Welcome back to the journal review. This month we're working through the July 2026 issue of the Journal of the American Academy of Dermatology, and we've got four pieces on the docket — a feasibility study on peer mentorship around Mohs surgery, a large international reader study comparing dermatoscopy against clinical close-up photography, a clinical review on the cutaneous fallout of transplant immunosuppression, and a letter-and-response exchange on imaging in high-risk squamous cell carcinoma. Let's get into it.

First up is a feasibility study out of a single academic center, looking at one-to-one peer mentorship — essentially a buddy system — for patients undergoing Mohs micrographic surgery for facial skin cancer. The background here is something most of you already sense anecdotally: quality of life for these patients doesn't move in a straight line around the time of surgery, it fluctuates, dips and recovers unpredictably in the days and weeks after resection and reconstruction. Peer support programs have shown value in general oncology, but nobody had tested a structured buddy program specifically in the facial Mohs population, where the psychosocial stakes are arguably higher because of visible, midline, identity-relevant scarring.

So what did they actually do. Between late 2018 and mid-2019, they recruited two kinds of participants. Mentors were prior Mohs patients, at least a year out from surgery for nonmetastatic facial skin cancer, and current patients were then randomized into one of three arms — a presurgical mentee group getting peer support before their Mohs procedure plus standard counseling, a same-day mentee group getting peer contact on the day of surgery plus standard counseling, or a control group getting standard counseling alone. Mentors were paired to mentees based on tumor location and anticipated reconstruction type, so a patient facing an interpolation flap on the nose was matched with someone who'd been through something similar, and mentors underwent standardized training from a clinical psychologist. Contact was scripted — within a day of matching, at one week, then every two weeks out to the three-month follow-up. The reason for a three-arm randomized design with two different timing strategies is presumably to test not just whether peer support helps, but when it helps most, before the anxiety of surgery or on the day itself — a sensible design question even in a small pilot, though the paper frames the whole thing explicitly as a feasibility study rather than a powered efficacy trial.

The numbers are small, as you'd expect from a feasibility study — eleven mentors, thirteen presurgical mentees, eleven same-day mentees, and twenty-five controls. Quality of life was tracked with the Skin Cancer Index across three time points, before surgery, one to two weeks after, and three months after. Across every group, quality of life scores were actually highest at three months, which tells you these patients recover meaningfully as scars mature, regardless of study arm. The same-day mentee group did trend toward the highest scores at essentially every time point, particularly by three months, but none of these between-group differences reached statistical significance. Satisfaction with the program itself was solid, averaging about five out of seven on their feedback scale.

The honest limitation here, which the authors state plainly, is sample size — this cohort simply wasn't powered to detect a real between-group difference, and with groups this small you also have to worry about baseline imbalances, and indeed there was a significant sex imbalance across arms that complicates interpretation. So where does this leave you as a practicing Mohs surgeon. This is interesting, not practice-changing. It tells you a peer buddy program is feasible, logistically doable, and well received — that's a genuine contribution. But it does not yet tell you that it moves the needle on quality of life, and you shouldn't read the same-day-mentee trend as proof of anything. If your practice has the infrastructure — engaged prior patients, willing to be trained and matched — this is a low-risk, low-cost thing to pilot informally, but the evidence base for formally investing in it isn't there yet. A larger, adequately powered trial is really what's needed before anyone should call this standard of care.

Second article, and this one's a meatier original study — a large international reader study out of the MILK10k dataset, asking a deceptively simple question that turns out to be under-studied: on its own, without the other, how good is dermatoscopy compared to a good clinical close-up photograph, and how much does each add when you combine them. The gap they're addressing is that most prior literature only tested the complementary value of dermatoscopy — does adding it to a clinical image help — without ever isolating the standalone diagnostic performance of each modality by itself, and without ever testing this across a broad range of lesion types and a broad range of reader expertise rather than just melanocytic lesions read by expert academic dermoscopists. That matters enormously for teledermatology, where the images you get and the order you see them in can vary.

