Milea Timbergen

171 Pain, with or without radiological evidence of progression, functional symptoms, or patient request, were frequently mentioned reasons for shifting to an ‘active’ treatment 10 . A total of 402 patients (reported in twenty studies) shifted to ‘active’ treatment. The median percentage of patients shifting in these studies was 29% (IQR: 17-40%). The type of ‘active’ treatment was systemic treatment in 195 cases, surgery in 107 cases, radiotherapy in 18 cases, a combination of therapies (e.g., systemic treatment with surgery, and systemic treatment with radiotherapy) in 8 cases and local therapy (e.g., radiofrequency and cryotherapy) in 4 cases. In 69 cases it was reported that patients shift to an active form of treatment but the type was unspecified (Table 3). Progression and change in treatment strategy The median follow-up time of patients with the AS approach was reported by twelve studies and ranged between 8 months and 73 months (Table 4). Most studies reported the median time to progression (n = 5) 9, 22, 28, 29, 32 , and solely two studies reported median time to shifting from AS to ‘active’ therapy 31, 36 . Other studies used PFS 14, 30, 33, 38 or EFS 6, 28 to express the success rates of the AS approach. Two studies described time to SD 27, 37 . Van Broekhoven et al. described that the median duration of the AS approach was 22 months (IQR: 13-46) for patients with CR or PR 32 . Kim et al. reported that age younger than 40 and a recurrent tumour were significant predictive factors of longer time to disease stabilisation (p = 0.014 and p = 0.036, respectively) 37 . Penel et al. reported that 30.1% of patients in the AS group experienced an event (progression during AS, change in treatment strategy and/or disease-related death) 6 . Briand et al. reported a cumulative probability of dropping out from the AS approach of 5.7% (95% CI 1.5%-14.2%) at one year, and 9.6% (95% CI 3.5%- 19.6%) at 2, 5, and 10 years 8 . Bonvalot et al. stated that the percentage of patients shifting to another treatment was 33% (95% CI 24-43) at 1-year, and 41% at 3 years (95% CI 31%-52%) 10 . Fiore et al. reported that 89% of patients progressed within the first two years after referral, and reported a 5-year PFS rate of 47% (standard error [SE] 10.3%) for primary tumours and 54% (SE 11.6%) for recurrent tumours (p = 0.48) 14 (Table 4). A description of the risk factors for progression or a change in treatment strategy is reported in Table 5. A larger tumour size, >5 cm vs. ≤5 cm, was associated with a shorter time to intervention (6.9 months versus 32.6 months, p = 0.02) 31 , and shift to ‘active’ treatment was more likely in patients with “larger” tumours (≥ 7cm) with a hazard ratio (HR) of 2.0 (95% CI 1.3%- 3.2%, p = 0.002) 36 , and >3.5 cm, p = 0.004 10 . Furthermore, the initiation of ‘active’ treatment was more likely for patients with PD or SD than for patients with PR 6