Transthyretin Stabilizers, Silencers, Gene Editing and Amyloid Depleters in Transthyretin Amyloidosis: A Systematic Review and Meta-Analysis of Randomized and Pivotal Clinical Trials

Arndt Rolfs^1,2,*, Christian Tanislav³, Ameen Al-Refai⁴, Volha Skrahina⁵

¹Medical Faculty, University of Rostock, 18057 Rostock, Germany

²RCV GmbH, Institute for Rare Disease Diagnostics, Leibnizstrasse 58, 10629 Berlin, Germany

³Department of Geriatrics and Neurology, Diakonie Klinikum Jung-Stilling, 57074 Siegen, Germany

⁴Tahleel Consultation Company LLC, Dammam, al Nakhil dist. 32244, Saudi Arabia

⁵RareGenetix GmbH, Dianastrasse 7, 16515 Oranienburg, Germany

^*Corresponding author: Arndt Rolfs, Medical Faculty, University of Rostock, 18057 Rostock, Germany.

Received: 01 October 2026; Accepted: 09 October 2026; Published: 19 October 2026

Introduction

Transthyretin (TTR) is a 127-amino-acid homotetrameric plasma protein synthesized predominantly by the liver that transports thyroxine and retinol-binding protein. Dissociation of the tetramer into monomers is the rate-limiting step of amyloidogenesis; misfolded monomers and oligomers aggregate into fibrils that deposit in peripheral nerves, myocardium and other tissues [1,2]. Two disease forms are distinguished. Hereditary (variant) ATTR (ATTRv) is an autosomal dominant disorder caused by more than 130 pathogenic TTR variants with variable penetrance and a genotype-dependent spectrum ranging from length-dependent sensorimotor and autonomic polyneuropathy (ATTRv-PN) to cardiomyopathy (ATTR-CM) and mixed phenotypes. Wild-type ATTR (ATTRwt) is an age-related cardiomyopathy that predominantly affects men older than 70 years [1,3]. The global prevalence of ATTRv-PN was estimated at approximately 10,000 patients [4]. ATTR-CM is substantially more frequent: ATTRwt was identified in 13% of patients hospitalized for heart failure with preserved ejection fraction and increased left ventricular wall thickness [5], and the p.Val142Ile (V122I) variant is carried by 3–4% of individuals of African ancestry [6]. In the multicentre TRAM study conducted in Germany, Austria and Switzerland, 1.1% of 5,141 patients with polyneuropathy and/or cardiomyopathy of no obvious aetiology carried a pathogenic TTR variant [7]; the successor programme has characterized the first 3,167 participants [8]. Untreated ATTR-CM has a median survival of approximately 2–6 years from diagnosis, stratified by NT-proBNP and estimated glomerular filtration rate [9,10]. Therapeutic development has progressed through four mechanistic stages. Orthotopic liver transplantation, introduced in 1990, removes the source of variant TTR and was the first disease-modifying option for early-onset p.Val50Met (V30M) disease [11]. Kinetic stabilizers (tafamidis, diflunisal, acoramidis) bind the thyroxine-binding sites of the tetramer and reduce dissociation; diflunisal, long used off-label, obtained an EU marketing authorization as Attrogy on 17 July 2025 for stage 1 and stage 2 ATTRv-PN [12]. Silencers reduce hepatic TTR synthesis by RNA interference (patisiran, vutrisiran) or antisense oligonucleotides (inotersen, eplontersen). Investigational approaches comprise next-generation silencers with extended dosing intervals, single-dose in vivo CRISPR-Cas9 editing of TTR, and monoclonal antibodies that target misfolded TTR to promote phagocytic clearance of deposited amyloid [1,13,14]. Existing fibrils additionally act as templates that accelerate further fibril formation (amyloid seeding), which provides a rationale for combining substrate reduction with deposit removal [15]. Approved agents are summarized in Table 1.

Three developments require an updated quantitative synthesis. The CARDIO-TTRansform trial of eplontersen in ATTR-CM, the largest trial in the field (n = 1,432), did not meet its primary endpoint, and a prespecified analysis showed effect modification by background stabilizer therapy [16,17]. Diflunisal, previously used off-label, became the sixth agent licensed for ATTRv-PN in the European Union, an indication extension within the stabilizer class rather than a new mechanism [12,18]. In parallel, amyloid-depleter antibodies and CRISPR-based editing have entered phase 3. The objectives of this review were (i) to pool efficacy and safety data from randomized and pivotal trials across phenotypes, (ii) to compare stabilizer and silencer effects and quantify the interaction between silencers and background stabilizers, and (iii) to provide a structured overview of approved and investigational therapies including phenotype, genotype, population, sample size, endpoints and trial duration.

Table 1: Approved TTR-directed therapies (status 30 September 2026).

Note: Registration numbers are given in Tables 2, 3 and 6. Abbreviations: CM, cardiomyopathy; GalNAc, N-acetylgalactosamine; IV, intravenous; MOE, methoxyethyl; NSAID, non-steroidal anti-inflammatory drug; PN, polyneuropathy; SC, subcutaneous; siRNA, small interfering RNA; v, variant; wt, wild-type. Approval dates refer to first authorization for the respective indication.

2. Methods

2.1 Protocol and reporting

The review was conducted and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement [32]. Eligibility criteria, outcomes and analysis methods were defined before data extraction. The protocol was not registered.

2.2 Information sources and search strategy

PubMed/MEDLINE, Embase and the Cochrane Central Register of Controlled Trials were searched from inception to 30 September 2026 without language restriction. ClinicalTrials.gov, the EU Clinical Trials Register/Clinical Trials Information System, European Public Assessment Reports, US prescribing information and abstracts of the European Society of Cardiology, American Heart Association, American College of Cardiology, Heart Failure Association, International Society of Amyloidosis and Peripheral Nerve Society meetings (2018–2026) were screened. The search combined disease terms, class terms and substance names, the last including discontinued and terminated agents: ("transthyretin" OR "ATTR" OR "familial amyloid polyneuropathy") AND ("transthyretin stabili*" OR "gene silenc*" OR "antisense" OR "RNA interference" OR "siRNA" OR "gene editing" OR "amyloid deplet*" OR "monoclonal antibody" OR "tafamidis" OR "acoramidis" OR "diflunisal" OR "doxycycline" OR "tauroursodeoxycholic" OR "tolcapone" OR "patisiran" OR "vutrisiran" OR "revusiran" OR "ALN-TTRSC" OR "inotersen" OR "eplontersen" OR "nucresiran" OR "nexiguran ziclumeran" OR "NTLA-2001" OR "cliramitug" OR "ALXN2220" OR "NI006" OR "coramitug" OR "NNC6019" OR "PRX004" OR "AT-02") AND ("randomized" OR "randomised" OR "trial"). The class terms were added specifically to capture agents whose development has been discontinued and which are therefore unlikely to be retrieved by substance name alone. Database-specific strings including MeSH and Emtree terms, search dates and per-database yields are provided as Supplementary Table S1. Reference lists of included trials and reviews were hand-searched, and the complete ClinicalTrials.gov set of interventional phase 3 studies in ATTR amyloidosis was screened against the eligibility criteria.

