Association of Patisiran, an RNA Interference Therapeutic, With Regional Left Ventricular Myocardial Strain in Hereditary Transthyretin Amyloidosis: The APOLLO Study

Key Points

Question  Compared with placebo, does patisiran improve regional left ventricular (LV) strain in hereditary transthyretin-mediated (hATTR) amyloidosis?

Findings  In this secondary analysis of a cardiac subpopulation from the phase 3 APOLLO study, a randomized clinical trial in patients with hATTR amyloidosis, patisiran prevented deterioration in LV global longitudinal strain predominantly in the basal longitudinal strain compared with placebo at 18 months and no significant differences were found in the mid and apical regions between the patisiran and placebo groups.

Meaning  These findings highlight a potential role of the LV basal longitudinal strain as a sensitive marker to evaluate the association with cardiac manifestation in hATTR amyloidosis.

Importance  Patients with cardiac amyloidosis demonstrate reduced myocardial strain with associated sparing of the cardiac apex. In the APOLLO randomized clinical trial, patisiran, an RNA interference therapeutic that inhibits transthyretin synthesis, improved left ventricular (LV) global longitudinal strain (LV GLS) compared with placebo in patients with hereditary transthyretin-mediated (hATTR) amyloidosis with polyneuropathy and evidence of cardiac involvement.

Objective  To evaluate the treatment association of patisiran with regional LV myocardial strain in cardiac manifestation in hATTR amyloidosis.

Design, Setting, and Participants  This exploratory analysis of APOLLO, a randomized, double-blind, placebo-controlled, phase 3, multicenter international clinical trial that was conducted from December 2013 to January 2016, included patients with hATTR amyloidosis with polyneuropathy who were randomized 2:1 to receive patisiran or placebo. The prespecified cardiac subpopulation (126 of 225 [56%]) comprised patients with a baseline LV wall thickness of 13 mm or more and no history of hypertension or aortic valve disease. This post hoc data analysis was performed between September 2018 and January 2019.

Intervention  Placebo or patisiran, 0.3 mg/kg, via intravenous infusion once every 3 weeks for 18 months.

Main Outcomes and Measures  The association of patisiran with LV regional longitudinal strain at 18 months.

Results  Of the 126 patients included in the prespecified cardiac subpopulation, 36 patients (28.6%) received placebo (median [interquartile range] age, 62 [57-72] years) and 90 patients (71.4%) received patisiran (median [interquartile range] age, 60 [54-66] years); 98 (77.8%) were men, 28 (22.2%) were from North America, and 43 (34.1%) were from Western Europe. At baseline, LV GLS was impaired and regional longitudinal strains were lowest in the basal segments with apical sparing. There were no differences in regional longitudinal strains between the treatment groups at baseline. Patisiran improved the absolute GLS (least-squares mean [SE] difference, 1.4% [0.6%]; 95% CI, 0.3%-2.5%; P = .02) compared with placebo at 18 months, with the greatest differential increase observed in the basal region (overall least-squares mean [SE] difference, 2.1% [0.8%]; 95% CI, 0.6%-3.6%; P = .006) and no significant differences in the mid and apical regions among groups.

Conclusions and Relevance  Patisiran prevented the deterioration of LV GLS over 18 months, driven primarily by attenuating disease progression in the basal region, suggesting that basal longitudinal strain may be a more sensitive marker of treatment associations with the cardiac manifestation in hATTR amyloidosis and that basal region may be influenced by disease-modifying therapies more than other ventricular regions.

Trial Registration  ClinicalTrials.gov identifier: NCT01960348

Hereditary transthyretin-mediated (hATTR) amyloidosis is an autosomal dominant, progressive, and life-threatening disease that is caused by mutations in the gene encoding transthyretin protein during which most patients develop a mixed phenotype of polyneuropathy and cardiomyopathy. In hATTR amyloidosis, the presence of cardiac involvement is associated with a worse prognosis. The current treatment options for hATTR amyloidosis are limited. Although liver transplant and some pharmacological treatment strategies, such as transthyretin (TTR) tetramer stabilizers (eg, tafamidis or diflunisal), have been available, many patients have continued disease progression.^1-6 In the United States and Europe, inotersen, an antisense oligonucleotide, and patisiran, a small interfering RNA, were recently approved for this condition.

