The Most-Cited Case in This Field Has No Medical Record

Every article about melanoma under the nail arrives at the same place. A lesion on a right great toe in July 1977, an amputation refused, and a death from metastatic disease in May 1981 at the age of 36.

It is the reason a great many people know this cancer exists, and I do not think that is a bad thing. But I went looking for the medical record behind it, and there isn’t one.

What is actually in the literature

There is no case report, no autopsy record, no clinician-authored contemporaneous account of Bob Marley’s melanoma in the peer-reviewed literature. Every mention I could locate is a historical or educational vignette written decades afterwards by authors with no stated access to any records.

The fullest of them, a 2022 essay in Clinics in Dermatology, gives the diagnosis month and year, the site, the refusal, the age at death and the date. It carries no Breslow thickness, no stage, no histology, no treating institution, no named physician, and no primary citation. The rest are a French pair with no indexed abstracts, an Indian piece with no abstract, and an author-reflections page.

That essay is honest about what it is. Its own closing observation is that Marley has served as the posthumous poster child for skin cancer in people with skin of color for decades and has raised public awareness of this rare form of melanoma.

What has happened since is that biography has been laundered into the medical literature by repetition through review articles, and now functions as evidence. Specifically, it functions as evidence for a proposition — amputate early or die — that the surgical literature does not support.

None of this is a criticism of a man who was ill and made his own decisions about his own foot. It is a criticism of a field that has been teaching from a story for forty years without ever writing down the data.

What the surgery evidence says

The question the story implies is whether taking the digit saves the patient. Several retrospective series have looked, and they keep landing in the same place.

A Japanese series of 151 nail apparatus melanomas across five institutions found no local recurrence after any of the surgical procedures used, and survival analysis showed that the type of procedure was not significantly associated with disease-free survival or with overall survival. The authors’ framing is that the operation does not influence survival as long as total resection of the primary tumor is accomplished. Their five-year overall survival by thickness runs 100 percent for in situ, 94.4 for a millimeter or less, 91.7 up to two millimeters, 72.7 up to four, and 47.6 above four.

Thickness is doing the work in that table. Not the amputation.

An Italian institution reviewing twenty years and 39 patients found that function-preserving surgery compared with disarticulation did not correlate with a better prognosis — and the p-value, 0.08, tells you the comparison was underpowered rather than settled.

A Colombian cohort of 45 nail unit melanomas treated between 2007 and 2023 had 13 amputations and 32 function-preserving operations. Over a median follow-up of 52 months there was no regional recurrence, no metastasis and no death. The single local recurrence occurred in a patient who had been treated with amputation.

And a US single-institution series of 123 patients, in which every subungual case had a distal joint or proximal amputation, found that proximal amputations were associated with significantly worse overall and recurrence-free survival than distal ones. That is almost certainly selection — bigger tumors get bigger operations — but it is the opposite of a dose-response argument for cutting more.

These are retrospective cohorts. None of them randomized anybody, none can fully separate the operation from the tumor that prompted it, and a 2026 systematic review of function-preserving surgery for invasive lesions found no study defining clear indications for it. So the honest statement is not that amputation is unnecessary. It is that the evidence base contains no demonstration that the more destructive operation buys survival, and several series in which it did not.

The site that turns out not to be the problem

Subungual melanoma is generally described as carrying a worse outlook than melanoma on the sole or the palm.

The largest comparison tested it directly: 627 primary acral lentiginous melanomas, 45 of them subungual. Overall survival at one, five and ten years was 81, 40 and 28 percent for the subungual tumors against 94, 59 and 38 for the rest, with a significantly higher risk of death.

But disease-specific survival — deaths from the melanoma rather than from anything else — was 94, 56 and 48 percent against 96, 69 and 55, and that difference did not reach significance.

The authors’ conclusion is stated flatly: subungual anatomic site is not an independent prognostic factor for acral lentiginous melanoma.

