Treatment response and neurofilament light chain levels with long-term patisiran in hereditary transthyretin-mediated amyloidosis with polyneuropathy: 24-month results of an open-label extension study

Abstract Background Longitudinal changes in neurofilament light chain (NfL) levels were evaluated alongside prespecified clinical assessments 24 months into the patisiran Global open-label extension (OLE) study in patients with ATTRv amyloidosis with polyneuropathy. Methods All patients enrolled in the Global OLE, from phase III APOLLO and phase II OLE parent studies, received patisiran. Assessments included measures of polyneuropathy (modified Neuropathy Impairment Score+7 (mNIS+7)), quality of life (QOL; Norfolk QOL-Diabetic Neuropathy questionnaire (Norfolk QOL-DN)), and plasma NfL. Results Patients receiving patisiran in the parent study (APOLLO-patisiran, n = 137; phase II OLE-patisiran, n = 25) demonstrated sustained improvements in mNIS+7 (mean change from parent study baseline (95% confidence interval): APOLLO-patisiran −4.8 (−8.9, −0.6); phase II OLE-patisiran −5.8 (−10.5, −1.2)) and Norfolk QOL-DN (APOLLO-patisiran −2.4 (−7.2, 2.3)), and maintained reduced NfL levels at Global OLE 24 months. After initiating patisiran in the Global OLE, APOLLO-placebo patients (n = 49) demonstrated stabilized mNIS+7, improved Norfolk QOL-DN, and significantly reduced NfL levels. Patisiran continued to demonstrate an acceptable safety profile. Earlier patisiran initiation was associated with a lower exposure-adjusted mortality rate. Conclusions Long-term patisiran treatment led to sustained improvements in neuropathy and QOL, with NfL demonstrating potential as a biomarker for disease progression and treatment response in ATTRv amyloidosis with polyneuropathy.

ATTRv amyloidosis is associated with substantial morbidity and mortality, yet diagnosis is difficult and often delayed [8,9].As approved therapies appear to be more effective when given early [10,11], the need for earlier diagnosis is apparent.Further to the diagnostic challenges, monitoring disease progression and treatment response is also difficult, due to the multisystem nature and heterogeneous clinical presentation [12][13][14][15][16]. Hence, identification of reliable biomarkers for ATTRv amyloidosis would be of benefit.
In the phase III APOLLO study in patients with ATTRv amyloidosis with polyneuropathy, the RNA interference (RNAi) therapeutic patisiran demonstrated improvements in the primary endpoint of neuropathy, measured by modified Neuropathy Impairment Score þ 7 (mNISþ7) [30], and all secondary endpoints, including quality of life (QOL; Norfolk Quality of Life-Diabetic Neuropathy questionnaire (Norfolk QOL-DN)), vs. placebo [6].During APOLLO, patisiran treatment also led to a decrease in NfL levels from baseline over 18 months, whereas an increase was seen with placebo [28].Notably, changes in mNISþ7 in APOLLO were significantly correlated with changes in NfL levels, suggesting that NfL may act as a biomarker in ATTRv amyloidosis [28].
The ongoing patisiran Global open-label extension (OLE) study [10], which enrolled eligible patients from APOLLO [6] and the patisiran phase II OLE study [31], is evaluating the long-term safety and efficacy of patisiran.The first interim data analysis at 12 months has been presented previously [10].Here we present the second interim analysis at 24 months, together with an evaluation of longitudinal changes in NfL levels during patisiran treatment.

Materials and methods
The primary objective of the current analysis was to present updated efficacy and safety results at 24 months in the patisiran Global OLE study.In addition, an analysis of NfL levels was undertaken to: confirm the previously observed decrease in NfL levels upon patisiran treatment; describe longitudinal changes in NfL, alongside clinical assessments, with longterm patisiran treatment; and better understand the temporal relationship between NfL lowering and patisiran treatment.

Trial design
The patisiran Global OLE (NCT02510261) is an ongoing, international, multicenter trial [10], which enrolled patients with ATTRv amyloidosis with polyneuropathy who had completed the phase III APOLLO (NCT01960348) [6] or phase II OLE (NCT01961921) [31] parent studies, and, in the opinion of the investigator, tolerated the study drug.Methods for the Global OLE study have been reported previously [10] and are summarized below.In the Global OLE study, patients will receive patisiran 0.3 mg/kg intravenously every 3 weeks for up to 5 years, although patients had not completed the full duration of the study at the time of this 24-month interim analysis.

