What question did this study set out to answer?

February 19, 2026

Abstract PS4-08-24: The Mechanisms and Predictors of Aromatase Inhibitor-associated MusculoSkeletal Symptoms: The ZAP Trial

Key Points

The aim is to investigate the mechanisms leading to aromatase inhibitor-associated musculoskeletal symptoms (AIMSS) and identify predictive biomarkers.
Conducted a prospective single arm trial among postmenopausal women with breast cancer on letrozole and zoledronic acid.
Measured blood-based biomarkers including cytokines and vitamin D at baseline and 6 months.
AIMSS assessed via Health Assessment Questionnaire-Disability Index and Visual Analog Scale.
Cytokines TNF-beta and MCP4 decreases were significantly linked to the onset of AIMSS.
ROC analysis indicated high predictive values for TNF-beta and several interleukins in identifying women at risk for AIMSS.
Cumulative increases in AIMSS scores correlated with specific cytokine changes, suggesting early interventions may be beneficial.

Abstract

Abstract Background: Patients receiving treatment with aromatase inhibitors (AIs) often suffer from Aromatase Inhibitor-Associated Musculoskeletal Symptoms (AIMSS), defined as myalgias, arthralgias, or joint stiffness. These often lead to discontinuation and could impact breast cancer outcomes. We conducted the prospective single arm Zoledronic Acid Prophylaxis (ZAP) trial, which demonstrated that zoledronic acid administered prior to initiating letrozole and after 6 months was associated with a reduced incidence of AIMSS compared to historical controls (BCRT 2018). Our correlative analysis aims to elucidate the mechanism by which AIs contribute to the development of AIMSS. We hypothesize that alterations in muscle injury, vitamin D deficiency, inflammation, and estrogen deprivation may be associated with AIMSS. Methods: Participants from ZAP were postmenopausal women with stage 0-III breast cancer receiving adjuvant letrozole, who also received zoledronic acid at study entry and 6-months. Blood samples were obtained at baseline and 6 months. Blood-based correlatives included serial assessments of 25-OH vitamin D, CK-MB, a multiplex cytokine array, and estradiol (ultrasensitive). At baseline we also collected whole blood for pharmacogenomic analysis SNPs in a target genes list. AIMSS was assessed using the Health Assessment Questionnaire-Disability Index (HAQ-DI) and Visual Analog Scale (VAS) at 6 months. AIMSS was defined as an increase of 0.22 in a scale of 0-3 in the HAQ-DI and/or an increase of 2.0 cm in a scale of 10 cm in the VAS. We also evaluated other patient reported outcomes (PROs): depression (CESD), anxiety and depression (HADS-A), hot flashes (HFRDI), menopausal symptoms (NSABP-MSQ), sleep quality (PSQI), and quality of life (EuroQol). ROC analysis evaluated the predictive accuracy of biomarker changes for AIMSS. Pearson correlation assessed inter-biomarker correlations. Results: From 2011 to 2013, 59 postmenopausal women enrolled in ZAP. Samples were collected in 59 and 49 participants at baseline and 6 months, respectively. Of the cohort included in this analysis, the median age was 59.0. Among 59 women, 5(8.5%) were Black, 59(100%) were non-Hispanic. The median BMI was 28.5 kg/m2 and baseline arthritis was present in 30 (50.8%). Decreases in TNF-b and MCP4 were significantly associated with cumulative incidences of AIMSS (p=0.023, p=0.03) where those who developed AIMSS had a 16% median decrease in TNF-b, 11% median decrease in MCP4 versus a 13% median increase in TNF-b and 12% median increase in MCP4 in those who did not develop AIMSS. Increase in VAS cumulative incidence was significantly associated with decreases in IL27 (p=0.025), IL21 (p=0.026), and VEGFA (p=0.034). Increase in HAQ was significantly associated with decrease in TNF-b (p=0.01), IL17B (p=0.03), but increase in IL1RA (p=0.043). ROC curve-based analyses show that TNF-b exhibits high negative predictive values (NPVs) in classifying AIMSS (0.824; 95%CI 0.566, 0.962), HAQ (1; 95%CI 0.824, 1), and VAS (0.824; 95%CI 0.566, 0.962). IL1RA (0.97; 95%CI 0.82,1) and IL17B (0.93; 95%CI 0.79, 0.98) show high NPV for HAQ classification. IL27 (0.815; 95%CI 0.619, 0.937), IL21 (0.821; 95%CI 0.631, 0.939), VEGFA (0.8; 95%CI 0.593, 0.932), and MCP4 (0.81; 95%CI 0.581, 0.946) achieve similarly good discriminatory ability for VAS classification. Other cytokines and vitamin D levels were not significantly associated with AIMSS. We will report pharmacogenomic associations of genes and estradiol with AIMSS, and biomarker associations with PROs at the conference. Conclusions: Decreases in TNF-beta and MCP4 were associated with the reporting of AIMSS. Early changes in cytokines profile may identify those at greater risk for AIMSS and who may derive greater benefit from earlier symptom management to prevent discontinuation. Citation Format: J. Sheng, R. Chen, J. Zacharski, D. Tsang, A. Yende, H. Qi, v. stearns, D. sharma, C. Santa-Maria. The Mechanisms and Predictors of Aromatase Inhibitor-associated MusculoSkeletal Symptoms: The ZAP Trial abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-08-24.

Mark Helpful

Bookmark

Relay