RESULTS OF A RANDOMIZED, PROSPECTIVE, MULTICENTER TRIAL OF MYCOPHENOLATE MOFETIL VERSUS AZATHIOPRINE IN THE PREVENTION OF ACUTE LUNG ALLOGRAFT REJECTION¹

Abstract

Background. 

Although the use of mycophenolate mofetil (MMF) has reduced the incidence of acute rejection in heart and kidney allograft recipients, its role in lung transplantation remains controversial. Therefore, we conducted a randomized, prospective, open-label, multicenter study in lung transplant recipients to determine whether MMF decreases episodes of acute allograft rejection when compared with azathioprine (AZA).

Methods. 

Between March of 1997 and January of 1999, 81 consecutive lung transplant recipients from two centers were prospectively randomized to receive cyclosporine, corticosteroids, and either 2 mg/kg per day of AZA or 1 g twice daily of MMF. The primary study endpoint was biopsy-proven acute allograft rejection over the first 6 months posttransplant. Secondary endpoints included clinical rejection, cytomegalovirus (CMV) infection, adverse events, and survival. Surveillance bronchoscopies were performed at 1, 3, and 6 months, or if clinically indicated. Pathologists interpreting the biopsy results were blinded to the randomization. Results were analyzed according to intention-to-treat. Between group comparisons of means and proportions were made by using two sample t tests and Fisher’s exact tests, respectively. Six-month survival was calculated by the Kaplan-Meier method and compared by the log rank test.

Results. 

Thirty-eight patients were prospectively randomized to receive AZA, and 43 MMF. The incidence of biopsy proven grade II or greater acute allograft rejection at 6 months was 58% in the AZA group and 63% in the MMF group (P =0.82). The 6-month survival rates in the MMF and AZA groups were 86% and 82%, respectively (P =0.57). Rates of CMV infection and adverse events were not significantly different between the two groups.

Conclusions. 

Acute rejection rates and overall survival at 6 months are similar in lung transplant recipients treated with either MMF- or AZA-based immunosuppression.

Mycophenolate mofetil (MMF) has been viewed as a promising alternative to azathioprine (AZA) for use in solid organ transplantation. MMF selectively inhibits the proliferation of T and B cells by blocking de novo purine biosynthesis, a pathway critical for lymphocyte cell division ⁽¹⁾. Because other cell lines use both de novo and salvage pathways in purine biosynthesis, MMF may offer increased selectivity with decreased toxicity compared with azathioprine. Three large multicenter, randomized, prospective, double-blind studies in kidney transplant recipients have confirmed a significant reduction in acute rejection with MMF when compared with AZA or placebo ^(2–4).

Experience with MMF after lung transplantation remains quite limited. Only a few small retrospective comparative studies of MMF and AZA have been published ^(5–7). Although each study suggests that MMF is associated with a lower rate of acute allograft rejection than AZA, several methodologic shortcomings make it impossible to draw any definite conclusions from these studies. Most importantly, none of the studies included a randomized, prospective design. Therefore, we conducted a randomized, prospective, multicenter study in lung transplant recipients to determine whether MMF, as part of a triple drug regimen including cyclosporine and corticosteroids, leads to a decreased rate of acute rejection as compared with AZA.

METHODS

Study Design

A prospective, randomized, open-label study was conducted at Duke University Medical Center (Durham, NC) and University of Florida (Shands Hospital, Gainesville, FL) to determine the efficacy of MMF versus AZA in the prevention of acute allograft rejection in lung transplant recipients. The primary study endpoint was the incidence of biopsy-proven grade II or higher acute rejection over the first 6 months posttransplant. Pathologists interpreting the biopsy results were blinded to the randomization. Secondary endpoints included the combined incidence of biopsy-proven or clinical rejection, grade III or higher rejection, any CMV infection, invasive CMV disease (as defined by positive histology results), adverse events leading to drug discontinuation or dosage reduction, and survival over the first 6 months posttransplant. All endpoints were analyzed according to the intention-to-treat principle. A separate subset analysis of the primary study endpoint was conducted with only those patients who tolerated medication without dose reduction or discontinuation over the 6-month study period.

