Researchers who tracked breast cancer rates in Norwegian women proposed the controversial notion on Monday that some tumors found with mammograms might otherwise naturally disappear on their own if left undetected.

But leading cancer experts expressed doubt about the findings and urged women to continue to get regular mammograms, saying this screening technique unquestionably saves lives by finding breast cancer early on when it is most treatable.

Dr. Per-Henrik Zahl of the Norwegian Institute of Public Health in Oslo and Norwegian and U.S. colleagues examined invasive breast cancer rates among nearly 120,000 women age 50 to 64 who had a mammogram — an X-ray of the breast used to find evidence of cancer — every two years over a six-year period.

They compared the number of breast cancers detected with another group of about 110,000 Norwegian women of the same age and similar backgrounds who were screened just once at the end of the six-year period.

The researchers said they expected to find no differences in breast cancer rates but instead found 22 percent more invasive breast tumors in the group who had mammograms every two years.

This raises the possibility that some cancers somehow disappear naturally, although there is no biological reason to explain how this might be, according to Zahl, whose findings were published in the Archives of Internal Medicine.

"We are the first ones to publish such a theory," Zahl said in a telephone interview. "What we say is many cancers must spontaneously disappear or regress because we cannot find them at later screenings. I have no biological explanation for this."

Mammography and breast self-examination for tumors are standard methods used for early detection of breast cancer, the leading cause of cancer deaths among women worldwide.

The American Cancer Society estimated that about 465,000 women die of breast cancer globally each year, and 1.3 million new cases are diagnosed.

"I think generally when we look at studies like this it is important to keep in mind there are some studies that change practice and others that make us think a little bit more, said Dr. Eric Winer, director of the Breast Oncology Center at Dana-Farber Cancer Institute in Boston.

"The idea that somehow these cancers go away entirely is, I would say, an intriguing hypothesis, but one we don't have a lot of evidence to support," said Winer, who was speaking on behalf of the American Society of Clinical Oncology.

In much of Europe women undergo mammograms every two years after age 50 except for in Britain where it is every three years, Zahl said. The American Cancer Society recommends that women get an annual mammogram beginning at age 40.

Bob Smith, director of cancer screening for the American Cancer Society, said Zahl's team misinterpreted the data, and expressed doubt about the idea that a significant number of breast tumors "spontaneously regress."

"I imagine there are still some people who believe the Earth is flat, but there are not very many of them," Smith said in a telephone interview. "It's not usual — it happens every day that research is published that gets it wrong."

The researchers acknowledged many doctors might be skeptical of the idea but they cited 32 reported cases of a breast cancer regressing, a small number for such a common disease.

The researchers said their findings provide new insight on what is "arguably the major harm associated with mammographic screening, namely, the detection and treatment of cancers that would otherwise regress."