Dabigatran etexilate for the treatment and secondary prevention of deep vein thrombosis and/or pulmonary embolism

The Committee considered the 4 dabigatran trials included in the company's submission: RE‑COVER, RE‑COVER II, RE‑MEDY and RE‑SONATE. The Committee was aware that although the marketing authorisation for dabigatran does not require routine monitoring, press reports and an article published in the British Medical Journal had suggested that lower rates of bleeding could be achieved if dabigatran levels in the blood were monitored with subsequent dose‑adjustment if needed. The company representatives stated that there were no other data on the safety or efficacy of dabigatran etexilate that were relevant to the Committee's appraisal of dabigatran etexilate for the treatment and secondary prevention of DVT and PE, and that it had fully disclosed all relevant data to the European Medicines Agency. The Committee accepted the company's statement that all relevant data had been submitted for the appraisal.

The Committee considered the trials and their generalisability to clinical practice in England. It noted that people taking warfarin in the RE‑COVER trials and RE‑MEDY had an INR in the therapeutic range 57% to 60% of the time, and that 60% is considered to be at the lower limit of the acceptable time in therapeutic range. However, it also noted that the time in therapeutic range had been calculated early on in the warfarin treatment phase when the dose was still being titrated, which may have resulted in the average time in therapeutic range being lower. The Committee further noted the Evidence Review Group's (ERG's) concerns that people in the trials were younger than would be expected in clinical practice in England. The Committee went on to note that all people in the dabigatran arms of the 4 trials had the 150 mg twice‑daily dose, despite the summary of product characteristics stating that people aged over 80 years and people taking verapamil should have a lower dose of 110 mg twice daily. The Committee was concerned that there were no clinical trial data on the 110 mg dose for the treatment and secondary prevention of DVT and PE. It heard from the company that the European Medicines Agency requested pharmacokinetic data on plasma levels of dabigatran in people having the 150 mg or 110 mg doses in the RE‑LY trial for atrial fibrillation, and pharmacokinetic data from the trials for DVT and PE. Safety data for the 110 mg dose were available from the RE‑LY trial. The company stated that the European Medicines Agency had approved the use of the lower dose for certain people based on these data. The Committee concluded that the dabigatran trials were generalisable to people who would have the 150 mg dose of dabigatran etexilate in clinical practice in England. Although there was some uncertainty as to whether the 110 mg dose would be equally effective in treating and preventing recurrent VTE in those people for whom it is recommended, the Committee concluded that dabigatran etexilate should be appraised in accordance with its marketing authorisation.

The Committee considered the effectiveness of dabigatran etexilate for the treatment and prevention of recurrent VTE (secondary prevention). It noted that in the RE‑COVER and RE‑MEDY trials similar numbers of people had a recurrent VTE in the warfarin and dabigatran arms, and that dabigatran etexilate was statistically non‑inferior to warfarin for both acute treatment and secondary prevention. The Committee noted that in the RE‑SONATE trial (which included people for whom there was uncertainty about their need for continued anticoagulation), statistically significantly fewer people had a recurrent VTE when taking dabigatran etexilate compared with placebo. The Committee noted that there were no head‑to‑head trials comparing dabigatran etexilate with rivaroxaban and that the company had performed an adjusted indirect comparison. The Committee noted that there was no clear difference between dabigatran etexilate and rivaroxaban in preventing recurrent VTE during acute treatment or when continued longer term for secondary prevention, because the 95% confidence intervals surrounding the hazard ratio were wide and crossed 1. The Committee also noted that the ERG had performed a network meta‑analysis which showed no difference between dabigatran etexilate and rivaroxaban and that the 95% credible intervals surrounding the estimate were also wide. The Committee concluded that no difference had been demonstrated between dabigatran etexilate, warfarin and rivaroxaban in treating VTE and preventing recurrent events.

The Committee considered the effectiveness and safety of dabigatran etexilate in people with active cancer. It was aware that in clinical practice in England, standard care for people with active cancer who have a DVT or PE is at least 6 months' treatment with LMWH, with a review about whether to continue treatment at that point. It was also aware that LMWH is used rather than warfarin, because LMWH has been demonstrated to be more effective than warfarin in people with cancer. The Committee noted that only around 4% of people in the RE‑COVER and RE‑MEDY trials had active cancer, and agreed with the ERG's view that the definition of active cancer used in the trials was broader and would include more people than the definition of active cancer used in clinical practice in England. The Committee further noted that there were no head‑to‑head data comparing dabigatran etexilate with LMWH in people with active cancer available. The Committee concluded that there were insufficient data to assess the effectiveness and safety of dabigatran etexilate in people with active cancer who had a DVT or PE, and that it was not possible to make a specific recommendation for this group of people.

