A recent study (Weedon-Fekjær et al., carefully dissected and reviewed by Orac over at Respectful Insolence) has emphasized the tremendous variability in cancer growth rate. Even among otherwise outwardly similar breast cancers, the authors found, tumors grow at wildly different speeds, with doubling times ranging from a few months to a few years. These variations are hugely important with respect to both prognosis and therapy.

Why the variation? Some of the reasons have to do with the series of mutations that led to a particular tumor in the first place: certain genomic changes are associated with more or less aggressive cancers. I think it’s also likely that individual genetic differences play a major role — in other words, two tumors identical with respect to their founding mutations might grow at different rates in genetically distinct individuals, owing to differences in the tissue microenvironment, immune response, and other factors. From the relatively new field of pharmacogenetics, we’re learning that genetic differences also contribute to the variability in response to treatment: to take a simple example, some people’s liver enzymes process foreign compounds more efficiently than do others’, thus decreasing the half-life of pharmaceutical compounds in the bloodstream.

These observations are consistent with the growing awareness that in order to be effective, cancer treatment must ultimately be personalized. From Waters et al. (emphasis mine):

…We define personalized cancer prevention as a strategy that will enable each person to reduce his or her risk for lethal cancer by matching the dose, duration, and timing of an intervention with their own cancer risk profile. Most research studies provide us with data on the average person. But who is the average person anyway? The central tenet of personalized cancer prevention is that average is overrated.

In this article, we frame what are the major obstacles to developing personalized cancer-reducing interventions: the lack of validated, non-invasive stratifiers of risk; the U-shaped dose response between cancer-fighting nutrients (e.g., selenium) and DNA damage, meaning that more of a good thing is not necessarily a good thing; the relatively brief duration of interventions evaluated in human prevention trials; the challenge of finding populations in which the impact of early life interventions on the incidence of cancers affecting older adults can be studied; and the interindividual differences in gene expression that may influence a person’s response to a particular nutrient.

That second paragraph — in which the authors are focusing on personalized prevention, rather than treatment — got me thinking about aging. To what extent is aging, like cancer, an individual disease, whose idiosyncrasies in a given individual might be just as important as the features that are shared among individuals? In considering preventive measures and therapeutic corrections to age-related damage, how heavily will we need to weight individual genetic variation? Clearly, individuals age at different rates from one another, and tissues within individuals probably do too. How does this change the tactics one would bring to bear in designing a hypothetical anti-aging or life-extension pharmacopoeia?

As with cancer, there are enough universals in aging that it’s reasonable to imagine broad benefits being reaped from preventives and therapeutics designed based on these common features. But as with cancer, we must acknowledge that one size does not necessarily fit all. There will be great value in understanding how the aging process differs from person to person, and in adapting this knowledge into personalized programs to prevent and treat aging.