Measures
In the Global OLE study, mNISþ7, Norfolk QOL-DN, and polyneuropathy disability (PND) scores were assessed at Global OLE baseline, 12 months, and 24 months.mNISþ7 is a composite clinical measure of motor, sensory, and autonomic neuropathy, which uses standardized assessments to quantify nerve conduction, muscle weakness, muscle stretch reflexes, sensory loss, and autonomic impairment (scored 0 À 304; higher scores indicate greater impairment) [32].Norfolk QOL-DN is a 35-item, patient-reported questionnaire with five domains comprizing activities of daily living, physical functioning/large-fiber neuropathy, small-fiber neuropathy, autonomic neuropathy, and symptoms (scored À4 to 136; higher scores indicate lower QOL) [33].PND score, which is based on ambulatory function, is also used to assess neuropathy in patients with ATTRv amyloidosis (scored I, II, IIIa, IIIb, and IV; higher scores indicate more limited ambulatory function).
NfL levels were measured at 12 and 24 months from Global OLE baseline.NfL plasma measurements were done in duplicate at a central location (PPD Bioanalytical Lab, Richmond, VA, USA) using the Quanterix Simoa platform.NfL plasma level assessment is a post hoc exploratory analysis in the Global OLE.
Patients returned to the clinical site for safety evaluations, including laboratory assessments, at 12, 26, and 52 weeks after the first patisiran dose in the Global OLE study and yearly thereafter.Patients were continuously assessed for adverse events (AEs).

Statistical analysis
The sample size of this study was not prespecified [10].Patients who enrolled in the Global OLE are presented in groups based on their participation in the parent studies: the APOLLO-patisiran group comprizes patients who received patisiran for 18 months in APOLLO and continue receiving patisiran in the Global OLE, the APOLLO-placebo group comprizes patients who started receiving patisiran for the first time in the Global OLE, and the phase II OLEpatisiran group comprizes patients who received patisiran for 24 months in the phase II OLE and continue receiving patisiran in the Global OLE.mNISþ7 and Norfolk QOL-DN efficacy data are reported as an integrated analysis including all patients who had !1 dose of study treatment in the parent studies and had an efficacy assessment at a given timepoint.Data for the Global OLE 24-month timepoint are given for patients who had completed the 24-month efficacy assessment by the 24month interim cut-off for the Global OLE on 7 October 2019.mNISþ7 and Norfolk QOL-DN are described as change from parent study or Global OLE baseline.
NfL levels were analysed in a subset of consenting patients who had received ! 1 dose of patisiran and had available samples [28].NfL data are described as plasma levels or change in plasma levels from parent study or Global OLE baseline.
Improvement in mNISþ7 score, Norfolk QOL-DN score, and NfL levels was defined as a negative mean change from the baseline value.Unless otherwise stated, baseline was defined as the initiation of patisiran treatment.
A safety analysis for the total Global OLE population is presented.A post hoc analysis assessed integrated exposureadjusted mortality rate from the onset of patisiran treatment in each group, using pooled data from the parent and Global OLE studies.The integrated exposure-adjusted mortality rate was calculated overall and by treatment group as the total number of deaths per total patient-years of exposure multiplied by 100.Deaths occurring between the first patisiran dose and up to 90 days after the last dose were included in the calculation.
Number and percentage of patients were reported for categoric variables.Mean (standard deviation (SD)) or median (interquartile range) were calculated for continuous variables.Comparisons of NfL levels between groups were tested for significance using linear regression, controlling for age, using R (R Foundation, Vienna, Austria).Sex was not found to affect NfL levels and was not included in the regression model.Comparisons of NfL levels in a given group at different timepoints were tested for significance using a Wilcoxon signedrank test.Statistical analyses of efficacy data were made using SAS, version 9.4 (SAS Institute Inc, Cary, NC, USA).
The study protocol is publicly available [34].