Methodologically, this was a crowdsourced online reader study run through the International Dermoscopy Society on a platform called Dermachallenge, over about a year. Two hundred eighty-three readers from forty-five countries participated — a real mix, dermatology residents, board-certified dermatologists, general practitioners, and others — and before anyone touched real study cases, each reader took a standardized thirty-case baseline test that stratified them into low, medium, or high expertise tertiles. That's a clever design choice worth flagging explicitly, because it lets the analysis separate out expertise as a variable rather than just assuming everyone with an MD reads dermoscopy the same way. For the actual study, readers worked through pairs of images — a clinical close-up and a dermatoscopic image of the same lesion, drawn from over fifteen hundred cases across melanoma, basal cell carcinoma, squamous cell carcinoma and keratoacanthoma, actinic keratosis, benign keratinocytic lesions, dermatofibroma, nevi, and vascular lesions — with ground truth mostly by histopathology and otherwise by expert consensus or follow-up. Crucially, the order of the two images was randomized, and readers gave an initial diagnosis after the first image, then were allowed to revise after seeing the second. That two-step response is the methodological trick that lets them tease apart standalone accuracy — using only first-image responses — from complementary value, meaning how much accuracy improved once the second image was added. That's a genuinely elegant way to answer two different clinical questions with one dataset.

The results are clinically meaningful and worth remembering. Dermatoscopy alone had noticeably higher sensitivity than a clinical close-up alone, roughly eighty-five percent versus seventy-four percent, a real and statistically significant gap. But it came at a real specificity cost — around sixty-seven percent versus seventy-two percent for the clinical close-up — so dermatoscopy alone is catching more cancers but also generating more false alarms. When you break it down by lesion type, dermatoscopy's advantage held up clearly for melanoma, basal cell carcinoma, actinic keratosis, dermatofibroma, and benign keratinocytic lesions. But for nevi, the clinical close-up alone actually outperformed dermatoscopy alone — a specificity trade-off playing out exactly where you'd expect, in the lesion category where dermatoscopy's pattern complexity can trigger overcall. For squamous cell carcinoma and keratoacanthoma, there was no real difference between the two.

The expertise breakdown is the part I'd flag as most practically interesting. Among low- and medium-expertise readers, there was no significant difference between reading dermatoscopy alone versus a clinical close-up alone — the dermatoscopy advantage essentially wasn't there yet. It only became statistically significant among high-expertise readers. In other words, the diagnostic edge of dermatoscopy as a standalone tool is itself expertise-dependent, at least in this teledermatology-style format. Separately, and importantly, when you looked at the complementary value — does adding the second image help, regardless of who's reading — the answer was yes for everyone: adding dermatoscopy to a clinical image raised the odds of a correct diagnosis by about half, and adding a clinical image to a dermatoscopic one raised the odds by about forty percent, both solid, statistically significant effects.

Limitations are important to keep in mind — this was an image-based simulation of teledermatology, not a live clinical encounter with palpation, patient history, or dynamic examination, and there's the usual specter of selection and verification bias in how cases were curated. So what do you do with this. The practical takeaway is that neither modality alone is sufficient, and this data gives you a quantified reason to insist on pairing high-quality clinical close-ups with dermatoscopic images in any teledermatology or triage workflow rather than relying on dermatoscopy in isolation — particularly because dermatoscopy-alone assessment of pigmented nevi by less experienced readers is exactly where you'll see unnecessary biopsy referrals generated. For those of you supervising trainees or building teledermatology triage pathways, this is close to practice-relevant guidance; for your own day-to-day in-person Mohs and dermatologic oncology practice, it's more a confirmation of what good practice already assumes.

Third, a clinical review on cutaneous adverse effects of immunosuppressive therapy in solid organ transplant recipients — a practical, agent-by-agent synthesis rather than a study, so no methods or results to walk through, just the state of the evidence organized by drug class. The organizing logic is induction versus maintenance therapy. On the induction side, antithymocyte globulin causes serum sickness in something like a quarter or so of renal transplant recipients, typically seven to fourteen days after first exposure, presenting as a symmetric urticarial or morbilliform eruption on the trunk and proximal extremities, often with fever, arthralgia, and lymphadenopathy — the skin finding is frequently the first clue, diagnosis is clinical, and management is discontinuation plus systemic corticosteroids, with plasma exchange reserved for refractory cases. Alemtuzumab, by contrast, causes cutaneous problems less often but can trigger autoimmune phenomena during immune reconstitution — vitiligo, alopecia areata, chronic urticaria — usually mild and manageable topically. Basiliximab is largely quiet dermatologically, though nonspecific pruritic rashes and herpes reactivation turn up in a meaningful minority.