2.3 Eligibility criteria

Population: adults with ATTRv-PN, ATTR-CM (wild-type or variant) or mixed phenotype. Interventions: any TTR-directed therapy (stabilizer, silencer, gene editing, amyloid depleter). Comparator: placebo, either concurrent or external (placebo arm of a preceding randomized trial of the same sponsor, as accepted by regulatory authorities). Outcomes: as defined in Section 2.5.

Designs: phase 3 randomized controlled trials (RCTs) and pivotal externally controlled phase 3 trials for quantitative synthesis; phase 1-2 trials of novel classes for narrative synthesis. Observational studies, uncontrolled open-label extensions (used descriptively only) and trials of supportive heart-failure therapy were excluded. Trials of agents whose development has been discontinued were eligible and were assessed individually; reasons for exclusion from quantitative synthesis are stated in Section 3.1.

2.4 Study selection and data extraction

Records were screened by title and abstract, followed by full-text assessment. Data were extracted into a standardized form and verified against primary publications, supplementary appendices, regulatory documents and registry entries: design, randomization ratio, sample size, genotype distribution, phenotype and key inclusion criteria, New York Heart Association (NYHA) class, background stabilizer use, treatment duration, primary and secondary endpoints, effect estimates with 95% confidence intervals (CI) and adverse events.

2.5 Outcomes

For ATTR-CM, the primary outcome was all-cause mortality (ACM). Secondary outcomes were the composite of death and cardiovascular (CV) events as defined in each trial, change in 6-minute walk distance (6MWD), change in Kansas City Cardiomyopathy Questionnaire overall summary score (KCCQ-OS) and safety. For ATTRv-PN, the primary outcome was change in the modified Neuropathy Impairment Score +7 (mNIS+7); the secondary outcome was change in Norfolk Quality of Life–Diabetic Neuropathy (Norfolk QOL-DN) score. Estimates were taken at the longest prespecified time point at which a randomized comparison was reported. For HELIOS-B this is the 42-month analysis of ACM, which includes vital status after transition to the open-label extension and is therefore randomized but not fully double-blind; the 36-month double-blind analysis was used in a sensitivity analysis [28]. For CARDIO-TTRansform, ACM through week 140 is reported separately for patients with and without baseline stabilizer therapy [17]. Because these are strata of one trial and differ systematically in age, disease duration and stage, they were combined into a single trial-level estimate by inverse-variance weighting for the primary analysis; the stratum-level analysis is reported separately in Section 3.5. Each trial therefore contributes one unit to each pooled estimate.

2.6 Risk of bias and certainty of evidence

Risk of bias was assessed with the Cochrane RoB 2 tool for RCTs [33] and with ROBINS-I for externally controlled trials [34]. Certainty of evidence per outcome was graded with the GRADE approach [35].

2.7 Statistical analysis

Hazard ratios (HR) and rate ratios (RR) were log-transformed; standard errors were derived from reported 95% CI. For ENDEAVOUR, mortality was entered as a risk ratio calculated from the crude on-treatment death proportions (18/140 versus 2/66), with its standard error derived from the event counts; this estimate was used only in the sensitivity analysis, in which it was pooled with the hazard and rate ratios of the other trials. Mean differences were pooled on the original scale; where only an exact P value was reported, the standard error was derived from the corresponding normal deviate. Random-effects models with the Paule–Mandel estimator of between-trial variance were used, because the DerSimonian–Laird estimator [36] underestimates between-trial variance when few trials are pooled, whereas the Paule–Mandel estimator [37] performs better in this setting [38]. Heterogeneity was quantified with Cochran Q, I² and τ² [39]. For outcomes with three or more trials, the Hartung–Knapp–Sidik–Jonkman correction, in its modified form with the variance scaling factor truncated at 1, was applied as a prespecified sensitivity analysis and is reported alongside the primary interval [40,41]; at two trials the correction is not interpretable and was not applied. Prespecified subgroup comparisons were therapeutic class (stabilizer versus silencer) and, for silencers, baseline stabilizer use (monotherapy versus add-on); differences were tested with a z-test on pooled log estimates, which is low-powered with two trials per group and is therefore reported alongside the within-trial interaction test of CARDIO-TTRansform. Three limitations of the pooled estimates were handled explicitly. First, the composite outcome combines a hazard ratio for time to first event (ATTRibute-CM) with rate ratios for recurrent events (HELIOS-B, CARDIO-TTRansform); these estimate different parameters, so an analysis restricted to the recurrent-event rate ratios is reported. Second, HELIOS-A and NEURO-TTRansform used the placebo arms of APOLLO and NEURO-TTR as external controls, so these control participants would contribute twice to a pool containing all four trials; the primary neuropathy analysis is therefore restricted to the two concurrently randomized trials, and the four-trial estimate is reported as a secondary analysis with this dependency stated. Third, the mNIS+7 versions differ between sponsors (range 0–304 points in Alnylam trials; 0–346.3 points in Ionis trials), and NIS+7 as used in the diflunisal trial is a further, non-equivalent scale; cross-scale comparisons are descriptive only. Further sensitivity analyses comprised fixed-effect models, leave-one-out analyses and inclusion of the terminated ENDEAVOUR trial. Funnel-plot assessment was not performed (fewer than 10 trials per outcome); publication and reporting bias were instead assessed qualitatively within GRADE on the basis of trial registration, availability of results for terminated programmes and sponsor involvement. Analyses were performed in Python 3.11 (NumPy 1.26, SciPy 1.11) using standard formulae; the analysis script is provided as Supplementary File S2. Two-sided P < 0.05 defined statistical significance.

