Prospective Identification of Cardiac Amyloidosis by Cardiac Magnetic Resonance Imaging

RESEARCH PROTOCOL

Hypothesis:

The presence of amyloid protein in the myocardium changes its function and tissue
characteristics. These changes are responsible for the poor prognosis of patients with
cardiac amyloidosis. Cardiac magnetic resonance imaging (CMR) offers a novel, non-invasive
approach to identify cardiac involvement that may impact patient management. This study will
include the prospective validation of CMR parameters qualified as being abnormal in the
retrospective study (study ID 2012-3315) and the current literature. The myocardium/blood
pool inversion time null point ratio (Myo/BlP TI0 ratio), has been qualified as being
significantly different when compared to controls in the retrospective study.

Specific Aims:

1. Validate the diagnostic accuracy of CMR parameters in a prospective manner for cardiac
amyloidosis patients.

2. CMR parameters, including the Myo/BlP TI0 ratio, will be compared to serum biomarkers
(TroponinT, NT-proBNP, serum lambda/kappa free light chain concentration), established
to have prognostic value in cardiac amyloidosis. Subjects will be followed via death
registry and/or medical records to compare the prognostic value of the CMR parameter
with the biomarkers.

3. CMR parameters will be used to assess for the presence of early cardiac involvement in
amyloidosis patients without clinically apparent cardiac involvement (as determined by
symptoms, biomarkers and/or cardiac imaging).

4. CMR will be used to follow patients undergoing treatment, when available, to determine
whether CMR parameters consistent with the presence of cardiac amyloidosis, change
during treatment.

Background and Significance:

Matthias Schleiden, a German botanist and co-creator of the cell theory, used the term
amyloid in 1834 to characterize the waxy starch in plants. Today, amyloid is used to describe
any of a number of small proteins which when aggregated lead to insoluble fibrillar deposits.
Many mechanisms of protein dysfunction contribute to amyloidogenesis, including
"non-physiologic proteolysis, defective physiologic proteolysis, mutations involving changes
in thermodynamic or kinetic properties, and pathways that are yet to be defined". Amyloid
deposits are identified on the basis of their apple-green birefringence under a polarized
light microscope after staining with Congo red. They can also be identified based on the
presence of rigid, non-branching fibrils using electron microscopy. Amyloidosis describes the
infiltration of organs by these insoluble deposits. In humans, about 27 different unrelated
proteins are known to form amyloid fibrils in vivo. The organs involved and clinical
presentations are dependent on the precursor protein. A small number of the amyloidoses can
have cardiac involvement with a significant impact on mortality. The clinically most
significant cardiac amyloidoses include immunoglobulin light chain amyloidosis (AL) and the
transthyretin amyloidoses (ATTR); ATTR being either hereditary secondary to mutations in the
transthyretin gene or acquired secondary to wild type transthyretin protein.

AL amyloidosis, the most common type of systemic amyloidosis, is associated with plasma cell
dyscrasias. In AL amyloidosis, fifty percent of affected patients will have cardiac
involvement. Despite the potential for amyloid deposition in multiple organs, cardiac
involvement continues to dictate the poorest prognosis. Once congestive heart failure
develops in a patient with AL amyloidosis, survival is less than 6 months if untreated.
Therefore, patients with AL amyloidosis are screened, by assessing serum troponins and brain
natriuretic peptides, to determine whether cardiac involvement is present. Patients with ATTR
amyloidosis typically has a more insidious onset in comparison. ATTR amyloidosis also
responds better to traditional congestive heart failure management. However, neither AL nor
the ATTR amyloidoses are managed like traditional heart failure making appropriate diagnosis
critical for best care. Furthermore, early diagnosis of cardiac involvement may improve
outcomes but requires heightened suspicion and a systematic clinical approach to evaluation
[3].

In cardiac amyloidosis, there can be evidence of disease on routine testing. Restrictive
physiology by echocardiography, low EKG voltage as well as additional EKG abnormalities are
often identified in advanced cardiac amyloidosis. Routine nuclear cardiac imaging has not
proven to be helpful in diagnosis, although 123I-MIBG does appear to demonstrate
co-localization of denervation with amyloid deposition. Unfortunately, no routine cardiac
testing is specific for cardiac amyloidosis. Therefore, in the absence of high clinical
suspicion for cardiac amyloidosis, the appropriate diagnosis will often be missed.

