Lupus nephritis, MMF and maintenance therapy

I love it when I have a clinical question and I'm able to find a well executed study that's exactly fits my question. It's like fitting the last piece of a puzzle.

A year ago I was referred a patient with heavy proteinuria. Initial assessment showed 7 grams of proteinuria, a cholesterol over 300, edema and an albumin of 0.7. Classic nephrotic syndrome.

Before he returned for his first follow-up appointment disaster struck. He developed chest pain and shortness of breath from a pulmonary embolism. This was a patient my age, two kids, professional. Looking at him was like looking in the mirror, but for the grace of G-d that could be me sitting in that exam chair.

black arrow points to tubuloreticular inclusions
seen in SLE and HIV.
After a month of anti-coagulation I was able to convince pulmonology and hematology to reverse the Coumadin for a few hours to get a kidney biopsy. It was membranous nephropathy with endothelial tubuloreticular inclusions. Along with consistent ANA and DS DNA we made a diagnosis of SLE WHO V and began mycophenolate mofetil. We titrated the MMF up to 3g a day and after 6 months he was in remission.

After a few months of sub nephrotic proteinuria, a normal albumin and a year of anti-coagulation he stopped his Coumadin. He has weaned his prednisone to 10 mg every other day and stopped the alendronate and rosuvastatin. Now he wants to get off the mycophenolate. Given how frightening the PE was, my preference would be to treat him forever. I casually surveyed my peers and got answers as varied as:
  • I never lower the MMF, every time I do the patient relapses
  • I taper it off after 6 months of remission
When I consulted the literature to look into this I found this paper:

Bingo. They maintained patients on MMF for 3 years at 2 grams a day with excellent results. 

Importantly there were no deaths, only 10% had serious infections and no cases of cancer occurred with the mycophenolate. 

Looks like an acceptably benign therapy with good outcomes.

Aside: While looking up how to spell endothelial tubuloreticular inclusions, I came across this paper (how I love you so Google) showing a significant number of patients with endothelial tubuloreticular inclusions that did not have lupus or HIV. I was taught that these EM findings were essentially pathognomonic for lupus. Interesting.

UPDATE: a second trial, the MAINTAIN trial (PDF), showed almost identical results. Included for completeness.