Challenging the myths of cardiac amyloidosis

This editorial refers to ‘Diagnostic sensitivity of abdominal fat aspiration in cardiac amyloidosis’†, by C.C. Quarta et al., on page 1905 and ‘Clinical characteristics of wild-type transthyretin cardiac amyloidosis: disproving myths’‡, by E. González-López et al., on page 1895. The scientific history of systemic amyloidosis is replete with reports of incremental advancements reflecting developments in diagnostic tools and therapies, punctuated by frequent paradigm shifts that challenge the conventional wisdom of the day. First identified as a carbohydrate in the 19th century by Virchow, amyloid was subsequently reclassified as a protein, and then later as a category of proteins.1 Clinically too, the field of amyloidosis has evolved to represent a growing spectrum of disease manifestations. While the principal proteins that cause systemic amyloidosis were initially thought to be derived from immunoglobulin light chain fragments (AL amyloidosis) or from acute phase reactants (AA amyloidosis), it was soon discovered that pre-albumin (now known as transthyretin amyloid or ATTR) was the culprit in familial amyloidotic polyneuropathy2 and cardiomyopathy.3 We now know that ATTR amyloidosis can result from genetically abnormal TTR protein (ATTRm) and, at present, >100 TTR mutations have been identified.4 ATTR amyloidosis can also result from genetically normal or wild-type TTR, resulting in ATTRwt (formerly known as senile cardiac) amyloidosis. Cardiac amyloidosis is an infiltrative cardiomyopathy characterized by increased biventricular wall thickness, restrictive left ventricular (LV) filling, and, often, a non-dilated LV cavity with preserved or mildly depressed LV systolic function.5 AL amyloidosis typically affects multiple organ systems including renal and gastrointestinal systems, with cardiac involvement seen as a common manifestation in up to 70% of cases.6 In contrast, cardiac involvement in ATTRm amyloidosis is variable and depends on the specific mutation.7 While AL cardiac amyloidosis is rightly considered a rare disease (estimated incidence 1:100 000),8 recent evidence suggests that ATTRwt amyloidosis is probably much more common than is widely appreciated in elderly patients with heart failure with preserved ejection fraction (HFpEF).9 , 10 The diagnosis and appropriate typing of cardiac amyloidosis is critical for the clinical management of these patients so that life-extending chemotherapy can be administered in AL amyloidosis, and the TTR-stabilizing agent tafamidis (Vyndaqel, Pfizer Inc., available in Europe but not the USA) in ATTR amyloidosis. With the development of promising new therapies for systemic amyloidosis, early and accurate diagnosis of the precursor protein has tremendous potential to prolong survival and improve outcomes. Diagnosis is challenging, however, because features of cardiac amyloidosis must be disentangled from other more common ‘hypertrophic’ processes including hypertensive remodelling, hypertrophy in response to aortic stenosis, or hypertrophic cardiomyopathy (HCM). The diagnostic algorithm requires different clinical inputs including history, physical findings, and imaging, including echocardiography, cardiac magnetic resonance imaging (CMR), and, most recently, the highly ATTR-specific bone avid nuclear tracers 99mTc-PYP (pyrophosphate) and 99mTc-DPD (3,3-diphosphono-1,2-propanodicarboxylic acid).11 , 12 This issue of the journal features two distinct but complementary studies that separately refine the methodology employed for diagnosis of cardiac amyloidosis and identify heretofore less appreciated clinical features of the most common form of the disease (ATTRwt amyloidosis). In the first study, Quarta et al. assessed the diagnostic performance of an abdominal fat aspirate, perhaps the most easily accessible tissue biopsy, for the identification of cardiac amyloidosis.13 While advances in imaging have made it possible to definitively diagnose ATTR cardiac amyloidosis non-invasively, a tissue diagnosis remains important, particularly in AL disease, to confirm the type of the protein involved. In the early 1980s, Libbey and colleagues at Boston University first reported that Congo Red staining of abdominal fat obtained by fine needle aspiration diagnosed amyloidosis in 95% of patients with AL amyloidosis and 86% of patients with ATTRm amyloidosis.14 Unlike