Effectiveness of patisiran after switching from tafamidis for the treatment of hereditary transthyretin‐mediated amyloidosis with polyneuropathy

Abstract

Background and purpose

Hereditary transthyretin‐mediated amyloidosis with polyneuropathy (ATTRv‐PN [v for variant]) is a rare, progressive disease associated with multisystemic impairments. This study assessed the real‐world outcomes of patients with ATTRv‐PN who switched from tafamidis to patisiran, as well as the reasons for the treatment switch.

Methods

This was a retrospective chart review study at a large expert referral center. Data were extracted from medical charts of patients with ATTRv‐PN who switched from tafamidis to patisiran on or before 30 August 2019. Data elements included demographic and clinical characteristics, rationale for switch, and disease measures evaluated from tafamidis initiation through the 12‐month patisiran treatment period.

Results

Among the 24 patients with ATTRv‐PN included in the study, 50.0% had a V30M variant, and the mean (SD) age was 67.3 (8.0) years. During tafamidis treatment (mean [SD] = 30.1 [17.5] months) before switching to patisiran, patients worsened across multiple polyneuropathy measures, including walking ability, Neuropathy Impairment Score, and autonomic function. Neuropathic disease progression on tafamidis was the principal reason for switching to patisiran. After 12 months on patisiran (mean [SD] = 11.7 [1.4] months), patients experienced attenuated disease progression or improvement in the aforementioned measures of polyneuropathy.

Conclusions

Switching from tafamidis to patisiran attenuated the rate of functional decline, and most patients experienced stabilization or improvement of at least one polyneuropathy measure within 12 months of patisiran treatment. Timely switch from tafamidis to patisiran can be beneficial to avoid rapid disease progression in patients with ATTRv‐PN.

INTRODUCTION

Hereditary transthyretin‐mediated (ATTRv [v for variant]) amyloidosis is a rare, autosomal‐dominant, rapidly progressive, and multisystem disease associated with substantial disability, morbidity, and mortality [1, 2]. This adult onset disease is caused by mutations in the transthyretin (TTR) gene that result in the production of misfolded TTR proteins and amyloid fibril accumulation in tissues and organs, including the peripheral nerves, heart, eyes, and gastrointestinal tract [3, 4]. ATTRv amyoidosis has variable penetrance [5, 6], and most patients develop a mixed phenotype with variable impacts on sensorimotor, cardiovascular, and autonomic functions (e.g., diarrhea, sexual dysfunction, orthostatic intolerance) [2]. ATTRv amyloidoisis with polyneuropathy (ATTRv‐PN) has an aggressive course, as patients progress from being fully independent and ambulatory, with symptoms limited to the distal extremities, to being unable to care for themselves, confined to wheelchairs, or bedridden over several years [1, 7]. In addition to severely impairing patients' quality of life (QoL) [6, 8], polyneuropathy can ultimately be fatal [9].

Several therapies have been approved to treat the polyneuropathy of ATTRv amyloidosis. Tafamidis, a TTR stabilizer, was assessed (20 mg) in a randomized, placebo‐controlled phase III trial among a population with primarily ambulatory (i.e., stage 1), early onset (i.e., aged <50 years) transthyretin familial amyloid polyneuropathy, now known as ATTRv amyloidosis, with V30M mutation in the TTR gene [10]. The trial did not meet its coprimary efficacy endpoints in the intent‐to‐treat population, although significance was reached among a subpopulation who completed the trial per protocol. Tafamidis subsequently received approval from the European Medicines Agency (EMA) in 2011 for the treatment of transthyretin amyloidosis in adults with stage 1 symptomatic polyneuropathy, with approvals in other countries but not the United States (US) [11]. Since its approval outside the US, tafamidis has demonstrated differential real‐world effectiveness across patient groups, and the greatest benefit may be relegated to early onset patients with a V30M mutation, women, and those with minimal neurological impairment [12, 13, 14, 15, 16, 17, 18]. Most patients experience polyneuropathy progression within 1–1.5 years of diagnosis [12, 13, 14, 15, 16, 17, 18]. Patients with a higher Neuropathy Impairment Score (NIS) at baseline are associated with higher risk of neuropathy progression than those with a lower score [19].

