The Pediatric Rheumatology International Trials Organization/American College of Rheumatology provisional criteria for the evaluation of response to therapy in juvenile systemic lupus erythematosus: Prospective validation of the definition of improvement, Arthritis Care and Research,Volume 55, Issue 3, June 2006, Pages 355-363, N. Ruperto, A. Ravelli, S. Oliveira, et. al.

Thirty-seven pediatric rheumatologists from 27 countries studied information about 128 children with lupus and came to an agreement about whether each child was clinically improved or not improved after treatment. The goal of this was to develop precise definitions of improvement for childhood lupus that can be used in clinical trials and to help doctors better evaluate their patients in clinic.

In this, its third report, The Pediatric Rheumatology International Trials Organization (PRINTO) proposes a reproducible definition of disease improvement in childhood lupus that reflects agreement among experienced clinicians. PRINTO’s work is one of several efforts now underway to develop better scientific measurements for lupus disease activity that would help measure outcomes in clinical trials of new treatments and also be of value to doctors in charting their lupus patients’ progress.

The need for better measurements of lupus disease activity is great. Knowledge about the immune system disorders that underlie lupus has made great progress, and promising new targets for treatment have been identified. However, the actual development of new treatments for patients has lagged behind, in part because of the lack of agreed upon measures to determine whether patients with various symptoms are experiencing a flare, staying the same, or improving.

This is not a trivial problem because of how variable the symptoms of lupus can be. For example, if two patients enter a study of a new drug and one has a rash and arthritis and the other has mouth sores and arthritis, how can the the following outcomes be compared? 1.) If both patients have equal improvement of arthritis, but patient number one gets a worsening rash? 2.) If one patient gets an improvement in rash, the other gets an improvement of arthritis, but neither improves in their other symptom?

Without agreement on exactly what to measure, how to interpret the measurements, and how to decide whether taking a given drug would be worth it for the majority of patients (or whether it should only be used in some clear subset of lupus patients), pharmaceutical companies that want to do clinical trials to develop new treatments for lupus are left without clear benchmarks to see whether the agent will work or not. This is a particularly vexing problem for research on lupus, owing to the large variety of symptoms that patients can get, the way these symptoms can come and go on their own, and the very different responses that two patients can have to the same treatment, even patients who have the same types of symptoms.

Established in 1996, PRINTO is a group of lupus clinicians and researchers, primarily from the European Union, but with representation from North and South America and Asia as well. Earlier phases of the work, reported in 2003 and 2005, led to the development of a list of types of evaluations that might be useful to measure patients’ responses to treatments. These included a range of ways in which lupus can be looked at, from complicated scales that doctors have been using for some time, to questionnaires about the patient’s quality of life, to laboratory tests, or to simple scores given by the patient, the parents of a child-age patient, or the doctor that are known as global scores. These are done by making a mark through a line on a piece of paper to reflect how sick (how far towards the right side of the line) or how well (how far towards the left side of the line) they feel the patient is overall.

Having established those areas of assessment, the PRINTO turned its attention to optimizing the way these measures of improvement are used. Information on 128 children with lupus was evaluated by 37 experienced pediatric rheumatologists from 27 countries. A special scientific method known as “nominal group technique” was used to help the doctors discover their best areas of agreement in order to define what the overall group could accept as “clinically importantly improved” or “not improved.”

The patient profiles selected were those that were judged by conference organizers to be near a questionable threshold of improvement and most open to interpretation. That was a logical criterion, since including patient profiles at either end of the spectrum—100% improvement in all variables, or 0% improvement in any variables, for example—would easily be categorized and thus offer no real discussion or evaluation about the clinical significance of particular measurements. And these are not the problem situations that make clinical trials in lupus so difficult.

Consensus (≥80% agreement) was achieved for 109 (85%) of the 128 patients, with 70 of the 109 being judged as showing clinically important improvement. After a great deal of work the group was able to define improvement in childhood lupus as improvement in any 2 of the 5 types of assessments by at least 50% with no more than 1 of the remaining variables worsening by more than 30%.

The PRINTO is to be applauded for its work at providing a workable, reproducible definition of disease improvement for juvenile lupus.