Background: In patients with recently diagnosed multiple myeloma, the HOVON-50 phase 3 trial showed improved event-free survival for thalidomide-containing induction and maintenance regimens (in conjunction with high-dose melphalan and autologous stem cell transplantation [auto-SCT]) after a median of 52 months of follow-up, by comparison with regimens containing classical cytotoxic drugs. In this follow-up analysis, we aimed to determine the long-term effects of thalidomide in induction and maintenance therapy in multiple myeloma. Methods: In this open-label, phase 3 randomised controlled trial, patients with recently diagnosed multiple myeloma were recruited from 44 Dutch and Belgian hospitals. Participants had been diagnosed with multiple myeloma of Durie-Salmon stage II or III and were aged 18–65 years. Patients were randomly assigned (1:1) either to receive three 28-day cycles of vincristine (0·4 mg, intravenous rapid infusion on days 1–4), doxorubicin (9 mg/m2, intravenous rapid infusion on days 1–4) and dexamethasone (40 mg, orally on days 1–4, 9–12, and 17–20; control group); or to receive the same regimen, but with thalidomide (200–400 mg, orally on days 1–28) instead of vincristine (thalidomide group). No masking after assignment to intervention was used. Patients were randomly assigned to groups, stratified by centre and treatment policy (one vs two courses of high-dose melphalan and auto-SCT). After stem cell harvest, patients received one or two courses of 200 mg/m2 melphalan intravenously with auto-SCT. Patients with at least a partial response to high-dose melphalan and auto-SCT were eligible for maintenance therapy, starting 2–3 months after high-dose melphalan. Patients in the control group received maintenance therapy with interferon alfa (3 × 106 international units, subcutaneously, three times weekly). Patients in the thalidomide group received thalidomide as maintenance therapy (50 mg, orally, daily). Maintenance therapy was given until relapse, progression, or the occurrence of adverse events. The primary endpoint of the study was event-free survival (EFSc; censored at allogeneic stem cell transplantation), analysed by intention to treat. The study is closed for enrolment and this Article represents the final analysis. This trial was registered with the Netherlands Trial Register, number NTR238. Findings: Between Nov 27, 2001 and May 31, 2005, 556 patients were enrolled in the study, of whom 536 (96%) were eligible for evaluation and were randomly allocated (268 [50%] to the control group and 268 [50%] to the thalidomide group). These 536 patients were assessed for the primary endpoint of EFSc. At an extended median follow-up of 129 months (IQR 123–136), EFSc was significantly longer in the thalidomide group compared with the control group (multivariate analysis hazard ratio [HR] 0·62, 95% CI 0·50–0·77; p<0·0001). Thalidomide maintenance was stopped because of toxicity in 65 (42%) of 155 patients in the thalidomide group (neuropathy in 49 [75%] patients, skin reactions in four [6%] patients, fatigue in two [3%] patients, and as other symptoms [such as abdominal pain, pancreatitis, and dyspnoea] in ten [15%] patients). 24 (27%) of 90 patients in the control group discontinued protocol treatment during maintenance therapy with interferon alfa because of toxicity (five [21%] patients with psychiatric side-effects, five [21%] patients with flu-like symptoms, four [17%] patients with haematological toxicity [thrombocytopenia and leucocytopenia], three [13%] patients with skin reactions, and seven [29%] patients with other symptoms [such as infections, cardiomyopathy, and headache]). The frequency of second primary malignancies was similar in both groups. There were 23 second primary malignancies in 17 patients in the control group and 29 second primary malignancies in 24 patients in the thalidomide group. There were 19 treatment-related deaths in the control group, and 16 treatment-related deaths in the thalidomide group. Interpretation: Our data indicate that thalidomide-based treatment could be a treatment option for patients with multiple myeloma who are eligible for auto-SCT who live in countries without access to proteasome inhibitors or lenalidomide. However, careful follow-up and timely dose adjustments are important to prevent the development of thalidomide-induced neurotoxicity. Funding: The Dutch Cancer Foundation.