3. Results

3.1 Study selection and characteristics

Database searching, registry screening and hand-searching yielded 1,097 records; after removal of 211 duplicates, 886 records were screened by title and abstract and 839 were excluded, 47 full texts were assessed and 32 were excluded with reasons (PRISMA 2020 flow diagram, Supplementary Figure S1; per-database yields and search dates, Supplementary Table S1). Fifteen trials met the inclusion criteria: six pivotal trials in ATTR-CM (3,726 participants) [16,20,21,26,28,42], six pivotal trials in ATTRv-PN (987 participants) [19,23,25,27,29,31] and three early-phase trials of novel classes (180 participants) [14,43,44]. Characteristics of the pivotal trials and of the two early-phase amyloid-depleter trials are shown in Tables 2 and 3; the phase 1 trial of nexiguran ziclumeran [44] is summarized in Section 3.9 and Table 6. One eligible trial was identified that is absent from previous syntheses in this field. ENDEAVOUR (NCT02319005) randomized 206 patients with hereditary ATTR-CM 2:1 to the GalNAc-conjugated siRNA revusiran (n = 140) or placebo (n = 66) and was terminated after a median of 6.71 months because of a mortality imbalance: 18 of 140 patients (12.9%) on revusiran and 2 of 66 (3.0%) on placebo died during the on-treatment period, corresponding to a risk ratio of 4.24 (95% CI 1.01–17.75); no causative mechanism was identified and the programme was discontinued [42]. ENDEAVOUR meets every stated eligibility criterion and is therefore reported in Table 2 and in the flow diagram. It was excluded from the primary quantitative synthesis on three grounds: median exposure of 6.71 months is incompatible with the 30- to 42-month horizon of the other trials; termination was event-driven, so the mortality estimate is conditioned on the stopping rule; and revusiran is not a current or candidate therapy. The exclusion is nevertheless not neutral, and the estimates with ENDEAVOUR included are reported as a sensitivity analysis in Section 3.3. A phase 3 trial of doxycycline plus tauroursodeoxycholic acid (NCT03481972, n = 102) was excluded because the combination is not a TTR-directed therapy within the definition of Section 2.3 and no peer-reviewed report of results was available. ATTR-CM populations were older (mean age 72–77 years), predominantly male (≥90%) and, with the exception of ENDEAVOUR, predominantly wild-type (approximately 76% in ATTR-ACT to 90% in ATTRibute-CM). Background stabilizer use increased across trial eras: none in ATTR-ACT, drop-in permitted after month 12 in ATTRibute-CM, 25% in APOLLO-B, 40% in HELIOS-B (capped by protocol) and 57% in CARDIO-TTRansform, where a further 57% of patients without baseline stabilizer started one during follow-up [17]. Placebo-arm ACM declined from 42.9% over 30 months in ATTR-ACT to 25.7% over 30 months in ATTRibute-CM [20,21]. ATTRv-PN trials enrolled patients with Coutinho stage 1–2 disease; the proportion carrying p.Val50Met ranged from approximately 43% (APOLLO) to 100% (Fx-005).

Table 2: Trials in ATTR cardiomyopathy: design, population, genotype, endpoints and results. Rows 1–6 are pivotal phase 3 trials; rows 7–8 are early-phase trials reported narratively.

Note: Values in parentheses are 95% CI. ENDEAVOUR is reported for completeness and is excluded from the primary quantitative synthesis (see Section 3.1). ACM, all-cause mortality; BID, twice daily; CV, cardiovascular; CVH, cardiovascular hospitalization; ECV, extracellular volume; eGFR, estimated glomerular filtration rate; IVS, interventricular septum; KCCQ-OS, Kansas City Cardiomyopathy Questionnaire overall summary (points); mITT, modified intention-to-treat; q3w/q4w/q12w, every 3/4/12 weeks; 6MWD, 6-minute walk distance; v, variant; wt, wild-type.

3.2 Risk of bias

Risk-of-bias judgements are shown in Table 4. Two of the four cardiomyopathy trials contributing to the primary outcome were rated at some concerns because of stabilizer drop-in during follow-up, and the overall judgement follows the RoB 2 algorithm, under which any domain at some concerns precludes an overall rating of low. The diflunisal trial was at high risk of bias because of attrition. ENDEAVOUR was rated at some concerns because termination for a mortality imbalance makes the reported estimate conditional on the stopping rule. Both externally controlled trials (HELIOS-A, NEURO-TTRansform) were at serious risk of bias on ROBINS-I, driven by confounding from the use of a historical placebo population and by open-label outcome assessment.

3.3 All-cause mortality in ATTR-CM

Four phase 3 trials contributed one unit each (3,160 participants). Active TTR-directed therapy reduced ACM (pooled HR 0.73 (95% CI 0.62–0.85); P < 0.001) without detectable heterogeneity (Q = 0.76, P = 0.86, I² = 0%, τ² = 0) (Figure 1A). The Hartung–Knapp–Sidik–Jonkman interval was 0.56–0.93. Stabilizers (ATTR-ACT, ATTRibute-CM; 0.73 (95% CI 0.58–0.92)) and silencers (HELIOS-B, CARDIO-TTRansform; 0.72 (95% CI 0.59–0.89)) yielded equivalent estimates (P for subgroup difference = 0.94). In leave-one-out analyses the pooled estimate ranged from 0.70 to 0.75, and all four estimates excluded unity. Long-term extension of ATTR-ACT showed persistent survival benefit with continuous versus delayed tafamidis [45]. Three features of this estimate limit its interpretation. First, including the terminated ENDEAVOUR trial, in which revusiran was associated with excess mortality, moves the pooled estimate to 0.82 (95% CI 0.52–1.30) with I² = 39% and P = 0.40; for the silencer subgroup alone the estimate becomes 1.06 (95% CI 0.37–3.03) with I² = 69%. The grounds for excluding ENDEAVOUR from the primary analysis are stated in Section 3.1, but the sensitivity analysis shows that the consistency of the primary estimate rests on that exclusion. Second, the HELIOS-B 42-month estimate includes vital status after transition to open-label vutrisiran; with the 36-month double-blind estimate (HR 0.69 (95% CI 0.49–0.98)) in place of the 42-month estimate, the pooled HR is 0.74 (95% CI 0.63–0.86; I² = 0%; Hartung–Knapp–Sidik–Jonkman interval 0.57–0.95) [28]. Third, the CARDIO-TTRansform contribution is the inverse-variance combination of its two randomized strata (HR 0.84, 0.60–1.18 without baseline stabilizer and 0.68, 0.46–1.01 with baseline stabilizer; combined 0.77 (95% CI 0.59–0.99)) [17].

Figure 1: Random-effects meta-analyses in ATTR cardiomyopathy. (A) All-cause mortality; one unit per trial, CARDIO-TTRansform entered as the inverse-variance combination of its two baseline-stabilizer strata at week 140, HELIOS-B at 42 months; the open diamond is the sensitivity analysis including ENDEAVOUR. (B) Composite of death and cardiovascular events as defined in each trial. Squares are proportional to weight; horizontal lines are 95% CI; diamonds are pooled estimates.

3.4 Composite outcomes, cardiovascular events, functional capacity and quality of life

The composite of death and CV events (three trials, 2,719 participants) was reduced (0.75 (95% CI 0.63–0.89); I² = 44%; Hartung–Knapp–Sidik–Jonkman interval 0.51–1.09) (Figure 1B). The three estimates are not equivalent measures: ATTRibute-CM reports a hazard ratio for time to first event, whereas HELIOS-B and CARDIO-TTRansform report rate ratios for recurrent events. Restricted to the two recurrent-event rate ratios the pooled estimate was 0.80 (95% CI 0.68–0.95) (I² = 17%). CV hospitalization was reduced by tafamidis (RR 0.68, 0.56–0.81) and acoramidis (first CVH HR 0.60, 0.45–0.80; annualized frequency 0.22 vs 0.45) [20,22]. Across five trials, 6MWD decline was attenuated by 34.5 m (95% CI 13.2–55.7; I² = 88%) and KCCQ-OS decline by 7.1 points (3.2–10.9; I² = 84%); both intervals widen substantially under the Hartung–Knapp–Sidik–Jonkman correction (6MWD 4.4–64.6 m; KCCQ-OS 1.6–12.5 points), which is the appropriate reading given the heterogeneity. That heterogeneity reflects follow-up duration (12 months in APOLLO-B versus 30 months in ATTR-ACT) and background stabilizer use: in CARDIO-TTRansform, 6MWD differed by 29.9 m (14.3–45.5) without and by 12.0 m (−1.0 to 25.0) with baseline stabilizer [17].

