Background Venetoclax plus obinutuzumab has been established as a fixed-duration treatment regimen for patients with chronic lymphocytic leukaemia. We compared the long-term efficacy after treatment cessation of the combination of venetoclax plus obinutuzumab with chlorambucil plus obinutuzumab in patients with previously untreated chronic lymphocytic leukaemia. Methods CLL14 is a multicentre, randomised, open-label, phase 3 trial done at 196 sites in 21 countries. Eligible patients were aged 18 years or older, had untreated chronic lymphocytic leukaemia, and coexisting conditions with a cumulative illness rating scale greater than 6, a creatinine clearance of 30–69 mL/min, or both. Patients were randomly assigned (1:1) via a web and voicemail system with allocation concealment and based on a computer-generated randomisation schedule with a block size of six and stratified by Binet stage and geographical region. Patients received either venetoclax plus obinutuzumab (oral venetoclax initiated on day 22 of cycle 1 [28-day cycles], with a 5-week dose ramp-up [20 mg, 50 mg, 100 mg, and 200 mg, then 400 mg daily for 1 week], thereafter continuing at 400 mg daily until completion of cycle 12; combined with intravenous obinutuzumab for six cycles starting with 100 mg on day 1 and 900 mg on day 2 [or 1000 mg on day 1], 1000 mg on days 8 and day 15 of cycle 1, and subsequently 1000 mg on day 1 of cycles 2 through 6) or chlorambucil plus obinutuzumab (oral chlorambucil at 0·5 mg/kg bodyweight on days 1 and 15 of each cycle for 12 cycles combined with the same obinutuzumab regimen). The primary endpoint was investigator-assessed progression-free survival in the intention-to-treat population. Safety was assessed in all patients who received at least one dose of study treatment. Patient enrolment is complete, and the study is registered with ClinicalTrails.gov, NCT02242942. Findings Between Aug 7, 2015, and Aug 4, 2016, 432 patients were enrolled and randomly assigned to receive either venetoclax plus obinutuzumab (n=216) or chlorambucil plus obinutuzumab (n=216). All patients had been off treatment for at least 24 months at data collection. At a median follow-up of 39·6 months (IQR 36·8–43·0), patients given venetoclax plus obinutuzumab had a significantly longer progression-free survival than did patients given chlorambucil plus obinutuzumab (HR 0·31, 95% CI 0·22–0·44; p<0·0001). Median progression-free survival was not reached (95% CI not estimable to not estimable) in the venetoclax plus obinutuzumab group vs 35·6 months (33·7–40·7) in the chlorambucil plus obinutuzumab group. The most common grade 3 or 4 adverse event in both groups was neutropenia (112 [53%] of 212 patients in the venetoclax plus obinutuzumab group versus 102 [48%] of 214 patients in the chlorambucil plus obinutuzumab group). Serious adverse events occurred in 115 (54%) of 212 patients in the venetoclax plus obinutuzumab group and 95 (44%) of 214 patients in the chlorambucil plus obinutuzumab group. Venetoclax or chlorambucil treatment-related deaths were reported in one (1%) of 212 patients in the venetoclax plus obinutuzumab group (n=1 sepsis) and two (1%) of 214 patients in the chlorambucil plus obinutuzumab group (n=1 septic shock, n=1 metastatic skin squamous carcinoma). Interpretation 2 years after treatment cessation, venetoclax plus obinutuzumab continues to significantly improve progression-survival compared with chlorambucil plus obinutuzumab, thereby providing a limited duration treatment option for patients with previously untreated chronic lymphocytic leukaemia. Funding F Hoffmann-La Roche and AbbVie.