Welcome to this month's journal review — we're covering JAMA Dermatology's December 2025 issue, and given the density of the piece, we're spending our full time on a single Viewpoint that's likely to generate some real discussion in tumor boards and journal clubs: "Navigating Melanoma In Situ in the Age of Overdiagnosis," from the NYU Langone group — Shaked, Swearingen, Stein, and Polsky.

Let's be clear about what this is: it's an opinion piece, a Viewpoint, not an original study. So there's no methods section to critique, no cohort, no p-values. What we're getting is an argued position, grounded in cited literature, with a set of explicit calls to action. I'll walk through the argument the way the authors built it, and flag exactly where they're citing someone else's data versus where they're just asserting an opinion.

The setup is the now-familiar overdiagnosis argument, but sharpened specifically for melanoma in situ. The authors' framing of overdiagnosis is the standard one from the cancer screening literature — detecting something that's histologically real but would never have gone on to cause symptoms or shorten life. Their argument is that MIS is the prime example of this in our field. Unlike invasive melanoma with real metastatic potential, MIS carries essentially no mortality risk. And they point to the epidemiologic signal we all already know: MIS diagnoses have climbed substantially over recent decades without any corresponding drop in melanoma mortality or in invasive melanoma incidence. If MIS detection were meaningfully intercepting future lethal disease, you'd expect to see that curve bend. It hasn't. That's their core evidentiary anchor, and it's not new to this paper, but it's the premise everything else hangs on.

They're careful, though, to note this isn't just an academic numbers problem — they cite a study of melanoma survivors, including MIS patients, showing that fear of cancer recurrence significantly undermines psychological well-being, even when the original diagnosis carried no real mortality threat. So the harm isn't hypothetical — it's anxiety, it's the recurrence-fear burden, and downstream, it's whatever morbidity comes from treating a lesion that surgery might not have been necessary for.

Then they walk through why MIS numbers have gotten so high — screening pressure driving more total body exams and more biopsies of clinically atypical but benign pigmented lesions, possibly overly sensitive histopathologic thresholds for calling something melanoma, and better imaging — total body photography and dermoscopy — feeding more biopsies into the pipeline. None of this is presented as new data; it's synthesis of prior arguments, credited accordingly.

From there the piece pivots to solutions, structured as three explicit calls to action, and this is really the substance of the piece.

The first call to action is about reframing MIS conceptually and in patient conversations. They point to the updated MPATH-Dx version 2.0 framework — that's the Melanocytic Pathology Assessment Tool and Hierarchy for Diagnosis — which now uses a four-tier system, splitting melanoma into two classes based on depth, essentially separating pT1a from anything pT1b or deeper, and splitting benign lesions into two classes as well, with severely atypical nevi and MIS both landing in the "high-grade benign" category rather than being called cancer outright. The authors' argument is that this effectively reclassifies MIS as a non-obligate precursor with low progression risk — a precancer, not a bona fide malignancy — which is actually consistent with how MIS was viewed prior to 1985. Their ask here is explicitly for us to adopt this framing when we talk to patients, and — importantly — they acknowledge this is currently just a reasonable hypothesis. They call for actual clinical studies to measure whether reframing the diagnosis this way genuinely reduces psychological harm without introducing some other harm, like undertreatment or loss of follow-up compliance. So this is a proposed communication strategy awaiting validation, not an established practice.

The second call to action targets the front end of the pipeline — reducing biopsies of atypical but benign pigmented lesions in the first place. Here they do cite actual outcomes data, the Menzies et al. Australian primary care study, and this is worth stating precisely because the number is the point. Sixty-three primary care physicians trained in dermoscopy used three-month sequential digital dermoscopic imaging — watching flat, atypical pigmented lesions rather than biopsying immediately — across roughly 370 lesions. That strategy cut the number of biopsies performed by about half, and importantly, they still caught the cancers that mattered: forty-two skin cancers total, thirty-four of them melanomas. Only one lesion that went unbiopsied under the watch-and-wait protocol later turned out to be MIS. So the message the authors draw is that short-interval dermoscopic monitoring can meaningfully cut biopsy volume without an obvious safety cost, particularly because MIS by nature behaves indolently enough to tolerate a short observation window. They're upfront, though, that operationalizing this — training, cost, follow-up compliance, and reimbursement models that currently reward biopsying over watching — are real barriers, and they call for larger studies addressing exactly those implementation obstacles, plus AI-assisted dermoscopy tools that could formally incorporate watch-and-wait as a validated management arm rather than an off-protocol judgment call.

The third call to action is about treatment itself, once MIS is diagnosed — and this is likely the part most directly relevant to our daily practice. Reframing MIS as a precancer, they argue, should widen the acceptable management menu beyond surgery, especially for lentigo maligna on the face and neck. They're not arguing against surgery as the default — they explicitly state surgery remains the recommended treatment for MIS on trunk and extremities. But for cosmetically or surgically difficult lentigo maligna — thinking of your patients with significant ADL limitations, limited life expectancy, large lesion size, or surgical risk factors — they argue topical imiquimod, off-label but supported by multiple existing studies, radiation therapy, or continued SDDI surveillance are reasonable alternatives to discuss, even acknowledging these options come with somewhat lower published clearance rates than surgery. Their honest caveat here is important: they explicitly say there isn't sufficient high-quality comparative data to confidently counsel patients on these tradeoffs, and they call for a proper trial head-to-head against surgery, using clearance, recurrence, and patient-centered endpoints like cosmetic outcome and satisfaction.

So — what's the practical takeaway here for a Mohs surgeon reading this? I'd separate it cleanly into what changes tomorrow versus what's aspirational.

Nothing here is practice-changing in the sense of altering your surgical indications — surgery remains standard of care for MIS on trunk and extremities in this piece, full stop, and even for lentigo maligna, the alternatives are framed as options for a specific subset of surgically or cosmetically challenging cases, not a general shift away from excision. What is immediately actionable is the communication piece — there's a real, low-cost opportunity to discuss MIS with patients using precancer language rather than defaulting to "you have skin cancer," particularly leaning on the updated MPATH-Dx V2.0 framework, since the authors argue this may meaningfully reduce needless anxiety, even though it hasn't yet been formally tested.

Everything else is interesting but not yet actionable: the SDDI watch-and-wait approach is compelling data from a single Australian primary care trial, but it wasn't studied in a dermatology specialty setting, and the barriers to deploying it — training, reimbursement, follow-up compliance — are real and unresolved. And the nonsurgical treatment alternatives for lentigo maligna remain conversations to have selectively with appropriate patients, grounded in the authors' own acknowledgment that comparative high-quality trial data doesn't yet exist. This is a call for that evidence, not a substitute for it.

That wraps our December 2025 review. Thanks for listening, and we'll be back next month with more from the literature.