DEA Wants ‘Gas Station Heroin’ Banned. History Says the Shelf Won’t Stay Empty.

The DEA’s own proposed rule doesn’t give the person a name. It gives a scene: a Pennsylvania overdose death, a digital scale, white residue on a nightstand, and a toxicology report that came back clean for everything except tianeptine. No fentanyl. No alcohol. No polysubstance asterisk to hide behind. Just a molecule that, as Filter’s Kastalia Medrano reported on July 8, you can currently buy at a gas station in that same state, today, for cash, no ID, no prescription.

If you have ever grabbed something off a convenience-store rack — a shot, a pouch, a pill — because the label promised focus or calm and you didn’t ask too many questions about what was actually in it, you are not naive. You are exactly the customer the label was built for.

The federal government has known tianeptine was dangerous for years and is only now getting around to deciding whether that matters.

That is the plain version of what happened on July 8, when DEA Administrator Terrance Cole’s agency published a Notice of Proposed Rulemaking to permanently place tianeptine into Schedule I of the Controlled Substances Act — the same legal category as heroin. The public has until August 7, 2026 to comment or request a hearing. Then the DEA decides, on its own timeline, whether the drug sold next to the beef jerky and the energy shots gets banned outright.

A depression drug that works like heroin if you take enough of it

Here is the layman’s version, because the pharmacology matters and nobody explains it plainly: tianeptine is a real antidepressant. In France, Mexico, and several other countries, doctors have prescribed it in small doses — 12.5 milligrams, three times a day — for decades, under brand names like Stablon and Coaxil. At that dose, it behaves like a conventional antidepressant, nudging serotonin reuptake in ways not too different from drugs already on U.S. pharmacy shelves.

But tianeptine has a second face. At high doses — the kind sold in unregulated U.S. products, not the kind in a doctor’s prescription pad — it binds to the same mu-opioid receptors that heroin, fentanyl, and oxycodone bind to. Take enough of it, and your brain cannot tell the difference between tianeptine and an opioid. That is not a side effect discovered by accident; it is why people take it. The unregulated market sells it as “Tianaa,” “Zaza,” “Neptune’s Fix,” and “TD Red” — pouches and capsules marketed for focus, mood, or opioid withdrawal relief, stocked in gas stations, vape shops, and smoke shops across the country, never named an opioid on the label because federally, until now, it has never legally been one.

The math on how fast this moved is stark. America’s Poison Centers and the FDA both cite the same trajectory: tianeptine-related poison control calls went from 4 cases in 2013 to roughly 350 cases in 2024 — an eighty-fold increase in the space of a decade, without the drug ever changing its legal status. The FDA’s consumer warning lists the harms in the flat language of a poison-control readout: agitation, drowsiness, confusion, sweating, rapid heartbeat, high blood pressure, nausea, vomiting, slowed or stopped breathing, coma, death. Withdrawal from high-dose tianeptine use, clinicians who treat it report, can look like opioid withdrawal — because pharmacologically, that is close to what it is.

The 2018 cluster the industry outran

This is not the first time tianeptine has forced a public reckoning, and the DEA’s proposed rule leans on the earlier one to make its case. In 2018, New Jersey’s Department of Health issued a health alert after a cluster of cases tied to a specific product, Neptune’s Fix, marketed as a dietary supplement. CDC and state health departments jointly investigated what turned into a multistate outbreak — and confirmed something worse than a single bad batch: some Neptune’s Fix products tested positive for synthetic cannabinoid receptor agonists that weren’t on the label at all, meaning people buying what they thought was a mood supplement were unknowingly dosing an entirely different, unregulated compound class.

In 2018, New Jersey’s Department of Health issued a health alert after a cluster of cases tied to a specific product, Neptune’s Fix, marketed as a dietary supplement.

The response, then as now, was warnings, not a ban. The FDA sent letters. Poison centers logged calls. The product stayed on shelves under different names. Six years later, tianeptine cases had grown roughly eightyfold, and the agency writing the 2026 rule is citing the same 2018 outbreak as evidence the danger was foreseeable all along — which raises the obvious question of why the rulemaking is happening in year eight of a known problem instead of year one.