Patients

Before beginning the study, approval was obtained from each hospital’s Institutional Review Board for Human Studies. Informed written consent was obtained from each patient before participation in the study. All adult (18 years of age or older) lung transplant recipients providing informed consent were eligible. Over a 2-year period, from March of 1997 to January of 1999, 81 consecutive lung transplant recipients were enrolled in a prospective manner and randomly assigned to treatment with MMF or AZA.

Immunosuppression Protocol

All patients received triple immunosuppressive therapy with cyclosporine, corticosteroids, and either MMF or AZA according to randomization. None of the 81 patients received antilymphocyte induction therapy. Preoperative immunosuppression consisted of 2 to 2.5 mg/kg of cyclosporine given orally 4 hr before transplantation, 500 mg of methylprednisolone and 1 g of MMF or 2 mg/kg of AZA given intravenously at the time of reperfusion. Postoperative immunosuppression consisted of 125 mg of methylprednisolone given intravenously every 12 hr for four doses, followed by 20 mg/day prednisone, tapered by 5 mg every 3 months if the patient remained free of rejection. Postoperative cyclosporine was given either intravenously or orally with the dose strictly adjusted to maintain a level between 250 to 300 ng/ml (by high pressure liquid chromatography) during the first 6 months after transplantation. Postoperative MMF was given at a dose of 1 g twice daily (by nasogastric tube until oral medications were tolerated) and AZA was given at a dose of 2 mg/kg daily (intravenously until oral medications were tolerated). The dose of MMF or AZA was routinely held or reduced if the white blood cell count decreased below 4000 cells/mm³.

Infection Prophylaxis

All patients received antimicrobial prophylaxis with either cefepime and clindamycin or ceftazidime and vancomycin for 7 to 10 days postoperatively. Antibiotic coverage was modified in patients with cystic fibrosis or other septic lung disease to cover any known recipient preoperative pathogens and modified to cover any organisms identified from donor bronchial washings. All patients at risk for cytomegalovirus (CMV) infection (donor or recipient serology positive for CMV) received prophylaxis after transplantation with twice daily intravenous ganciclovir at 5 mg/kg for at least 2 weeks. Patients at high risk for CMV disease (donor positive/recipient negative) received prophylaxis with at least 4 weeks of intravenous ganciclovir and CMV hyperimmune globulin (CytoGam) at 150 mg/kg per dose for a total of five doses over 8 weeks. In addition, high-risk patients were continued on oral ganciclovir after completion of intravenous therapy. Patients with negative donor and recipient CMV serologic status received CMV-negative packed red blood cells (PRBC) only. Patients also received Pneumocystis carinii pneumonia (PCP) prophylaxis with trimethoprim/sulfamethoxazole three times per week or monthly aerosolized pentamidine if allergic to sulfa drugs. Patients received additional antifungal prophylaxis with inhaled liposomal amphotericin B in the early posttransplant period.

Acute Rejection

All patients underwent surveillance transbronchial biopsies at 1, 3, and 6 months after transplantation. Patients also underwent a follow-up surveillance biopsy 6 weeks after grade III or higher acute rejection was detected. Patients underwent diagnostic bronchoscopy as clinically indicated at any time after transplantation (i.e., if any clinical or radiographic signs of allograft dysfunction occurred). Biopsy-proven acute rejection was diagnosed according to International Society for Heart and Lung Transplantation criteria (ISHLT) ⁽⁸⁾. Although patients and investigators were not blind to the protocol, the biopsy results were read by a pathologist blinded to the patient’s treatment arm. Minimal acute rejection (ISHLT grade I) was not treated. Mild (ISHLT grade II) or higher grade acute rejection was treated with 500 mg of intravenous methylprednisolone daily for 3 days followed by an oral prednisone taper starting at 60 mg. Clinically diagnosed rejection was diagnosed by a decline in pulmonary function, radiographic abnormalities consistent with rejection, and a negative bronchoscopy for infection. Transbronchial biopsy was not performed in certain cases because of decreased platelet count, prolonged coagulation studies, or low arterial oxygenation.

Statistical Analyses

Continuous variables were reported as means and standard deviations. Categorical variables were reported as the number and percent with the trait. Comparisons between the patients given MMF and AZA were made using two sample t tests and Fisher’s exact tests for continuous and categorical variables, respectively. Six-month survival was analyzed using Kaplan Meier methodology, and the groups compared using the log rank statistic. All patients still alive at 6 months posttransplant were censored at 6 months. The study was designed to have a power of 80% to detect a 30% reduction in acute rejection (α=0.05) with at least 38 patients in each group.

RESULTS

Patients

A total of 81 patients were entered into the trial: 43 randomly assigned to MMF and 38 to AZA. As shown in Table 1, there were no significant differences between the baseline characteristics of the two groups with respect to gender, ethnicity, donor or recipient age, or donor or recipient CMV status. A higher percentage of CMV mismatch (donor positive/recipient negative) patients were in the MMF group. Native diseases were similar among both groups, with chronic obstructive pulmonary disease (COPD) and cystic fibrosis (CF) the most common indications for transplantation in either group. The number of single and bilateral lung transplants included in each group were also similar, although heart-lung transplant recipients (n=3) were represented only in the MMF group.

Table 1: Patient demographics^a

Clinical Outcomes

As shown in Table 2, biopsy-proven acute rejection occurred in 63% of MMF-treated patients versus 58% of AZA-treated patients (P =0.82). The rates for the combined endpoint of biopsy-proven or clinically diagnosed rejection as well as for grade III or higher rejection were also similar between the two groups (Table 2). The time to the first biopsy proven rejection or death was similar between MMF- and AZA-treated patients (P =0.61) and occurred approximately 1 month after transplantation in each group.

Table 2: Acute rejection 6 months after lung transplantation^a

Adverse Events

As shown in Table 3, rates of CMV infection and disease were not significantly different between the groups, although CMV infection and disease occurred less frequently among the MMF patients (P =0.24). Also shown in Table 3, a similar number of patients had either AZA or MMF dose reduced during the study period. Drug discontinuation, however, occurred more commonly in the MMF group (P =0.19). The reasons for dose reduction or drug discontinuation are illustrated in Table 4. Sixty-one percent (8 of 13) of MMF patients who discontinued drug subsequently tolerated AZA, and 67% (4 of 6) of AZA patients who discontinued drug subsequently tolerated MMF.

Table 3: Adverse events 6 months after lung transplantation^a

Table 4: Episodes of dose reduction or drug discontinuation^a

Subset Analysis

Rates of biopsy-proven acute rejection were also examined in the subset of patients (n=45) who tolerated either full-dose MMF or AZA without dosage reduction or discontinuation over the 6-month study period. In these patients, the rate of biopsy-proven acute rejection was not significantly different between MMF- and AZA-treated patients (61% MMF vs. 64% AZA, P =1.00).

Survival Analysis

Six-month survival was 84% for all enrolled patients and was similar between the two groups (86% MMF vs. 82% AZA, P =0.57), as shown in Figure 1. In the seven patients who died in the AZA group, death was attributed to ischemia reperfusion injury (n=4), bacterial sepsis (n=2), and posttransplant lymphoma (n=1). In the six patients who died in the MMF group, death was attributed to ischemia reperfusion injury (n=3), bacterial sepsis (n=2), and sudden death (n=1).

Figure 1: Survival for lung transplant recipients treated with mycophenolate mofetil (MMF) and azathioprine (AZA).

DISCUSSION

Initial reports of successful lung or heart-lung transplantation generally used a triple drug regimen of cyclosporine, azathioprine, and prednisone, with or without the use of antithymocyte antibodies, leading to the widespread acceptance of this combination as standard therapy ⁽⁹⁾. Despite this immunosuppressive regimen, acute allograft rejection occurs in 60–100% of all lung transplant recipients ^(10,11). Although most episodes of acute rejection resolve with augmented immunosuppression, repeated and/or high grade episodes of acute allograft rejection remain the most important identified risk factors for the development of a syndrome of chronic allograft dysfunction known as bronchiolitis obliterans syndrome (BOS) ^(12–14). BOS is characterized by progressive allograft dysfunction and is generally thought to reflect a manifestation of chronic allograft rejection. Once BOS develops, quality of life and expected survival are significantly diminished ^(12,15,16). Therefore, to improve the long-term outcomes of lung transplant recipients, alternative immunosuppressive therapies are needed which decrease the incidence of acute and/or chronic allograft rejection.

MMF is a novel immunosuppressive agent that selectively inhibits the proliferation of T- and B- cells by blocking de novo purine biosynthesis, a pathway critical for activated lymphocyte cell division. Previous studies in renal transplant recipients demonstrated that initial therapy with MMF, in conjunction with a cyclosporine-based immunosuppressive regimen, is associated with significantly decreased acute allograft rejection compared with azathioprine ^(2,3). Based on this experience, MMF has replaced azathioprine as a standard for immunosuppression in kidney allograft recipients. In addition, based on this renal data, many thoracic organ transplant programs have also shifted from azathioprine to MMF.

Recently, a double-blind, active controlled, multicenter trial comparing MMF with AZA in heart transplant recipients was completed by Kobashigawa et al ⁽¹⁷⁾. Six hundred fifty patients were randomized to receive mycophenolate mofetil or azathioprine in addition to cyclosporine and steroids. In the enrolled patients, the primary analysis was intention to treat and there was no significant difference in mortality between the two groups (42 deaths/re-transplants with mycophenolate versus 49 deaths/re-transplants with azathioprine, P =0.460). Similarly, among enrolled patients, there was no difference in the number of patients with treated rejection, biopsy-proven rejection, or biopsy-proven rejection with hemodynamic compromise.

Eleven percent of the patients were randomized but could not be given the active medication. An analysis was performed on the 578 patients who actually received one dose of study medication. For these patients, there was a significant reduction in mortality with mycophenolate (18 deaths, or 6.2%, vs. 33 deaths, or 11.4%) at 12 months (P =0.031). Similarly, the percent of patients rejection free was 34.3% in the mycophenolate group vs. 26.3% in the azathioprine group at 6 months (P =0.039). Thus, the authors conclude MMF seems superior to AZA in heart transplant patients, based on the as treated analyses and not intention to treat analyses. The use of this approach to analyze the data, however, has prompted criticism from the United States Food and Drug Administration, which argues the study demonstrates only the equivalence and not the superiority of MMF with AZA in this patient population ⁽¹⁸⁾.

Our results demonstrated acute rejection rates, rates of CMV infection, adverse events, and survival at 6 months are similar between MMF- and AZA-treated lung transplant recipients. Adverse events led to drug discontinuation more frequently in MMF-treated patients, which may explain the trend toward less CMV infection and disease in MMF-treated patients. In a separate subset analysis of patients who tolerated full-dose medication without dose reduction or discontinuation, rejection rates remained similar in each group.

Prior experience with MMF after lung transplantation is quite limited, and only a few comparative studies of MMF and azathioprine have been published. Ross et al. described a two-center nonrandomized concurrent cohort study of 22 patients ⁽⁵⁾. All patients received induction with ATG, cyclosporine, and prednisone, and either AZA or MMF. Patients treated with MMF experienced significantly fewer acute rejection episodes over the first 12 months than those treated with AZA (0.26 vs. 0.72 per 100 patient days P <0.001). The retrospective nature of the study, small numbers of patients, and differences in care of each cohort limit the strength of any conclusions. The MMF cohort, for example, underwent surveillance bronchoscopies on a regular basis, whereas the azathioprine cohort underwent bronchoscopy only for clinical indications. Although the utility of surveillance bronchoscopies is controversial, the use of this approach in one cohort, but not the other, might have contributed to differences in outcome measures observed between the two groups. In addition, the composition of the two cohorts differed significantly, with only single lung transplants in the AZA group, but 45% bilateral transplants in the MMF group.

In a retrospective cohort study that included 13 patients treated with MMF, O’Hair and colleagues found biopsy proven rejection was reduced over the first 3 months posttransplantation from 1.49 per 100 patient days in the AZA cohort to 0.85 per 100 patient days in the MMF cohort ⁽⁶⁾. The small number of treated patients, retrospective study design, and variability in immunosuppressive agents limit the ability to draw any definite comparison between the two cohorts. The authors specify that cyclosporine or tacrolimus were used for maintenance immunosuppression, but the rate of tacrolimus usage was not specified between the MMF and azathioprine group, therefore, confounding any comparison of rejection rates between the two groups.

Zukermann and colleagues also conducted a nonrandomized historical cohort study of MMF versus AZA in 38 patients ⁽⁷⁾. In this study, the rate of acute rejection per patient over the first 6 months posttransplant was reduced from 1.53 in the AZA group to 0.29 in the MMF group. Six-month survival for all patients, however, was only 71%. Given the high numbers of death and small numbers of patients, rejection rates must be interpreted with caution. In addition, 38% of all episodes of acute rejection were diagnosed clinically and not biopsy proven, introducing the possibility for bias in this nonrandomized, nonblinded retrospective study.

In contrast, our multicenter, prospective, randomized trial of MMF versus azathioprine demonstrated similar rates of biopsy-proven or clinical rejection at 6 months follow-up in patients treated with either agent. Although our results contrast with previous studies in lung transplant recipients, prior studies were limited by small numbers of patients, retrospective study design, and other flaws. The level of evidence provided by a randomized prospective clinical trial is superior to that provided by retrospective analyses and prospective clinical trials often produce results which conflict with earlier retrospective analyses ⁽¹⁹⁾ Moreover, our results are consistent with the intention to treat analyses of the large MMF study in heart transplant recipients.

Our results conflict with previous well-designed, prospective, randomized studies in kidney transplant recipients. We hypothesize that the higher baseline rate of rejection in lung transplant recipients, compared with renal transplant recipients, may contribute to the observed differences in efficacy of MMF in these different transplant populations ⁽¹⁰⁾. In one large prospective double-blind study of renal transplant recipients treated with triple immunosuppression, biopsy-proven acute rejection rates at 6 months were reduced from 35.5% with AZA to 19.9% with 2 g per day of MMF ⁽³⁾. In contrast, in lung allograft recipients treated with triple drug AZA-based immunosuppression, biopsy proven or clinical rejection occurs in 70–100% of patients within 6 months ^(9,11).

In addition, variation in absorption of MMF, especially in cystic fibrosis lung transplant recipients, may also have contributed to decreased efficacy in the lung population. Although the clinical significance of monitoring plasma MMF levels is controversial, we are currently performing additional analyses to determine the relationship between trough MMF levels and acute rejection in our MMF-treated lung transplant recipients. Preliminary studies in heart transplant recipients suggest that allograft rejection did not occur in patients with MMF trough levels greater than 3 μg/ml ⁽²⁰⁾. Although our results do not exclude a possible benefit for higher doses of MMF in the lung population, the increased rate of drug discontinuation in MMF-treated lung transplant recipients suggests tolerance of higher dosages would be poor.

In summary, MMF or AZA in conjunction with cyclosporine-based immunosuppression after lung transplantation offers equivalent rates of acute allograft rejection and survival at 6 months of follow-up. Compared with the AZA group, MMF-treated patients experienced less frequent CMV infections but more frequent adverse events leading to drug discontinuation. Although additional research is ongoing to determine whether either agent is associated with a decreased rate of BOS, the similar rates of acute rejection in either group make significant differences in BOS rates unlikely.