If gabapentin ignores GABA receptors, why does the 300 mg capsule still carry that name?
GABA was the bait, not the dock
Native GABA stays in blood. That was the design problem, not a marketing line.
Gamma-aminobutyric acid is the main inhibitory transmitter in mammalian brain. Raising its tone can quiet a seizure circuit. The practical snag is simple: GABA is hydrophilic and does not cross the blood-brain barrier in useful amounts. 1970s medicinal chemists grafted a cyclohexane ring onto the GABA backbone to add lipophilicity. On paper that is a brain-penetrant GABA mimic.
The name gabapentin encodes that plan. Every early assumption pointed at GABA receptors or GABA enzymes. None of those assumptions survived a clean binding panel. I still write that in the margin when a student treats the INN as a mechanism. It is a chemistry souvenir.
You will meet the same molecule as a yellow 300 mg capsule printed Neurontin on the cap. That strength is a real US label option - 100, 300, and 400 mg capsules sit beside 600 and 800 mg tablets. The 300 mg unit is how most adult titrations begin. It is not proof the drug works like GABA. Full handling notes live in the Neurontin 300 mg medicine page.
Binding work that emptied the original theory
The FDA label is blunt on this point. Gabapentin is structurally related to GABA and has no effect on GABA binding, uptake, or degradation. No meaningful GABA-A occupancy. No GABA-B occupancy. No conversion into GABA. No clinically relevant GABA-transaminase block. A GABA-named capsule that does not do GABA work.
What it does bind, with high affinity, is the alpha-2-delta auxiliary subunit on voltage-activated calcium channels. Occupancy at that site reduces calcium influx at overactive terminals and cuts excitatory transmitter release. That story fits injured sensory axons better than it fits a generic 'calming GABA' pitch. The relationship between binding and the clinical effect is still formally unknown - the label says so - but the GABA story is finished.
Peer-line I keep for residents: if a patient asks whether Neurontin 300 mg is a benzodiazepine cousin, the answer is no. It does not sit on GABA-A, and it does not sit on opioid or cannabinoid receptors either. Misuse still happens. The receptor card is not a benzodiazepine card.
What Parke-Davis actually received in December 1993
Developer: Parke-Davis, then under Warner-Lambert. The first US stamp was narrow on purpose.
December 1993: FDA cleared Neurontin as adjunctive therapy for partial seizures in adults. Adjunctive means add-on. Not monotherapy. Not first-line. Not a pain drug. The approved market was modest. The commercial hope was larger than the label.
Inside that lane the capsule had real appeal. No routine serum level. Almost no hepatic metabolism, so few CYP fights with other antiepileptics. Renal excretion of unchanged drug. Tolerability that let people stay on an add-on. Those traits later made off-label spreading easy. They did not expand the licence by themselves.
Adult epilepsy still starts, on the current label, at 300 mg three times daily in people 12 and older, with room to 600 mg three times daily. The 300 mg capsule is the unit of that start. Maximum gap between doses should not exceed 12 hours - half-life is only 5 to 7 hours when kidneys work. That PK detail already belongs in a history note, because the same short half-life later shaped pain titration.
How a 300 mg capsule walked out of the epilepsy lane
US law splits two acts. A clinician may prescribe off-label. A manufacturer may not promote off-label. Parke-Davis and Warner-Lambert promoted Neurontin for neuropathic pain, bipolar illness, migraine, ADHD, restless legs, and epilepsy monotherapy that the 1993 stamp did not cover. The Department of Justice later listed those pitches in the 2004 resolution.
Tactics were not ordinary detailing. Funded articles. Paid speaker tours. CME steering. Literature seeding. David Franklin, a former Parke-Davis medical liaison, filed the whistleblower action that opened the Massachusetts case. By 2004 Warner-Lambert - then a Pfizer subsidiary - pleaded guilty and paid more than $430 million in criminal and civil resolution. About $240 million of that was a criminal fine for misbranding.
I file that number because students still treat 'Neurontin 300 mg' as a harmless yellow vitamin. The capsule is a real drug. The reputation inflation was a marketing project. Evidence that later held up is a separate page: what Neurontin 300 mg actually treats.
The 2002 PHN add that later fit the mechanism
One heavily sold off-label use later earned a legitimate second stamp.
1970s
Chemists build a lipophilic GABA analog so a polar transmitter can enter brain.
1980s-1990s
Binding work shows no GABA receptor, uptake, or degradation effect; alpha-2-delta later becomes the high-affinity site.
Dec 1993
FDA clears Neurontin as adjunctive therapy for partial seizures in adults.
Late 1990s
Promotion spreads into pain, mood, migraine, and other uses the 1993 stamp did not cover.
2002
FDA adds postherpetic neuralgia in adults after two placebo-controlled trials.
