Appraisal Consultation Document: Immunosuppressive therapy for renal transplantation

The Department of Health and the National Assembly for Wales have asked the National Institute for Clinical Excellence (NICE or the Institute) to conduct an appraisal of immunosuppressive therapy for renal transplantation and provide guidance on its use to the NHS in England and Wales. The Appraisal Committee has had its first meeting to consider both the evidence submitted and the views put forward by the representatives nominated for this appraisal by professional organisations and patient/carer and service user organisations. The Committee has developed preliminary recommendations on the use of basiliximab, daclizumab, mycophenolate mofetil, tacrolimus and sirolimus.

This document has been prepared for consultation with the formal consultees. It summarises the evidence and views that have been considered and sets out the preliminary recommendations developed by the Committee. The Institute is now inviting comments from the formal consultees in the appraisal process (the consultees for this appraisal are listed on the NICE website).

Note that this document does not constitute the Institute's formal guidance on this technology. The recommendations made in Section 1 are preliminary and may change after consultation.

The process the Institute will follow after the consultation period is summarised below. (For further details, see the Guide to the Technology Appraisal Process on the Institute's website).

• The Appraisal Committee will meet again to consider the original evidence and this Appraisal Consultation Document in the light of the views of the formal consultees.
• At that meeting, the Committee will also consider comments made on the document by people who are not formal consultees in the appraisal process.
• After considering feedback from the consultation process, the Committee will prepare the Final Appraisal Determination (FAD) and submit it to the Institute.
• Subject to any appeal by consultees, the FAD may be used as the basis for the Institute's guidance on the use of the appraised technology in the NHS in England and Wales.

The key dates for this appraisal are:

Closing date for comments: Tuesday 4 November 2003
Second Appraisal Committee meeting: Tuesday 11 November 2003

Details of membership of the Appraisal Committee are given in Appendix A and a list of the sources of evidence used in the preparation of this document is given in Appendix B.

Note that this document does not constitute the Institute's formal guidance on this technology. The recommendations made in Section 1 are preliminary and may change after consultation.

Basiliximab or daclizumab are recommended as options for induction therapy in the prophylaxis of acute organ rejection in individuals undergoing renal transplantation. They should be used concomitantly with calcineurin inhibitor-based immunosuppression. In considering the use of these products for people at high immunological risk, clinicians should ensure that patients are aware that less is known about the outcomes of treatment than for people at low to moderate risk. The regimen with the lowest cost should be considered first.

Tacrolimus is recommended as an option when a calcineurin inhibitor is indicated as initial or maintenance immunosuppression in renal transplantation. The relative appropriateness of tacrolimus or ciclosporin should reflect the importance of their different side effect profiles in different patients.

Sirolimus is not recommended in combination with calcineurin inhibitors for the prophylaxis of rejection in renal transplantation. It is recommended as an option in cases of severe adverse reactions to calcineurin inhibitors and mycophenolate mofetil necessitating withdrawal of these treatments and substitution of an alternative immunosuppressant.

Over the past three decades renal transplantation has become established as the treatment of choice for many patients with end stage renal failure (ESRF). The only alternative is renal dialysis. The establishment of transplantation has been made possible by the introduction of immunosuppressants. Immunosuppression has classically been composed of triple therapy with 1) a calcineurin inhibitor (that is, ciclosporin) 2) an antiproliferative agent (that is, azathioprine) and 3) a corticosteroid. However, some treatment centres use a policy of initial monotherapy with a calcineurin inhibitor, adding in other agents if necessary.

Ciclosporin for oral administration has been available in two forms. It was originally available as an oily solution (Sandimmun) but is now only marketed as an oral solution/microemulsion (Neoral). However, the two are not necessarily bioequivalent.

In England in 2001, there were approximately 13,000 patients receiving immunosuppression following kidney transplantation and approximately 900 in Wales. In 2001, approximately 1500 new renal transplants were performed in England and Wales with about 21% of organs coming from live donors.

The median age of all adults receiving a kidney transplant in 2001 was 49 years. There is a 7-10% annual increase in the UK dialysis population and the number of people needing a transplant is expected to rise over the next decade.

