MRI for the diagnosis of cardiac and liver iron overload in patients with transfusion‐dependent thalassemia: An algorithm to guide clinical use when availability is limited

Iron overload (IOL) in the heart and liver is a common complication in transfusion-dependent thalassemia (TDT) and is associated with increased risk of morbidity and mortality.1 Owing to the different mechanisms of iron accumulation and the wide scope of disease severity, iron-loading rate varies between organs and, consequently, different organs are affected unevenly. Nowadays, several measurement methods are available to diagnose and quantify IOL in the various organs.1 Serum ferritin (SF) has been the standard IOL measure in patients with TDT, especially through serial assessment. Although inexpensive, practical, and commonly available, SF does not always predict body iron as it is an indirect indicator of tissue iron burden and is influenced by several factors such as infection and inflammation.1 Its correlation with liver iron concentration (LIC), specifically, is significant but varies with assessment method and disease,2 while its correlation with cardiac iron is poor.3 SF < 1000 ng/mL has been associated with improved survival, while levels >2500 ng/mL indicate increased risk of cardiac morbidity and mortality.1 Alternatively, liver biopsy is a direct measure of iron in the liver; nevertheless, it is invasive and associated with risks and sampling and measurement errors.1, 4 Direct tissue-iron measurement in the heart is not possible. Magnetic resonance imaging (MRI) has emerged as a non-invasive alternative for quantifying both liver (by R2 or T2* MRI) and heart (by cardiac T2* MRI) iron by overcoming shortcomings of SF measurement and liver biopsy. The technique is sensitive, validated, and internationally reproducible and enables better tailoring of iron-chelation therapy (ICT) by allowing identification of patients at risk of organ-specific morbidity and monitoring of ICT effectiveness in the liver and heart.1, 4 Moreover, specific thresholds of LIC and cardiac T2* have been associated with clinical morbidities and mortality1, 5 (LIC >7 mg/g dry weight associated with clinical morbidity; cardiac T2* <20 ms associated with arrhythmias, <10 ms associated with heart failure and death). Therefore, monitoring of liver and heart iron by MRI is considered a standard of care in thalassemia.1 Despite the established utility and widespread use of MRI, it is not always available or affordable, especially in resource-poor regions where TDT prevalence is highest. Moreover, in healthcare centers with large patient volumes, assessment for all patients at regular intervals is not always feasible because of financial constraints. Furthermore, although international guidelines recommend standard monitoring of liver and cardiac iron by MRI (every 6 months to 2 years, depending on several factors and IOL status),1, 6 they do not provide sufficient guidance on when or whom to assess with MRI, especially when resources are limited and patient prioritization should be considered. To address the gaps in MRI utilization for TDT management in resource-poor settings, a panel of thalassemia experts from the Middle East, North Africa, and Asia Pacific regions met and proposed a simple and practical algorithm based on their experience of limitations and obstacles encountered with MRI utilization in those regions. The algorithm can guide the use of MRI for the detection and monitoring of IOL in TDT while considering its availability and accessibility for patients. Three scenarios were proposed based on availability and accessibility of MRI. Scenario 1: MRI available and accessible for all patients Age is the main determinant for initial MRI measurement. Consequently, in an ideal situation whereby MRI is easily accessible, all TDT patients aged ≥10 years, regardless of all other characteristics, would be candidates for baseline IOL screening by liver and heart MRI (based on validated methods and standard calibration). Of note, and in view of the recently published reports describing the early occurrence of liver and cardiac iron loading in children with TDT,7, 8 MRI could still be requested in patients <10 years old informed by physician discretion, based on transfusion burden, SF levels, clinical complication profile, and MRI availability (Figure 1). Algorithm to guide MRI use in TDT across countries with limited resources. †Based on serial measurements of at least three SF readings; ‡Irrespective of SF levels and heart-failure status. LIC, liver iron concentration; MRI, magnetic resonance imaging; Q, every; SF, serum ferritin; T2*, cardiac T2* Scenario 2: MRI available but not accessible for all patients Where MRI is not accessible for all patients, those aged 10–18 years with SF >2500 ng/mL should be prioritized, followed by any patient with SF >2500 ng/mL to optimize chelation therapy and prevent complications. In addition, patients with an increasing SF trend despite iron chelation should be considered for MRI screening as increasing SF levels could reflect an increase in organ iron loading. Liver and heart examination should be performed simultaneously when possible. The frequency of follow-up testing depends on the baseline reading: repeat MRI annually when cardiac T2* <20 ms or LIC >7 mg/g dry weight by either R2 or T2* MRI, or every 6 months when cardiac T2* <10 ms; otherwise, repeat every 2 years. The proposed thresholds and frequency should be considered irrespective of SF levels and heart-failure status. Notably, those with cardiac T2* <6 ms might require more frequent evaluation with MRI, depending on individual clinical profile. Patients with heart failure, however, would need to be monitored and managed according to standard of care for heart disease (Figure 1). It should be noted that recent evidence supports a role for pancreatic iron monitoring with MRI (done as part of abdominal MRI imaging for LIC) in predicting cardiac iron loading.9, 10 If technical expertise is available, the use of pancreatic iron measurement during abdominal MRI as a surrogate for cardiac T2* may be considered especially that it also saves time and burden of undergoing a cardiac MRI when access is limited. It should be noted, however, that the use of pancreatic MRI is still experimental in the absence of validation studies and experience from iron chelator clinical trials. Scenario 3: MRI unavailable Finally, where MRI is unavailable, serial SF values obtained every 3 months should be used to monitor IOL (Figure 1). The use of MRI does not come without challenges as values from MRI measurements depend not only on tissue iron concentration but also on the quality of data acquisition and analysis. Therefore, calibration and validation of MRI techniques must be carried out to obtain appropriate and accurate measurements of iron concentration and, consequently, to make the right clinical decisions. It is equally important that technicians and analysts are well trained on the scanner and aware of its potential pitfalls.11 In the absence of specific guidelines for assessing and monitoring IOL in TDT patients in resource-poor countries, and with the persistence of challenges in availability and access to MRI in such regions with high patient volumes and high associated costs, the development of an algorithm for MRI in IOL assessment herein is a first step toward improving the management of these patients. This guide might also be useful in well-developed settings in countries facing challenges related to any recent massive influx of immigrants that might have exhausted their healthcare services. The consolidation of this algorithm in the clinical setting is essential to optimize the use of MRI and, consequently, clinical outcomes among this population. Authors thank Sarah Keyrouz of AMICULUM, Ltd. for medical writing assistance. Financial support for medical writing assistance was provided by Novartis Pharma AG. VV reports receiving research funding from Novartis, Roche Pharmaceuticals, Celgene, Roche Diagnostics, Sebia, and Bio-Rad, consultancy fees from Novartis, Roche Pharmaceuticals, Celgene, Agios, and La Jolla, and participating in advisory boards sponsored by Novartis, Roche Pharmaceuticals, and Celgene. AA reports participating in advisory boards sponsored by Novartis. YA reports receiving honoraria from Cerus and Novartis and research funding from Cerus, Novartis, and Celgene. AT reports receiving honoraria and research funding from Novartis and research funding from Celgene and Roche. BAE, HD, ASI, LR, OT, ASW, and LW have no conflicts of interest to disclose.