Saxagliptin (3) – Onglyza®

• Clinical trials
  • The double-blind, multicentre, randomised and controlled SAVOR-TIMI 53 trial investigated the administration of saxagliptin versus placebo in patients with type 2 diabetes mellitus, in addition to antidiabetic and cardiovascular therapy, which was to be administered according to the patient’s needs and in accordance with regional standards.
  • The D1680L00002 study was a 52-week, directly comparative, randomised, double-blind Phase IIIstudy in which only older patients (aged ≥ 65 years) who were inadequately controlled on metformin monotherapy (HbA1c ≥ 7.0% to ≤ 9.0%) took part.
  • The directly comparative, randomised, double-blind study D1680C00001 investigated adult patients with type 2 diabetes mellitus who had not achieved adequate blood glucose control despite metformin monotherapy at a daily dose of ≥ 1500 mg.
  • According to the pharmaceutical manufacturer, this study (CV181057) included adult patients with type 2 diabetes mellitus whose blood glucose levels could not be adequately controlled with insulin or with a combination of insulin and metformin.

a) In combination with metformin, where metformin monotherapy, together with diet and exercise, does not adequately control blood glucose

• Overall, the G-BA therefore concludes that the additional benefit of saxagliptin in dual combination therapy with metformin, In adult patients aged 18 years and over whose blood glucose levels are not adequately controlled with the maximum tolerated dose of metformin alone, proof that this treatment is superior to the appropriate comparator therapy has not been demonstrated.
• Mortality and morbidity
  • The results on all-cause mortality and on cardiovascular or cerebrovascular events in studies D1680C00001 and D1680L00002 are derived from data on adverse events. In the overall population, no differences were observed between the treatment groups.
• Morbidity – Health status (EQ-5D VAS)
  • In study D1680L00002, health status was assessed using the EQ-5D questionnaire. The pharmaceutical manufacturer’s dossier does not analyse the results for this endpoint in further detail. The study data showed a statistically significant difference in mean values to the detriment of saxagliptin (-1 [-1.1; -0.9], p < 0.0001).
• quality of life
  • Data on health-related quality of life were not collected in studies D1680C00001 and D1680L00002. It is therefore not possible to draw any conclusions regarding an improvement or deterioration in quality of life with saxagliptin compared with the appropriate comparator therapy.
• Side effects – hypoglycaemia
  • In study D1680L00002, confirmed symptomatic hypoglycaemia (blood glucose < 50 mg/dl) occurred statistically significantly less frequently overall with saxagliptin plus metformin compared with glimepiride plus metformin (0 vs. 43 (10.0 %); Peto OR = 0.12; 95% CI: [0.07; 0.23]; p < 0.001).
  • There were no statistically significant differences between the saxagliptin plus metformin arm and the glimepiride plus metformin arm for the endpoint ‘severe hypoglycaemia’ (1 (0.3%) vs. 6 (1.7%), Peto OR: 0.24; 95% CI: [0.05; 1.05]; p = 0.061).
• Side effects – Serious adverse events (SAEs), therapy discontinuations due to AEs, pancreatitis, renal dysfunction
  • The results for the endpoints ‘Serious adverse events (SAEs)’, ‘Therapy discontinuations due to AEs’, ‘pancreatitis’ and ‘renal dysfunction’ showed no statistically significant differences between the treatment groups in either of the two studies, D1680C00001 and D1680L00002.
• Morbidity – weight gain of at least 7%
  • In the saxagliptin plus metformin arm of study D1680C00001, a statistically significantly lower number of patients experienced a weight gain of at least 7% compared with the glipizide/metformin arm (6 vs. 37; RR: 0.16; 95% CI: [0.07; 0.38]; p < 0.001).
  • In study D1680L00002, there were no statistically significant differences between the groups with regard to the endpoint ‘weight gain of at least 7%’.
• Morbidity – HbA1c
  • The primary endpoint selected in study D1680C00001, HbA1c (change in HbA1c level compared with baseline at the start of the study at week 52 and week 104) is a surrogate parameter in the treatment of diabetes mellitus. The results for this endpoint show no statistically significant differences between the treatment groups at weeks 52 and 104, respectively.
  • In study D1680L00002, the mean HbA1c level in the saxagliptin/metformin arm was not reduced as much as in the comparator arm (change from baseline: -0.44% vs. -0.64%; MD = 0.20; 95% CI: [0.10; 0.30]).
  • The primary endpoint of study D1680L00002 was to to demonstrate the superiority of saxagliptin plus metformin in achieving an HbA1c target of < 7 % without confirmed or severe hypoglycaemia, compared with glimepiride plus metformin. The difference between the two groups (37.9% for saxagliptin and 38.2% for glimepiride) was not statistically significant.
• Conclusion
  • Taking the results of the two studies, D1680C00001 and D1680L00002, into account as a whole, the minor reduction in the number of confirmed, non-severe hypoglycaemic episodes with saxagliptin results in a moderate improvement compared with the appropriate comparator therapy.
  • Based on the data presented, the SAVOR-TIMI 53 study reveals additional disadvantages regarding the additional benefit of saxagliptin, which are attributable in particular to the negative results for the endpoint ‘hospitalisation due to heart failure’.
  • Overall, therefore, no additional benefit for saxagliptin can be inferred from the data presented.

