University of Wisconsin School of Medicine and Public Health

Two years ago, I met a woman who had been living with breast cancer for nearly 4 decades. I saw her for a colleague, who had taken on her case from 2 previous oncologists over the years. I reviewed the most recent imaging, which showed a slight progression in a metastatic mass, indicating the current therapy was no longer working.

Reviewing her chart, I learned she originally had breast cancer in 1978; it recurred in the 1980s, and she had a biopsyproven pelvic metastasis that was treated with radiation but never disappeared. She had received multiple medical treatments over the years, including multiple hormonal therapies and a few chemotherapies. Although she was reaching her eighth decade of life, her performance status remained excellent.

But the day I met her, there was evidence of progression of the metastatic mass, a gradual growth that had occurred very slowly over years. Now this woman, incredibly knowledgeable about her history and the unusual aspects of her cancer, wanted my advice on what to do next. My training and guidelines made it clear that I could select from a number of possibilities, including other chemotherapies considered standard and the more esoteric endocrine therapies she had not yet received.

One of the classic papers I assign to oncology fellows is “Natural History of Untreated Breast Cancer” by HJG Bloom.1 It describes a series of women in the 19th century who were not treated for their breast cancer, 4% of whom lived a decade after the breast cancer was found. In 1 case, a woman lived 35 years with breast cancer.

Now I had met a woman who had lived nearly 4 decades with breast cancer that, in retrospect, was not cured despite surgery and adjuvant therapy in the 1970s. Did other women live decades with incurable cancer? After paying careful attention to other unusual case histories and speaking with colleagues, I became convinced that this woman was not alone—others had survived an extreme length of time with incurable breast cancer, although 4 decades remained highly unusual.

In a typical case history, a woman with estrogen receptor—positive lobular breast cancer was treated with surgery and adjuvant therapy and then developed metastasis years later. One woman developed a pancreas mass more than 10 years after primary treatment of lobular cancer. Her physicians were concerned about pancreatic cancer, but the resected mass turned out to be metastatic breast cancer. Further endocrine therapy was provided for a few years. The woman had no evidence of disease for another decade, until she developed lower gastrointestinal bleeding and a biopsy revealed the same cancer invading the colon, 2 decades from initial presentation.

Other patterns became apparent: solitary lung nodules, resected metastatic lesions of nodes, and lesions of the bone with minimal progression over long periods. For example, 1 woman had an ovarian mass treated with hysterectomy and bilateral salpingo-oophorectomy for what was thought to be primary ovarian cancer but was in fact recurrent breast cancer. She remained without evidence of disease for many years, until she developed another lobular breast cancer.

Although the most common—and longest&mdash;extreme survivors had hormone-sensitive cancers, we also identified extreme survivors with HER2-positive cancer metastatic for more than a decade, as well as several with indolent triple-negative breast cancer with survival over 5 years. Still, these cases were usually less extreme than the decade-long histories identified with hormone-positive breast cancers.

As I spoke with colleagues, I found it important to be clear what I meant by “extreme survivors.” Many colleagues were familiar with the National Cancer Institute project looking for “extraordinary responders” and expected that I was looking for the same. However, few of my extreme survivors had ever had an extraordinary response to therapy. They had received many therapies with evidence of slow progression. Thus, there was little overlap between extraordinary responders and extreme survivors.

After speaking with advocates and students, I saw there also was some confusion about who qualifies as an extreme survivor. We are fortunate to live in an age when many patients’ breast cancer is eliminated with surgery and adjuvant treatment, and many survivors have no further signs of breast cancer. However, we were finding more unusual cases of survivors with metastatic cancer distant from the site of primary tumor—either stage IV or recurrent, metastatic tumors. The textbook answer is that these individuals will not be cured, with typical survival of a few years.

Another question arose from benchmarking this extreme survival. Typically, individuals living with metastatic breast cancer benchmarked their survival to the time the metastatic cancer became medically apparent to the present. However, in retrospect, for many of these women who originally received diagnoses of stage I to III cancer, the disease was present and not apparent between the primary therapy and metastatic recurrence. Thus, we felt it best to benchmark their survival from the time of original diagnosis. By our definition, a woman who had breast cancer 20 years ago and recurred at a metastatic site just 3 years ago would be considered an extreme survivor. The cancer had been present for 2 decades and did not take her life.