3.5 Silencers with and without background stabilizer

Two outcome trials reported silencer effects by baseline stabilizer use (Figure 2A). Without baseline stabilizer, the pooled estimate for the primary composite was 0.69 (95% CI 0.56–0.85) (I² = 0%). With baseline stabilizer (tafamidis in all cases), the pooled estimate was 0.98 (95% CI 0.69–1.40) (I² = 47%). Each subgroup contains two trials, so no small-sample correction was applied and the across-trial test is low-powered: P = 0.09. The within-trial interaction test of CARDIO-TTRansform, which is the better-powered comparison, gave P = 0.017 across 1,432 patients [17]. Serum TTR on eplontersen at week 140 was similar in both strata (median 6.5 and 7.1 mg/dL), which places the effect modification downstream of TTR lowering [17]. In HELIOS-B, the composite of CV death and recurrent CV events in patients on baseline tafamidis was RR 0.78 (0.50–1.22) [28]; the slightly different value in Table 2 (HR 0.79, 0.51–1.21) is the corresponding estimate for the trial primary composite of all-cause mortality and recurrent CV events [28].

The mortality strata point in the opposite direction and this must be stated. For ACM at week 140, CARDIO-TTRansform reported HR 0.84 (0.60–1.18) without and 0.68 (0.46–1.01) with baseline stabilizer — that is, the more favourable estimate in the stratum in which no benefit was seen on the primary composite [17]. The two results are not contradictory in a strict sense: they are different endpoints with different event counts, both stratum estimates include unity, and the authors of the primary analysis attribute the pattern to patients with advanced cardiac involvement surviving longer while remaining susceptible to recurrent events, as previously observed with tafamidis in NYHA class III [17,20]. The conclusion of this review that silencer benefit is concentrated in patients without a stabilizer therefore applies to the composite and functional endpoints, for which the interaction was formally tested, and not to all-cause mortality, for which no interaction was demonstrated (P = 0.44) [17]. Combination therapy in the reverse sequence (stabilizer added to an established silencer) has not been tested in randomized trials.

Figure 2: (A) TTR silencers in ATTR cardiomyopathy by baseline stabilizer use; primary composite endpoints (HELIOS-B: all-cause mortality and recurrent CV events; CARDIO-TTRansform: CV mortality and recurrent CV events). Open diamond: silencer added to stabilizer. (B) Change in mNIS+7 versus placebo in ATTRv-PN silencer trials. The primary estimate is restricted to the concurrently randomized trials; HELIOS-A and NEURO-TTRansform used the placebo arms of APOLLO and NEURO-TTR as external controls, so those participants contribute twice to the secondary four-trial estimate. HELIOS-A standard error derived from the exact P value.

3.6 Neuropathy outcomes in ATTRv-PN

In the two concurrently randomized silencer trials (APOLLO, NEURO-TTR; 397 participants), the pooled mNIS+7 mean difference versus placebo was −26.9 points (95% CI −41.0 to −12.9; I² = 90%) and the Norfolk QOL-DN difference −16.5 points (−25.7 to −7.3; I² = 76%) (Figure 2B). Adding the two externally controlled trials gives −26.9 points (−32.8 to −21.0; I² = 72%) and −18.5 points (−22.8 to −14.2; I² = 46%). The point estimates are almost unchanged, but the four-trial intervals are narrower than the data support, because the external control arms are the placebo arms of the two concurrently randomized trials and therefore contribute twice; the four-trial estimates are reported for completeness and are not used for inference. Improvement of mNIS+7 relative to baseline occurred in 56% of patisiran-treated and 48% of vutrisiran-treated patients versus 4% on placebo [25,27]. Stabilizers were tested only in early-stage disease and on a different scale, so no numerical comparison with the silencer estimates is possible. Tafamidis did not meet the co-primary endpoint in the intention-to-treat population of Fx-005 [19]. Diflunisal reduced progression of NIS+7 — a scale that is not equivalent to mNIS+7 — by an adjusted least-squares mean difference of 16.3 points over 24 months (observed means +8.2 versus +26.3 points), with concordant benefit on NIS, NIS-LL, the Kumamoto score and the SF-36 physical component [23]. This trial supported the EU marketing authorization of diflunisal as Attrogy on 17 July 2025 for stage 1 and stage 2 ATTRv-PN [12,24]. A network meta-analysis of pharmacological interventions in ATTRv-PN ranked both stabilizers below the silencers and did not separate diflunisal from tafamidis [46].

Table 3: Pivotal trials in hereditary ATTR polyneuropathy: design, population, genotype, endpoints and results.

Note: Values in parentheses are 95% CI; mean differences are least-squares differences versus placebo (negative values favour active treatment). mNIS+7, modified Neuropathy Impairment Score +7; NIS-LL, NIS–lower limbs; Norfolk QOL-DN, Norfolk Quality of Life–Diabetic Neuropathy; ITT, intention-to-treat; NSAID, non-steroidal anti-inflammatory drug.

3.7 Safety

Serious adverse events did not differ from placebo for tafamidis, acoramidis, vutrisiran and eplontersen [16,20,21,28]. Acoramidis caused an early, non-progressive decrease in estimated glomerular filtration rate [21]. Patisiran requires premedication; infusion-related reactions occurred in 19% of APOLLO participants [25]. Inotersen caused thrombocytopenia (3%, including one fatal intracranial haemorrhage) and glomerulonephritis (3%), requiring weekly platelet and regular renal monitoring [29]; NICE withdrew its guidance after the manufacturer ceased marketing in the United Kingdom [47], and in the European Union marketing ceases from 1 December 2026, with withdrawal of the marketing authorization planned from the same date, for commercial reasons and explicitly not because of manufacturing, quality or safety concerns [30]. Eplontersen showed no thrombocytopenia or glomerulonephritis signal [31]. Silencers reduce serum retinol through loss of the TTR–retinol-binding protein complex, and vitamin A supplementation is required by the product information of each agent [18]. As an analgesic and anti-inflammatory agent, diflunisal is given at 500–1,000 mg/day in two divided doses for osteoarthritis and rheumatoid arthritis and, for mild to moderate pain, as a 1,000 mg loading dose followed by 500 mg every 12 hours; maintenance doses above 1,500 mg/day are not recommended [48]. At these doses it carries the established NSAID risks of gastrointestinal bleeding, ulceration and perforation, renal impairment, fluid retention and oedema, and cardiovascular thrombotic events [48]. Attrogy is licensed for ATTRv-PN at 250 mg twice daily (500 mg/day), at the lower end of this dose range [12], and in the pivotal trial adverse events did not differ between the diflunisal and placebo groups [23]. The non-steroidal anti-inflammatory drug class risks (renal dysfunction, fluid retention, bleeding) nevertheless require assessment in patients with cardiac involvement [3,12,23]. The most consequential safety observation in the field remains the terminated ENDEAVOUR trial, in which revusiran was associated with 12.9% versus 3.0% on-treatment mortality without an identified mechanism [42]. In the phase 3 MAGNITUDE programme of nexiguran ziclumeran, grade 4 transaminase elevation with increased bilirubin in one participant led to a clinical hold on 29 October 2025, lifted on 27 January 2026 (MAGNITUDE-2) and 2 March 2026 (MAGNITUDE) with additional exclusion criteria [49]. The amyloid depleters coramitug and cliramitug showed no drug-related serious adverse events in phase 1–2 [14,43].