Patients
The study population for this analysis comprized a subset of patients from the Global OLE and its parent studies; details of enrolment have been described previously [10].Of the 148 APOLLO patients treated with patisiran, 138 (93.2%) completed the parent study and 137 enrolled in the Global OLE (APOLLO-patisiran group).Of the 77 patients treated with placebo in APOLLO, 55 (71.4%) completed the parent study and 49 enrolled in the Global OLE (APOLLO-placebo group) (Figure 1).There were 27 patients treated with patisiran in the phase II OLE study, 25 of whom completed the study and enrolled in the Global OLE (Figure 1).  1) had similar characteristics to the total population of the respective patient group [31,35].
Rapid polyneuropathy progression in the APOLLO-placebo group was observed during placebo treatment in the parent study and at Global OLE baseline.However, polyneuropathy progression in the APOLLO-placebo group was stabilized upon initiation of patisiran treatment in the Global OLE.This stabilization of polyneuropathy progression was sustained over 24 months of patisiran treatment in the Global OLE (mean change (95% CI) from Global OLE baseline 1.2 (-5.6, 8.0)).However, the mNISþ7 values for APOLLO-placebo patients remained worse than their parent study baseline at Global OLE 24 months, likely due to polyneuropathy progression while on placebo in APOLLO (mean change (95% CI) from APOLLO baseline 27.0 (17.0, 37.0)).
The majority of patients in the APOLLO-patisiran (53.4%) and phase II OLE-patisiran (66.7%) groups had an improved or stable PND score at Global OLE 24 months compared with parent study baselines (Table 2).In the APOLLO-placebo group, more patients with evaluable data had a worsened PND score (29.9%) than a stable score (14.3%) at Global OLE 24 months compared with APOLLO baseline, and no patients with evaluable data in this group had an improved PND score compared with APOLLO baseline.However, when compared with Global OLE baseline, the majority of the APOLLO-placebo patients had a stable or improved PND score (57.1%) at Global OLE 24 months, reflecting the effect of patisiran initiation on entry to the Global OLE (Table 2).
After 24 months of additional patisiran treatment in the Global OLE, durable improvement in Norfolk QOL-DN was observed in the APOLLO-patisiran group compared with the parent study baseline (mean change (95% CI) from APOLLO parent study baseline À2.4 (À7.2, 2.3); Figure 2(C)).For patients in the APOLLO-placebo group, switching to patisiran treatment led to an improvement in Norfolk QOL-DN over 24 months in the Global OLE (mean change (95% CI) from Global OLE baseline À4.1 (-10.8, 2.6)).Similar to the mNISþ7 results, Norfolk QOL-DN scores in APOLLO-placebo patients did not return to their parent study baseline by Global OLE 24 months, likely due to progression on placebo during APOLLO (mean change (95% CI) from APOLLO baseline 15.9 (6.7, 25.0)).

Global OLE safety
Safety of, and total exposure to, patisiran at 24 months in the Global OLE are summarized in Table 5. AEs were reported in 209/211 (99.1%) patients in the Global OLE, with the majority of AEs in the APOLLO-patisiran and phase II OLE-patisiran groups being mild or moderate in severity.Serious AEs were more common in the APOLLOplacebo group (34/49 (69.4%) patients with ! 1 event) compared with the APOLLO-patisiran group (59/137 (43.1%)) or phase II OLE-patisiran group (11/25 (44.0%)).The most common treatment-related AEs were mild or moderate infusion-related reactions (IRRs).The frequency of IRRs decreased over time, consistent with observations in APOLLO [6], and there were no serious IRRs or discontinuations due to IRRs between the Global OLE baseline and 24-month data cut-off.Over the first 24 months of the Global OLE, the APOLLO-patisiran and APOLLO-placebo groups reported alanine aminotransferase elevation AEs in one (0.7%) patient and three (6.1%)patients, respectively, and aspartate aminotransferase elevation AEs in one (0.7%) patient and two (4.1%) patients, respectively.Neither AE was reported in the phase II OLE-patisiran group.
Deaths were reported in 29 (13.7%)patients in the Global OLE, and all were considered unrelated to patisiran by the investigators; causes of death were consistent with the natural history of ATTRv amyloidosis.The proportion of deaths in the APOLLO-placebo group (16 (32.7%)) was higher than in the APOLLO-patisiran (13 (9.5%)) and phase II OLE-patisiran (zero) groups.
Throughout the parent studies and Global OLE, the exposure-adjusted mortality rate for patients who received !1 dose of patisiran was 4.3 per 100 patient-years (95% CI 3.1, 5.9), based on 35 deaths and 808.5 patient-years of cumulative exposure (Table 6), which is below the estimated range for patients with ATTRv amyloidosis (6.8 À 29 per 100 patient-years) [36][37][38][39][40].The exposure-adjusted mortality was lowest for patients in the phase II OLE, who were treated with patisiran for the longest duration and from the earliest stage of disease, and lower in the APOLLO-patisiran group compared with the APOLLO-placebo group.