Several years later, patisiran, an RNA interference (RNAi) therapeutic targeting variant and wild‐type TTR mRNA and resulting in rapid TTR knockdown, was evaluated in the randomized, placebo‐controlled phase III APOLLO study [20, 21]. APOLLO enrolled 225 patients with ambulatory abilities ranging from walking unassisted to requiring various assistive devices (Polyneuropathy Disability [PND] score of 1–3B), representing 39 different pathological TTR variants. The results demonstrated that patisiran improved polyneuropathy in most patients, as measured by a modified version of the NIS (mNIS+7), nutritional status, QoL, and activities of daily living from baseline to 18 months. Furthermore, subgroup analyses revealed consistent treatment effects across all prespecified subpopulations, including by baseline NIS (<50 or ≥50), V30M variant (early or late onset), and previous use of a TTR stabilizer such as tafamidis. Patisiran was subsequently approved in 2018 for the treatment of adults with ATTRv‐PN by the US Food and Drug Administration and the EMA (for stages 1 and 2), with later approvals in other countries. An open‐label study of APOLLO‐enrolled patients supported the long‐term (12 months) efficacy and safety of patisiran for ATTRv‐PN [22].

In APOLLO, approximately one third of enrolled patients were previously treated with or started on tafamidis prior to other therapies becoming available, yet still chose to participate in the trial of patisiran [23]. Multiple trials in ATTRv‐PN have enrolled substantial proportions of patients with prior TTR stabilizer use, primarily tafamidis [20, 21, 24]. With approval of patisiran as an additional treatment option for ATTRv‐PN, it is important to understand its real‐world effectiveness among patients who switch from tafamidis to patisiran, given that prior use tafamidis use is expected to be common. Accordingly, this single‐center chart review study evaluated the real‐world outcomes over the course of treatment among patients with ATTRv‐PN who switched from tafamidis to patisiran, and the clinical rationale for the treatment switch.

METHODS

Data source

Data were extracted from medical charts of eligible patients with ATTRv amyloidosis treated at the Centre Hospitalier Universitaire (CHU) Bicêtre in France. Participating physicians abstracted data using a predefined standardized chart abstraction form in Excel; all data were anonymized and nonidentifiable.

This study was approved by the ethics review committee and was conducted with informed consent for research purposes at CHU Bicêtre, Assistance Publique‐Hôpitaux de Paris (AP‐HP). All research was performed in accordance with the Declaration of Helsinki.

Patient inclusion and exclusion criteria

Selected patient charts met the following inclusion criteria: (i) confirmed diagnosis of ATTRv amyloidosis with a documented TTR variant; (ii) documented initiation and discontinuation of tafamidis for the treatment of ATTRv‐PN prior to patisiran initiation; (iii) switched from tafamidis to patisiran on or before 30 August 2019 (within 1 year of EU approval of patisiran); and (iv) stage 1 or 2 polyneuropathy, per the Coutinho Familial Amyloid Polyneuropathy (FAP) staging system, at patisiran initiation. Patients were excluded if they received treatment with diflunisal or inotersen between tafamidis discontinuation and patisiran initiation or had received an orthotopic liver transplant prior to patisiran initiation.

Data collection

All data elements were extracted from patients' medical records based on clinical assessments at routine visits, per the site's standard of care (SoC). Data were extracted for the assessment period, defined as from tafamidis initiation (or earliest clinical visit available at CHU Bicêtre AP‐HP during the tafamidis treatment period or the earliest visit history available from the neurologic referral center database) to 12 months after patisiran initiation (Figure 1). During the assessment period, the following SoC time points were used for data extraction: tafamidis initiation, 12 months prior to tafamidis discontinuation, tafamidis discontinuation, patisiran initiation, and 12 months after patisiran initiation. Data were extracted from the SoC evaluation that occurred nearest to but within ±3 months of the prespecified milestones.

FIGURE 1.

Study design. ATTRv, hereditary transthyretin‐mediated (variant); NIS, Neuropathy Impairment Score; PND, Polyneuropathy Disability. ^aIf data from the tafamidis treatment initiation visit were unavailable, data from the first clinical visit available at the center during the tafamidis treatment period were abstracted. ^bAll data were collected per the site's standard of care. Additional measures to assess progression and treatment effectiveness included, but were not limited to, manual grip strength, diarrhea severity, orthostatic intolerance severity, and erectile dysfunction severity.

A patient was considered lost to follow‐up if they did not return for any SoC visit within 12 (+3) months of initiating patisiran. If exact dates were not available, they were estimated based on the abstractor's best judgment.

Data collected

Patient demographics, disease characteristics, treatment history, and concomitant medications at patisiran initiation were recorded. The reasons for discontinuing tafamidis to initiate patisiran were collected and included subjective reasons based on patients' symptoms and physicians' clinical examination findings, and changes on neuropathy and cardiomyopathy severity measures. The multisystem and widespread nature of polyneuropathy due to ATTRv amyloidosis was evaluated, encompassing upper and lower body function, ambulatory ability, and autonomic dysfunction. These measures, assessed at each SoC visit, included PND score, FAP stage, site‐developed physician‐assigned categorization of a patient's walking difficulty, NIS‐Total score (upper and lower limb function), manual grip strength, and severity of diarrhea, orthostatic intolerance, and erectile dysfunction. As a secondary aim, the incidence of safety events occurring during the tafamidis and patisiran treatment periods requiring hospitalization or leading to fatality (patisiran treatment period only) was recorded.

