3 studies alter approach to some cancer screening

Two big studies suggest possible new ways to screen healthy people for cervical or prostate cancers, but a third disappointed those hoping for a way to detect early signs of deadly ovarian tumors.

Researchers found:

• For women 30 and over, a test for the virus, HPV, is better than a Pap smear for predicting cervical-cancer risk, and those who test negative on both can safely wait three years to be screened again.

• A single PSA blood test at ages 44 to 50 might help predict a man's risk of developing advanced prostate cancer or dying of it up to 30 years later. The PSA test is notoriously unreliable, but using it this way separates men who need a close watch from those who are so low-risk that they can skip testing for five years or more.

• Screening women with no symptoms for ovarian cancer with a blood test and an ultrasound exam is harmful. It didn't prevent deaths and led to thousands of false alarms, unneeded surgeries and serious complications.

The last study is a warning to people who get screening tests that aren't recommended, or who question whether screening can ever hurt.

"The answer is, it could hurt a lot," said Dr. Allen Lichter, chief executive of the American Society of Clinical Oncology. The group published these and 4,000 other studies Wednesday, ahead of next month's annual meeting.

Cervical cancer is easy to prevent. It's very slow-growing, and screening finds precancerous cells and allows early treatment. The new study was the first big one to examine a newer screening tool, HPV tests, with or without Pap smears in routine practice.

For a Pap test, cells scraped from the cervix, the gateway to the uterus, are checked under a microscope. But this can miss problems or raise false alarms.

HPV tests detect the human papillomavirus, which causes most cases of cervical cancer. But HPV is "the common cold" of the nether regions — most sexually active young people have been exposed to it, said Debbie Saslow, the American Cancer Society's director of breast and gynecologic cancer. Most infections go away on their own; they're only a cancer risk if lasting a year or more.

The new study gives "very, very solid support" for screening less often, Lichter said.

Hormuzd Katki of the National Cancer Institute studied more than 330,000 women getting HPV and Pap tests through Kaiser Permanente Northern California for five years.

Only about three of 100,000 women each year developed cervical cancer after negative HPV and Pap tests. HPV tests were twice as good as Paps for predicting risk. Adding a Pap after a negative HPV test did little to improve risk prediction.

However, if an HPV test was positive, a Pap test helped confirm or rule out the need for follow-up.

The prostate study sought a better way to use PSA tests, which are troublesome because PSA can be high for many reasons besides cancer, and doctors don't know which cancers need treatment or whether screening saves lives. Most groups don't recommend PSA tests, but most men over 50 get them anyway.

Researchers at Memorial Sloan-Kettering Cancer Center in New York used stored blood samples that 12,000 Swedish men gave for a heart study decades ago, when most were 44 to 50 years old. They also had second samples from some of them six years later, and samples from other 60-year-old men.

Looking 27 years later, researchers saw that 44 percent of cancer deaths occurred in men whose initial PSAs had been in the top 10 percent when they were 44 to 50.

Conversely, scoring below the median meant very little cancer risk years later.

The results are "provocative," but this type of study can't prove that screening prevents deaths, said Dr. Matthew Smith of the Massachusetts General Hospital Cancer Center. Few of the Swedish men were treated for prostate cancer as most men are today, which can affect survival.

The government-funded ovarian-cancer study involved nearly 80,000 women. Half were screened annually with an ultrasound for four years and a blood test for six years. The blood test looked for CA-125, a substance often elevated in ovarian cancer.

After 13 years, there were no major differences in ovarian cancers found or deaths from the disease. Screening found only 212 cancers while giving 3,285 false alarms that led to 1,080 unneeded biopsy surgeries and 163 serious complications.