Discussion
The second interim analysis of the Global OLE study presents the results at 24 months, including data from up to 48 months of patisiran treatment, which represents the longest treatment period reported to date for an RNAi therapeutic.These data indicate that the improvements in mNISþ7 and Norfolk QOL-DN scores seen with patisiran in parent studies were durable over 24 months of additional patisiran treatment.Furthermore, the overall safety of patisiran during this additional treatment period in the Global OLE was consistent with that previously reported up to 12 months [10], and patisiran continued to demonstrate an acceptable safety profile at Global OLE 24 months.
Of interest for this post hoc assessment was the significant and sustained reduction in NfL levels observed with patisiran treatment in patients with ATTRv amyloidosis with polyneuropathy, supporting the potential utility of NfL as a biomarker for this disease.
In the APOLLO and phase II OLE studies, patisiran treatment lowered levels of NfL and led to improvements in neuropathy measured by mNISþ7 compared with their parent study baselines.Additionally, patisiran also improved QOL vs. parent study baseline, with changes in Norfolk QOL-DN score mirroring the pattern seen in the mNISþ7 results.The lowering of NfL levels, as well as the improvements in the clinical efficacy assessments of mNISþ7 and Norfolk QOL-DN, were sustained over an additional 24 months of treatment in the Global OLE study, demonstrating that the effect of patisiran on NfL levels corresponds with the clinical effectiveness of patisiran.In the Global OLE, initiation of patisiran in the APOLLO-placebo group also led to significant reduction in NfL levels and prevented further worsening of polyneuropathy and QOL, which had been seen during APOLLO [6], further supporting an association between NfL levels and clinical outcomes with patisiran.
At parent study baseline, patients from both APOLLO groups had more severe disease and higher NfL levels compared with the phase II OLE-patisiran group.This observation supports the potential relationship between increasing disease severity (as measured by, for example, worse PND score) and higher NfL that was observed by Kapoor et al. and Maia et al. in ATTRv amyloidosis [27,41,42], and suggests the possibility that NfL levels, in addition to being a marker of disease progression or treatment response, could also serve as an indicator of disease severity.An association between increased disease severity and higher NfL levels has also been observed in other neurologic diseases, such as Huntington's disease [20], chronic inflammatory demyelinating polyneuropathy [22], and Charcot-Marie-Tooth disease [24].Interestingly, however, a prior analysis of the APOLLO data showed that there was no correlation between an individual's NfL levels and mNISþ7 scores at study baseline, regardless of age adjustment [28].This lack of correlation may relate to differences in patient populations and/or outcomes assessed in previous studies on NfL in ATTRv amyloidosis compared with APOLLO, which included a broader range of TTR variants and patients at different disease stages [6,27,[41][42][43].These different study groups may have included patients with involvement of varying nerve fibres, which could differentially affect NfL levels, as NfL is not equally expressed in all peripheral nerve types affected by ATTRv amyloidosis [42].In addition, correlations between NfL and PND score were assessed in two of the other studies that investigated NfL, rather than between NfL and mNISþ7 [27,42].
It is of interest that the pattern of change in NfL and the scores measuring polyneuropathy (mNISþ7) and QOL (Norfolk QOL-DN) demonstrate certain differences.For example, in the Global OLE, while initiation of patisiran treatment led to a clear reduction in NfL in the APOLLOplacebo group, there was a relatively small improvement in Norfolk QOL-DN, and mNISþ7 remained stable.These differences may be because NfL is a biologic indicator of active nerve damage, while scores such as mNISþ7 reflect the burden of cumulative nerve damage on the patient's overall clinical picture.Thus, NfL may have the potential to serve as an 'acute-phase' marker of nerve damage in ATTRv amyloidosis.Indeed, it has been referred to as 'the neurologist's troponin' and 'the neurologist's C-reactive protein' in the literature [44,45].Notably, the differences in the behaviours of different assessment methods to demonstrate treatment response indicate the value of using multiple measures to monitor a multisystem disease like ATTRv amyloidosis.
The decrease in NfL levels in APOLLO with patisiran treatment became significant between Day 21 and Day 126.Although this reduction in NfL levels occurred later than the reduction in TTR levels in the APOLLO study, in which the first documented TTR reduction took place around Week 3 of treatment in the patisiran group, eventually levels of both biomarkers demonstrated a similar trend, namely a significant decrease from baseline [6].Although RNAi has been shown to lead to a reduction in TTR levels within days, it takes several months for the beneficial effects of treatment to become clinically measurable [46].
