A total of 482 participants enrolled in the study, and 205 (42.5%) completed it; 344 participants used the final version of STS101 at least once during the study and were included in the final device safety population (Fig. 2). The all subjects safety population included a total of 446 participants who used either the first or the final version of STS101 at least once during the study to treat a total of 9091 migraine attacks. Most early study discontinuations were due to an inadequate number of treated migraine attacks (defined as < 2 attacks per month per the FDA requirement [20]; 97/482 [20.1%] enrolled participants) and withdrawal of consent [83/482 (17.2%) enrolled participants]. A total of 34/446 (7.6%) participants experienced ≥ 1 TEAE that led to study drug discontinuation. No significant differences in demographic or clinical characteristics were noted between the all subjects safety population and final device safety population (Supplementary Table 1).

Study populations and disposition

The 344 participants included in the final device safety population used 8234 doses of study medication to treat a total of 6610 migraine attacks during the study. Of the 6610 migraine attacks treated, 1315 (19.9%) were treated with a second dose of study medication. A majority of participants had a monthly average of 2–4 treated attacks [170/344 (49.4%)].

The expansion population comprised 172 participants who used the final version of STS101 exclusively, treating 3394 migraine attacks during the study.

Baseline demographic and clinical characteristics of study participants are summarized in Table 1. The average duration of migraine prior to study participation was 18.5 years, with participants reporting an average of 7.5 headache days and 5.5 migraine attacks per month during the 3 months prior to screening (Table 2). Typical migraine symptoms included photophobia, phonophobia, and nausea for the majority of participants (97.4%, 95.1%, and 84.6%, respectively). Nearly half (47.4%) reported aura, and nearly one-third (31.7%) reported allodynia.

TEAEs in the final device safety population are summarized by participant in Table 3. The most commonly reported TEAEs by participant [n/N (%)] were nasal discomfort [39/344 (11.3%)], dysgeusia [26/344 (7.6%)], nasal congestion [18/344 (5.2%)], nasopharyngitis [18/344 (5.2%)], and nausea [17/344 (4.9%)]. The majority of TEAEs reported were of mild or moderate severity. Only 9 (2.6%) participants in this population experienced ≥ 1 severe TEAE. A severe TEAE of migraine was reported in 2 (0.6%) participants. Severe TEAEs of abdominal pain, acute myocardial infarction, ligament rupture, nasal discomfort, nasopharyngitis, rhinalgia, sinus pain, throat irritation, upper abdominal pain, and vomiting were reported by 1/344 (0.3%) participant each.

A total of 89/344 (25.9%) participants experienced an event considered related to study treatment. The most commonly reported treatment-related TEAEs were nasal discomfort [38/344 (11.0%)], dysgeusia [26/344 (7.6%)], rhinalgia [16/344 (4.7%)], nasal congestion [15/344 (4.4%)], and rhinorrhea [9/344 (2.6%)]).

Of the total 6610 migraine attacks treated with the final version of STS101, 945 (14.3%) were associated with a TEAE. Events occurring in ≥ 2% of attacks were nasal discomfort [425 (6.4%)], nasal congestion [253 (3.8%)], and dysgeusia [196 (3.0%)] (Table 3). A total of 777/6610 (11.8%) treated attacks were associated with TEAEs considered related to study treatment, the most common (occurring in ≥ 1% of attacks) being nasal discomfort [421/6610 (6.4%)], nasal congestion [250/6610 (3.8%)], dysgeusia [196/6610 (3.0%)], and rhinorrhea [68/6610 (1.0%)]. Most TEAEs were mild [844/6610 (12.8%)] or moderate [142/6610 (2.1%)] in severity. Severe TEAEs were reported for only 13/6610 (0.2%) treated attacks. Local TEAEs occurred at similar rates between the 1315 migraine attacks treated by participants who administered a second STS101 dose to treat ≥ 1 attack relative to the 5295 migraine attacks treated by participants who only used a single dose for treatment (Table 4). The incidence of local TEAEs was highest in the first 3-month period and decreased over time. There were no deaths reported.

One participant of 344 (0.3%) experienced an SAE. A 45-year-old Caucasian male with a prior myocardial infarction and a diagnosis of bipolar disease, which were not disclosed to the investigator at study enrollment, experienced a serious TEAE of non-ST elevation myocardial infarction. Both his previous myocardial infarction and his active bipolar disease were trial exclusions and, as noted, not disclosed. The participant developed intermittent chest pain approximately 2 h after use of STS101 and was admitted to the emergency department in the evening after experiencing sharp chest pain. The event was not considered life-threatening but was assessed as related to study medication. Unknown to the investigator, the participant used several more doses of study medication after the SAE without experiencing any additional TEAEs.

Five pregnancies were reported during the study; 4 of the pregnant participants were removed from the study, and 1 was lost to follow-up. Two participants delivered healthy babies; the outcomes of the other pregnancies remain unknown.

A similar safety profile was seen in the all subjects safety population and can be found in Supplementary Tables 2 and 3.

Exploratory efficacy data from the 172 participants who participated in the expansion part of the study, and exclusively used the final device, are presented here. These participants treated a total of 3394 migraine attacks during the study.

Treatment was associated with rapid increases in pain freedom, from 12.7% (338/2665) and 36.6% (1014/2768) of treated attacks at 1 and 2 h post-dose, respectively, to 67.1% (1507/2247), 85.5% (1701/1990), and 88.8% (1521/1713) of treated attacks at 4, 24, and 48 h post-dose (Fig. 3).

Proportion of migraine attacks with pain freedom over time following the administration of STS101 (expansion population*). *Expansion population included data from 172 participants who treated a total of 3394 migraine attacks; data were available for 2665, 2768, 2247, 1990, and 1713 attacks at 1, 2, 4, 24, and 48 h post-dose, respectively

Treatment also led to rapid increases in MBS freedom, from 26.6% (710/2665) and 54.3% (1504/2768) of treated attacks at 1 and 2 h post-dose, respectively, to 79.6% (1789/2247), 91.3% (1816/1990), and 91.5% (1567/1713) of treated attacks at 4, 24, and 48 h post-dose (Fig. 4).

Proportion of migraine attacks with MBS freedom over time (photophobia, phonophobia, or nausea) following the administration of STS101 (expansion population*). *Expansion population included data from 172 participants who treated a total of 3394 migraine attacks; data were available for 2665, 2768, 2247, 1990, and 1713 attacks at 1, 2, 4, 24, and 48 h post-dose, respectively. MBS most bothersome symptom

Headache relief was reported for 37.9% (1011/2665) and 66.5% (1840/2768) of treated attacks at 1 and 2 h post-dose, respectively, increasing to 89.1% (2002/2247), 94.3% (1877/1990), and 93.5% (1602/1713) of treated attacks at 4, 24, and 48 h post-dose (Fig. 5).

Proportion of migraine attacks with headache relief over time following the administration of STS101 (expansion population*). *Expansion population included data from 172 participants who treated a total of 3394 migraine attacks; data were available for 2665, 2768, 2247, 1990, and 1713 attacks at 1, 2, 4, 24, and 48 h post-dose, respectively

For many attacks [718/2768 (25.9%)], treatment was associated with sustained pain freedom from 2 to 24 h post-dose (Supplementary Fig. 1). The proportion of attacks with sustained pain freedom increased from 68/356 (19.1%) at month 1 to 13/55 (23.6%) at month 12, with some month-to-month variability.

Similarly, treatment was associated with sustained freedom from MBS from 2 to 24 h post-dose in 1063/2768 (38.4%) of migraine attacks, with a rate of 141/356 (39.6%) at month 1, which was generally sustained with some month-to-month variability, to 19/55 (34.5%) at month 12 (Supplementary Fig. 1).

Rescue medication was used in 3.5% and 4.6% of migraine headache attacks within 24 and 48 h after the initial STS101 dose, respectively. A second dose of study medication was used in 19.3% and 19.4% of migraine headache attacks within 24 and 48 h after the initial STS101 dose.

Most participants rated treatment as good or very good at month 3 [104/124 (83.9%)], month 6 [87/96 (90.6%)], and month 12 [113/142 (79.6%)].

Similarly, most rated treatment as easy or very easy to use at month 3 [113/124 (91.1%)], month 6 [90/96 (93.8%)], and month 12 [132/142 (93.0%)] (Fig. 6).

Participant ease of use impression (expansion population*). *n = 165

Finally, most participants indicated that they would be likely or very likely to use the treatment again at month 3 [95/124 (76.6%)], month 6 [75/96 (78.1%)], and month 12 [83/142 (58.5%)]. At month 12, most participants agreed or strongly agreed [89/140 (63.6%)] that STS101 treated their migraine attacks more consistently than their prior medications.