Endomyocardial biopsy remains the gold standard for making the diagnosis of cardiac
involvement. However, given the invasive nature of endomyocardial biopsy and associated risks
(including death), the diagnosis of cardiac amyloidosis is usually made by non-invasive
means; this diagnosis being supported by a non-cardiac biopsy demonstrating amyloid deposits.
The most frequently biopsied site is the abdominal fat pad, usually positive in patients with
AL amyloidosis. However, ATTR amyloidosis is not reliably identified with fat pad biopsy.
Therefore, a negative fat pad biopsy requires further work up including endomyocardial biopsy
with associated immunostaining and genetic testing for transthyretin mutations. Supporting
the diagnosis of cardiac amyloidosis by noninvasive techniques such as EKG, echocardiography
and serum amyloid component scintigraphy has been shown to have significant limitations. A
novel non-invasive approach is required to ensure accurate diagnosis. Given the limitations
of previously used non-invasive techniques, there has been interest over the last decade in
using cardiovascular magnetic resonance (CMR) imaging in the diagnosis of cardiac
amyloidosis.

The utility of CMR in evaluating patients with cardiac amyloidosis was first proposed in the
1990s. However, evaluation was limited to the assessment of myocardial morphology and
function. It was not until the turn of the century that the powerful diagnostic potential of
delayed gadolinium enhancement (DGE) was realized. The first utility of DGE was in the
evaluation of myocardial viability in patients with coronary artery disease. It quickly
became apparent that DGE could also be used to evaluate patients with non-ischemic
cardiomyopathies. The characterization of cardiac amyloidosis with DGE was first described as
global sub-endocardial enhancement. Abnormal T1 transmyocardial maps were used to both
describe the abnormality and to assist in the determination of prognosis. However, these
studies were limited by inclusion of only patients with biopsy-proven amyloidosis who also
met the echocardiographic criteria for restrictive physiology, common to advanced cardiac
amyloidosis. More recent work precluding the use of echocardiographic diagnosis supported the
concept that the inability to null the myocardium relative to the blood pool may be an early
sign of cardiac amyloidosis. In addition, to the DGE data, studies have utilized T1 and T2
weighted imaging to identify abnormal extracellular volume (ECV) and myocardial edema in the
cardiac amyloidosis population.

It has been the culmination of these findings, in addition to our own experience studying
cardiac amyloidosis that led to the hypothesis that the presence of amyloid protein, known to
increase the ECV, may also change the inversion time curves of both the blood pool and the
myocardium. These features, as well as abnormal functional parameters, will help detect
cardiac amyloidosis earlier than would otherwise be possible by morphologic assessment. The
possibility of diagnosing sub-clinical disease has implications for screening patients with
known amyloidosis for early cardiac involvement. Also, CMR parameters may be useful in
assessing response to treatment of the plasma cell dyscrasias, such as cardiac amyloidosis,
that may be complicated by heart failure and ventricular tachycardia. Lastly, the
relationship of the MRI findings to cardiac biomarkers has yet to be described and will be
assessed in this study.

Preliminary Data:

The investigators have observed a significant difference between the Myo/BlP TI0 ratio in
patients with cardiac amyloidosis when compared with controls both with preserved LV function
(CTL) and non-amyloid cardiomyopathies (CTL-CMP). The Myo/BlP TI0 ratio of the CA group is
close to 1 to 1 (0.95 +/- 0.16) leading to poor myocardium and blood pool contrast in the
corresponding delayed enhancement images acquired following the TI scout (Figure 7). The
Myo/BlP TI0 ratio provides a quantitative method to support the diagnosis of cardiac
amyloidosis in patients presenting with cardiac symptoms.

Estimated Period of Time to Completion:

Estimated time to completion of aim 1, 2 and 4 will depend on power calculations following
the acquisition of pilot prospective data. The estimated timing of completion of aim 3 will
be 12-18 months from the start of data acquisition.

Inclusion Criteria:

- Subjects meeting diagnostic criteria for the target population will be recruited for
the study. Age range will be 18 to 85 years of age.

Exclusion Criteria:

- Exclusion criteria will include history of heart transplantation as well as any
contraindications for 1.5T CMR. For the use of GBCA, patients with acute renal failure
or stage IV/V chronic renal failure will be excluded. Patients with end stage renal
failure receiving peritoneal dialysis will also be excluded. Women who are pregnant or
breast feeding will also be excluded.