SUN-070 USP8 genotype is associated with recurrence risk in Cushing’s disease: an international, retrospective, multicenter cohort study

Abstract Disclosure: Q. Zhang: None. Y. Cai: None. A. Abreu: None. M. Buchfelder: None. B. Yao: None. F. Hanzu: None. Y. Liu: None. J. Thorsteinsdottir: None. Z. Ma: None. J. Flitsch: None. S. Zhang: None. H. Ye: None. J. Honegger: None. R.S. Carroll: None. Z. Zhang: None. U.B. Kaiser: None. M. Reincke: None. M. Theodoropoulou: None. Y. Zhao: None. L. Gustavo Perez-Rivas: None. Background: The recurrence rate of Cushing’s disease (CD) after surgery varies between 5-50%. About 30-60% of CD tumors carry somatic USP8 variants. Prior studies on small cohorts have reported opposite associations between USP8 variants and recurrence. Here, we analyzed USP8 genotype and recurrence risk in a large, international cohort of patients with postoperative remission and long follow-up. Methods: Retrospective analysis across seven neuroendocrine centers from China, Europe, and USA. All patients had USP8 genotype, demographic, clinical, biochemical, and follow-up data. Patients with prior pituitary intervention or bilateral adrenalectomy were excluded. We estimated the recurrence risk with Kaplan-Meier curve, log-rank test, and Cox regression. Results: Among 435 patients (82.8% female; median follow-up 5.61 years), USP8 variants were identified in 195 (44.8%) tumors. Variants were more prevalent in younger (p<0.001) and female (p<0.001) patients and were associated with a lower incidence of cavernous sinus invasion (p=0.021) and reduced post-operative cortisol levels (p=0.004). Of these, 371 initially reached post-op remission, but 71 recurred during follow-up. USP8 variants significantly increased recurrence risk in microadenomas (HR 2.36, p=0.025), while in macroadenomas, no significant association was observed (HR 0.58, p=0.126). Stratification by USP8 genotype and tumor size produced four subgroups. However, due to similar recurrence risks (HR 1.08, p=0.824) and comparable clinical characteristics between USP8 variant microadenomas and macroadenomas, these two variant subgroups were merged into a single category, resulting in three distinct categories: wild-type microadenomas, USP8 variant adenomas, and wild-type macroadenomas, with 5-year cumulative recurrence rates of 6.45%, 12.53%, and 31.16%, respectively. The log-rank test revealed significant differences in recurrence risk among these 3 categories (p<0.001). Further evaluation of postoperative 1-week nadir cortisol levels showed category-specific recurrence risks. For wildtype microadenomas, cortisol levels ≥ 2 µg/dL were associated with increased recurrence risk (HR 7.26, p=0.016). In USP8 variant adenomas, recurrence risk was elevated only at cortisol levels ≥ 5 µg/dL (HR 6.11, p<0.001). Postoperative cortisol levels were not significantly associated with recurrence in wildtype macroadenomas. We further identified additional, varied factors associated with recurrence risk across the three categories. Conclusions:USP8 genotype, combined with tumor size, is associated with different risk of recurrence in CD. This study highlights the heterogeneity of CD recurrence patterns and underscores the importance of considering both tumor size and somatic USP8 genotype to guide follow-up strategies. Presentation: Sunday, July 13, 2025