Cardiac Tumors and Cardio-oncology

Quantitative Assessment of LV Ejection Fraction in Cardio-Oncology

Echocardiography is frequently used for sequential measurement of EF in the assessment of potential cardiotoxicity caused by chemotherapy or immune therapy for patients with malignancies ( Table 18.1 ). In most oncology practices, the LVEF is followed closely and given a great deal of clinical importance.

In clinical practice, the most commonly accepted definition of cardiac toxicity comes from a retrospective review by the independent Cardiac Review and Evaluation Committee (CREC) of cardiotoxicity of patients enrolled in a variety of trastuzumab clinical trials.

In practice, given the use of singular measures for any particular cardiotoxicity definition, it has been common for echocardiography clinicians to report single numbers and to avoid reporting EF ranges. Oncologists make critical clinical decisions based on a 5% or 10% EF change or a drop to less than 50% or 55%. When the value reported from one study to the next is 55% to 60% followed by 50% to 55%, confusion, if not inappropriate clinical consequences from decisions to halt anti-cancer therapy, may ensue.

Single-digit measures of LVEF have a long history in oncology. Cardiac imaging was established in the late 1970s and early 1980s, when a number of publications supported the use of different modalities for assessing cardiac toxicity. Measurement of LVEF by nuclear methods (i.e., multigated acquisition [MUGA]) soon became the established practice and was considered the gold standard for LV function assessment during chemotherapy. LVEF by radionuclide imaging proved to be sensitive, specific, and reproducible and was reported as a single measure.

Measurement of LVEF by MUGA as a sole indicator of cardiotoxicity has significant limitations, including image quality and the technical difficulty of the measurement. For instance, MUGA is subject to variations related to the use of a particular cardiac cycle, operator experience, and volume drawing style. The EF is the relative volume ejected in systole, a measurement that does not necessarily reflect intrinsic myocardial systolic function, particularly because it can be load dependent.

The ready availability, ease, and improvements in image quality of transthoracic echocardiography (TTE) have made it a more attractive option for the assessment of LVEF in cancer patients. The most commonly used LVEF measures in routine practice are two-dimensional (2D) methods, and among the 2D options, the most common method for volume calculations is the biplane method of disks summation (i.e., modified Simpson rule). It is the recommended 2D echocardiographic method by consensus and by current published LV quantification guidelines (see Chapter 4 ).

The literature is clear that the use of microbubble enhancement offers the best results in terms of intraobserver and interobserver variability, and contrast agents are recommended when needed to improve endocardial border delineation, specifically when two or more contiguous LV endocardial segments are poorly visualized in apical views. Microbubble-enhanced images provide larger volumes than unenhanced images. Volumes obtained in this fashion are closer to those obtained with cardiac magnetic resonance (CMR) (see Chapter 3 ).

Three-dimensional echocardiography (3DE) is more accurate than 2D methods for ventricular volume and EF measurements compared with CMR imaging, and 3DE should be used when available ^, (see Chapter 1 ). This method, which is still not routine in most centers, has been shown to offer the lowest temporal variability for EF and ventricular volumes, on the basis of multiple echocardiograms done over a period of 1 year in women with breast cancer receiving chemotherapy ( Fig. 18.1 ).

Doppler Methods for Detection of Cardiotoxicity

Earlier detection of cardiotoxicity allows a time advantage in risk stratification. Measurements of diastolic function by Doppler echocardiography could represent a marker for the early detection of toxicity. One study found that the isovolumetric relaxation time was significantly prolonged (from 66 ± 18 to 84 ± 24 ms; P < 0.05) after a cumulative doxorubicin dose of 100 to 120 mg/m ² . [CR] An increase of more than 37% in isovolumetric relaxation time was 78% sensitive (7 of 9 cases) and 88% specific (15 of 17 cases) for predicting the subsequent development of doxorubicin-induced systolic dysfunction. Overall, however, the results regarding the value of diastolic dysfunction as an indicator of the diagnosis of cardiotoxicity have been inconsistent. Because of the influence of hypertension and other risk factors on diastolic function, this signal appears to be nonspecific.

The myocardial performance (Tei) index is another important Doppler-derived tool. This index expresses the ratio of the sum of isovolumetric contraction time and isovolumetric relaxation time divided by the ejection time. This formula combines systolic and diastolic myocardial performance without geometric assumptions, and it correlates well with the results of invasive measurements. It is appealing for use with cancer patients because it appears to be independent of heart rate, mean arterial pressure, and degree of mitral regurgitation. It is sensitive and accurate in detecting subclinical cardiotoxicity associated with anthracycline therapy. Studies have shown that the Tei index is better than EF in detecting anthracycline-induced deterioration in LV function among adults: it detects deterioration earlier in the course of treatment and is more likely to detect statistically significant differences.

Speckle Tracking Characteristics of Cardiac Mechanics in Cardio-Oncology

New techniques have aimed at detecting cardiotoxicity before the onset of a measurable decrease in LVEF or symptoms. These methods include echocardiographic assessment for strain using speckle tracking imaging and testing for elevations in cardiac biomarkers, including troponin.

Speckle tracking takes full advantage of the capacity for image acquisition at higher frame rates. Use of this particular technology in the realm of cardiac dysfunction related to cancer therapeutics has yielded important results, particularly the use of longitudinal deformation measures, including global longitudinal strain (GLS).

It was first reported in 2009 that changes in tissue deformation assessed by myocardial strain and strain rate were able to identify LV dysfunction earlier than LVEF in women undergoing treatment with trastuzumab for breast cancer. Two subsequent reports produced comparable findings. ^,

A multicenter collaboration investigated the cooperative use of troponin and longitudinal strain measures to predict the development of cardiotoxicity in patients treated with anthracyclines and trastuzumab. Patients who demonstrated decreases in longitudinal strain measures or elevations in hypersensitive troponin had a ninefold increase in risk of cardiotoxicity at 6 months compared with those who had no changes in either of these markers. LVEF alone and diastolic function parameters failed to predict cardiotoxicity.

In a review that included more than 30 studies, although the best GLS cutoff value to predict cardiotoxicity was not clear, an early relative change of between 10% and 15% appeared to have the best specificity. Similar studies, however, found a stronger correlation from ventricular-arterial coupling and circumferential strain than from longitudinal measures. A consensus statement on the evaluation of adult patients during and after cancer therapy that was published by the American Society of Echocardiography and the European Association of Cardiovascular Imaging; based on the current literature, it recommended that a relative reduction in GLS of more than 15% is very likely to be abnormal, whereas a change of less than 8% appears to be of no clinical significance ( Fig. 18.2 ).

The same consensus statement recommended that an abnormal GLS value be confirmed by a repeat study. The repeat study should be performed 2 to 3 weeks after the initial abnormal study result. These recommendations have been reported mostly for the breast cancer population, and it remains to be seen whether the same cardiac imaging benefit extends to other malignancies and their treatment regimens ( Fig. 18.3 ).

Chemotherapeutic Impact on the RV

Right ventricular (RV) function is a prognostic indicator in patients with LV systolic dysfunction and various LV pathologies, but there is limited literature concerning the specific effects of cardiotoxic agents on the RV. For low-dose anthracycline chemotherapy regimens, it has been reported that a differential effect on ventricular function is detected using the myocardial performance index measure; a significant negative impact on LV function is produced, but RV function is spared. ^, A later study challenged that notion. Using a comprehensive list of measures, a retrospective analysis showed statistically significant decreases in RV fractional area change and longitudinal deformation in patients exposed to relatively low doses of anthracyclines.