Oral Presentation ESA-SRB-NZSE-CaSR 2026 in conjunction with ENSA

Pituitary tumour classification and risk stratification: predicting clinical outcomes in acromegaly (143780)

Prishila Fookeerah 1 2 , Winny Varikatt 3 4 , Sue Lynn Lau 1 2 , Mark Mclean 1 2
  1. Diabetes and Endocrinology, Westmead Hospital, Sydney, NSW, Australia
  2. School of Medicine, Western Sydney University, Sydney, NSW, Australia
  3. Department of Tissue Pathology and Diagnostic Oncology, Westmead Hospital, Sydney, NSW, Australia
  4. Westmead Clinical School, University of Sydney, Sydney, NSW, Australia

Pituitary tumours causing acromegaly are morphologically diverse and can be categorised into seven histological subtypes according to the 2022 WHO classification. Tumour subtyping may assist with choice of therapy for individual patients.  Sparsely granulated somatotroph and immature PIT1-lineage tumour subtypes have been designated as “high-risk” but there is limited clinical data comparing tumour variants associated with acromegaly. We assessed the clinical outcomes of high and low risk tumour subtypes to determine if risk status is useful in prognostication.

In a retrospective observational study of 245 pituitary tumours histologically classified using transcription factor immunohistochemistry, 51 tumours belonging to subtypes that can cause acromegaly were identified. Clinical and histological variables were compared for high and low-risk subtypes. Cox proportional hazards regression was used to evaluate predictors of clinical outcome.

There were 24 high-risk and 27 low-risk tumours. High-risk subtypes were larger (23.0mm vs 18.0 mm, p=0.002), were associated with lower IGF-1 (1.9 vs 2.9 x ULN, p=0.008) and less frequently exhibited clinical features of acromegaly (41.7% vs 88.9%, p<0.001) at presentation. There was no significant difference in proliferative markers [Ki-67>3% (29.2% vs 29.6%, p=0.971); mitotic count >2 hpf (17.4% vs 3.8%, p=0.17)]. Although SSTR2 expression was strong in both groups [immunoreactivity score (IRS) 9 vs 12, p=0.11], SSTR5 IRS was greater in high-risk tumours (8 vs 4, p=0.02). High-risk subtypes were more likely to require further intervention with repeat surgery and/or radiotherapy after primary resection (HR 3.3, 95% CI 1.0-10.4, p=0.04), with clinical trajectories diverging after the first 12 months. Over a median follow-up duration of 68 (47-106) months, tumour recurrence/progression only occurred in high-risk subtypes (23.8% vs 0%, p=0.01) with time to detection ranging from 5-89 months.

Risk based assessment using histological subtype may predict recurrence and future therapeutic intervention in tumours that are associated with acromegaly.