What is prognostic is how thick the tumor is when it is found, and subungual tumors are found thick. A Brazilian cohort of 157 subungual melanomas had a median Breslow depth of 6.0 millimeters and ulceration in six of ten. A Polish series put median thickness at 4.0 millimeters for subungual against 2.2 for melanoma overall. A 124-case pathology series found 79 percent locally advanced at presentation and a median of 3.2 millimeters.

Six millimeters is a very late tumor. The problem is not where it grows. The problem is when it is caught, which is what the piece on nail bands is about.

The two registry analyses that disagree

Here is a discrepancy that anybody writing about this should have to confront.

The 2009 analysis reported melanoma-specific survival highest in non-Hispanic white patients and intermediate in Black patients — with the lowest figures in Hispanic white and Asian and Pacific Islander patients.

Read the ten-year column again. Black patients, 71.5 percent. Non-Hispanic white patients, 69.4.

The 2020 analysis, covering the next decade of the same registry, put five-year survival lowest in Black patients — with the difference surviving adjustment for stage at both stage I and stage III.

Two analyses of the same national registry, consecutive periods, opposite orderings. I cannot reconcile them from the published abstracts, and neither can anybody quoting one of them as settled. What can be said is that a 2024 analysis of 4,188 cases across 2000 to 2020 found significantly elevated melanoma-specific mortality for Hispanic, Asian and Pacific Islander, and Black patients, with the Black adjusted hazard ratio at 2.19.

The direction of the modern evidence is clear enough. The specific survival percentage anybody quotes is not.

And one more figure from that 2024 analysis, because it goes against how this disease is described. Of the 4,188 US cases, 2,786 — two-thirds — were in non-Hispanic white patients. Incidence rose over the period only among Hispanic and non-Hispanic white patients, and stayed statistically stable among Black and among Asian and Pacific Islander patients.

The proportional argument in the piece on where melanoma appears still stands, and it is the argument that matters for what gets examined at an appointment. But most people who get acral melanoma in the United States are white, and a paper that describes it as constituting 50 to 75 percent of melanoma in populations of color — as one 2020 introduction does — is at odds with the measured figures of 36 and 32.6 percent from the registry itself.

The nail plate is not a perfect shield

The standard line is that acral melanoma has nothing to do with sun, and at the level of the whole subtype that is right. Whole-genome work found acral and mucosal melanomas dominated by structural changes and mutation signatures of unknown etiology rather than by ultraviolet damage.

The nuance almost always gets dropped. A whole-genome study of 87 acral tumors reported a subset — mostly subungual — carrying an ultraviolet radiation signature. An earlier analysis put a number on it: three of 35 acral melanomas, 8.6 percent, had a predominant UV mutation burden, all verified as arising at acral or subungual sites, and no histopathological feature predicted which ones.

The authors’ conclusion is that the nail plate and acral skin do not provide complete protection from ultraviolet radiation.

That does not make acral melanoma a sun disease, and it does not mean sunscreen on your hands prevents it. It means the confident version of “this one has nothing to do with UV” is confident about a subset it has not checked.

Thickness is the number. The digit is the argument.

If a nail lesion is being discussed and the word amputation is in the air, the published questions are: how thick is it, is it in situ or invasive, can it be completely excised without taking the joint, and what does the surgeon’s own recurrence experience look like.

Complete excision is the thing every one of these series has in common. How much of the finger goes with it is the part where the evidence stops being clear, and where the story of a refused amputation in 1977 has been doing the arguing instead.

Anybody who wants to invoke that story should have to say, in the same breath, that no medical record of it has ever been published — and that in the series we do have, the operation was not what separated the survivors from the rest. The thickness at diagnosis was. Which puts the whole weight of this disease back where it belongs: on the months between the streak appearing and somebody taking it seriously.

This article is educational and is not a substitute for care from someone who can look at your skin. Anything new, changing, painful or spreading should be assessed rather than researched.