Section 1 of 9What is GSB-106?
What is GSB-106?
GSB-106 (ГСБ-106) is a synthetic dimeric dipeptide, bis(N-monosuccinyl-L-seryl-L-lysine) hexamethylenediamide, designed to mimic part of BDNF. It is a distinct molecule, not the full BDNF protein. Early rodent studies reported antidepressant-like activity in behavioral tests. The BDNF–TrkB guide explains the endogenous signal; it does not establish the efficacy of this compound. C01 (opens in a new tab)
Section 2 of 9Cell protection has a concentration-dependent limit
Cell protection has a concentration-dependent limit
In a 2021 study of serum-deprived human neuroblastoma cells, GSB-106 promoted survival, with a maximum response about 37% of the maximum BDNF response in that assay. Higher concentrations tended to reduce viability. Signaling and inhibitor experiments supported TrkB-related mechanisms, but this study did not measure direct receptor binding. Human-derived tumor cells are not human participants, and these concentration-response results do not define a human therapeutic window. C02 (opens in a new tab)
Section 3 of 9What did the later binding experiment show?
What did the later binding experiment show?
A 2022 Russian surface-plasmon-resonance study found a roughly 30% reduction in subsequent BDNF binding signal after preincubating the TrkB extracellular domain with GSB-106. This supports a possible direct interaction. However, the experiment used two measurements, and its significance rule relied on non-overlapping standard deviations. It did not establish a binding-affinity constant for the compound. The result adds evidence beyond cell signaling, but should not be presented as a complete receptor-pharmacology characterization. C03 (opens in a new tab)
Section 4 of 9Oral activity does not establish human efficacy
Oral activity does not establish human efficacy
A 2018 rat forced-swim study reported activity after oral administration of both the substance and a tablet formulation. Some tested doses were inactive; more was not uniformly better. The lowest reported active doses differed between formulations. This is not a human dose conversion or proof of superiority to amitriptyline in patients. The article contains an inconsistency between one table row, its reported ratio and the prose; that numerical comparison is not reproduced here as a verified effect size. C04 (opens in a new tab)
Section 5 of 9Positive stress-model findings, with an important statistical limit
Positive stress-model findings, with an important statistical limit
Two 2021 mouse studies reported benefits in chronic stress models. In social defeat stress, oral treatment improved sucrose preference and hippocampal protein markers. The claimed locomotor improvement depended on the statistical test: one analysis found only a trend, while another reached the significance threshold. In unpredictable mild stress, repeated oral treatment reduced forced-swim immobility and restored reduced brain BDNF levels. These behavioral and biochemical outcomes do not demonstrate recovery from human depression, and overlapping research teams do not constitute independent replication. C05 (opens in a new tab)
Section 6 of 9What did the toxicology study test?
What did the toxicology study test?
A 2019 Russian study tested a specific tablet mixture containing 1% active compound. It reported no deaths after the tested single administrations in mice and rats, and no toxic effects in the assessed laboratory, organ and tissue endpoints after one month of repeated administration in rats and rabbits. Active-compound mass and total tablet mass must not be confused. These are bounded animal findings, not evidence of no side effects in humans or proof of reproductive, cancer or long-term safety. C06 (opens in a new tab)
Section 7 of 9What does the withdrawal experiment leave unanswered?
What does the withdrawal experiment leave unanswered?
In a 2018 rat study, the behavioral effect persisted between 14 and 28 days of administration. Selected anxiety, exploration and movement tests found no withdrawal-related changes 36 hours after cessation. That observation cannot exclude later effects or human dependence and withdrawal. The paper also contains an unresolved inconsistency between the stated total animal count and its group description, which limits confidence in the reporting. C07 (opens in a new tab)
Section 8 of 9The newer half-life result is not a human measurement
The newer half-life result is not a human measurement
The Russian abstract of a 2025 pharmacokinetic study reports plasma half-lives of 0.6 hours in rats and 1.3 hours in rabbits after a single oral administration. Its approximately 3-hour human half-life is an allometric prediction from animals, not a measured human result. Only the abstract was reviewed here. Neither these half-lives nor oral activity in rodents establish a human dosing interval, bioavailability or benefit. C08 (opens in a new tab)
Section 9 of 9What remains to be established?
What remains to be established?
A separate Russian paper from 2025 optimizes tablet granulation and physical quality. It is pharmaceutical development, not a patient trial. The reviewed sources establish neither clinical efficacy nor a human adverse-event profile. A standardized formulation, reproducible results from independent teams and adequately controlled human studies are separate evidence needs. The earlier papers’ use of the word antidepressant should therefore be read in its experimental context. C09 (opens in a new tab)