The list emerged from a two-year effort, similar to a project other medical specialties are undertaking, to identify procedures that do not help patients live longer or better or that may even be harmful, yet are routinely prescribed.

As much as 30 percent of health-care spending goes to procedures, tests, and hospital stays that do not improve a patient's health, according to a 2008 analysis by the nonpartisan Congressional Budget office.

"Our goal was to improve care and improve the value of the care we deliver," said Dr. Lowell Schnipper, a cancer physician at Beth Israel Deaconess Medical Center who led the task force assembled by the American Society of Clinical Oncology (ASCO). The group of more than 200 oncologists released the list from a report in its Journal of Clinical Oncology.

Although the task force emphasized that its recommendations -- winnowed from about 10 suggestions by oncologists -- were driven by medical considerations, the report makes clear that expense was a major factor. A number of cancer drugs cost nearly $100,000 but extend life a few months or not at all. Widely-used imaging tests cost up to $5,000 yet do not benefit patients.

The list has been closely guarded, with public announcements scheduled for Wednesday. Patients, advocacy groups, and policy experts contacted by Reuters were mixed in their reaction to the recommendations.

"The American people have a much higher opinion of doctors than of government bureaucrats," said Kate Nix, a policy analyst at the free-market Heritage Foundation. Whether the ASCO recommendations to withhold some tests and treatments will be seen as rationing "depends on how they are used. Will they inhibit the ability of doctors and patients to make the best decision in each case?"

PATIENTS VS. FINANCIAL INTERESTS

Advocates for cancer patients applauded the recommendations.

"I think this is a great effort from ASCO," said Dr. Otis Brawley, chief medical officer of the American Cancer Society. "They are putting their patients ahead of their own financial interests," since in the existing fee-for-service health-care system physicians make money on every test and treatment they order.

Brawley does not see the effort as leading to health-care rationing. "This is the rational use of health-care, not rationing," he said.

Other advocates agreed. "I am shocked, surprised, and pleased," said Fran Visco, president of the National Breast Cancer Coalition and a breast-cancer survivor. "Shocked because ASCO has long been loath to say, 'do less.' Pleased because it appears that they are trying to follow the science, which is nice."

Following the science, however, can lead to conclusions that do not sit well with all patients. Shari Baker, who was diagnosed with metastatic breast cancer eight years ago and has been in remission since 2007, has already felt the brunt of one of the recommendations: that patients who have been successfully treated for breast cancer and have no symptoms of cancer not undergo CT, PET, other imaging, or bone scans to check for a recurrence or spread of the disease, known as metastasis.

"As a survivor of bone metastasis -- the breast cancer had spread to my spine -- after my doctor told me I was clear those PET scans were my peace of mind," said Baker, a jewelry designer in Scottsdale, Ariz. "By the time you have symptoms it's too late."

Her insurer, guided by the same kind of studies that served as the basis for the ASCO list, would not cover the scan, citing that it provides no benefit.

ASCO recommends against routine use of four other procedures: chemotherapy for patients with advanced cancers who are unlikely to benefit; advanced imaging technologies such as CT and PET or bone scans to determine the precise stage of both early breast and prostate cancers at low risk for metastasis; and drugs to stimulate white blood cell production in patients receiving chemotherapy if they have a risk of febrile neutropenia, an often-fatal condition marked by fever and abnormally low numbers of certain white blood cells.

The supporting evidence for each recommendation is expected to surprise patients and even some physicians, since these very widely-used tests and treatments have little or no scientific basis, said Schnipper.

WITHHOLDING CHEMOTHERAPY

One recommendation likely to stir controversy, and even revive charges of "death panels," is to not use chemotherapy and other treatments in patients with advanced solid-tumor cancers such as colorectal or lung who are in poor health and did not benefit from previous chemo.

Such treatment is widespread. At one large health maintenance organization, for instance, 49 percent of patients with a common form of lung cancer but with poor "performance status" (they were largely confined to a bed or chair and capable of only limited self-care) received chemo. Research shows, however, that it is unlikely to extend their life or improve its quality.

Similarly, many patients receive three and even four kinds of chemo after not responding to earlier rounds. Yet in the largest study of its kind, only 2 percent of lung-cancer patients responded to a third form of chemo; 0 percent responded to a fourth form. In another study, withholding fourth-line chemotherapy from patients with non-small-cell lung cancer or colorectal cancer did not shorten their lives compared to that of similar patients receiving last-ditch treatments.

ASCO therefore recommends that such very ill, weak, and non-responsive patients receive only palliative and other end-of-life care. An important exception: a cancer whose molecular fingerprint makes it vulnerable to a targeted therapy such as AstraZeneca's Iressa for non-small-cell lung cancer.

"In no way do we want to deprive a desperately-ill patient of something that might be helpful," said Schnipper. And every oncologist should continue to base medical decisions on the individual patient, he said.

"But if nature is telling us that something will not help, it is our obligation to do no harm," he added. Giving chemo to a seriously ill patient whose cancer has not responded to earlier treatment "might let grandma live another few weeks. But she is very likely to be made toxic by the chemo and suffer big time. There is a real risk it will make a patient worse."

The challenge for physicians as well as insurers will be balancing the needs of the vast majority of patients against the rare cases. Studies have shown that when very ill lung-cancer patients were given a common chemotherapy, the average survival was half that of stronger patients, but nevertheless, 3.4 months.

FINDING METASTASIS

The recommendation that physicians not perform advanced imaging or bone scans in patients with early-stage breast or low-grade prostate cancer that is unlikely to spread also reflects research that contradicts what many patients believe.

In these patients, imaging such as PET and CT to search for metastasis does not improve detection of metastases or extend survival. Moreover, said Schnipper, "finding metastases early does not improve survival in breast cancer." But the expensive imaging ($2,500 to $5,000) does misdiagnose some harmless changes as cancer, leading to unnecessary invasive procedures or treatments that can shorten lives.

Again, there are exceptions. The most authoritative review found that 99 percent of patients with early-stage, seemingly low-risk prostate cancer do not benefit from imaging or bone scans. That suggests 1 percent might.

In breast-cancer patients who have been successfully treated, such as Shari Baker, neither the advanced imaging nor blood tests for molecules associated with tumors improve survival. (The tests do help patients treated for colorectal cancer, however.) Most recurrences are found through a physical exam or mammogram. Yet the advanced tests are routine.

"Why do doctors keep doing them?" Schnipper asks. "Most of them are aware of the evidence that these tests don't benefit patients. But in my own experience treating breast-cancer patients, sometimes when I go through the explanations of how these things don't help a patient will say, 'but it will help me sleep at night.' And I do it."

It is way too soon to know whether the recommendations will affect how ASCO's 30,000-plus members treat cancer patients. The group is launching a program to educate physicians as well as patients to understand which tests and treatments are more likely to be harmful than helpful.

(Reporting By Sharon Begley; Editing by Bernard Orr)

• A Permanent Economic Emergency