Statistical analysis

Patient demographics, disease characteristics, treatment history, concomitant medications, reasons for switching, and clinical assessments for effectiveness and safety were summarized descriptively. Categorical variables were reported using frequency and proportion, and continuous variables were reported using mean, SD, and median. Analyses of continuous variables were conducted among evaluable patients with data at both the start and end of an assessment period. Data were presented as bar graphs depicting the proportions of patients with improvement (by ≥1 level), no change, and worsening (by ≥1 level) outcomes for PND score and severity of diarrhea, orthostatic intolerance, and erectile dysfunction during the 12 months prior to tafamidis discontinuation and after patisiran initiation. Stable walking difficulties since the last evaluation were defined as no new (or worsening of) walk distance decrease, falls, or balance disorder. The NIS‐Total was graded on a scale of 0–244, with a higher score indicating greater impairment. Change in NIS‐Total score at visits associated with tafamidis and patisiran treatment was calculated as the difference from NIS‐Total score at tafamidis and patisiran initiation, respectively. The annual rate of change in NIS‐Total score during the 12 months prior to tafamidis discontinuation and after patisiran initiation was calculated for each patient as the change divided by follow‐up time in years and averaged across patients with available data.

RESULTS

Patient demographics, disease characteristics, and concomitant treatments

Demographics

Twenty‐four patients with ATTRv‐PN who switched from tafamidis to patisiran by 30 August 2019 were included. Two thirds of patients were male (66.7%) and the mean (SD) age at patisiran initiation was 67.3 (8.0) years (Table 1).

TABLE 1.

Patient demographic and disease characteristics (N = 24).

Characteristic
Age at patisiran initiation, years, mean (SD) [median]	67.3 (8.0) [69.0]
Sex, n (%)
Male	16 (66.7%)
Female	8 (33.3%)
Mutation type, n (%)	24 (100.0%)
Early onset V30M [≤50 years of age at onset]	1 (4.2%)
Late onset V30M [>50 years of age at onset]	11 (45.8%)
Non‐V30M	12 (50.0%)
Ser77Tyr	6 (25.00%)
Ile107Val	2 (8.33%)
Thr69Ile	1 (4.17%)
Thr59Arg	1 (4.17%)
Thr49Ala	1 (4.17%)
Thr60Ala	1 (4.17%)
Timing of key study milestones, months, mean (SD) [median]
Duration from ATTRv amyloidosis diagnosis to tafamidis initiation	3.7 (4.5) [1]
Duration of tafamidis use	30.1 (17.5) [26.5]
Duration between tafamidis discontinuation and patisiran initiation	0.3 (1.1) [0.0]
Duration from ATTRv amyloidosis diagnosis to patisiran initiation	34.1 (18.4) [28.7]
Duration of patisiran use during the 12‐month follow‐up	11.7 (1.4) [12.0]
Concomitant medication use at the time of tafamidis initiation, n (%)
Any symptomatic medications for ATTRv amyloidosis	12 (50.0%)
Nerve pain medicine	12 (50.0%)
Antidiarrhea medication	2 (8.3%)
Concomitant medication use at the time of patisiran initiation, n (%)
Any symptomatic medications for ATTRv amyloidosis	17 (70.8%)
Nerve pain medicine	14 (58.3%)
Antidiarrhea medication	5 (20.8%)
Select comorbidities at the initiation of patisiran, n (%)
Heart arrhythmia/conduction disturbance	7 (29.2%)
Related to ATTRv amyloidosis	4 (57.1%)
Vision problem(s)	2 (8.3%)
Related to ATTRv amyloidosis	0 (0.0%)
Procedures received any time prior to patisiran initiation, n (%)
Cardiac pacemaker	9 (37.5%)
Carpal tunnel release surgery	9 (37.5%)
None of the above	9 (37.5%)

Abbreviations: ATTRv, hereditary transthyretin‐mediated (variant); V30M, valine 30 methionine variant.

Disease characteristics and treatment history

The mean (SD) duration of ATTRv amyloidosis from diagnosis to tafamidis initiation was 3.7 (4.5) months. Patients used tafamidis for a mean (SD) of 30.1 (17.5) months before switching to patisiran. The mean (SD) time between tafamidis discontinuation to patisiran initiation was 0.3 (1.1) months, and at patisiran initiation, patients were diagnosed for 34.1 (18.4) months. During the 12‐month follow‐up period, patients received patisiran for 11.7 (1.4) months (Table 1).

Half (50.0%) of the patients had a V30M variant (45.8% and 4.2% with late and early onset V30M, respectively). At patisiran initiation, 29.2% of patients had heart arrythmia, 8.3% had vision problems, 37.5% had a pacemaker, and 37.5% had ≥1 carpal tunnel release surgery (Table 1).

