Conference Coverage

Cancer screening in elderly: When to just say no


 

EXPERT ANALYSIS FROM THE ANNUAL INTERNAL MEDICINE PROGRAM

– A simple walking speed measurement over a 20-foot distance is an invaluable guide to physiologic age as part of individualized decision making about when to stop cancer screening in elderly patients, according to Jeff Wallace, MD, professor of geriatric medicine at the University of Colorado at Denver.

“If you have one measurement to assess ‘am I aging well?’ it’s your gait speed. A lot of us in geriatrics are advocating evaluation of gait speed in all patients as a fifth vital sign. It’s probably more useful than blood pressure in some of the older adults coming into our clinics,” he said at a conference on internal medicine sponsored by the University of Colorado.
 

 

Dr. Wallace also gave a shout-out to the ePrognosis cancer-screening decision tool, available free at www.eprognosis.org, as an aid in shared decision-making conversations regarding when to stop cancer screening. This tool, developed by researchers at the University of California, San Francisco, allows physicians to plug key individual patient characteristics into its model, including comorbid conditions, functional status, and body mass index, and then spits out data-driven estimated benefits and harms a patient can expect from advanced-age screening for colon or breast cancer.

Of course, guidelines as to when to stop screening for various cancers are available from the U.S. Preventive Services Task Force, the American Cancer Society, and specialty societies. However, it’s important that nongeriatricians understand the serious limitations of those guidelines.

“We’re not guidelines followers in the geriatrics world because the guidelines don’t apply to most of our patients,” he explained. “We hate guidelines in geriatrics because few studies – and no lung cancer or breast cancer trials – enroll patients over age 75 with comorbid conditions. Also, most of these guidelines do not incorporate patient preferences, which probably should be a primary goal. So we’re left extrapolating.“

Regrettably, though, “it turns out most Americans are drinking the Kool-Aid when it comes to patient preferences. It’s amazing how much cancer screening is going on in this country. We’re doing a lot more than we should,” said Dr. Wallace.

Dr. Jeff Wallace, professor of geriatric medicine, University of Colorado
Dr. Jeff Wallace
He highlighted a University of North Carolina study of more than 27,000 participants aged 65 years or older in the population-based National Health Interview Survey. Among those deemed at very high risk of mortality within 9 years, 55% of men had recently undergone prostate cancer screening, and 53% of women had recently had a mammogram. Up to 56% of women who underwent a hysterectomy for benign reasons had a Pap test within the previous 3 years. Moreover, more than one-third of women with less than a 5-year life expectancy had a recent mammogram (JAMA Intern Med. 2014 Oct;174[10]:1558-65).

All of that is clearly overscreening. Experts unanimously agree that if someone is not going to live for 10 years, that person is not likely to benefit from cancer screening. The one exception is lung cancer screening of high-risk patients, where there are data to show that annual low-dose CT screening is beneficial in those with even a 5-year life expectancy.

As part of the Choosing Wisely program, the American Geriatric Society has advocated that physicians “don’t recommend screening for breast, colorectal, prostate, or lung cancer without considering life expectancy and the risks of testing, overdiagnosis, and overtreatment.”

That’s where gait speed and ePrognosis come in handy in discussions with patients regarding what they can realistically expect from cancer screening at an advanced age.

The importance of gait speed was highlighted in a pooled analysis of nine cohort studies totaling more than 34,000 community-dwelling adults aged 65 years and older with 6-21 years of follow-up. Investigators at the University of Pittsburgh identified a strong relationship between gait speed and survival. Every 0.1-m/sec made a significant difference (JAMA. 2011 Jan 5;305[1]:50-8).

A gait speed evaluation is simple: The patient is asked to walk 20 feet at a normal speed, not racing. For men age 75, the Pittsburgh investigators found, gait speed predicted 10-year survival across a range of 19%-87%. The median speed was 0.8 m/sec, or about 1.8 mph, so a middle-of-the-pack walker ought to stop all cancer screening by age 75. A fast-walking older man won’t reach a 10-year remaining life expectancy until he’s in his early to mid-80s; a slow walker reaches that life expectancy as early as his late 60s, depending upon just how slow he walks. A woman at age 80 with an average gait speed has roughly 10 years of remaining life, factoring in plus or minus 5 years from that landmark depending upon whether she is a faster- or slower-than-average walker, Dr. Wallace explained.

