Unfortunately, while all of these drugs represent advances for MS management, none is fully effective. The pivotal trials of all of these agents show that only limited numbers of patients were free of disease activity over each study’s duration, that this proportion was only modestly larger than that found in the trial’s placebo arm, and that for most subjects treatment was only partially effective in controlling the clinical and magnetic resonance imaging (MRI)-monitored expressions of their disease.Whatever the relative merits of these drugs, all can only be considered partially effective agents. This reality raises the difficult problem of the identification of a sub-optimal response or treatment failure in an individual case and, once identified, leads to consideration of the appropriate avenues for alternative treatments. Regrettably, primary data for evidence-based recommendations on these important concerns do not exist. Given the pressing nature of these issues, the Medical Advisory Board believes that some expert advice would be useful to help guide decision-making for the general physician confronted with this problem.

Background

Treatment failure is readily recognized when the expected effect is the rapid reversal of an obvious abnormality. Failure to reduce symptoms of bladder infection and sterilize the urine within several days of initiating antibiotic therapy is one example. Prevention of the development of a late and variable complication of a disease can be more difficult, such as when stroke occurs as a complication of hypertension. However, anticipation of future drug failure might be recognized by such treatment’s inability to reduce hypertension. The goal of current disease-modifying treatments in MS is to prevent further disability, not to reverse existing deficits.When the clinical state that the drug is expected to prevent is delayed and not precisely defined, substitute markers for treatment efficacy are often used. In relapsing MS, these include clinical attacks, which in and of themselves are of concern and importance for patients, and acute subclinical activity as monitored by MRI.While both likely contribute to disability over time, neither is highly correlated with either disability or even accumulated, persisting neurological deficits within most clinical trials of only a few years’ duration. Nevertheless, the effects of current therapies on attack rates and MRI measures of newly accumulated lesion burdens are the outcome measures that are best described by modern treatment trials, and are the events that are most readily available to the clinician when considering treatment failure or sub-optimal response in an individual patient. In relapsing MS, clinical events occur relatively infrequently, making it unlikely that treatment failure can be declared with any assurance within six months of compliant drug exposure.

The concept of ‘rescue therapy’ is deceptive as applied to MS. It implies that treatment failure or sub-optimal response can be defined and consistently identified in the individual patient, even though these concepts derive from the results of grouped data reported in relevant clinical trials.³ It also assumes that the rescue treatment is either too toxic to be considered for all patients with relapsing MS, or is itself only partially effective for the majority of patients so treated.Were rescue treatment safe, universally effective, and its protection sustained, that treatment should be the definitive first-line therapy. Even if toxic, risk–benefit considerations might also favor the ‘rescue’ treatment as a primary therapy were it highly effective in preventing disability for the vast majority of patients in a sustained manner. Rescue therapy also implies a sense of urgency, a step that if not taken expeditiously will result in irreparable harm. For MS, this suggests that the level of recent disease activity observed, despite therapy, predicts a high likelihood of impending disability if not aggressively managed. Such strong outcome predictors remain to be defined for MS.