13 May 2004
DOJ announces Warner-Lambert will pay more than $430 million and plead guilty over off-label promotion.
Postherpetic neuralgia is burning, often lingering pain after the shingles rash heals. The label trials enrolled adults with pain lasting more than three months after the skin healed. Two randomized, double-blind, placebo-controlled multicenter studies put 563 people in the intent-to-treat set. That is a nerve-injury model. Alpha-2-delta occupancy has a place there.
FDA added the PHN indication in 2002. Adult PHN titration on the current label starts with a single 300 mg dose on day one, 300 mg twice on day two, 300 mg three times on day three, then climb toward 1800 mg per day as 600 mg three times daily. Trials looked at 1800 to 3600 mg per day. Extra benefit above 1800 mg was not shown. The 300 mg capsule is the first brick, not the destination.
Verdict I write in the margin: some promoted uses never grew a dataset. PHN did. Prescribing habits from the marketing years still outrun both stamps. A history of illegal promotion does not erase a later RCT. It also does not bless every leftover script. Practical numbers sit in the eGFR-first 300 mg titration note.
Why the GABA name still misleads a new 300 mg script
Gabapentin fills a huge number of US scripts. Many of those scripts are not PHN and not add-on partial seizures. That sprawl is a leftover of the 1990s campaign plus a low perceived risk. A 300 mg start looks gentle. Kidneys that have already slowed do not treat 300 mg as gentle.
I tell students to read the name as a fossil. Then read the label indications. Then read creatinine clearance. The order matters. People who skip the middle step treat Neurontin as a universal nerve soother. People who skip the last step stack sedation on a reduced eGFR and call it 'sensitivity.'
File this: useful in defined nerve pain and as an add-on for partial-onset seizures. Not a universal analgesic. Not an anxiolytic with a stamp. Not a sleep tablet with a stamp. The history explains why those extra jobs piled up. It does not make them evidence.
Pregabalin as the later, intended sibling
Once alpha-2-delta was on the map, pregabalin (Lyrica) was built for that subunit. Absorption is nearer to complete and closer to linear. Doses are smaller in milligrams. Schedules are often twice daily. Gabapentin absorption saturates: about 60% bioavailability at 900 mg per day, falling toward 27% at 4800 mg per day in three divided doses. That is why a 300 mg capsule is a useful unit early and a leaky unit late.
Same target class. Not a milligram-for-milligram swap. Pregabalin is Schedule V federally in the US. Gabapentin is not federally scheduled, though several states now watch it. The history note is this: gabapentin found the target by accident. Pregabalin was aimed. Patients still get both. Switching needs a prescriber, not a ratio from a forum.
PK the archive often skips and a 300 mg start still needs
History notes love the lawsuit. Ward notes need the clearance.
Absorption
Saturable gut uptake; ~60% at 900 mg/day, lower as the daily total climbs; food effect small; wait 2 h after Al/Mg antacid.
Distribution
Vd ~58 L after 150 mg IV; <3% protein bound; CSF trough about 20% of plasma in epilepsy.
Metabolism
Not metabolized; no hepatic-impairment study; negligible CYP inhibition at clinical levels.
Excretion
Unchanged in urine; t1/2 5-7 h if CrCl is fair, ~52 h if CrCl <30; dialyzable.
Gabapentin is not metabolized. Protein binding is under 3%. Apparent volume of distribution after 150 mg IV is about 58 L. Food barely moves AUC and Cmax (about 14%). An aluminum-magnesium antacid cuts bioavailability about 20%; the label wants at least two hours after Maalox before the next gabapentin dose. Hepatic impairment studies were not done because there is nothing for the liver to do.
Half-life stays 5 to 7 hours when creatinine clearance is decent. It stretches toward 52 hours when clearance is under 30 mL/min. In anuric adults it is about 132 hours on non-dialysis days and about 3.8 hours during hemodialysis. That is why a copied 300 mg three-times-daily habit from a young adult is a different drug in a frail kidney.
What I file after reading the archive
Wrong design premise. Narrow 1993 add-on. Illegal expansion. Partial later vindication for PHN. A 300 mg yellow capsule that still starts most adult climbs. That is the arc. I do not need a morality play on top of it. I need the next prescriber to know which part of the arc they are standing on.
If the script is PHN or add-on partial-onset seizures, the history is background. If the script is leftover bipolar, ordinary backache, or 'something for sleep,' the history is the main text. Ask what the 300 mg capsule is for. Then decide whether the file supports staying, tapering, or never starting.
Primary label text sits at DailyMed Neurontin. The 2004 resolution is in the Department of Justice release. Patient-facing wording is on MedlinePlus gabapentin.
