Background/Purpose: Asian women accounted for more than 40% of incident breast cancer (BC) worldwide. While aromatase inhibitors (AIs) significantly reduce recurrence of breast cancer (BC) and improve disease-free survival in post-menopausal women with hormone receptor-presenting BC, over one-third of AI users in randomized controlled trials developed arthralgia. Little is known about the risk of AI-induced arthralgia among Asian women with BC in real-world settings. Our objective was to compare the risk of AI-induced arthritis/musculoskeletal problems (A/MSK) between use of AIs vs. tamoxifen in a large Taiwan administrative claims dataset having >99% national coverage and ~23 million enrollees. Methods: In a retrospective cohort study using Taiwan Catastrophic Illness Patient Datasets and Taiwan National Health Insurance Research Datasets, we identified BC women (ICD-9 code: 174) that newly initiated AIs (anatrazole, exemestane and letrozole) or tamoxifen (TAM) from 2007 to 2012. We examined three A/MSK -related outcomes of interest within a year after therapy initiation, including women having 1) prescription pain relievers (i.e., non-steroidal anti-inflammatory drugs [NSAIDs], acetaminophen and opioids), 2) diagnosis of any type of arthritis (ICD-9: 710-719), or 3) carpal tunnel syndrome (CTS) or trigger fingers (TF; ICD-9: 354.0, 727.03). Individuals having outcome episodes that occurred within 3 months prior to AI/TAM initiation were excluded. We applied Greedy methods to match AI users with up to 5 TAM users on age and prior use of taxane therapy. To ensure the robustness of our results, we estimated the association between A/MSK risk and AI use compared to TAM use using conditional logistic regression, cause-specific Cox regression and competing risk regression models. Results: Among 40,761 BC women initiating endocrine therapy from 2007 to 2012, 15.5% were AI users and 84.5% were TAM users. The mean age was 54.1 (SD±12.1) years and 19.6% had history of taxane exposure prior to AI/TAM initiation. There were significant differences (p <0.0001) in the crude rates of pain reliever use (68.7% vs. 65.0%), diagnosed arthritis (16.8% vs. 11.7%) and CTS/TF (2.6% vs. 1.1%) within a year after AIs or TAM initiation. After matching, AI users had higher odds of receiving pain relievers (OR: 1.13, 95%CI: 1.02-1.24), having a diagnosis of arthritis (OR: 1.09, 95%CI: 1.00-1.18) and a diagnosis of CTS/TF (OR: 2.19, 95%CI: 1.76-2.74) compared to TAM users (Table 1). Similar findings were seen using cause-specific Cox regression and competing risk regression (Table 1). Conclusion: We found that AI users have a higher risk of A/MSK within the first year of treatment compared with TAM users among Taiwanese BC women. Timely pain management and close follow-up for AI-induced A/MSK is suggested to prevent poor adherence or discontinuation of AI therapy.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/use-of-aromatase-inhibitors-and-risk-of-arthritis-and-musculoskeletal-problems-among-taiwanese-women-with-breast-cancer-a-nationwide-claims-analysis/