cardiac or gastrointestinal biopsy, the procedure is easily performed in the clinical exam room, with local anaesthesia, and minimal complication. Six hundred patients (an exceptionally large cohort) with an established diagnosis of cardiac amyloidosis referred to the National Amyloidosis Centre in London, UK, underwent fine needle aspiration of their abdominal fat. Congo Red staining identified amyloidosis in 84% of patients with AL amyloidosis, as compared with 45% of patients with ATTRm amyloidosis, and only 15% of those with ATTRwt amyloidosis. Immunohistochemistry of the fat aspirates correctly identified light chain fragments in 56% of AL amyloidosis patients who were Congo Red positive (Figure 1). While much smaller studies have examined the feasibility and diagnostic yield of immune techniques as a means to type precursor protein from abdominal fat aspirates,15 the present report is by far the largest to assess the diagnostic sensitivity of Congo Red staining and immunohistochemical typing of abdominal fat aspirates in patients with cardiac amyloidosis. The sensitivity of amyloid detection by fat aspirate in AL amyloidosis patients was proportionate to the burden of systemic amyloidosis by [123I]serum amyloid P (SAP) scintigraphy, suggesting that a larger whole-body amyloid burden increased the likelihood of detection by abdominal fat aspirate. The authors also found that in ATTRm amyloidosis, the specific mutation was associated with the likelihood of positive needle aspirate. As discussed by the authors, this probably reflects the higher systemic burden of amyloid in patients with mutations that cause polyneuropathy and cardiomyopathy (such as Thr60Ala) as opposed to those that cause cardiac-restricted amyloidosis (Val122Ile). Consistent with this observation, patients with cardiac-restricted ATTRwt amyloidosis were unlikely to have positive aspirates, suggesting that for this particular type of disease, further testing, such as CMR, 99mTc-PYP, or 99mTc-DPD, is a necessity. Furthermore, in equivocal cases, for example patients with a positive DPD scan and monoclonal gammopathy (seen in 10% of cases),16 , 17 an endomyocardial biopsy may be necessary to establish the correct diagnosis. Finally, while the authors report successful typing of the precursor amyloidogenic protein in 47% of all AL cases, 33% of all ATTRm amyloidosis cases, and 10% of all ATTRwt amyloidosis cases by immunohistochemistry, it is important to remember that this technique requires sufficient pathological experience and expertise, and as such a similar diagnostic yield may not be attainable in other centres. Salient features of the two studies reviewed in this Editorial. The left panel describes the diagnostic yield from abdominal fat fine needle aspiration in patients with AL, mutant ATTR, and wild-type ATTR cardiac amyloidosis from the study by Quarta et al. The right panel summarizes the main results from the study by Gonzalez-Lopez et al. that revise our present understanding of wild-type ATTR cardiac amyloidosis. In the second study, Gonzalez-Lopez et al. challenge some of the currently held ‘myths’ of ATTRwt cardiac amyloidosis with the report of a retrospective cohort of 108 patients from two amyloid referral centres in Europe (Figure 1).17 Myth #1: ATTRwt amyloidosis is typically held to be most commonly a disease of elderly males, with the average proportion of females in the largest cohorts published to date between 2% and 11%.16 , 18 , 19 However, in the current report, 18.5% of the ATTRwt amyloidosis cohort studied were female, suggesting that ATTRwt amyloidosis may be almost twice as common in women than is presently appreciated. Women with ATTRwt amyloidosis were more likely to have been diagnosed by non-invasive 99mTc-DPD scintigraphy (75% of women were diagnosed in this manner), suggesting that the advent and accessibility of this new imaging test may permit a more accurate determination of the true gender distribution of this disease. Interestingly, female cases were also older at the time of symptom onset and diagnosis (by ∼6 years), arguing for a delayed presentation or ‘protective’ effect of gender. Myth #2: cardiac amyloidosis is widely held to be a symmetric wall thickening process as distinct from HCM, which is more commonly asymmetric. In contrast, Gonzalez-Lopez et al. found that LV wall thickening was asymmetric in 23% of patients, as