Reader mail
Reader questions on this article
Answered by Dr. Helena Vasquez, PharmD · Clinical pharmacology & drug safety
These notes came back with questions in the margin. Helena answered the ones that kept repeating.
The INN is a chemistry souvenir. Researchers wanted a brain-penetrant GABA derivative - GABA plus a cyclohexane ring - and named the analog for that plan. Later assays showed no GABA receptor binding, no conversion to GABA, and no relevant effect on GABA uptake or breakdown. High-affinity binding sits on the alpha-2-delta calcium-channel subunit. I tell people the yellow 300 mg Neurontin capsule records a failed hypothesis, not a receptor. Use the name to find the bottle. Use the label to find the job.
What did FDA actually allow in 1993, before anyone talked about nerve pain?
One adult job: adjunctive treatment of partial seizures. December 1993. Add-on only - another antiepileptic already on board, control still incomplete. Not monotherapy. Not first choice. Not PHN. The 300 mg three-times-daily start that still sits on today's epilepsy section grew from that first stamp. Almost every other modern use arrived later, either as the 2002 PHN add or as off-label habit. I send people to FDA and to DailyMed when they want the primary wording, not a forum summary.
I keep seeing a Neurontin lawsuit. Was the drug fake, or was the selling fake?
The selling was illegal. The molecule was real. Manufacturers may not promote uses the label does not carry. Clinicians may still prescribe off-label. Parke-Davis pushed pain, bipolar illness, migraine, ADHD, restless legs, and off-label monotherapy. David Franklin's whistleblower suit opened the case. In May 2004 Warner-Lambert pleaded guilty and paid more than $430 million, including a $240 million criminal fine for misbranding. The leftover is a generation of 300 mg scripts written from reputation. That is a marketing scar, not a proof the capsule is inert.
So is a Neurontin 300 mg start a legitimate medicine or a leftover from that campaign?
Both, depending on the line on the script. Legitimate and labeled: adult postherpetic neuralgia, and add-on partial-onset seizures (adults and children 3 and older on the current stamp). Leftover zone: broad mood, ordinary backache, 'something for sleep,' migraine prevention. I do not tell a PHN patient the history means they should refuse the capsule. I do tell a back-pain patient the history explains why the bottle appeared. Match the 300 mg unit to a supported job, then titrate with kidneys in view.
How does pregabalin sit next to this older 300 mg capsule?
Pregabalin is the aimed follow-on. Same alpha-2-delta family, more complete and linear absorption, smaller milligram numbers, often twice daily. Gabapentin absorption saturates - a 300 mg capsule early in a climb is a fair unit, and the same milligram late in a 3600 mg day is a leakier unit. Pregabalin is Schedule V federally. Gabapentin is not, though some states treat it as controlled. Do not swap milligram for milligram. A switch is a new prescription, not a kitchen-table ratio.
Does the messy archive mean I should refuse gabapentin if my clinician just started 300 mg?
No. The archive is about promotion, not about a poison capsule. If the target is PHN or add-on seizures, the trials and the stamp are on your side. If the target is vague, ask what the 300 mg is supposed to change in two to four weeks, and ask for a creatinine or eGFR before the climb. Stopping cold is a separate mistake - withdrawal and, in epilepsy, extra seizures. History is a reason to ask a precise question. It is not a reason to throw the bottle in the bin tonight.
Why do you keep mentioning kidneys in a history note?
Because the same molecule that missed GABA is cleared unchanged by the kidney. Half-life is 5 to 7 hours with fair clearance and about 52 hours when clearance is under 30 mL/min. A copied 300 mg three-times-daily habit from a young adult is a different exposure in an older person with a creeping creatinine. Few CYP interactions is the pleasant half of that story. Accumulation is the other half. History without renal math is a museum card. Ward work needs both.
Was PHN just the company cleaning up after the lawsuit?
The 2002 PHN stamp rests on two randomized, placebo-controlled multicenter trials - 563 adults in the intent-to-treat set, pain lasting more than three months after the zoster rash healed. That is a real dataset, not a press release. Promotion of other pain and mood uses was the illegal part. I keep those two facts on separate lines. PHN earned a label. Bipolar and mechanical backache did not. A 300 mg day-one start for PHN is label language. A 300 mg start for 'nerves' is a different claim.
Where should I read the primary file instead of a recap?
DailyMed for the current Neurontin prescribing information - indications, the 300 mg titration, renal table, taper rule, respiratory warning. The May 2004 DOJ release for the guilty plea and the $430 million figure. MedlinePlus if you want patient wording without the legal PDF. On this desk the medicine page is Neurontin 300 mg, the use map is what it treats and what it does not, and the numbers live in the eGFR-first titration note. Read those before you trust a social-media timeline.
General education from a clinician, not personal medical advice. Bring your own history to your own prescriber.
Last Updated