July 6, 2026 · Dr. Ray Okafor, Cognitive Neuroscience~6 min read
A woman in my surgery last week slid a newspaper clipping across the desk — "Is Alzheimer's a kind of cancer now?" — and her hand was shaking a little. Take a breath, I told her, because that headline is doing a great deal of frightening work that a careful reading simply doesn't support. There is a real, and rather elegant, piece of science sitting underneath it, out of Boston this summer, and it's well worth understanding properly. But it's a clue about a mechanism — about how the disease might get going — not a verdict on anybody's future. So let me give you the honest version, the one I gave her.
What the Boston team actually found
Christopher Walsh, Alice Eunjung Lee and August Yue Huang — working at Boston Children's Hospital with Harvard Medical School and the Broad Institute — went looking in brain tissue for the sort of DNA mutations we normally file under "cancer". They read 149 cancer-driving genes across the brains of 190 people who had died with Alzheimer's, and set them beside 121 brains from people who hadn't. The Alzheimer's brains carried more mutations in five of those same cancer-driver genes — those are the researchers' figures, and I'd hold them lightly until others repeat the work. Then came the detail that made everyone sit up straight: the very same mutations turned up in the patients' blood cells. The same typo, written into two different notebooks. That's the sort of coincidence that stops being a coincidence — it hints that whatever went wrong in the brain may have started, quite literally, in the blood.
First, meet the brain's cleaning staff
To see why that's interesting, borrow a frame from Cognitive Neuroscience, the slim Oxford introduction. Your brain keeps its own resident immune cells, called microglia, and the honest job description is janitor. They patrol the tissue, sweep up debris and dead cells, and quietly prune the synaptic connections you've stopped using. That pruning isn't damage, by the way — it's part of how a brain learns, keeping the wiring you use and clearing the wiring you don't. When they're doing their work well you never notice them — a good janitor is invisible. Hold on to that picture, because the whole story turns on what happens when one of them stops sweeping and starts striking matches.
How a janitor picks up a box of matches
Here's the chain the Boston team propose, and I want to be plain that "propose" is the operative word. As we age, or after an injury, the blood–brain barrier — the border wall that normally keeps blood cells out of brain tissue — grows leaky. A mutant immune cell from the blood slips across. Once inside, it transforms into a microglia-like cell; but it's carrying a cancer-type mutation, the kind that hands a cell a growth advantage, so it doesn't just settle in — it multiplies and spreads. And this expanding population behaves badly. It makes its patch of brain more inflammatory than healthy microglia ever would, and that inflammation can damage, and even kill, the very neurons the cell was meant to be tidying around. The janitor, in short, has picked up a box of matches — and it's breeding more like itself.
'A little like cancer' — and where the honesty lives
Walsh put it in a line worth quoting: "Alzheimer's disease is a little like cancer — driven by the same mutations that drive blood cancers like lymphoma and leukaemia." Huang added that "the blood's immune cells with cancer mutations are likely getting into the brain and contributing to disease." Now, the honest version, because this is where headlines and science part company. What the team have firmly shown is an association — more of these mutations in Alzheimer's brains, and the same ones sitting in the blood. The border-crossing, arsonist-microglia story is a proposed mechanism: a well-argued, biologically sensible hypothesis, not yet a filmed chain of events. The evidence here is genuinely promising, but it's early, and two things found together is not the same as one thing proven to drive the other. Read it as a strong lead, not a closed case.
A good janitor is invisible; the trouble starts when it lights a fire
Microglia are meant to clear debris and protect neurons — the worry is that a cancer-type mutation flips a small population into an inflammatory, self-multiplying crowd that harms the very cells it was there to tend.
So what do you do with this on a Tuesday?
For now, in the clinic, almost nothing changes — and that's the honest answer, not a disappointing one. The tantalising long-term possibility is that a blood test for these mutations might one day flag people at raised Alzheimer's risk independently of APOE4, the risk gene we currently lean on. But "one day" is carrying real weight in that sentence: no such test exists for you to ask for yet, and I'd be wary of anyone selling one. What genuinely shifts is how we think about the disease. It turns the spotlight onto neuroinflammation — onto the brain's own housekeeping cells — as something that might help drive the damage, rather than just rubble left in its wake. And re-framing a disease as something being actively done, by identifiable cells, is usually where new treatments start looking for a foothold.
So, to the woman with the clipping: no, you haven't got cancer of the brain, and no test is going to hand you your future this year. What we actually know is enough to be genuinely interested, and nowhere near enough to lose a night's sleep over. If you're worried about your own memory, or a parent's, that worry is worth a proper conversation — just have it with the person who knows your history, not with a headline. Take a breath. Then take it to your own GP.
Framing drawn from Cognitive Neuroscience (Oxford Very Short Introduction): microglia are the brain's resident immune cells — its housekeepers, pruning synapses and clearing debris. The study is from Christopher Walsh, Alice Eunjung Lee and August Yue Huang (Boston Children's Hospital / Harvard Medical School / Broad Institute), published June 2026; all figures are the researchers' own. These findings are preliminary — an association plus a proposed mechanism, not a proven cause, and an association is not the same as causation. This is popular science, not medical advice — for anything about your own memory or risk, talk to your own doctor.
A proposed mechanism from Boston Children's Hospital (Walsh, Lee & Huang, 2026): a blood immune cell carrying a cancer-driver mutation may cross a weakened blood–brain barrier, turn into a microglia-like cell that multiplies, and make its patch of brain inflammatory enough to harm neurons — the brain's janitor becoming an arsonist. The same mutations also sat in patients' blood (190 Alzheimer's vs 121 control brains; 5 of 149 cancer-driver genes). Frame: Cognitive Neuroscience (microglia as the brain's resident immune cells). Findings are preliminary — an association plus a proposed mechanism, not proven cause; this is popular science, not medical advice.
ボストン小児病院が示した一つの機序(Walsh・Lee・Huang、2026):がん駆動変異を持つ血中の免疫細胞が、緩んだ血液脳関門を越えてミクログリア様細胞になり、増殖し、その一帯を神経を傷つけるほど炎症に傾ける——脳の掃除係が放火犯になる、というわけだ。同じ変異は患者の血液にも座っていた(アルツハイマー脳190例 vs 対照121例;149遺伝子中5つ)。枠組:『認知神経科学』(ミクログリアは脳の常駐免疫細胞)。これは初期段階の知見で、関連と推定機序であって証明された因果ではない。本稿は科学解説であり医療助言ではない——気になるときはかかりつけ医に相談を。