Blood pressure and heart risk emerge as multiple myeloma clues

A review of 11 studies involving about 130,000 patients with multiple myeloma has identified hypertension, older age, pre-existing cardiovascular disease and exposure to some proteasome-inhibitor therapies as potential warning signs of cardiovascular complications during cancer care. The most clearly quantifiable signal was an association between hypertension and cardiovascular adverse events: pooling two observational studies produced a hazard ratio of 1.90, meaning that patients with high blood pressure had an estimated 90% higher event risk than those without hypertension. The evidence nevertheless remains preliminary because the review found only one factor suitable for quantitative synthesis and rated the certainty of that estimate as low.

The systematic review and exploratory meta-analysis, reported in Frontiers in Cardiovascular Medicine, examined multivariable-adjusted predictors of cardiovascular events in adults with multiple myeloma. Researchers searched PubMed, Embase, Web of Science, the Cochrane Library and Ovid MEDLINE from their inception through August 2025, without language restrictions during database searching, although only English-language full texts were considered for eligibility. The protocol followed PRISMA 2020 guidance and was registered in PROSPERO as CRD420251084727. Studies involving cardiac amyloidosis were excluded because the review treated that condition as a distinct, high-risk clinical entity.

Eleven eligible studies were identified from 4,786 records: nine were retrospective cohorts, one was a prospective observational study and one was a retrospective case-control comparison. Five analyses used population-based administrative or cancer-registry data, while six were hospital-based investigations. The studies were conducted in the United States, China, South Korea and Germany, and their sample sizes ranged from 47 to 88,328. Mean or median ages ranged from about 56 to 72 years, reported cardiovascular event counts ranged from 23 to 1,892, and event incidence ranged from 8.3% to 37.6%. Some registry populations may have overlapped, meaning the approximate total of 130,000 patients should not be interpreted as 130,000 fully independent individuals.

Definition of the outcome was one of the review’s main sources of inconsistency. Six studies focused on composite cardiovascular adverse events, three assessed cardiovascular mortality and two examined particular cardiovascular complications. Composite endpoints variously included heart failure, hypertension, ischemic heart disease, myocardial infarction, arrhythmia, pulmonary hypertension, venous thromboembolism, dyspnea, edema, syncope and chest pain. Some events were identified through systematic echocardiography, electrocardiography or cardiac-biomarker testing, whereas others came from claims data, clinical records or standardized diagnostic codes. These differences limited comparability and prevented a broad pooled analysis of all candidate predictors.

Hypertension was the only predictor that met the review’s prespecified requirements for exploratory meta-analysis. Two studies with sufficiently comparable adjusted hazard ratios for incident composite cardiovascular adverse events were combined in a random-effects model. Their pooled hazard ratio was 1.90, with a 95% confidence interval of 1.57 to 2.29; statistical heterogeneity was not detected, with an I-squared value of 0% and a heterogeneity P value of 0.83. A separate hospital cohort reported a consistent direction of association, with an adjusted odds ratio of 2.50 and a 95% confidence interval of 1.10 to 5.90, but it was not pooled because it used an odds ratio rather than a hazard ratio.

Several conventional demographic and clinical signals recurred, although differences in populations, endpoints and statistical methods prevented formal pooling. Male sex was associated with higher cardiovascular mortality in two population-based studies, while older age was linked to cardiovascular events in four analyses. Smoking history was associated with adverse outcomes in two studies: one reported an adjusted odds ratio of 2.80 for composite events, and another reported a hazard ratio of 2.13 for mortality. Low body mass index and obesity were each associated with increased cardiovascular risk in studies that examined both measures. A history of cardiovascular disease pointed toward higher event risk in all four studies that assessed it, reinforcing the distinction between pre-existing vulnerability and treatment-emergent complications.

