Six melanoma patients, six weeks, one common variable: the question oncology is choosing not to study

The six were supposed to be the success stories.

Angus Dalgleish had been treating stage-4 melanoma patients for the better part of forty years, and the six in question had between them somewhere between three and eighteen years of stable disease behind them. The kind of patients an oncologist mentions when someone asks whether the immunotherapy work was worth it. Late 2021 into early 2022, all six relapsed within about six weeks of one another. The one variable they shared was that each had recently received a COVID booster.

Dalgleish says he raised the pattern with the British medicines regulator as early as 2021. The MHRA did not act on it. Four years later he sat in front of Senator Ron Johnson’s Permanent Subcommittee on Investigations and described it under oath, and a few days after that he laid it out at length in an interview with Dr. John Campbell, the British nurse-educator whose YouTube channel has become a clearinghouse for the kind of clinical observation the medical-society apparatus would rather not field.

Dalgleish is not a wellness influencer or a podcast doctor. He co-discovered the CD4 receptor as the entry point for HIV in the mid-1980s, a finding that underwrites the architecture of modern HIV treatment, and he spent the rest of his career on tumor immunology, helping move cancer immunotherapy from theory into clinical practice in the United Kingdom long before checkpoint inhibitors had brand names. When a man with that résumé says he is watching something in his own clinic the field told him should not happen, you can dismiss him, but you cannot do so without a counter-explanation.

The counter-explanation the American Society of Clinical Oncology sent to the subcommittee was that “RNA breaks down quickly in the body and doesn’t enter a person’s DNA.” It came from ASCO’s chief medical officer, Dr. Julie Gralow. It is a comforting sentence and a true one about ordinary mRNA. It is also an answer to a different question. Some witnesses at the hearing did press the nuclear-integration concern, and Gralow addressed that. The argument Dalgleish himself has spent the last two years pressing is narrower and harder to wave away: that repeated dosing does two things to the immune system that the regulators never had to think about when they cleared a single-shot product, and that those two things together would, mechanistically, do exactly what he is now watching happen at the bedside.

The first is profound T-cell exhaustion, the same kind of immune suppression oncologists spend careers trying to reverse, now being induced iatrogenically at the population level. The second is a class-switch in the antibody response, away from the neutralizing IgG1 the boosters were marketed on and toward an antibody subclass called IgG4 that does not kill what it binds. The shift is dose-dependent. Irrgang and colleagues, writing in Science Immunology in late 2022, measured the IgG4 fraction of the spike-specific antibody response at 0.04 percent two weeks after a second mRNA dose and 19.27 percent late after a third. A 2023 review in the journal Vaccines by Uversky and colleagues pulled that finding into a broader case that this kind of repeated-antigen IgG4 switch is associated with immune tolerance rather than immune clearance. A subsequent paper looking specifically at solid-cancer patients found the booster-driven IgG4 shift was present in them too, with the wrinkle that further fourth and fifth doses did not push it higher than they pushed it in healthy controls.

Put the two findings together and you have a plausible chain a working oncologist can actually run in his head. Repeated dose. Subclass switch. The tumor-killing IgG1 antibodies the immune system was using to keep a residual melanoma quiet get crowded by IgG4 antibodies that bind the same target and do nothing. The T-cells that would have stepped in are themselves exhausted. The brakes come off. A patient who has been stable for a decade relapses. Then five more do, within six weeks.

It is also not only Dalgleish’s clinic. Wafik El-Deiry of Brown’s Legorreta Cancer Center, testifying at the same hearing, told the panel he had identified nearly seventy published reports across twenty-seven countries describing more than three hundred cancer cases following COVID vaccination. The All-Party Parliamentary Group on Pandemic Response and Recovery, summarizing analyses from Japan, South Korea, and Italy, reported a dose-response pattern: cancer mortality climbing in Japan after the third, fourth, and fifth doses; cancer incidence rising in the other two. A 2025 review in the MDPI oncology journal Cancers set out the plausible mechanistic pathways at length: immune-surveillance failure, IgG4 tolerance, persistent spike-protein production, trace DNA impurities left from the plasmid-based manufacturing process whose biological significance, the authors note, is not yet known. None of those mechanisms requires the mRNA to “enter a person’s DNA” the way ASCO’s prepared line implies.

That review is the one to watch, because it sits in a mainstream oncology journal and because the response to it from the field’s leadership has been, broadly, to not engage. ASCO’s posture, repeated by Gralow at the hearing, is that there is “no clinical evidence” linking the boosters to cancer outcomes. The work being done by the word “clinical” in that sentence is what the audience is meant to miss. A peer-reviewed mechanism paper, a hundred-plus case reports across two dozen countries, international epidemiology pointing in the same direction, and a forty-year oncologist’s bedside pattern do not, by ASCO’s reading, count as clinical evidence. What would count is a randomized controlled trial of boosters versus placebo in cancer patients. The argument of this piece is that such a trial will never be run, because the evidentiary standard being held out has been chosen for its impossibility, not its rigor.

This is not proof that mRNA boosters reactivate dormant cancers. It is not a stray anecdote either, and that is the move the official answer keeps trying to make. Dalgleish has a mechanism that is published, a class-switch finding that is replicated, a case series that is corroborated by independent series on three continents, and a clinical pattern that the country’s flagship oncology society chose to meet by answering a question he did not ask. The honest verdict on the evidence is that the question is live and the institutions whose job it is to study it are choosing not to.

The six patients are still in his case notes.

Six melanoma patients, six weeks, one common variable: the question oncology is choosing not to study

Sources

Lena Marshall

Six melanoma patients, six weeks, one common variable: the question oncology is choosing not to study

Sources

Lena Marshall

Stories like this, every Sunday.

More to read

A Korean lab found cancer's DNA-repair off-switch. In a mouse.

ALTAIR Missed Its Primary Endpoint. The Press Tour Found Something Else.

Ozempic's 30 Percent Breast-Cancer Headline, and the Trial Evidence It Walked Past