Patisiran is specific for TTR and is formulated in a hepatotropic lipid nanoparticle for intravenous administration. Following lipid nanoparticle–mediated delivery to hepatocytes, the main site of TTR production, patisiran targets TTR messenger RNA for degradation, resulting in the potent and sustained reduction of the mutant and wild-type TTR protein.⁷ The recently completed phase 3 APOLLO study demonstrated that patisiran improved neuropathy and quality of life in patients with hATTR.⁸

Two-dimensional speckle tracking echocardiography (2-D–STE) has played a role in the diagnosis and risk stratification of cardiac amyloidosis. In contrast to left ventricular (LV) ejection fraction, which may remain normal until the late stages of the disease, LV global longitudinal strain (GLS) by STE, a sensitive measure of systolic function, is universally impaired and has been shown to be an independent predictor of mortality in patients with cardiac amyloidosis.⁹ Furthermore, LV apical segments are typically less affected, a finding consistently observed in ATTR and light chain amyloidosis.^9,10 Although patisiran reduced LV wall thickness and improved LV GLS compared with placebo in a prespecified subpopulation of patients with cardiac involvement, to our knowledge, the association of any amyloid treatment, including patisiran, with regional LV myocardial strain is not known.^8,11 Therefore, the objective of this study was to compare the change in regional myocardial strain evaluated by 2-D–STE between the patisiran and placebo groups in the cardiac subpopulation.

APOLLO (NCT01960348) was a phase 3, multicenter, randomized, double-blind, placebo-controlled clinical trial comparing patisiran with a placebo in patients with hATTR amyloidosis with polyneuropathy. APOLLO was approved by institutional review boards or local ethics committees at 44 sites in 19 countries. All the participants provided written informed consent. The details of the study have been published.^8,11 Eligible patients were age 18 to 85 years, had a diagnosis of hATTR amyloidosis with a documented TTR mutation and symptomatic neuropathy, were ambulatory, and had adequate liver and renal function. Patients with a previous liver transplant or who were planning to undergo liver transplant during the trial period or who had a New York Heart Association class of III or IV were excluded. Patients were randomized 2:1 to receive patisiran, 0.3 mg/kg, or placebo via intravenous infusion once every 3 weeks for 18 months. A cardiac subpopulation was prespecified in the statistical analysis plan and comprised patients with a baseline LV wall thickness of 13 mm or more in the absence of a history of hypertension or aortic valve disease. The assessment of cardiac structure and function via 2-D echocardiography were exploratory end points in the APOLLO study and assessments were conducted at baseline, 9 months, and 18 months; echocardiogram results were analyzed in a masked core laboratory. Left ventricular myocardial strain was assessed with speckle tracking using vendor-independent software (Tom Tec Image Area, version 4.6; TOMTEC).^9,11 The peak longitudinal strain (LS) values were computed automatically, generating regional data from 6 segments and an average value for apical 4-chamber and 2-chamber views. Peak LS is a negative value, but for ease of interpretation in comparing serial LS values, we have expressed LS as an absolute value. The relative regional strain ratio, a measure of the degree of apical sparing of LS, was defined as the average apical LS divided by the average mid plus basal LS values.¹⁰ Reproducibility was assessed by repeated measurements in a subset of patients with an average of coefficient of variation for basal LS of less than 10% (eMethods in the Supplement).

The cardiac parameters were analyzed in the prespecified cardiac subpopulation, using a mixed-effects model that included the change from baseline as an outcome variable with corresponding baseline values as a continuous covariate. The primary comparison was the difference in the least squares mean between the patisiran and placebo groups at 18 months.^8,11 A P value of <.05 was considered to indicate statistical significance. All analyses were performed using Stata, version 14.1 (StataCorp) (eMethods in the Supplement).

