Although the combination of rituximab and MMF was not beneficial in a controlled clinical trial, the use of anti-TNF therapy in combination with MMF for new onset nephritis might be beneficial, given the much different kinetics of anti-TNF therapy and MMF, as long as time to response would count in assessing efficacy.13 However, as with the situation with essentially all biological response modifiers, the available experience is based on patients with nephritis refractory to standard therapy. Therefore, we advocate a randomized controlled trial of TNF blockade in class III, IV, or V lupus nephritis refractory to standard therapy.
TNF and Organ Inflammation
In MRL/lpr mice, TNF highly increases with time in the organs affected by lupus inflammation, in association with disease activity.21 Even in NZB/W mice, TNF is upregulated in late organ disease.22 TNF blockade with various agents improved organ inflammation in various mouse models, including arthritis and pneumonitis in MRL/lpr mice; arthritis, pneumonitis, and skin disease in motheaten mice; and nephritis in antibody-induced lupus in C3H.SW mice.23-25
The Role of Trials
Getting a prospective trial funded and successfully completed in the future will be a significant challenge. First, TNF blockers have gained such widespread popularity that it will likely be very difficult to convince any of the manufacturers to take a chance by studying these agents in a disease as risky and potentially problematic as SLE. Second, current standard therapy often includes cyclophosphamide, and there is strong reluctance to allow patients who have ever received cyclophosphamide into a TNF blocker trial. This reluctance is mostly based on the negative experience of Wegener’s Granulomatosis Etanercept Trial, in which an increased number of solid malignancies was found in those patients who received etanercept in combination with cyclophosphamide.14 However, we do not believe that concomitant therapy with both drugs is the same as sequential therapy.
Cyclophosphamide is an alkylating agent that produces DNA strand breaks. If malignant cells avoid death because TNF blockade takes away inflammation, tumors might result. This might even resemble the situation with ultraviolet light and TNF blockade in skin cancer, the only consistent pattern in TNF-induced cancer. In contrast, if a patient were to receive TNF blockade weeks or months after treatment with cyclophosphamide, malignant cells should have theoretically been removed, or they would have survived anyway. Only by resolving this important issue will a pivotal TNF blocker trial in lupus nephritis refractory to standard therapy have the chance to adequately recruit patients. Failure to recruit appropriate patients has already stopped the two trials in which we have participated: the European-based new-onset class V nephritis trial and the U.S.-based trial in proliferative nephritis, which excluded any patients who had ever used cyclophosphamide. Nevertheless, we still believe that an appropriate trial can and should be done.
