skip navigation nih record
Vol. LX, No. 12
June 13, 2008

next story

Era of Personalized Prevention Comes with Challenges

On the front page...

Although the era of personalized cancer prevention is already under way, the decisions people will have to make regarding their own risk of disease are likely to get no easier.

Epidemiologists are getting better at predicting who will come down with which ailment when—which is an undoubted benefit. But their studies, while offering a clearer picture of what is likely to occur, will require patients to exhibit the wisdom of Solomon in making health care choices.

That was the take-home message of Stanford epidemiologist Dr. Alice Whittemore’s recent Wednesday Afternoon Lecture. Her talk, “Personalized Cancer Prevention,” was the 14th Robert S. Gordon, Jr. Lecture, held annually to commemorate the late Dr. Gordon, an esteemed NIH scientist.


  Dr. Alice Whittemore  
  Dr. Alice Whittemore  

In a series of vignettes involving three hypothetical patients, Whittemore illustrated just how savvy patients will have to be when confronted with their own matrix of risks.

Consider just one, a 71-year-old political scientist, female, with a risk of stroke, breast cancer and bone fracture who must choose between therapies that will either boost or lower her risk of each of the three ailments. She must decide, along with her physician, whether it’s worth it to take a drug that will lower the breast cancer risk a bit, but also raise her risk of stroke. Or she could opt to lower the stroke risk (she had already made it clear that her brain is her source of livelihood) at a cost of up-tweaking her risk of breast cancer.

Ultimately, in Whittemore’s story, the professor elects simply to address the risk of fracture with a medication that will have no [known] effect on her risk of cancer or stroke. That seems the safest, most rational course.

But one man gathers what another man spills, acknowledges Whittemore, and what may seem a rational choice to Patient A might be unacceptable to Patient B.

Mammography guidelines offer an example of Whittemore’s point: Up until 5 years ago, a “one size fits all” mentality prevailed with respect to the usefulness of the procedure—most medical authorities recommended mammograms at least to age 70, if not for the entire lifecourse.

“Now, the recommendations are based on life expectancy and personal preferences,” Whittemore said. “We’re not all the same on these.”

She cited a paper from 2001 showing the senselessness of state-mandated mammograms in populations that routinely include patients who would never live to see the benefit of mammography, due to other ailments. There is, after all, such a thing as a “clinically unimportant cancer.”

Whittemore accepts the Gordon Lecture memento from Dr. Barry Kramer, NIH associate director for disease prevention. Behind them is the poster commemorating the event.
Whittemore accepts the Gordon Lecture memento from Dr. Barry Kramer, NIH associate director for disease prevention. Behind them is the poster commemorating the event.

Addressing the current popularity of personal genome sequencing, Whittemore noted that the practice “is here and will be increasingly prevalent. But what will happen when everyone gets their genome sequenced for $1,000 and then want their physicians’ advice on what to do? This is a huge problem. We don’t know what to tell these people and we need more information to guide them.”

Whittemore believes that, armed with this information, personalized cancer prevention can be effective. Already there are a number of ways to combat a heightened risk of cancer due to family history or genetic susceptibility, including prophylactic surgery in the case of breast and ovarian cancer, MRI to detect breast cancer early and PSA screening for men at risk of prostate cancer. Patients have to decide “what is the outcome with or without intervention,” she said. “Personal preferences are very important…Science needs to provide the data, but it’s up to patients to balance choices.”

Having one’s genome sequenced is not entirely without merit, Whittemore noted. “It can offer more accurate and precise risk estimates, and allows the adoption of prevention strategies, especially in those at highest risk. But,” she concludes, “the promise of personalized prevention currently outstrips the data and risk models. Patients must often pit one adverse outcome against another and weigh multiple imprecise probabilities.”

She said patients “need decision tools” and conceded that everyone, including epidemiologists “has difficulty balancing probabilities.” In calculating benefit vs. harm, there will always be a “moveable fulcrum” owing to varying personal values.

During a brief Q&A following Whittemore’s talk, she was able to underscore some important points: there will be a need for quality control (“possibly an NIH imprimatur”) as more risk tools are developed; the profit motive will indeed affect the depiction of individual risk; and “patients will have to be much more active in decisions about their own care.” NIHRecord Icon

back to top of page