Table 4: Risk-of-bias assessment: RoB 2 for randomized trials (panel A), ROBINS-I for externally controlled trials (panel B).

3.8 Certainty of evidence

Certainty was moderate for all-cause mortality in ATTR-CM and for the composite and neuropathy outcomes, and low for the functional outcomes (6MWD, KCCQ-OS) and for silencers as monotherapy and added to a background stabilizer (Table 5). The rating for all-cause mortality was downgraded for risk of bias, because two of the four contributing trials are at some concerns, and the publication-bias domain was rated serious for no outcome but was explicitly considered: results of the terminated revusiran programme are published, the registered phase 3 trials in the field have reported, and all included trials were registered before enrolment, but every included trial is sponsor-funded and the primary estimate depends on a defensible exclusion of the one trial with an adverse mortality result.

Table 5: Summary of findings and GRADE certainty.

Note: All ratings start at high (randomized evidence); no upgrading was applied. HR, hazard ratio; for CARDIO-TTRansform and recurrent-event analyses the estimate is a rate ratio; MD, mean difference. Negative mNIS+7 and Norfolk QOL-DN values and positive 6MWD and KCCQ-OS values favour active treatment. The externally controlled trials HELIOS-A and NEURO-TTRansform are not included in the neuropathy rows because their control arms are those of APOLLO and NEURO-TTR (Section 3.6).

3.9 Investigational therapies

Six programmes are in active clinical development (Table 6). Nucresiran, a next-generation GalNAc-siRNA dosed twice yearly, is evaluated in TRITON-CM, an event-driven trial expanded to approximately 1,750 patients, and in TRITON-PN [50,51]. Nexiguran ziclumeran achieved a mean serum TTR reduction of approximately 90% after a single infusion in phase 1, sustained over at least 12 months in ATTR-CM [13,44]. Two depleter antibodies are in phase 3 outcome trials: cliramitug (ALXN2220, formerly NI006) in DepleTTR-CM [52] and coramitug in CLEOPATTRA [53]. In the phase 2 coramitug trial, 84% of participants were on tafamidis; NT-proBNP decreased by 48% versus placebo at 60 mg/kg without a significant 6MWD effect over 52 weeks [43].

Table 6: TTR-directed therapies in active clinical development (phase 3 programmes and the phase 1 pan-amyloid depleter AT-02), and the completed CARDIO-TTRansform programme for reference (registry status 1 October 2026).

Note: ASO, antisense oligonucleotide; ECV, extracellular volume; LNP, lipid nanoparticle; other abbreviations as in Tables 1–3. ClinicalTrials.gov registration numbers are given for every trial in Tables 2, 3 and 6; the NI006 phase 1 trial is identified by its sponsor protocol number. Sample sizes and status are taken from the ClinicalTrials.gov records cited in the respective row, accessed 1 October 2026; where a planned sample size is given by the sponsor rather than the registry, this is stated. The final row is a completed programme included in the quantitative synthesis and is shown here for comparison.

4. Discussion

4.1 Principal findings

This meta-analysis includes all phase 3 outcome trials in ATTR-CM reported through September 2026, including CARDIO-TTRansform and the terminated revusiran trial ENDEAVOUR. Four findings emerge. First, TTR-directed therapy reduces all-cause mortality in ATTR-CM by about 27% among the four current programmes, without detectable heterogeneity; the estimate is attenuated and the consistency is lost when ENDEAVOUR is added, so the result holds for the therapies in current use and not for TTR-directed therapy as an undifferentiated class. Second, stabilizers and silencers produce equivalent mortality effects in indirect comparison. Third, on the composite and functional endpoints the benefit of silencers is concentrated in patients not receiving a stabilizer, while the pooled estimate for silencers added to tafamidis is centred on the null; for all-cause mortality no such interaction was demonstrated. Fourth, in ATTRv-PN, silencers attenuate or reverse neuropathy progression.

4.2 Stabilization versus silencing

Both classes reduce the flux of amyloidogenic monomer: stabilizers by lowering the dissociation rate of the tetramer, silencers by lowering substrate concentration, with placebo-adjusted reductions of approximately 80% or more in the pivotal trials [25,27,31]. Neither class removes established deposits. The near-identity of the pooled mortality estimates for the two classes is consistent with a shared downstream mechanism. No head-to-head trial exists, and differences in trial era, background therapy, disease severity and follow-up preclude a ranking. Across trials, benefit emerged earlier in less advanced disease; in ATTR-ACT, survival curves separated after approximately 18 months and patients in NYHA class III had more CV hospitalizations on tafamidis, whereas the HELIOS-B primary endpoint separated after approximately 6 months in a population with lower NT-proBNP [20,28]. Current guidance recommends tafamidis for ATTR-CM with NYHA class I–II symptoms [54,55]; the subsequent approvals of acoramidis and vutrisiran extend the choice of first-line therapy. Selection between classes in practice follows phenotype (neuropathy favours silencers), route and adherence, monitoring requirements, renal function, cost and access.

4.3 Combination therapy

The central new finding is the interaction between silencer effect and background stabilizer. In CARDIO-TTRansform, eplontersen reduced the primary composite by 29% in patients without a stabilizer and had no effect in the 819 patients on tafamidis, despite similar serum TTR suppression in both strata [17]. HELIOS-B showed a point estimate favouring vutrisiran in the tafamidis stratum with wide confidence intervals [28]. The pooled add-on estimate is centred on unity, which is compatible with a ceiling effect of combined substrate reduction and stabilization, although the across-trial subgroup test at two trials per group has little power and the within-trial interaction test of CARDIO-TTRansform (P = 0.017) carries the inferential weight. Four constraints apply: the evidence is subgroup-based; baseline stabilizer users differed systematically (younger, longer disease duration, less advanced stage); 57% of patients without baseline stabilizer started one during CARDIO-TTRansform, which biases the monotherapy estimate towards the null [17]; and the interaction was demonstrated for the composite and functional endpoints but not for mortality, where the stratum estimates run in the opposite direction (Section 3.5). The data do not support routine addition of a silencer to an established stabilizer in ATTR-CM. Combination remains a rational option in mixed phenotypes with progressive neuropathy, where silencers carry independent evidence.