The key question we want to answer is: What allows some people with breast cancer to survive for extreme durations of time? The first idea was that this is an intrinsic feature of the cancer, perhaps related to a specific set of genes that drive the cancer’s growth. If so, we could collect tumor samples and identify the genetic drivers, which could be used to identify individuals who are likely to be extreme survivors.

Clearly, however, alternative possibilities exist. By talking with scientists, physicians, and survivor-advocates, I learned that there are disparate ideas of what drives extreme survival. Some proposed that specific cancer treatments selected or rendered in a particular order allow for extreme survival. Others suggested that survival is prolonged by medical treatments for comorbid conditions; unique features of the immune system or tumor immunogenicity in these survivors; or habits such as diet, exercise, and alternative medical practices.

In fact, a number of possibilities—some that appear more plausible than others&mdash;provide hope that, if realized, could empower others living with metastatic breast cancer to alter their fate. Thus, we concluded that all were worth a detailed evaluation.

After obtaining institutional review board approval, I reviewed charts of extreme survivors at our institution. I gathered information on individuals indicated by colleagues and through searches of our cancer registry (it turned out later that these approaches still missed a number of survivors). We identified 53 individuals who qualified as extreme survivors who are living with metastatic breast cancer more than 10 years after initial diagnosis (>5 years for hormone receptor—negative patients). The 4 longest-term survivors had lived from an original diagnosis between 1978 and 1980 and developed distant metastases in 1982, 1996, 2000, and 2007. All are alive today with distant and persistent metastatic disease. We enrolled 15 of the longest-term survivors with available tissue into a companion study to perform genome analyses of tumor and blood and examine their tissues for unusual cancer features. Thus far, we have found no specific reason for their survival, but some variation between these individuals made it clear that our analysis would have to extend to a larger cohort. At the same time, word got out about our project, and we were contacted by other survivors across the United States and beyond.

To expand our study, I teamed up with Gabrielle Rocque, MD, at the University of Alabama Birmingham School of Medicine. Dr Rocque has extensive experience evaluating Surveillance, Epidemiology, and End Results (SEER)—Medicare data and felt it would be possible to determine treatment patterns for large numbers of long-term metastatic survivors. In addition, she has developed innovative approaches to evaluate the treatment patterns of our extreme survivors.

We submitted a grant proposal to the Avon Breast Cancer Crusade to enroll large numbers of extreme survivors in the United States and internationally to complete a survey and to allow us to contact them for possible genomic and immunologic analyses. This project was supported by the Dr Susan Love Research Foundation’s Army of Women, and Metastatic Breast Cancer Alliance leaders offered critical feedback on our survey. These and other advocates insisted that we share deidentified patient-level data to ensure the maximal utility of these data in learning about metastatic breast cancer and helping other survivors.

Many physicians have extreme survivors with metastatic breast cancer in their practice. We invite these physicians to participate by notifying their patients of this project (

). People living with metastatic breast cancer need not be long-term survivors to participate, but we anticipate at least 1000 long-term survivors will do so. Those who fill out the survey will allow us to gain critical information about treatment patterns, comorbid illnesses, diet, exercise, and habits. Participation is initiated by survivors who choose to visit our website (outliers.cancer.wisc.edu) and select “Participate Now.”

We plan to recontact about 50 of the most extreme longterm survivors with available archived tumor samples to see if they would be willing to participate in the study’s second part, in which we will analyze the genes in their tumor and in their body that control immunity and tumor biology.

If we identify common features of the long-term survivors, we hope to make this information available to other people with metastatic breast cancer, leading to more longterm survivors. This information may also be useful to identify likely long-term survivors at the outset of a breast cancer diagnosis—it may turn out that these individuals should be treated in a distinct way. For example, if the intrinsic features of the tumor or immune system dictate a very indolent type of breast cancer, it may be possible to deintensify upfront treatments and rely primarily on endocrine and/or targeted therapies.

We may identify a group for whom, like the woman described at the outset, observation is a valid treatment approach—even for metastatic breast cancer.

Bloom HJG, Richardson WW, Harries EJ. Natural history of untreated breast cancer (1805-1933). Comparison of untreated and treated cases according to histological grade of malignancy. 

Articles

Mark Burkard, MD, PhD, discusses an unusual study launched at the University of Wisconsin. 

Mark Burkard, MD, PhD

Articles

What Can We Learn From "Extreme Survivors" of Breast Cancer?