Could a Urine Test Predict Response to Biologics?



GLASGOW, SCOTLAND – Urine analysis might help to predict if patients with rheumatoid arthritis are likely to respond to biologic treatments, the results of a small metabolomics study suggest.

Pretreatment levels of histamine, glutamine, xanthurenic acid, and ethanolamine were consistently correlated to response to anti–tumor necrosis factor–alpha (anti-TNF-alpha) therapy at 12 weeks in the 36-patient evaluation.

Further research is needed, of course, before any practical application of the research can be confirmed, but the finding does raise the possibly that a simple urine-based dipstick test could one day be available in the physician’s office.

"TNF has huge effects on metabolism," said Dr. Sabrina Kapoor at the annual meeting of the British Society for Rheumatology. These include effects on rheumatoid cachexia, angiogenesis, the acute phase response, and the recruitment of leukocytes to sites of inflammation.

"Metabolomics assess many metabolites together in biological samples, such as urine or blood," explained Dr. Kapoor, an Arthritis U.K. clinical research fellow in the Rheumatology Research Group at the University of Birmingham (England). The questions were, could such metabolites be found in the urine of patients with arthritic disease, and if they were present, did the metabolites change in response to treatment?

Dr. Kapoor and her associates obtained frozen urine samples from 16 patients with RA and 20 with psoriatic arthritis (PsA).

The samples had been taken during a randomized, three-center clinical study investigating patient responses to treatment with infliximab or etanercept (Rheumatology 2012;51[suppl.3]:abstract O15).

Nuclear magnetic resonance (NMR) spectroscopy was used to analyze the metabolomic profiles of the urine samples that were taken before and 12 weeks after anti-TNF treatment.

All patients in the study received methotrexate and had a disease duration of more than 6 months. RA patients were rheumatoid factor (RF) positive, anti–cyclic citrullinated protein (CCP) antibody positive, or both, and had a DAS28 (Disease Activity Score based on a 28-joint count) greater than 4. PsA patients were negative for RF and anti-CCP antibodies, with three or more swollen or tender joints.

Only the samples from the patients with RA could be linked to treatment effect. All the patients with PsA had a good response to treatment according to EULAR criteria (defined as improvement in two or more of the following: tender joint score, swollen joint score, patient global score, and physician global score).

Changes in the DAS28 were used to identify patients with RA who did (n = 7) or did not (n = 9) respond to anti-TNF therapy at 12 weeks. Good responders were those who achieved a DAS28 lower than 3.2, or an improvement in score greater than 1.2.

The baseline clinical characteristics of "good responders" and "not good responders" were similar: The mean age was about 50 years, all patients were women, and all had a similar history of steroid or nonsteroidal anti-inflammatory drug use.

RA patients who responded well to anti-TNF therapy had a distinct metabolomic profile compared with those who did not exhibit a good response to treatment.

"There was a significant [P = .04] correlation between baseline metabolomic profiles in the urine samples and the extent of change in DAS28," Dr. Kapoor said.

Baseline metabolomic analysis of the urine in RA patients had 85.9% sensitivity and 85.7%, specificity to detect treatment response.

"Urine is actually a cleaner biofluid than other biofluids, such as serum," Dr. Kapoor noted in the discussion that followed her presentation, noting that there are fewer proteins involved that could interfere with the NMR.

The samples tested were randomly obtained, but there is no evidence that any special collection protocols (such as early-morning collection or dietary restrictions) would be needed, she said. There is also no suggestion that urine would need handling in a particular way, although perhaps antibacterial treatment might be needed to keep specimens fresh.

These data, of course, need to be confirmed in a much larger, independent cohort of patients before any subsequent investigation of specific collection or storage requirements.

Dr. Kapoor had no financial disclosures. Merck sponsored the original study, but the company did not sponsor the metabolomics analysis reported here.

Next Article: