Abstract Disclosure: D.J. Handelsman: None. M. Grossmann: None. B.B. Yeap: Consulting Fee; Self; Bayer, Inc., Lawley. Research Investigator; Self; Lawley. B.G. Stuckey: Speaker; Self; Lawley. N. Shankara-Narayana: None. A.J. Conway: None. W. Inder: None. R.I. McLachlan: None. C.A. Allan: None. A. Jenkins: None. D.R. Jesudason: None. K. Bracken: None. K. Robledo: None. G.A. Wittert: Research Investigator; Self; Bayer, Inc., Lilly USA, LLC. The T4DM study randomized 1007 men (age 50-74 yr, waist ≥95 cm, serum T ≤14.0 nmol/L, no pathological hypogonadism) with impaired glucose tolerance or newly diagnosed type 2 diabetes to T undecanoate (1000 mg) or matching placebo injections every 12 weeks for 24 months with a lifestyle program. At study’s end, T treatment reduced OGTT diabetes diagnosis by 40% without change in HbA1C. After a median of 5 years since last injection, a follow-up email survey obtained 705 responses (70%) comprising 599 completed surveys (316 T, 283 P, 10 deceased, 95 declining). Participants in follow-up survey were similar to non-participants in 23 anthropometric and demographic variables at entry to T4DM, but more in follow-up study were randomized to T (53 vs 46%%, p=0.038), had lower entry weight (107 vs 109 kg, p=0.026) and older school leaving age (16.7 vs 16.5 years, p=0.026) but remaining well matched for work status (46% retired, 35% full-time work, p=0.60), alcohol intake (69% nil or light, 3% heavy; p=0.12) and smoking (3% smoking, p=0.61). At long-term follow-up, randomization to T treatment was associated with stronger belief in benefits during study (64% vs 49%, p<0.001) with less for post-study benefits (44% vs 40%, p=0.07) but overall high interest in future studies (surveys 93%, clinical 76%). At entry, 35% had sleep apnea, most (71%) diagnosed pre-study, with new diagnosis more frequent on T during (9% vs 1%, p=0.03), but not before or after, the study. After study, T treatment was prescribed at a higher rate for men who had study T treatment (6% vs 2.8%, p=0.03), mostly using T injections with 81% continuing at 24 months (median) on post-study T treatment. In the long-term, study T treatment did not influence self-reported new post-T4DM study diagnosis of diabetes (19%, p=0.65) or diabetes drug treatments (22%, p=0.31; oral 18%, injectable 2%, both 2%), prostate disease (p=0.49; cancer 3%, non-cancer 8%) or cardiovascular disease (p=0.95; heart disease 13%, stroke 1%). Both groups had similar weight maintenance (maintained 24%, lost 22%, gain 24%, up/down 30%, p=0.91) and further attempts to lose weight (76%, p=0.29; diet 68%, exercise 57%, drugs 6%, surgery 2%). We conclude that randomized T (vs placebo) treatment for 24 months in men with impaired glucose tolerance or new diabetes but without pathological hypogonadism, was associated with higher rates of (a) self-reported benefits during, but not after the study, (b) of resuming T treatment after the study, consistent with androgen dependence due to withdrawal (androgen deficiency) symptoms and/or recalling perceived benefits of study T treatment, and (c) of sleep apnea diagnosis during, but not before or after the study. Five years after T treatment stopped, there was no difference in the long-term rates of self-reported new diagnosis of diabetes, prostate or cardiovascular disease nor change in weight maintenance or weight loss behaviors. Presentation: Friday, June 16, 2023