The Committee considered the rates of adverse events in people taking dabigatran etexilate. It noted that fewer people taking dabigatran etexilate in the RE‑COVER and RE‑MEDY trials had a major bleeding event compared with warfarin, but the confidence intervals surrounding the estimates crossed 1. The rate of major or clinically relevant bleeding, or any bleeding, was lower in these trials in people having dabigatran etexilate compared with warfarin (tables 1 to 3, section 3.5). The Committee also highlighted that there were fewer people in the trials who had an intracranial haemorrhage while taking dabigatran etexilate than while taking warfarin. It noted, and the clinical expert agreed, that the trial data suggest there is a tendency towards lower risk of intracranial haemorrhage in people taking the new oral anticoagulants (dabigatran etexilate, rivaroxaban, and apixaban) compared with warfarin across the indications for these drugs. The Committee noted that there were no significant differences between rates of major bleeds in either the company's adjusted indirect comparison or the ERG's network meta‑analysis comparing dabigatran etexilate with rivaroxaban. The Committee observed that more people had an acute coronary syndrome event when having dabigatran etexilate compared with warfarin in RE‑MEDY, but understood that this has been thought to reflect a protective effect of warfarin, rather than an adverse effect of dabigatran etexilate. It concluded that dabigatran etexilate had an acceptable safety profile compared with warfarin and rivaroxaban.

The Committee noted that the company had presented 2 base‑case analyses: 1 for acute treatment and 1 for treatment and prevention of recurrent VTE ('secondary prevention'). It noted that the ERG had made a series of corrections to errors in the model which the company had agreed were appropriate. The Committee noted that in the acute treatment base case, the company had assumed that treatment would last for 6 months. In the treatment and secondary prevention base case, people were assumed to receive anticoagulation treatment for 6 to 18 months, depending on whether rivaroxaban or warfarin was the comparator. The Committee heard from the clinical expert that the risks and benefits of continuing treatment vary between people, but that a significant proportion of people continue treatment for the rest of their lives (see section 4.3). It considered that a large percentage of people needing long‑term anticoagulation would have it indefinitely, but the company had not tested the effect of life‑long treatment in its base case or sensitivity analyses. The Committee noted the ERG's scenario for life‑long treatment, which increased the company's base‑case incremental cost‑effectiveness ratio (ICER) for dabigatran etexilate compared with warfarin (with ERG model corrections included) from £9973 to £15,634 per quality‑adjusted life year (QALY) gained. The Committee further noted that the company had not included the cost and utility decrement of myocardial infarction and dyspepsia (indigestion) in its base case, but tested the effect of cardiac events in a sensitivity analysis which resulted in only a modest increase in the ICER for dabigatran etexilate compared with warfarin. The Committee concluded that it was appropriate to present base cases for acute treatment and treatment with secondary prevention separately, but that it should be assumed in the secondary prevention base case that treatment would be life‑long for most people whose condition needs treatment beyond 6 months.

The Committee noted the company's assumption that there would be a disutility associated with warfarin therapy of -0.012. It noted that this disutility had also been applied for warfarin therapy in NICE technology appraisal guidance on rivaroxaban for the treatment of DVT and prevention of recurrent DVT and PE and rivaroxaban for treating PE and preventing recurrent VTE. The Committee understood the ERG's concerns that the utility decrement was derived solely from a small study including just 48 people. The Committee heard that qualitative research on warfarin treatment suggested that people were anxious about being in the appropriate INR therapeutic range and whether they were taking the correct dose of warfarin. They also found warfarin monitoring visits to be inconvenient, and the dietary considerations associated with taking warfarin had a detrimental effect on their quality of life. The Committee noted that in the company's model, the utility decrement associated with warfarin was greater than that associated with injections of LMWH. It heard from the patient expert that this was reasonable, because people are willing to accept the negative effects of LMWH as a part of initial short‑term treatment. In contrast, because warfarin is taken for a longer time, the cumulative effect on quality of life would likely be greater. The Committee concluded that warfarin treatment, particularly if life‑long, could be expected to reduce quality of life but the extent to which it did so was uncertain. It further concluded that although the company's estimate of utility decrement was based on limited evidence, it was the best estimate available and had been accepted as reasonable in previous appraisals.

