Expanding indications for non-biopsy diagnosis of transthyretin amyloid cardiomyopathy

Graphical Abstract

Graphical Abstract.

An updated algorithm for non-biopsy diagnosis of transthyretin amyloid cardiomyopathy using bone-avid tracer cardiac SPECT/CT incorporating renally adjusted κ/λ ratios.

This editorial refers to ‘Tc-99m labelled bone scintigraphy in suspected cardiac amyloidosis’, by M.U. Rauf et al., https://doi.org/10.1093/eurheartj/ehad139.

Bone-avid tracer cardiac single photon emission computed tomography (SPECT) has revolutionized the evaluation and management of transthyretin amyloid cardiomyopathy (ATTR-CM). This cardiomyopathy, which results from cardiac interstitial deposition of misfolded transthyretin protein as amyloid fibrils, is an increasingly recognized cause of heart failure (HF) in older adults.¹ Until recently, endomyocardial biopsy (EMB) was required to confirm a diagnosis of ATTR-CM, but bone-avid tracer cardiac scintigraphy using technetium-99m-labelled pyrophosphate (PYP), ^99mTc-3,3-diphosphono-1,2-pyrophosphate (DPD), or hydroxymethylene diphosphonate (HMDP) has emerged as a non-invasive means to diagnose ATTR-CM without biopsy. Gillmore and colleagues² established the high specificity and positive predictive value of ^99mTc-PYP/DPD/HMDP scintigraphy to diagnose ATTR-CM non-invasively. This non-biopsy approach to diagnose ATTR-CM, using bone-avid tracer SPECT, is now supported by multiple societies^1,3,4 and incorporated into HF guidelines.^4,5 The availability of novel approved treatments⁶ for ATTR-CM and ATTR-neuropathy has sparked a tremendous clinical interest in diagnosing cardiac amyloidosis. The option of diagnosis without need for biopsy has opened the door for its use in screening for ATTR-CM in high-risk cohorts.⁷ Exponential growth in the use of ^99mTc-PYP/DPD/HMDP SPECT is leading to an era of earlier diagnosis, diagnosis of patients with a less severe disease phenotype,⁸ and novel insights into early disease pathology in ATTR-CM.⁸ Indeed, in the contemporary era, more patients are being diagnosed non-invasively than by biopsy,⁹ except in specific scenarios where EMB is till preferred.

The current algorithms do not recommend the use of ^99mTc-PYP/DPD/HMDP SPECT for diagnosing ATTR-CM in patients with monoclonal gammopathy (MG); EMB is recommended.^1,4 The reason is that in this situation myocardial uptake of ^99mTc-PYP/DPD/HMDP is not specific for ATTR-CM and a positive scan in patients with MG can represent either ATTR-CM or light chain amyloid cardiomyopathy (AL-CM). In the Gillmore study,² high grade myocardial tracer uptake (Grade 2/3, myocardial uptake equal to or greater than rib uptake) is far more common in ATTR-CM than in AL-CM, but is also seen in 21% of the patients with AL amyloid on EMB.² The presence of an MG substantially reduced the specificity of high grade uptake for ATTR-CM from 100% to 87%.² More importantly, misdiagnosis of myocardial uptake of ^99mTc-PYP/DPD/HMDP as ATTR-CM in a patient with AL amyloidosis can result in missed or delayed diagnosis of AL amyloidosis—a problem with serious consequences.¹⁰ Therefore, evaluation for MG is one of the first steps in most current algorithms for non-biopsy diagnosis of ATTR-CM¹¹ and, if MG is identified, additional evaluation is recommended, including cardiac magnetic resonance imaging (MRI) and/or biopsy.³

MG is diagnosed by abnormal serum free light chain (FLC) assay and serum and urine immunofixation electrophoresis.¹¹ Unlike serum and urine immunofixation electrophoresis which are both abnormal with a plasma cell clonal process, the serum FLC assay measures both normal and clonal (abnormal) light chains of kappa or lambda type (κ, λ).¹¹ A κ/λ ratio of 0.26 to 1.65 is normal. This ratio can become abnormal with overproduction of κ or λ light chains from a clonal process or by reduced renal clearance of FLCs in patients with impaired glomerular filtration rate (GFR).¹¹ Thus, interpretation of the κ/λ ratio in patients with renal dysfunction can be challenging. This is particularly relevant for patients with suspected ATTR-CM as renal function progressively decreases with age and with HF. Also, MG increases with age; 5.3% of people after age 70 years have an MG,¹² and 19% of patients with ATTR amyloidosis had an MG in one study.² This renders the non-biopsy diagnosis of ATTR-CM less useful in the cohort most needing it, subjecting them to risks of invasive EMB, challenges of histological evaluation, and costs. Better interpretation of the κ/λ ratio to distinguish a clonal process from decreased elimination in those with renal dysfunction is much needed.

Hutchison et al.¹³ showed that as renal function decreases, the serum and urinary concentrations of polyclonal FLC can increase up to five-fold. They proposed a higher cut-off of the FLC ratio of 0.37 to 3.1 in patients with an estimated GFR (eGFR) of < 60 mL/min/1.73 m² to increase the specificity of the FLC assays for detecting monoclonal FLC production in patients with severe renal failure.¹³ Further increasing the normal limits of κ/λ ratios in renal failure (to 0.82–3.6) has been shown to improve the specificity of diagnosis in multiple myeloma.¹⁴ Thus, renal adjustment of κ/λ ratios can avoid false-positive diagnosis of MG in patients with renal dysfunction.¹⁵ Amyloidosis experts with vast experience are currently able to evaluate the κ/λ ratio in the context of renal function and avoid EMB, but this is not recommended by the societal algorithms.