Renal transplants are unsuccessful for a number of reasons including: technical failures, recurrence of original renal disease in the allograft, chronic allograft nephropathy (CAN, formally called chronic rejection - that is, long-term deterioration of the graft), acute rejection and death of the recipient with a functioning graft.

CAN is arguably the most common cause of late graft loss. It is usually a gradual process, although both the time of onset and the rate of progression vary. CAN may develop as early as within a few months of the transplant or emerge after several years. The course is generally unremitting and ultimately leads to total loss of graft function necessitating a return to dialysis or re-transplantation.

Episodes of acute rejection are most frequently observed during the first few weeks after transplantation, but can occur at any time if the level of immunosuppression becomes inadequate. The response is cell-mediated and leads to injury to or destruction of the functioning cellular structures of the transplanted organ. Occasionally, the response may be more aggressive and include an antibody-mediated vascular component.

Clinically, acute rejection tends to occur as acute episodes heralded by a reduction in graft function (seen on biochemistry) and clinical features such as fluid retention and occasionally graft tenderness or fever.

People who undergo renal transplantation are required to receive life-long (or at least, long-term) treatment with immunosuppressive drugs. When selecting these treatments, the risk of immunological-mediated graft failure for any donor-recipient pair needs to be balanced against the drug's associated side effects for the recipient. The ultimate aim of treatment is to prolong patient and graft survival.

Complications of immunosuppression include increased risk of developing infections (including viral infections such as cytomegalovirus, herpes simplex and zoster, and Epstein Barr virus; and opportunistic protozoal, fungal and bacterial infections). As immunosuppression is usually at its highest level in the first 6 months post-transplantation, this is also the peak period for infections in patients. Although modern immunosuppressive agents direct their activity principally towards the components of the rejection response recipients are at higher risk for infections than the general population throughout their post-transplant life. Some drugs also cause bone marrow suppression.

Suppression of the immune system is also associated with an increase in the development of cancers, especially lymphoproliferative disorders. This appears to be directly related to the total exposure to immunosuppression.

The risk of premature death due to cardiovascular disease is well-documented in renal transplant recipients. Much of this is due to previous damage incurred during chronic renal failure. Dyslipidaemia is common in patients with ESRF and some immunosuppressive drugs are thought to be associated with adverse lipid profiles. Hypertension and weight gain are also among the side effects of immunosuppressive drugs.

De novo post-transplant diabetes mellitus is a potentially serious treatment-related side effect. Some patients are at increased risk of this complication, for example, because of ethnic background, obesity or family history of the condition.

Nephrotoxicity is a particular complication of some immunosuppressive regimens, notably the calcineurin inhibitors, which may increase the risk of chronic graft nephropathy.

Other treatment side effects, depending on the drugs used, may include hirsutism, alopecia, tremors, mood swings or gastrointestinal intolerance. Some side effects are temporary and resolve as the body adjusts to the medication and some will continue throughout treatment.

Most treatment centres attempt to categorise donor-recipient pairs by the degree of perceived immunological risk and offer a different intensity of immunosuppression. Most adopt a different strategy for patients with delayed graft function, or recipients of non-heart-beating or live donor kidneys. Other risk factors (for acute rejection episodes) include poor HLA (Human Leukocyte Antigen) matching, high levels of antibody sensitisation, prolonged graft cold ischaemia times, black donor-recipients and whether the patient has received a previous kidney transplant.

Immunosuppression can be categorised as follows:

• prevention of graft rejection
  - induction therapy
  - initial therapy
  - maintenance therapy

• treatment of established acute rejection episodes.

Induction therapy is a short course of intensive immunosuppression immediately post-operatively (and often started pre-operatively) with the aim of 'switching off' the immune system post-transplantation to reduce the likelihood of accelerated rejection and acute rejection (treatment is for approximately 2 weeks). It has also been used as a means of reducing exposure to calcineurin inhibitors in the early stages post-transplantation when the graft may be particularly vulnerable to their nephrotoxic effects. Traditionally, the term induction therapy has been linked with the use of the following agents - the polyclonal antibodies antithymocyte immunoglobulin (ATG) and antilymphocyte immunoglobulin (ALG) and the monoclonal antibody muromonab-CD3 (previously known as OKT3). Induction therapy with these agents has been used extensively in the USA but its use has been more limited in the UK where the agents' side effects are considered unacceptable. For this reason the scope stated that 'placebo' or 'no induction drug' would be an acceptable comparison for the newer induction therapies in addition to the three drugs listed above.