The Committee discussed the company's assumptions relating to the frequency and cost of warfarin monitoring. It noted that the company assumed all people having warfarin would have an initial appointment with a consultant, 4 monitoring visits while their dose of warfarin was titrated, and then monthly visits to check that their INR was within the therapeutic range for the remainder of treatment. The company had assumed that the initial consultant‑led visit would cost £62.56, and estimated that follow‑up INR visits would cost on average £28 per visit. This resulted in a monitoring cost of £482.41 for the first year of treatment. The Committee noted that the ERG had assumed that the cost of INR follow‑up visits would be lower (£10.61 per visit) because they would all be done by a nurse, rather than some being consultant‑led. The ERG's assumptions on INR visit costs resulted in a first‑year cost of £221.71, less than half the company's original estimate. The Committee also noted that the ERG believed that people would have fewer warfarin monitoring visits (once every 3 months) than the company had suggested during the secondary prevention period. The Committee heard from both the patient expert and clinical expert that it was difficult to give a precise estimate of the cost of warfarin monitoring, because the structure of warfarin monitoring services varies widely and there is no definitive average monitoring cost available for the NHS. The Committee understood from previous technology appraisals of dabigatran etexilate, rivaroxaban and apixaban that the estimates of warfarin monitoring costs from the companies and ERGs varied widely. It noted that in NICE technology appraisal guidance on rivaroxaban for the treatment of DVT and prevention of recurrent DVT and PE, the ERG's estimate of £320 for 15 visits in the first year had been considered more reasonable than the company's estimate of £656 for 24 visits. Similarly, in NICE technology appraisal guidance on rivaroxaban for treating PE and preventing recurrent VTE, the ERG's estimate of £304 to £379 for 12 to 15 visits had been considered more reasonable than the company's estimate of £607 for 24 visits in the first year. The Committee noted that some people need more frequent monitoring than others because their INR is more difficult to control. This combined with the variable warfarin monitoring arrangements throughout the NHS meant that estimating the average cost was very difficult. The Committee concluded that the company's estimate was higher than figures previously accepted as reasonable, but that the ERG's was lower.

The Committee discussed the ERG's exploratory base case for acute treatment, noting that it included 13 scenarios. The ICER for dabigatran etexilate compared with warfarin increased from £831 per QALY gained in the company's corrected base case to £18,240 per QALY gained in the ERG's exploratory analysis (incremental cost £223, incremental QALYs 0.012). The main reason for this higher ICER was the warfarin monitoring costs assumed by the ERG (see section 4.12). The Committee concluded that the most plausible ICER could not be determined for dabigatran etexilate compared with warfarin for acute treatment, but even if it accepted the ERG's exploratory analysis the ICER remained in the range which could be considered a cost-effective use of NHS resources. In the comparison with rivaroxaban, both the company and the ERG calculated that dabigatran etexilate dominated rivaroxaban (that is, dabigatran etexilate was both less costly and more effective than rivaroxaban). However, the Committee noted that neither the company nor the ERG had found any significant difference in efficacy between the 2 treatments in their indirect comparisons, and that the costs were also very similar. This would result in the ICER estimates being sensitive to small changes in the costs or QALYs having a large effect on the ICER. Therefore, the Committee accepted that dabigatran etexilate could be recommended as an option for the acute treatment of DVT and PE as an alternative to warfarin or rivaroxaban.

For the combined treatment and secondary prevention of VTE the ERG presented an exploratory base case and an incremental analysis. The Committee noted that the ICER for dabigatran etexilate compared with warfarin was £9973 per QALY gained in the company's corrected base case, and £35,786 per QALY gained in the ERG's exploratory analysis. The Committee was aware that the ERG had included 16 scenarios in its exploratory base case, and the main factors increasing the ICER were: assuming life‑long secondary prevention for all patients (section 4.9), resulting in an ICER of £15,634 per QALY gained; assuming that warfarin monitoring in the secondary prevention period was less frequent (once every 3 months rather than monthly), resulting in an ICER of £15,208 per QALY gained; and assuming a lower cost of each warfarin monitoring visit (section 4.12), resulting in an ICER of £17,419 per QALY gained. The Committee noted that the scenarios included in the ERG exploratory base case interacted, and it was difficult to determine the ICER if the Committee did not accept all of the assumptions that the ERG had included. In particular, the Committee considered that the ERG's assumptions surrounding frequency and cost of warfarin monitoring visits were more conservative than assumptions accepted as reasonable in previous appraisals (see section 4.12). Combining these assumptions had a cumulative effect, driving the ICER towards £35,000 per QALY gained, but applying them separately resulted in ICERs of less than £20,000 per QALY gained. The Committee concluded that the ICER for dabigatran etexilate compared with warfarin for the treatment and secondary prevention of VTE was uncertain because of the lack of an average NHS warfarin monitoring cost, as well as uncertainty about the proportion of people who would stay on treatment for the rest of their lives. Although the company's base case was likely to be too low, the ERG's exploratory base case for treatment and secondary prevention, including conservative assumptions surrounding warfarin monitoring costs, may have overestimated the ICER. The Committee was prepared to accept that the ICER probably lay somewhere between the 2 estimates. In the comparison with rivaroxaban, the Committee noted that rivaroxaban was extendedly dominated by dabigatran etexilate. The Committee also noted that dabigatran etexilate and rivaroxaban had not been shown to have different efficacy, and their costs were very similar. This resulted in an ICER that was highly sensitive to changes in costs and QALYs. The Committee concluded that, on balance, dabigatran etexilate could be considered a clinically and cost-effective option for the treatment and secondary prevention of VTE.