Whole Sella vs Targeted Stereotactic Radiosurgery for Acromegaly: A Multicenter Matched Cohort Study

Low-rank fusion convolutional neural network for prediction of remission after stereotactic radiosurgery in patients with acromegaly: a proof-of-concept study

Artificial intelligence approaches to analyze pathological images (pathomic) for outcome prediction have not been sufficiently considered in the field of pituitary research. A total of 5,504 Hematoxylin & Eosin-stained pathology image tiles from 58 acromegalic patients with a good or poor outcome were integrated with other clinical and genetic information to train a low-rank fusion convolutional neural network (LFCNN). The model was externally validated in 1,536 patches from an external cohort. The primary outcome was the time to the first endocrine remission after SRS. The median time of initial endocrine remission was 43 months [IQR: 13-60 months] after SRS, and the 24-month initial cumulative remission rate was 57.9% [IQR: 46.4-72.3%]. The patient-wise accuracy of the LFCNN model in predicting the primary outcome was 92.9% in the internal test dataset, and the sensitivity and specificity were 87.5% and 100.0%, respectively. The LFCNN model was a strong predictor of initial cumulative remission in the training cohort (HR 9.58, 95% CI 3.89-23.59; p < 0.001) and was higher than that of established prognostic markers. The predictive value of LFCNN model was further validated in an external cohort (HR 9.06, 95% CI 1.14-72.25; p = 0.012). In this proof-of-concept study, clinically and genetically useful prognostic markers were integrated with digital images to predict endocrine outcomes after SRS in patients with active acromegaly. The model considerably outperforms established prognostic markers and can potentially be used by clinicians to improve decision-making regarding adjuvant treatment choices. This article is protected by copyright. All rights reserved.

Stereotactic radiosurgery for acromegaly: an international systematic review and meta-analysis of clinical outcomes

INTRODUCTION

We performed a systematic review and meta-analysis of clinical outcomes for patients with acromegaly treated with stereotactic radiosurgery (SRS).

METHODS

Primary outcomes were 5- and 10-year endocrine remission (ER) and endocrine control (EC). Secondary outcomes were 10-year radiographic local control (LC), visual toxicity, and hypopituitarism rates. Weighted random effects meta-analyses using the DerSimonian and Laird methods were conducted to characterize and compare effect sizes. Mixed effects regression models were used to examine correlations between potential prognostic factors and primary and secondary outcomes.

RESULTS

In total, 1533 patients across 20 published studies with acromegaly treated with SRS were included. At 5-years, estimated ER and EC rates were 43.2% (95% CI 31.7-54.6%) and 55.0% (95% CI 27.6-82.4%), respectively. At 10-years, estimated ER and EC rates were 56.9% (95% CI 47.5-66.4%) and 69.7% (95% CI 47.7-91.8%), respectively. The estimated 10-year LC rate was 92.8% (95% CI 83.0-100%). Visual toxicity and hypopituitarism following SRS were estimated to be 2.7% (95% CI 1.3-4.2%) and 26.8% (95% CI 16.9-36.7%), respectively. Every 1 Gy increase in margin prescription dose beyond 17 Gy was estimated to result in a 0.41% increased risk of visual toxicity (p = 0.03). No prognostic factors were associated with EC, ER, LC, or hypopituitarism.

CONCLUSIONS

SRS was well-tolerated in the management of pituitary acromegaly resulting in gradually improving ER and EC rates over time that approached 60% and 70%. SRS-related visual loss is an uncommon treatment-related side effect, and patient-specific clinical decision making remains critical.