b) In combination with a sulphonylurea in patients for whom the use of metformin appears unsuitable, if sulphonylurea monotherapy, together with diet and exercise, does not adequately control blood glucose

• Additional benefit from saxagliptin for patient population b) is not proven.
• As already explained, the SAVOR-TIMI 53 study can only be used in its entirety to assess the additional benefit of saxagliptin compared with the respective comparator groups. Taken together, the results of the study do not indicate any additional benefit of saxagliptin compared with the comparator therapy.

c) As triple oral therapy in combination with metformin and a sulphonylurea, when this treatment alone, together with diet and exercise, does not adequately control blood glucose

• Additional benefit from saxagliptin for patient population c) is not proven.
• As already explained, the SAVOR-TIMI 53 study can only be used in its entirety to assess the additional benefit of saxagliptin compared with the respective comparator groups. Taken together, the results of the study do not indicate any additional benefit of saxagliptin compared with the comparator therapy.

d) In combination with insulin (with or without metformin), where this treatment alone, together with diet and exercise, does not adequately control blood glucose

• Overall, the G-BA concludes that for saxagliptin in combination with insulin and metformin, or with insulin alone—where this treatment alone, together with diet and exercise, does not adequately control blood glucose— no additional benefit has been identified compared with the appropriate comparator therapy (metformin + human insulin or human insulin alone).
• As already explained, the SAVOR-TIMI 53 trial included patients with different prior treatments. Even if the comparison groups formed by the pharmaceutical manufacturer for patient population d) have a high probability of not lacking structural equivalence, there is no guarantee that the appropriate comparator therapy was implemented adequately.
• Mortality, morbidity, quality of life, side effects
  • No data were provided on mortality (overall mortality). The data provided on cardiac events (as part of the study’s adverse event monitoring) are not meaningful for assessing additional benefit, as the study was not designed to validly collect these data over the long term.
  • No data were provided on quality of life either. Consequently, it is not possible to draw any conclusions regarding an improvement or deterioration in quality of life with saxagliptin plus metformin in combination with insulin compared with the appropriate comparator therapy.

