Standard treatment for untreated advanced-stage follicular lymphoma is rituximab plus chemotherapy. The incidence of histological transformation of follicular lymphoma has been reported only in heterogeneously treated populations and rarely with long-term follow-up. Additionally, the incidence of secondary malignancies after treatment, without high-dose therapy for follicular lymphoma, is largely unknown. The aim of our study was to assess progression-free survival, overall survival, incidence of secondary malignancies, and incidence of histological transformation in a 10-year follow-up analysis of the JCOG0203 trial. In the phase 2-3 randomised JCOG0203 trial, previously untreated patients with stage III or IV indolent B-cell lymphoma, including grades 1-3 follicular lymphoma, from 44 hospital centres in Japan, were randomly assigned (1:1) by use of a minimisation method to receive six cycles of R-CHOP (rituximab [375 mg/m2], given on day 1, plus cyclophosphamide [750 mg/m2], doxorubicin [50 mg/m2], vincristine [1·4 mg/m2, capped at 2·0 mg] given intravenously on day 3, and oral prednisone [100 mg once daily on days 3-7]) every 3 weeks (R-CHOP-21) or every 2 weeks (enabled by mandatory granulocyte-colony stimulating factor administration once daily for 6 days, starting on day 8; R-CHOP-14) without rituximab maintenance. Age, bulky disease (nodal or extranodal mass ≥10 cm in diameter on CT), and institution were used as adjustment factors. Investigators enrolled participants, and assignment to trial groups was done with a computer-assisted randomisation allocation sequence that took place centrally at the Japan Clinical Oncology Group Data Center, without the intervention of investigators. Interventions were not masked for patients or investigators. Data were collected 10 years after enrolment of the last patient. The primary endpoint of the phase 3 part of the study was progression-free survival, and the primary endpoint of the phase 2 part of the study was the proportion of patients who achieved a complete response. Accrual was 4·5 years, and follow-up was 3 years after registration was closed. Data were updated on the cutoff date of Feb 28, 2017. Intention-to-treat analyses (ie, progression-free survival, overall survival, and incidence of secondary malignancies) were predefined, to be done at 10 years after the last patient was enrolled. An additional analysis of the incidence of histological transformation was defined 15 years after the protocol, on May 8, 2017, in a supplementary analysis plan, and assessed at 10 years after the last patient was enrolled. Follow-up is ongoing. This trial is registered with ClinicalTrials.gov, number NCT00147121. Between Sept 1, 2002, and Feb 28, 2007, 300 patients were enrolled, and 149 (50%) were assigned to the R-CHOP-21 group and 151 (50%) were assigned to the R-CHOP-14 group. After eligibility was assessed, one patient was excluded from the R-CHOP-21 group. 10-year progression-free survival was not different between groups (R-CHOP-21 33%, 95% CI 25-41; R-CHOP-14 39%, 31-47; hazard ratio 0·89, 95% CI 0·67-1·17). In 248 patients with grade 1-3a follicular lymphoma, progression-free survival was 39% (33-45) at 8 years and 36% (30-42) at 10 years. The cumulative incidence of histological transformation was 3·2% (95% CI 1·5-6·0) at 5 years, 8·5% (5·4-12·4) at 8 years, and 9·3% (6·1-13·4) at 10 years after enrolment. At 10 years, the cumulative incidence of secondary malignancies was 8·1% (5·1-12·0) and the cumulative incidence of haematological secondary malignancies was 2·9% (1·3-5·5). R-CHOP is a viable option for first-line treatment in patients with newly diagnosed advanced follicular lymphoma. Clinicians choosing a first-line treatment for patients with follicular lymphoma should be cautious of secondary malignancies caused by immunochemotherapy and severe complications of infectious diseases in the long-term follow-up-both of which could lead to death. National Cancer Center and Ministry of Health, Labour and Welfare of Japan.