June 2011 Archives

This time German physicians are proposing that higher CSF concentrations of the soluble amyloid precursor protein beta (sAPPβ), not the endproduct amyloid beta 1-42, are better at predicting which patients with MCI will progress onto AD. Furthermore the marker appears to be useful for distinguishing AD from other dementias, like frontotemporal dementia (FTD). This new iteration of seemingly endless studies of CSF biomarkers in AD appears online in an early publication from Neurology.

The best predictive model for the development of AD considered CSF levels of sAPPβ and tau, along with age—for a sensitivity and specificity in the 80% range. For the purposes of differentiation, the 2 CSF biomarkers provided a sensitivity of 95% and a specificity of, again, about 80%. Notably CSF levels of amyloid beta 1-42, which have been used previously by others (see here, for example), "did not contribute significantly to the models."

The authors caution that the "modest number of patients" in their study, 58, and the "relatively short follow-up period," an average of ~33 months, "may have resulted in an underestimation of the predictive value of sAPPβ" [emphasis added].

CSF = cerebrospinal fluid; MCI = mild cognitive impairment.

Photograph of atrophied brain from person with AD: National Institute on Alcohol Abuse and Alcoholism.

The Human Resources Manager (2010): When an immigrant employee of Jerusalem's largest bakery is killed in a suicide bombing, the bakery's HR manager (Mark Ivanir) goes above and beyond to counter the public's perception of corporate indifference. To this end, he orchestrates a company-funded effort to return the employee's body to her remote Romanian village. Although not overtly comical, the odyssey becomes increasingly absurd, evoking farcical elements of the road trip in Little Miss Sunshine (2006).

Merck KGaA (the US subsidiary of which is Merck Serono) announced yesterday that it will cease late-stage development of oral cladribine for the treatment of multiple sclerosis. The company cites the FDA's safety (probably cancer-related) concerns about the drug. In addition, Merck KGaA will pull oral cladribine from markets in Russia and Australia, where the compound has been available under the ridiculous trade name Movectro.

Merck KGaA implies in a press release that its decision to kill oral cladribine has something to do with forthcoming safety data (or lack thereof) from continuing trials of the drug—namely, the extension phase of the CLARITY trial, the ORACLE MS study, and/or the ONWARD study. In March, the FDA issued a complete response letter, indicating that it would not approve oral cladribine on the basis of the drug's safety profile; but Merck KGaA put an exceptionally rosy spin on the news...which it has evidently now rescinded.

Oral cladribine was, at one time, in a nail-biting horse race for FDA approval with fingolimod (Gilenya), Novartis's orally administered challenger. Fingolimod was approved by the FDA for the treatment of relapsing-remitting MS in September of last year and is expected to take a sizable chunk of the market* from other disease-modifying drugs, like frontrunners interferon beta (eg, Avonex; Biogen Idec) and glatiramer acetate (Copaxone; Teva).

* Despite its staggering cost.

A high saturated-fat, high glycemic-index diet may promote age-related cognitive impairment, if CSF levels of beta amyloid 1-42 are any indication. This conclusion—and it may be a considerable leap—is based on data from a newly published study in the Archives of Neurology (subscription required). The general assumption of the study authors, from the University of Washington, is that a declining CSF level of beta amyloid 1-42 is indicative of the road to Alzheimer disease (AD)—an assumption that is generally supported by animal models and some clinical data.

In this randomized, double-blind study, 49 elderly adults consumed a high or low saturated-fat diet for 4 weeks. In those with normal cognition (n = 20), the "low diet" decreased CSF levels of beta amyloid 1-42, and the "high diet" increased CSF beta amyloid. In the 29 subjects with amnestic mild cognitive impairment (aMCI), the CSF levels of beta amyloid rose with the low diet and didn't change with the high diet. CSF levels of apolipoprotein, which (putatively) clears beta amyloid from the CNS, increased with the low diet and declined with the high diet.

