Preclinical studiesSix-month NHP IV study

Administration of 1.1 × 1014 vg/kg OA with or without prednisolone to juvenile cynomolgus monkeys via a single IV infusion was generally well tolerated for up to 6 months post IV infusion and was not associated with mortality, clinical observations, or any pathology (gross or microscopic) findings in the heart at 6 week or 6 month postdose.

No OA-related increases were observed in plasma cTnI. Any other differences in plasma cTnI, compared to baseline and/or control data, were not attributed to test article because they were sporadic, consistent with biologic variation, negligible in magnitude, and/or lacked dose relationship. Additionally, no changes in PR interval, QRS duration, QT interval, corrected QT (QTc) interval, heart rate, and abnormal waveforms or arrhythmias were observed during Week 1, 6, 13, or 26 in any of the animals dosed with or without prednisolone.

12-week GLP mouse studies Study 1 (a single dose temporal vein injection study in the neonatal FVB/NJ mouse followed by a 12-week observation period)

Administration of a single IV dose of OA to PND1 FVB/NJ mice was clinically tolerated at doses of 7.9 × 1013 and 2.37 × 1014 vg/kg. Moribundity and unscheduled mortality were recorded in mice administered 3.91 × 1014 vg/kg, primarily during the weeks 4-7 of the study. Of the 29 mice presented for examination at unscheduled intervals, morbidity/death of 13 mice was attributed to OA-related atrial thrombi in the heart (Table 2). Thrombi were not observed macroscopically in the heart, but noted microscopically and were characterized by variable collections of fibrin, cellular debris, and inflammatory cells in the atria/auricles that were often adjacent to or attached to plump, reactive endothelium. No evidence of thrombus or embolization was noted in any other organ. The cause of morbidity/death for the remaining 16 OA-treated mice examined that died prior to scheduled sacrifice was undetermined. In addition, while no gross/macroscopic lesions were noted in the heart of any of the preterminal mice whose moribundity/mortality were attributed to the test article, these mice were noted with other microscopic heart findings including minimal or mild microscopic degeneration/regeneration in the heart. These findings, together with thrombi in the heart, may have contributed to the deteriorating clinical condition. OA-related mortality and adverse clinical observations and body weight changes did not occur at doses ≤2.37 × 1014 vg/kg.

Table 2 Summary of microscopic findings in mice-early deaths.

OA-related microscopic findings in the heart at scheduled sacrifice intervals comprised of degeneration/regeneration observed at 3-, 6- and/or 12-weeks postdose across all treatment groups, and thrombi at week 6 in mice dosed at 3.91 × 1014 vg/kg. Due to the unscheduled mortality up to weeks 6/7 and the scheduled euthanasia at week 3 and weeks 6/7, no mice administered 3.91 × 1014 vg/kg were available for the week 12 time point. The thrombi in the heart appeared morphologically similar to those described in the mortality section above. The degeneration/regeneration was characterized by shrunken, wavy cardiomyocytes and/or clear spaces between cardiomyocytes (degeneration), associated slight mononuclear cell hypercellularity, crowded residual myocyte nuclei, or occasional cardiomyocyte nuclei that were enlarged 2–4 times those of adjacent cells (regeneration). In general, the change was most prominent in the myocardium near the endocardial (luminal) surfaces of the left and right ventricular free walls, interventricular septum, and bases of papillary muscles, although the atrial myocardium was also sometimes involved. At week 12 the heart degeneration/regeneration was minimal, in contrast to mild heart degeneration/regeneration seen in some mice at weeks 3 and 6/7 (less severe overall at week 12 than at other time-points), which was consistent with repair/healing over time (Table 2).

At 3-weeks postdose, degeneration//regeneration in the heart was seen at all doses with severity being mild at doses ≥2.37 × 1014 vg/kg, and minimal at 7.9 × 1013 vg/kg, which supported a dose-response. While still present at 6 week postdose at minimal or mild severity, only one mouse administered 7.9 × 1013 vg/kg was noted with degeneration//regeneration in the heart. The finding was noted more at doses ≥2.37 × 1014 vg/kg with average severities in these groups being slightly less than the average severities of the same groups at week 3, possibly showing a trend toward lesion resolution over time. Thrombi (moderate severity) in the heart were present only in mice administered 3.91 × 1014 vg/kg at this interval. At 12 weeks postdose, degeneration//regeneration in the heart was noted in mice administered 7.9 × 1013 or 2.37 × 1014 vg/kg at a severity of minimal, in contrast to mild changes seen at weeks 3 and 6/7. This indicated a trend toward lesion resolution over time.

Minimal to mild clinical pathology changes consistent with possible inflammation (increased lymphocytes, monocytes, eosinophils and/or neutrophils, decreased reticulocytes) were observed at all dose levels as early as Week 3. Increased creatine kinase activity was noted as early as Week 3 in mice given 3.91 × 1014 vg/kg and then at later timepoints in mice dosed at ≥7.9 × 1013 vg/kg. These changes were possibly associated with the microscopic findings of degeneration/regeneration in the heart.

Study 2 (a single dose of AVXS-101 (Lot Number 600443) via temporal vein injection in neonatal FVB/NCrl mice followed by a 12-week observation period)

Administration of OA in mice resulted in mortality between weeks 4 and 12 and when the cause of death could be ascribed, it was associated with atrial thrombosis at ≥2.4 × 1014 vg/kg. OA-related microscopic findings were dose-dependent, noted in the atria and ventricular myocardium and mostly in mice administered ≥2.4 × 1014 vg/kg. OA-related ventricular myocardial findings included minimal to slight myocardial degeneration/necrosis, and/or minimal to slight mononuclear cell inflammation, whereas, the atrial findings consisted of minimal to moderate thrombus formation, slight to marked dilation, minimal to slight fibroplasia. Atrial thrombosis and associated atrial wall changes were dose-related and present at higher incidences and/or severities in unscheduled death animals administered ≥2.4 × 1014 vg/kg, which suggested that atrial lesions may have been associated with morbidity (in moribund sacrifice animals) and mortality. The microscopic findings in the heart were sometimes associated with macroscopic findings of large size, abnormal shape, and/or large atrium, and increased heart weights in some animals administered 3.0 × 1014 vg/kg. There was no evidence of thrombi in the cardiac chambers upon gross pathology assessments. The only OA-related hematology finding was minimally higher neutrophil count on PND 42 and 84 in males administered ≥2.4 × 1014 vg/kg, suggestive of an inflammatory response (Tables 3–7).

Table 3 Summary of microscopic findings in mice—terminal euthanasia (weeks 3, 6/7 and 12).Table 4 Incidence and severity of OA-related microscopic findings in mice—unscheduled sacrifices.Table 5 Incidence and severity of OA-related microscopic heart findings in mice—week 3 (PND 21) terminal sacrifice.Table 6 Incidence and severity of OA-related microscopic heart findings in mice—week 6 (PND 42) terminal sacrifice.Table 7 Incidence and severity of OA-related microscopic heart findings in mice—week 12 (PND 84) terminal sacrifice. Clinical trial data

A total of 101 patients were dosed with IV OA during the clinical trials; of whom 99 received the therapeutic dose and are included in this analysis. The remaining 3 patients did not receive a dose lower than the therapeutic dose, which was not developed further.

Adverse events

Cardiac adverse events by preferred term are presented in Table 7. Cardiac AEs were reported in 17 patients (17.2%) and consisted mainly of elevated enzymes (cardiac troponin and CK-MB) and changes in heart rate (tachycardia and bradycardia) without clinical significance. All other AEs as presented in Table 8 were considered to be related to the underlying SMA, not a cardiac etiology. No adverse events of cardiac thrombi were reported. No patient died of a cardiac AE.

