Remembering Cokie Roberts and Understanding Breast Cancer

Seventeen years, mostly on the air

Cokie Roberts was diagnosed with breast cancer in 2002, at 58. She had a lumpectomy that July, then covered the midterm elections while having chemotherapy, wearing a wig on air. STAT reported that timeline in 2019.

She had already been talking publicly about screening before her own diagnosis. "I am very religious about annual mammograms and annual visits to the gynecologist," she said. She also put it this way: "Fortunately, in the course of my efforts to inform others about the disease, I learned about the benefits of early detection. Now I am the beneficiary of that information."

Her cancer came back in 2016. In the summer of 2019 she acknowledged that she was being treated, said "I am doing fine," and kept planning her work. She died in Washington, D.C., on September 17, 2019, at 75. ABC News reported the cause as complications from breast cancer. Her family's statement thanked the staff at the National Institutes of Health for her care.

That is the record. Nothing else about her treatment is public, and nothing here guesses. What her seventeen years illustrate is something the phrase "breast cancer survivor" hides: for a lot of people, this is a long relationship, not an episode.

What recurrence means

Breast cancer that comes back is described by where it comes back.

Local recurrence means the cancer has returned in the same breast or chest wall. Regional recurrence means it has returned in nearby lymph nodes. Distant recurrence means it has appeared somewhere else in the body, and that is the same thing as metastatic breast cancer.

The National Cancer Institute describes stage IV breast cancer as cancer that has spread to other parts of the body, such as the bones, liver, or lungs, and notes that this is also called metastatic breast cancer.

One point trips people up constantly. Breast cancer in the bones is not bone cancer. It is breast cancer, made of breast cells, and it is treated with breast cancer drugs. The name of a cancer comes from where it started, not where it ends up.

Why treatment can last years

NCI splits treatment into local and systemic. Local treatment, meaning surgery and radiation, is aimed at one area. Systemic treatment travels in the bloodstream and reaches cells anywhere: chemotherapy, hormone therapy, targeted therapy, and immunotherapy.

Hormone therapy explains a lot about long timelines. It works only when tumor cells carry estrogen or progesterone receptors. NCI lists tamoxifen and toremifene, aromatase inhibitors including anastrozole, letrozole, and exemestane, the receptor degrader fulvestrant, and ovarian suppression drugs such as goserelin and leuprolide.

NCI states that hormone therapy taken for five years or more can greatly reduce the risk of a new breast cancer, of recurrence, and of dying from breast cancer. That is years of daily medication long after surgery is a memory. NCI lists the usual side effects as hot flashes, night sweats, and reduced interest in sex, and notes it can disrupt the menstrual cycle before menopause.

For metastatic disease, the goal shifts. Treatment aims to control the cancer, protect function, and keep symptoms manageable for as long as possible, changing drugs when one stops working.

When to get checked

Two different lists matter here, depending on where you are starting.

If you have never had breast cancer, NCI says to contact a doctor about:

• A lump, or a thick or firm area, in or near the breast or under the arm.
• A change in the size or shape of one breast.
• Nipple discharge that is not breast milk.
• A nipple that flattens, inverts, or changes direction.
• Skin that becomes scaly, swollen, red, or darker on the breast, nipple, or areola.
• Dimpling, puckering, or a rash on the breast.
• Swelling of the whole breast with no lump.

NCI adds two warnings. Breast cancer does not usually cause pain, so a painless change is not a reassuring change. And follow up on a breast change even if your last mammogram was normal.

If you have had breast cancer before, add this second list. NCI names these as possible signs of spread: back pain, bone pain or fractures, shortness of breath, a constant dry cough, abdominal pain or swelling, jaundice, severe headaches, seizures, and vision changes. Any of them means calling the oncology team rather than waiting for the next scheduled appointment.

For screening in people without symptoms, the U.S. Preventive Services Task Force recommends mammography every two years for women aged 40 to 74, a grade B recommendation issued April 30, 2024.

What screening does, honestly

Roberts credited early detection, and the evidence supports the general claim while putting limits on it.

NCI's guidance for clinicians states that screening mammography reduces breast cancer deaths, with solid evidence in women aged 60 to 69 and fair evidence in women aged 50 to 59. It also gives the numbers needed to screen to prevent one death: about 1,904 women aged 39 to 49, about 1,339 aged 50 to 59, and about 377 aged 60 to 69.

The costs are real too. About 10% of women screened are recalled for more testing, while only 0.5% of those tested have cancer. NCI estimates that 20% to 50% of screen-detected cancers represent overdiagnosis, meaning disease that would never have caused harm.

Screening lowers the odds of dying from breast cancer across a population. It does not guarantee anything for one person, and Roberts's own long course shows why both halves of that sentence are true.

Population numbers, not personal ones

For 2026, SEER lists 321,910 new female breast cancers in the United States — about 15.2% of new cancer diagnoses — and 42,140 deaths, but the projection itself is the American Cancer Society's. Five-year relative survival across all stages is 91.9% for women diagnosed from 2016 to 2022. The median age at diagnosis is 64.

By stage, and again from the 2016 to 2022 group, SEER records five-year relative survival of 100.0% for localized disease, 87.5% for regional disease, and 33.8% for distant disease. About 64% of cases are found localized, 27% regional, and 6% distant.

Those figures average over enormous numbers of people. They cannot tell any individual how long she will live with breast cancer, and they were not built to. Seventeen years of reporting from someone who was living with it is a better reminder of that than any table.

Sources

• NCI: Breast Cancer Stages
• NCI: Hormone Therapy for Breast Cancer
• NCI: Treatment for Breast Cancer
• NCI: Signs and Symptoms of Breast Cancer
• NCI: Breast Cancer Screening (PDQ) — Health Professional Version
• SEER Cancer Stat Facts: Female Breast Cancer
• USPSTF: Breast Cancer Screening, final recommendation, April 30, 2024
• STAT: Cokie Roberts, a beacon for cancer survivors like me
• ABC News: Legendary journalist and political commentator Cokie Roberts dies at 75
• American Cancer Society: Cancer Facts & Statistics