There is a reason to think even Schedule I won’t be the last word. Fourteen years ago to the day this rule published in the Federal Register, Congress signed the Synthetic Drug Abuse Prevention Act of 2012 into law — the response to the last “gas station drug” panic, when the DEA used emergency scheduling powers in 2011 to control the synthetic cannabinoids sold as K2 and Spice and the synthetic cathinones sold as bath salts. Congress followed up by permanently scheduling 26 specific compounds. Within a few years, chemists had already reformulated around the named list, selling new analogues that weren’t technically banned because they weren’t named — a whack-a-mole dynamic the Congressional Research Service documented in its own review of synthetic drug policy. Scheduling a molecule does not schedule the incentive to invent the next one; it only raises the price of staying one step ahead of the DEA’s list.

Even Filter, no fan of scheduling, isn’t defending the shelf

It would be a mistake to read this as a clean story of an overreaching agency versus a harmless supplement. Filter’s own reporting — a publication that has spent years arguing against criminalization as a response to drug harm — pushes back hard on the DEA’s evidentiary framing, not on whether tianeptine is dangerous. Filter’s reporting notes that tianeptine “hasn’t been widely linked to fatal overdose” despite the agency’s language, questions the strength of evidence behind the two Pennsylvania deaths the DEA cites, and points out that the DEA’s filing omits an inconvenient fact: an earlier investigational drug application for tianeptine as a treatment was dropped for ineffectiveness, not danger.

That distinction matters, and it is the honest caveat a clinician reading this deserves before trusting the DEA’s framing wholesale: the agency’s case for permanent Schedule I status rests on documented harm and a rising poison-control curve, not on a large body of confirmed fatal-overdose cases with tianeptine as the sole cause. Two deaths, cited with incomplete toxicology detail, is a thin evidentiary reed for the CSA’s most restrictive category — even when the underlying concern about opioid-receptor activity and product contamination is well founded. Scheduling can be the right call for the wrong stated reasons, and both things can be true in the same rule.

What isn’t contested, by Filter or anyone else covering this: tianeptine products are unregulated, routinely adulterated, sold to anyone with cash, and marketed in a way that hides an opioid-receptor drug behind a wellness label. That is true whether or not the DEA’s two-death citation holds up to scrutiny, and it’s the part of this story that should worry a case manager or a parent more than the scheduling mechanics.

What Schedule I actually changes, and what it doesn’t

Scheduling tianeptine does a few concrete things: it makes manufacturing, distributing, and possessing it a federal crime, gives DEA authority to seize existing retail stock, and — in theory — removes it from gas station shelves nationally. It does not retroactively treat anyone currently dependent on high-dose tianeptine, and it does not fund the withdrawal management or treatment access that dependent users will need once the product they’ve been using daily disappears from the market. A ban without a treatment pathway on the other side of it is a policy half-finished, and nothing in the current NPRM addresses that gap.

A ban without a treatment pathway on the other side of it is a policy half-finished, and nothing in the current NPRM addresses that gap.

One thing any intake clinician can do this week, ban or no ban: tianeptine doesn’t show up on a standard urine drug panel or opioid immunoassay, so a patient can be actively dependent on it and screen clean. Adding one intake question — have you used a gas-station mood, energy, or “tianna/zaza” product in the past month? — will surface use that the lab work won’t.

It also, per the pattern above, likely does not end the product category. If the last fifteen years of designer-drug policy have taught regulators anything, it’s that a named ban creates market pressure toward an unnamed substitute — a new compound, a new blend, a new brand name selling the same opioid-receptor effect one legal step ahead of the schedule. Watching what a formulator does with tianeptine’s newly restricted status over the next twelve months will tell you more about whether this rule actually protects anyone than the scheduling vote itself.

For now, the docket is open. Public comments and hearing requests close August 7, 2026, and anyone — a treatment provider, a parent, a person who has used tianeptine themselves — can submit one directly to the Federal Register. If you or someone you know is using a tianeptine product and wants to stop, the FDA’s own guidance points to real, working options right now, ban or no ban: call the poison control hotline at 1-800-222-1222 for suspected overdose, or search findtreatment.gov for a provider who treats opioid-receptor dependence — because whatever the label says, that’s the kind of dependence this is.

The DEA got the 2018 cluster on the record, the poison-control curve, and Filter’s skepticism, all in the same proposed rule — and still needs until fall to decide. The rulemaking process moving through federal channels is the same one that took three years to schedule 7-OH kratom after that product had already saturated the same gas station shelves. The store shelf moves in weeks. The federal scheduling process moves in years. Nobody who wrote this rule is confused about which clock the next unnamed compound is racing against.