Concomitant treatments

Use of ≥1 concomitant medications to treat ATTRv amyloidosis symptoms was common, and their use increased between the time of tafamidis initiation (50.0% of patients) and the switch to patisiran (70.8%; Table 1). Nerve pain medication was used by 50.0% and 58.3% of patients at tafamidis and patisiran initiation, respectively; antidiarrhea medication was used by 8.3% and increased to 20.8%.

Reasons for switching from tafamidis to patisiran

The clinical rationale for switching from tafamidis to patisiran was neuropathic disease progression for all patients (100.0%); concurrent cardiac disease progression was also reported for three (12.5%; Table 2). Among patients with neuropathic disease progression, 21 (87.5%) had documented new/worsening sensory (70.8%), motor (70.8%), and autonomic (50.0%) symptoms. Sixteen (66.7%) patients had progression documented via objective neuropathy measures such as PND score (45.8%) and NIS‐Total score (41.7%).

TABLE 2.

Reasons for discontinuing tafamidis to initiate patisiran (N = 24). ^a

Disease progression with neuropathy, n (%)	24 (100.0%)
With neuropathy symptoms b	21 (87.5%)
Sensory symptoms	17 (70.8%)
Motor symptoms	17 (70.8%)
Dysautonomia symptoms	12 (50.0%)
Other, i.e., reduction of walking distance, new lipothymia, and new hand gesture disability	3 (12.5%)
With objective measures for neuropathy	16 (66.7%)
Worsening PND score	11 (45.8%)
Increased by 1 level	8 (72.7%)
Increased by 2 level	3 (27.3%)
NIS‐Total score increase	10 (41.7%)
Increase in NIS‐Total score, mean (SD)	28.9 (16.2)
Manual grip strength worsening in left hand	3 (12.5%)
Manual grip strength worsening in right hand	2 (8.3%)
Worsening FAP stage	1 (4.2%)
Disease progression with cardiomyopathy, n (%)	3 (12.5%)
With cardiomyopathy symptoms b , c	2 (8.3%)
With objective measures for cardiomyopathy, using echocardiogram	1 (4.2%)

Abbreviations: FAP, familial amyloid polyneuropathy; NIS, Neuropathy Impairment Score; PND, Polyneuropathy Disability.

^a Reasons for discontinuing tafamidis to initiate patisiran were not mutually exclusive.

^b Including new onset symptoms or worsening of existing symptoms.

^c Symptoms included heart failure and syncopes with cardiac dysautonomia.

Effectiveness of tafamidis and patisiran

Assessment time for treatment effectiveness

The effectiveness of tafamidis was assessed at around tafamidis initiation, 12 months prior to tafamidis discontinuation, and tafamidis discontinuation, with a mean (SD) of 30.1 (17.5), 12.8 (1.7), and 0.1 (0.4) months prior to tafamidis discontinuation, respectively. The effectiveness of patisiran was assessed at around 12 months (mean [SD] = 12.4 [2.0] months) after patisiran initiation.

Neuropathy Impairment Score‐Total Score

The mean (SD) NIS‐Total score increased (worsened) from 36.9 (17.3) at tafamidis initiation to 53.4 (16.6) at 12 months prior to tafamidis discontinuation, and further increased to 61.7 (20.0) at tafamidis discontinuation after a mean (SD) of 30.1 (17.5) months of treatment (Table 3). Conversely, the mean (SD) NIS‐Total score decreased (improved) to 59.1 (22.7) after 12 months on patisiran. NIS‐Total score increased by a mean (SD) of 14.3 (15.2) points per year during the last 12 months on tafamidis; it increased at a slower rate of mean (SD) 2.9 (12.4) points per year during the 12 months following patisiran initiation (Figure 2).

TABLE 3.

Summary of ATTR amyloidosis clinical assessments from the time of tafamidis initiation to 12 months after patisiran initiation ^a