Initial therapy is the treatment given to all recipients (except where the donor is an identical twin) for 0-3 months post-transplantation. Initial therapy is usually 'triple therapy', in which a calcineurin inhibitor (traditionally ciclosporin) is used as the 'primary agent' in combination with a corticosteroid (prednisolone) and azathioprine. Occasionally dual therapy (ciclosporin plus corticosteroid) is used. Both of these regimens were stated as relevant comparators in the scope.

Maintenance therapy is the treatment patients receive long-term, during the entire survival of the allograft. Often maintenance therapy is identical to initial therapy but at a reduced dosage because the transplanted kidney becomes immunologically more stable with increasing time. However, it is also not uncommon for agents used in maintenance therapy to be altered in response to the development of acute rejection, severe infections or toxicity. Poor tolerability leading to non-adherence with drugs may be another reason for changing therapies.

Acute rejection therapy. Maintenance therapies are sometimes adjusted either temporarily or permanently following acute rejection and especially following multiple rejection episodes. However, short courses of high-dose corticosteroids are the standard treatment for episodes of acute rejection. In the majority of cases corticosteroids will treat the problem quickly and effectively, although it is not unusual for two courses of corticosteroids to be required. If acute rejection does not resolve after treatment with corticosteroids it will be defined as 'corticosteroid-resistant acute rejection'. Corticosteroid-resistant acute rejection may be treated with polyclonal antibodies ALG or ATG or monoclonal antibody muromonab-CD3.

Basiliximab

Basiliximab is a CD25 monoclonal antibody. It is licensed as an induction therapy for the prophylaxis of acute organ rejection in de novo allogenic renal transplantation in adult and paediatric patients. It is used concomitantly with ciclosporin microemulsion and corticosteroid-based immunosuppression in patients with panel reactive antibodies less than 80%, or in a triple maintenance immunosuppressive regimen containing ciclosporin microemulsion, corticosteroids and either azathioprine or mycophenolate mofetil. The standard total dosage is 40 mg given in two doses of 20 mg each. In paediatric patients weighing less than 35 kg, the recommended total dosage is 20 mg given in two doses of 10 mg each. In paediatric patients weighing 35 kg or more, the recommended dosage is the adult dosage, i.e. a total of 40 mg, given in two doses of 20 mg each.

One dose of basiliximab costs approximately £840 (excluding VAT; British National Formulary, 44th edition). A two-dose course therefore costs approximately £1680 (excluding VAT).

Daclizumab

Daclizumab is also a CD25 monoclonal antibody used as an induction agent in the prophylaxis of acute rejection. It is licensed as an induction therapy for the prophylaxis of acute organ rejection in de novo allogenic renal transplantation used concomitantly with an immunosuppressive regimen, including ciclosporin and corticosteroids in patients who are not highly immunised. The recommended dose for daclizumab in adult and paediatric patients is 1 mg/kg. It should initially be given at least 24 hours before transplantation. The next, and each subsequent, dose is given at intervals of 14 days, for a total of five doses.

The Appraisal Committee considered evidence from a number of sources (see Appendix B).

Basiliximab

A total of eight randomised controlled trials (RCTs) were included in the Assessment Report. Four of the RCTs compared basiliximab with placebo within regimens comprising either ciclosporin/corticosteroid dual therapy (two trials) or ciclosporin-based triple therapy (two trials) with either azathioprine or mycophenolate mofetil as the antiproliferative component. One trial compared basiliximab with no induction agent in a ciclosporin/corticosteroid/azathioprine triple regimen with a third group receiving ciclosporin/corticosteroid/mycophenolate mofetil triple therapy. The three remaining trials compared basiliximab with another induction agent (either ATG or muromonab-CD3); one of these also included a no-induction-therapy arm.

