A new post hoc analysis from the STARMEN trial suggested that the combined assessment of 24-hour proteinuria, creatinine levels, and immunologic response at 3 months post-immunosuppressive therapy (IST) initiation can reliably predict clinical remission of nephrotic syndrome at 2 years. The combined model outperformed individual variables, regardless of the IST regimen.

The findings, which were published in Clinical Kidney Journal, could pave the way for earlier, more personalized clinical decisions in treating patients with nephrotic syndrome who have membranous nephropathy (MN).

“We found that changes in biomarkers, as early as 3 months after starting treatment, highly reliably predict clinical response at 2 years and do so better than baseline or 6-month data,” said lead author Jorge E. Rojas-Rivera, MD, of The Ohio State University Division of Nephrology Glomerular Diseases, Columbus. “Any change in therapeutic decisions will have a greater and better impact the sooner it is made. MN is no exception.”

Moving Toward More Personalized Treatment

Current prognostic markers in MN, such as baseline proteinuria or albumin levels, have limited predictive value and are often inconsistent. Previous research has relied mainly on observational and retrospective data, leaving uncertainty about which combinations of markers are most predictive and when they should be measured.

The STARMEN trial included 86 patients with biopsy-proven MN (median age, 55 years). For this post hoc analysis, investigators used logistic regression modeling to test combinations of standard clinical markers, assessing their predictive performance using receiver operating characteristic curve analysis. They found that higher baseline proteinuria or more severe hypoalbuminemia were associated with lower probability of achieving clinical remission, even after IST.

“With this information, we can decide to change the type of treatment administered early, without having to wait more months for proteinuria and positive serum antibodies, which, in the long term, are deleterious to kidney tissue and the course of the disease,” Rojas-Rivera told Medscape Medical News.

The optimal multivariate model excluding the specific immunosuppression regimen to predict complete or partial remission at 2 years included 3-month 24-hour proteinuria, serum creatinine, and immunologic response. This model achieved an area under the curve (AUC) of 0.87 (95% CI, 0.76-0.94) and 87% efficiency. For complete remission alone at 2 years, the predictive performance was lower (AUC, 0.74; 95% CI, 0.62-0.85; efficiency, 70%).

More Evidence Required

The authors emphasized that the findings are exploratory and still require formal validation.

Vimal K. Derebail, MD, MPH, professor of medicine and vice chief of clinical operations at the UNC Kidney Center in Chapel Hill, North Carolina, who was not involved in the study, said the approach is promising but not yet definitive. Derebail told Medscape Medical News that current therapies for MN can induce complete and partial remissions, but these outcomes often take a prolonged time to occur. This can make studying these outcomes in a clinical trial difficult.

“Using this combination of measures does seem to offer better and earlier predictive capacity than any one of these alone and could potentially identify patients who will have a later response and may not need continued immunosuppression.”

However, Derebail cautioned, “these outcome measures may not apply for all therapies for MN and would need to be evaluated in the context of their mechanism of action.”

Rojas-Rivera and Derebail reported having no financial disclosures.