Following the relatively steep reduction in NfL levels in the early stage of treatment in the APOLLO-patisiran group, the subsequent decrease in NfL levels with continuing treatment became more gradual.Once NfL levels had decreased to a certain point, they appeared to remain stable over an additional treatment period, and during this follow-up period they did not decrease to levels seen in healthy controls [28].The NfL levels of patients in the APOLLOplacebo group decreased to a similar value to the APOLLOpatisiran group at Global OLE 24 months, but the overall clinical burden of the APOLLO-placebo group remained higher, as demonstrated by higher mNISþ7 and Norfolk QOL-DN scores.Of the three treatment groups, the NfL level at Global OLE 24 months was lowest in phase II OLEpatisiran patients, although they did not decrease to levels seen in healthy controls either [28].This suggests that the lowest level of NfL reached in patients with ATTRv amyloidosis receiving patisiran treatment may depend on pretreatment levels of NfL and/or disease severity.It is unclear whether this lowest base level of NfL in patients with ATTRv amyloidosis is due to an ongoing, low level of axon injury or axon injury accumulated prior to treatment; for  some patients, this may also represent central nervous system involvement.However, it is also possible that with longer-term continued patisiran treatment, NfL levels may decrease further towards those seen in healthy controls [28].
The stabilization or improvement of neuropathy and QOL during long-term patisiran treatment indicates the value of early intervention.Importantly, the onset of treatment stabilized the deterioration in ambulation, which is part of the natural history of the disease [47,48].These results are further supported by the post hoc analysis of mortality, in which the lowest exposure-adjusted mortality rates were observed in phase II OLE-patisiran patients, who were treated for the longest time and from the earliest stage of disease.
The results presented here collectively support the value of monitoring NfL levels in ATTRv amyloidosis, although some limitations of the study should be recognized.The healthy control group referenced was not age-, sex-, and race-matched to the phase II OLE-patisiran patients and no variables other than age were adjusted for when making between-group comparisons, which limits the comparisons made between both groups.Furthermore, although NfL is a promising potential biomarker for monitoring disease progression and treatment response, it is not specific to ATTRv amyloidosis.However, it may still have potential utility for clinicians treating the disease as either a marker of disease severity, a measure of disease progression, or a method of assessment of response to treatment.NfL testing could be especially valuable since existing assessment methods for polyneuropathy, such as mNISþ7 [30], can be cumbersome to perform, require training, and some are not routinely used in clinical practice.
The results of this study suggest several promising areas for further investigation.An evaluation of potential correlations between NfL levels and baseline measures of disease (such as the mNISþ7 or PND score) may also be of interest.However, interpretation of such an analysis would be challenging given that NfL levels may reflect active or continuing neuronal damage at a single time point and mNISþ7 or PND scores would reflect the clinical impact of the cumulative nerve damage over time.It may also be useful to conduct studies to examine whether NfL can be used as a biomarker for early detection of disease onset in patients carrying TTR variants, enabling earlier intervention and potentially preventing polyneuropathy progression.Further studies to elucidate NfL levels in patients with ATTR amyloidosis with predominant cardiomyopathy could also be conducted.In addition, measuring NfL levels over time in patients who receive treatments for ATTR amyloidosis other than patisiran could provide insights into treatment response/non-response for these medications.The integrated safety population encompasses all patients exposed to patisiran.Data are recorded from first patisiran dose in either the APOLLO, phase II OLE, or Global OLE studies until data cut-off (7 October 2019).b For each patient, exposure in years is defined as: (last dose date of study drugfirst dose date of study drug þ 91)/365.25.Total patient-year exposure time is calculated as the sum of each patient's time using minimum of exposure in years or follow-up in years.c Only deaths from the period of first dose of patisiran to 90 days after last dose are included.d Exposure-adjusted mortality rate is calculated as: (total number of deaths/total patient-years of exposure) Â 100.CI: confidence interval; OLE: open-label extension.

