Apixaban (2) – Eliquis®

• Clinical trials
  • To demonstrate additional benefit compared with the appropriate comparator therapy, ‘vitamin K antagonists’, the pharmaceutical manufacturer’s dossier cites the results of a randomised, controlled trial involving patients suitable for treatment with vitamin K antagonists (ARISTOTLE trial; n=18,201).

Patients with non-valvular atrial fibrillation for the prevention of strokes and systemic embolisms

• For the prevention of strokes and systemic embolisms in adult patients with non-valvular atrial fibrillation (NVAF) and one or more risk factors, such as a history of stroke or TIA (transient ischaemic attack), age ≥75 years, hypertension, diabetes mellitus, symptomatic heart failure (NYHA class ≥II), there is an indication for a minor additional benefit for apixaban compared with the appropriate comparator therapy.
• The certainty of the evidence (probability of additional benefit) is classified as ‘indication’.
• The G-BA classifies the extent of the additional benefit of apixaban (new therapeutic indication) for the VKA population as minor, based on the criteria in Section 5(7) of the AM-NutzenV, taking into account the severity of the disease and the therapeutic goal in the treatment of the disease.
• Compared with the appropriate comparator therapy, vitamin K antagonists (VKAs), this represents, in accordance with Section 5(7) in conjunction with Section 2(3) of the AM-NutzenV, a moderate and not merely minor improvement in treatment-related benefit that has not previously been achieved.
• Mortality (overall survival)
  • With regard to overall survival, there is a statistically significant difference between apixaban and the appropriate comparator therapy, warfarin, for the overall population of the ARISTOTLE study.
  • The statistically significant difference, with a p-value of 0.047, is described by the EMA as borderline significance.
  • The hazard ratio is 0.89 [CI 0.80; 1.00], [proportion of events 6.6% vs. 7.4%], with an absolute risk reduction (ARR) of 0.8%.
  • No superiority of apixaban was demonstrated in the individual endpoints of cardiovascular and non-cardiovascular mortality.
  • In study centres where a high TTR was achieved, the effects were less pronounced or lost altogether.
  • TTR values for the centres, categorised into quartiles (TTRc), show an increase in the HR from 0.81 in the first quartile (TTRc < 58.1 per cent) to an HR of 1.04 (TTRc ≥ 72.2%) in the fourth quartile.
  • The results were not statistically significant within the individual quartiles.
• Morbidity – Combined endpoint (primary endpoint): stroke and systemic embolisms (symptomatic)
  • With regard to the primary endpoint, a statistically significant result was observed for the overall population of the ARISTOTLE study, with an HR of 0.79 (95% CI 0.66–0.95).
  • TTR values for the centres (TTRc), categorised into quartiles, show an increase in the HR from 0.78 in the first quartile (TTRc < 58.1%) to an HR of 0.81 (TTRc ≥ 72.2%) in the fourth quartile.
  • The results were not statistically significant within the individual quartiles.
  • In the ARISTOTLE trial, there were far more strokes than systemic embolisms, so it is to be expected that the result for this endpoint is largely determined by the stroke component.
• Morbidity – stroke (ischaemic, haemorrhagic or of unknown cause) (symptomatic)
  • For the endpoint of stroke (ischaemic, haemorrhagic or of unknown cause), there is a statistically significant difference between apixaban and the appropriate comparator therapy, warfarin (HR 0.79 [CI 0.65; 0.95], ARR 0.6% [event rate 2.2% vs. 2.8%].
  • The difference in favour of apixaban was primarily due to the minor decrease in the rate of haemorrhagic strokes (40 (0.4%) vs. 78 (0.9%) patients with an event; HR 0.51 [0.35; 0.75]; ARR 0.5%).
  • There were no statistically significant differences in the incidence of ischaemic strokes or strokes of unknown cause.
  • For the endpoint of stroke leading to disability (ischaemic, haemorrhagic or of unknown cause), there was no statistically significant difference between apixaban and warfarin (HR = 0.84; 95% CI [0.58; 1.20]).
• Health-related quality of life
  • The endpoint of health-related quality of life was not assessed in the ARISTOTLE trial.
  • Consequently, no usable data were available for this endpoint.
• Side effects – bleeding events
  • Both major bleeding events (3.6% vs. 5.1%; ARR 1.5%; HR 0.69 [0.60; 0.80]) and clinically relevant non-major bleeding events (3.5% vs. 4.9%; ARR 1.4%; HR 0.70 [0.60; 0.80]) occurred less frequently with apixaban than with warfarin.
  • The result was statistically significant in each case, and this also applied to the combined endpoint of major and clinically relevant non-major bleeding (6.8% vs. 9.7%; ARR 2.9%; HR 0.68 [0.61; 0.75]).
  • In the overall analysis, proof of an effect modification (p = 0.029) by the characteristic of geographical region remained for the combined bleeding endpoint with regard to relevant effect modification.
  • Such bleeding events occurred more frequently in all regions among patients treated with warfarin than among those treated with apixaban.
  • The result was statistically significant for all regions.
  • The effect was strongest in Asia/Pacific and weakest in North America.
  • The effect was minor in Europe (HR 0.74 [0.62; 0.88]) compared to the overall population (HR 0.68 [0.61; 0.75]); 5.9% vs. 7.8%; ARR = 1.9%).
  • With regard to major bleeding, it is evident that study centres with better INR control showed numerically more favourable effects for apixaban, although the effects in these centres were minor compared to centres with poor INR control.
  • When centres were divided into quartiles based on INR values, significant differences were observed only for those centres where INR control was below the median (INR control < 1st quartile [HR 0.44 (95% CI: 0.27 – 0.72); INR control ≥ 1st quartile to < median [HR 0.60 (95% CI: 0.47 – 0.76)].
  • There are no longer any significant differences for centres where INR control was above the median (INR control ≥ median to < 3rd quartile [HR 0.87 (95% CI: 0.70 – 1.08; INR control ≥ 3rd quartile [HR 0.70 (95% CI: 0.48 – 1.03)].
  • It should be noted that, with regard to the operationalisation of the ‘bleeding’ outcome, ‘major bleeding’ (intracranial and extracranial) also included haemorrhagic strokes, meaning that there was an overlap between the endpoints.
  • For major bleeding at other sites (extracranial, including gastrointestinal), there was a statistically significant difference in favour of apixaban versus warfarin, with an HR of 0.79 (3.0% vs. 3.8%; ARR 0.8 per cent; HR 0.79 [0.68; 0.93]).

Courtesy translation only, please refer to the German original.