Introduction
Anthracycline antibiotic is one of the most effective anti-tumor drugs used to manage certain types of breast cancers, lymphomas, and leukemias [
1]. However, anthracyclines induce a dose-dependent cardiotoxicity that may progress to heart failure [
2]. In addition, in patients with depressed heart function at baseline, lower dosages of anthracyclines are used in full awareness that antineoplastic activity is dose dependent [
3]. Furthermore, clinical monitoring of left ventricular (LV) function during chemotherapy has little proven value since many patients eventually progress to heart failure from global depression of cardiac ejection fraction after chemotherapy has been halted or completed [
4]. Interestingly, while the incidence of cancer has increased significantly worldwide over the past decade, the rate of survival has increased due to improvements in diagnoses and therapies. However, the number of patients at risk for anthracycline cardiotoxicity is obviously increasing [
5,
6].
Equally troubling is the fact that anthracycline-mediated decline in LV systolic function does not completely normalize when conventional heart failure therapy is initiated 4 months after an initial decline in left ventricular ejection fraction (LVEF) in patients with or without heart failure symptoms [
7]. Additionally, LVEF does not completely recover in one-third of patients when heart failure therapy was initiated within 1–2 months after cardiac insult [
7]. Thus, there is an obvious clinical need to detect anthracycline cardiotoxicity before the initial decline in LVEF [
8]. Indeed, using a sensitive predictor of early cardiac dysfunction in patients treated with anthracyclines can help in the early detection of subclinical cardiac dysfunction and help in initiating interventions to protect these patients [
8]. Given the cardiotoxicity in patients treated with anthracycline, protecting the heart from this side-effect would improve the risk-to-benefit ratio of anthracycline chemotherapy and lead to better quality and duration of post-chemotherapy life.
Among parameters of myocardial measure, cardiac magnetic resonance (CMR)-measured native myocardial T1 mapping is considered a sensitive and accurate quantitative measure of early subclinical cardiac changes [
9,
10]. Unlike other myocardial measures, native myocardial T1 mapping can detect the changes at the myocardial tissue level likely to precede functional changes [
10]. However, to understand the quality and the validity of the current evidence supporting the use of these measures in patients treated with anthracyclines, it is imperative to perform a systematic review and meta-analysis of studies investigating CMR-measured native myocardial T1 mapping as a predictor of anthracycline-related early subclinical cardiac dysfunction. Thus, we aim to synthesize results from clinical studies in an attempt to provide a conclusive answer on the validity of these measures to detect early myocardial changes in cancer patients treated with anthracyclines.
Methods
Search Strategy
This study followed the Cochrane Review Protocol (
https://training.cochrane.org/handbook). In addition, we registered our study in PROSPERO—International Prospective Register of Systematic Reviews (CRD42022331267). Our study is based on a systematic review of the clinical trials on the role of native myocardial T1 mapping in the early detection of anthracycline-induced cardiotoxicity in patients with cancer. The systematic review of this project was performed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol. The following databases were used to extract relevant data: PubMed, Embase, Web of Sciences, and Scopus. The search terms were created according to the population, intervention, comparison, and outcome (PICO) question: “
What is the value of native myocardial T1 mapping for the early detection of anthracycline-induced cardiotoxicity in patients with cancer?”. The key search term for the population was according to the domain “cancer,” whereas for the intervention was “anthracyclines.” The comparator terms include “myocardial T1 mapping,” and finally, the outcomes were mainly based on the results from the clinical studies (Supplementary Table 1) [
11‐
19].
Selection Criteria
All studies were selected according to our inclusion criteria which include the articles in English language, peer-reviewed articles and reporting clinical studies on myocardial T1 mapping after exposure to any cumulative dose of anthracyclines used in adult cancer patients and compared the outcome with that of a healthy control group who were not treated with anthracyclines or baseline till November 2022. Our exclusion criteria include clinical studies where patients exposed to multiple cardiotoxic pharmacotherapies or less than 50% of patients’ cohort had anthracycline, if there is no baseline or healthy control, case reports, conference abstracts, duplicate studies, or reviews as well as non-clinical studies, such as animal studies and lab studies. The authors independently evaluated the studies to ensure that they matched the inclusion criteria. In cases of discrepancy, general agreement was attained by discussion between the authors.