Disclosure statement
ST, EA, JV, and PN are all employed by Alnylam Pharmaceuticals and report ownership of shares in Alnylam Pharmaceuticals.CH was employed by Alnylam Pharmaceuticals at the time of this analysis and reports ownership of shares in Alnylam Pharmaceuticals.MP has participated in clinical trials sponsored by Akcea, Alnylam Pharmaceuticals, and Pfizer, and has received consulting fees from Akcea, Alnylam Pharmaceuticals, Biogen-Idec, Pfizer, and Vertex Pharmaceutical.DA has participated in clinical trials sponsored by Akcea and Alnylam Pharmaceuticals, and has received consulting fees advisory from Alnylam Pharmaceuticals, Bridgebio, Pfizer, and AstraZeneca.TC has participated in clinical trials sponsored by Akcea, Alnylam Pharmaceuticals, Eidos, Ionis, Prothena, and Pfizer.MU reports grants, personal fees, and non-financial support from Alnylam Pharmaceuticals and Pfizer; grants from Prothena; and personal fees from Janssen Pharmaceutical K.K., outside the submitted work.

CFigure 2 .
Figure 2. Integrated changes in mNISþ7 during the parent studies and the Global OLE.(A) APOLLO.(B) Phase II OLE.Data are integrated change from parent study baseline in mNISþ7.(C) Integrated change in Norfolk QOL-DN score during the parent studies and the Global OLE from the APOLLO study.a For APOLLO patients initiating alternative ATTRv amyloidosis treatment, mNISþ7 assessments after alternative treatment are treated as missing.The APOLLO-placebo arm began patisiran treatment at the start of the Global OLE.APOLLO mNISþ7 parent study baseline (mean (SD)): APOLLO-placebo ¼ 74.6 (37.0);APOLLO-patisiran ¼ 80.9 (41.5) [6].b Phase II OLE mNISþ7 parent study baseline (mean (SD)): 53.0 (35.6).c Data are integrated change from parent study baseline in Norfolk QOL-DN.The APOLLO-placebo arm began patisiran treatment at the start of the Global OLE.APOLLO Norfolk QOL-DN parent study baseline (mean (SD)): APOLLO-placebo ¼ 55.5 (24.3);APOLLO-patisiran ¼ 59.6 (28.2) [6].ATTRv: hereditary transthyretin (v for variant); BL: baseline; CI: confidence interval; mNISþ7: modified Neuropathy Impairment Score þ 7; Norfolk QOL-DN: Norfolk Quality of Life-Diabetic Neuropathy questionnaire; OLE: open-label extension; SD: standard deviation.

Figure 3 .
Figure 3. Integrated change in NfL levels during the parent studies and the Global OLE.(A) APOLLO.(B) Phase II OLE.Data are integrated change from parent study baseline in NfL level.The APOLLO-placebo arm began patisiran treatment at the start of the Global OLE.BL: baseline; CI: confidence interval; NfL: neurofilament light chain; OLE: open-label extension.

Table 1 .
Demographics and disease characteristics at parent study baseline of patients with NfL measurements.

Table 2 .
Change in PND score from parent study baseline and Global OLE enrolment to Global OLE 24 months.

Table 3 .
Mean NfL levels (SD) in the APOLLO and Global OLE studies.The APOLLO-placebo arm began patisiran treatment at the start of the Global OLE.b Statistically significant difference compared with APOLLO-placebo parent study baseline.c Statistically significant difference compared with APOLLO-patisiran parent study baseline.d Statistically significant difference compared with APOLLO-placebo Global OLE baseline.NfL: neurofilament light chain; OLE: open-label extension; SD: standard deviation. a

Table 4 .
Mean NfL levels (SD) in the phase II OLE and Global OLE studies.Patients in the phase II OLE study continued with patisiran treatment at the start of the Global OLE study.NfL: neurofilament light chain; OLE: open-label extension; SD: standard deviation. b

Table 5 .
Exposure and overall safety in the Global OLE.Patients with !1 event APOLLO-placebo (n ¼ 49) APOLLO-patisiran (n ¼ 137) Phase II OLE-patisiran (n ¼ 25) Global OLE all patients (n ¼ 211) All deaths summarized, including deaths due to AEs that are not treatment emergent.Data as of interim cut-off 7 October 2019.Exposure extends 21 days past the final dose of patisiran.AE: adverse event; IRR: infusion-related reaction; OLE: open-label extension; SAE: serious adverse event. a

Table 6 .
Integrated exposure-adjusted mortality rates in the parent and Global OLE studies.