Tenecteplase real-world data: A three phase sequential comparison

Introduction: The New Zealand (NZ) Central Region Stroke Network, serving 1.17 million catchment population, changed to tenecteplase for stroke thrombolysis in 2020 but was forced to revert to Alteplase in 2021 due to a sudden cessation of drug supply. We used this unique opportunity to assess for potential before and after temporal trend confounding. Patients and methods: In NZ all reperfused patients are entered prospectively into a national database for safety monitoring. We assessed Central Region patient outcomes and treatment metrics over three time periods: alteplase use (January 2018–January 2020); during switch to tenecteplase (February 2020–February 2021) and after reverting to alteplase (February 2021–December 2022) adjusting regression analyses for hospital, age, onset-to-needle, NIHSS, pre-morbid mRS and thrombectomy. Results: Between January 2018 and December 2022, we treated 1121 patients with Alteplase and 286 with tenecteplase. Overall, patients treated with tenecteplase had greater odds of favorable outcome ordinal mRS [aOR = 1.43 (95% CI = 1.11–1.85)]; shorter door-to-needle (DTN) time [median 52 (IQR 47–83) vs 61 (45–84) minutes, p < 0.0001] and needle to groin (NTG) times [118 (74.5–218.5) vs 185 (118–255); p = 0.02)]. Symptomatic intracerebral hemorrhage (sICH) rate was lower in tenecteplase group [aOR 0.29 (0.09–0.95)]. Findings similarly favored tenecteplase when comparing tenecteplase to only the second alteplase phase. There was no inter-group difference when comparing the two alteplase phases. Conclusions: Our results suggest that previously reported benefits from tenecteplase in a real-world setting were not likely attributable to a temporal confounding.


Introduction
2][3][4][5] Due to tenecteplase's practical advantages of a single bolus injection, non-inferiority alone is a compelling reason to make the change although some evidence suggests TNK might be superior especially in patients with large vessel occlusion (LVO). 1,57][8] The optimal stroke TNK dose is 0.25 mg/kg as there is clinical trial evidence of increased bleeding risk and worse outcomes with 0.4 mg/kg dose. 9

Tenecteplase real-world data: A three phase sequential comparison
However, real-world observational data are limited by potential temporal or inter-cluster confounding as beforeand after or treatment cluster (e.g. by center or stroke type such as LVO vs non-LVO) comparison are the usual employed methodologies. 2,4,7The NZ Central Region Stroke Network was an early universal adopter, but had to revert to alteplase following a global tenecteplase shortage.The resultant three different treatment phases provide a unique opportunity to address potential temporal confounding.

Methods
We extracted data from a compulsory national stroke reperfusion register comparing all adult patients treated in Central NZ with alteplase from 1 January 2018 to 1 March 2020 (phase 1) to those treated with tenecteplase from 2 March 2020 to 14 February 2021 (phase 2) and those treated with from 15 February 2021 to 31 December 2022 (phase 3).This register is pragmatic and primarily for assessing the type, efficacy, safety and time metrics of reperfusion therapy and hence comprehensive data on comorbidities and other variables such as risk factors are not routinely collected.The Central Region of NZ comprises 1,166,333 people serviced by one tertiary, four provincial hospitals, and four rural hospitals.Population-based thrombolysis rates were calculated per 100,000 person years.The primary efficacy outcome was 3-month modified Rankin Scale (mRS) (10-16 weeks; doctors/nurses collected; unblinded; 15% retrospective chart-review) using ordinal shift analysis.The main safety outcome was sICH using ECASS 3 criteria.Secondary patient outcomes included mRS dichotomized (0-2vs 3-6), death by day seven, and angioedema.We also assessed four process outcomes: doorto-needle time overall and limited to non-thrombectomy centers (where patients require transfer to another center to undergo thrombectomy), and needle-to-groin time for those who did undergo thrombectomy both overall and by nonthrombectomy center.For the main analysis we compared all alteplase treated patients with tenecteplase treated patients.In addition, we completed two sensitivity analyses: (1) tenecteplase compared to only the second alteplase phase and (2) first alteplase to second alteplase phase.
Data was analyzed using descriptive statistics, Mann-Whitney U test for non-parametric, and logistic regression for dichotomous and ordinal variables.Regression models included hospital, age, pre-morbid mRS, baseline National Institute of Health Scale (NIHSS), onset-to-needle time, and thrombectomy.Statistical analysis was performed using Stata IC 17.
NZ Health and Disability Ethics Committee approved registry data use without individual patient consent (HDEC 19STH/55).Authors (AR/AT) could access the complete centrally cleaned registry dataset and the RECORD reporting checklist was used.Study data are available from the corresponding author upon reasonable request.
See Supplemental Tables 1 and 2 for additional baseline variable comparisons.
Comparing tenecteplase to only the most recent alteplase (phase 3), similar results were noted except the dichotomous 3-month mRS result, which became non-significant (Table 3).There was no difference in patient outcome between the two alteplase phases (Table 4).

Discussion
In this opportunity real world study of unselected consecutively thrombolysed patients at tertiary, urban and non-urban secondary, and small rural hospitals, we found no evidence of harm related to tenecteplase use compared with alteplase and overall improved patient outcomes, with fewer adverse events, and reduced treatment delays.We previously presented similar findings looking at only the first two phases, 5 however, potential other intervening temporal confounders presented a limitation especially given the significant rise in thrombectomy rates between phases 1 and 2. This new data arising from the subsequent forced reversion to alteplase provides significant reassurance regarding the validity of the original results.In addition, the overall increased sample size has resulted in greater study power around the reduction of adverse events.
One finding that differs between our and other observational and trial data is the association of tenecteplase with reduced treatment delays.In the real-world setting, the 1-h alteplase infusion can present a barrier to in-CT thrombolysis due to concerns around a long gap between bolus and infusion, especially where tenecteplase boluses are routinely given before advanced imaging, and barriers to helicopter transport with an infusion pump.The demonstrated reduction in time delays with tenecteplase is even more significant consdiering that the tenecteplase phase coincided with the peak health system impact of the covid pandemic in New Zealand.It is possible that this real-world benefit in treatment speed contributes to reduced adverse and overall better patient outcomes with tenecteplase.Our sICH rates in tenecteplase and Alteplase groups aligns with other published data although higher sICH rates have been      reported in some tenecteplase trials. 2,3,9This difference may be attributable to higher doses, including use of a weighttiered dosing schemes in relevant trials. 3,9e ability to compare with alteplase cases both pre-and post-tenecteplase implementation is a unique feature of this study that addresses temporal confounding.The complete -needle Ɵme by study phase Door-to-needle Ɵme Door-to-needle Ɵme (non-

Figure 1 .
Figure 1.Door-to-Needle and Groin times by phase; overall and for non-thrombectomy centers.