• mortality
  • Overall mortality / cardiovascular mortality There are no significant differences between the treatment groups with regard to overall mortality and the endpoint ‘cardiovascular death’. In the study, 420 deaths (5.1 per cent) occurred in the saxagliptin arm and 378 deaths (4.6 per cent) in the control arm. Cardiovascular-related deaths occurred in 3.2% of patients in each study arm (saxagliptin arm 269 vs. control arm 260).
• Morbidity – Major Adverse Cardiac Events (MACE)
  • There were no statistically significant differences between the treatment groups for the composite endpoint ‘Major Adverse Cardiac Events (MACE)’. A total of 613 events occurred in the saxagliptin arm and 609 events in the control arm (7.4% of patients in each group). Furthermore, the individual endpoints also showed no statistically significant differences between the treatment groups.
• Morbidity – Hospitalisation due to heart failure
  • In the SAVOR-TIMI 53 trial, 3.5% of patients in the saxagliptin arm and 2.8% of patients in the control arm were hospitalised due to heart failure. This effect is statistically significant to the detriment of saxagliptin (HR = 1.27 [95% CI 1.07; 1.51]; p = 0.007).
• Morbidity – Other morbidity endpoints
  • For all other endpoints – ‘all myocardial infarctions’, ‘all strokes (ischaemic)’, ‘all strokes (ischaemic, haemorrhagic and unspecified)’, ‘all myocardial infarctions’, ‘transient ischaemic attack (TIA)’ and ‘other cerebrovascular events’, ‘laser treatment for diabetic retinopathy’, ‘other local treatment for retinopathy’, ‘initiation of chronic dialysis and/or kidney transplantation and/or detection of a serum creatinine concentration of > 6.0 mg/dl’ and “doubling of the serum creatinine concentration”, no statistically significant differences were observed between the saxagliptin arm and the control arm. This indicates that saxagliptin offers neither advantages nor disadvantages compared with the comparator therapy.
• Morbidity – Health status (EQ-5D VAS)
  • As a significant proportion (16.9 %) of patients were not included in the analyses of the endpoint ‘Health status EQ-5D (VAS)’ at the end of treatment, the potential for bias for this endpoint was assessed as high. It is unclear which method was used and on the basis of which patients the effect estimator was calculated. For this reason, no usable data are available for the ‘health status’ endpoint.
• quality of life
  • Data on health-related quality of life were not collected in the study.
• Side effects – serious adverse events (SAEs), discontinuation due to AEs
  • The results for the endpoints ‘serious adverse events (SAEs)’ and ‘discontinuation due to AEs’ do not differ statistically significantly between the treatment groups.
• Side effects – Symptomatic hypoglycaemia
  • Overall, symptomatic hypoglycaemia occurred in 703 patients (8.5 %) in the saxagliptin arm and in 578 patients (7.0 %) in the control arm. This effect differs statistically significantly between the treatment groups (RR = 1.21 [95% CI 1.09; 1.34]; p < 0.001). This indicates a disadvantage of saxagliptin compared with the control group with regard to the endpoint ‘symptomatic hypoglycaemia’.
• Side effects – severe hypoglycaemia, hospitalisation due to hypoglycaemia
  • The results for the endpoints ‘severe hypoglycaemia’ and ‘hospitalisation due to hypoglycaemia’ do not differ statistically significantly between the treatment groups.
• Side effects – Pancreatitis
  • In both the saxagliptin arm and the control arm of the SAVOR-TIMI 53 study, 0.3% of patients developed pancreatitis. There were therefore no statistically significant differences between the treatment groups for the endpoint ‘pancreatitis’.
• Conclusion
  • The pharmaceutical manufacturer is submitting the data from the SAVOR-TIMI 53 cardiovascular outcomes study for the benefit assessment of saxagliptin following the expiry of the deadline, in order, firstly, to address the outstanding issues regarding the limitation of the resolution of 1 October 2013 and, secondly, to present new data for the derivation of the additional benefit of saxagliptin.
  • Admittedly, for the reasons already outlined, no comparison can be made between saxagliptin and the respective appropriate comparator therapy for the individual patient groups. Nevertheless, given its duration, size and the collection of patient-relevant cardiovascular endpoints, the SAVOR-TIMI 53 study as a whole provides new insights for the benefit assessment of saxagliptin and is therefore considered relevant to this assessment.
  • The study shows no statistically significant differences between the treatment groups for the primary endpoint MACE and thus neither advantages nor disadvantages for saxagliptin compared with the control group for the MACE endpoint. However, in the saxagliptin arm, a statistically significant disadvantage was observed compared with the control arm for the endpoint ‘hospitalisation due to heart failure’ (HR = 1.27 [95% CI 1.07; 1.51]; p = 0.007), with no effect modifications observed in the a priori defined patient populations of region or cardiovascular risk group. In addition, symptomatic hypoglycaemia was observed more frequently in the saxagliptin arm (RR = 1.21 [95% CI 1.09; 1.34]; p < 0.001).

Courtesy translation only, please refer to the German original.