Of the 225 patients enrolled in APOLLO, 126 (56%) were included in the prespecified cardiac subpopulation, of whom 36 (29%) received placebo and 90 (71%) received patisiran. The patients’ baseline clinical characteristics and echocardiographic parameters are listed in Table 1. Conventional echocardiography results showed the typical features of cardiac amyloidosis, including increased LV wall thickness and preserved LV ejection fraction, and the LV GLS was impaired. Regional deformation analyses showed a basal-to-apical gradient, with average LS values lowest in the basal segments compared with mid and apical segments. At baseline, there were no differences in STE parameters between the treatment groups.

In the cardiac subpopulation, patisiran improved LV GLS (least squares mean [SE] difference, 1.4% [0.6%]; 95% CI, 0.3%-0.5%; P = .02) compared with placebo at 18 months. The placebo group demonstrated the deterioration in the LV GLS and regional longitudinal strains (the absolute decrease indicating worsening in systolic function) at 18 months, whereas the patisiran group showed stability in these variables. The treatment association of patisiran with regional LS parameters was only observed in the basal regions (overall least squares mean [SE] difference, 2.1% [0.8]; 95% CI, 0.6%-3.6%; P = .006), and no significant differences in the mid and apical regions between the treatment groups (Table 2 and Figure). The magnitude of the change from baseline was approximately 21% in the basal region compared with 6% in the apical region. The treatment association in the basal regions persisted after adjustment for baseline and the change in the mid and apical values (overall least squares mean [SE] difference, 1.9% [0.7]; 95% CI, 0.5%-3.3%; P = .01) (eTable 1 in the Supplement). Improvement in basal LS was not associated with improvement in N-terminal pro b-type natriuretic peptide levels over 18 months in the cardiac subgroup. We observed a significant interaction between patisiran treatment and baseline basal LS in which patients with the most preserved basal strain at baseline derived the greatest benefit (P < .001) (eTable 2 in the Supplement). In the noncardiac subpopulation (n = 99), there was no significant association with LV regional parameters between the treatment groups.

In this post hoc analysis of the cardiac subpopulation from the phase 3 APOLLO study, we observed that there was a specific pattern of LV LS that was characterized by worse strain in the basal and mid LV with comparative sparing of the apex at baseline. Furthermore, treatment with patisiran for up to 18 months resulted in improvement compared with placebo in LV GLS, which was primarily driven by improvement in the basal region.

This study showed the specific regional LV strain pattern with comparative sparing of the apex, consistent with the previous studies.^9,10 A basal-to-apical gradient in LV LS is well described in cardiac amyloidosis and has been postulated to be due to regional differences in comparative amounts of amyloid deposition.^10,12-14 In this study, LV longitudinal strain in patients with hATTR amyloidosis with relatively mild cardiac involvement demonstrated deterioration in the placebo group over 18 months, predominantly in the basal segments, and this deterioration was attenuated by patisiran. Indeed, the placebo group showed the deterioration in LV basal LS values by 1.4% at 18 months, whereas the patisiran group showed an attenuation of worsening in regional strain by 0.7%. We further found that the treatment association of patisiran with the basal segments was most profound in patients with larger absolute basal LS values (better systolic function) at baseline. This hypothesis-generating finding suggests that the greatest potential benefit of disease-modifying therapy might be for those patients whose cardiac function has not deteriorated too greatly, a finding that will need to be confirmed prospectively. The fact that the comparative improvement in LV GLS due to patisiran was driven by the basal segments suggests that this region might be affected more than other regions by disease-modifying therapies and that basal regional LS may be a more sensitive marker to assess responses to treatment in cardiac amyloidosis.

There are several limitations in our study. First, the cardiac subpopulation was defined by LV wall thickness rather than a more definitive confirmation of cardiac involvement. Second, the quantitative assessment of the magnitude of LV regional strain is limited because of a lack of reference values, vendor differences, and suboptimal reproducibility.¹⁵ Third, this trial did not include patients who were New York Heart Association class III or IV at baseline, and whether these findings are applicable to those patients in more advanced stages of disease is unknown.⁶

Treatment with the RNA interference therapeutic patisiran prevented progressive deterioration in LV GLS, which was driven primarily by attenuating disease progression in the basal region, suggesting that this region may be differentially affected by disease-modifying therapies and that basal LS may be a more sensitive marker to evaluate the treatment effects of cardiac amyloid therapies.