Data Collection
The authors uploaded search results from PubMed, Embase, Web of Sciences, and Scopus to the Rayyan software program (
https://www.rayyan.ai/). The Rayyan software helped remove duplications and manually screened searched abstracts and articles. The data extracted from the selected clinical studies includes information about the number of patients, year and country, study population, sex, age, study design, intervention and outcome. The PRISMA protocol was used to analyze the data qualitatively and descriptively.
Quality Assessment
The quality assessment of our collected reports was performed by all authors independently and any disagreement between them was resolved by consensus. The Joanna Briggs Institute (JBI) critical appraisal checklist was used to assess the quality of the studies.
Statistical Analysis
Publication bias was assessed with visual inspection of funnel plot analysis, and Egger’s test was used to determine the funnel plot asymmetry. The meta-analysis was done using jamovi version 2.4.11 (Edelsbrunner, Peter (2017-03-23). The data were presented using the forest plot. Also, the data were pooled to calculate the standardized mean difference with a 95% confidence interval using a fixed-effects model. In our study, we utilized a standardized mean difference as our meta-analysis assessed native T1 mapping in various studies that have been measured differently. Standardized mean difference is used to measure the effect size of the study. Also, it is important to standardize the results of the studies to a uniform scale before they can be combined in this meta-analysis. A p value < 0.05 is considered significant. Heterogeneity was assessed using the I2 parameter by Chi-squared test (high heterogeneity if I2 is ≥ 50%). Leave-one-out sensitivity analysis was also performed to assess the level of heterogeneity.
Discussion
Anthracyclines are very effective anti-cancer agents used in the treatment of several types of cancer, including breast cancer, leukemias, and lymphomas [
1]. However, anthracycline therapy triggers cardiotoxicity that may progress to cardiac failure later in life [
2]. Thus, using sensitive measures of cardiotoxicity in patients treated with anthracyclines can help in the early detection of cardiac dysfunction and help in initiating interventions to protect these patients [
8]. Of those sensitive tools, CMR-measured native myocardial T1 mapping is considered a sensitive and precise quantitative measure of early subclinical cardiac changes, particularly cardiac inflammation and fibrosis [
21]. However, the quality and validity of the current evidence supporting the use of these measures in patients treated with anthracyclines is still unclear. Thus, we performed a systematic review and meta-analysis to synthesize results from clinical studies to provide a conclusive answer on the validity of this measure as an early detector of anthracycline-induced myocardial changes. To our knowledge, this study is the first systematic review and meta-analysis to address the quality and the validity of native myocardial T1 mapping in the early detection of anthracycline-induced cardiotoxicity.
In the present study, our systematic review and meta-analysis showed a significant increase in the native myocardial T1 mapping in cancer patients treated with anthracyclines from baseline as well as compared to healthy controls. We also show that this increase in the native myocardial T1 mapping was associated with low heterogeneity among the included studies, suggesting that the outcome is consistent. Given that the LVEF in cancer patients treated with anthracyclines was within the normal range, the native myocardial T1 mapping can likely detect anthracycline-induced myocardial changes before any functional abnormalities. In support of this, recent clinical and preclinical findings demonstrated that native myocardial T1 mapping is a valuable tool for detecting chemotherapy-induced cardiotoxicity early [
16,
17]. Thus, our systematic review and meta-analysis suggest that the native myocardial T1 mapping may have a significant clinical utility for the early detection of anthracyclines-induced cardiotoxicity.
