Introduction

The use of rituximab as maintenance treatment of antineutrophil cyto plasm antibody (ANCA) associated vasculitides (AAVs) was supported by MAINRITSAN trial that demonstrated rituximab superiority to azathioprine¹. Moreover, a recent publication concluded that maintenance treatment by rituximab instead of azathioprine could be cost-effective for preventing re lapses in patients with AAVs².

MAINRITSAN2 was a randomised, open-label, multicenter, phase III trial conducted in France which evaluated the difference between an individually tailored and a fixed-schedule maintenance therapy with rituximab³. The trial included 162 patients (81 per group) with newly diag nosed or relapsing granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA) in complete remission after induction therapy and those general characteristics were comparable. ANCA and circulating CD19+ B cells were the biological parameters monitored to re-infuse rituximab in tailored regimen. The conclusion was that AAVs relapse rates did not differ significantly between the two groups over a period of 28 months (18 months of treatment plus 10 months of follow-up). At month 28, 21 pa tients had suffered 22 relapses: 14/81 (17.3%) in 13 tailored-infusion recipients and 8/81 (9.9%) in 8 fixed-schedule patients (p = 0.22). Further more, individually tailored-arm patients received fewer rituximab infusions, 248 vs 381 infusions, with medians (interquartile range (IQR)) of 3 (2-4) vs 5 (5-5) administrations, respectively. An extension of MAINRITSAN2 is currently underway (MAINRITSAN3) to assess a long-term maintenance therapy with rituximab⁴.

The fact that fewer rituximab infusions were required with the indivi dually tailored regimen is important in terms of cost. Although, cost-mini misation analysis can not be conducted unless two alternative treatments are demonstrated to be equivalent, in clinical practice, this type of economic evaluation is frequently applied when a clinical relevant di fference has not been demonstrated between two alternative options. Based on this, the present study used a cost-minimization analysis as the base approach to examine the real world costs of an individually tailored-therapy compared to a fixed-schedule therapy with rituximab for remission maintenance of AAVs to facilitate decision making in clinical practice. A cost-effectiveness analysis has been considered in the sen sitivity analysis.

Methods

Drug acquisition

Drug acquisition costs (Table 1) included cost of rituximab and premedi cation: intravenous methylprednisolone (100 mg), intravenous dexchlorphe niramine (5 mg) and oral acetaminophen (1,000 mg) as reported by the MAINRITSAN2 clinical trial³.

Table 1. Drug unit costs 

ANCA: antineutrophil cytoplasm antibody.

For base case analysis, we adopted our hospital's drug prices including 4% value added tax (VAT) and 75% discount for rituximab biosimilar (Truxima® 500 mg⁵, available on the European market since 2017 and indicated in GPA and MPA treatment), as indicated in Royal Decree of Law 8/2010⁶.

Preparation, administration and monitoring

We based the decision on which resources to use in preparation, ad ministration and monitoring (Table 2) on the opinion of clinical experts. The unit costs associated with these resources were an average obtained from the health costs database eSalud⁷.

Table 2. Preparation, administration and monitoring costs 

ANCA: antineutrophil cytoplasm antibody.

Preparation costs consisted of consumable materials (500 mL saline so lution for dilution, 50 mL syringe, vented spike, alcohol, gauze, gown, latex gloves, disposable surgical cap and mask, and plastic bag for transport) and rituximab reconstitution and dilution in aseptic conditions by pharmacy nursing staff (31.34 euros per preparation). Equipment costs such as main tenance contracts of electromedical equipment (laminar flow cabinet) were considered negligible.

Costs for administration process included the patient's stay in the treatment room receiving specialized care provided by nurses (252.24 eu ros per stay) and auxiliary nurses during the first 30 minutes (5.94 euros per hour). Intravenous sets and solutions for premedication dilutions (50 mL saline solutions) were included as consumable materials.

In the MAINRITSAN2 trial, therapy monitoring consisted of extraction of blood samples and ANCA and CD19+ B lymphocytes tests. In addition, costs for Internal Medicine consult, Pharmacy consult before rituximab ad ministration and other complementary analyses were included in keeping with routine clinical practice. Internal Medicine consults, extraction of blood samples and blood tests were trimestral in the tailored-infusion group and before infusion in the other arm, except for the first two administrations (days 0 and 14) that were scheduled in the first visit. Other complementary analy ses involved complete blood count, kidney function test, liver function test, reactive C protein, globular sedimentation rate, serum immunoglobulins and urine test. Costs from routine complementary tests before the beginning of treatment were excluded from the model.

Results

In the base case, the per-patient total cost for the tailored-infusion regi men with rituximab for remission maintenance of AAVs was 6,049 euros vs 7,850 euros for the fixed-schedule regimen. This represents a saving of 1,801 euros (23%) per patient in direct costs for the health system (Table 3).

Table 3. Results of base case cost-minimization analysis 

Savings were primarily due to lower drug acquisition costs for the tai lored-infusion regimen (2,861 euros vs 4,768 euros) in addition to lower preparation and administration costs (892 euros vs 1,486 euros) due to the lower number of infusions per patient in the tailored-infusion regimen. The tailored-infusion regimen presented higher monitoring costs (2,296 euros vs 1,596 euros).

The results of the sensitivity analysis S1 to S9 are shown in Table 4.

Table 4. Sensitivity analysis of base case 

^aPrice in our hospital: MabThera® 500 mg 1,091.32 euros.