	Tafamidis initiation	12 months prior to tafamidis discontinuation	Tafamidis discontinuation b	Patisiran initiation b	12 months after patisiran initiation
Timing of visit prior to tafamidis discontinuation or after patisiran initiation (months), mean (SD)	30.1 (17.5)	12.8 (1.7)	0.1 (0.4)	–	12.4 (2.0)
PND score, n (%)	24	16	24	24	19
0—No symptoms	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
I—Sensory disturbances, preserved walking capability	7 (29.2%)	2 (12.5%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
II—Impaired walking capacity	10 (41.7%)	8 (50.0%)	13 (54.2%)	13 (54.2%)	8 (42.1%)
IIIA—Walking with the help of one stick or crutch	5 (20.8%)	4 (25.0%)	5 (20.8%)	5 (20.8%)	7 (36.8%)
IIIB—Walking with the help of two sticks or crutches	1 (4.2%)	2 (12.5%)	6 (25.0%)	6 (25.0%)	4 (21.1%)
IV—Confined to a wheelchair or bedridden	1 (4.2%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
Change in level of walking difficulties since last assessment c , n (%)	17	16	24	24	18
Stable since last evaluation	5 (29.4%)	5 (31.3%)	9 (37.5%)	8 (33.3%)	12 (66.7%)
NIS d , n	24	14	23	23	15
Total NIS, mean (SD)	36.9 (17.3)	53.4 (16.6)	61.7 (20.0)	61.7 (20.0)	59.1 (22.7)
Change in NIS, mean (SD) e , f	–	+17.0 (18.4)	+24.9 (17.1)	–	+2.36 (12.8)
Rate of change in NIS per year, mean (SD) g	–	14.3 (15.2)		2.9 (12.4)
Diarrhea severity, n (%)	24	13	19	19	15
None	17 (70.8%)	9 (69.2%)	12 (63.2%)	12 (63.2%)	11 (73.3%)
Once a month	3 (12.5%)	2 (15.4%)	4 (21.1%)	4 (21.1%)	2 (13.3%)
Once a week	1 (4.2%)	1 (7.7%)	1 (5.3%)	1 (5.3%)	1 (6.7%)
More than twice a week but less than daily	2 (8.3%)	1 (7.7%)	2 (10.5%)	2 (10.5%)	0 (0.0%)
Daily	1 (4.2%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	1 (6.7%)
Orthostatic intolerance severity, n (%)	24	14	20	20	15
None	10 (41.7%)	7 (50.0%)	5 (25.0%)	5 (25.0%)	6 (40.0%)
Asymptomatic	7 (29.2%)	6 (42.9%)	8 (40.0%)	7 (35.0%)	6 (40.0%)
Lipothymia	6 (25.0%)	1 (7.1%)	7 (35.0%)	8 (40.0%)	3 (20.0%)
Postural syncopes	1 (4.2%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
Bedridden	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
Erectile dysfunction severity (males), n (%) h	15	8	12	12	9
None	5 (33.3%)	1 (12.5%)	2 (16.7%)	2 (16.7%)	2 (22.2%)
Difficulties	3 (20.0%)	5 (62.5%)	9 (75.0%)	7 (58.3%)	3 (33.3%)
Impotency	7 (46.7%)	2 (25.0%)	1 (8.3%)	3 (25.0%)	4 (44.4%)

Abbreviations: NIS, Neuropathy Impairment Score; PND, Polyneuropathy Disability; SD, standard deviation; SoC, standard of care.

^a For each outcome, data were extracted from the SoC evaluation that occurred nearest to, but within ±3 months of the pre‐specified milestone (e.g., 6 months prior to discontinuation of tafamidis).

^b In instances where data were not available from two distinct SoC evaluations for discontinuing tafamidis and initiating patisiran, measures from a visit closest to either tafamidis discontinuation or patisiran initiation were included under both milestones.

^c Stable was defined as no walk distance decrease, fall or balance disorder, or no worsening of these events if they had already happened.

^d NIS‐Total is graded on a scale of 0–244, with a higher score indicating greater impairment.

^e Change in NIS at visits associated with tafamidis treatment was calculated as the difference from NIS at tafamidis initiation.

^f Change in NIS at visits associated with patisiran treatment was calculated as the difference from NIS at patisiran initiation.

^g The rate of change was calculated for each patient divided by each patient's follow‐up time in years and averaged across patients who had data available at both the SoC evaluations. The rate of change for the tafamidis treatment period was calculated during the 12 months period prior to tafamidis discontinuation. The rate of change for the patisiran treatment period was calculated for the 12 months period after patisiran initiation.

^h The proportion of patients with erectile dysfunction is reported among male patients.

FIGURE 2.

Change in patients' Neuropathy Impairment Score (NIS)‐Total score before and after patisiran initiation. NIS‐Total is graded on a scale of 0–244, with a higher score indicating greater impairment. The tafamidis discontinuation visit did not overlap with the eve of patisiran initiation for four of 24 patients. For 11 of 24 patients, the NIS‐Total score measurement date closest to the patisiran initiation date was before the patisiran initiation date.

Walking ability

At tafamidis initiation, 29.2% of patients had preserved walking ability (PND score 0 or I; Table 3). At tafamidis discontinuation, 95.8% of patients had no improvement in PND score (i.e., reduction by ≥1 level) and 0% had preserved walking ability. Similarly, during the last 12 months on tafamidis, no patients improved in PND score, 68.8% experienced no change, and 31.3% worsened in ambulatory ability (Figure 3). Comparatively, after 12 months on patisiran, 10.5% had PND score improvement, 73.7% experienced no change, and 15.8% worsened. At tafamidis discontinuation, 37.5% of patients were considered by their clinician to have a stable walking status since the previous evaluation (Table 3). At 12 months on patisiran, 66.7% of patients were considered to be stable since their last evaluation.