Table 8 Summary of treatment-emergent cardiac adverse events by preferred term. Echocardiogram and electrocardiogram

With respect to echocardiogram data, baseline LVEF was normal or borderline high for all but 2 patients who had borderline low values. Two patients each had a single post-treatment occurrence of LVEF values that were slightly below the normal range (51.0% and 55.4%, respectively) which were not considered clinically meaningful. At baseline, the mean (±SD) LVEF was 65.1% (±5.19), and the median LVEF was 64.9% (N = 85). Six months after treatment (N = 73), the mean LVEF was 68.8% (±5.90), and the median LVEF was 69.3%. At 12 months (N = 56) and 18 months (N = 23), mean LVEF was similar to the means at baseline and at 6 months after infusion. Baseline LVFS was normal or borderline high for all patients except for a borderline low value (27.1%) in one patient from SPRINT. No patient had a low post-treatment LVFS value. At baseline, the mean (±SD) LVFS was 36.1% (±4.37) and the median LVFS was 35.4%. Six months after treatment the mean LVFS was 38.5% (±4.83) and the median LVFS was 38.0%. At 12 months and 18 months post dosing, the mean LVFS values were similar to the means at baseline and at 6 months after dosing. Upon review of the echocardiogram results, no patients had evidence of cardiac thrombi.

No patient in any of the clinical trials was reported to have persistent potentially clinically significant ECG abnormalities, nor were any findings associated with abnormal clinical signs or symptoms. No abnormalities in PR interval, QRS complex or QTc were present.

Laboratory data

Cardiac troponin I (cTnI) as a marker of potential cardiac toxicity was collected in START, and was not collected initially in STR1VE-EU, STR1VE-AP, or SPRINT; however, these protocols were amended to include measurement of cTnI in patients enrolled following the amendment. Cardiac troponin I was not collected in STR1VE. A total of 22 patients had at least one cardiac troponin value obtained and 11/101 (10.9%) patients had baseline values obtained, however, not all patients had cTnI performed; some had cardiac troponin I and/or T performed at local laboratories as compared to central laboratory as this was allowed per protocol. Therefore, the data presented is a mix of high sensitive and normal cTn values. A value of >0.05 ng/mL was considered elevated; the % changes from baseline could not be calculated due to many missing values. The limited data show that cardiac troponin was elevated in some patients at baseline; while in other patients, some increases were seen after treatment with OA and all but one elevation normalized by the Month 2 visit (Table 9). Due to limitations associated with archival, cardiac troponin values from START are not available.

Table 9 Summary of patients with document troponin values in OA clinical trials.

Creatine Kinase in Blood-Muscle/Brain (CK-MB) was obtained and all patients had elevations noted. No consistent changes from baseline in mean CK-MB values were observed in any of the studies, and most patients with elevated post-treatment CK-MB test results had elevated baseline values.

Data from long-term follow up studies, registry and post-marketing reports

Approximately 2269 patients have been dosed cumulatively since OA’s first approval in May 2019 through global commercial and early access programs as of 23-May-2022. A total of 418 ICSRs containing at least one preferred term indicative of a cardiac event were identified from all sources including the long-term follow up study from START (5 reports in 5/13 patients) and no cardiac AEs were identified from the other long-term follow study. Atrial or ventricular septal defect was reported in 9 of 418 cases.

The most commonly reported cardiac adverse events were elevated cardiac troponin I, troponin T or unspecified troponin. Overall, 191 (45%) case reports included elevated cardiac troponin as a reported cardiac event. In 54% of reported cases (227/418), cardiac events were reported without associated elevations in cardiac troponin values. One hundred thirty one of 418 (31%) cases had reports of elevated cardiac troponin without any other cardiac abnormalities associated with the cardiac troponin elevation. In other words, they were isolated, asymptomatic elevations in cardiac troponin. The degree of elevations is not necessarily known due to limitation in completeness of data provided given the real-world setting. For the remainder of cases, cardiac troponin elevations with other signs/symptoms were reported in 60/418 (14%) but none of those signs/symptoms were of cardiac etiology (further discussed below). The 418 case reports are summarized in Table 10.

