Profiling the Risk for Cardiovascular Adverse Events in Patients with Relapsed/Refractory Multiple Myeloma Treated with Carfilzomib Using Topological Analysis

Author(s)

Quach H1, Ludwig H2, Chari A3, Richter J4, Goldrick A5, Yusuf A5, Abbasi S5, Gnacadja G5, Mikhael J6
1University of Melbourne, St Vincent’s Hospital, Melbourne, VIC, Australia, 2Wilhelminen Cancer Research Institute, Wilhelminenspital, Vienna, CA, Austria, 3Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York City, NY, USA, 4Tisch Cancer Institute at Mount Sinai Medical Center, New York City, NY, USA, 5Amgen Inc., Thousand Oaks, CA, USA, 6Translational Genomics Research Institute, City of Hope Cancer Center, Phoenix, AZ, USA

OBJECTIVES : Patients with multiple myeloma (MM) often have cardiovascular risk factors and/or established cardiovascular disease. Carfilzomib is approved for managing relapsed/refractory (RR) MM and may be associated with increased risk of cardiovascular adverse events (AEs). We leveraged machine learning to identify risk factors for developing cardiovascular AEs in patients with RRMM treated with carfilzomib.

METHODS : Retrospective data were pooled from carfilzomib phase 3 trials (ASPIRE, ARROW, FOCUS, ENDEAVOR). Variables analyzed included baseline patient demographics, vital signs, electrocardiogram results, laboratory values, and comorbidities. Topological data analysis (an analytic approach akin to dimensionality reduction and distinct from traditional multivariate modeling analysis) was used to reveal AE occurrence micropatterns. A decision algorithm accounting for both posterior and prior AE probability with respect to the variables was used to find clinically interpretable risk factors.

RESULTS : In the pooled population (n=1,484), grade ≥3 cardiovascular AEs of interest were cardiac failure (5.1%), ischemic heart disease (2.2%), and hypertension (9.2%). The modeling approach successfully grouped patients into high-risk %/low-risk % groups (cardiac failure: 31.1%/68.9%; ischemic heart disease: 38.5%/61.5%; hypertension: 32.1%/67.9%), with AE rates enriched in the high-risk group (cardiac failure AE rate was 4x higher in the high-risk than low-risk group [10.8% vs 2.4%]). Risk factors for grade ≥3 cardiac failure were baseline comorbidity of congestive heart failure, moderate-to-severe chronic kidney disease, mean QRS ≥120 msec, glucose ≥7.9 mmol/L, and Asian race; for grade ≥3 ischemic heart disease, risk factors were comorbidity of prior myocardial infarction, glucose ≥7.9 mmol/L, activated partial thromboplastin time ≤20 seconds and creatinine clearance ≤60 mL/min; for grade ≥3 hypertension, risk factors were systolic blood pressure ≥140 mmHg and Asian or African race.

CONCLUSIONS : Baseline risk factors identified for cardiovascular AEs were detectable by standard clinical assessment or laboratory tests. These findings may provide direction for future research into cardiovascular AE risk factors in carfilzomib-treated patients.

Conference/Value in Health Info

2021-11, ISPOR Europe 2021, Copenhagen, Denmark

Value in Health, Volume 24, Issue 12, S2 (December 2021)

Code

POSB15

Topic

Methodological & Statistical Research

Topic Subcategory

Artificial Intelligence, Machine Learning, Predictive Analytics

Disease

Oncology

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×