OBJECTIVES: Currently, there is no specific characterization of acromegaly control from the patient’s perspective. This analysis utilizes principal component analysis and latent class analysis to define a dichotomous patient-centered acromegaly control (PCAC: well controlled versus poorly controlled).

METHODS: The statistical analyses utilized data from an acromegaly study (NCT# 03613623) utilizing the ACRO-TSQ tool, a validated patient reported outcome (PRO) regarding injectable acromegaly treatment, and additional clinical measures such as insulin grown hormone (IGF)-1 value which determines biochemical control. A PCA was conducted to determine the best combination of components across survey responses and IGF-1 score to conduct LCA. Out of 9 PCA component sets, 2 with IGF-1 score were selected and compared via LCA across two potential PCAC measures.

RESULTS: Model 1 LCA coefficients showed that IGF-1 levels are not significantly specific towards control level between groups (5.32, 95% Confidence Interval (CI) [5.09, 5.56] vs 5.12, 95%CI [5.09, 5.56]). However, treatment convenience (0.49, 95%CI [0.29, 0.71] vs 1.34, 95%CI [1.20, 1.47]), how treatment affected plans (0.34, 95%CI [0.14, 0.53] vs 1.19, 95%CI [1.08, 1.30]), and patient satisfaction (0.63, 95%CI [0.41, 0.84] vs 1.35, 95%CI [1.23, 1.47] were significant indicators. Model 2 LCA showed that IGF-1 levels (5.42, 95%CI [5.06, 5.79] vs 5.36, 95%CI [5.13, 5.58]), patient satisfaction (1.33, 95%CI [1.11, 1.55] vs 1.14, 95%CI [1.01, 1.27]), and headache (0.55, 95%CI [0.27, 0.83] vs 0.33, 95%CI [0.15, 0.51]) were not significantly specific towards control between groups. However, acute injection site reactions (1.33, 95%CI [1.25, 1.41] vs 0.39, 95%CI [0.27, 0.51]) and how bothered the patient was by the reactions (1.27, 95%CI [1.16, 1.37] vs 0.57, 95%CI [0.46, 0.70]) were significant indicators.

CONCLUSIONS: Overall, IGF-1 value does not appear to be a differentiating indicator for PCAC. Patient convenience and injection site reaction related measures however indicate significant differences through LCA.