In addition to the native T1 mapping, cardiac global longitudinal strain (GLS) is an echocardiographic parameter that can also detect early changes in cardiac function by measuring the longitudinal myocardial deformation [
22]. However, GLS values are usually influenced by technical factors, such as image acquisition as well as patient factors, like sex and age [
22]. For instance, GLS values are lower in females compared to males and decrease with age in both [
22,
23]. Conversely, technical factors like a foreshortening LV image may overestimate the apical strain values [
22]. Unlike GLS, the CMR technique, native T1 mapping, quantitatively measures the relaxation time (T1) of myocardial tissue, which reflects the myocardial tissue composition, including the presence of water content (inflammation and edema) and/or collagen (fibrosis) [
24]. Thus, it shows diffuse abnormal changes in the cardiac tissue that may precede myocardial dysfunction [
24]. Nevertheless, it should be noted that the use of both measures together, GLS and native T1 mapping, may provide a stronger prognostic association with cardiac changes [
25].
Given that (1) the native myocardial T1 mapping is a marker of cardiac inflammation, tissue water content, and/or fibrosis [
21] and (2) the elevation of the native myocardial T1 mapping occurs before any cardiac functional abnormalities, we speculate that anthracyclines may induce cardiac injury, at least in part, via the activation of cardiac inflammation and fibrosis. This postulation is congruent with the notion that cardiac fibrosis and inflammation play a vital role in cardiotoxicity in several preclinical and clinical studies [
26,
27]. In support of this, suppressing inflammation and lessening cardiac fibrosis reduces myocardial injury in numerous pre-clinical models [
28‐
32]. Thus, it is likely that cardiac inflammation and fibrosis may contribute to impaired cardiac and represent a vital mechanism for anthracyclines-induced cardiotoxicity.
While the molecular mechanism of anthracycline-induced cardiac fibrosis and inflammation is still unclear, mounting evidence suggests a pivotal role of nucleotide-binding domain-like receptor protein-3 (NLRP3) inflammasome and myeloid differentiation primary response protein (MyD88) myddosome in the development of cardiac inflammation and fibrosis induced by anthracycline administered either alone or in combination with immune therapy, such as immune checkpoint inhibitors [
33,
34]. The activation of NLRP3 inflammasome and MyD88 myddosome increases the secretion of pro-inflammatory cytokines, such as IL-β, thereby contributing to chemotherapy-induced cardiac inflammation and fibrosis [
33,
34]. Thus, curtailing the activation of NLRP3 inflammasome and MyD88 myddosome may be a potential therapeutic approach to treat chemotherapy-induced cardiac inflammation and fibrosis. Interestingly, several recent studies have shown that sodium-glucose cotransporter 2 (SGLT2) inhibitors such as empagliflozin or dapagliflozin can reduce cardiac inflammation and fibrosis as well as improve cardiac function by suppressing the activation of NLRP3 inflammasome and MyD88 myddosome [
33,
35]. In addition to SGLT2 inhibitors, proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors have also demonstrated a potential protective effect against chemotherapy-induced cardiotoxicity by downregulating NLRP3 inflammasome and MyD88 myddosome [
36]. Thus, it is likely that strategies that target NLRP3 inflammasome and MyD88 myddosome, such as SGLT2 inhibitors and potentially PCSK9 inhibitors, may hold promise in mitigating anthracycline-induced cardiotoxicity.
An important limitation of this systematic review and meta-analysis is that we have a relatively small number of cancer patients treated with anthracyclines as well as healthy individuals. In addition, the included studies adopted a single-center quasi-experimental design. Thus, multicenter randomized controlled trials with a more significant number of patients and longer-term follow-up are required in order to provide fulsome data about the clinical utility of native T1 mapping for the early detection of anthracycline-induced cardiotoxicity. Also, only non-contrast MOLLI imaging has been used to measure native T1 mapping in the included studies. Nevertheless, non-contrast MOLLI imaging has more clinical feasibility for measuring native T1 mapping over the other imaging tools such as saturation recovery single-shot acquisition (SASHA) due to decreased motion artifact, higher reproducibility, and increased spatial resolution [
37]. Lastly, different strength magnets have been used to detect native myocardial T1 mapping suggesting that the meta-analysis is not relatively uniform despite low heterogeneity among the included studies. Given that we do not have access to the raw patient information data, we could not perform a meta-regression analysis.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.