defined by echocardiography. The pattern of asymmetry was not defined further, but is illustrated in figure 4 of this report. While asymmetry mimicking HCM has been reported in AL amyloidosis by echocardiography,20 asymmetry is not well appreciated in ATTRwt amyloidosis, and therefore its presence should no longer be utilized to favour HCM strongly over amyloidosis. In addition, other observations made in this cohort reaffirm the challenge of identification of ATTRwt amyloidosis by standard testing that includes the findings of lower overall LVEF (in fact, 9% of patients had an LVEF <30%) and LV hypertrophy by ECG (10%), both of which are not commonly associated with cardiac amyloidosis. Another important point made by the authors is the high degree of misdiagnosis (35%) prior to accurate ATTRwt amyloidosis diagnosis. Not surprisingly, hypertensive heart disease and HCM were the most frequently made diagnoses prior to the accurate one. On one hand, misdiagnosis clearly results in delay of definitive treatment, and probably negatively impacts mortality. On the other, given the age-dependent penetrance of ATTRwt amyloidosis, it is interesting to consider that prior disease diagnoses were indeed accurate, but that ATTRwt amyloidosis later ensued with advancing age, changing the cardiac pathophysiology. This is likely to be the case with low-gradient aortic stenosis and the observed 6% ATTRwt amyloidosis incidence as has been recently reported.21 With respect to survival, Gonzalez-Lopez et al. observed a median time to death of ∼6.1 years, which is roughly twice that of the Boston University (3.9 years),16 London (2.7 years),18 and Mayo (3.6 years) cohorts.19 This is probably due to a skew of this cohort toward earlier disease, with incidentally identified cases (11%) and a sizeable proportion of asymptomatic patients (18%). As in any chronic disease, the true survival depends upon the time of diagnosis. Through this lens, the findings of Gonzalez-Lopez et al. are encouraging, because they suggest that if diagnosed earlier, survival is indeed likely to be better. Finally, and with more reason for optimism, we can anticipate that administration of ATTR amyloidosis-stabilizing therapy might result in even better outcomes when instituted at earlier stages of disease. Cardiac amyloidosis (particularly ATTRwt and ATTRm amyloidosis) is an important cause of heart failure in the elderly that, as informed by the studies published in this issue, is marked by clinical heterogeneity. Fine needle aspiration of abdominal fat is an easy and relatively reliable method for obtaining a tissue diagnosis, especially in the case of AL amyloidosis. This method is variably sensitive for diagnosis of ATTRm amyloidosis, and particularly insensitive for ATTRwt amyloidosis, yet is certainly advisable as a first site of biopsy before proceeding to more invasive procedures. Integration of the findings reported here with a currently accepted clinical diagnostic algorithm can be found in Figure 2. Early diagnosis and proper classification of cardiac amyloidosis are especially important given that the specific treatments for AL22 and ATTR cardiac amyloidosis23 hold great promise for improving survival in these otherwise fatal diseases. Investigations such as those reported in this issue of the journal advance our understanding of the clinical features of cardiac amyloidosis and provide validation of tools that clinicians can use to diagnose this disease accurately. Diagnostic algorithm for cardiac amyloidosis. CAD, coronary artery diseas; DPD, 99mTc 3,3-diphosphono-1,2-propanodicarboxylic acid scan; LC/MS2, liquid chromatography–tandem mass spectrometry; CR, Congo red; LGE, late gadolinium enhancement; LV, left ventricular; MRI, magnetic resonance imaging; Pyrophosphate, 99mTc pyrophosphate scan; SIFE, serum immunofixation electrophoresis; UIFE, urine immunofixation electrophoresis. *Grading scale for pyrophosphate/DPD scans: positive = heart uptake: contralateral lung uptake ratio ≥1.5 (after 1 h) or Perugini visual score of 2 or 3; negative = heart uptake: contralateral lung uptake ratio <1.0 or Perugini visual score of 0; intermediate = heart uptake: contralateral lung uptake ratio between 1.0 and 1.5 or Perugini visual score of 1. Conflicts of interest: F.L.R. is a consultant to Prothena Pharmaceuticals and Alnylam Pharmaceuticals. O.K.S. has no conflicts to declare.