Clinical measurements supplied some of the largest reported associations, although their precision and interpretation varied. Peripheral arterial occlusive disease was associated with cardiovascular events at a hazard ratio of 10.4, but its wide 95% confidence interval of 1.23 to 89.1 indicates substantial uncertainty. Chronic obstructive pulmonary disease had a smaller reported association, with an adjusted hazard ratio of 1.40 and a 95% confidence interval of 1.03 to 1.90. More advanced disease, represented by Durie–Salmon stage B, was associated with events at a hazard ratio of 2.28 and a 95% confidence interval of 1.30 to 4.00. These estimates came from individual observational studies and should not be treated as interchangeable measures of the same outcome.

Treatment-related findings centered on proteasome inhibitors. In one retrospective comparison of 96 patients receiving bortezomib- or carfilzomib-based therapy, a longer duration of proteasome-inhibitor treatment was associated with higher odds of cardiovascular events, yielding an adjusted odds ratio of 1.94 and a 95% confidence interval of 1.21 to 3.09. A separate prospective observational study of 95 patients with relapsed disease reported a higher cardiovascular-event hazard with carfilzomib-based therapy than with bortezomib-based therapy, with a hazard ratio of 3.00 and a 95% confidence interval of 1.10 to 8.40. These were not pooled because the studies differed in design, treatment context, endpoints and effect measures.

The most striking biomarker signal also came from a single study and has not been independently replicated. Among 95 patients in the prospective multicentre analysis, elevated baseline B-type or N-terminal pro-B-type natriuretic peptide levels were associated with subsequent cardiac events at a hazard ratio of 4.10 and a 95% confidence interval of 2.10 to 8.10. A marked rise in these peptides during the first treatment cycle had an even larger reported association, with a hazard ratio of 9.50 and a 95% confidence interval of 4.30 to 20.70. The result suggests that natriuretic peptides may carry more prognostic information when measured during treatment than at baseline, but the review could not establish monitoring intervals, actionable thresholds or a routine testing strategy.

Methodological limitations constrain how confidently any of these associations can be interpreted. Most studies were judged to have a moderate risk of bias, with recurring problems involving retrospective design, inconsistent cardiovascular definitions and incomplete adjustment for confounders such as baseline cardiovascular disease, renal function, disease stage and cumulative treatment exposure. Several reports also gave unclear accounts of how missing data were handled. Administrative datasets may not capture delivered doses, adherence, outpatient events or subclinical cardiac injury, while small single-center cohorts are difficult to generalize broadly. Publication bias could not be assessed reliably because fewer than 10 studies contributed to a meta-analysis. The findings therefore identify candidate prognostic signals rather than a validated cardiovascular risk score for individual patients.

Overall, the review portrays multiple-myeloma cardiovascular risk as a combination of baseline susceptibility, disease-related effects, treatment exposure and potentially changing markers of cardiac stress. Hypertension is the strongest meta-analytic signal in the available evidence, but age, smoking, prior cardiovascular disease, body size, treatment duration, carfilzomib exposure and dynamic natriuretic peptides may also matter. The authors call for prospective, multicenter studies with standardized cardiovascular endpoints, consistent adjustment for confounding and external validation of prediction models. Such work would be needed before the signals could support individualized monitoring, treatment modification or other clinical decisions. For now, the results sharpen the case for studying cardiovascular risk throughout repeated myeloma therapy rather than treating it as an isolated consequence of a single drug or a one-time assessment at diagnosis.

Subject of Research: Systematic review of cardiovascular-event predictors in adults with multiple myeloma

Article Title: Predictors of cardiovascular events in multiple myeloma: a systematic review and exploratory meta-analysis

Article References: Lv, T., Liu, M., Zhang, W., Zhang, J., & Fang, Y. (2026). Predictors of cardiovascular events in multiple myeloma: a systematic review and exploratory meta-analysis. Frontiers in Cardiovascular Medicine, 13, Article 1925889. https://doi.org/10.3389/fcvm.2026.1925889

Image Credits: AI Generated

DOI: 10.3389/fcvm.2026.1925889

Keywords: multiple myeloma, cardiovascular risk, hypertension, cardio-oncology, carfilzomib, proteasome inhibitors, natriuretic peptides, systematic review, meta-analysis, Predictors, cardiovascular, events