AUTHORS
Maria Daniel Santiago Estevez, M.D, Richard C. Godby, MD2, Nelson Leung, MD3, Ronald Go, MD4, Thomas M. Habermann, MD5, Yoshito Nishimura, M.D., PhD, MPH6
AFFILIATIONS
1 Mayo Clinic, Rochester, Minnesota, United States
2 Mayo Clinic, Rochester, Minnesota, United States
3 Mayo Clinic, Rochester, Minnesota, United States
4 Mayo Clinic, Rochester, Minnesota, United States
5 Mayo Clinic, Rochester, Minnesota, United States
6 Mayo Clinic, Rochester, Minnesota, United States
ABSTRACT
Introduction: Mechanistic target of rapamycin (mTOR) inhibitors (mTORi), including sirolimus, everolimus, and temsirolimus, are used in various areas of practice such as hematology, solid-organ transplantation, and for immune or inflammatory disorders. Thrombotic complications reported with these agents include thrombotic microangiopathy (TMA), venous thromboembolism/pulmonary embolism (VTE/PE), and arterial thrombosis, which may significantly impact patient outcomes. Previous pharmacovigilance studies have evaluated adverse events associated with individual mTORi or combined clinically different thrombotic events into a single outcome. In clinical practice, mTORi are also frequently used with calcineurin inhibitors (CNIs), both of which may cause vascular endothelial dysfunction. This study aimed to evaluate reporting of thrombotic adverse events associated with systemic mTORi, compared with reports without mTORi, and to determine whether reporting was further increased with concomitant use of mTORi and CNIs.
Methods: A total of 15,901,334 deduplicated reports from the FDA Adverse Event Reporting System (FAERS), including 63,958 with transplant, were analyzed from 2014 Q3 through 2026 Q1. Systemic mTORi exposure included sirolimus, everolimus, temsirolimus, and nab-sirolimus, whereas topical formulations and drug-eluting stents were excluded from the analysis. Outcomes included individual reports of TMA, VTE/PE, arterial thrombotic or ischemic events, and a secondary composite thrombosis outcome. The primary analysis compared reports listing an mTORi as a primary or secondary suspect with reports not listing an mTORi using reporting odds ratios (RORs) with 95% confidence intervals (CIs). Data regarding timing of the thrombotic event was not available. Reports were classified into four mutually exclusive groups: mTORi plus CNI, mTORi without CNI, CNI without mTORi, or neither drug class. Excess joint reporting with both drug classes was assessed using Ω shrinkage, which compares the observed number of reports listing both drug classes and the event to the number expected from the individual reporting of each drug class. An Ω025, representing the lower 95% credibility limit, > 0 was considered positive for excess joint reporting.
Results: Among 15,901,334 reports, 53,215 listed an mTORi as a primary or secondary suspect, 8,282 listed both mTORi and CNI, 44,933 mTORi without CNI, and 160,016 CNI without mTORi. Compared with those without mTORi, mTORi reports had higher reporting of TMA (ROR 15.0, 95% CI 13.8-16.4), VTE/PE (ROR 1.7, 95% CI 1.5-1.8), and the combined thrombosis outcome (ROR 2.6, 95% CI 2.4-2.7), but not arterial thrombotic or ischemic events (ROR 1.1, 95% CI 0.89-1.3). Compared with reports having neither mTORi nor CNI, TMA RORs were 4.4 (95% CI 3.6-5.3) for mTORi without CNI, 34.3 (95% CI 32.9-35.9) for CNI without mTORi, and 109.4 (95% CI 99.1-120.7) for mTORi plus CNI, with 442 combination reports versus 158 expected (Ω 1.5; Ω025 1.3). Corresponding VTE/PE RORs were 1.6 (95% CI 1.5-1.8), 1.1 (95% CI 1.0-1.2), and 2.0 (95% CI 1.6-2.4), with 95.0 combination reports versus 82.8 expected (Ω 0.20; Ω025 -0.093). Corresponding arterial event RORs were 0.67 (95% CI 0.53-0.86), 1.6 (95% CI 1.4-1.7), and 3.3 (95% CI 2.5-4.2), with 58.0 combination reports versus 27.9 expected (Ω 1.0; Ω025 0.67).
Conclusions: Systemic mTORi were associated with disproportionate reporting of TMA and VTE/PE, whereas arterial thrombotic or ischemic events were not increased overall. Reporting with combined mTORi and CNI exceeded that expected from the individual drug classes for TMA and arterial events, but not for VTE/PE. The results suggest that clinicians should monitor for thrombotic events while using mTORi, particularly in combination with CNIs. It is worth noticing that the data was obtained from the FAERS database which is a spontaneous reporting system with varying reporting practices by regions, and cannot determine incidence or causality. Further clinical cohort studies are needed to confirm the association, extent of thrombotic events, and whether prophylactic anticoagulation or antiplatelet therapies may mitigate thrombotic risk in patients treated with mTORi, with or without CNIs.
DISCLOSURES
Maria Daniel Santiago Estevez, M.D: No relevant conflicts of interest to declare.