^bHighest price in Europe: MabThera® 500 mg 12,305.15 kr (Denmark) equivalent to 1,650.23 euros.

^cLowest price in Europe: MabThera® 500 mg 9,703.61 kr (Sweden) equivalent to 920.85 euros.

Considering the uncertainty about the difference in relapse rate, in the S10 scenario the tailored-infusion regimen would be a dominant strategy. It would be associated with lower rate of relapses, 3% lower, and lower cost, 1,801 euros less per patient. In the S11 and S12 scenarios, the fixed-schedule regimen would be an alternative more costly and more effective, with an incremental cost effectiveness ratio (ICER) of 24,338 euros per avoi ded relapse in S11. This means that if the fixed schedule is used instead of the tailored-infusion regimen, it would cost 24,338 euros per avoided relapse. Similarly, in the S12 scenario, the ICER would be 10,061 euros per avoided relapse.

Discussion

The tailored-therapy had higher monitoring costs than the fixed-schedu le maintenance therapy, which had higher drug acquisition, preparation and administration costs owing to patients undergoing this regimen recei ving more cycles of rituximab. Overall, tailored-therapy costs less than the fixed regimen. This result was replicated in all assumptions considered in the sensitivity analysis of cost-minimization approach, emphasizing its robustness.

Regardless of the global reduction of drug acquisition price due to biosi milar commercialization, drug acquisition cost will still represent the main cost in the overall attention of these patients. Therefore, regimens able to reach the same effectiveness with fewer drug infusions will be less onerous. Further more, the reduction in the number of infusions in the tailored-infusion arm de creased the theoretical probability of the occurrence of adverse events and severe adverse events (SAEs). However, the clinical trial found no statistically significant differences between the two groups regarding safety³.

We found a large number of studies evaluating rituximab as a mainte nance treatment in AAVs, but only a few looked at economic considerations. Richard A. Watts et al. carried out a study to quantify the global burden of AAVs. They declared that the introduction of biological therapies increased drug costs, but some of those increases might be offset by better disease control¹². Turner-Stokes T. et al. studied the induction treatment of AAVs with a single dose of rituximab instead of several doses. They questioned the need to repeat dosing and adopted a standard single-dose protocol to treat active AAVs¹³. Pentek M. et al. evaluated the impact of biosimilars on chro nic immune-mediated inflammatory diseases. Their results suggested that, gi ven the lower price of biosimilars, formerly established biological treatment sequence practices and the eligibility criteria for biological treatment should be reconsidered. They concluded that biosimilars may contribute to better patient-access and provide savings to governments¹⁴.

The MAINRITSAN2 study³ has shown that either of the two treatment modalities could be implemented in clinical practice with similar efficacy. The main strength of our study is that it provides added information for deci sion-making in clinical practice over the choice of the maintenance treatment of ANCA associated vasculitides.

The main limitation of the present analysis is that the economic evaluation is based on the results of a single study, the MAINRITSAN2, with a limited number of patients, and, therefore, limited statistical power to detect as signi ficant some clinically relevant differences between groups in relapse rate or in the incidence of serious adverse effects. However, this study provides the best evidence available to choose in clinical practice between a tailored or fixed rituximab regimen, and this economic evaluation adds costs information for decision making. In the base-case scenario, similar efficacy and safe ty was assumed and therefore a cost-minimization analysis was performed, and in the sensitivity analysis, a cost-effectiveness analysis was performed to consider the uncertainty in effectiveness. The latter has also the limitation of not including survival or quality of life data; however, this information is not currently available. In addition, cost of relapses were not included. In the cost-effectiveness analysis, the fixed dose strategy goes from being dominated by the tailored strategy to being more effective and more costly. To know whether it is an efficient strategy is difficult as there is uncertainty regarding long term or final outcome consequences of relapses and serious adverse effects. When this information will be available the best strategy will have to be reconsidered, until then this economic evaluation can guide us as the best alternative to select.

A discount rate could not be applied because of the absence of informa tion about when the successive treatments occurred in the tailored therapy arm. However, this should not have been significant because the total pe riod of the study was only 18 months. We were also unable to find data on the cost of patient hospital care depending on the duration of the treatment administration; we had to use the cost of a full visit with an average duration calculated among many other types of treatments and medical procedures. This fact would not seem to have influenced results because of the small contribution of this cost to the total cost. Our study was affected by the lack of transparency regarding official prices in the western world and, above all, by the opacity of the real purchase prices of public health systems and large health maintenance organizations or insurers. We have applied the prices available in the databases^{8 9-10}, although some of them may not be completely updated. In any case, this CMA is easily reproducible for those who wish to apply it in other scenarios.

In this analysis, we have established a point of view of the health system without considering patients' preferences, or indirect and intangible costs (although these are very difficult to measure). The tailored-infusion option requires more medical visits and blood extractions. It also generates uncer tainty in patients since their therapeutic plan is not defined from the outset. For these reasons, in the future we need to establish which treatment option generates less expense for patients, less discomfort and less interference in their daily life.

This economic evaluation based on available data shows the cost-sa ving option between a tailored-therapy and a fixed-schedule regimen with rituximab for the maintenance treatment of AAVs. From the perspective of the health system, the tailored-therapy regimen seems to be the preferable option in terms of direct costs. Further studies assessing all the effects and costs associated to AAVs maintenance treatment with rituximab are needed to support clinical management and healthcare planning.