SEOUL, SOUTH KOREA – A panel of urinary biomarkers may do better than measuring proteinuria in predicting which patients with lupus nephritis are going to respond to treatment, according to a presentation at an international congress on systemic lupus erythematosus.
Physician-scientist Andrea Fava, MD, of the division of rheumatology at Johns Hopkins University, Baltimore, presented data from a study using urine proteomics to identify biomarkers present in the urine of patients with lupus nephritis at 3 months after starting treatment that were linked to better outcomes from that treatment at 1 year.
While proteinuria is the standard measure used to guide decisions about whether to do a kidney biopsy and how to treat lupus nephritis, it doesn’t always correlate with what’s actually going on inside the kidney in terms of histology and inflammation, Dr. Fava said.
Bianca Nogrady/MDedge News
Dr. Andrea Fava
He pointed to an earlier study in which researchers did kidney biopsies 6 months after patients with lupus nephritis started treatment with mycophenolate. This suggested that around half of patients who showed a clinical response to treatment – defined as proteinuria below 500 mg/day – still had significant histologic disease activity. Another study suggested that this elevated histologic disease activity is associated with a risk of flare, which can result in significant nephron loss. On the flip side, nearly two-thirds of patients in complete histologic remission still had elevated proteinuria.
Unfortunately, it’s not possible or practical to biopsy patients on a regular basis, Dr. Fava said. “So we need better biomarkers, and to do so, we need better knowledge of the pathophysiology because if we have biomarkers that reflect tissue biology in real-time, that may surely guide personalized treatments,” he said at the congress.
Dr. Fava and colleagues enrolled 225 patients with SLE who were undergoing kidney biopsy and 10 healthy controls and used proteomics to quantify the urinary levels of around 1,200 proteins at baseline, 3, 6, and 12 months after initiating treatment.
The team then analyzed these data to look for protein signatures that correlated with histologic phenotypes – particularly the amount of inflammation in the kidney – and clinical features such as response to treatment.
They found several protein biomarkers that appeared to be linked to histologic activity in the kidney, including interleukin (IL)-16, CD163, and neutrophil granule proteins.
Initially, the team looked at baseline levels of these proteins to see if they predicted who responded to treatment, but found no difference between responders and nonresponders.
However, when they looked at levels at 3 months after treatment, a pattern emerged. “We found that in patients who were not responding, there were no changes after 3 months of treatment in the urine proteome,” Dr. Fava said. Among those who did respond to treatment, the levels of these proteins – IL-16, CD163, galectin-1, and CD206 – decreased significantly.
“So the proteins that are linked to renal activity decrease only in responders, suggesting that effective immunosuppression is effective in reducing intrarenal inflammation, which eventually results in low proteinuria at 1 year.”
The decline in these biomarkers persisted at 1 year, and the study suggested it was a better predictor of which patients would respond to treatment at 1 year than proteinuria.
Dr. Fava said in an interview that better biomarkers could revolutionize the treatment and management of lupus nephritis.
“First of all, it can shift the management strategy from treatment to prevention, because at the very beginning we can nip it in the bud maybe with very gentle treatment,” he said. Different panels of urine biomarkers could identify patients at risk of treatment failure, and also help patients to taper off their immunosuppressive therapy without an increased risk of flare. “If we have a way to tell us there’s still inflammation that needs treatment, that could change the way we do it,” he said.
He acknowledged there are significant challenges to developing these biomarkers for clinical use; one is the decision of how to define disease activity without relying on proteinuria as a measure. “Why do I want a biomarker that can predict another biomarker?” he said.
Another presentation during the same session, by Huihua Ding, MD, of Shanghai Jiao Tong University in Shanghai, China, reported on the use of urinary L-selectin to assess renal disease activity and response to treatment in a multiethnic cohort.
This study, involving 474 patients with SLE with or without renal involvement in the United States and China, found levels of urinary L-selectin were elevated only in patients with active lupus nephritis and showed patterns that correlated with renal histologic characteristics.
Dr. Eric F. Morand
Clinical rheumatologist Eric F. Morand, MD, PhD, and head of the School of Clinical Sciences at Monash University in Melbourne said one challenge with using urinary biomarkers was that it was not yet clear what these biomarkers reveal about the kidney. “It will be important to see whether this proteomic data actually link to renal outcomes,” Dr. Morand said in an interview. “I think predicting the response to treatment should be based around GFR [glomerular filtration rate] preservation, and I don’t think I’ve seen data yet that the urine biomarkers are going to tell us how to do that better.”
Dr. Morand is optimistic that urine biomarkers will one day be able to achieve that, but he stressed the importance of having urine biomarker tests available in the field at low cost. “You’re going to be doing the tests repeatedly, so therefore, you’re probably going to need to come down to a smaller list of proteins that you measure.”
Dr. Fava reported receiving support from Sanofi and Annexion Bio.