Sequential Therapies for Proliferative Lupus Nephritis
http://www.100md.com
《新英格兰医药杂志》
To the Editor: The article by Contreras et al. (March 4 issue)1 on the treatment of proliferative lupus nephritis omitted some important issues. Only the median duration of treatment was provided, but the interquartile ranges would have been helpful, because it is unusual for a trial to allow for a wide variation in the duration of treatment (one to three years). Furthermore, it is not clear when the decision to stop therapy was made in the absence of side effects. The duration of follow-up should have been stated, since the power calculation was based on the development of chronic renal failure over a 5.5-year period.
The Bonferroni correction for multiple comparisons should have been applied and would have rendered most differences in the end points among the three treatment groups not significant. However, the result is still important, because it shows the noninferiority in efficacy of azathioprine and mycophenolate mofetil as compared with cyclophosphamide but an improved safety profile.
Finally, clinical trials of the treatment of proliferative lupus nephritis should include data on lupus disease activity collected with the use of one of the validated disease-activity indexes, not just data on renal outcomes.2
Chee-Seng Yee, M.R.C.P.
Caroline Gordon, M.D.
University of Birmingham
Birmingham B15 2TT, United Kingdom
p.c.gordon@bham.ac.uk
Editor's note: Dr. Gordon reports having received honoraria for consulting on the design of clinical trials in systemic lupus erythematosus from Human Genome Sciences, Bristol-Myers Squibb, Genentech, Roche, and Celltech.
References
Contreras G, Pardo V, Leclercq B, et al. Sequential therapies for proliferative lupus nephritis. N Engl J Med 2004;350:971-980.
Merrill JT. Measuring disease activity in systemic lupus: progress and problems. J Rheumatol 2002;29:2256-2257.
To the Editor: Caring for patients with organ-damaging and life-threatening disorders, which are typified by proliferative lupus nephritis, is a great challenge. In treating young women, the potential health threats posed by cyclophosphamide therapy, including the risks of infertility and malignant disease, are daunting.
We applaud Contreras et al. for examining three sequential maintenance regimens of cyclophosphamide, azathioprine, or mycophenolate mofetil. Two issues require clarification. First, the definition of remission as a decrease in the ratio of urinary protein to creatinine of less than 3 is inadequate. Reporting the proportion of patients in whom protein excretion was less than 0.3 g,1 or at least less than 1 g,2 ought to be obligatory. Second, this report calls for a multivariate analysis to dissect the contribution of prognostic factors from that of the treatment assignment in the prediction of outcome. The cyclophosphamide group fared much worse. However, this group appeared to include a disproportionate number of subjects with World Health Organization (WHO) class IV lupus nephritis, with higher double-stranded DNA antibody (anti-dsDNA) titers before induction and greater proteinuria after induction.3 Is it possible that this group had more severe disease, which, rather than only the treatment assignment, was responsible for producing the unfavorable outcome?
Allan C. Gelber, M.D., M.P.H.
Lisa Christopher-Stine, M.D.
Derek M. Fine, M.D.
Johns Hopkins University School of Medicine
Baltimore, MD 21205
References
Chan TM, Li FK, Tang CSO, et al. Efficacy of mycophenolate mofetil in patients with diffuse proliferative lupus nephritis. N Engl J Med 2000;343:1156-1162.
Gourley MF, Austin HA III, Scott D, et al. Methylprednisolone and cyclophosphamide, alone or in combination, in patients with lupus nephritis: a randomized, controlled trial. Ann Intern Med 1996;125:549-557.
Fraenkel L, MacKenzie T, Joseph L, Kashgarian M, Hayslett JP, Esdaile JM. Response to treatment as a predictor of longterm outcome in patients with lupus nephritis. J Rheumatol 1994;21:2052-2057.
To the Editor: The current unblinded study has surprising results, especially in view of the ethnic and racial characteristics of the subjects, given that black and Hispanic patients, who predominated among the subjects, are known to have severe disease and a poor outcome. The azathioprine and mycophenolate mofetil groups unexpectedly did not require higher doses of prednisone, which diminished their risk of steroid-related morbidity such as infection. The rates of relapse in the cyclophosphamide group and in the azathioprine group did not differ significantly, but there were 50 percent fewer relapses in the mycophenolate mofetil group, a finding that may be consistent with this agent's more specific renal immunosuppressive properties.
Yolanda Farhey, M.D.
Evelyn Hess, M.D.
University of Cincinnati
Cincinnati, OH 45267
The authors reply: In our study, sequential therapies consisting of intravenous cyclophosphamide induction followed by either azathioprine or mycophenolate mofetil maintenance appeared to offer better benefit-to-risk profiles than did long-term intravenous cyclophosphamide. In response to the editorial on our article by Balow and Austin,1 the use of a composite end point of death or chronic renal failure avoided an informative censoring bias resulting from the use of two competing outcomes. In renal-survival analysis, censoring death could lead to overestimation of the benefit or underestimation of the risk associated with the use of immunosuppressive regimens. We do not share the concern of Farhey and Hess with regard to the study population owing to the higher incidence of systemic lupus erythematosus and higher risk of death and chronic renal failure among Hispanic and black patients as compared with white patients.2,3,4 In an intention-to-treat analysis, higher relapse rates resulted in higher corticosteroid requirements in the intravenous-cyclophosphamide group. This result may indicate the need for higher doses of cyclophosphamide in the maintenance phase to achieve adequate disease control, which probably would come at the expense of increased adverse events.
To answer the questions of Yee and Gordon, the interquartile ranges for treatment were 20 to 33 months, 22 to 39 months, and 21 to 45 months, and the median length of follow-up was 31, 37, and 39 months in the intravenous-cyclophosphamide, mycophenolate mofetil, and azathioprine groups, respectively (not a significant difference). With the use of the Bonferroni correction for multiple comparisons and the patients receiving intravenous-cyclophosphamide as the control group, the corrected P was 0.025. Disease-activity indexes will be a very useful tool for future trials, but they had not been clearly established when our study was conceived.
In response to the comments of Gelber et al.: our study was a maintenance study, and remission was not defined as a complete resolution of renal dysfunction. However, complete remission occurred in 11 patients (3 in the intravenous-cyclophosphamide group, 4 in the azathioprine group, and 4 in the mycophenolate mofetil group) during induction (urine protein-to-creatinine ratio, <0.3; normal serum creatinine; and inactive urine sediment). Given the limited number of events, we did not report the results of the Cox multivariate analysis. Of the factors included in the analysis — group status, maintenance group, anti-dsDNA titer, and WHO class (IV vs. others) — only group status independently decreased the risk of death or chronic renal failure (azathioprine, P=0.008; mycophenolate mofetil, P=0.013).
Gabriel Contreras, M.D., M.P.H.
Oliver Lenz, M.D.
David Roth, M.D.
University of Miami School of Medicine
Miami, FL 33101
gcontrer@med.miami.edu
References
Balow JE, Austin HA III. Maintenance therapy for lupus nephritis -- something old, something new. N Engl J Med 2004;350:1044-1046.
Alarcon GC, McGwin G Jr, Bastian HM, et al. Systemic lupus erythematosus in three ethnic groups. VII. Predictors of early mortality in the LUMINA cohort. Arthritis Rheum 2001;45:191-202.
Alarcon GS, McGwin G Jr, Petri M, Reveille JD, Ramsey-Goldman R, Kimberly RP. Baseline characteristics of a multiethnic lupus cohort: PROFILE. Lupus 2002;11:95-101.
Trends in deaths from systemic lupus erythematosus -- United States, 1979-1998. MMWR Morb Mortal Wkly Rep 2002;51:371-373.
The Bonferroni correction for multiple comparisons should have been applied and would have rendered most differences in the end points among the three treatment groups not significant. However, the result is still important, because it shows the noninferiority in efficacy of azathioprine and mycophenolate mofetil as compared with cyclophosphamide but an improved safety profile.
Finally, clinical trials of the treatment of proliferative lupus nephritis should include data on lupus disease activity collected with the use of one of the validated disease-activity indexes, not just data on renal outcomes.2
Chee-Seng Yee, M.R.C.P.
Caroline Gordon, M.D.
University of Birmingham
Birmingham B15 2TT, United Kingdom
p.c.gordon@bham.ac.uk
Editor's note: Dr. Gordon reports having received honoraria for consulting on the design of clinical trials in systemic lupus erythematosus from Human Genome Sciences, Bristol-Myers Squibb, Genentech, Roche, and Celltech.
References
Contreras G, Pardo V, Leclercq B, et al. Sequential therapies for proliferative lupus nephritis. N Engl J Med 2004;350:971-980.
Merrill JT. Measuring disease activity in systemic lupus: progress and problems. J Rheumatol 2002;29:2256-2257.
To the Editor: Caring for patients with organ-damaging and life-threatening disorders, which are typified by proliferative lupus nephritis, is a great challenge. In treating young women, the potential health threats posed by cyclophosphamide therapy, including the risks of infertility and malignant disease, are daunting.
We applaud Contreras et al. for examining three sequential maintenance regimens of cyclophosphamide, azathioprine, or mycophenolate mofetil. Two issues require clarification. First, the definition of remission as a decrease in the ratio of urinary protein to creatinine of less than 3 is inadequate. Reporting the proportion of patients in whom protein excretion was less than 0.3 g,1 or at least less than 1 g,2 ought to be obligatory. Second, this report calls for a multivariate analysis to dissect the contribution of prognostic factors from that of the treatment assignment in the prediction of outcome. The cyclophosphamide group fared much worse. However, this group appeared to include a disproportionate number of subjects with World Health Organization (WHO) class IV lupus nephritis, with higher double-stranded DNA antibody (anti-dsDNA) titers before induction and greater proteinuria after induction.3 Is it possible that this group had more severe disease, which, rather than only the treatment assignment, was responsible for producing the unfavorable outcome?
Allan C. Gelber, M.D., M.P.H.
Lisa Christopher-Stine, M.D.
Derek M. Fine, M.D.
Johns Hopkins University School of Medicine
Baltimore, MD 21205
References
Chan TM, Li FK, Tang CSO, et al. Efficacy of mycophenolate mofetil in patients with diffuse proliferative lupus nephritis. N Engl J Med 2000;343:1156-1162.
Gourley MF, Austin HA III, Scott D, et al. Methylprednisolone and cyclophosphamide, alone or in combination, in patients with lupus nephritis: a randomized, controlled trial. Ann Intern Med 1996;125:549-557.
Fraenkel L, MacKenzie T, Joseph L, Kashgarian M, Hayslett JP, Esdaile JM. Response to treatment as a predictor of longterm outcome in patients with lupus nephritis. J Rheumatol 1994;21:2052-2057.
To the Editor: The current unblinded study has surprising results, especially in view of the ethnic and racial characteristics of the subjects, given that black and Hispanic patients, who predominated among the subjects, are known to have severe disease and a poor outcome. The azathioprine and mycophenolate mofetil groups unexpectedly did not require higher doses of prednisone, which diminished their risk of steroid-related morbidity such as infection. The rates of relapse in the cyclophosphamide group and in the azathioprine group did not differ significantly, but there were 50 percent fewer relapses in the mycophenolate mofetil group, a finding that may be consistent with this agent's more specific renal immunosuppressive properties.
Yolanda Farhey, M.D.
Evelyn Hess, M.D.
University of Cincinnati
Cincinnati, OH 45267
The authors reply: In our study, sequential therapies consisting of intravenous cyclophosphamide induction followed by either azathioprine or mycophenolate mofetil maintenance appeared to offer better benefit-to-risk profiles than did long-term intravenous cyclophosphamide. In response to the editorial on our article by Balow and Austin,1 the use of a composite end point of death or chronic renal failure avoided an informative censoring bias resulting from the use of two competing outcomes. In renal-survival analysis, censoring death could lead to overestimation of the benefit or underestimation of the risk associated with the use of immunosuppressive regimens. We do not share the concern of Farhey and Hess with regard to the study population owing to the higher incidence of systemic lupus erythematosus and higher risk of death and chronic renal failure among Hispanic and black patients as compared with white patients.2,3,4 In an intention-to-treat analysis, higher relapse rates resulted in higher corticosteroid requirements in the intravenous-cyclophosphamide group. This result may indicate the need for higher doses of cyclophosphamide in the maintenance phase to achieve adequate disease control, which probably would come at the expense of increased adverse events.
To answer the questions of Yee and Gordon, the interquartile ranges for treatment were 20 to 33 months, 22 to 39 months, and 21 to 45 months, and the median length of follow-up was 31, 37, and 39 months in the intravenous-cyclophosphamide, mycophenolate mofetil, and azathioprine groups, respectively (not a significant difference). With the use of the Bonferroni correction for multiple comparisons and the patients receiving intravenous-cyclophosphamide as the control group, the corrected P was 0.025. Disease-activity indexes will be a very useful tool for future trials, but they had not been clearly established when our study was conceived.
In response to the comments of Gelber et al.: our study was a maintenance study, and remission was not defined as a complete resolution of renal dysfunction. However, complete remission occurred in 11 patients (3 in the intravenous-cyclophosphamide group, 4 in the azathioprine group, and 4 in the mycophenolate mofetil group) during induction (urine protein-to-creatinine ratio, <0.3; normal serum creatinine; and inactive urine sediment). Given the limited number of events, we did not report the results of the Cox multivariate analysis. Of the factors included in the analysis — group status, maintenance group, anti-dsDNA titer, and WHO class (IV vs. others) — only group status independently decreased the risk of death or chronic renal failure (azathioprine, P=0.008; mycophenolate mofetil, P=0.013).
Gabriel Contreras, M.D., M.P.H.
Oliver Lenz, M.D.
David Roth, M.D.
University of Miami School of Medicine
Miami, FL 33101
gcontrer@med.miami.edu
References
Balow JE, Austin HA III. Maintenance therapy for lupus nephritis -- something old, something new. N Engl J Med 2004;350:1044-1046.
Alarcon GC, McGwin G Jr, Bastian HM, et al. Systemic lupus erythematosus in three ethnic groups. VII. Predictors of early mortality in the LUMINA cohort. Arthritis Rheum 2001;45:191-202.
Alarcon GS, McGwin G Jr, Petri M, Reveille JD, Ramsey-Goldman R, Kimberly RP. Baseline characteristics of a multiethnic lupus cohort: PROFILE. Lupus 2002;11:95-101.
Trends in deaths from systemic lupus erythematosus -- United States, 1979-1998. MMWR Morb Mortal Wkly Rep 2002;51:371-373.