The U.S. Preventive Services Task Force currently recommends colon cancer screening routinely for 50- to 75-year-olds, declaring in accord with other groups that this strategy has a high certainty of substantial net benefit. But the USPSTF also recommends selective screening for those aged 76-85, with a weaker C recommendation (JAMA. 2016 Jun 21;315[23]:2564-75).

What are the practical implications of that recommendation for selective screening after age 75?

Investigators at Harvard Medical School and the University of Oslo recently took a closer look. Their population-based, prospective, observational study included 1,355,692 Medicare beneficiaries aged 70-79 years at average risk for colorectal cancer who had not had a colonoscopy within the previous 5 years.

The investigators demonstrated that the benefit of screening colonoscopy decreased with age. For patients aged 70-74, the 8-year risk of colorectal cancer was 2.19% in those who were screened, compared with 2.62% in those who weren’t, for an absolute 0.43% difference. The number needed to be screened to detect one additional case of colorectal cancer was 283. Among those aged 75-79, the number needed to be screened climbed to 714 (Ann Intern Med. 2017 Jan 3;166[1]18-26).

Moreover, the risk of colonoscopy-related adverse events also climbed with age. These included perforations, falls while racing to the bathroom during the preprocedural bowel prep, and the humiliation of fecal incontinence. The excess 30-day risk for any adverse event in the colonoscopy group was 5.6 events per 1,000 patients aged 70-74 and 10.3 per 1,000 in 75- to 79-year-olds.

In a similar vein, Mara A. Schonberg, MD, of Harvard Medical School, Boston, has shed light on the risks and benefits of biannual mammographic screening for breast cancer in 70- to 79-year-olds, a practice recommended in American Cancer Society guidelines for women who are in overall good health and have at least a 10-year life expectancy.

She estimated that 2 women per 1,000 screened would avoid death due to breast cancer, for a number needed to screen of 500. But roughly 200 of those 1,000 women would experience a false-positive mammogram, and 20-40 of those false-positive imaging studies would result in a breast biopsy. Also, roughly 30% of the screen-detected cancers would not otherwise become apparent in an older woman’s lifetime, yet nearly all of the malignancies would undergo breast cancer therapy (J Am Geriatr Soc. 2016 Dec;64[12]:2413-8).

Dr. Schonberg’s research speaks to Dr. Wallace.

“It’s breast cancer therapy: It’s procedures; it’s medicalizing the patient’s whole life and creating a high degree of angst when she’s 75 or 80,” he said.

As to when to ‘just say no’ to cancer screening, Dr. Wallace said his answer is after age 65 for cervical cancer screening in women with at least two normal screens in the past 10 years or a prior total hysterectomy for a benign indication. All of the guidelines agree on that, although the American Congress of Obstetricians and Gynecologists recommends in addition that women with cervical intraepithelial neoplasia 2 be screened for the next 20 years.

For prostate cancer, Dr. Wallace recommends his colleagues just say no to screening at age 70 and above because harm is more likely than benefit to ensue.

“I don’t know about you, but I have a ton of patients over age 70 asking me for PSAs. That’s one place I won’t do any screening. I tell them I know you’re in great shape for 76 and you think it’s a good idea, but I think it’s bad medicine and I won’t do it. Even the American Urological Association says don’t do it after age 70,” he said.

For prostate cancer screening at age 55-69, however, patient preference rules the day, he added.

He draws the line at any cancer screening in patients aged 90 or over. Mean survival at age 90 is another 4-5 years. Only 11% of 90-year-old women will reach 100.

“Everybody has to die eventually,” he mused.

Dr. Wallace reported having no financial conflicts regarding his presentation.
 

Next Article: