific metabolic fluorescence signal across melanocytic proliferations generally, not a diagnostic hallmark of malignancy, and its value as a standalone diagnostic marker is probably limited. They call for larger comparative cohorts including Spitz nevi, congenital and dysplastic nevi, paired with biochemical work to actually identify the fluorophore.

Nothing here is practice-changing yet — ultraviolet-induced fluorescent dermoscopy isn't a validated standalone diagnostic tool for you to adopt regardless. But it's worth filing away: if this technology does make its way toward clinical use, don't anchor on ochre color as a melanoma-specific red flag, particularly in a pediatric or spitzoid lesion, where this letter suggests you'd expect a meaningful false-positive signal.

Last, a letter exchange: "Response to Zhang et al, 'Mixing Povidone-Iodine and Erythromycin Ointment in Skin Wound Management: A Reconsidered Practice.'" This is the original authors — Xu, Song, and colleagues — responding to a prior critical letter about their previously published technique for dressing large skin defects, the kind you see in pemphigus vulgaris or toxic epidermal necrolysis with extensive epidermal detachment. Zhang and colleagues had apparently raised concerns, presumably about chemical compatibility or safety, of combining povidone-iodine with erythromycin ointment and petroleum jelly gauze.

The authors clarify their rationale: the erythromycin ointment isn't chosen for antibacterial potency here at all — it's chosen for its oily vehicle, which combined with petroleum jelly gauze creates a moist, non-adherent barrier that avoids traumatizing fragile re-epithelializing skin during dressing changes. Povidone-iodine is layered on separately for its antimicrobial role. They note their mixing sequence — petroleum jelly gauze blended with erythromycin ointment first, povidone-iodine added last — is designed so the iodine sits predominantly on the outer surface, limiting direct chemical interaction, and they point out that Zhang's laboratory data used isolated "erythromycin A reagent" rather than the commercial ointment formulation, which may not translate to real-world clinical conditions. On the safety concern specifically about thermal or irritant reactions, they state that in their clinical observations, no patients reported burning or increased pain.

They close by describing a refined, staged protocol — starting with a low-concentration 0.5% povidone-iodine combination during the acute phase, with pre-treatment screening for iodine-intolerant patients such as those with hyperthyroidism, routine hepatic and renal monitoring, and a taper down to plain petroleum jelly-based dressings as the patient stabilizes, with the option to skip erythromycin altogether for those worried about the iodine interaction.

This is a technique-and-rationale exchange, not a study — there's no new data here beyond their unquantified clinical observations, so there's nothing to statistically evaluate. The practical takeaway is modest: if you or colleagues manage extensive epidermal detachment and have seen the original technique, this letter mainly reassures on the compatibility question and adds a reasonable staged, safety-screened protocol — including iodine contraindication screening and organ function monitoring — worth adopting if you already use a similar approach, but this remains low-level, single-group clinical experience rather than validated evidence.

That wraps our four articles for this August 2026 issue. To summarize the signal versus noise: the multidisciplinary care cohort study is the one with an actionable, quantifiable operational finding — a real reduction in time to treatment — while the Merkel cell trial landscape letter and the ultraviolet fluorescence rebuttal are both valuable for calibrating how you interpret emerging literature, without yet changing what you do at the bedside. Thanks for listening, and I'll see you next month.