Introduction: In the international randomised phase 3 'Watch & Wait' (W&W) trial patients with asymptomatic low tumour burden follicular lymphoma (LTBFL) were randomised between (1) W&W, (2) rituximab induction (RI) or (3) RI followed by rituximab maintenance (RM). The primary endpoints were time to initiation of new treatment (TTNT) and quality of life (QoL) at 7 months. After a median follow up of ~4 years 88% of RM had not initiated new treatment at 3 yrs versus 46% with W&W. There was no significant difference in TTNT between the 2 rituximab (R) arms at 3yrs. There was no difference in overall survival (OS) between the arms. Selected QoL measures were shown to be superior for patients in the RM arm compared to W&W. Follow up of the study was extended to assess if the improvement in TTNT was sustained, and to collect the time to initiation of 2nd new treatment (TT2NT) to explore if the early use of R impacted patients when R-chemotherapy was subsequently given. Here we report the final analysis of the W&W study with median follow-up of 12.3 years (IQR 11.4-13.3). Methods: From Oct 2004 patients age ≥18 with stage II-IV, asymptomatic, LTBFL (grades 1-3a) were randomly assigned 1:1:1 to RI (R 375 mg/m2 weekly for 4 weeks), RM (RI followed by R every 2 months for 12 doses) or W&W. Recruitment to the RI arm was closed early in Sept 2007 and a 2-arm study then continued to recruit until March 2009. The primary endpoint was TTNT (including chemotherapy or radiotherapy). Secondary endpoints included response at 25 months, PFS, OS, cause-specific survival (CSS, including deaths due to lymphoma or treatment) and rates of spontaneous remission and high-grade transformation (HT). TT2NT was measured to compare the efficacy of the composite of (observation +1st systemic therapy) with (rituximab +1st systemic therapy) Results: Of 463 patients recruited from 118 centres in the UK, Australia, New Zealand, Turkey, and Poland, 84 were randomised to the RI arm, 192 to the RM arm and 187 to the W&W arm. New treatment was initiated in 242 patients (W&W 133, RI 42, RM 67). Median TTNT was 2.7 years (95% confidence interval (CI) 2.2-4.0) for the W&W arm, 9.9 years (5.7-not reached) for the RI arm and was not reached for the RM arm. At 10 years 28.8% (22.1%-35.8%) of patients in the W&W arm had not started a new treatment compared to 49.4% (37.4%-60.3%) in the RI arm (p<0.0001) and 64.5% (56.7%-71.2%) in the RM arm (p<0.0001). TTNT was also significantly longer for patients in the RM arm compared to the RI arm (HR = 0.61 (0.41-0.90), p=0.012). Second new treatment was initiated in 75 patients (W&W 35, RI 13, RM 27). The median TT2NT was not reached in any of the 3 arms, with no significant differences observed between the groups. At 25 months the overall response rate (ORR) was 52.4% in the RI arm (complete response (CR) 31.0%, CRu 9.5%) and 76.0% in the RM arm (CR 63.0%, CRu 5.7%). The rate of disease progression by 25 months was 57.8% for the W&W arm, 36.9% for the RI arm and 15.1% for the RM arm. 35 patients in the W&W arm (18.7%) had a confirmed spontaneous remission at any point during follow-up with median time to spontaneous CR of 1.6 years (1.2-1.8) with 24/35 subsequently relapsing or requiring further treatment. The 4-year PFS rates for both treatment arms were superior when compared to W&W (W&W 32.1% (95% CI 25.5%-38.9%) RI - 56.5% (45.2%-66.4%, p=0.0001), RM - 78.4% (71.9%-83.6%, p<0.0001). PFS for the RM arm was also superior compared to the RI arm (HR = 0.41 (0.27-0.65), p<0.001). No significant differences in OS or CSS were observed with 10yr OS rates of 76.4%, 80.3% and 82.5%, and 10 yr CSS rates of 87.1%, 87.9% and 89.7% in W&W, RI and RM respectively. HT occurred in 34 patients (18.2%) in the W&W arm, 9 (10.7%) in the RI arm and 26 (13.5%) in the RM arm, with no significant differences in the time to HT between the 3 groups. Rates of second primary malignancy between the 3 groups were also comparable at 18.7%, 17.9% and 15.6% for the W&W, RI and RM arms respectively. Conclusion: After a long median follow-up of 12.3 years we have shown that rituximab monotherapy is highly effective at deferring TTNT in patients with asymptomatic LTBFL, with RM being superior to RI. Initial treatment with rituximab did not result in inferior outcomes following 1st new treatment compared with an initial period of observation. Upfront rituximab should therefore be considered an effective treatment option for patients who wish to delay chemoimmunotherapy. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal