Mycophenolate Mofetil vs. IV Cyclophosphamide as Induction Therapy for Severe Lupus Nephritis

For approximately three decades, the standard of care for lupus nephritis has been the "NIH regimen," consisting of intravenous cyclophosphamide, 1.0 g per meter squared, monthly for six consecutive months then every three months, in conjunction with high dose prednisone (initial, then tapered) for two more years. This regimen is based on studies comparing intravenous and oral cyclophosphamide, azathioprine, and prednisone alone.

Few people recall that the outcome measure of the NIH studies was doubling of serum creatinine, not end stage renal failure, and that overall survival (as opposed to renal survival) did not differ among the treatment groups. Few people also recall that the azathioprine plus prednisone group did approximately as well as the intravenous cyclophosphamide plus prednisone group. Most people do know that the differences in outcomes were not seen at two or five years, but took more than 10 years to be demonstrated. It is this 10-year outcome that has justified adoption of the NIH regimen.

Because of toxicity and because, in the real world outside NIH, failures are much more common than one might expect, physicians have sought alternatives to intravenous cyclophosphamide for some time. Thus the British have been exploring lower dose, shorter cyclophosphamide regimens, and the continental Europeans (primarily the French) have explored using azathioprine after the first six months. An Asian group reported on mycophenolate mofetil. All of these groups have claimed equivalence of outcome, and less toxicity, than with the NIH regimen.

At the recent American College of Rheumatology scientific sessions in Orlando, Ellen Ginzler, MD, and a large group of collaborators, reported an American study[1] comparing mycophenolate mofetil to intravenous cyclophosphamide for induction of renal response in severe lupus nephritis. The 24-week study was randomized but not blinded and was a crossover design. The outcome measures were serum creatinine and urinary protein.

This study recruited 140 patients from 19 separate sites. Fourteen of 71 mycophenolate patients and 4 of 69 cyclophosphamide patients achieved complete remission; 21 additional mycophenolate patients achieved partial remission versus 10 cyclophosphamide patients, statistically highly significant results. Side effects were no worse, and were possibly better in the mycophenolate group. Although this study was cast as an equivalence study, the authors in their presentation implied that the study could be interpreted as an efficacy study (and were criticized for doing so). The authors concluded that mycophenolate should be considered first-line therapy for lupus nephritis.

The study was large, of good design, and had a clear outcome. In this sense, it merits wide attention. It is important to understand, however, that this is a 24-week induction study, not a long-term renal outcomes study. The study clearly says that in the first six months of treatment for severe lupus nephritis, it is probably acceptable to use the less toxic mycophenolate in lieu of intravenous cyclophosphamide.

We do not know, however, whether the two-year or five-year or 10-year outcomes will also be equivalent. Unfortunately, to learn this will take a decade. Thus, it is extraordinarily important that this group continue to follow their patients and report on them periodically. Given the initial results of this study, and combining these results with those being reported in less well-organized studies from abroad, it is very likely that physicians treating patients with severe lupus nephritis can be less bound by the NIH regimen than they have been in the past.