The Trouble with Markers

FEATUREThe Longevity Dividend The Trouble with MarkersBY MICHAEL O'NEILLEvaluating a potential anti-aging therapeutic poses a unique challenge. When the endpoint is natural death, assessing efficacy in a realistic timeframe requires a surrogate, but biomarkers for aging have been elusive. ARTICLE EXTRASRelated Articles: In Pursuit of the Longevity DividendWhat should we be doing to prepare for the unprecedent

Mar 1, 2006
Nigel O'Neill
FEATURE
The Longevity Dividend

The Trouble with Markers

Evaluating a potential anti-aging therapeutic poses a unique challenge. When the endpoint is natural death, assessing efficacy in a realistic timeframe requires a surrogate, but biomarkers for aging have been elusive.

"There are no biomarkers that are very good at predicting of subsequent longevity, and that is the gold standard of what a biomarker is," says Tom Johnson of the University of Colorado at Boulder. While clinical research still rests largely on physiological status such as diastolic function, some argue that genomics or proteomics may provide more precise molecular markers.

Researchers at the Washington University School of Medicine recently showed that diastolic function in members of the Calorie Restriction Society resembled that normally seen in people 15 years younger.1 Members of this society voluntarily adhere to a nutritionally balanced, low-calorie diet (1,300 to 2,000 per day). Diastolic function generally declines as people age, says Luigi Fontana, who headed the study. Whether such physiologic indicators translate into extended life is unclear.

Don Ingram, a senior investigator at the US National Institute on Aging, says that identification of molecular biomarkers for aging is crucial both for understanding the aging process in humans and for evaluating potentially effective interventions.

Stephen Spindler, at the University of California, San Diego, uses microarray-based gene-expression analyses in experiments on long-term calorie restriction (CR) in Caenorhabditis elegans.2 A drug candidate that recapitulates the gene-expression changes seen in long-term CR might well be working. Spindler has shown that both short-term CR and the diabetes drug metformin do this to some degree. (see An Aging Drug in our Midst?)

Richard Miller, professor of pathology at the University of Michigan, sets the criteria higher. "You could call something a biomarker of aging if it were documented that those people who show rapid change in the marker also show rapid change in most other traits that change with age, and those who show slow change in the marker also change slowly in a wide range of other age-sensitive traits. As far as I know, there aren't any studies in mice or in people that meet these tough criteria."

Worms at least provide consistent results. Monica Driscoll at Rutgers University has shown that so-called age pigments correlate with lifespan in C. elegans.3 Clonal worms of the same chronological age differed markedly in their apparent health. "Some looked very healthy; others looked decrepit," Driscoll says. The decrepit worms had higher cellular amounts of age pigments, fluorescent molecules such as lipofuscin, and glycation end products that accumulate in lysosomes and generally increase with age in various metazoans, including humans.

Nevertheless, finding something definitive will remain problematic. Ingram says that the ultimate need for human trials of aging interventions "demands that the field ... become more sophisticated, using a variety of measures that can be shown to be reliable and valid."

References
1. T.E. Meyer et al., "Long-term caloric restriction ameliorates the decline in diastolic function in humans," J Am Coll Cardiol, 47:398-402, Jan. 17, 2006.
2. J.M. Dhahbi et al., "Identification of potential caloric restriction mimetics by microarray profiling," Physiological Genomics, 23:343-50, Nov. 17, 2005.
3. B. Gerstbrein et al., "In vivo spectrofluorimetry reveals endogenous biomarkers that report healthspan and dietary restriction in Caenorhabditis elegans," Aging Cell, 4:127-37, 2005.