The impact of insulin-like growth factor index and biologically effective dose on outcomes after stereotactic radiosurgery for acromegaly: Cohort study

BACKGROUND: Stereotactic radiosurgery (SRS) is a safe and effective treatment for acromegaly. OBJECTIVE: To improve understanding of clinical and dosimetric factors predicting biochemical remission. METHODS: A single-institution cohort study of nonsyndromic, radiation-naïve patients with growth hormone-producing pituitary adenomas (GHA) having single-fraction SRS between 1990 and 2017. Exclusions were treatment with pituitary suppressive medications at the time of SRS, or <24 mo of follow-up. The primary outcome was biochemical remission-defined as normalization of insulin-like growth factor-1 index (IGF-1_i) off suppression. Biochemical remission was assessed using Cox proportional hazards. Prior studies reporting IGF-1_i were assessed via systematic literature review and meta-analysis using random-effect modeling. RESULTS: A total of 102 patients met study criteria. Of these, 46 patients (45%) were female. The median age was 49 yr (interquartile range [IQR] = 37-59), and the median follow-up was 63 mo (IQR = 29-100). The median pre-SRS IGF-1_i was 1.66 (IQR = 1.37-3.22). The median margin dose was 25 Gy (IQR = 21-25); the median estimated biologically effective dose (BED) was 169.49 Gy (IQR = 124.95-196.00). Biochemical remission was achieved in 58 patients (57%), whereas 22 patients (22%) had medication-controlled disease. Pre-SRS IGF-1_i ≥ 2.25 was the strongest predictor of treatment failure, with an unadjusted hazard ratio (HR) of 0.51 (95% CI = 0.26-0.91, P =.02). Number of isocenters, margin dose, and BED predicted remission on univariate analysis, but after adjusting for sex and baseline IGF-1_i, only BED remained significant-and was independently associated with outcome in continuous (HR = 1.01, 95% CI = 1.00-1.01, P =.02) and binary models (HR = 2.27, 95% CI = 1.39-5.22, P =.002). A total of 24 patients (29%) developed new post-SRS hypopituitarism. Pooled HR for biochemical remission given subthreshold IGF-1_i was 2.25 (95% CI = 1.33-3.16, P <.0001). CONCLUSION: IGF-1_i is a reliable predictor of biochemical remission after SRS. BED appears to predict biochemical outcome more reliably than radiation dose, but confirmatory study is needed.