The study of Rauf et al.¹⁵ in this issue of the European Heart Journal aimed to (i) validate the specificity of the prior non-biopsy diagnostic algorithm for ATTR-CM and (ii) to develop new reference ranges for serum κ/λ FLC levels adjusted for eGFR. They studied 3354 patients at UK and multiple Italian amyloidosis centres. All patients underwent cardiac ^99mTc-DPD or HMDP SPECT/CT, evaluation for MG, and echocardiography, and many underwent biopsy (n = 300 EMB; n = 1129 any organ biopsy). Serum κ/λ FLC levels were interpreted using (i) the standard method (κ/λ ratio normal limits: 0.26–1.26); (ii) the Hutchinson method (κ/λ ratio normal limits: 0.37–3.1); and (iii) a refined method based on four estimated GFR categories mL/min/1.73 m². For the refined method, the lower limit of the κ/λ ratio was maintained at 0.26 but the upper cut-off limit was increased from 1.65 to 2.00, 2.50, and 3.10, for eGFR > 90, 60–90, 30–60, and < 30, respectively. Both Chronic Kidney Disease Epidemiology Collaboration (CKD-API)- and Modification of Diet in Renal Disease (MDRD)-based eGFR were evaluated.

Amyloidosis was prevalent in this cohort (62%). Evidence of MG was present in 21% of patients with Grade 2/3 uptake (myocardial radiotracer uptake equal to or greater than rib uptake), highlighting the need to screen for AL amyloidosis. Grade 2/3 scans mostly represented ATTR-CM, but 17% were seen with AL amyloidosis. Notably, ∼1 in 4 patients with Grade 0 scans (no myocardial radiotracer uptake) had biopsy-proven amyloidosis, predominantly (97%) AL amyloidosis. Of the patients with AL amyloidosis, the majority (64%) had Grade 0 scans, underscoring Rauf et al.’s conclusions that this test should not be used to diagnose AL amyloidosis. This study confirmed the high specificity (97% using EMB or 99.6% for any biopsy as reference) of the previously published diagnostic algorithm for non-biopsy diagnosis of ATTR-CM, using eGFR-based refined κ/λ ratio reference ranges. This approach eliminated the need for biopsy in 23% (369/1636) of patients who, by conventional κ/λ ratio criteria, would have been classified as having an MG; but this came at the expense of 9% of patients who were incorrectly labelled as having MG and requiring biopsy. Renal adjustment for the κ/λ ratio using either CKD-API- or MDRD-based eGFR was better than the Hutchinson method which misclassified two AL amyloidosis patients as having no MG. Current algorithms recommend screening for AL amyloidosis for non-biopsy diagnosis of ATTR-CM using the standard reference range of 0.26–1.65, and the findings of this study support the use of eGFR-adjusted refined κ/λ ratios to expand the indications for bone-avid tracer SPECT to diagnose ATTR-CM (Graphical Abstract). Rauf et al. appropriately prioritized high specificity to avoid unnecessary EMB. However, they found only a modest sensitivity, ranging from 46% to 63% (compared with 70% to 74% in a prior study²), of Grade 2/3 ^99mTc-DPD or HMDP SPECT/CT to diagnose ATTR-CM, underscoring the importance of further evaluation for patients with negative bone-avid tracer SPECT scans, including cardiac MRI and biopsy.

I applaud Rauf et al. for conducting this large study utilizing state of the art imaging with ^99mTc-DPD or HMDP SPECT/CT, including biopsy confirmation in many patients, having a median follow-up of 25.5 months, and using two methods of determination of eGFR. However, this observational study included patients with a high prevalence of ATTR-CM, and patients with known cardiac amyloidosis. In clinical practice, bone-avid tracer cardiac SPECT will be used in patients with a much lower prevalence of ATTR-CM, and in patients with suspected ATTR-CM. Furthermore, while the survival models illustrate the prognostic value of MG using the refined κ/λ reference ranges, multivariable risk-adjusted analyses were not performed. Nevertheless, the primary results of this study remain highly significant and show that bone-avid tracer cardiac SPECT should be combined with MG testing and using renal-specific κ/λ FLC ratio reference ranges.

In summary, Rauf et al. have successfully validated the high specificity of the current algorithm for the non-biopsy diagnosis of ATTR-CM. Through the use of eGFR-adjusted κ/λ ratios, their results have expanded the indications for ^99mTc-PYP/DPD/HMDP cardiac SPECT to diagnose ATTR-CM in patients with isolated FLC abnormality, and have supported the current recommendations of further evaluation, including cardiac MRI and EMB, despite a negative scan in patients with clinical and imaging findings suggestive of amyloidosis. In the near future, emerging techniques such as PET/CT with amyloid-targeting tracers are expected to provide non-biopsy diagnosis in patients with suspected cardiac amyloidosis of any type, including ATTR, AL, and rare forms of amyloidoses.

Funding

This work was supported by the National Institutes of Health grants R01 HL 150342; R01 HL 159987; and K24 HL 157648

Data availability

No new data were generated or analysed in support of this research.