The Scorch of Silence: Death by Foam Board
The federal government is burning millions to learn whether cannabis is the savior or the executioner of the HIV-ravaged brain. The findings are real. The HIV press is asleep at the wheel.
By Felipe Recalde
In roughly nine days I’ll be standing in the Poster Hall at AIDS 2026. It’s a fluorescent-lit purgatory of white foam boards and academic jargon, the exact spot where high-stakes science goes to die in public. The air is going to taste like lukewarm coffee and industrial-grade desperation. You’ve got Ph.D.s in polyester suits guarding their data like religious relics, waiting for a nod from the gods that never comes. Right now the federal government is incinerating millions of dollars to work out whether cannabis is the savior or the executioner of the HIV-ravaged brain, and you’d never know it from the coverage. The findings are heavy. The stakes are visceral. And the HIV media is face-down in a gutter, asleep at the wheel. Nobody’s writing it down, and that’s the goddamn scandal.
Scientific priorities don’t talk. They bleed cash. Enter SCORCH, the Single-Cell Opioid Responses in the Context of HIV. It sounds like a black-ops weapons program. It’s actually a bureaucratic leviathan built to map the brain one twitching cell at a time. The thing is steered out of NIDA’s Division of Neuroscience in Rockville by the high priests of the federal checkbook, Susan Wright and John Satterlee. Back in 2023 they cut an $11.6 million check to Shomi Ndhlovu’s crew at Weill Cornell to see how cannabis warps or welds infected tissue. Tens of millions poured into the void, and the output is a run of posters pinned to the back wall of a convention center. Twenty-three thousand people will swarm IAS/AIDS 2026 and not one of
On paper, NIDA and NIAID are shaking hands. In the trenches, the SCORCH findings are hitting a brick wall. They aren’t translating to the AIDS Clinical Trials Group, the ACTG, the heavy hitters who actually run the trials that decide whether you live or rot. The ACTG’s Neurology Collaborative Science Group, chaired by Felicia Chow of UCSF, sits on the far side of a bridge nobody has built yet. No one on the executive committee is forcing the connection. So we watch tens of millions of people march toward cognitive decline while the researchers refuse to read off the same map. It’s a clinical failure dressed up in the robes of “process.”
And don’t let anyone tell you the data isn’t there. It’s piled to the ceiling. It just never leaves the building. Out at UC San Diego, Ronald Ellis and his colleagues pulled spinal fluid from 263 people and found that the daily cannabis users living with HIV had the calmest brains on paper. Their inflammatory chemokines, MCP-1 and IP-10, dialed down toward the levels of people who never carried the virus at all, and the calmer the fluid, the better they learned. It’s a cohort study, not a randomized trial, so association is the asterisk the scientists will always wave at you. But the signal is loud, and Ellis’s own review lays out the mechanism: cannabinoids tamping down the smoldering, body-wide inflammation that drives heart attacks, dementia, and death in people aging with the virus.
Then there’s the pill problem. Amy Justice at Yale has spent a career quantifying what it costs to grow old with HIV, and her polypharmacy data is a quiet horror show. Stack six drugs in an aging HIV patient’s medicine cabinet and you get roughly seven documented interaction pairs, against about one you’d expect by chance. This is the population SCORCH is supposed to protect. People already drowning in prescriptions, already aging a decade ahead of schedule. And the one intervention half of them are already self-administering, cannabis, is the one nobody in the clinical networks will study with any urgency.
Here’s the true absurdity. The system treats “cannabis” like a single grey blob. Most studies, including the elegant UC San Diego work on CBD and the brain’s TREM2 repair signal, use sterile CBD isolate. It’s safe, it’s boring, and it’s nothing like what humans actually put in their bodies: CBN, CBG, THC, and terpene profiles that would make a chemist weep. Ndhlovu’s Cornell trial is our best shot at a real answer, but here’s the catch. Nobody knows which cultivar they’re actually using. Not the dispensaries, not the regulators, not the industry.
And that’s the part that should make you throw your coffee. The tools to answer “which cannabis” already exist. Germplasm banks and DNA-marker fingerprinting can pin a plant to its genetic identity and its chemical profile. We know in black and white that the folk names are a lie. What sells as “Blue Dream” at the dispensary counter is, genetically, a coin toss. “Which cultivar?” isn’t an unanswerable question. It’s an unasked one. Answering it is exactly the applied science that outfits like Buffalo’s Dent Neurologic Institute, and our own affiliated ventures at GEM and Compound Genetics, exist to force onto the record.
Dent is worth pausing on, because it’s proof the clinical signal survives contact with real patients. In a chart review of 204 of their neurology patients, average age 81, roughly seven in ten reported that medical cannabis helped, and a third of them cut their opioid dose. It’s a retrospective review, not a gold-standard trial, but it’s exactly the kind of grounded, human evidence that the SCORCH machine, for all its millions, still refuses to connect to the people who need it.
Understand the scale of what’s being ignored. HIV-associated neurocognitive disorder isn’t some rare complication. The largest meta-analysis on record puts it at 42.6% of people with HIV, some 16 million human beings. And the population is graying fast. By 2021 more than half of Americans living with diagnosed HIV had already crossed 50, and the CDC now projects the ranks of the 75-and-older could swell as much as sixfold by 2040. Every one of those people runs roughly $420,000 in lifetime care, a bill that climbs, not falls, the longer they live. A brain you can’t keep is a patient who can’t adhere, can’t work, can’t live alone. This isn’t a footnote to the epidemic. It’s the next front of it, the same convergence of frailty, inflammation, and neglect this publication mapped in When the Levy Breaks.
The shift of medical marijuana toward Schedule III has kicked the doors open, but that window won’t stay propped forever. Federal funding is a political knife fight, and anything touching drug use and HIV is always first onto the chopping block. In the spring of 2025 the administration terminated $1.8 billion in NIH grants in a single month, and more than 200 HIV-related grants were killed outright. Congress preserved the topline NIH budget for 2026, but hundreds of terminated grants were never restored. If this science doesn’t mature into real therapies, FDA-approved drugs that reach a human hand, it’ll get swept away in the next budget brawl. We have to drag these findings out of the poster hall and onto the podium before the machine pulls the plug. In nine days I'll be at the microphone in Rio asking why. Somebody has to.