Variable	Normal Cognition		aMCI
	Low Diet	High diet	Low Diet	High Diet
Beta-amyloid 42	↓	↑	↑	—
Apolipoprotein E	↑	↓	↑	↓

The authors conclude that the observed pattern of a rising CSF level of beta amyloid 1-42 with the high-fat diet in cognitively normal subjects is a sign of these individuals are approaching a "tipping point" toward the CNS deposition of amyloid and the formation of amyloid-laden plaques in the brain. Conversely the rising beta-amyloid level in aMCI patients who consumed a low-fat diet indicates that this group is moving back toward the "tipping point," the authors argue, and away from the AD-consistent pattern of lower CSF beta amyloid.

The authors' explanation for their results amounts to considerable speculation, which nevertheless may be correct. What is potentially troubling, however, about their data is that CSF levels of tau (both total and phosphorylated) did not change in any group. Rising CSF levels of tau, along with declining levels of beta amyloid 1-42, have been proposed as an "AD signature." This pattern was observed in 72% of subjects with MCI in a heavily publicized study from last year (which was also printed in the Archives of Neurology).

CSF = cerebrospinal fluid.

Photograph of atrophied brain from person with AD: National Institute on Alcohol Abuse and Alcoholism.

More favorable publicity for the Mediterranean diet re the reduction of vascular risks; but way-big caveat emptor.

"Intensive use" of olive oil—meaning its use in cooking and dressing*—appeared to lower the risk of stroke over the course of ~5 years, by as much as 40% or so, in a newly published cohort study of 7625 elderly (>65 years) French residents. However, the 95% confidence interval for this impressive and statistically significant risk reduction (comparing intensive users with nonusers) was gaping: 6%, 63%.

In addition, the study authors attempted to correlate the mean plasma oleic acid proportion,** as an indirect indicator of olive oil consumption, with a reduction of stroke risk among the cohort. But the differences in mean proportions among the nonusers, moderate users, and intensive users—although statistically significantly different—were tiny. Specifically the respective proportion values for each of the subgroups were 19.9% vs 20.5% vs 20.7%. Nevertheless the authors were evidently able to pound out statistical significance among these values, despite the fact that the biggest absolute difference was only 0.6%. And they then were able to associate the higher values with a statistically reduced stroke risk. If nothing else, I applaud them for their statistical prowess and/or luck.

Official commenters on the study, Scarmeas and Dauchet, provide a number of caveats when interpreting these results (as do, appropriately, the study authors themselves)—not the least of which is the inability to tease out the single contribution of olive oil, which is necessarily consumed with other, potentially risk-reducing, foods.

* As opposed to one or the other or neither.
** Meaning plasma oleic acid as a proportion of plasma fatty acids.

Ulzana's Raid (1972): A poorly known and, consequently, unappreciated depiction of Native American terrorism that builds on John Ford's classic The Searchers and portends more complex character-driven westerns, like Larry McMurtry's Lonesome Dove. A US Cavalry lieutenant (a very young Bruce Davison) struggles with his Christian beliefs in the Arizona territory while chasing a highly destructive and clever Apache war chief (the titular Ulzana). A nearly unrecognizable Burt Lancaster, as the military's hired scout, provides the crusty character-based counterpoint to the lieutenant's internal conflict. The story and dialog, written by Alan Sharp, are sufficiently thoughtful to merit a remake that transcends, among other things, the movie's 1970s-western score.

Where: Mostly Germany (concentration in northwest Germany), with about 30 cases in Sweden, 12 in Denmark, and a handful in various other European countries. The CDC has reported 4 suspected or confirmed cases of infection in the United States (3 with HUS), and Canada has reported 1 suspected case (without HUS).

A more-than-respectable early effort from Muscovite Alexei Popogrebski, How I Ended This Summer (2010) follows the contentious (to say the least) relationship between a slacker intern and a hotheaded veteran at a low-tech weather station on Russia's Arctic circle. The isolated atmosphere—shown in exquisite, lingering shots from DP Pavel Kostomarov—harkens back to the tense narrative threads of TV's "Lost" and, more authentically, Lord of the Flies—but on a 2-person scale. Particular props should be handed to the younger of the 2 lead actors (although both are excellent), Grigory Dobrygin, who effectively conveys a childlike, but realistic, fear of his superior (on whom he is almost entirely dependent) and a tenacious will to survive absurd hardship.

