14 June 2010. A firm diagnosis of ALS takes many months while doctors eliminate possible causes for symptoms. In the research realm, this delay impedes many people from entering clinical trials until it may be too late for drugs to take effect. Four years ago, researchers met in Japan to revise ALS diagnostic criteria with changes that should get people into trials faster. Now, scientists at the Royal Hallamshire Hospital in Sheffield, UK, have shown that the new criteria do indeed provide better diagnoses. They published their results in this month’s Journal of Neurology, Neurosurgery and Psychiatry. Christopher Douglass at the hospital is first author of the paper; Christopher McDermott of the University of Sheffield is senior author. In their retrospective study, they found that the new checklist achieved more than double the diagnostic sensitivity than the old method, picking up many more people with ALS at their very first visit to the hospital.
Since 1990, clinical researchers have relied on diagnostic criteria developed at a conference in El Escorial, Spain (Brooks, 1994). These classify people into categories of possible, probable, or definite ALS, based on a bedside neurological exam and how widespread symptoms are. In 1998, scientists convened again in Warrenton, Virginia, to add electromyography, in a limited fashion, to the symptoms included in a probable ALS diagnosis (Brooks et al., 2000).
Researchers rely on these diagnostic criteria to screen people for trials and put together a cohort with similar severity of symptoms. However, many scientists found the El Escorial categories to be restrictive. Occasionally, people whose ALS is quite obvious do not meet all the criteria, noted Kerry Mills of King’s College London in an editorial accompanying the new paper. Some people with ALS succumb to their disease before even reaching “probable ALS” status (Traynor et al., 2000).
There is a need to get people into trials faster, said Richard Bedlack of the Duke ALS Clinic in Durham, North Carolina. “If you wait, nothing works,” Bedlack, who was not involved in the current study, noted it takes more than a year between first symptoms and a diagnosis that meets a trial’s rigid requirements. By then, he said, half of the person’s motor neurons are already gone.
In 2006, scientists convened again in Awaji-shima, Japan, to revise ALS diagnostic criteria (de Carvalho et al. 2008). They made two crucial changes to the El Escorial system. First, EMG analysis is no longer a secondary consideration, but carries the same weight as a physical exam. Second, the Awaji-shima group expanded the kinds of EMG data that can indicate ALS. Previously, an EMG had to show fibrillations—a signal indicating nerves have pulled away from muscle tissue—in order to count. Now, fasciculations, which indicate an earlier stage of denervation, count, too. This is an important issue, according to Angela Genge of the Montréal Neurological Institute and Hospital in Québec. “Outside of clinical trial inclusion criteria, fasciculations are considered a very strong indication of ALS,” she wrote in an e-mail to ARF. “There are times when fasciculations are so numerous on an EMG that it is difficult to find other features.” Genge was not part of the current study.
The new criteria do not make much difference to doctors diagnosing ALS, Bedlack said, because it still takes time to eliminate other possible causes of limb weakness. What they do is to minimize the time between a likely ALS diagnosis and a person’s entry into a trial. However, the new criteria are not yet widely used, wrote Merit Cudkowicz, of the Massachusetts General Hospital in Boston, in an e-mail to ARF. Cudkowicz also was not involved in the current study.
Researchers have analyzed the Awaji-shima criteria before and found they improve diagnosis specificity (Carvalho and Swash, 2009). The current study, Mills writes, describes similar results in a more routine clinical setting without extensive EMG use.
The Sheffield researchers collected clinical records and EMG data from 205 people who came to the hospital with a suspected motor neuron disease. Using information from each person’s initial exam, they categorized people according to both the El Escorial and Awaji-shima criteria. They then compared those diagnoses to the person’s final diagnosis to determine the sensitivity and specificity of each method. For the El Escorial, sensitivity was 28 percent; using Awaji-shima guidelines, it jumped to 60.7 percent. That meant identifying 35 more people with probable or definite ALS, who not only could have joined a clinical trial earlier, but also would likely have benefited psychologically from receiving a more certain diagnosis. Specificity was the same—95.9 percent—between the two methods.
These criteria help, but it will still take months to diagnose ALS, Bedlack said, because doctors must eliminate many other causes for weak limbs. What the field truly needs, he said, is a simple biomarker. For example, a blood test that determines the presence of ALS would be easy for primary care doctors to order, and could considerably speed a person’s access to therapies and trials. Such a biomarker has not been easy to find, although many researchers are looking. Prize4Life, a nonprofit based in Cambridge, Massachusetts, is offering $1 million in prize money to anyone who can come up with a valid biomarker, and has already given away some money for progress in that direction (see ARF related news story).—Amber Dance.
Douglass CP, Kandler RH, Shaw PJ, McDermott CJ. An evaluation of neurophysiological criteria used in the diagnosis of motor neuron disease. J Neurol Neurosurg Psychiatry. 2010 Jun;81(6):646-9. Abstract
Mills KR. New guidelines for the diagnosis of amyotrophic lateral sclerosis validated. J Neurol Neurosurg Psychiatry. 2010 Jun;81(6):589. Abstract