4.4 Stabilizers in polyneuropathy after the licensing of diflunisal

Diflunisal is the oldest TTR stabilizer with randomized evidence and was used off-label in ATTRv-PN for more than a decade. Its EU marketing authorization in July 2025 is an indication extension within an established class rather than a new mechanism, but it has two practical consequences. Tafamidis is licensed in the European Union for Coutinho stage 1 only, and acoramidis has not been studied in polyneuropathy [18,56]; diflunisal is therefore the only stabilizer with an approved indication covering stage 2. In addition, dosage, product quality, pharmacovigilance and reimbursement are now defined, which removes the practical barrier of prescribing an unapproved analgesic in a progressive orphan disease. Neither point implies comparative efficacy: the German expert consensus continues to recommend a second-generation silencer as first-line therapy for most patients with symptomatic ATTRv-PN and to regard stabilizers as a bridging option with immediate switch on any sign of progression, and an indirect comparison does not separate diflunisal from tafamidis [18,46]. Although Attrogy is given at 250 mg twice daily, at the lower end of the analgesic and anti-inflammatory dose range of diflunisal [12,48], and adverse events did not differ from placebo in the pivotal trial [23], the NSAID class risks — renal dysfunction, fluid retention and gastrointestinal bleeding — require assessment and monitoring, particularly in patients with cardiac involvement [12,23]. A second setting is patients who progress despite adequate TTR suppression by a silencer. Adding a stabilizer is mechanistically plausible, because tetramer dissociation is concentration-dependent and tetramers become more dissociation-prone at the low TTR concentrations produced by silencing [57]; under silencer therapy the drug-to-target ratio therefore shifts in favour of the stabilizer. Retrospective cohort data report that a substantial proportion of patients receive stabilizer plus silencer without an identified safety signal [58], and the German consensus considers combination justified in severely progressive cases after multidisciplinary review [18]. This sequence is the reverse of the one tested in CARDIO-TTRansform, where a silencer was added to patients already on a stabilizer, and the negative result there does not transfer to it. No randomized trial has tested stabilizer add-on in either phenotype. A placebo-controlled add-on trial in patients with ATTRv-PN progressing under a second-generation silencer, with mNIS+7 and Norfolk QOL-DN as endpoints and prespecified renal and gastrointestinal monitoring, would close this gap; any of the licensed stabilizers could serve, with the approved stage-2 indication favouring diflunisal. Until such data exist, combination remains an individual decision taken in a specialized centre [3,18].

4.5 Declining event rates and trial design

Placebo-arm mortality fell from 42.9% (ATTR-ACT) to 25.7% (ATTRibute-CM) over 30 months [20,21]. Contributing factors are earlier diagnosis by bone scintigraphy, lower NT-proBNP at enrolment, the entry of sodium–glucose cotransporter 2 inhibitors into heart failure with preserved ejection fraction [59] and background disease-modifying therapy. The consequences are larger and longer trials (CARDIO-TTRansform 1,432 patients over 140 weeks; TRITON-CM expanded to approximately 1,750 patients) [16,51], recurrent-event and hierarchical endpoints, and growing interest in validated surrogates (serum TTR, NT-proBNP, extracellular volume, scintigraphic uptake). Where an approved disease-modifying therapy is reimbursed, placebo-controlled monotherapy designs are increasingly difficult to justify, and future trials of novel agents will test add-on effects on top of an active standard of care, for which the CARDIO-TTRansform result sets a demanding benchmark.

4.6 Genotype and phenotype

ATTR-CM trials enrolled predominantly ATTRwt; ATTRv subgroups were small, and p.Val142Ile carriers of African ancestry are under-represented relative to the carrier frequency in that population, which limits genotype-specific inference [6,21]. In ATTRibute-CM, the treatment effect on the composite of death or first CV hospitalization was consistent across subgroups including genotype [22]. ATTRv-PN trials were dominated by p.Val50Met; late-onset and non-p.Val50Met genotypes with mixed cardiac and neurological involvement are the populations in which silencers offer dual-organ benefit. Genetic testing is indicated in every patient with ATTR-CM [3]; it identifies ATTRv, enables cascade testing of relatives and defines candidates for pre-symptomatic surveillance and for the ACT-EARLY prevention trial, whose primary endpoint is the time to onset of either cardiomyopathy or polyneuropathy [60]. Screening programmes in unexplained polyneuropathy and cardiomyopathy yield a diagnostic rate of approximately 1% [7,8], which supports systematic genetic testing in these populations.

4.7 Novel modalities

In vivo CRISPR-Cas9 editing offers permanent TTR reduction after a single dose [13,44]. Irreversibility, the hepatotoxicity signal in MAGNITUDE and the need for long-term surveillance of off-target effects define the benefit–risk evaluation [49]. Amyloid depleters address the component of disease that no approved therapy targets. Phase 1 data for cliramitug and phase 2 data for coramitug show biomarker and imaging effects consistent with deposit removal, without functional benefit at 52 weeks [14,43]. Outcome data from DepleTTR-CM and CLEOPATTRA will determine whether deposit removal adds clinical benefit to substrate control. Given the CARDIO-TTRansform result, the depleter class represents the most plausible route to incremental benefit in patients already on a stabilizer.

4.8 Limitations

The analysis is based on aggregate data; individual participant data were not available, and extracted estimates were verified against primary publications and supplementary appendices but not recomputed from source data. Composite endpoints, follow-up durations and analysis methods differ between trials, and the composite pool mixes a first-event hazard ratio with recurrent-event rate ratios. The HELIOS-B 42-month mortality estimate includes vital status after transition to open-label treatment. The primary neuropathy estimate rests on two trials with I² of 90%, and the four-trial estimate contains dependent control arms. The exclusion of ENDEAVOUR is defensible but consequential, as Section 3.3 shows. With four or fewer trials per outcome, between-trial variance is estimated imprecisely whichever estimator is used, and the Hartung–Knapp–Sidik–Jonkman intervals should be read as the conservative bound. Study selection and data extraction were performed by a single reviewer; no second reviewer independently screened or extracted, which is a departure from standard systematic review methodology, and the protocol was not registered. Funnel-plot analysis was not feasible and publication bias was assessed qualitatively only. Information on pipeline programmes derives from registries and sponsor communications and was not independently verified beyond the cited records.

5. Conclusions

Among the therapies in current use, TTR stabilizers and silencers reduce all-cause mortality in ATTR cardiomyopathy by approximately one quarter, with equivalent effects between classes; the estimate is attenuated when the terminated revusiran trial is included. Silencers are effective as monotherapy; added to an established stabilizer, incremental benefit on composite and functional endpoints is not demonstrated. In hereditary polyneuropathy, silencers attenuate or reverse neuropathy progression and remain first-line; the European licensing of diflunisal extends the approved indications of the stabilizer class to Coutinho stage 2. Stabilizer add-on in patients progressing under a silencer is untested and warrants a randomized trial. CRISPR-based editing and amyloid depleters are investigational; phase 3 outcome trials will report from 2027 onwards.

Declarations

Acknowledgements: None.

Funding: This work received no specific funding.

Conflicts of interest: A.R. is managing director of RCV GmbH (Arcensus) and RareGenetix GmbH, co-founder and chief executive officer of subrosa OmicsAI AG, founder and former chief executive officer of CENTOGENE, and investigator of the TRAM studies. A.R. has received no honoraria, consulting fees or research support from Purpose Pharma. V.S. is an employee of RareGenetix GmbH and investigator of the TRAM studies. A.A.-R. is affiliated with Tahleel Consultation Company LLC, Dammam, Saudi Arabia. C.T. declares no conflicts of interest. No author has any advisory relationship with Purpose Pharma, the marketing authorization holder of diflunisal.