Accepted for Publication: February 27, 2019.

Corresponding Author: Scott D. Solomon, MD, Cardiovascular Division, Brigham and Women’s Hospital, Harvard Medical School, 75 Francis St, Boston, MA 02115 (ssolomon@bwh.harvard.edu).

Published Online: March 16, 2019. doi:10.1001/jamacardio.2019.0849

Open Access: This article is published under the JN-OA license and is free to read on the day of publication.

Author Contributions: Drs Minamisawa and Solomon had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Kristen, Falk, Karsten, Chen, Vest, Solomon.

Acquisition, analysis, or interpretation of data: Minamisawa, Claggett, Adams, Kristen, Merlini, Slama, Dispenzieri, Shah, Karsten, Sweetser, Chen, Riese, Vest, Solomon.

Drafting of the manuscript: Minamisawa, Claggett.

Critical revision of the manuscript for important intellectual content: Claggett, Adams, Kristen, Merlini, Slama, Dispenzieri, Shah, Falk, Karsten, Chen, Riese, Vest, Solomon.

Statistical analysis: Minamisawa, Claggett, Dispenzieri, Chen, Solomon.

Obtained funding: Solomon.

Administrative, technical, or material support: Shah, Falk, Riese.

Supervision: Merlini, Shah, Riese, Solomon.

Other - Clinical Trial Oversight: Vest.

Conflict of Interest Disclosures: Dr Minamisawa reported receiving support from the Japanese Circulation Society, the Japanese Society of Echocardiography, and the Uehara Memorial Foundation Overseas Research Fellowship. Dr Adams reported personal fees from Alnylam outside the submitted work. Dr Slama reported personal fees from Alnylam during the conduct of the study and personal fees from Pfizer and grants from Ionis outside the submitted work. Dr Dispenzieri reported grants from Alnylam and Pfizer during the conduct of the study and grants from Celgene, Takeda, and Prothena outside the submitted work and serves on the advisory board for Janssen. Dr Shah reported research support from Alnylam during the conduct of the study and personal fees from Philips Ultrasound and Bellerophon and research support from Novartis outside the submitted work. Dr Falk reported personal fees from Alnylam and grants and personal fees from Akcea outside the submitted work. Dr Sweetser reported being an employee of Alnylam and owning stock and stock options. Drs Chen, Karsten, Vest, and Riese reported being employees of Alnylam. Dr Solomon reported grants from Alnylam during the conduct of the study and grants from Alnylam, Amgen, AstraZeneca, Bellerophon, Bayer, BMS, Celladon, Cytokinetics, Eidos, Gilead, GSK, Ionis, Lone Star Heart, Mesoblast, MyoKardia, National Institutes of Health/National Heart, Lung, and Blood Institute, Novartis, Sanofi Pasteur, and Theracos and personal fees from Akros, Alnylam, Amgen, AstraZeneca, Bayer, BMS, Cardior, Corvia, Cytokinetics, Gilead, GSK, Ironwood, Merck, Myokardia, Novartis, Roche, Takeda, Theracos, Quantum Genetics, Cardurion, AoBiome, Janssen, Cardiac Dimensions, and Tenaya outside the submitted work. No other disclosures were reported.

Funding/Support: This study was funded by Alnylam Pharmaceuticals.

Role of the Funder/Sponsor: Alnylam Pharmaceuticals was involved in the design and conduct of the APOLLO study; they had no role in the design and conduct of the secondary analysis or the collection, management, analysis, and interpretation of the data. Named authors from Alnylam were involved in the preparation, review, or approval of the manuscript. The decision to submit the manuscript for publication was solely that of the named authors.

Meeting Presentation: This study was presented at the American College of Cardiology 2019 Annual Meeting; March 16, 2019; New Orleans, Louisiana.

Additional Contributions: We thank all the participants (patients, caregivers, and staff members) of the APOLLO trial for providing invaluable contributions. These individuals were not compensated for their contributions.