On maintenance therapy, the calcineurin inhibitors are the ones you'll see constantly. Cyclosporine brings early hypertrichosis and gingival hyperplasia, later sebaceous hyperplasia, acneiform eruptions, and folliculitis, but the oncologic point that matters most to this audience is that chronic cyclosporine use raises cutaneous squamous cell carcinoma risk through PI3K-AKT pathway activation — which is the mechanistic rationale behind recommending dermatologic surveillance roughly every six to twelve months, tightened up for anyone with prior skin cancer or heavy actinic damage. Tacrolimus tends toward a diffuse nonscarring alopecia instead. Among the antimetabolites, mycophenolate mofetil causes a nonspecific maculopapular rash in roughly a quarter of patients and aphthous ulcers that respond well to dexamethasone rinses and supportive mouth care, while azathioprine is notable for a photosensitive dermatitis mimicking chronic actinic damage, plus a dose-independent hypersensitivity syndrome in the first month of therapy that mandates immediate discontinuation with no rechallenge — and azathioprine, like cyclosporine, meaningfully raises squamous cell carcinoma risk with long-term use, again prompting six-to-twelve-month surveillance and reinforced photoprotection. The mTOR inhibitors, sirolimus and everolimus, flip the oncologic story somewhat, offering a protective effect against squamous cell carcinoma, though they bring their own baggage — aphthous stomatitis in maybe ten to twenty percent of patients, acneiform eruptions in a much larger share, occasional erosive pustular dermatosis of the scalp, and wound-healing complications that can run remarkably high depending on the series.

The review's overarching practical message, and the one worth carrying into clinic, is that dermatologists should be using a structured risk-stratification approach — they specifically point to the skin and ultraviolet neoplasia transplant risk assessment calculator, or SUNTRAC — to time referral and surveillance intensity, and that close, ongoing coordination with the transplant team is essential any time you're considering dose reduction or a switch, say from a calcineurin inhibitor to an mTOR inhibitor, in a patient accumulating aggressive squamous cell carcinomas. There's nothing here that changes your Mohs technique, but it's a genuinely useful framework refresher for how you triage and time surveillance in this population, and for recognizing which acute eruptions warrant an urgent call to the transplant service versus routine topical management.

Last piece is a letter — an author response to a critique of an earlier retrospective cohort study on radiologic imaging in high-risk cutaneous squamous cell carcinoma. The original study had reported that imaging frequently detected occult disease and changed management in high-risk patients; Xie and colleagues wrote in with concerns, essentially arguing that the impressive imaging yield reflected selection bias — sicker, higher-risk tumors were the ones that got imaged in the first place — and that the survival comparisons were unadjusted and therefore fragile, and separately noting that ultrasound seemed underused as a nodal staging tool relative to CT and PET-CT. The authors' response here is a good example of appropriately humble but firm defense. On selection bias, they concede the point outright — yes, the imaged cohort had significantly worse baseline features, greater diameter, depth, poorer differentiation — but note their sample size didn't support propensity matching or multivariable adjustment, and reframe their actual primary aim as diagnostic yield and immediate management impact, not a survival analysis; they also clarify that only management changes made directly in response to imaging findings were counted, which shields that particular claim from confounding by indication. On survival, they acknowledge the caveat about lack of adjustment but point out that overall survival was similar between imaged and non-imaged groups despite the imaged group starting with worse tumors — which they interpret, appropriately cautiously, as consistent with imaging-driven upstaging and management changes offsetting some of that baseline risk, while explicitly calling for prospective multi-institutional work to actually isolate an independent survival benefit. On ultrasound, they simply agree, noting their data predates the broader shift toward nodal ultrasound in the United States, and they encourage further comparative work on modality-specific performance.

There's no new data here, so there's nothing practice-changing to extract beyond what the original study already offered — but the exchange itself is a useful reminder for how to read retrospective imaging-outcomes literature in high-risk cutaneous squamous cell carcinoma: treat the diagnostic yield and management-change findings as fairly solid, since they're less vulnerable to selection bias by design, and treat any survival comparison in a retrospective, non-adjusted cohort as hypothesis-generating only, no matter how reassuring the topline numbers look.

That wraps our four articles for July. To summarize the through-line — a small but well-intentioned pilot on peer mentorship that's feasible but not yet proven effective, a robust international reader study confirming that dermatoscopy and clinical close-up imaging are complementary rather than interchangeable, a practical transplant immunosuppression review worth keeping as a reference for surveillance timing, and a letter exchange that's really a lesson in interpreting retrospective imaging data honestly. Thanks for listening, and we'll see you next month.