Background: Between 4%- 30% of patients with clinical stage 1 (CS I) seminoma testis (ST) in active surveillance (AS) relapse depending on pathological risk factors tumor size (TS) and rete testis invasion (RTI). The level of evidence supporting these pathological risk factors in clinical decision-making is low due to heterogeneous study design and reporting and the difficulty to analyze patient subgroups according to combination of these factors. The objectives of the present study were to identify the most important pathological prognostic factors predicting relapse in CS I seminoma patients with normal post-orchidectomy serum tumor marker (STM) levels in AS and to construct risk-groups for guiding treatment decision-making and follow-up. Methods: Individual patient data from 1016 CS I-ST patients diagnosed between February 1994 and January 2019 in AS were collected from 9 institutions. Central pathology review was not routinely performed in all institutions; therefore, pagetoid and stromal RTI were not differentiated explicitly in most cases. Assessing patient age, pre-orchidectomy STM b-human chorionic gonadotropin and lactate dehydrogenase, pathological TS, RTI, lympho-vascular invasion (LVI), multi-focality, and GCNIS, multivariable Cox proportional hazards regression models were fit to identify the most important prognostic factors for the time to first relapse by imaging and/or markers (primary endpoint). Probabilities of relapse were estimated using Kaplan-Meier curves. Results: After median follow-up of 7.7 years, 149 (14.7%) patients relapsed, 104 identified by imaging alone, 44 by imaging with elevated STMs and 1 by elevated STMs alone. Excluding 18 patients with unknown LVI from the multivariable analyses, TS (≤ 2 cm, between 2 and 5 cm, > 5 cm), presence of RTI and presence of LVI were used to form three risk groups: very low, low and high-risk (Table). Five-years probability of relapse varied from 8% in the very low risk-group to 44% in the high risk-group. The new model outperformed the current model with TS < 4cm vs ≥ 4 and RTI (Harrell's C index 0.65 vs 0.61) and identifies a subgroup of patients with a higher risk of relapse. Conclusions: The new risk-group stratification for CS I -ST patients in AS outperforms the histopathological model based on TS and RTI. It will serve to better inform patients on the risk of relapse during follow-up after orchidectomy but requires an independent validation.