FIGURE 3.

Change in patients' Polyneuropathy Disability (PND) score during the tafamidis and patisiran treatment periods. Pink refers to the change from 12 months prior to tafamidis discontinuation to tafamidis discontinuation among patients without missing data (n = 16). Blue refers to the change from initiation of patisiran to 12 months after patisiran initiation among patients without missing data (n = 19). “Improved” refers to reduction in PND score by ≥1 level, and “Worsened” refers to increase in PND score by ≥1 level.

Diarrhea severity

At tafamidis discontinuation, 63.2% of patients reported no diarrhea, compared with 73.3% at 12 months of patisiran (Table 3). During the 12 months prior to tafamidis discontinuation, no patients experienced improvement in diarrhea symptoms, 69.2% had no change, and 30.8% experienced worsened symptoms (Figure 4a). During the 12 months following patisiran initiation, 14.3% experienced improvement, 71.4% had no change, and 14.3% worsened.

FIGURE 4.

Change in patients' (a) diarrhea, (b) orthostatic intolerance, and (c) erectile dysfunction severity during the tafamidis and patisiran treatment periods. Pink refers to the change from 12 months prior to tafamidis discontinuation to tafamidis discontinuation among patients without missing data (n = 13 in a, n = 14 in b, n = 8 in c). Blue refers to the change from patisiran initiation to 12 months after patisiran initiation among patients without missing data (n = 14 in a, n = 14 in b, n = 8 in c). “Improved” refers to improvement in severity by ≥1 level, and “Worsened” refers to worsening in severity by ≥1 level. The assessment levels for diarrhea severity (a) were as follows: none, once per month, once per week, more than twice per week, daily. The assessment levels for orthostatic intolerance severity (b) were as follows: none, asymptomatic, lipothymia, postural syncopes, bedridden. The assessment levels for erectile dysfunction severity (c) were as follows: none, difficulties, impotency.

Orthostatic intolerance severity

At tafamidis discontinuation, 65.0% of patients had no orthostatic intolerance or were asymptomatic, compared with 80.0% at 12 months on patisiran (Table 3). During the 12 months before tafamidis discontinuation, 7.1% of patients experienced improvement in orthostatic intolerance, 57.1% had no change, and 35.7% worsened (Figure 4b). During the 12 months following patisiran initiation, 28.6% of patients experienced improvement in orthostatic intolerance, 57.1% had no change, and 14.3% worsened.

Erectile dysfunction

Among males, 16.7% had no erectile dysfunction at tafamidis discontinuation, compared with 22.2% at 12 months on patisiran (Table 3). During the 12 months before tafamidis discontinuation, 25.0% of patients experienced improvement in erectile dysfunction, 75.0% had no change, and 0% worsened (Figure 4c). During the 12 months following initiation of patisiran, no patients experienced improvement in erectile dysfunction, 87.5% had no change, and 12.5% worsened.

Safety

Safety events requiring unplanned hospitalization or extension of an existing hospitalization were reported in two patients (8.3%) during the tafamidis treatment period (pelvic bone fracture and syncope, each n = 1; Table 4). During the patisiran treatment period, three patients (12.5%) had safety events. One patient with a non‐V30M variant, who received a cardiac pacemaker during the tafamidis treatment period, had a fatal event of heart failure. Two patients had events of left foot tarsal dislocation or vitreous hemorrhage (both n = 1).

TABLE 4.

Safety events that led to potential unplanned hospitalizations or extension of existing hospitalizations.

	During tafamidis treatment period, N = 24	During patisiran treatment period, N = 24
Duration of treatment use, months, mean (SD)	30.1 (17.5)	11.7 (1.4)
Patients with safety events leading to an unplanned hospitalization or extension of an existing hospitalization, n (%) a	2 (8.3%)	3 (12.5%)
Outcome of the event, n (%)	2 (8.3%)	3 (12.5%)
Recovered/resolved without sequelae	2 (100.0%)	0 (0.0%)
Ongoing as of the date of data extraction	0 (0.0%)	2 (66.7%)
Fatality b	NA	1 (33.3%)

Note: Per protocol, there was no attribution of relatedness to treatment of any safety event.

Abbreviation: NA, not applicable.

^a Safety events during the tafamidis treatment period included bone pelvic fracture (n = 1, 4.2%) and syncope (n = 1, 4.2%). Safety events during the patisiran treatment period included a fatal event of heart failure (n = 1, 4.2%), left foot tarsal dislocation (n = 1, 4.2%), and vitreous hemorrhage due to severe amyloidosis of the retinous vessels (n = 1, 4.2%).