Author contributions: A.R. conceived the review, performed the search, data extraction and analyses, and wrote the manuscript. C.T. and A.A.-R. contributed to data interpretation and critical revision. V.S. supervised the work and critically revised the manuscript. All authors approved the submitted version.

Ethics approval and consent: Not applicable (analysis of published aggregate data).

Data availability: All data are derived from published sources cited in this article. The analysis script that reproduces every pooled estimate, heterogeneity statistic and sensitivity analysis reported here is provided as Supplementary File S2; the data extraction sheet is available from the corresponding author on request.

Supplementary material: Supplementary Table S1, database-specific search strings, search dates and per-database yields; Supplementary Figure S1, PRISMA 2020 flow diagram; Supplementary File S1, completed PRISMA 2020 checklist; Supplementary File S2, analysis script (Python).

References

1. Ruberg FL, Maurer MS. Cardiac Amyloidosis Due to Transthyretin Protein: A Review. JAMA 331 (2024): 778-791.
2. Adams D, Koike H, Slama M, et al. Hereditary Transthyretin Amyloidosis: A Model of Medical Progress for a Fatal Disease. Nat Rev Neurol 15 (2019): 387-404.
3. Garcia-Pavia P, Rapezzi C, Adler Y, et al. Diagnosis and Treatment of Cardiac Amyloidosis: A Position Statement of the ESC Working Group on Myocardial and Pericardial Diseases. Eur Heart J 42 (2021): 1554-1568.
4. Schmidt HH, Waddington-Cruz M, Botteman MF, et al. Estimating the Global Prevalence of Transthyretin Familial Amyloid Polyneuropathy. Muscle Nerve 57 (2018): 829-837.
5. González-López E, Gallego-Delgado M, Guzzo-Merello G, et al. Wild-Type Transthyretin Amyloidosis as a Cause of Heart Failure with Preserved Ejection Fraction. Eur Heart J 36 (2015): 2585-2594.
6. Buxbaum JN, Ruberg FL. Transthyretin V122I (pV142I) Cardiac Amyloidosis: An Age-Dependent Autosomal Dominant Cardiomyopathy Too Common to Be Overlooked as a Cause of Significant Heart Disease. Genet Med 19 (2017): 733-742.
7. Skrahina V, Grittner U, Beetz C, et al. Hereditary Transthyretin-Related Amyloidosis Is Frequent in Polyneuropathy and Cardiomyopathy of No Obvious Aetiology. Ann Med 53 (2021): 1787-1796.
8. Rösner S, Pardo LM, Bertoli-Avella AM, et al. Hereditary Transthyretin-Related Amyloidosis Ongoing Observational Study: A Baseline Report of the First 3167 Participants. J Clin Med 13 (2024): 6197.
9. Lane T, Fontana M, Martinez-Naharro A, et al. Natural History, Quality of Life, and Outcome in Cardiac Transthyretin Amyloidosis. Circulation 140 (2019): 16-26.
10. Gillmore JD, Damy T, Fontana M, et al. A New Staging System for Cardiac Transthyretin Amyloidosis. Eur Heart J 39 (2018): 2799-2806.
11. Ericzon BG, Wilczek HE, Larsson M, et al. Liver Transplantation for Hereditary Transthyretin Amyloidosis: After 20 Years Still the Best Therapeutic Alternative? Transplantation 99 (2015): 1847-1854.
12. European Medicines Agency. Attrogy (Diflunisal): European Public Assessment Report. Marketing Authorisation Valid Throughout the EU on 17 July 2025. Amsterdam: EMA (2025).
13. Gillmore JD, Gane E, Taubel J, et al. CRISPR-Cas9 In Vivo Gene Editing for Transthyretin Amyloidosis. N Engl J Med 385 (2021): 493-502.
14. Garcia-Pavia P, Aus dem Siepen F, Donal E, et al. Phase 1 Trial of Antibody NI006 for Depletion of Cardiac Transthyretin Amyloid. N Engl J Med 389 (2023): 239-250.
15. Saelices L, Chung K, Lee JH, et al. Amyloid Seeding of Transthyretin by Ex Vivo Cardiac Fibrils and Its Inhibition. Proc Natl Acad Sci USA 115 (2018): E6741-E6750.
16. Fontana M, Masri A, Solomon SD, et al. Eplontersen for Transthyretin Amyloid Cardiomyopathy. N Engl J Med (2026). doi: 10.1056/NEJMoa2608510.
17. Garcia-Pavia P, Cappelli F, Davis MK, et al. Eplontersen with and without Background Transthyretin Stabilizers in Transthyretin Amyloid Cardiomyopathy: Secondary Analysis of a Phase 3, Randomized Controlled Trial. Nat Med (2026). doi: 10.1038/s41591-026-04670-6.
18. Weiler M, Dohrn MF, Gingele S, et al. Disease Modifying Treatment of Hereditary Transthyretin Amyloidosis with Polyneuropathy in Germany - Expert Consensus of the German Society of Amyloid Diseases (DGAK) and the German Neurological Society (DGN). Neurol Res Pract 8 (2026): 63.
19. Coelho T, Maia LF, Martins da Silva A, et al. Tafamidis for Transthyretin Familial Amyloid Polyneuropathy: A Randomized, Controlled Trial. Neurology 79 (2012): 785-792.
20. Maurer MS, Schwartz JH, Gundapaneni B, et al. Tafamidis Treatment for Patients with Transthyretin Amyloid Cardiomyopathy. N Engl J Med 379 (2018): 1007-1016.
21. Gillmore JD, Judge DP, Cappelli F, et al. Efficacy and Safety of Acoramidis in Transthyretin Amyloid Cardiomyopathy. N Engl J Med 390 (2024): 132-142.
22. Judge DP, Alexander KM, Cappelli F, et al. Efficacy of Acoramidis on All-Cause Mortality and Cardiovascular Hospitalization in Transthyretin Amyloid Cardiomyopathy. J Am Coll Cardiol 85 (2025): 1003-1014.
23. Berk JL, Suhr OB, Obici L, et al. Repurposing Diflunisal for Familial Amyloid Polyneuropathy: A Randomized Clinical Trial. JAMA 310 (2013): 2658-2667.
24. Purpose Pharma International AB. European Commission Approval of ATTROGY (Diflunisal) for Hereditary Transthyretin-Mediated Amyloidosis with Stage 1 or Stage 2 Polyneuropathy. Press release, 4 August 2025.
25. Adams D, Gonzalez-Duarte A, O'Riordan WD, et al. Patisiran, an RNAi Therapeutic, for Hereditary Transthyretin Amyloidosis. N Engl J Med 379 (2018): 11-21.
26. Maurer MS, Kale P, Fontana M, et al. Patisiran Treatment in Patients with Transthyretin Cardiac Amyloidosis. N Engl J Med 389 (2023): 1553-1565.
27. Adams D, Tournev IL, Taylor MS, et al. Efficacy and Safety of Vutrisiran for Patients with Hereditary Transthyretin-Mediated Amyloidosis with Polyneuropathy: A Randomized Clinical Trial. Amyloid 30 (2023): 1-9.
28. Fontana M, Berk JL, Gillmore JD, et al. Vutrisiran in Patients with Transthyretin Amyloidosis with Cardiomyopathy. N Engl J Med 392 (2025): 33-44.
29. Benson MD, Waddington-Cruz M, Berk JL, et al. Inotersen Treatment for Patients with Hereditary Transthyretin Amyloidosis. N Engl J Med 379 (2018): 22-31.
30. European Medicines Agency. Tegsedi (Inotersen): Supply Shortage – Permanent Discontinuation in the European Union. EMA Medicine Shortage Communication, 3 July 2026.
31. Coelho T, Marques W Jr, Dasgupta NR, et al. Eplontersen for Hereditary Transthyretin Amyloidosis with Polyneuropathy. JAMA 330 (2023): 1448-1458.
32. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews. BMJ 372 (2021): n71.
33. Sterne JAC, Savović J, Page MJ, et al. RoB 2: A Revised Tool for Assessing Risk of Bias in Randomised Trials. BMJ 366 (2019): l4898.
34. Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: A Tool for Assessing Risk of Bias in Non-Randomised Studies of Interventions. BMJ 355 (2016): i4919.
35. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: An Emerging Consensus on Rating Quality of Evidence and Strength of Recommendations. BMJ 336 (2008): 924-926.
36. DerSimonian R, Laird N. Meta-Analysis in Clinical Trials. Control Clin Trials 7 (1986): 177-188.
37. Paule RC, Mandel J. Consensus Values and Weighting Factors. J Res Natl Bur Stand 87 (1982): 377-385.
38. Veroniki AA, Jackson D, Viechtbauer W, et al. Methods to Estimate the Between-Study Variance and Its Uncertainty in Meta-Analysis. Res Synth Methods 7 (2016): 55-79.
39. Higgins JPT, Thompson SG, Deeks JJ, et al. Measuring Inconsistency in Meta-Analyses. BMJ 327 (2003): 557-560.
40. IntHout J, Ioannidis JPA, Borm GF. The Hartung-Knapp-Sidik-Jonkman Method for Random Effects Meta-Analysis Is Straightforward and Considerably Outperforms the Standard DerSimonian-Laird Method. BMC Med Res Methodol 14 (2014): 25.
41. Röver C, Knapp G, Friede T. Hartung-Knapp-Sidik-Jonkman Approach and Its Modification for Random-Effects Meta-Analysis with Few Studies. BMC Med Res Methodol 15 (2015): 99.
42. Judge DP, Kristen AV, Grogan M, et al. Phase 3 Multicenter Study of Revusiran in Patients with Hereditary Transthyretin-Mediated (hATTR) Amyloidosis with Cardiomyopathy (ENDEAVOUR). Cardiovasc Drugs Ther 34 (2020): 357-370.
43. Fontana M, Garcia-Pavia P, Grogan M, et al. Coramitug, a Humanized Monoclonal Antibody for the Treatment of Transthyretin Amyloid Cardiomyopathy: A Phase 2, Randomized, Multicenter, Double-Blind, Placebo-Controlled Trial. Circulation 153 (2026): 214-225.
44. Fontana M, Solomon SD, Kachadourian J, et al. CRISPR-Cas9 Gene Editing with Nexiguran Ziclumeran for ATTR Cardiomyopathy. N Engl J Med 391 (2024): 2231-2241.
45. Elliott P, Drachman BM, Gottlieb SS, et al. Long-Term Survival with Tafamidis in Patients with Transthyretin Amyloid Cardiomyopathy. Circ Heart Fail 15 (2022): e008193.
46. Duarte GS, Machado T, Rodrigues FB, et al. Pharmacological Interventions for Hereditary Transthyretin-Related Amyloidosis with Polyneuropathy: Systematic Review and Network Meta-Analysis. Eur J Neurol 33 (2026): e70627.
47. National Institute for Health and Care Excellence. Inotersen for Treating Hereditary Transthyretin Amyloidosis (HST9): Guidance Withdrawn. London: NICE (2026).
48. Merck & Co. Dolobid (Diflunisal) Tablets: Prescribing Information. Whitehouse Station, NJ: Merck & Co. (2007). Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2007/018445s058lbl.pdf (accessed 01 October 2026).
49. Intellia Therapeutics. Intellia Therapeutics Announces FDA Lift of Clinical Hold on MAGNITUDE Phase 3 Clinical Trial in ATTR-CM. Press release, 2 March 2026.
50. gov. TRITON-CM: A Study to Evaluate Nucresiran in Patients with Transthyretin Amyloidosis with Cardiomyopathy. NCT07052903 (accessed 01 October 2026).
51. Alnylam Pharmaceuticals. Second Quarter 2026 Financial Results and Business Update. Company communication, 30 July 2026.
52. gov. DepleTTR-CM: Efficacy and Safety of ALXN2220 in Adult Participants with Transthyretin Amyloid Cardiomyopathy. NCT06183931 (accessed 01 October 2026).
53. gov. CLEOPATTRA: Effects of NNC6019-0001 versus Placebo on Cardiovascular Outcomes in Participants with Transthyretin Amyloid Cardiomyopathy. NCT07207811 (accessed 01 October 2026).
54. McDonagh TA, Metra M, Adamo M, et al. 2021 ESC Guidelines for the Diagnosis and Treatment of Acute and Chronic Heart Failure. Eur Heart J 42 (2021): 3599-3726.
55. Kittleson MM, Ruberg FL, Ambardekar AV, et al. 2023 ACC Expert Consensus Decision Pathway on Comprehensive Multidisciplinary Care for the Patient with Cardiac Amyloidosis. J Am Coll Cardiol 81 (2023): 1076-1126.
56. European Medicines Agency. Vyndaqel (Tafamidis): Summary of Product Characteristics. Amsterdam: EMA (accessed 01 October 2026).
57. Hurshman Babbes AR, Powers ET, Kelly JW. Quantification of the Thermodynamically Linked Quaternary and Tertiary Structural Stabilities of Transthyretin and Its Disease-Associated Variants: The Relationship Between Stability and Amyloidosis. Biochemistry 47 (2008): 6969-6984.
58. Qarni TN, Jones FJ, Drachman B, et al. Treatment Characteristics of Patients with Hereditary Transthyretin Amyloidosis: A Cohort Study. Orphanet J Rare Dis 19 (2024): 191.
59. Solomon SD, McMurray JJV, Claggett B, et al. Dapagliflozin in Heart Failure with Mildly Reduced or Preserved Ejection Fraction. N Engl J Med 387 (2022): 1089-1098.
60. gov. ACT-EARLY: A Study of Acoramidis in Asymptomatic Carriers of a Pathogenic Variant in the Transthyretin Gene. NCT06563895 (accessed 01 October 2026).
61. Suhr OB, Grogan M, Silva AMD, et al. PRX004 in Variant Amyloid Transthyretin (ATTRv) Amyloidosis: Results of a Phase 1, Open-Label, Dose-Escalation Study. Amyloid 32 (2025): 14-21.