Rethinking the Long‐Term Follow‐Up Strategy for Non‐Functioning Pituitary Neuroendocrine Tumours in a Real‐World Setting

OBJECTIVE: To assess long-term tumour behaviour after surgery for non-functioning pituitary neuroendocrine tumours (NF-PiTNETs) and to propose a risk-adapted postoperative MRI surveillance strategy based on real-world data. METHODS: This single-centre retrospective cohort study included 174 adults with histologically confirmed NF-PiTNETs who underwent first-line pituitary surgery at our tertiary referral centre between 2005 and 2020, with a minimum follow-up of 36 months. Serial MRI was used to assess postoperative residual tumour and recurrence. The primary outcomes were time to progression of residual tumour and time to recurrence after complete resection. Secondary outcomes included management strategies and cumulative MRI burden. Predictors of residual progression and recurrence were evaluated using multivariable Cox proportional hazards models. RESULTS: Residual tumour was present in 107 patients (61.5%) and progressed in 29 (27.1%) after a median of 47 months. Among 67 patients with complete resection, recurrence occurred in 15 (22.4%) after a median of 67 months. Most residual growths (62.1%) occurred within 5 years, while two-thirds of recurrences appeared after 5 years. Residual growth was independently associated with younger age (HR 0.94 per year, 95% CI 0.91-0.97, p < 0.001) and larger tumour size (HR 1.05 per mm, 95% CI 1.01-1.09, p = 0.015), whereas Ki-67 showed a non-significant trend. In the complete resection cohort, Cox regression did not identify robust baseline predictors of recurrence, despite the predominance of late events. Applying a risk-adapted surveillance schedule-MRI at 2, 3, 7, 10 years for complete resection and 1, 2, 3, 5, 7, 10 years for residual tumour-would reduce imaging by 36% without missing any clinically relevant event. CONCLUSIONS: Residual tumour progression and recurrence display distinct temporal patterns. Early imaging is warranted for patients with residual tumour, whereas those with complete resection may safely omit MRI within the first 2 years. A risk-adapted follow-up protocol can reduce unnecessary imaging while maintaining patient safety.