Daclizumab

The side effect profile of mycophenolate mofetil is different to that of azathioprine. In comparative clinical trials, there was a higher incidence of gastrointestinal adverse events (diarrhoea and bleeding) and cytomegalovirus infection in the mycophenolate mofetil groups, but a lower incidence of nausea, thrombocytopenia and jaundice than in the azathioprine groups.

Four of the mycophenolate mofetil RCTs reported longer-term results up to 3-years follow-up. Results were pooled from three of the trials. Although there was some evidence of a reduction in graft loss and all cause mortality with mycophenolate at 3 years, the differences were not statistically significant (OR 0.62, 95% CI 0.34 to 1.13, and OR 0.77 95% CI 0.47 to 1.26 respectively).

Tacrolimus

Thirteen RCTs comparing tacrolimus with either of the two ciclosporin formulations were included in the Assessment Report. Six of these RCTs assessed the use of tacrolimus against the older oily formulation of ciclosporin (Sandimmun) whereas the remaining seven assessed its use against the newer microemulsion formulation (Neoral). Most of the RCTs assessed the use of ciclosporin or tacrolimus in combination with an antiproliferative agent (azathioprine or mycophenolate mofetil) and a corticosteroid. In some of the studies, induction therapy with antilymphocyte agents was also used.

Pooling the 1-year results for the RCTs that compared tacrolimus with the oily formulation of ciclosporin showed that the probability of biopsy-confirmed acute rejection favoured treatment with tacrolimus (OR 0.46; 95% CI 0.35 to 0.61). However, there was no statistically significant difference in the probability of all cause mortality or graft loss.

The comparisons with ciclosporin microemulsion also favoured tacrolimus for the endpoint of biopsy-confirmed acute rejection (OR 0.44; 95% CI 0.33 to 0.58). Again, there was no statistically significant difference in the probability of all cause mortality or graft loss.

Across the trials there was evidence at 12 months of an increase in the incidence of tremor and post-transplant diabetes mellitus with tacrolimus. Conversely there was a significant increase in hirsutism, hyperlipidaemia, and gingivitis with ciclosporin.

Only one of the 13 RCTs collected information on health-related quality of life, using the generic SF-36 and a disease-specific measure the 'Bergner Appearance Scale'. No statistically significant difference in SF-36 was reported between tacrolimus and the oily ciclosporin groups. However, results on the Bergner Scale statistically favoured tacrolimus.

Sirolimus

There are no RCTs comparing the licensed regimen for sirolimus with a standard calcineurin-based dual or triple therapy. The two pivotal studies compared two regimens that both included sirolimus (n = 525 and n = 246 respectively). One arm received a regimen of sirolimus initially combined with ciclosporin, followed by tapering of the ciclosporin dose to discontinuation after 2-3 months with a concomitant increase in the dose of sirolimus adjusted according to whole blood concentrations (this regimen is now licensed). The other arm received sirolimus nominally 2 mg per day with continued ciclosporin (the marketing authorisation specifically prohibits this regimen). In both studies, no statistically significant differences in the incidences of biopsy-confirmed acute rejection were found. However, in both studies, renal function was significantly better in the group in which ciclosporin was withdrawn from the regimen. Also, in both studies hypertension was reported to be significantly less frequent in the group that discontinued ciclosporin. The results from a third RCT, in which the use of sirolimus plus low-dose steroids as a maintenance regimen with or without low-dose ciclosporin adjunctive therapy was evaluated, were marked 'commercial in confidence' as the manuscript was unpublished.

Other studies have compared sirolimus 2 mg or 5 mg daily versus either azathioprine 2-3 mg/kg per day (one study, n = 719) or placebo (one study, n = 576). All patients received concomitant ciclosporin and corticosteroids. In both studies, the incidence of acute rejection was lower in the groups that received sirolimus. However, since there was no attempt to withdraw ciclosporin, these studies are not directly relevant to estimating the effectiveness of the licensed regimen.

Two smaller RCTs compared sirolimus-based triple therapy with ciclosporin-based triple therapy. The rate of biopsy-confirmed acute rejection and levels of serum creatinine were not statistically significantly different in either trial at 1 year. Neither sirolimus-treatment arm included an initial phase where people received sirolimus and ciclosporin as per the marketing authorisation.

Cost effectiveness

Basiliximab

Eight economic evaluations of basiliximab were included in the Assessment Report.

Six of the eight evaluations were published studies (at the time of submission). Two of these six compared the addition of basiliximab or ATG to initial therapy regimens. The remaining four published evaluations assessed the addition of basiliximab alone to ciclosporin-based immunosuppressant regimens. Four of the published studies were cost-consequence analyses, one was a cost-effectiveness analysis and one a cost-utility analysis. One cost-consequence study was conducted from an NHS perspective; the remaining five studies were all non-UK based evaluations.

The Assessment Group re-ran this model with a number of different assumptions, notably the inclusion of the results from the systematic review of clinical effectiveness. When basiliximab was added to ciclosporin, azathioprine and a corticosteroid, it remained less costly and more effective. However, the addition of basiliximab to ciclosporin, mycophenolate mofetil and a corticosteroid produced a cost per incremental QALY of about £176,000.

The Assessment Group's own economic model showed the adjunctive use of basiliximab to be less costly and more effective than ciclosporin-based triple therapy alone.

Daclizumab

Four economic evaluations of daclizumab were available to the Committee, including two from the Assessment Group.

One (non-UK) published economic evaluation for daclizumab was identified. It compared the cost effectiveness of adding daclizumab to a variety of different drug combinations, including ciclosporin-based triple therapy.

Few details of the results are available in the Assessment Report but it states that at 10 years, the cost effectiveness of daclizumab plus triple therapy was 'superior' compared with triple therapy alone.

The sponsor estimated the 1-year cost effectiveness of the licensed five-dose daclizumab plus ciclosporin-based triple therapy regimen compared with ciclosporin-based triple therapy alone to be £153,000 per QALY gained. However, for a two-dose regimen, they provided an estimate of £8400 per QALY gained; again at 1 year. Costs were restricted to those incurred by the NHS.

The Assessment Group's original evaluation predicted the licensed five-dose daclizumab plus ciclosporin-based triple therapy regimen compared to ciclosporin-based triple therapy alone to be more clinically effective and less costly compared with no induction therapy. The Assessment Group's second economic evaluation produced the same result.

Tacrolimus

Ten economic evaluations of tacrolimus were contained within the Assessment Report. Six of the ten evaluations were published studies, two were from the drug sponsor and the Assessment Group undertook two evaluations.

All of the published economic evaluations assessed the use of tacrolimus versus ciclosporin, both in combination with other agents. Although all six studies included healthcare costs only, four of them were performed specifically from an NHS perspective. Four of the published studies evaluated the oily formulation of ciclosporin (Sandimmun) and the remaining two the microemulsion formulation (Neoral).

Three of the six evaluations were either cost-effectiveness or cost-utility analyses. The first of the three studies modelled the cost-utility of tacrolimus over 1- and 10-year periods, producing incremental cost-effectiveness ratios of £120,000-£220,000 and £75,000 per QALY respectively. The second evaluation produced a cost-effectiveness estimate of £30,000 per additional graft saved or patient death avoided, although the time period for this evaluation is unclear.

The remaining published cost-effectiveness analysis was based on 6 months' clinical data, which was collected retrospectively from a European multi-centre clinical trial. The differing cost structures for each country resulted in variable per patient cost differences, but all analyses suggested tacrolimus was the least costly treatment option.

The three remaining published studies were cost analyses. One of these studies reported that there was no cost difference between tacrolimus and ciclosporin whereas the remaining two studies suggested that tacrolimus was the less costly treatment option.

The sponsor provided two cost-effectiveness analyses. The first evaluation was performed alongside a RCT at a single UK treatment centre. An NHS perspective was used and follow-up was for a minimum of 1 year. The results of the analysis showed that tacrolimus cost about £200 more over the follow-up period than ciclosporin but it was also associated with fewer episodes of acute rejection, leading to an incremental cost per rejection-free graft of about £6000.

The sponsor's second evaluation had a longer time horizon- ten years instead of one. The analysis, which was essentially based on a model, was performed from the perspective of a UK Transplant Unit. The results from the analysis suggested that the mean cost per additional patient death avoided was approximately £8000.

The Assessment Group's original economic evaluation suggested that the incremental cost per QALY of substituting tacrolimus for ciclosporin is approximately £80,000 to £400,000. The results from the Assessment Group's second economic model suggested that tacrolimus, azathioprine and corticosteroids are less costly and more effective than ciclosporin, azathioprine and corticosteroids. The main reason for the difference between the two estimates is that they used different costs for tacrolimus-based therapies.

Mycophenolate mofetil

Ten economic evaluations of mycophenolate mofetil versus azathioprine were included in the Assessment Report. Seven of these evaluations were published studies, one was performed by the sponsor and two by the Assessment Group.

Four of the seven published evaluations were cost-consequence analyses whereas the remaining three were cost-effectiveness analyses. All of the published evaluations included healthcare costs only, although their exact content varied, but none were UK-based evaluations. Only one of the published studies included a time horizon that was greater than 1 year.

All but two of the published cost analyses found the costs of mycophenolate mofetil to be greater at 6 months or 1 year compared with azathioprine. The one cost analysis that considered the outcomes of treatment over a longer period suggested that this cost difference was maintained at 10 years. The remaining two cost analyses estimated the short-term costs of mycophenolate mofetil to be lower than those associated with azathioprine.

Of the three published cost-effectiveness analyses, one estimated the costs and effects of mycophenolate mofetil to be superior than with azathioprine at 10 years, another analysis arrived at the same conclusion when considering outcomes over 1 year. The third produced an incremental cost per QALY estimate of approximately CAN$50,000.

The sponsor submitted a cost-utility analysis, which was performed from an NHS perspective over 1- and 10-year periods. Both analyses compared the use of mycophenolate mofetil to azathioprine in ciclosporin-based treatment regimens. Over 1 year, the estimated incremental cost per QALY of using mycophenolate mofetil instead of azathioprine was about £4500. At 10 years it was estimated to be approximately £23,000 wheciclosporin doses were also reduced by 30%. Switching to mycophenolate mofetil but not reducing the dose of ciclosporin produced an incremental cost per QALY of approximately £40,000. After critiquing the model, the Assessment Group re-ran it using alternative parameters and assumptions and estimated the incremental cost per QALY to be at least £44,000 when ciclosporin doses were not reduced.

In the Assessment Group's original economic evaluation, mycophenolate mofetil or azathioprine both combined with a calcineurin inhibitor and a corticosteroid were compared. The result showed that it cost approximately £77,000 per QALY gained if mycophenolate mofetil is used instead of azathioprine.

The results from the second evaluation showed the cost per QALY of replacing azathioprine with mycophenolate mofetil in a ciclosporin-based treatment regimen was approximately £130,000.

Sirolimus

Three economic evaluations of sirolimus were considered by the Committee. One was a published study; one was performed by the Assessment Group and the third submitted by the sponsor. All three studies included healthcare costs only.

The published study was a US-based cost analysis that compared the 1-year costs of using sirolimus instead of azathioprine. Costs included the consequences of treatment but did not include the actual cost of the study drugs. The study reported the costs of these treatments to be similar (US$122,033 versus US$126,627).

The sponsor's economic evaluation compared the cost-effectiveness of sirolimus with ciclosporin withdrawal and a corticosteroid to a standard calcineurin inhibitor-based treatment regimen. As there has not been a RCT of these treatments, treatment effects were estimated by incorporating the results from a number of other studies. The results from this analysis suggested that sirolimus was a more effective and less costly treatment option.

The Assessment Group's economic model compared the cost effectiveness of sirolimus with ciclosporin withdrawal and a corticosteroid to sirolimus, ciclosporin and a corticosteroid. The results from the analysis suggested that the withdrawal regimen was a more effective and less costly trea