^b Safety events with fatal outcomes were only collected during the patisiran treatment period. The study enrolled patients who could switch treatment to patisiran, thus it was not designed to collect fatal outcomes in tafamidis‐treated patients.

DISCUSSION

Prior to the approval of patisiran in 2018, tafamidis was the only approved therapy available for patients with ATTRv‐PN across any geographic region. In the APOLLO trial of patisiran, approximately one third of those enrolled had previously used tafamidis, and most had discontinued it either due to disease progression (34%) or to participate in APOLLO (62%). The rate of polyneuropathy progression among these tafamidis‐treated patients before entering APOLLO is unknown, precluding assessment of how their polyneuropathy progression changed after switching to patisiran. However, the results from this single‐center experience systematically describe, for the first time, how a switch from tafamidis to patisiran can impact polyneuropathy disease progression. Additionally, the results of this study confirmed what earlier real‐world evidence has shown, that some patients experience substantial worsening on polyneuropathy measures with tafamidis, requiring alternative therapeutic options.

The results indicated that patients started tafamidis quickly following diagnosis and were treated with tafamidis for nearly the entire period since diagnosis (30 of 34 months). During the tafamidis treatment period, this cohort experienced substantial disease progression, including the loss of unimpaired walking ability in all patients and lack of any PND score improvement in 96%. Additionally, a substantial proportion of patients experienced worsening autonomic function (i.e., increased severity of diarrhea and orthostatic intolerance) while on tafamidis. Annual change in NIS‐Total, a combined measure of motor function, sensory function, and tendon reflexes, was 14.3 points per year during the last 12 months on tafamidis. This rate is similar to the natural history of progression of polyneuropathy from ATTRv amyloidosis estimated by Adams et al. at 14.3 points annually for a population with only 6% receiving a TTR stabilizer [7]. Furthermore, the 2022 guideline for monitoring ATTR amyloidosis describes a change in NIS‐Total score of 7–16 points over 12 months as indicative of neuropathic progression [25]. Notably, in early onset ATTRv amyloidosis with V30M, the proportion of NIS‐Lower Limbs responders at 18 months was not significantly different between the tafamidis or placebo groups [10]. Subsequent follow‐up studies have shown that tafamidis may have more modest effects on disease progression in patients with late onset V30M and non‐V30M mutations [25], a finding mirrored in real‐world studies [12, 13, 14, 15, 16, 17, 18].

In this study, all patients switched to patisiran by the first year after its EU approval due to worsening polyneuropathy while on tafamidis. The most common reasons for switching were worsening of PND score or NIS‐Total score (66.7%) and new onset and/or worsening neuropathy symptoms (87.5%), including autonomic dysfunction. Autonomic impairments are tangible measures correlated with patients' QoL and daily functioning [26, 27, 28]. Gastrointestinal function is correlated with survival, orthostatic intolerance impacts independence and safety, and erectile dysfunction is a common early disease manifestation in males [27, 28, 29, 30]. However, autonomic impairments may be underappreciated in routine clinical evaluations and are challenging to improve. Thus, it is important to note that not all aspects of the patient's course of treatment can be captured with PND score or NIS assessments, and that the addition of tools assessing autonomic dysfunction (e.g., the Compound Autonomic Dysfunction Test) may enhance evaluations of progression [31, 32]. Furthermore, recent expert consensus guidance is aligned with using NIS, PND score, autonomic dysfunction, and other outcome measures, depending on the site's SoC, to assess progression and prompt an evaluation for treatment switch if clinically warranted [25].

As patisiran had only recently become available for the study cohort, the switch from tafamidis to patisiran may have occurred later compared to current clinical practice, thus patients may have experienced more disease progression on tafamidis prior to switching. This study did not require patients to have a minimum duration of patisiran use, although only one patient discontinued within the 12‐month period. During the first year on patisiran, polyneuropathy stabilized in most patients across multiple measures. The increase in NIS‐Total was greatly attenuated, and the mean annual NIS‐Total progression during this period fell well below the average reported in natural history studies (2.9 vs. 14.3 points per year) [7, 33]. Most patients were stable in their walking ability measured by both the site's walking ability tool and standard PND scoring. Indeed, 84.2% of evaluable patients had no change or improved in PND score after 12 months on patisiran. These results are broadly similar to those reported for patisiran in APOLLO [27, 34]. On measures of autonomic dysfunction, 86% of patients had improvement or no change in the severity of orthostatic intolerance or diarrhea after 12 months on patisiran. Patisiran was also well tolerated, and no patients discontinued patisiran to switch to alternative therapies during the study.