Still of more lighthearted moment for a Russian intern at an Arctic weather station from How I Ended This Summer.

A variant of magnetic resonance (MR) imaging may be more sensitive than conventional MR for detected blast-related axonal brain injury, according to a new study published online in the NEJM. The imaging variant, called diffusion tensor imaging, or DTI, measures anistrophy, or the directional diffusion of water in the brain—the idea being that water is more likely to diffuse along the path of intact axons. An abnormal DTI study would indicate a lack of directional flow, thereby suggesting some interruption or damage to axons—for instance, a shear injury from a concussive blast.

Among 63 military subjects who sustained a mild, uncomplicated, concussive-type brain injury (self-reported) in the recent line of duty, 18 (~30%) demonstrated abnormalities on DTI in 2 or more brain regions in this study. In all of these positive DTI cases except one, conventional MR images were normal. Observed DTI abnormalities, in contradistinction to those found historically in civilian cases, were more likely to be in the middle cerebellar peduncles (located in the back of the brain) and the orbitofrontal white matter (located in front of the brain)—areas "predicted to be vulnerable to primary blast injury," the study authors* reasonably conclude.

A major drawback of this study, however, is that none of the subjects had isolated primary blast exposure (meaning that they sustained other, extracranial injuries). Therefore the sole contribution of blast exposure to the DTI-detected brain abnormalities could not be determined. In addition, and most important, only about one third of subjects showed DTI abnormalities in the context of a normal MR study. Consequently the diagnosis of traumatic brain injury from a concussive blast remains a clinical one. It was also assumed, and the study results bear out this assumption in a qualified way (see the first sentence of this paragraph), that blasts, through their "shock energy" (as neurologist Allan Ropper writes in an accompanying editorial), are in and of themselves capable of producing brain injury.

Nevertheless, this study suggests that DTI is a promising method for detecting and studying concussive-type brain injury. Moreover, it can be "performed relatively quickly on the MRI scanners at US military and civilian hospitals." On a follow-up basis, specifically, DTI may aid the diagnosis and management of military personnel with concussive-type brain injuries, and it may provide some organic-based insight into post-traumatic stress disorder, the study authors (and Ropper) suggest.

* From Washington University, the Landstuhl Regional Medical Center in Germany, and the Walter Reed Army Hospital in DC.
Photo of aftermath of car bombing in Baghdad, August 2003, from the FBI web site.

A new post-hoc analysis of the RE-LY trial—the pivotal trial of Boehringer Ingelheim's Pradaxa (dabigatran)—suggests that the lower, 110-mg tablet should be available for elderly patients. In a recent online report at the Circulation website, reported yesterday by heartwire, the risk of extracranial bleeding was found to be significantly higher with the 150-mg bid dosage in patients older than 74 years of age (table below adapted from heartwire).

Endpoint	Warfarin, %	Pradaxa 110 mg bid, %	Pradaxa 150 mg bid, %	P
Major bleeding
All ages	3.57	2.87	3.31	.002*
Age <75	3.04	1.89	2.12	—
Age ≥75	4.37	4.43	5.10	<.001*
Extracranial bleeding
Age <75	2.44	1.76	1.91	—
Age ≥75	3.44	4.10	4.68	.001*
Intracranial bleeding
Age <75	0.61	0.14	0.26	—
Age ≥75	1.00	0.37	0.41	.28*

* P for interaction.

Consequently the lead author of the analysis, Eikelboom, concludes to heartwire that these new data "support the need for both doses of dabigatran to be made available in the US. The 150-mg dose will be the dose of choice in the under-75s but in older patients clinicians will want to consider the 110-mg dose to reduce the risk of extracranial bleeding."

In October of last year, the FDA approved only the 150-mg bid dosage of Pradaxa, and the agency attempted to justify its perplexing decision by providing a not-terribly-convincing argument in an April issue of the NEJM. Beasley et al rationalized that the FDA approval "was based on our inability to identify any subgroup in which use of the lower dose [110 mg bid] would not represent a substantial disadvantage." However, Eikelboom's subanalysis provides a compelling reason for approving the lower dosage (which is already available in Canada) in the United States.