Table 10 Summary of cardiac events from post-marketing sources.

Of the 418 ICSRs, 32 cases included an outcome of patient death. Of these 32 reports, the causes of death in 29 patients were respiratory in nature, consistent with complications of the underlying SMA. Of the 3 remaining cases, one patient (age 19 months) died of complications after acute shock (including cardiovascular collapse) resulting from prolonged bleeding events due to traumatic insertion of nasogastric tube in the setting of thrombocytopenia. Another patient (age 4 months) initially was reported to have thrombotic microangiopathy (TMA) approximately one week after OA administration. While TMA was noted to have recovered, the patient ultimately developed septic shock resulting in multi-organ system failure (including cardiac failure) and death. For the third patient, (age 15 months) reported clinical details indicated pre-existing and recurrent infections progressed to sepsis. Multiple organs including the heart were involved, with reported myocarditis, with a fatal outcome. The reported cardiac troponin, post-cardiopulmonary resuscitation was 8210 ng/mL.

The second commonly reported “cardiac event”, other than elevated cardiac troponin, identified by the given search criteria was dyspnea (reported in 95 ICSRs) which were assessed as more likely secondary to the underlying SMA. The other commonly reported events were tachycardia (in 54 case reports) or increased heart rate (in 20 case reports) which were more reported as likely secondary to compromised respiratory function, feeding difficulties or respiratory infections commonly seen in children with SMA.

While events of depressed cardiac function were reported, none of these events had a primary cardiac etiology. Such events included cardiac arrest (in 16 cases), cardio-respiratory arrest (in 5 cases), cardiac failure (in 1 cases), Cardiac failure congestive (in 1 case) and left ventricular dysfunction (in 1 case). In some cases, more than one such events were reported. In each of the reports, clinical conditions contributing to the events of depressed cardiac function included a respiratory event, multiorgan failure following delayed medical intervention of bleeding events due to thrombocytopenia (discussed for fatal cases), and fluid overload.

Myocarditis was reported in 3 cases, one of which is presented above (death event). In the second case, the patient had recurrent bacterial infection, respiratory infections, “persistent tachycardia and respiratory distress” requiring antibiotics and intubation. Baseline cardiac troponin was not available. Post-treatment cardiac troponin (not specified) was 63.5 pg/mL (reference: lower than 20) and then decreased to 3.3 pg/mL 10 days later. In the third case, the patient had a “lower respiratory tract infection” with elevated cardiac troponin I (80 ng/l, reference range not provided). Electrocardiogram and echocardiogram were reported as “normal”. The reporter indicated that “presumably this reflected a viral infiltration causing a myopericarditis or similar”.

Pericardial effusion and myopericarditis were reported in three and one cases, respectively. In the first case, baseline cardiac troponin was not reported. After OA administration, cardiac troponin (unspecified) values were available starting from week-1 with values fluctuating between 50 s – 60 s (reference <19, unit was not reported) for 2 months. Two and half months after OA administration, rhinorrhea was reported with cardiac troponin increased to 84, One week later, followed by pericardial effusion No infection was reported and COVID-19 testing was negative. Leukocyte counts were available and stable through the onset of rhinorrhea and were within normal range. No other information was available. In the second case, pericardial effusion was reported with ascites and edema. Total protein was reported as “abnormal”. Liver and renal function tests were reported as normal. In the third case, pericardial effusion was reported on an unknown date with body temperature of 38 degree Celsius. Troponin T was 0.085 ng/mL (reference range 0-0.14). In the fourth case, myopericarditis was reported on an unknown date. Troponin I was 80 ng/L (no reference available). Due to lack of clinical details, these four cases could not be further evaluated.

Based on further analysis, two-thirds of the reports of elevated cardiac troponin had concurrent reports of hepatic adverse events, most of which were abnormal liver functional tests (Table 11).

Table 11 Description of cardiac events with and without concurrent hepatic events.