It is encouraging to observe that the real‐world rates of disease progression associated with tafamidis and patisiran mirror those of previous studies, including their respective clinical trials. The substantial slowing of the rate of NIS‐Total worsening observed in this cohort after switching to patisiran, as well as a reduction in the accumulation of further autonomic disability, has important implications for patient care. Specifically, delaying the switch from tafamidis to patisiran may place patients at heightened risk of progression on multiple measures of functioning and aspects of polyneuropathy, as the accumulating neuropathy impairment may not be entirely reversable.

This study benefits from several strengths, including a cohort that was representative of a wide variety of mutations including both early and late onset ATTRv amyloidosis, which supports the broad generalizability of the findings across global regions where tafamidis is available. Additionally, the study describes the real‐world treatment experience at one of the largest expert referral centers for ATTRv amyloidosis worldwide. The data represent a unique contribution to the literature for a rare, understudied disease, and provide an enhanced understanding of whether disease progression changes as a result of a treatment switch.

The results of this study should also be considered in the light of some limitations associated with a single center, single‐arm study. Notably, the reasons underlying the decision to initiate a switch from tafamidis to patisiran may differ across centers, geographic locations, and patient populations. These results reflect the experience of a single center and may not be representative of all patients with ATTRv‐PN, so evidence from other centers could be useful for a broader perspective. Second, as is often the case in real‐world studies of rare diseases, the limited cohort size and missing data at certain visits may impact the generalizability of the findings. Third, it should also be noted that this study collected secondary data on safety events that led to potentially unplanned hospitalizations and was not designed to characterize the overall safety profile of tafamidis or patisiran. The safety events reported were consistent with those that frequently occur in patients with ATTRv amyloidosis, and no new safety concerns were observed. Finally, the intention of this analysis was to observe treatment switches between tafamidis and patisiran. Thus, switches involving other agents for ATTRv‐PN, like the antisense oligonucleotide inotersen or RNAi therapeutic vutrisiran (EMA approvals 2018 and 2022, respectively) were not evaluated in this study [35, 36, 37], precluding a comparison of switching to patisiran versus switching to other treatments. Previous analyses have reported that subcutaneously administered vutrisiran and intravenously infused patisiran are clinically comparable, and that both have substantial benefits for patients previously treated with TTR stabilizers [38, 39, 40]. Future research is warranted to examine the evolution of switching patterns and the reasons for switching with therapeutic advances and changes in patient care.

CONCLUSIONS

This retrospective study summarizes the real‐world treatment experience in a single‐center cohort of patients with ATTRv‐PN who received treatment with tafamidis and switched to patisiran. Over the course of tafamidis treatment, patients worsened across multiple measures of polyneuropathy, including walking ability, NIS‐Total score, and autonomic functions, resulting in a switch to patisiran. On average, patients experienced attenuation of neurologic impairment or improvement of these same measures of polyneuropathy within 1 year on patisiran. These findings emphasize the importance of monitoring disease progression and considering a timely switch to patisiran in patients who experience worsening neuropathy with tafamidis to avoid unnecessary progression.

AUTHOR CONTRIBUTIONS

Conceptualization: All authors. Methodology: All authors. Data curation: Sakshi Sethi, Lyuba Popadic, Hongbo Yang, and Celine Labeyrie. Formal analysis: Sakshi Sethi, Lyuba Popadic, and Hongbo Yang. Investigation: All authors. Validation: Sakshi Sethi, Lyuba Popadic, and Hongbo Yang. Visualization: Sakshi Sethi, Lyuba Popadic, and Hongbo Yang. Project administration: All authors. Writing—original draft: All authors. Writing—review and editing: All authors.

FUNDING INFORMATION

This work was supported by Alnylam Pharmaceuticals.

CONFLICT OF INTEREST STATEMENT

M.M., M.T.S., and H.L. are employees of Alnylam Pharmaceuticals and report ownership of shares in Alnylam Pharmaceuticals. S.S., L.P., and H.Y. are employees of Analysis Group, which received funding for this research from Alnylam Pharmaceuticals. C.L. has no business positions, no investments in pharmaceutical companies; C.L. has received fees for consulting and symposiums in the area of ATTRv amyloidosis from Alnylam and Pfizer, and for data extraction and study design for this study from Alnylam Pharmaceuticals. D.A. has participated in clinical trials sponsored by Akcea and Alnylam Pharmaceuticals and has received consulting fees from Alnylam Pharmaceuticals, Bridgebio, Pfizer, and AstraZeneca.

Labeyrie C, Merkel M, Sethi S, et al. Effectiveness of patisiran after switching from tafamidis for the treatment of hereditary transthyretin‐mediated amyloidosis with polyneuropathy. Eur J Neurol. 2024;31:e16384. doi: 10.1111/ene.16384

DATA AVAILABILITY STATEMENT

The authors confirm that the data supporting the findings of this study are available within the article.